NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Hall KK, Shoemaker-Hunt S, Hoffman L, et al. Making Healthcare Safer III: A Critical Analysis of Existing and Emerging Patient Safety Practices [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2020 Mar.
Elizabeth Schoyer , M.P.H., Kendall K. Hall , M.D., M.S., and Eleanor Fitall , M.P.H.
Preventing Clostridioides difficile infection (CDI) in healthcare settings is an important U.S. public health priority and has led to new research, guidelines, and reporting requirements that have emerged since the last version of this report, Making Health Care Safer II (MCHS II). While many of the patient safety practices (PSPs) that help prevent a range of healthcare-associated infections (HAIs) also help to prevent the transmission of CDI (e.g., contact precautions), several CDI-specific practices address the unique risk factors, pathology, and transmission of CDI.
Antimicrobial Stewardship Hand Hygiene Environmental Cleaning SurveillanceWe retrieved and screened studies that evaluated these PSPs and were published in English from 2008 onward. Many studies were quasi-experimental with a pre-post design, and most were in hospital settings (although some research was in long-term care facilities [LTCFs]).
Multicomponent CDI prevention Interventions.Multicomponent studies show outcomes associated with different combinations of CDI PSPs. They also offer insight into implementation methods, as well as challenges and facilitators of CDI prevention interventions.
Other CDI PSPs such as contact precautions and patient isolation continue to be recommended by experts 1 and were addressed briefly in the last MHCS report. Communication and staff education were also identified in the CDI PSP guidelines and are often important components of the reviewed PSPs (e.g., clinician education about revised antimicrobial prescribing guidelines and communication of CDI status after testing). Since these are cross-cutting practices and little research focused on these practices and this harm area specifically, they are discussed separately in the cross-cutting chapter of the report.
C. difficile is a contagious bacterium that can cause diarrhea, fever, colitis (an inflammation of the colon), toxic megacolon (a dilated colon that may be accompanied by septic shock), and, in some cases, death. The C. difficile bacterium colonizes in the large intestine. In infected patients, toxins produced by the organism cause CDI symptoms, primarily diarrhea and colitis. The most common risk factors for CDI are antimicrobial use, advanced age, hospitalization, and a weakened immune system. C. difficile is transmitted through the fecal-oral route and acquisition is most frequently attributed to the healthcare setting. 2 , 3
Complications are common in patients age 65 and older and an estimated 1 in 11 patients 65 and older with healthcare-associated CDI dies within 30 days of CDI diagnosis. 4 Patients with a healthy immune response to the organism can be carriers of C. difficile (and contagious) but asymptomatic. These patients are considered “colonized” and are at higher risk of developing CDI. 5
Research on CDI prevention practices has evolved and expanded over the last decade. Therefore, to address C. difficile prevention, this report dedicates an entire chapter to CDI PSPs; in the last report, much of the information on HAI PSPs was grouped together, in a more “horizontal” approach to prevention. In addition, the previous report noted the emergence of hypervirulent C. difficile strains and briefly discussed research on CDI risk prediction tools. That report noted that CDI PSPs with good supporting evidence were wearing gloves and antimicrobial stewardship. Alternatively, the current review found strong evidence that supports not just contact precautions and antimicrobial stewardship, but also environmental cleaning practices, surveillance, and testing as effective PSPs for preventing CDI.
The research reviewed in this report reflects not only new knowledge, but also new technologies and policies now in widespread use. For example, electronic health records (EHRs) are now commonly used and are valuable for antimicrobial stewardship efforts and CDI surveillance. Research on no-touch decontamination technology has grown in the last 10 years, as has understanding of CDI transmission pathways. Testing methods have also evolved, with Food and Drug Administration (FDA) approval of nucleic acid amplification tests (NAATs) in 2009. There are increased mandates for surveillance of CDI and the standard interim CDI case definitions that the CDC published in 2007 have been revised in recent years. 1 , 6 Facilities have implemented new automated surveillance systems, and CDI data collection at the national level is now standardized, with the advent of the National Healthcare Safety Network’s (NHSN’s) LabID Event reporting in 2013.
CDI is among the most common HAIs, representing roughly 12 percent of all HAIs. 7 According to a recent estimate, approximately half a million incident clinical infections occur (with more than 100,000 in U.S. nursing homes) per year in the United States, with around 30,000 deaths per year as a result of the pathogen. 3 , 4 The financial cost of CDI is also high; in recent years, CDI has resulted in about $5 billion a year in healthcare costs. 8 , 9 Costs attributable to primary and recurrent CDI are $24,205 and $10,580 per case, respectively. 10 CDI colonization is also a concern, and two U.S. studies found that around 10 percent of admitted hospital patients were colonized with C. difficile. 11 , 12
CDI incidence nearly tripled in the first decade of the 21 st century, 13 and data from 2010 to 2016 showed CDI rates plateauing. However, after falling short of 2013 reduction goals, the Department of Health and Human Services set a target reduction of 30 percent in hospital-onset CDI from 2015 to 2020. 14 Healthcare-associated CDI has been decreasing slightly, while community-associated (CA) CDI is stable or increasing slightly; according to CDC estimates, in 2015, almost half of CDI cases were CA. 15
The clinical severity of the infection has also evolved since the last report. Increasingly virulent strains were a concern roughly 10 years ago. 1 However, a 10-year study of a sample of inpatient data found CDI-related mortality rates declined from 2005 to 2014. 16 Other CDI incidence outcomes, including rates of recurrent CDI, have increased. 17 It is notable that healthcare-associated CDI incidence trends differ based on setting, with a greater decline seen in nursing homes versus hospitals and other healthcare facilities. 18
Reimbursement policies have increasingly mandated and reinforced the reduction of CDI. CDI LabID Event reporting began in January 2013 for all acute care hospitals facilitywide using the NHSN. The Centers for Medicare & Medicaid Services (CMS) Inpatient Quality Reporting program’s CDI reporting requirements became mandatory as of January 1, 2013. Since 2017, CDI rates are among the hospital-acquired complications CMS uses to penalize the lowest performing hospitals. Many States also now mandate CDI data submission by hospitals to NHSN as part of State HAI public reporting programs. 19 In the future, participation in surveillance reporting will increase and include a broader spectrum of settings. For example, data from a larger group of LTCFs will be used to establish national benchmarks and track achievement of prevention goals. 20
To identify the PSPs for inclusion in this report, we started by reviewing the consensus guidelines for CDI prevention published by government agencies and reputable organizations. From this review, we developed an initial list that was reviewed by AHRQ and the Technical Expert Panel. The focus of this review was to identify practices that combat a prevalent harm in the U.S. healthcare system or a harm that has a high impact (e.g., high mortality). After this review and a narrowing of practices, we conducted a literature search in two databases (CINAHL and MEDLINE) and reviewed resulting abstracts for relevance. As noted, some CDI PSPs (e.g., staff training) spanned multiple harm areas, so they were moved to cross-cutting chapters (and some CDI PSP searches yielded too few articles to warrant a review [e.g., communication, contact precautions]).
Five PSPs had sufficient research in the last 10 years to conduct a review. While screening articles, we found several studies of interventions that included more than one CDI PSP (i.e., multicomponent prevention interventions). Due to the number of studies on multicomponent interventions that included patient outcomes, we decided to include an addendum on this topic.
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Mada PK, Alam MU. Clostridium difficile. Statpearls [internet]: StatPearls Publishing; 2019.Lessa FC, Winston LG, McDonald LC. Burden of Clostridium difficile infection in the United States. N Engl J Med. 2015;372(24):2369–70. doi: 10.1056/NEJMc1505190. [PubMed : 26061850 ] [CrossRef]
Centers for Disease Control and Prevention. Nearly half a million Americans suffered from Clostridium difficile infections in a single year. https://www .cdc.gov/media /releases/2015/p0225-clostridium-difficile.html. Accessed December 10, 2019.
Tschudin-Sutter S, Carroll KC, Tamma PD, et al. Impact of toxigenic Clostridium difficile colonization on the risk of subsequent C. difficile infection in intensive care unit patients. Infect Control Hosp Epidemiol. 2015;36(11):1324–9. doi: 10.1017/ice.2015.177. [PubMed : 26223207 ] [CrossRef]
McDonald LC, Coignard B, Dubberke E, et al. Recommendations for surveillance of Clostridium difficile-associated disease. Infect Control Hosp Epidemiol. 2007;28(2):140–5. doi: 10.1086/511798. [PubMed : 17265394 ] [CrossRef]
Magill SS, Edwards JR, Bamberg W, et al. Multistate point-prevalence survey of health care-associated infections. N Engl J Med. 2014;370(13):1198–208. doi: 10.1056/NEJMoa1306801. [PMC free article : PMC4648343 ] [PubMed : 24670166 ] [CrossRef]
Dubberke ER, Olsen MA. Burden of Clostridium difficile on the healthcare system. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2012;55 Suppl 2(Suppl 2):S88–S92. doi: 10.1093/cid/cis335. [PMC free article : PMC3388018 ] [PubMed : 22752870 ] [CrossRef]
Desai K, Gupta SB, Dubberke ER, et al. Epidemiological and economic burden of Clostridium difficile in the United States: Estimates from a modeling approach. BMC Infect Dis. 2016;16:303. doi: 10.1186/s12879-016-1610-3. [PMC free article : PMC4912810 ] [PubMed : 27316794 ] [CrossRef]
Zhang D, Prabhu VS, Marcella SW. Attributable healthcare resource utilization and costs for patients with primary and recurrent Clostridium difficile infection in the United States. Clin Infect Dis. 2018;66(9):1326–32. doi: 10.1093/cid/cix1021. [PMC free article : PMC5905590 ] [PubMed : 29360950 ] [CrossRef]
Curry SR, Muto CA, Schlackman JL, et al. Use of multilocus variable number of tandem repeats analysis genotyping to determine the role of asymptomatic carriers in Clostridium difficile transmission. Clin Infect Dis. 2013;57(8):1094–102. doi: 10.1093/cid/cit475. [PMC free article : PMC3783061 ] [PubMed : 23881150 ] [CrossRef]
Leekha S, Aronhalt KC, Sloan LM, et al. Asymptomatic Clostridium difficile colonization in a tertiary care hospital: Admission prevalence and risk factors. Am J Infect Control. 2013;41(5):390–3. doi: 10.1016/j.ajic.2012.09.023. [PubMed : 23622704 ] [CrossRef]
Hall AJ, Curns AT, McDonald LC, et al. The roles of Clostridium difficile and norovirus among gastroenteritis-associated deaths in the United States, 1999–2007. Clin Infect Dis. 2012;55(2):216–23. doi: 10.1093/cid/cis386 [PubMed : 22491338 ] [CrossRef]
US Department of Health and Human Services Office of Disease Prevention and Health Promotion. National Action Plan To Prevent Health Care Associated Infections: Road Map to Elimination. US Department of Health and Human Services; 2019. https://health .gov/hcq /prevent-hai-action-plan .asp#actionplan_development.
Luo R, Barlam TF. Ten-year review of Clostridium difficile infection in acute care hospitals in the USA, 2005–2014. J Hosp Infect. 2018;98(1):40–3. doi: 10.1016/j.jhin.2017.10.002. [PubMed : 29017933 ] [CrossRef]
Ma GK, Brensinger CM, Wu Q, et al. Increasing incidence of multiply recurrent Clostridium difficile infection in the United States: A cohort study. Ann Intern Med. 2017;167(3):152–8. doi: 10.7326/m16-2733. [PubMed : 28672282 ] [CrossRef]
Guh AY, Mu Y, Baggs J, Winston LG, et al. Trends in incidence of long-term-care facility onset Clostridium difficile infections in 10 U.S. geographic locations during 2011–2015. Am J Infect Control. 2018;46(7):840–2. doi: 10.1016/j.ajic.2017.11.026. [PMC free article : PMC6489441 ] [PubMed : 29329918 ] [CrossRef]
Reagan J, Herzig CT, Pogorzelska-Maziarz M, et al. State law mandates for reporting of healthcare-associated Clostridium difficile infections in hospitals. Infect Control Hosp Epidemiol. 2015;36(3):350–2. doi: 10.1017/ice.2014.61. [PMC free article : PMC4514439 ] [PubMed : 25695178 ] [CrossRef]
Palms DL, Mungai E, Eure T, et al. The national healthcare safety network long-term care facility component early reporting experience: January 2013–december 2015. Am J Infect Control. 2018;46(6):637–42. doi: 10.1016/j.ajic.2018.01.003. [PubMed : 29478758 ] [CrossRef]
This review includes a summary of evidence published from 2008 to 2018 for antimicrobial stewardship d as a practice to prevent CDI. After a brief overview of the foundational elements of antimicrobial stewardship programs (ASPs) as recommended by the CDC, this review explains how antimicrobial stewardship is believed to work as a safety practice for preventing CDI and discusses implications of recent policy changes. We examine the evidence for the estimated effect of ASPs on CDI incidence rates, starting with meta-analyses and followed by individual studies in hospitals and LTCFs. We then provide a summary of common ASP components and explores additional implementation and contextual factors, including settings, resources, and provider buy-in. Finally, we discuss research gaps and future directions for ASPs and CDI prevention.
Most studies showed statistically significant or statistically nonsignificant decreases in facility or ward-level CDI after a period of antimicrobial stewardship.
The most common ASP interventions are formulary restrictions, audit and feedback, and education.In the reviewed studies, significant reductions in CDI were associated with higher baseline CDI rates/outbreaks, ASPs developed specifically to reduce CDI (as opposed to ASPs focused on other clinical and microbiological outcomes), and ASPs that included restrictions of high-risk antimicrobials and/or a preauthorization component.
Research is needed on the impact of different ASP components, financial costs/savings of ASPs, and ASPs in a variety of healthcare settings.
ASPs require staffing, technological resources, and provider buy-in.In the future, ASPs and ASP research will benefit from improved study design and a regional perspective on CDI prevention.
ASPs are intended to limit and optimize antimicrobial prescribing, reduce the evolution of antibiotic-resistant bacteria, and improve patient outcomes. To meet these goals, the CDC provides the “Summary of Core Elements of Hospital Antibiotic Stewardship Programs.” The elements outlined below provide a basic framework of recommendations for hospital settings. (The CDC also provides core elements for nursing homes, outpatient settings, and small and critical access hospitals, and resource-limited settings). 1
Leadership Commitment: Dedicating necessary human, financial, and information technology resources.Accountability: Appointing a single leader responsible for program outcomes. Experience with successful programs shows that a physician leader is effective.
Drug Expertise: Appointing a single pharmacist leader responsible for working to improve antibiotic use.
Action: Implementing at least one recommended action, such as systemic evaluation of ongoing treatment needs after a set period of initial treatment (e.g., “antibiotic time out” after 48 hours).
Tracking: Monitoring antibiotic prescribing and resistance patterns.Reporting: Regularly reporting information on antibiotic use and resistance to doctors, nurses, and relevant staff.
Education: Educating clinicians about resistance and optimal prescribing.These elements are foundational and meant to complement additional ASP guidelines. The CDC notes that no template exists for an ASP, and ASPs can be effective in a variety of settings and under a diverse set of conditions. While the ASPs studied in the papers selected for this report included these foundational elements to varying degrees, they take many different forms based primarily on a particular facility’s resources and needs. Frequently, the ASPs were developed and executed by a multidisciplinary team with medical, pharmaceutical, and/or microbiological expertise.
The studied ASPs required tracking and reporting of data (at minimum quantifying antimicrobial use and CDI rates), as well as staff education and outreach. The “Action” element was operationalized through different strategies, the most common of which were patient case reviews, audits of antimicrobial use, restrictions on high-risk antimicrobials, and provider education. The Infectious Diseases Society of America and Society for Healthcare Epidemiology of America (IDSA/SHEA) guidelines 2 recommend minimizing the frequency and duration of high-risk antimicrobials and using local epidemiology to determine which antimicrobials to address in an ASP. The guidelines further state that ASPs should consider reducing/restricting the use of drugs including fluoroquinolones, clindamycin, and cephalosporins.
Antimicrobial exposure is widely considered one of the most significant and modifiable risk factors for CDI. In the last two decades, at the population level, increasing rates of CDI have been linked to increases in antimicrobial prescribing, particularly in older patients. 3 Patients receiving, or having recently received, antimicrobial therapy are more susceptible to colonization or infection with pathogenic bacteria such as C. difficile because antimicrobials alter gastrointestinal tract flora, destroying the bacteria that help to protect against C. difficile.
The length and type of regimen also impacts CDI risk. Several broad-spectrum antimicrobials have been most strongly linked to CDI, 4 and certain outbreaks appear to be associated with heavy prescribing of particular antimicrobials. 5 Therefore, many CDI ASPs are designed to reduce the use of particular “high-risk” antimicrobials. The CDC found that people receiving high-risk antimicrobials had a three times higher risk of CDI than did people with low-risk or no antibiotic use. 6
There is increasing urgency about reducing overreliance on antimicrobials). 7 The CDC estimates that between 30 and 50 percent of antimicrobial prescriptions are clinically inappropriate. 8 In 2015, the White House released a National Action Plan that included goals to implement antimicrobial stewardship in healthcare facilities. In 2016, CMS implemented a rule requiring nursing homes and LTCFs to have ASPs to monitor the use of antimicrobial drugs; and in 2017, The Joint Commission began assessing ASPs as part of their accreditation standards. Other countries have similar efforts, 9 and a number of resources are designed to help facilities implement ASPs. We highlight some of these resources later in this section.
This section describes literature search and review methods specific to the CDI PSPs; general methods will be described in a Methods chapter for the whole report.
The question of interest for this review is: Do ASPs reduce the risk of CDI?
To answer this question, we searched two English language databases (CINAHL, MEDLINE) for papers published from 2008 through 2019 for “Clostridium difficile” and other related Medical Subject Heading (MeSH) terms and synonyms, as well as “Antimicrobial Stewardship” or “Antibiotic Stewardship” or “Antibiotic Prescribing Practices.” The search string also included all healthcare settings, including “hospitals,” “inpatient,” “ambulatory care,” “long-term care,” “nursing homes,” “transitional care,” and “home health.” The search included both “prevention” and “treatment.”
The initial search of databases yielded 134 results and 16 papers from other sources. After duplicates were removed, 126 papers were screened for inclusion. From these papers, 43 full-text articles were retrieved. Of those, 17 studies, 3 meta-analyses, and 2 systematic reviews were selected for this review. Reference lists of included articles were also screened to ensure thoroughness. Articles were excluded at each stage if they were not primary studies, systematic reviews, or meta-analyses; treatment variables or outcomes were not relevant; or study design was insufficient. Studies in which antimicrobial stewardship implementation was accompanied by other significant infection control practices (e.g., changes in environmental cleaning) were ruled out for this section and are considered in Section 4.6, Multicomponent CDI Prevention Interventions.
General methods for this report are described in the Methods section of the full report.
For this patient safety practice, a PRISMA flow diagram and evidence table, along with literature-search strategy and search-term details, are included in the report appendixes A through C.
We reviewed the evidence from 3 meta-analyses and 17 individual studies that examined ASPs and CDI. Three meta-analyses found significant decreases in CDI following implementation of ASPs. Six individual studies on CDI outcomes showed statistically significant decreases in CDI following ASP implementation 5 , 10 – 14 , 1 showed borderline significance, and 9 showed statistically nonsignificant decreases in CDI following ASP implementation. One additional study reviewed local strategies for determining high-risk antimicrobials. 15 Study designs were generally quasi-experimental (pre-post analyses).
Three meta-analyses of ASP studies in hospital settings found that studies collectively show that antimicrobial stewardship is effective in reducing CDI rates. 16 – 18 Feazel et al. (2014) analyzed studies published between 1997 and 2012 on ASPs in hospitals during non-outbreak situations. When the results of all studies were pooled in a random effects model, ASPs conferred a significant 52 percent risk reduction (pooled risk ratio 0.48; 95% confidence interval [CI], 0.38 to 0.62; p <0.00001) on CDI incidence. Of note, geriatric patients had the largest risk reduction for CDI following implementation of an ASP.16
Similarly, in their meta-analysis of hospital ASPs in 11 articles going back several decades, Baur et al. (2017) determined that following ASP implementation periods, the incidence of CDI decreased 32 percent (incidence rate 0.68, 95% CI, 0.53 to 0.88; p=0.0029). 17 Davey et al. (2017) reviewed seven studies published up to January 2015 on hospital antimicrobial stewardship and CDI. They found a range of CDI rate reductions related to antimicrobial stewardship (median 48.6%, interquartile range 19.2% to 80.7%). They note that across all antimicrobial stewardship studies (including those that measured impact on other infections), antimicrobial stewardship generally reduced hospital stay and did not appear to impact patient mortality. 18
ASPs developed specifically to reduce CDI (as opposed to ASPs focused on other clinical and microbiological outcomes), and
ASPs that included a formulary restriction component.The majority of the studies on CDI outcomes (13/16) examined ASPs in hospitals or hospital units. The duration of the ASP period ranged from 6 months to a little over 6 years (mean 19.3 months; standard deviation [SD] 16.7). Most studies were quasi-experimental (interrupted time series or before and after design) and lacked a control or comparison group. All included studies measured the amount of prescribed antimicrobials (e.g., defined daily dose, or DDD, as defined by the World Health Organization [WHO], per 1,000 patient days) and CDI rates pre- and post-ASP implementation.
While many of the studies controlled for other contemporaneous prevention initiatives, the study designs may not account for potential covariates and confounders such as previous infection prevention efforts (e.g., hand hygiene, environmental cleaning), patient risk factors, changes in testing method, or seasonal, regional CDI fluctuations. This finding is consistent with the findings of two systematic reviews by Louh and colleagues (2017) and Pitiriga et al. (2017), which both indicated that the diversity in ASPs and weaknesses in study design undermine the strength of the evidence. 20 , 21
Another example of significant reductions in CDI after a period of ASP was at an acute general hospital with over 500 beds in the United Kingdom. 5 This ASP consisted of removal of “high-risk” antibiotics such as fluoroquinolones, cephalosporins, clindamycin, and broad-spectrum penicillins such as amoxicillin/clavulanate, from ward stocks in order to reduce their availability. These antimicrobials were targeted because they were associated with antimicrobial resistance and CDI. New prescribing guidelines with low-risk alternatives were featured in educational sessions and hospital posters and distributed to clinicians as laminated pocket-sized guides. In addition, an antibiotic management team performed regular ward rounds five times a week (compared with irregular rounds 3x/week) to optimize adherence to revised antibiotic guidelines and control the use of high-risk antibiotics. These changes corresponded to a 58.5 percent drop in fluoroquinolone use and a 45.8 percent drop in cephalosporin use. A negative binomial regression showed a significant decrease in CDI associated with the ASP (incidence rate ratio [IRR] 0.34; 95% CI, 0.20 to 0.58, p <0.0001). The researchers found no significant differences in clinical outcomes (as measured by length of stay and readmission rate for elderly patients treated for urinary tract and lower respiratory tract infections) associated with the change in prescribing practices.5
One study at a 48-bed orthopedic ward in Mexico showed borderline significant reductions in CDI 22 after restricting clindamycin (i.e., only patients with a previous infectious disease consult could receive clindamycin). After a 7-month baseline period, there was a 16-month ASP period in which clindamycin use, measured in mean DDDs per 1,000 patient days, decreased by 92.61 percent (p=0.0002). CDI rates went from 1.07 per 1,000 patient days during the baseline period to 0.12 per 1,000 patient days during the ASP period, constituting a decrease of 88.78 percent (p=0.056). 22
The reductions in CDI were generally greater in studies with higher pre-ASP (i.e., baseline) CDI rates. This finding could be because those hospitals had more room to improve than hospitals where rates were already low. Another possibility is that studies that report ASPs in the context of an outbreak could find reductions that reflect a natural regression to the mean as the outbreak wanes, rather than a result of the intervention. 23
Nine studies in hospital settings showed statistically nonsignificant changes or no decrease in CDI associated with ASP implementation. 19 , 24 – 31 In one example, antimicrobial stewardship practices were enhanced at a 525-bed public safety-net hospital, where CDI and antimicrobial prescribing rates were declining and already low, relative to other hospitals in the region. 24 New ASP practices included a preauthorization requirement for select broad-spectrum, toxic, or costly antibiotics, retrospective audit and feedback, and revised prescribing guidelines. After the changes, Jenkins et al. (2015) found total antimicrobial and high-risk antimicrobial use declined, and antimicrobial expenditures decreased, but CDI rates did not change. 24 While there are confounding factors, such as a switch to more sensitive testing methods, the authors point out that in the context of relatively low CDI rates and low antimicrobial prescribing, there may have been little room for additional decreases, since a minimal level of antimicrobial use is necessary to maintain optimal clinical outcomes.
Hospital ASPs in which CDI was not the primary clinical/microbiological target also showed nonsignificant changes or no decrease in CDI rates. 25 – 29 For example, Taggart et al. (2015) examined an ASP in two intensive care units (ICUs) in a 465-bed teaching hospital in Toronto, Canada. The ICUs included a trauma and neurosurgery ICU and a medical/surgical ICU. In both units, following a 12-month audit and feedback ASP, there were no significant changes in the CDI rate. Mean total monthly antimicrobial use declined in the trauma/neuro ICU but increased in the medical/surgical ICU. The authors speculate that the baseline prescribing practices in the medical/surgical unit were more appropriate (with more room to improve in the trauma/neuro ICU). 25
While most of the studies included in this review examined ASPs in hospitals, three studies evaluated ASPs in LTCFs. 11 , 14 , 31 LTCFs are important sites for antimicrobial stewardship due to the number of patient infections, frequent overuse of antimicrobials, and numerous transfers to and from the hospital. 31 ASPs that centered on outside infectious disease consultation showed promising results in LTCFs. 11 , 14 For example, Jump et al. (2012) measured antimicrobial use and CDIs 36 months before and 18 months after bringing in a Long-Term Care Infectious Disease consult team to a 160-bed Veterans Affairs (VA) LTCF. The team was composed of an infectious disease physician and a nurse practitioner who examined residents at the facility once each week and provided case review, feedback, and antimicrobial prescribing recommendations. In contrast to the pre-ASP period, total systemic antibiotic administration decreased by 30 percent (p<0.001), with steeper decreases in use of certain broad-spectrum antimicrobials.
The rate of change of positive C. difficile tests in the pre-ASP period showed a trend toward increasing (p=0.09), whereas in the post-ASP period the trend was reversed (p=0.21). The difference between the slopes in pre- versus post-ASP period is significant (p=0.04). While the rate of change in positive C. difficile tests did not change significantly over time for the two individual time periods, the difference in the rates of change between the two time periods was significantly different. 14
Several common ASP interventions were studied in this review. To implement changes in prescribing practices, the ASPs use various strategies or interventions, which, as shown in Table 4.1, are typically grouped into the following categories: formulary restrictions, audit and feedback, and provider education. There is some research about outcomes associated with each individual strategy, but usually ASPs use more than one of the above interventions, making it difficult to assess each approach individually. Feazel et al. (2014) state that approaches that are “restrictive,” (i.e., restrict high-risk antimicrobials) are more effective than the “persuasive” strategies (i.e., audit and feedback, education, guidelines). 16 Pitiriga et al. (2017) made no such overarching distinction about the efficacy of different strategies. 21 There is no consensus on which interventions are most effective, and it is likely that the most effective approach may differ in different settings; effective programs are dynamic and can be adapted to facility needs. 32
Studies on Antimicrobial Stewardship and Clostridioides difficile Infection Outcomes Published 2008 to 2018.
An important first step in formulary restriction is determining which antimicrobials to target for restriction. In addition to reducing the high-risk antimicrobials outlined in current guidelines, facilities may use data on regional and facility associations between CDI and antimicrobials. In one example, an ASP team examined temporal associations between antimicrobial use and CDI cases in their facility to determine which antimicrobials to target for restriction. 19
Several studies examined the role of different CDI ribotypes (more common in certain regions) and certain antimicrobials. 5 , 13 Using case-control studies to identify antibiotics that should be restricted is one way to assess local associations between antimicrobial classes and CDI. In a multicenter study in New York, each hospital performed its own case-control study to determine CDI-associated antimicrobials. 28 The hospitals used odds ratios to compare case (CDIs) and control groups. Chung et al. (2014) describe this process in more detail and found that, while more complex matching strategies are preferable, using criteria such as admission date (to correct for variation in hospital CDI prevalence) and length of stay (as a surrogate for cumulative risk of developing CDI) may be sufficient to identify high-risk antibiotics associated with CDI. For more accurate associations between antimicrobials and CDI, the researchers included additional matching variables, such as age and comorbidities. 15
Once target antimicrobials have been identified, ASPs may use strategies such as preauthorization requirements and removing access to the target antimicrobials. In their review, Feazel et al. (2014) reported that interventions that included restricting high-risk antimicrobials (e.g., preauthorization requirements, restrictions on certain antibiotics except in unusual circumstances) were associated with the greatest reductions in CDI rates. 16
To assess the CDI associations with a formulary restriction, Dancer and colleagues (2013) measured the associations of an ASP education program and restriction policy separately. They attributed decreases in CDI (a decline of 6.59% per month [95% CI,−2.52% to 15.02%; p=0.169) to the educational component of the ASP, while the restriction policy was associated with a 45.22 percent reduction (95% CI, −4.79% to 72.05%; p=0.09) in the rate of CDIs (although neither intervention had a statistically significant effect at the 0.05 level.) This study was one of the few to measure the unique contributions of individual ASP interventions. 29
Audit and feedback include case reviews of patients receiving antimicrobial therapy, often involving a multidisciplinary team (e.g., prescribers, pharmacists, infectious disease experts, administrators) and feedback to providers, as well as audits of targeted antibiotics and other clinical measures both before and/or after treating the patient. Feedback to prescribers may include advice about switching to alternative antimicrobial agents (e.g., broad to narrow spectrum), discontinuation of antimicrobial treatment, shortened duration of microbial dose, higher or lower dose, and switch from intravenous to oral antibiotics. The latter recommendation is based on the idea that an earlier switch to oral therapy allows faster discharge from the hospital, thereby reducing exposure to CDI and drug-resistant organisms. 23
ASPs with an audit and feedback component were common in the studies we reviewed, and these are widely recommended antimicrobial stewardship practices; 17 , 21 however, ASPs based solely on an audit and feedback program showed no statistically significant reductions in CDI. 25 , 27 One benefit of audit and feedback is that the practice itself educates prescribers and other healthcare staff. 11 , 14 In most studies, audit and feedback are accompanied by a staff education component, making it difficult to find associations between audit and feedback alone and CDI rates.
Researchers suggest that education is important to provide context and convince physicians and other staff to participate in antimicrobial stewardship activities. 11 , 29 Jump et al. (2012) note that some rehabilitation physicians may be aware of the problem of antimicrobial resistance but unaware of local resistance patterns. The education programs described in the reviewed studies included information about antimicrobial resistance, local and facility antibiogram data, treatment guidelines, and/or CDI-specific education. Educational methods included the use of emails, pocket cards, presentations, and trainings. 14
In an attempt to isolate the CDI associations of an educational program (as part of a multicomponent strategy), Shea et al. (2017) assessed results associated with a 3-month education campaign, then, separately, the results following a subsequent 12 months of a fluoroquinolone restriction policy. The shorter education component appeared to have a significant impact, which was enhanced by the restriction policy. Compared with pre-ASP, the four hospitals experienced 48 percent and 88 percent average reductions in fluoroquinolone utilization (days of therapy per 1,000 patient days) after education and restriction, respectively. CDI rates decreased significantly (p=0.044) from 4.0 cases/10,000 patient days pre-ASP to 3.43 cases/10,000 patient days following staff education, and to 2.2 cases/10,000 patient days following restriction. 12
One potential consideration with ASPs is that they may encourage the use of (untargeted) broad-spectrum agents and/or alternative “lower-risk” antimicrobials, which, in turn, may lead to increased resistance to the unrestricted drugs. Pitiriga and colleagues (2017) promoted the restriction of quinolones but also warn against the so-called “squeezing the balloon” phenomenon, wherein restriction policies for use of one set of drugs leads to increased use of unrestricted alternatives, which leads to resistance. This practice runs counter to the goal of decreasing antimicrobial selection pressure. 21
One of the positive outcomes of a CDI-targeted ASP can be lower rates of MRSA (methicillin-resistant Staphylococcus aureus), ESBL (extended-spectrum beta-lactamases)-producing coliform infections, and other MDROs (multidrug-resistant organisms). For example, while the primary reason for the antimicrobial restrictions and revised prescribing guidelines in the ASP studied by Dancer et al. (2013) was to decrease CDI rates at the hospital, the researchers also found decreases in ESBL-producing coliforms following the ASP an 8.21 percent reduction [95% CI, −0.39% to 16.15%]). During the following 3 years, both ESBL-producing coliform infections and MRSA declined. 29
One additional benefit (or perhaps less identified outcome of an ASP) was an increase in the accuracy of patient diagnoses following audit and feedback interventions. Talpaert et al. (2011) found that, out of 386 interventions by the ASP team, on 75 occasions the clinicians changed the patient’s diagnosis. 5 Similarly, Lowe et al. (2017) describe how virology results tied to ASP consults helped facilitate appropriate antimicrobial treatment. Many patients in that study (17/19) who were on empiric oseltamivir were found not to have proven influenza, and following proper diagnosis, oseltamivir was promptly discontinued. 27
ASPs require resources, and sometimes creative mechanisms to address resource gaps. Researchers noted challenges with staffing limitations (when additional staff were not hired for the ASP) and a need for technical resources to track antimicrobial use. 28 In addition, the lack of EHRs in many LTCFs can make it hard to track the exact indication for antimicrobial use. 30 , 31 However, even with limited means, antimicrobial stewardship can produce meaningful benefits. 26 For example, Yam et al. (2012) described the challenges of resource constraints in a small rural hospital. The ASP team decided to use scheduled and as-needed consultations with a remote infectious disease specialist physician. After the ASP worked with the remote specialist for 13 months, the researchers found nosocomial CDI decreased from an average of 5.5 cases per 10,000 patient days to an average of 1.6 cases per 10,000 patient days, and antibiotic purchase costs decreased nearly 50 percent. 30
Using nontraditional staff types to lead the ASP (e.g., infection control nurses, clinical microbiologists, or pharmacists without infectious disease training);
Using telehealth for advising on prescribing decisions; Identifying a single priority hospital unit (e.g., ICU) in which to implement an ASP; orChoosing and implementing a single prescribing practice (e.g., reviewing the need for antibiotics after 48 hours, or improving adherence to guidelines for empiric treatment for CA pneumonia or sepsis).
There are several examples of ASP collaborations that overcame resource and expertise gaps. Lowe et al. (2017) described an efficient collaboration between the ASP physician or pharmacist and the virology laboratory for polymerase chain reaction (PCR) testing on respiratory tract infection, in order to optimize antiviral and antimicrobial use. 27 LTCFs often lack appropriate personnel, funding, and electronic resources, and face a paucity of well-validated strategies for their sector. 14
To implement an ASP in an LTCF, Rahme et al. (2016) document a hospital that collaborated with an LTCF for antimicrobial stewardship in part because the facilities shared patients and there was concern about interfacility HAI transmission. 31 The hospital ASP team provided microbiology data, provider education and treatment guidelines, and a 24-hour hotline for LTCF prescribers. Some LTCFs collaborated with outside consultants to implement audit and feedback ASPs. 11 , 14 , 30
Resistance on the part of providers is a major barrier to ASP implementation that is described in the literature; conversely, a facilitator to implementation is a good relationship between the ASP team and prescribers. 17 Educating physicians and providing proof of ASP safety and efficacy are essential to garnering support. 19 Dancer et al. (2013) found that gaining support for their ASP was challenging at the outset, especially when ASP recommendations for prescribing conflicted with previously published guidelines for a specific infection. For example, gastroenterologists initially refused to curtail ciprofloxacin prescribing for spontaneous bacterial peritonitis. 29 After being educated about the microbiological etiology of the infection, the gastroenterologists were persuaded to change prescribing practices. This observation aligns with the findings of Libertin and colleagues (2017), who noted that development of a “collegial environment for a health care provider’s growth in ASP knowledge was important in achieving acceptance of the program” (p. 981). 10
The following are resources for implementing an ASP, starting with a CDI-specific resource and followed by ASP resources in general:
There is a notable absence of research on the implementation of ASPs in settings other than hospitals. Of the 16 studies included in this review, we only found 3 ASP studies in LTCFs. 11 , 14 , 31 In these three studies, facilities worked with outside consultants to provide expertise and feedback. Researchers commented on the challenges of ASP implementation in LTCF settings due to high rates of infection 25 and a “treat-first” culture. 34 At the same time, ASPs in these settings could potentially have a large impact as they serve high-risk patients and share patients with other facilities. In addition, ASPs in outpatient settings warrant attention, since according to 2016 data reported to the NHSN, CA CDI is on the rise. 8 Our search found no studies on CDI and ASPs in outpatient settings. This is an important gap in the literature and an area for further exploration, especially given the links between antimicrobial prescribing in the outpatient setting and CA CDI. 35
The reviewed articles had little information on financial outcomes and antimicrobial stewardship. While Jenkins et al. (2015), Libertin et al., (2017) and Taggart et al. (2015) show total reductions in the cost of antibiotics, particularly from reductions in use of costly broad-spectrum antibiotics, 10 , 24 , 25 other financial outcomes are not examined in these or other ASP studies. It has been speculated that the financial savings of ASPs measured in cost of antimicrobials and expenses associated with CDI management outweigh the costs of investing in infectious disease expertise to support an ASP. 11
On a national level, it is believed that antimicrobial stewardship is extremely cost effective in terms of prevention of healthcare costs. 36 However, there is a need for more economic information for healthcare systems and facilities to determine costs and savings. 37 More robust and nuanced cost-effectiveness analyses would help staff in various settings, particularly those with resource limitations, to consider how to best invest in support for an ASP.
Despite the methodological, technological, and resource challenges of research on ASPs, many researchers noted a need for more rigorous study design, including randomized controlled trials (in addition to pre-post) study design. 16 There is also a need for studies that consider the costs and benefits of antimicrobial stewardship over the course of multiple years, to measure longer term associations that may not be evident in shorter study periods. 17
Researchers have pointed out that reducing antimicrobial use is not always equivalent to improved prescribing and antibiotic appropriateness is as important as counts of prescriptions. 38 One of the issues that comes up in systematic reviews and studies of ASPs is the heterogeneity in process measures, which, in addition to study design, makes comparison and generalization difficult. 38 As noted by Ostrowsky et al. (2014), the prescribed daily doses relative to WHO DDDs may vary between hospitals. 28 DDDs are based on standard dosing and therefore may not accurately capture administered doses that are lower than the routine dose. Point prevalence (accurate surveys taken at particular points in time that can compared) has been suggested as a low-cost way to understand antimicrobial consumption. 39 Finally, there are different measures of clinical and microbiological outcomes, 38 as evidenced in the studies in this review.
Some future directions for ASPs to reduce CDI include patient and family education on antimicrobial stewardship. The ASP described by Rahme et al. (2016) included an education component to address the pressure on prescribers from patients’ families in an LTCF. It was theorized that including a focus on family education would lessen the pressure on prescribers to treat symptoms unnecessarily with antibiotics. 31 Findings of qualitative provider surveys confirm that family pressure can be a challenge. For example, Cole (2014) found that 55 percent of doctors felt under pressure—mainly from patients—to prescribe antibiotics. 40 Similarly, Sanchez et al. (2014) reported a major reason for nonadherence to prescribing guidelines is a concern for patient or family satisfaction. 41
In LTCFs, doctors report being influenced by family pressure to prescribe antimicrobials, especially in situations when they are undecided about whether to prescribe an antimicrobial. 42 Greater public awareness could help patients and families to better understand why judicious use of antimicrobials is important, thereby lessening pressure on prescribers and promoting better prescribing practices.
The use of technology for more accurate and rapid diagnosis of viral versus bacterial infections is another area for future ASP improvement. Lowe et al. (2017) point out how rapid diagnostics can help decrease antimicrobial use, as in the case of PCR testing to help determine if antibiotic treatment is required. 27 Pitiriga et al. (2017) also endorse “diagnostic stewardship programs” incorporating rapid molecular diagnostics, genomic pathogen profiling, and estimation of patient–pathogen–treatment interactions to help individualize prescribing practices. 21 A more detailed review of the use of improved diagnostics can be found in the Section 4.5, Testing.
Finally, regionally and ecologically informed antimicrobial stewardship is another direction for the future. CDI is transferred across settings in a region, and regional resistance patterns and CDI strains are important prescribing considerations. 14 Regional, multifacility, and collective ASP efforts could be especially effective strategies. As ASPs become more common due to increasing regulations, more LTCFs will be involved, intervening with a population at high risk of CDI and providing an opportunity for an increased understanding of ASPs.
Centers for Disease Control and Prevention. Core elements of hospital antibiotic stewardship programs. https://www .cdc.gov/antibiotic-use /core-elements/hospital .html. Accessed December 10, 2019
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Ticinesi A, Nouvenne A, Folesani G, et al. Multimorbidity in elderly hospitalised patients and risk of Clostridium difficile infection: A retrospective study with the cumulative illness rating scale (CIRS). BMJ Open. 2015;5(10):e009316. doi: 10.1136/bmjopen-2015-009316. [PMC free article : PMC4636642 ] [PubMed : 26503394 ] [CrossRef]
Vardakas KZ, Trigkidis KK, Boukouvala E, et al. Clostridium difficile infection following systemic antibiotic administration in randomised controlled trials: A systematic review and meta-analysis. Int J Antimicrob Agents. 2016;48(1):1–10. doi: 10.1016/j.ijantimicag.2016.03.008. [PubMed : 27216385 ] [CrossRef]
Talpaert MJ, Gopal Rao G, Cooper BS, et al. Impact of guidelines and enhanced antibiotic stewardship on reducing broad-spectrum antibiotic usage and its effect on incidence of Clostridium difficile infection. J Antimicrob Chemother. 2011;66(9):2168–74. doi: 10.1093/jac/dkr253. [PubMed : 21676904 ] [CrossRef]
Centers for Disease Control and Prevention. Projected Incidence of Infections and Deaths Due to Antibiotic Resistant Healthcare-Associated Infections. https://www .cdc.gov/hai /data/archive/stop-spread-technical-appendix.html. Accessed December 10, 2019.
Gould IM. Controversies in infection: Infection control or antibiotic stewardship to control healthcare-acquired infection? J Hosp Infect. 2009;73(4):386–91. doi: 10.1016/j.jhin.2009.02.023. [PubMed : 19596494 ] [CrossRef]
European Center for Disease Prevention and Control. Guidelines for the prudent use of antimicrobials in human health. 2017. https://ec .europa.eu /health/amr/sites/amr /files/amr_guidelines_prudent_use_en .pdf.
Libertin CR, Watson SH, Tillett WL, et al. Dramatic effects of a new antimicrobial stewardship program in a rural community hospital. Am J Infect Control. 2017;45(9):979–82. doi: 10.1016/j.ajic.2017.03.024. [PubMed : 28526311 ] [CrossRef]
Tedeschi S, Trapani F, Giannella M, et al. An antimicrobial stewardship program based on systematic infectious disease consultation in a rehabilitation facility. Infect Control Hosp Epidemiol. 2017;38(1):76–82. doi: 10.1017/ice.2016.233. [PubMed : 27745559 ] [CrossRef]
Shea KM, Hobbs ALV, Jaso TC, et al. Effect of a health care system respiratory fluoroquinolone restriction program to alter utilization and impact rates of Clostridium difficile infection. Antimicrob Agents Chemother. 2017;61(6). doi: 10.1128/aac.00125-17. [PMC free article : PMC5444144 ] [PubMed : 28348151 ] [CrossRef]
Wenisch JM, Equiluz-Bruck S, Fudel M, et al. Decreasing Clostridium difficile infections by an antimicrobial stewardship program that reduces moxifloxacin use. Antimicrobial agents and chemotherapy. 2014;58(9):5079–83. doi: 10.1128/AAC.03006-14. [PMC free article : PMC4135825 ] [PubMed : 24936597 ] [CrossRef]
Jump RL, Olds DM, Seifi N, et al. Effective antimicrobial stewardship in a long-term care facility through an infectious disease consultation service: Keeping a lid on antibiotic use. Infect Control Hosp Epidemiol. 2012;33(12):1185–92. doi: 10.1086/668429. [PMC free article : PMC4370223 ] [PubMed : 23143354 ] [CrossRef]
Chung P, Currie B, Guo Y, et al. Investigation to identify a resource-efficient case-control methodology for determining antibiotics associated with Clostridium difficile infection. Am J Infect Control. 2014;42:(10 Suppl):S264–8. doi: 10.1016/j.ajic.2014.05.001. [PubMed : 25239720 ] [CrossRef]
Feazel LM, Malhotra A, Perencevich EN, et al. Effect of antibiotic stewardship programmes on Clostridium difficile incidence: A systematic review and meta-analysis. J Antimicrob Chemother. 2014;69(7):1748–54. doi: 10.1093/jac/dku046. [PubMed : 24633207 ] [CrossRef]
Baur D, Gladstone BP, Burkert F, et al. Effect of antibiotic stewardship on the incidence of infection and colonisation with antibiotic-resistant bacteria and Clostridium difficile infection: A systematic review and meta-analysis. Lancet Infect Dis. 2017;17(9):990–1001. doi: 10.1016/s1473-3099(17)30325-0. [PubMed : 28629876 ] [CrossRef]
Davey P, Marwick CA, Scott CL, et al. Interventions to improve antibiotic prescribing practices for hospital inpatients. Cochrane Database Syst Rev. 2017;2:Cd003543. doi: 10.1002/14651858.CD003543.pub4. [PMC free article : PMC6464541 ] [PubMed : 28178770 ] [CrossRef]
Patton A, Davey P, Harbarth S, et al. Impact of antimicrobial stewardship interventions on Clostridium difficile infection and clinical outcomes: Segmented regression analyses. J Antimicrob Chemother. 2018;73(2):517–26. doi: 10.1093/jac/dkx413. [PubMed : 29177477 ] [CrossRef]
Louh IK, Greendyke WG, Hermann EA, et al. Clostridium difficile infection in acute care hospitals: Systematic review and best practices for prevention. Infect Control Hosp Epidemiol. 2017;38(4):476–82. doi: 10.1017/ice.2016.324. [PMC free article : PMC5560033 ] [PubMed : 28300019 ] [CrossRef]
Pitiriga V, Vrioni G, Saroglou G, et al. The impact of antibiotic stewardship programs in combating quinolone resistance: A systematic review and recommendations for more efficient interventions. Adv Ther. 2017;34(4):854–65. doi: 10.1007/s12325-017-0514-y. [PubMed : 28303388 ] [CrossRef]
Cruz-Rodriguez NC, Hernandez-Garcia R, Salinas-Caballero AG, et al. The effect of pharmacy restriction of clindamycin on Clostridium difficile infection rates in an orthopedics ward. Am J Infect Control. 2014;42(6):e71–3. doi: 10.1016/j.ajic.2014.02.018. [PubMed : 24837129 ] [CrossRef]
Chalmers JD, Al-Khairalla M, Short PM, et al, Winter JH. Proposed changes to management of lower respiratory tract infections in response to the Clostridium difficile epidemic. J Antimicrob Chemother. 2010;65(4):608–18. doi: 10.1093/jac/dkq038. [PubMed : 20179023 ] [CrossRef]
Jenkins TC, Knepper BC, Shihadeh K, et al. Long-term outcomes of an antimicrobial stewardship program implemented in a hospital with low baseline antibiotic use. Infect Control Hosp Epidemiol. 2015;36(6):664–72. doi: 10.1017/ice.2015.41. [PMC free article : PMC4836835 ] [PubMed : 25740560 ] [CrossRef]
Taggart LR, Leung E, Muller MP, et al. Differential outcome of an antimicrobial stewardship audit and feedback program in two intensive care units: A controlled interrupted time series study. BMC Infect Dis. 2015;15:480. doi: 10.1186/s12879-015-1223-2. [PMC free article : PMC4625716 ] [PubMed : 26511839 ] [CrossRef]
Carbo JF, Ruh CA, Kurtzhalts KE, et al. Male veterans with complicated urinary tract infections: Influence of a patient-centered antimicrobial stewardship program. Am J Infect Control. 2016;44(12):1549–53. doi: 10.1016/j.ajic.2016.04.239. [PubMed : 27388268 ] [CrossRef]
Lowe CF, Payne M, Puddicombe D, et al. Antimicrobial stewardship for hospitalized patients with viral respiratory tract infections. Am J Infect Control. 2017;45(8):872–5. doi: 10.1016/j.ajic.2017.03.025. [PubMed : 28526309 ] [CrossRef]
Ostrowsky B, Ruiz R, Brown S, et al. Lessons learned from implementing Clostridium difficile-focused antibiotic stewardship interventions. Infect Control Hosp Epidemiol. 2014;35: Suppl 3:S86–95. doi: 10.1086/677828. [PubMed : 25222903 ] [CrossRef]
Dancer SJ, Kirkpatrick P, Corcoran DS, et al. Approaching zero: Temporal effects of a restrictive antibiotic policy on hospital-acquired Clostridium difficile, extended-spectrum beta-lactamase-producing coliforms and meticillin-resistant staphylococcus aureus. Int J Antimicrob Agents. 2013;41(2):137–42. doi: 10.1016/j.ijantimicag.2012.10.013. [PubMed : 23276500 ] [CrossRef]
Yam P, Fales D, Jemison J, et al. Implementation of an antimicrobial stewardship program in a rural hospital. Am J Health Syst Pharm. 2012;69(13):1142–8. doi: 10.2146/ajhp110512. [PubMed : 22722593 ] [CrossRef]
Rahme CL, Jacoby HM, Avery L. Impact of a hospital’s antibiotic stewardship team on fluoroquinolone use at a long-term care facility. Ann Longterm Care. 2016;24(6):13–20.
Wenzler E, Mulugeta SG, Danziger LH. The antimicrobial stewardship approach to combating Clostridium difficile. Antibiotics (Basel). 2015;4(2):198–215. doi: 10.3390/antibiotics4020198. [PMC free article : PMC4790327 ] [PubMed : 27025621 ] [CrossRef]
Centers for Disease Control and Prevention. The core elements of human antibiotic stewardship programs in resource -limited settings: National and hospital levels. Atlanta, GA: US Department of Health and Human Services; 2018. https://www .cdc.gov/antibiotic-use /healthcare /pdfs/stewardship-resource-limited-508.pdf.
Van Schooneveld T, Miller H, Sayles H, et al. Survey of antimicrobial stewardship practices in nebraska long-term care facilities. Infect Control Hosp Epidemiol. 2011;32(7):732–4. doi: 10.1086/660855. [PubMed : 21666410 ] [CrossRef]
Dantes R, Mu Y, Hicks LA, et al. Association between outpatient antibiotic prescribing practices and community-associated Clostridium difficile infection. Open Forum Infect Dis. 2015;2(3):ofv113. doi: 10.1093/ofid/ofv113. [PMC free article : PMC4551478 ] [PubMed : 26509182 ] [CrossRef]
Scott RD, 2nd, Slayton RB, Lessa FC, et al. Assessing the social cost and benefits of a national requirement establishing antibiotic stewardship programs to prevent Clostridioides difficile infection in us hospitals. Antimicrob Resist Infect Control. 2019;8:17. doi: 10.1186/s13756-018-0459-1. [PMC free article : PMC6343309 ] [PubMed : 30680153 ] [CrossRef]
Naylor NR, Zhu N, Hulscher M, et al. Is antimicrobial stewardship cost-effective? A narrative review of the evidence. Clin Microbiol Infect. 2017;23(11):806–11. doi: 10.1016/j.cmi.2017.06.011. [PubMed : 28642146 ] [CrossRef]
Dik JW, Hendrix R, Poelman R, et al. Measuring the impact of antimicrobial stewardship programs. Expert Rev Anti Infect Ther. 2016;14(6):569–75. doi: 10.1080/14787210.2016.1178064. [PubMed : 27077229 ] [CrossRef]
Morris AM. Antimicrobial stewardship programs: Appropriate measures and metrics to study their impact. Curr Treat Options Infect Dis. 2014;6(2):101–12. doi: 10.1007/s40506-014-0015-3. [PMC free article : PMC4431704 ] [PubMed : 25999798 ] [CrossRef]
Cole A. Gps feel pressurised to prescribe unnecessary antibiotics, survey finds. BMJ. 2014;349:g5238. doi: 10.1136/bmj.g5238. [PubMed : 25143516 ] [CrossRef]
Sanchez GV, Roberts RM, Albert AP, et al. Effects of knowledge, attitudes, and practices of primary care providers on antibiotic selection, United States. Emerg Infect Dis. 2014;20(12):2041–7. doi: 10.3201/eid2012.140331. [PMC free article : PMC4257826 ] [PubMed : 25418868 ] [CrossRef]
van Buul LW, van der Steen JT, Doncker SM, et al. Factors influencing antibiotic prescribing in long-term care facilities: A qualitative in-depth study. BMC Geriatr. 2014;14:136. doi: 10.1186/1471-2318-14-136. [PMC free article : PMC4289541 ] [PubMed : 25514874 ] [CrossRef]
This review includes a summary of evidence published from 2008 to 2018 on hand hygiene as a prevention practice for CDI. After a brief practice description of hand hygiene, as recommended by IDSA, the review explains how hand hygiene is believed to work as a safety practice for preventing the transmission of C. difficile. Next, we examine evidence for the estimated effect of healthcare worker (HCW) and patient hand hygiene interventions on CDI incidence rates, and we provide a brief look at research on specific hand hygiene methods for C. difficile. The review then explores hand hygiene intervention implementation and contextual factors, including compliance strategies, sink location, and tailoring to staff needs. Finally, we explore research gaps and future directions for hand hygiene and CDI prevention. The review’s key findings are located in the box on the right.
Gloves and handwashing with soap and water are the recommended hand hygiene practices for C. difficile prevention.
Multiple experimental studies show ABHRs are not effective in eliminating C. difficile spores.Studies were quasi-experimental and showed large and mostly statistically nonsignificant decreases in CDI following implementation of hand hygiene programs that targeted multiple HAIs (statistical significance was impacted by small sample sizes).
Studies are needed that measure C. difficile-targeted hand hygiene initiatives, as well as financial outcomes, and hand hygiene programs in nonhospital settings.
Important contextual factors for CDI/hand hygiene include sink location, visibility, and accessibility.
Future directions for hand hygiene programs include patient hand hygiene, studies on glove compliance, electronic monitoring, and sustainable interventions.
In the 2017 clinical practice guidelines for preventing C. difficile, IDSA states that HCWs “must” use gloves while caring for CDI patients, including when entering a room with a CDI patient. In CDI outbreaks or hyperendemic settings (periods of persistently high levels of CDI), the guidelines include performing hand hygiene with soap and water before and after caring for a patient with CDI and after removing gloves. When working with CDI patients in routine or endemic situations, the guidelines recommend washing hands with soap and water or using alcohol-based hand rubs (ABHRs) for hand hygiene after removing gloves. 1 While ABHRs are the preferred means of disinfecting hands for most pathogens, alcohol is not active against C. difficile spores, and it is believed that the most efficacious way to eliminate C. difficile is via the mechanical action of handwashing. 2 , 3 Washing hands with soap and water is recommended after any contact with feces. 1
Before touching a patient Before clean/aseptic procedures After body fluid exposure/risk After touching a patient After touching patient surroundingsUse of proper handwashing technique is important for C. difficile spore removal. 5 When handwashing is indicated, both the CDC and WHO recommend vigorous and thorough washing of all surfaces for at least 15 seconds. 6 The entire process from start to finish should take between 40 and 60 seconds. 7 This technique has been tested against unstructured and alternative techniques and found to be most effective at removing C. difficile spores. 5
Concerning the type of soap to use during handwashing, the general CDC recommendations (for all HAIs) call for antibacterial soap over plain soap. However, in experimental studies, some researchers have found that plain soap is more effective for removing C. difficile spores. 2 , 8 This is one of several unresolved issues in hand hygiene for C. difficile that is explored in the research included in this review.
The CDC defines hand hygiene as “a general term that applies to either handwashing, antiseptic hand wash, antiseptic hand rub, or surgical hand antisepsis” (pp. 12–40). e As such, glove use was not included in most of the reviewed studies. However, C. difficile hand hygiene recommendations strongly recommend the use of gloves. 1 , 9 One study found that universal glove use (with emollients for skin care) at 78 percent compliance was more effective than standard contact precautions (use of gowns and gloves; 67% compliance) to avoid C. difficile transmission. 10
According to the WHO (2009), HCWs should conduct hand hygiene before and after wearing gloves. Appropriate technique helps prevent potential hand contamination when removing gloves. 11 , 12 Gloves should not be reused on more than one patient. 7 The 2009 WHO guidelines also provide guidance on proper skin and nail care. 7
Multiple studies have found C. difficile contamination on HCWs’ hands and several studies have linked cases of CDI and CDI outbreaks to HCW transmission. 11 Similarly, inadequate hand hygiene has been linked to higher incidence of CDI. 13 A study that looked specifically at HCW hand contamination after contact with CDI patients found that 24 percent of HCW hands (p <0.001) were contaminated with CDI (even when gloves were used in 356/386 of patient contacts). In addition, contact without the use of gloves was independently associated with hand contamination (adjusted OR, 6.26; 95% CI, 1.27 to 30.78; p=0.02).14
Tomas et al. (2016) found that HCWs may spread C. difficile directly from one patient to another or by touching contaminated surfaces in the environment. 15 Each hand-to-surface exposure can result in the hand transmission of microorganisms. 16 Cross-contamination of C. difficile originates in the feces of people who are infected, including in the form of spores (a resilient form of the bacterium), which, if not properly cleaned, can survive in the patient’s surroundings on any surface (e.g., toilet areas, clothing, sheets, furniture 7 ) for over 4 days. 17 C. difficile is transmitted when the spores found in feces are ingested via the fecal-oral route or into the colon directly through shared equipment. 18
Recent studies provide additional evidence supporting handwashing with soap and water over ABHRs for C. difficile prevention. 3 , 8 , 19 , 20 For example, Kundrapu et al. (2014) tested hands contaminated with C. difficile after several methods of hand hygiene. Before conducting hand hygiene, roughly half of the subjects were found to have C. difficile spores on their hands. Handwashing significantly reduced the percentage of positive cultures (from ~48% to 10%, n=62; p=0.0005), as well as the number of spores recovered from contaminated hands; conversely, ABHR did not significantly reduce positive cultures or spores (from ~51% to ~49% positive cultures, n=59; p=0.85). 19 While the in vitro evidence for handwashing is consistent across multiple studies, evidence is limited on the impact of handwashing on CDI rates in healthcare settings.
Due to concern about HAI rates and poor HCW hand hygiene compliance, hand hygiene (including use of ABHRs) has been heavily promoted over the last two decades. One systematic review found median hand hygiene compliance across 96 studies in a variety of healthcare settings was 40 percent, 21 and hand hygiene rates are potentially even lower at LTCFs. 22 Single-facility studies on compliance with CDI-specific guidelines also show the need for improved practice. Deyneko et al. (2016) found that, at a 637-bed tertiary care hospital in Canada, glove use compliance was 85.4 percent (211/247), but handwashing compliance after care of CDI patients was only 14.2 percent (35/247) and hand rubbing with ABHR was performed instead of handwashing in 33.2 percent of opportunities (82/247). 23 Similarly, in a study in a single surgical transplant unit, Zellmer et al. (2015) found that the baseline percentage of visitors and staff seeing CDI patients that did not practice hand hygiene was 72.5 percent (58/80) before entering the room and 54.6 percent (42/77) after exiting the room (11.7% of which was ABHR hygiene only). 24
Regulatory agencies have implemented hand hygiene and reporting requirements in an effort to improve compliance. In 2004, The Joint Commission required healthcare facilities to implement hand hygiene programs, and starting in 2018, observation by surveyors of individual staff failure to perform hand hygiene in the process of direct patient care began to be cited as a deficiency. CMS also identifies deficiencies in LTCFs that do not meet hand hygiene standards, and requirements for Medicare and Medicaid participation were revised in 2016 to reflect advances in the theory and practice of patient safety.
The question of interest for this review is: Is hand hygiene effective at preventing CDI?
To answer this question, we searched the databases CINAHL and MEDLINE from 2008 to 2018 for “Clostridium difficile” and related MeSH terms and synonyms, as well as “Hand Hygiene,” “Hand Disinfection,” or “anti-infective agents.” The initial search yielded 168 results, and, after duplicates were removed, 165 were screened for inclusion and 20 full-text articles were retrieved. Of those, 11 studies and one systematic review were selected for inclusion in this review. Reference lists of included articles were also screened to ensure thoroughness and four additional studies were retrieved via this method. Articles were excluded if the outcomes were not relevant or precisely reported or study design was insufficient. Studies in which hand hygiene was accompanied by other significant infection control practices (e.g., changes in environmental cleaning) were ruled out for this section and are considered in Section 4.6, Multicomponent CDI Prevention Interventions.
General methods for this report are described in the Methods section of the full report.
For this patient safety practice, a PRISMA flow diagram and evidence table, along with literature-search strategy and search-term details, are included in the report A through C appendixes.
We reviewed five quasi-experimental studies on HCW hand hygiene initiatives and CDI rates in real-world clinical settings. Most of the studies (4/5) showed statistically nonsignificant improvements in CDI rates after implementation of a hand hygiene intervention. In all the studies, the hand hygiene initiatives targeted multiple HAIs and not CDI specifically. In this review of the evidence, we first present important methodological considerations, followed by more detailed study outcomes. We then highlight one study on patient hand hygiene. Then we discuss an additional five in vitro studies that focus on methods for hand hygiene (e.g., type of cleaning agent, handwashing technique, glove removal) to reduce C. difficile hand contamination.
Consistent with the findings of others (e.g., Louh et al., 2017), the studies on hand hygiene and CDI were generally of low quality and did not address multiple confounding factors. 25 In some studies, the researchers failed to control for important variables, such as antimicrobial prescribing. 26 In addition, there were issues with internal validity when measuring hand hygiene compliance, such as observer reliability and the potential of workers to temporarily alter their behavior while being observed (i.e., Hawthorne effect). The studied hand hygiene interventions were intended to reduce transfer of multiple infectious agents; while the researchers state that the interventions followed established guidelines, it was not always clear how “compliance” was defined and measured and whether CDI-specific hand hygiene guidelines were included.
More specifically, the studied hand hygiene initiatives aimed to reduce multiple HAIs, and study authors reported that the interventions included the promotion of ABHRs (either through additional dispensers or by encouraging ABHR use). It is therefore important to consider the potential impact of ABHRs as a strategy on the incidence of CDI. While ABHRs work to eliminate many other pathogens that cause infection, ABHRs are shown to have limited effectiveness for CDI eradication. 2 , 3 However, several hospital studies that measured CDI rates after ABHR hand hygiene campaigns found that CDI rates decreased or remained stable.
For example, Knight et al. (2010) conducted a retrospective chart analysis following 5 years of a hospital ABHR policy (which included education and installation of ABHR dispensers) and found a significant decrease in CDI (3.98 per 10,000 patient days after implementation of the ABHR policy, compared with 4.96 per 10,000 patient days before implementation (p=0.0036). 27 Conversely, Silva et al. (2013) found that hospital CDI rates remained stable despite several years of increased use of ABHRs. 28 Researchers speculate that these findings may be attributable to improved compliance with CDI prevention strategies, increased awareness of the importance of hand hygiene in reducing infection, and the effect of hand rubbing in reducing the bacterial load on hands. It is because promotion of ABHRs has not been linked to increases in CDI that the CDC guidance promotes handwashing (not ABHRs) in cases of high endemic CDI or CDI outbreaks. 9
As noted, the studied hand hygiene initiatives were intended to reduce several HAIs and included some or all of the following components: staff education, compliance monitoring and feedback, incentives, promotion of guidelines, and, in some studies, new ABHR dispensers. Using p
Studies on HCW Hand Hygiene Initiatives and CDI Rates (Published 2008–2018).
Sickbert-Bennett et al. (2016) evaluated HCW hand hygiene compliance and HAIs following the implementation of “Clean In, Clean Out” in an 853-bed hospital in North Carolina. The hospital hand hygiene program included focus on cleaning hands before and after working with patients, covert observation of compliance, staff data collection, and feedback. After 17 months, the researchers found a 10 percent improvement in appropriate hand hygiene compliance and a 14 percent reduction in healthcare-acquired CDI (p=0.070), as well as decreases in other HAIs. The published article did not clarify what constituted hand hygiene compliance, and whether ABHR use or handwashing was considered compliant, making it difficult to determine which practice may have contributed to the CDI reduction. 32
Following a 3-year hand hygiene initiative in a 383-bed teaching hospital in rural New Hampshire, Kirkland et al. (2012) evaluated hand hygiene compliance and HAI rates. This study described promotion of published hand hygiene guidelines but did not specify whether handwashing for CDI was emphasized. The initiative included leadership endorsement, measurement and feedback on hand hygiene compliance, and education. Over the study period, observed hand hygiene compliance increased significantly from 41 percent to 87 percent (p <0.01), and the overall HAI rate declined significantly (from 4.8 to 3.3 per 1,000 inpatient days; p<0.01). The decline in CDI was not statistically significant (0.9 to 0.6 per 1,000 patient days, p=0.1); like other smaller studies, statistical significance was potentially due to small sample size.29 This was one of three studies that found statistically nonsignificant decreases in CDI following staff hand hygiene initiatives. 29 – 31
Several studies explored initiatives in which ABHR protocols were described as a key component. For example, in the only study in a nonhospital setting, Schweon et al. (2013) studied a hand hygiene program in a 174-bed skilled nursing facility. The program included installation of a number of new ABHR dispensers, staff education on handwashing guidelines, staff monitoring, and patient education on when to conduct hand hygiene. A monthly hand hygiene champion was recognized, and hand hygiene posters were placed around the facility. Following the year-long program, most HAIs decreased but only lower respiratory tract infections showed statistically significant decreases. CDI rate per 1,000 resident days decreased but was not significant (from 0.08 to 0.04; p=0.36). Again, it is not clear the degree to which the use of ABHRs was deemed an appropriate practice for hand hygiene. 31
Like the hand hygiene program studied by Schweon and colleagues (2013), 31 the regional initiative described by Stone et al. (2012) measured HAI rates following a hygiene initiative at acute care hospitals in England and Wales, which included ABHR promotion in addition to other strategies (although in year 4 of the study, the 2009 WHO protocols for hand hygiene were adopted). The initiative titled “Cleanyourhands” was informed by Habit-Forming Theory 33 , 34 and included installation of ABHR dispensers, materials promoting hand hygiene, and regular hand hygiene audits. After 4 years, the CDI rate decreased from 16.75 to 9.49 cases per 10,000 bed days, but the report did not mention statistical significance. Researchers found that increases in the amount of soap purchased by facilities was independently associated with reduced CDI throughout the study. The researchers also noted potential confounders that they did not study (e.g., antimicrobial prescribing rates). 26
In the past decade, patient hand hygiene has received increasing attention as a potential major source of C. difficile transmission in healthcare settings. Patients colonized with C. difficile often go undetected and may transmit C. difficile to HCWs’ hands directly, or indirectly through contaminated surfaces in the healthcare environment. Patient mobility, dexterity, and cognitive limitations can be barriers to patient hand hygiene. 20 , 35 One study found patient hand hygiene compliance rates as low as 10 percent. 36
Pokrywka et al. (2017) conducted a study in a 495-bed university-affiliated medical center on a patient handwashing program focused specifically on CDI reduction. In this intervention, hospital staff were educated about specific times when they should encourage and assist patients with handwashing and hand hygiene (i.e., practicing hand hygiene prior to meals, after using the toilet or bedpan, prior to touching dressings and incisions, after returning from testing or a procedure, before and after having visitors). After a trial conducted on four units in the hospital, the initiative was implemented hospitalwide.
Post-implementation patient survey results showed some improvement in staff assistance with patient hand hygiene, and the CDI standardized infection ratio (SIR) decreased in the first two quarters after implementation, from 0.834 to 0.572 and 0.497 (p≤0.05). (The NHSN uses SIRs to track HAIs over time; the SIR compares the actual number of HAIs at each hospital with the predicted number). Infection rates increased in the third and final quarters of the measurement period, which potentially shows the need for sustained staff education and reminders to consistently educate new patients. 35
It is believed that the mechanical action and friction from handwashing helps to remove C. difficile spores from hands. To explore this theory, Isaacson et al. (2015) experimented with the use of sand to remove C. difficile spores from hands and compared these results with washing with soap and water. In this study, 14 subjects each used five different hand hygiene methods following contamination with C. difficile (4 × 10 5 colony forming units). The hand hygiene methods were water rinse, water rub and rinse, water and antibacterial soap, oil/baking soda/dish detergent/water, and sand rub and water rinse. The use of sand and water resulted in the greatest reduction in spores, but results were not significant. Compared with antibacterial soap and water, which resulted in an average 1.84 log reduction (SD 0.46) or 98.5 percent, sand and water resulted in an average 2.34 log reduction (SD 0.33) or 99.5 percent. Compared with soap and water, the sand and water method removed a statistically significant greater average amount of C. difficile spores (−0.50; p=0.003). 37
Other studies examined the efficacy of handwashing with soap and water. To compare five practical strategies for hand hygiene, Oughton et al. (2009) conducted an experiment with 10 volunteers to measure the efficacy for C. difficile spore removal from the whole hand or just the surface of the palm. The researchers found that, using both whole hand and palmar surface protocols, washing with warm water with plain soap left the lowest amount of C. difficile spores, followed by cold water with plain soap, warm water with antibacterial soap, antiseptic hand wipe, ABHR, and no hand hygiene.
Perhaps the most interesting finding from this study was that plain soap performed better than antimicrobial soap in the whole-hand protocol. 2 Washing with non-antimicrobial soap and water was more effective for removing C. difficile than 4% chlorhexidine gluconate hand wash. The researchers speculate that this finding may be because a higher amount of organic matter is present on the whole hand than on the palm, and high levels of organic matter interfere with the activity of chlorhexidine. Edmonds et al. (2013) found similar results and noted that the most effective antibacterial products were too harsh to be used on human skin (e.g., peracetic acid surfactant prototype [Triton-X], commercial ink and stain remover, sodium tetraborate decahydrate powder [Borax]). 8
Tomas et al. (2015) explored preventing HCW hand contamination from the removal of gloves and other personal protective equipment. The study found that, after CDI patient care, 16 percent of HCWs had CDI spores on their hands after removing gloves and personal protective equipment (n=25). The frequency of contamination was reduced to 7 percent after an educational intervention on proper glove/gown removal (p=0.4) and further reduced to 0 percent after disinfection of gloves with bleach wipes (p=0.04). 12
Due to complaints of irritation from the bleach wipes, Tomas et al. (2016) conducted a second study in which HCWs used a sporicidal formula (of acidic ethanol) instead of bleach for glove decontamination (to use before glove removal). The findings suggest that the sporicidal properties of certain solutions could be useful for glove disinfection before removal, when caring for CDI patients. The reduction achieved by the sporicidal ethanol solution (70% ethanol pH 1.3) was equivalent to the 1:100 dilution of bleach on artificially contaminated gloves. Researchers tested glove contamination of HCWs following 159 CDI patient care episodes and found that the sporicidal ethanol resulted in significantly reduced glove contamination, whereas 70% ethanol did not. Despite the promise of glove decontamination as a prevention strategy, the authors stipulate that decontaminating gloves would not replace HCWs washing their hands after glove removal. 15
In general, the literature regarding hand hygiene indicates that the costs associated with preventing HAIs far outweigh the costs to improve hand hygiene compliance. 29 , 32 Sickbert-Bennett et al. (2016) reported that the cumulative prevention of HAIs saved approximately $5 million at their institution. 32 Although some cost-effectiveness analyses are available for hand hygiene programs in general, we could not find financial outcomes related to hand hygiene and CDI specifically. To better understand and encourage the implementation of hand hygiene initiatives, it would be beneficial to take into account the cost of a hand hygiene initiative (staffing, staff time, supplies, installation of sinks, etc.), as well as the costs of sustaining a program, and compare these totals with estimated savings in terms of medical costs from CDI prevention.
Among the review’s conclusions were recommendations that hand hygiene education be interactive and engaging and that interventions be tailored to the institution’s unique needs. 38 Researchers have assessed barriers to hand hygiene and report that hand hygiene interventions should be tailored to the particular classification/role of staff and that context and staff needs should be taken into account when designing hand hygiene interventions. For example, Kirkland et al. (2012) noted that regular review of data linking hand hygiene performance to HAIs was persuasive for physicians, but they were less likely to engage in educational programs geared toward staff with less medical knowledge. 29
In an example of addressing a facility’s unique needs, Al-Tawfiq et al. (2018) described positive experience using The Joint Commission Center for Transforming Healthcare’s web-based Targeted Solutions Tool ® (TST ® ) to improve hand hygiene and reduce HAIs in a 30-bed oncology/hematology inpatient unit in Saudi Arabia. The tool is designed to identify root causes of nonadherence to hand hygiene and improve process outcomes. Researchers found that housekeepers needed more help than other staff help with improving hand hygiene, but these workers were not fluent in either English or Arabic (the dominant languages) and their educational levels varied substantially. To address this issue, an extensive training program was developed for housekeeping staff using in-action learning tools and translators. After 1 year, the hand hygiene compliance rate increased from 75.4 percent at baseline to 88.6 percent (p <0.0001). Researchers found a decrease in CDIs from 7.95 (CI, 0.8937 to 28.72) to 1.84 (CI, 0.02411 to 10.26) infections per 10,000 patient days that was not significant (p=0.23) and cited sample size as a barrier to statistical significance.30
An interactive strategy to assist HCWs in improving glove and gown removal technique includes the use of fluorescent lotion. In the training described by Tomas et al. (2015), fluorescent lotions were used to help HCWs learn proper glove and gown removal to minimize hand contamination. The fluorescent lotion provides immediate visual feedback on contaminated sites. 12 A similar strategy includes the use of nonpathogenic RNA beads that fluoresce under ultraviolet (UV) light to help track contamination during removal of personal protective equipment. This practice can help HCWs see that glove use does not preclude the need for hand hygiene. 39
The design of the healthcare environment can affect hand hygiene compliance. Some researchers suggest a human factors engineering approach that calls for abundant, convenient, and available sinks, handwashing products, and ABHRs to improve compliance. 40 Several researchers found that longer distances to sinks, and sink visibility, were related to HCW handwashing compliance. For example, Zellmer et al. (2015) reviewed the practices of HCWs and visitors for CDI-positive patients on a transplant medical-surgical unit at a large academic medical center. While there were sinks in the patients’ rooms, these were not used due to the placement of furniture, patients’ personal items blocking access, and lack of foot pedals. Before the study began, the only two easily accessible sinks were at the end of a hallway. After the installment of two highly visible sinks in the unit, completion of proper hand hygiene on exiting the CDI patient room improved by 18 percent (p=0.03). 24
In another example, Deyneko et al. (2016) investigated the relationship between sink location and HCW compliance with handwashing; their multivariate analysis found that increased distance between the patient zone and the nearest sink was inversely associated with handwashing compliance. The median distance to the nearest sink was 7.6 meters when hand hygiene was correctly performed, but 14.9 meters when it was omitted (p <0.001). There was also a strong association between the number of 90° turns required to reach the sink and handwashing compliance.23
Sequence for putting on and removing personal protective equipment: https://www.cdc.gov/hai/pdfs/ppe/PPE-Sequence.pdf
Veterans Health Administration: Infection: Don’t Pass It On education and communication materials: https://www.publichealth.va.gov/infectiondontpassiton/index.asp
As already noted, there is a need for more real-world randomized and crossover hand hygiene studies in which CDI prevention is a primary focus. One of the most important omissions of the reviewed clinical/quasi-experimental studies was that compliance with hand hygiene practices specific to CDI was not distinctly measured and reported. In several of the reviewed studies, hand hygiene processes (end points) were clinician hand hygiene at the appropriate moments, not whether a CDI-appropriate method (e.g., use of gloves and washing hands in outbreak/hyperendemic settings) was used. 30 , 32 CDI-specific research would help improve understanding about the impact of using ABHRs versus handwashing when working with CDI patients. In addition, the strength of the research on hand hygiene in clinical settings and hand hygiene methods was limited by small sample sizes.
Research on hand hygiene interventions in a wide variety of setting types (and in multiple settings) is needed given that hand hygiene behaviors and challenges differ across settings. Neo et al. (2016) found in their review that most studies of hand hygiene interventions were in hospitals or ICUs. 38 As CDI disproportionately impacts elderly and immunocompromised patients, more research is needed on CDI and hand hygiene in LTCFs that serve these specific patient populations. In addition, LTCFs have unique staffing and environmental factors and require different types of patient contacts than hospitals do. Many nursing home facilities are designed to encourage social contact between patients, and patients move throughout the facility coming into contact with spaces outside their rooms (e.g., dining room, physical therapy room). In such settings, hand hygiene programs aimed at patients could be particularly impactful. Additional studies in the outpatient setting would also be useful.
Patient hand hygiene is a promising area of prevention and research. As the role of colonized patients is increasingly understood, patient hand hygiene analyses will likely account for patients with asymptomatic colonization in addition to those with CDI. As found by Kundrapu et al. (2014), the numbers of CDI colonies recovered from patients’ hands were similar for those diagnosed with CDI and asymptomatic carriers. 19 Due to some of the barriers for patient hand hygiene, including mobility, some have suggested more research into the potential of using skin-safe cleaning wipes with C. difficile eliminating agents (e.g., sporicidal electrochemically generated hypochlorous acid solution) for patients who cannot ambulate or be brought to sinks for routine handwashing. 19 , 41 Patient education about C. difficile is potentially important. Kundrapu et al. (2014) found that 73 percent of colonized and infected patients in their study were not aware that ABHR does not kill C. difficile spores. 19
Some research has been conducted to identify new ways to decontaminate HCWs’ hands. Researchers may continue to explore potential noncorrosive hand rubs that provide the convenience of a hand rub and are more effective at killing all pathogens, including C. difficile spores. 35 For example, an experimental study by Nerandzic et al. (2013) found that sporicidal electrochemically generated hypochlorous acid solution (Vashe), used to soak or as a wipe, is effective in reducing spores. Wiping with Vashe-soaked cloths significantly enhanced reduction of C. difficile spores by approximately 68 percent (0.5 log10 CFU [colony-forming unit]; p0.01). 41 Vashe is FDA approved for use on wounds, and more research is needed to determine safety for other uses. In addition, more real-world research is needed to determine efficacy for HCW exposure to C. difficile.
Direct and persistent observation is both a study technique and an intervention to encourage hand hygiene. There are some limits to in-person monitoring, including cost, feasibility of achieving sufficient sample size, sustainability, potential for HCWs to temporarily alter behavior while being observed, and lack of consistency (within and across studies) for measuring compliance. Monitoring by video is another observation strategy that eliminates the physical presence of the observer but has some of the same drawbacks as in-person monitoring. 41
Staats et al. (2017) studied the use of electronic monitoring, using radio frequency identification, in 71 hospital units. HCWs were given badges that communicated with a network of sensors throughout the hospital and at hand hygiene stations. Monitoring measured whether the HCWs used hand hygiene stations at the appropriate place and time. The researchers found that electronically monitoring individual compliance resulted in a large, positive increase in compliance that was not sustained. 43
One drawback of electronic monitoring and censors is cost, and more research is needed. Other strategies include use of electronic counters on ABHRs and measuring handwashing product use. The drawbacks of these strategies is they do not account for appropriate hand hygiene technique, hand hygiene moments, and person using the product (patients and visitors may also use these products). 42
The use of gloves for preventing transmission of CDI is strongly recommended in the guidelines yet not well studied in the healthcare setting. More research could be done on promoting HCW compliance with glove use, barriers and facilitators, and best practices for glove use when working with CDI patients.
Finally, interventions for hand hygiene will need to address issues of sustainability, as multiple studies reported declines in compliance after the hand hygiene intervention period. 35 , 43 For example, Pokrywka et al. (2017) report that sustainability requires ongoing leadership, continued staff reminders, education for new staff, and ongoing resources, without which hand hygiene compliance rates will fall. 35 Kirkland et al. (2012) report that understanding the hospital context, based on responses to the initiative across units and HCW types, helped sustain improved hand hygiene compliance rates for a year following a 3-year hand hygiene initiative. 29 Additional research concerning the sustainability of hand hygiene programs would be helpful to improve understanding.
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Oughton MT, Loo VG, Dendukuri N, et al. Hand hygiene with soap and water is superior to alcohol rub and antiseptic wipes for removal of Clostridium difficile. Infect Control Hosp Epidemiol. 2009;30(10):939–44. doi: 10.1086/605322. [PubMed : 19715426 ] [CrossRef]
Jabbar U, Leischner J, Kasper D, et al. Effectiveness of alcohol-based hand rubs for removal of Clostridium difficile spores from hands. Infect Control Hosp Epidemiol. 2010;31(6):565–70. doi: 10.1086/652772. [PubMed : 20429659 ] [CrossRef]
Centers for Disease Control and Prevention. Clean Hands Count Campaign. 2016. https://www .cdc.gov/handhygiene /campaign/index.html.
Deschenes P, Chano F, Dionne LL, et al. Efficacy of the World Health Organization-recommended handwashing technique and a modified washing technique to remove Clostridium difficile from hands. Am J Infect Control. 2017;45(8):844–8. doi: 10.1016/j.ajic.2017.04.001. [PubMed : 28526314 ] [CrossRef]
Boyce JM, Pittet D, Healthcare Infection Control Practices Advisory Committee, et al. Guideline for hand hygiene in health-care settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Infect Control Hosp Epidemiol. 2002 Dec;23(S12):S3–40. doi: 10.1086/503164. [PubMed : 12515399 ] [CrossRef]
World Health Organization. WHO guidelines on hand hygiene in health care: First Global Patient Safety Challenge, Clean Care Is Safer Care. Geneva: World Health Organization; 2009. [PubMed : 23805438 ]
Edmonds SL, Zapka C, Kasper D, et al. Effectiveness of hand hygiene for removal of Clostridium difficile spores from hands. Infect Control Hosp Epidemiol. 2013;34(3):302–5. doi: 10.1086/669521. [PubMed : 23388366 ] [CrossRef]
Dubberke E, Gerding D. Rationale for Hand Hygiene Recommendations After Caring For a Patient With Clostridium difficile Infection. A Compendium of Strategies to Prevent Healthcare-Associated Infections in Acute Care Hospitals; Fall 2011. https://www .shea-online .org/images/patients /CDI-hand-hygiene-Update.pdf.
Bearman G, Rosato AE, Duane TM, et al. Trial of universal gloving with emollient-impregnated gloves to promote skin health and prevent the transmission of multidrug-resistant organisms in a surgical intensive care unit. Infect Control Hosp Epidemiol. 2010;31(5):491–7. doi: 10.1086/651671. [PubMed : 20350197 ] [CrossRef]
Jullian-Desayes I, Landelle C, Mallaret MR, et al. Clostridium difficile contamination of health care workers’ hands and its potential contribution to the spread of infection: Review of the literature. Am J Infect Control. 2017;45(1):51–8. doi: 10.1016/j.ajic.2016.08.017. [PubMed : 28065332 ] [CrossRef]
Tomas ME, Sunkesula VC, Kundrapu S, et al. An intervention to reduce health care personnel hand contamination during care of patients with Clostridium difficile infection. Am J Infect Control. 2015;43(12):1366–7. doi: 10.1016/j.ajic.2015.07.017. [PubMed : 26654239 ] [CrossRef]
Ragusa R, Giorgianni G, Lupo L, et al. Healthcare-associated Clostridium difficile infection: Role of correct hand hygiene in cross-infection control. J Prev Med Hyg. 2018;59(2):E145–e52. [PMC free article : PMC6069405 ] [PubMed : 30083622 ]
Landelle C, Verachten M, Legrand P, et al. Contamination of healthcare workers’ hands with Clostridium difficile spores after caring for patients with C. difficile infection. Infect Control Hosp Epidemiol. 2014;35(1):10–5. doi: 10.1086/674396. [PubMed : 24334792 ] [CrossRef]
Tomas ME, Nerandzic MM, Cadnum JL, et al. A novel, sporicidal formulation of ethanol for glove decontamination to prevent Clostridium difficile hand contamination during glove removal. Infect Control Hosp Epidemiol. 2016;37(3):337–9. doi: 10.1017/ice.2015.289. [PubMed : 26679745 ] [CrossRef]
Sax H, Allegranzi B, Chraiti MN, et al. The World Health Organization hand hygiene observation method. Am J Infect Control. 2009;37(10):827–34. doi: 10.1016/j.ajic.2009.07.003. [PubMed : 20004812 ] [CrossRef]
Weaver L, Michels HT, Keevil CW. Survival of Clostridium difficile on copper and steel: Futuristic options for hospital hygiene. J Hosp Infect. 2008;68(2):145–51. doi: 10.1016/j.jhin.2007.11.011. [PubMed : 18207284 ] [CrossRef]
Weber DJ, Rutala WA, Miller MB, et al. Role of hospital surfaces in the transmission of emerging health care-associated pathogens: Norovirus, Clostridium difficile, and acinetobacter species. Am J Infect Control. 2010;38(5 Suppl 1):S25–33. doi: 10.1016/j.ajic.2010.04.196. [PubMed : 20569853 ] [CrossRef]
Kundrapu S, Sunkesula V, Jury I, et al. A randomized trial of soap and water hand wash versus alcohol hand rub for removal of Clostridium difficile spores from hands of patients. Infect Control Hosp Epidemiol. 2014;35(2):204–6. doi: 10.1086/674859. [PubMed : 24442089 ] [CrossRef]
Barker AK, Zellmer C, Tischendorf J, et al. On the hands of patients with Clostridium difficile: A study of spore prevalence and the effect of hand hygiene on c difficile removal. Am J Infect Control. 2017;45(10):1154–6. doi: 10.1016/j.ajic.2017.03.005. [PMC free article : PMC5731647 ] [PubMed : 28964347 ] [CrossRef]
Erasmus V, Daha TJ, Brug H, et al. Systematic review of studies on compliance with hand hygiene guidelines in hospital care. Infect Control Hosp Epidemiol. 2010;31(3):283–94. doi: 10.1086/650451. [PubMed : 20088678 ] [CrossRef]
Smith A, Carusone SC, Loeb M. Hand hygiene practices of health care workers in long-term care facilities. Am J Infect Control. 2008;36(7):492–4. doi: 10.1016/j.ajic.2007.11.003. [PubMed : 18786453 ] [CrossRef]
Deyneko A, Cordeiro F, Berlin L, et al. Impact of sink location on hand hygiene compliance after care of patients with Clostridium difficile infection: A cross-sectional study. BMC Infect Dis. 2016;16:203. doi: 10.1186/s12879-016-1535-x. [PMC free article : PMC4867524 ] [PubMed : 27184488 ] [CrossRef]
Zellmer C, Blakney R, Van Hoof S, et al. Impact of sink location on hand hygiene compliance for Clostridium difficile infection. Am J Infect Control. 2015;43(4):387–9. doi: 10.1016/j.ajic.2014.12.016. [PubMed : 25704256 ] [CrossRef]
Louh IK, Greendyke WG, Hermann EA, et al. Clostridium difficile infection in acute care hospitals: Systematic review and best practices for prevention. Infect Control Hosp Epidemiol. 2017;38(4):476–82. doi: 10.1017/ice.2016.324. [PMC free article : PMC5560033 ] [PubMed : 28300019 ] [CrossRef]
Stone SP, Fuller C, Savage J, et al. Evaluation of the national Cleanyourhands campaign to reducetaphylococcus aureus bacteraemia and Clostridium difficile infection in hospitals in England and Wales by improved hand hygiene: Four year, prospective, ecological, interrupted time series study. BMJ. 2012;344:e3005. doi: 10.1136/bmj.e3005. [PMC free article : PMC3343183 ] [PubMed : 22556101 ] [CrossRef]
Knight N, Strait T, Anthony N, et al. Clostridium difficile colitis: A retrospective study of incidence and severity before and after institution of an alcohol-based hand rub policy. Am J Infect Control. 2010;38:523–8. [PubMed : 20392538 ]
Silva M, Jr., Marra AR, Camargo TZ, et al. Secular trends in the epidemiology of Clostridium difficile infection (CDI): Relationship with alcohol gel and antimicrobial usage in a hospital. Int J Infect Dis. 2013;17(6):e418–21. doi: 10.1016/j.ijid.2012.12.022. [PubMed : 23419978 ] [CrossRef]
Kirkland KB, Homa KA, Lasky RA, et al. Impact of a hospital-wide hand hygiene initiative on healthcare-associated infections: Results of an interrupted time series. BMJ Qual Saf. 2012;21(12):1019–26. doi: 10.1136/bmjqs-2012-000800. [PubMed : 22822243 ] [CrossRef]
Al-Tawfiq JA, Treble M, Abdrabalnabi R, et al. Using targeted solution tools as an initiative to improve hand hygiene: Challenges and lessons learned. Epidemiol Infect. 2018;146(2):276–82. doi: 10.1017/s0950268817002758. [PMC free article : PMC9134745 ] [PubMed : 29235431 ] [CrossRef]
Schweon SJ, Edmonds SL, Kirk J, et al. Effectiveness of a comprehensive hand hygiene program for reduction of infection rates in a long-term care facility. Am J Infect Control. 2013;41(1):39–44. doi: 10.1016/j.ajic.2012.02.010. [PubMed : 22750034 ] [CrossRef]
Sickbert-Bennett EE, DiBiase LM, Willis TM, et al. Reduction of healthcare-associated infections by exceeding high compliance with hand hygiene practices. Emerg Infect Dis. 2016;22(9):1628–30. doi: 10.3201/eid2209.151440. [PMC free article : PMC4994356 ] [PubMed : 27532259 ] [CrossRef]
Lally P, Gardner B. Promoting habit formation. Health Psychol Rev. 2013;7(sup1):S137–S58. doi: 10.1080/17437199.2011.603640. [CrossRef]
The National Social Marketing Center. Cleanyourhands. https://www .thensmc.com /resources/showcase/cleanyourhands. Accessed December 11, 2019.
Pokrywka M, Buraczewski M, Frank D, et al. Can improving patient hand hygiene impact Clostridium difficile infection events at an academic medical center? Am J Infect Control. 2017;45(9):959–63. doi: 10.1016/j.ajic.2017.06.019. [PubMed : 28863811 ] [CrossRef]
Sunkesula VC, Knighton S, Zabarsky TF, et al. Four moments for patient hand hygiene: A patient-centered, provider-facilitated model to improve patient hand hygiene. Infect Control Hosp Epidemiol. 2015;36(8):986–9. doi: 10.1017/ice.2015.78. [PubMed : 25857700 ] [CrossRef]
Isaacson D, Haller B, Leslie H, et al. Novel handwashes are superior to soap and water in removal of Clostridium difficile spores from the hands. Am J Infect Control. 2015;43(5):530–2. doi: 10.1016/j.ajic.2015.02.008. [PubMed : 25952050 ] [CrossRef]
Neo JR, Sagha-Zadeh R, Vielemeyer O, et al. Evidence-based practices to increase hand hygiene compliance in health care facilities: An integrated review. Am J Infect Control. 2016;44(6):691–704. doi: 10.1016/j.ajic.2015.11.034. [PubMed : 27240800 ] [CrossRef]
Casanova L, Alfano-Sobsey E, Rutala WA, et al. Virus transfer from personal protective equipment to healthcare employees’ skin and clothing. Emerg Infect Dis. 2008;14(8):1291–3. doi: 10.3201/eid1408.080085. [PMC free article : PMC2600382 ] [PubMed : 18680659 ] [CrossRef]
Steinberg JP, Denham ME, Zimring C, et al. The role of the hospital environment in the prevention of healthcare-associated infections by contact transmission. Herd. 2013;7(1_suppl):46–73. doi: 10.1177/193758671300701s06. [CrossRef]
Nerandzic MM, Rackaityte E, Jury LA, et al. Novel strategies for enhanced removal of persistent bacillus anthracis surrogates and Clostridium difficile spores from skin. PloS One. 2013;8(7):e68706. doi: 10.1371/journal.pone.0068706. [PMC free article : PMC3699662 ] [PubMed : 23844234 ] [CrossRef]
Marra AR, Edmond MB. New technologies to monitor healthcare worker hand hygiene. Clin Microbiol Infect. 2014;20(1):29–33. doi: 10.1111/1469-0691.12458. [PubMed : 24245809 ] [CrossRef]
Staats BR, Dai H, Hofmann D, et al. Motivating process compliance through individual electronic monitoring: An empirical examination of hand hygiene in healthcare. Manag Sci. 2017;63(5):1563–85. Epub 5 May 2016. doi: 10.1287/mnsc.2015.2400. [CrossRef]
The most recommended cleaning and decontamination agents for manual use are chlorine-based solutions.
In many of the reviewed studies, the addition of hydrogen peroxide decontamination (HPD) or ultraviolet light decontamination (UVD) to standard cleaning was associated with significant reductions in facility-level CDI rates.
HPD and UVD have drawbacks, including expense and the time it takes to decontaminate a room. However, the process for UVD is shorter than for HPD.
The performance of environmental cleaning services staff is important and can be improved through the use of training, checklists, and audit and feedback.
There is a need for higher quality studies, multifacility studies, and studies that compare cleaning and decontamination methods.
Future directions include research and development of nontoxic decontamination agents, new technologies, and research on patient outcomes and environmental cleaning in diverse healthcare settings.
This review includes a summary of evidence published from 2008 to 2018 on environmental cleaning and decontamination as a prevention practice for CDI. We start with a definition of terms by the CDC and a brief practice description for environmental cleaning and decontamination for C. difficile from the 2017 guidelines by the IDSA and SHEA. The review then provides an overview of how environmental cleaning and decontamination work as a safety practice for preventing the transmission of C. difficile.
Next, we summarize the evidence for the impact of environmental cleaning and decontamination interventions on CDI rates and highlight some experimental research on cleaning agents for C. difficile. We then explore implementation factors, including monitoring and improving the performance of environmental service workers and challenges with the use of decontamination equipment. Finally, we explore gaps and future directions for environmental cleaning and decontamination for C. difficile. The review’s key findings are located in the box on the right.
The CDC (2008) in their guideline for sterilization and disinfection of healthcare facilities define the practice of cleaning in the healthcare environment as “the removal of visible soil (e.g., organic and inorganic material) from objects and surfaces” (page 9). 1 The CDC defines disinfection as the elimination of many or all pathogenic microorganisms from the environment, while sterilization refers to the elimination of all forms of microbial life.
Decontamination is the process to remove pathogenic microorganisms from objects for the purposes of safe handling and use. The CDC states that cleaning (i.e., removing visible material from surfaces) is a first step in the decontamination process so that organic or inorganic material does not interfere with decontamination. As outlined in this report, the use of sporicidal agents to manually clean healthcare environments is a form of both cleaning and decontamination. Use of touchless automated methods are solely for the purpose of environmental decontamination.
Terminal room cleaning (cleaning after a patient is discharged or transferred from a room) with a sporicidal agent should be considered in conjunction with other measures to prevent CDI during endemic high rates or outbreaks, or if there is evidence of repeated cases of CDI in the same room (weak recommendation, low quality of evidence).
Daily cleaning with a sporicidal agent should be considered in conjunction with other measures to prevent CDI during outbreaks or in hyperendemic (sustained high rates) settings, or if there is evidence of repeated cases of CDI in the same room (weak recommendation, low quality of evidence).
Measures of cleaning effectiveness should be incorporated to ensure quality of environmental cleaning (good practice recommendation).
Disposable patient equipment should be used when possible and reusable equipment should be thoroughly cleaned and disinfected, preferably with a sporicidal disinfectant that is equipment compatible (strong recommendation, moderate quality of evidence).
The IDSA/SHEA state in the guidelines that they have no recommendation for the use of automated touchless terminal (i.e., upon discharge) disinfection CDI prevention due to data limitations. The CDC guidelines for environmental cleaning and decontamination for C. difficile include the creation of daily and terminal cleaning protocols and checklists for patient-care areas and equipment. 3 Other guidelines from an earlier SHEA/IDSA report for acute care facilities recommend frequent education for environmental service personnel in the primary language of the cleaning team and the use of various techniques to help improve cleaning and decontamination practice as outlined by the CDC 4 (e.g., observation, fluorescent markers, and bioluminescence). 4 , 5
Safety practices for laundry, bedding, and other environmental services are included in the CDC’s “Guidelines for Environmental Infection Control in Health Care Facilities.” 6 Guidelines for specific facility types, including hospitals, nursing homes, long-term acute care facilities, and outpatient facilities, are available from the CDC and other healthcare agencies. We include some of these resources later in this chapter.
The healthcare environment is recognized as a primary source of C. difficile transmission. 7 C. difficile is spread through the feces of infected and colonized patients. Patients with contaminated hands may spread C. difficile by touching surfaces in the healthcare environment. Some evidence suggests C. difficile may be dispersed to surfaces near the patient through droplets in the air. 8 , 9 Transmission can occur when other patients, healthcare staff, or visitors touch contaminated surfaces and orally ingest C. difficile (e.g., while eating). 7 Those who take antimicrobials, are advanced in age, or have compromised immune systems are at high risk of getting CDI from exposure to the pathogen. Others may become asymptomatic carriers of C. difficile. 2
Both symptomatic and asymptomatic carriers have the potential to contaminate the environment. In one hospital, C. difficile was recovered from 59 percent of samples in rooms of asymptomatic carriers 10 and 75 percent of samples of rooms with patients with CDI. 11 Patients may continue to contaminate the environment after treatment. 12 The most contaminated areas, or “high-touch surfaces,” include the bed rails, bed surface, supply cart, over-bed table, and intravenous pumps. 13
In one study, CHWs’ hands were just as likely to be contaminated with C. difficile after touching high-touch surfaces as they were by touching a CDI patient. 14 C. difficile produces spores that are especially robust and may remain viable in the environment for over 4 days. 15 Shaughnessy et al. (2011) examined the potential role of environmental transmission of C. difficile through a prior room occupant and found that the prior occupant’s CDI status was a significant risk factor for acquiring CDI (p=0.01; hazard ratio, 2.35), after controlling for other risk factors (e.g., antimicrobial use, age, proton pump inhibitors). 16
Eliminating C. difficile in the healthcare environment requires specialized practices. Evidence shows that C. difficile spores are resistant to alcohol and many hospital disinfectants. 17 In one study, exposure of the bacteria to low levels of certain cleaning agents resulted in higher CDI sporulation capacity (the ability for vegetative cells to forms spores during unfavorable environmental conditions). 18
Among cleaning and decontamination agents for washing surfaces by hand, chlorine-releasing solutions (e.g., bleach), at sufficient concentration and with appropriate exposure time (at least 10 minutes), demonstrate the best evidence for killing C. difficile. 17 The CDC-recommended cleaning/decontamination agents for C. difficile can be found on EPA List K: Registered Antimicrobial Products Effective Against Clostridium difficile Spores. 19
Decontamination by hand is challenging and not always effective in reaching all contaminated surfaces in the healthcare environment. 12 , 20 Automated touchless methods have been developed and implemented to supplement cleaning by hand and prevent the spread of CDI and other HAIs. The two most commonly studied touchless methods for C. difficile decontamination are hydrogen peroxide decontamination (HPD)—including vaporized, aerosolized, atomized, and dry mist systems—and ultraviolet disinfection (UVD), which includes UV radiation and pulsed xenon UV light systems. In laboratory studies, both methods have shown effectiveness in almost entirely eliminating C. difficile contamination from targeted surfaces. 21 , 22
Although subject to some debate, it is generally recommended that surfaces be precleaned by hand prior to use of UVD or HPD, as organic matter is thought to reduce the efficacy of the UVD and HPD methods. 23 In their review, Doll et al. (2015) found that studies were mixed as to which no-touch method (UVD or HPD) was most effective at killing C. difficile. The UVD methods generally take less time than HPD to decontaminate a room. 23
There is increasing incentive for facilities to implement an effective environmental cleaning and decontamination program as facility rankings and CMS reimbursement rates are tied to reported rates of healthcare facility-acquired onset (HO CDI). The 2016 revised requirements for participation in Medicare and Medicaid outlined the specific components of an effective infection control program, including environmental cleaning and decontamination procedures. One review found that, among several PSPs, environmental cleaning and decontamination practices were the most cost effective for reducing facility-level CDI rates. 24
The question of interest for this review is: What are the most effective and feasible environmental cleaning and decontamination practices to prevent CDI?
To answer this question, we searched the databases CINAHL and MEDLINE from 2008 to 2018 for “Clostridium difficile” and related MeSH terms and synonyms, in combination with terms such as “Disinfection,” “Decontamination,” and “No-touch decontamination.” The search string also included a variety of healthcare settings, including “hospitals,” “inpatient,” “ambulatory care,” “long-term care,” and “transitional care.” After duplicates were removed, the initial search yielded 121 results that were screened for inclusion. Of these, 45 full-text articles were retrieved. Of those, 18 studies and 3 systematic reviews were selected for this review.
Reference lists of retrieved articles were also screened to ensure thoroughness, and five studies were retrieved that way. Articles from the searches were excluded if the outcomes were not relevant or precisely reported or study design was insufficient (e.g., opinion pieces, nonsystematic reviews). Due to the number of experimental studies on this topic, a select group are included in the evidence tables and cited in the review. Studies in which environmental cleaning and decontamination were accompanied by other significant infection control practices (e.g., changes in hand hygiene practices) were ruled out for this section and are considered in Section 4.6, Multicomponent CDI Prevention Interventions.
General methods for this report are described in the Methods section of the full report.
For this patient safety practice, a PRISMA flow diagram and evidence table, along with literature-search strategy and search-term details, are included in the report appendixes A through C.
In this evidence summary, we review 12 articles and 2 reviews on environmental cleaning and CDI patient outcomes. These studies were primarily (10/12) based in hospitals and examined CDI rates after a period of enhanced cleaning and decontamination. In our search of the literature, we also found numerous experimental studies published from 2008 to 2018 on environmental cleaning and disinfection methods and CDI. Among these were three studies that compared UVD or HPD with bleach cleaning. We also found two studies on alternatives to chlorine-based solutions for the manual elimination of C. difficile from healthcare surfaces. We include a review of these experimental studies and information from one qualitative study on concerns about the effects of bleach on HCWs. Two systematic reviews included studies on environmental cleaning and CDI rates, and a third examined research on cleaning agents used to eliminate the C. difficile organism.
As shown in Table 4.3, the evidence for environmental cleaning and decontamination and CDI patient outcomes includes 12 studies published from 2008 to 2018. Most studies showed statistically significant reductions in CDI rates after a period of an environmental cleaning intervention; however, study quality was low. These findings align with the review conducted by Louh et al. (2017) in their examination of studies on CDI prevention practices in acute care hospitals from 2009 to 2015. 24 We review five of the same studies here. 25 – 29
Louh et al. (2017) reported that environmental cleaning was the most cost effective of the multiple strategies they studied. 24 Khanafer et al. (2015) found nine studies on environmental cleaning and CDI published from 1982 to December 2013. 30 They concluded that environmental cleaning with a 10:1 bleach solution was both practical and effective. Of the nine studies, four are included here; 26 , 27 , 29 , 31 we excluded the remaining studies because they were published before 2008 or measured the combined effect of several PSPs.
The environmental decontamination strategies in this review fall into one of four categories: use of a chlorine-based agent, use of a chlorine-based agent plus the use of HPD, a chlorine-based agent plus the use of UVD, and one study about washable bed covers. Within these categories, certain variables differed, such as the frequency of cleaning (e.g., daily or at discharge) and the area of cleaning (e.g., CDI patient rooms, all patient rooms, communal spaces).
The studies reviewed here were primarily quasi-experimental with a before-after approach. The study by Anderson et al. (2017) was the only randomized trial in the group of studies. 32 The cleaning intervention period ranged roughly from 8 months 31 to 2 years. 26 Two of the studies on HPD no-touch decontamination methods received some financial support from the makers of the products, in the form of free use of equipment 33 and reduced cost to use the products. 31 Two UVD studies had more than one author who was an employees of Xenex, the company that sells the machines that were studied in the intervention. 34 , 35
Studies From 2008 to 2018 on Environmental Cleaning/Decontamination and CDI Patients.
Two of the reviewed studies examined patient outcomes after a period in which patient rooms were cleaned with bleach either daily or at patient discharge. Hacek et al. (2010) evaluated a cleaning intervention at three hospitals with a total of approximately 850 beds in which terminal cleaning of the rooms occupied by CDI patients was conducted with a bleach solution (5,000 ppm) as a replacement for quaternary ammonium compound. In addition to the switch to bleach, walls were added to a checklist of surfaces to clean after patient discharge. The change in cleaning practices was a response to increases in CDI at the hospitals. The cleaning initiative included periodic unannounced cleaning assessments by supervisory staff.
Following 2 years of the new cleaning procedures, the average number of CDI patients per 1,000 patient days decreased from 0.85 before the use of bleach to 0.45 during bleach cleaning. There was a 48 percent reduction in the prevalence density of CDI (95% CI, 36% to 58%, p <0.0001) compared with the 10 prior months. The researchers report that there were no other significant infection prevention practice changes during the cleaning intervention implementation period.26
Orenstein et al. (2011) measured CDI outcomes following a cleaning intervention on two hospital wards with high baseline incidences of CDI. The cleaning program included switching from the use of quaternary ammonium compound to that of germicidal bleach wipes (5,500 ppm active chlorine) for daily and terminal cleaning of patient rooms. To evaluate progress and cleaning performance, certain rooms were randomly assessed for cleanliness with the use of adenosine triphosphate bioluminescence, which detects organic matter on surfaces.
Three reviewed studies examined the use of HPD for patient room decontamination and found reductions in CDI rates. 29 , 31 , 33 The three cleaning and decontamination interventions all added the use of HPD to cleaning with bleach and were using bleach for terminal cleaning of CDI rooms prior to the intervention. The frequency of HPD varied across the studies, ranging from a one-time HPD deep clean of a ward, 33 to priority-based HPD terminal cleaning of rooms, 29 to a one-time deep HPD cleaning of five high-incidence wards followed by terminal HPD cleaning of CDI patient rooms. 31
Boyce et al. (2008) found that, following a deep cleaning of five wards with HPD, then 8 months of terminal cleaning of CDI-occupied rooms with bleach and HPD, the incidence of nosocomial CDI decreased from 2.28 to 1.28 cases per 1,000 patient days (p=0.047). 31 Manian et al. (2013) evaluated an intervention at a 900-bed community hospital, in which HPD was added to terminal cleaning of all rooms. When HPD decontamination was not possible, CDI rooms were cleaned with four rounds of bleach cleaning. After approximately 7 months, the rate of nosocomial CDAD dropped significantly, from 0.88 cases/1,000 patient days to 0.55 cases/1,000 patient days (rate ratio, 0.63; 95% CI, 0.50 to 0.79, p <0.0001). These results are somewhat difficult to interpret as approximately half of the CDI rooms were cleaned with HPD and half were cleaned with four rounds of bleach cleaning.29
Six studies selected for this review examined the use of UVD and CDI patient outcomes. Of these, four studies showed statistically significant decreases in CDI following a period of UVD added to standard terminal cleaning with bleach of CDI patient rooms 25 , 28 , 34 , 35 and one found borderline significant reductions in CDI. 36 In one example, Vianna et al. (2016) report on the addition of UVD to terminal cleaning with bleach in a 206-bed hospital. The terminal UVD procedure was implemented for all room discharges in the ICU and for rooms occupied by patients with C. difficile in the rest of the hospital.
Following 21 months of the UVD intervention, the researchers reported a 41 percent decrease in CDI (p=0.01). CDI reductions were greater in the ICU than in the rest of the hospital (61% vs. 29%). The results indicate that UVD is effective when deployed to higher risk/higher acuity settings (e.g., the ICU) and/or when used in all room discharges (not just for patients with C. difficile). One potential confounder was an ASP, implemented 11 months prior to adoption of UVD. However, this change was not statistically linked to the reduction in CDI rates during the UVD period. 34
Long-term acute care facilities have different environmental cleaning/decontamination needs than hospitals. For example, patient stays are longer than in the hospital, so patient rooms turn over less frequently. In a study of CDI patient outcomes and environmental cleaning in a long-term acute care facility, Miller et al. (2015) looked at the addition of UVD to standard procedures for cleaning patient rooms at discharge and for cleaning common areas on an approximately weekly basis. For rooms occupied by C. difficile patients, standard procedures also included cleaning with a bleach solution.
During a 15-month period of added UVD, CDI rates decreased from 19.3 per 1,000 patient days to 8.3 per 1,000 patient days, a 56.9 percent reduction (p=0.02). It is important to note that in the prior year, the facility had implemented additional infection prevention measures consisting of education for staff around hand hygiene for CDI, disposable equipment, additional handwashing sinks, reminders about equipment decontamination, and a checklist for terminal cleaning. It is possible that the reductions in CDI rates reflect the longer term impact of these measures. 35
In the most robust study, less favorable results were found in a broad cluster-randomized study of nine hospitals, in which terminal cleaning with bleach of all rooms occupied by CDI patients was compared with terminal cleaning with bleach plus UVD. In this crossover trial, Anderson et al. (2017) found that, comparing the strategies for 7 months each, the incidence of CDI infection among patients exposed to rooms previously occupied by patients with CDI was unchanged (n=38 vs 36; 30.4 cases vs 31.6 cases per 10,000 exposure days; relative risk 1.0, 95% CI 0.57 to 1.75, p=0.997). 32
Hooker et al. (2015) examined CDI rates associated with the introduction of launderable bed covers at two long-term acute care hospitals. The researchers note that prior studies had shown that HAIs could be spread through contaminated mattresses (which are difficult to clean without damaging) and bedframes (i.e., bed decks). To prevent this source of transmission, the cleaning intervention consisted of the use of washable bed covers that covered both the mattress and bed deck. (The covers consisted of the same material used in high-end mattresses and allow moisture transmission.) The washable covers were used on all patient beds, removed after every patient discharge, and replaced with a clean cover.
After 14 months of use of the bed covers, the rate of CDIs at one hospital decreased 47.8 percent (95% CI, 47.1 to 48.6), controlling for the rate of handwashing compliance and length of stay in days. At the second hospital, the rate of CDIs decreased by 50 percent (95% CI, 47.5 to 52.7), controlling for the rate of handwashing compliance and length of stay in days. Data were not available on antimicrobial use, so this variable was not factored into the analyses. Hooker and colleagues (2015) theorized that, in addition to reducing the spread of C. difficile, the use of bed covers could help to reduce room turnover time between patients as the bed surfaces did not require thorough cleaning. 37
A number of studies and one review compare the performance of different cleaning agents and methods in removal/eradication of the C. difficile organism. We provide a sample of studies in the next two segments.
Several experimental studies compared the touchless methods with bleach cleaning with mixed results. Ghantoji et al. (2015) examined whether, after cleaning with standard detergents, terminal cleaning with bleach solution or UVD was more effective at removing C. difficile. High-touch surfaces in rooms previously occupied by CDI patients were sampled after discharge and before and after the use of both methods. The researchers found that the difference in final contamination levels between the two cleaning protocols was not significant (p=0.98). 38 Similarly, Mosci et al. (2017) looked at hydrogen peroxide and silver ion solution compared with cleaning with bleach following standard cleaning for removing C. difficile on different surfaces in a hospital. After disinfection, 0 percent (p<0.001) of samples were contaminated with C. difficile after HPD, and 3 percent (p <0.001) of samples were contaminated after bleach cleaning. The differences between groups was not statistically significant and the time for each cleaning intervention was roughly the same.39
Barbut et al. (2009) found that an in situ hydrogen peroxide dry mist system was more effective than 0.5% sodium hypochlorite solution at eradicating C. difficile spores; samples taken from hydrogen peroxide-treated rooms showed a 91 percent decrease in C. difficile, whereas samples taken after hypochlorite decontamination showed a 50 percent decrease in C. difficile (p <0.005).40
While cleaning with bleach and chlorine-based solutions has been shown to be highly effective in eliminating C. difficile from surfaces, these agents can be corrosive to metals and irritating to skin and mucus membranes. 17 Housekeepers have reported respiratory irritation when using bleach and other chlorine-based disinfectants. 41 One reason for terminal cleaning rather than daily cleaning of CDI patient rooms is for environmental services staff to avoid excessive exposure to bleach. 26 Concerns for patients and employees include the appearance of bleach residue left on surfaces, odors, and respiratory tract irritation. 41 Due to the toxicity of bleach, the Occupational Safety and Health Administration recommends using gloves and eye protection, ventilating the room properly, preparing the bleach solution daily, and allowing the solution to stand at least 30 minutes after preparation before use.
Several studies have examined potential alternatives to bleach. For example, Alfa et al. (2008) looked at different formulations of hydrogen peroxide for cleaning toilets contaminated with C. difficile. The researchers found that one of the tested hydrogen peroxide alternatives was equivalent to bleach 1,000 ppm after 1 minute but was not as efficient as that achieved for bleach at 5,000 ppm (1:10 bleach to water). 42
Peracetic acid has performed similarly to bleach. 43 Kundrapu et al. (2012) studied the potential use of a peracetic acid-based disinfectant because preliminary studies indicated that it was as effective as bleach solution but less corrosive and irritating. The peracetic acid was associated with a significant reduction in the frequency of acquisition of pathogens on investigators’ hands after contact with the surfaces and in the mean number of colony-forming units acquired. Patients in the rooms reported no adverse effects during use of the product, and there were no complaints from the nursing staff. 44
In the reviewed studies, there was limited financial information on the studied cleaning and disinfection interventions. The article by Orenstein et al. (2011) was an exception, reporting that the cost of the bleach wipes used for the daily and terminal cleaning of two medical units was $12,684 per year. They estimated that 27 cases of healthcare-associated CDI were prevented in this study, resulting in healthcare savings of between $135,000 and $216,000. While additional staffing time for daily and terminal bleach cleaning was not factored into the analyses, the researchers say that “it added little extra time to the housekeepers’ daily routine” (page 1138), indicating that there were minor increases in room turnover time. 27
Other reviewed studies provided some information about the costs of UVD and HPD. These findings are summarized in Table 4.4. Specifically, Miller et al. (2015) and Vianna et al. (2016) reported that UVD was cost effective in terms of CDIs avoided. 34 , 35 Levin et al. (2013) reported that the cost to lease two UVD machines was less than $5,000 per month 28 and Doan et al. (2012) estimated the cost of HPD equipment was $1,154.98 per month. 43
Ghantoji et al. (2015) reported that UVD was more cost effective than HPD, primarily because of the time needed to use each device—HPD takes longer than UVD per room. Both methods require that rooms be vacant and items be placed in a manner that allows adequate contact with the hydrogen peroxide mist or UV light. Before the HPD process starts, all heating, ventilation, and air-conditioning ducts in the area need to be sealed. 38
Boyce et al. (2008) reported that the HPD process took approximately 3 to 4 hours per patient room and approximately 12 hours for an entire ward. Doll et al. (2015) stated the time per room for UVD depended on the type of UVD; pulsed xenon UV takes 15 to 20 minutes and UVC radiation takes 20 to 40 minutes. 31 Haas et al. (2014) reported that the time for UVD light exposure in their study was around 6 minutes, but it took close to a half hour for setup (including setting up blackout curtains), depending on the room. Haas et al. (2014) also reported that cleaning can be more efficient by using UVD first in the bathroom, while finishing cleaning the larger room by hand. 25
While UVD may be more time efficient than HPD, it has some limitations; the process has decreased effectiveness at higher distances (over 1.22 m) and cannot decontaminate items in shadow. 36 Finally, in their review of multiple cleaning methods, Doan et al. (2012) report that decontamination with bleach was cheaper than and as effective as touchless methods. 43
Cost, Decontamination Time, and Setup for HPD and UVD.
One of the challenges reported across several of the studies on HPD and UVD was being able to use the touchless machines in all intended cases. 28 , 29 For example, Levin et al. (2013) reported that the goal was to conduct terminal UVD on all contact precautions rooms but only 56 percent of discharged contact precautions rooms received the UVD treatment. This discrepancy was due to limited device availability or the presence of a second room occupant. 28
Similarly, Haas et al. (2014) reported 76 percent of contact precautions rooms received the UVD treatment, rather than the intended 100 percent. Reasons for not conducting the UVD included a second room occupant who could not be moved, an urgent need for the room, and labor constraints. 25 Manian et al. (2013) report that using a system that prioritized use of the HPD machine based on the HAI of the discharging patient (with CDI as the top priority) allowed the machine to be used for rooms not inhabited by CDI patients when possible. When the HPD machine was not available for a CDI room, the room was cleaned multiple times with bleach. 29
Compliance with cleaning procedures is essential for eliminating active C. difficile from the environment. Research shows that touchless methods require appropriate operation. For example, the UVD machine may require repositioning in order to be most effective. 23 , 36 Ways to assist with manual cleaning compliance include cleaning checklists and audit and monitoring. Khanafer et al. (2015) recommend the use of checklists to guide housekeepers on the cleaning sequence and provision of education and direct and immediate feedback to environmental services staff. 30
Denton et al. (2016) discussed survey results from cleaning staff and others following a period of use of an audit and monitoring tool. They reported positive responses about the tool, saying that education of—and investment by—the housekeeping staff, in addition to positive, approachable, and supportive leaders, helped make the tool effective. 45 The use of adenosine triphosphate bioluminescence 27 or fluorescent markers can be effective in auditing/monitoring the thoroughness of cleaning and a basis from which to provide feedback. 46
SHEA/APIC Guideline: Infection Prevention and Control in the Long-Term Care Facility: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3319407/
There are several gaps in the studies on environmental cleaning for CDI prevention. While much of the evidence is promising for the environmental cleaning interventions included in this review, there is a need for more high-quality (e.g., randomized, robust) studies in diverse healthcare environments and larger multifacility studies to better understand this PSP. The only randomized/crossover study, by Anderson et al. (2017), found no significant change in CDI incidence following the addition of UVD to bleach cleaning for room discharges at nine hospitals. 32 More randomized studies are needed to compare the evidence. In addition, more robust financial evaluations that investigate the various methods and combinations of methods and incorporate staff time, room turnover time, and cost of no-touch devices and other cleaning machines and supplies would be beneficial.
There is also a gap in the literature with regard to cleaning and CDI patient outcomes outside of the patient rooms in the larger facility environment. Only Miller et al. (2015) describe decontamination of common areas, 35 while Best et al. (2014) and Boyce et al. (2008) describe one-time “deep” cleaning of entire wards using HPD. 31 , 33 While patient rooms are the primary focus of most of the reviewed studies, C. difficile contamination has been found in nonisolation rooms, in physician and nurse work areas, and on portable equipment. 47
Finally, there is a shortage of studies on environmental cleaning/decontamination in long-term facilities, outpatient, and other nonhospital settings. We identified only two studies of sufficient sample size on environmental cleaning and CDI outcomes in long-term acute care settings. 35 , 37 Nursing home residents are at high risk for CDI due to frequent antimicrobial exposure and the relatively high number of colonized patients in LTCFs. A systematic review found that 14.8 percent (95% CI, 7.6% to 24.0%) of LTCF residents are asymptomatic carriers of toxigenic C. difficile. 48
CDI recurrence is also high in LTCFs due to new infection or recurrence of the original infection. Given longer patient stays and the presence of more patient belongings (creating additional possible transmission pathways), and that LTCFs are intended to promote social interaction, LTCFs have unique environmental decontamination needs that require further study. 49
Future directions for environmental cleaning practices to prevent C. difficile transmission include advances in hospital equipment and standard hospital items. 50 For example, research has explored the use of copper for hospital surfaces (e.g., cabinets, tables). Copper has been shown to provide a significant (>70 percent) reduction in survival of C. difficile vegetative cells and spores on copper alloys compared with stainless steel. 15 Sporicidal properties in common hospital items such as curtains has also been explored. 51 Installation of items such as toilet lids can help prevent the spread of CDI droplets when a contaminated toilet is flushed. 8 Some studies show that microfiber cloths (made of a combination of polyamide and polyester) perform better than standard cotton materials at removing C. difficile. 52
Future research could build on and enhance existing cleaning and decontamination technologies. One example is hand-held wands that can be used on items such as keyboards and portable medical devices to kill pathogens with UV radiation. 53 Another example involves rendering C. difficile spores more susceptible to UVD and increasing the efficacy of UVD by initiation of C. difficile germination. (The initiation of germination has been shown to make spores more susceptible to heat and radiation.) Application of germination solution to a contaminated surface prior to UVD was shown to increase the number of spores killed by UVD compared with UVD alone. 54 Finally, continued research on environmental services systems and efficacy of methods, as well as improved support and training of environmental services workers, will help to advance cleaning and decontamination practices in the future.
Rutala W, Weber D. Guideline for disinfection and sterilization in healthcare facilities. Atlanta, GA: Centers for Disease Control and Prevention; 2008. https://www .cdc.gov/infectioncontrol /pdf /guidelines/disinfection-guidelines-H .pdf.
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Centers for Disease Control and Prevention. CDI prevention strategies: Strategies to prevent Clostridioides difficile infection in acute care facilities. https://www .cdc.gov/cdiff /clinicians/cdi-prevention-strategies.html. Accessed December 11, 2019
Guh A, Carling P, Environmental Evaluation Wokrgroup. Options for evaluating environmental cleaning. Atlanta, GA: Centers for Disease Control and Prevention,; 2010. https://www .cdc.gov/hai /pdfs/toolkits/Environ-Cleaning-Eval-Toolkit12-2-2010.pdf.
Dubberke ER, Carling P, Carrico R, et al. Strategies to prevent Clostridium difficile infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35(6):628–45. doi: 10.1086/676023. [PubMed : 24799639 ] [CrossRef]
Sehulster L, Chinn R, Arduino M, et al. Guidelines for environmental infection control in health-care facilities. Recommendations from CDC and the Healthcare Infection Control Practices Committee (HICPAC). Chicago, IL: American Society for Healthcare Engineering/American Hospital Association; 2004. https://www .cdc.gov/infectioncontrol /pdf /guidelines/environmental-guidelines-P .pdf.
Weber DJ, Rutala WA, Miller MB, et al. Role of hospital surfaces in the transmission of emerging health care-associated pathogens: norovirus, Clostridium difficile, and acinetobacter species. Am J Infect Control. 2010;38(5 Suppl 1):S25–33. doi: 10.1016/j.ajic.2010.04.196. [PubMed : 20569853 ] [CrossRef]
Best EL, Sandoe JA, Wilcox MH. Potential for aerosolization of Clostridium difficile after flushing toilets: The role of toilet lids in reducing environmental contamination risk. J Hosp Infect. 2012;80(1):1–5. doi: 10.1016/j.jhin.2011.08.010. [PubMed : 22137761 ] [CrossRef]
Roberts K, Smith CF, Snelling AM, et al. Aerial dissemination of Clostridium difficile spores. BMC Infect Dis. 2008;8:7. doi: 10.1186/1471-2334-8-7. [PMC free article : PMC2245947 ] [PubMed : 18218089 ] [CrossRef]
Riggs MM, Sethi AK, Zabarsky TF, et al. Asymptomatic carriers are a potential source for transmission of epidemic and nonepidemic Clostridium difficile strains among long-term care facility residents. Clin Infect Dis. 2007;45(8):992–8. doi: 10.1086/521854. [PubMed : 17879913 ] [CrossRef]
Dubberke ER, Reske KA, Noble-Wang J, et al. Prevalence of Clostridium difficile environmental contamination and strain variability in multiple health care facilities. Am J Infect Control. 2007;35(5):315–8. doi: 10.1016/j.ajic.2006.12.006. [PubMed : 17577478 ] [CrossRef]
Sethi AK, Al-Nassir WN, Nerandzic MM, et al. Persistence of skin contamination and environmental shedding of Clostridium difficile during and after treatment of C. difficile infection. Infect Control Hosp Epidemiol. 2010;31(1):21–7. doi: 10.1086/649016. [PubMed : 19929371 ] [CrossRef]
Huslage K, Rutala WA, Sickbert-Bennett E, et al. A quantitative approach to defining “high-touch” surfaces in hospitals. Infect Control Hosp Epidemiol. 2010;31(8):850–3. doi: 10.1086/655016. [PubMed : 20569115 ] [CrossRef]
Guerrero DM, Nerandzic MM, Jury LA, et al. Acquisition of spores on gloved hands after contact with the skin of patients with Clostridium difficile infection and with environmental surfaces in their rooms. Am J Infect Control. 2012;40(6):556–8. doi: 10.1016/j.ajic.2011.08.002. [PubMed : 21982209 ] [CrossRef]
Weaver L, Michels HT, Keevil CW. Survival of Clostridium difficile on copper and steel: Futuristic options for hospital hygiene. J Hosp Infect. 2008;68(2):145–51. doi: 10.1016/j.jhin.2007.11.011. [PubMed : 18207284 ] [CrossRef]
Shaughnessy MK, Micielli RL, DePestel DD, et al. Evaluation of hospital room assignment and acquisition of Clostridium difficile infection. Infect Control Hosp Epidemiol. 2011;32(3):201–6. doi: 10.1086/658669. [PubMed : 21460503 ] [CrossRef]
MacLeod-Glover N, Sadowski C. Efficacy of cleaning products for C difficile: Environmental strategies to reduce the spread of Clostridium difficile–associated diarrhea in geriatric rehabilitation. Can Fam Physician. 2010;56(5):417–23. pmid: 20463269. [PMC free article : PMC2868609 ] [PubMed : 20463269 ]
Wilcox MH, Fawley WN. Hospital disinfectants and spore formation by Clostridium difficile. Lancet. 2000;356(9238):1324. doi: 10.1016/S0140-6736(00)02819-1. [PubMed : 11073024 ] [CrossRef]
Environmental Protection Agency. List K: EPA’s registered antimicrobial products effective against Clostridium difficile spores. Environmental Protection Agency; 2018. https://www .epa.gov/pesticide-registration /list-k-epas-registered-antimicrobial-products-effective-against-clostridium.
Carling PC, Parry MM, Rupp ME, et al. Improving cleaning of the environment surrounding patients in 36 acute care hospitals. Infect Control Hosp Epidemiol. 2008;29(11):1035–41. doi: 10.1086/591940. [PubMed : 18851687 ] [CrossRef]
Rutala WA, Gergen MF, Weber DJ. Room decontamination with UV radiation. Infect Control Hosp Epidemiol. 2010;31(10):1025–9. doi: 10.1086/656244. [PubMed : 20804377 ] [CrossRef]
Steindl G, Fiedler A, Huhulescu S, et al. Effect of airborne hydrogen peroxide on spores of Clostridium difficile. Wien Klin Wochenschr. 2015;127(11–12):421–6. doi: 10.1007/s00508-014-0682-6. [PubMed : 25527140 ] [CrossRef]
Doll M, Morgan DJ, Anderson D, et al. Touchless technologies for decontamination in the hospital: A review of hydrogen peroxide and UV devices. Curr Infect Dis Rep. 2015;17(9):44. doi: 10.1007/s11908-015-0498-1. [PubMed : 26252970 ] [CrossRef]
Louh IK, Greendyke WG, Hermann EA, et al. Clostridium difficile infection in acute care hospitals: Systematic review and best practices for prevention. Infect Control Hosp Epidemiol. 2017;38(4):476–82. doi: 10.1017/ice.2016.324. [PMC free article : PMC5560033 ] [PubMed : 28300019 ] [CrossRef]
Haas JP, Menz J, Dusza S, et al. Implementation and impact of ultraviolet environmental disinfection in an acute care setting. Am J Infect Control. 2014;42(6):586–90. doi: 10.1016/j.ajic.2013.12.013. [PubMed : 24837107 ] [CrossRef]
Hacek DM, Ogle AM, Fisher A, et al. Significant impact of terminal room cleaning with bleach on reducing nosocomial Clostridium difficile. Am J Infect Control. 2010;38(5):350–3. doi: 10.1016/j.ajic.2009.11.003. [PubMed : 20123150 ] [CrossRef]
Orenstein R, Aronhalt KC, McManus JE, et al. A targeted strategy to wipe out Clostridium difficile. Infect Control Hosp Epidemiol. 2011;32(11):1137–9. doi: 10.1086/662586. [PubMed : 22011546 ] [CrossRef]
Levin J, Riley LS, Parrish C, et al. The effect of portable pulsed xenon ultraviolet light after terminal cleaning on hospital-associated Clostridium difficile infection in a community hospital. Am J Infect Control. 2013;41(8):746–8. doi: 10.1016/j.ajic.2013.02.010. [PubMed : 23685092 ] [CrossRef]
Manian FA, Griesnauer S, Bryant A. Implementation of hospital-wide enhanced terminal cleaning of targeted patient rooms and its impact on endemic Clostridium difficile infection rates. Am J Infect Control. 2013;41(6):537–41. doi: 10.1016/j.ajic.2012.06.014. [PubMed : 23219675 ] [CrossRef]
Khanafer N, Voirin N, Barbut F, et al. Hospital management of Clostridium difficile infection: A review of the literature. J Hosp Infect. 2015;90(2):91–101. doi: 10.1016/j.jhin.2015.02.015. [PubMed : 25913648 ] [CrossRef]
Boyce JM, Havill NL, Otter JA, et al. Impact of hydrogen peroxide vapor room decontamination on Clostridium difficile environmental contamination and transmission in a healthcare setting. Infect Control Hosp Epidemiol. 2008;29(8):723–9. doi: 10.1086/589906. [PubMed : 18636950 ] [CrossRef]
Anderson DJ, Chen LF, Weber DJ, et al. Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the benefits of enhanced terminal room disinfection study): A cluster-randomised, multicentre, crossover study. Lancet. 2017;389(10071):805–14. doi: 10.1016/S0140-6736(16)31588-4. [PMC free article : PMC5935446 ] [PubMed : 28104287 ] [CrossRef]
Best E, Parnell P, Thirkell G, et al. Effectiveness of deep cleaning followed by hydrogen peroxide decontamination during high Clostridium difficile infection incidence. J Hosp Infect. 2014;87(1):25–33. doi: 10.1016/j.jhin.2014.02.005. [PubMed : 24746230 ] [CrossRef]
Vianna PG, Dale Jr CR, Simmons S, et al. Impact of pulsed xenon ultraviolet light on hospital-acquired infection rates in a community hospital. Am J Infect Control. 2016;44(3):299–303. doi: 10.1016/j.ajic.2015.10.009. [PubMed : 26684369 ] [CrossRef]
Miller R, Simmons S, Dale C, et al. Utilization and impact of a pulsed-xenon ultraviolet room disinfection system and multidisciplinary care team on Clostridium difficile in a long-term acute care facility. Am J Infect Control. 2015;43(12):1350–3. doi: 10.1016/j.ajic.2015.07.029. [PubMed : 26362699 ] [CrossRef]
Nagaraja A, Visintainer P, Haas JP, et al. Clostridium difficile infections before and during use of ultraviolet disinfection. Am J Infect Control. 2015;43(9):940–5. doi: 10.1016/j.ajic.2015.05.003. [PubMed : 26159499 ] [CrossRef]
Hooker EA, Bochan M, Reiff TT, et al. Decreasing Clostridium difficile health care–associated infections through use of a launderable mattress cover. Am J Infect Control. 2015;43(12):1326–30. doi: 10.1016/j.ajic.2015.07.002. [PMC free article : PMC4679628 ] [PubMed : 26498703 ] [CrossRef]
Ghantoji SS, Stibich M, Stachowiak J, et al. Non-inferiority of pulsed xenon UV light versus bleach for reducing environmental Clostridium difficile contamination on high-touch surfaces in Clostridium difficile infection isolation rooms. J Med Microbiol. 2015;64(Pt 2):191. doi: 10.1099/jmm.0.000004-0. [PMC free article : PMC4811651 ] [PubMed : 25627208 ] [CrossRef]
Mosci D, Marmo G, Sciolino L, et al. Automatic environmental disinfection with hydrogen peroxide and silver ions versus manual environmental disinfection with sodium hypochlorite: A multicentre randomized before-and-after trial. J Hosp Infect. 2017;97(2):175–9. doi: 10.1016/j.jhin.2017.06.010. [PubMed : 28610932 ] [CrossRef]
Barbut F, Menuet D, Verachten M, et al. Comparison of the efficacy of a hydrogen peroxide dry-mist disinfection system and sodium hypochlorite solution for eradication of Clostridium difficile spores. Infect Control Hosp Epidemiol. 2009;30(6):507–14. doi: 10.1086/597232. [PubMed : 19379098 ] [CrossRef]
Aronhalt KC, McManus J, Orenstein R, et al. Patient and environmental service employee satisfaction of using germicidal bleach wipes for patient room cleaning. J Healthc Qual. 2013;35(6):30–6. doi: 10.1111/j.1945-1474.2011.00202.x. [PubMed : 22530592 ] [CrossRef]
Alfa MJ, Dueck C, Olson N, et al. UV-visible marker confirms that environmental persistence of Clostridium difficile spores in toilets of patients with c. Difficile-associated diarrhea is associated with lack of compliance with cleaning protocol. BMC Infect Dis. 2008;8(1):64. doi: 10.1186/1471-2334-8-64. [PMC free article : PMC2390558 ] [PubMed : 18474086 ] [CrossRef]
Doan L, Forrest H, Fakis A, et al. Clinical and cost effectiveness of eight disinfection methods for terminal disinfection of hospital isolation rooms contaminated with Clostridium difficile 027. J Hosp Infect. 2012;82(2):114–21. doi: 10.1016/j.jhin.2012.06.014. [PubMed : 22902081 ] [CrossRef]
Kundrapu S, Sunkesula V, Jury LA, et al. Daily disinfection of high-touch surfaces in isolation rooms to reduce contamination of healthcare workers’ hands. Infect Control Hosp Epidemiol. 2012;33(10):1039–42. doi: 10.1086/667730. [PubMed : 22961024 ] [CrossRef]
Denton A, Topping A, Humphreys P. Evolution of an audit and monitoring tool into an infection prevention and control process. J Hosp Infect. 2016;94(1):32–40. doi: 10.1016/j.jhin.2016.04.017. [PubMed : 27313140 ] [CrossRef]
Sitzlar B, Deshpande A, Fertelli D, et al. An environmental disinfection odyssey: Evaluation of sequential interventions to improve disinfection of Clostridium difficile isolation rooms. Infect Control Hosp Epidemiol. 2013;34(5):459–65. doi: 10.1086/670217. [PubMed : 23571361 ] [CrossRef]
Dumford III DM, Nerandzic MM, Eckstein BC, et al. What is on that keyboard? Detecting hidden environmental reservoirs of Clostridium difficile during an outbreak associated with North American pulsed-field gel electrophoresis type 1 strains. Am J Infect Control. 2009;37(1):15–9. doi: 10.1016/j.ajic.2008.07.009. [PubMed : 19171247 ] [CrossRef]
Ziakas PD, Zacharioudakis IM, Zervou FN, et al. Asymptomatic carriers of toxigenic C. difficile in long-term care facilities: A meta-analysis of prevalence and risk factors. PloS one. 2015;10(2):e0117195. doi: 10.1371/journal.pone.0117195. [PMC free article : PMC4338134 ] [PubMed : 25707002 ] [CrossRef]
Jump RL, Donskey CJ. Clostridium difficile in the long-term care facility: Prevention and management. Curr Geriatr Rep. 2015;4(1):60–9. doi: 10.1007/s13670-014-0108-3. [PMC free article : PMC4322371 ] [PubMed : 25685657 ] [CrossRef]
Steinberg JP, Denham ME, Zimring C, et al. The role of the hospital environment in the prevention of healthcare-associated infections by contact transmission. HERD. 2013;7(1_suppl):46–73. doi: 10.1177/193758671300701S06. [CrossRef]
Kotsanas D, Lovegrove A, Sloane TL, Gillespie EE. Disposable biocidal cubical curtains: Can they prevent the transfer of bacterial pathogens? Healthc Infect. 2012;17(3):87–90. doi: 10.1071/HI12015. [CrossRef]
Trajtman AN, Manickam K, Alfa MJ. Microfiber cloths reduce the transfer of Clostridium difficile spores to environmental surfaces compared with cotton cloths. Am J Infect Control. 2015;43(7):686–9. doi: 10.1016/j.ajic.2015.03.002. [PubMed : 25907782 ] [CrossRef]
Nerandzic MM, Cadnum JL, Eckart KE, et al. Evaluation of a hand-held far-ultraviolet radiation device for decontamination of Clostridium difficile and other healthcare-associated pathogens. BMC Infect Dis. 2012;12(1):120. doi: 10.1186/1471-2334-12-120. [PMC free article : PMC3419611 ] [PubMed : 22591268 ] [CrossRef]
Nerandzic MM, Donskey CJ. Triggering germination represents a novel strategy to enhance killing of Clostridium difficile spores. PloS One. 2010;5(8):e12285. doi: 10.1371/journal.pone.0012285. [PMC free article : PMC2924393 ] [PubMed : 20808871 ] [CrossRef]
This review includes a summary of evidence published from 2008 to 2018 on surveillance practices for CDI. After a brief practice description from CDC, IDSA/SHEA, and others, the review explains how regional and facility-level surveillance work as safety practices for preventing the transmission of C. difficile. Next, we provide a review of studies on CDI surveillance methods and explore surveillance contextual factors, such as setting and CDI testing method. Finally, we discuss research gaps and future directions for CDI surveillance. The review’s key findings are listed in the box below.
Research has shown that automated surveillance systems are generally accurate and save time and resources, compared with manual case review.
Automated laboratory alerts have been shown to help expedite contact precautions for CDI patients.Classifying CDI cases using standard case definitions is important although some researchers have found that the current definitions over represent the number of nosocomial cases.
There is a need for research that evaluates and compares different facility-level CDI surveillance strategies and implementation barriers and facilitators.
Genotyping provides detail about differences in C. difficile virulence and has helped to identify transmission pathways and outbreaks.
Promising technologies include rapid molecular typing, integrated systems that can track CDIs across health systems and facilities, and facility-access to regional real-time surveillance data.
Use available standardized case definitions for surveillance of (1) healthcare facility-onset (HO) CDI; (2) community-onset, healthcare facility–associated (CO-HCFA) CDI; and (3) community-associated (CA) CDI (good practice recommendation).
At a minimum, conduct surveillance for HO CDI in all inpatient healthcare facilities to detect elevated rates or outbreaks (weak recommendation, low quality of evidence).
Express the rate of HO CDI as the number of cases per 10,000 patient days. Express the CO-HCFA prevalence rate as the number of cases per 1,000 patient admissions (good practice recommendation).
In settings of high endemic rates or outbreaks, stratify data by patient location to target control measures when CDI incidence is above national or facility reduction goals, or if an outbreak is noted (weak recommendation, low quality of evidence).
Facility C. difficile surveillance practices include conducting internal surveillance data collection and analyses and reporting to State and Federal agencies via CDC’s NHSN. The NHSN assists facilities in collecting data to help determine local, regional, and national infection prevention priorities. The NHSN also helps facilities meet quality benchmarks, identify areas for improvement, and comply with CMS infection reporting requirements. To track national CDI incidence and establish reduction targets, the NHSN calculates standardized infection ratios. The standardized infection ratio is a risk-adjusted summary measure used to track HAIs at a national, statewide, or local level over time and by facility type. The NHSN also collects information on certain infection safety practices and antimicrobial resistance. 5
Another national surveillance program is the CDC Emerging Infections Program, a network of 10 State health departments, academic institutions, Federal agencies, and other public health stakeholders that collect data and support research and training to inform policy and public health practice. The national Healthcare Cost and Utilization Project is a database resource sponsored by AHRQ that has been used to track and report C. difficile hospitalizations. 6 C. difficile is also among the conditions tracked in the AHRQ National Scorecard on Hospital-Acquired Conditions. 7
At the State level, CDI reporting requirements vary; some States require facilities to report on C. difficile (via the NHSN) either by adopting CMS’s quality reporting requirements as State law, or through State mandates. 8 Many States implemented reporting requirements in 2013, the year in which hospitals were first required to report HAIs via NHSN for the CMS Hospital Inpatient Quality Reporting program. 9
Internal facility surveillance practices vary depending on facility resources and local requirements. Facilities may use the NHSN system to conduct internal CDI surveillance using the MDRO/CDI Module. 10 The LabID option, introduced in 2013, uses admission date, laboratory test results, and patient care location to automatically estimate measures of CDIs. An incident case is defined as any CDI LabID event from a specimen obtained more than 56 days after the most recent CDI LabID Event. A recurrent case is any CDI LabID event from a specimen obtained >14 days and ≤56 days after the most recent CDI LabID event for that patient. The day of the first specimen collection is considered day 1. HO-CDI cases are those LabID events collected more than 3 days after admission to an inpatient facility (i.e., on or after day 4). The Infection Surveillance Reporting option for CDI is based on clinical case reviews to identify and report CDIs. Facilities may report at the facility level or by different units within the facility.
Calculate CDI measures (e.g., prevalence at admission, CO prevalence, facility or unit incidence), Create charts, Filter data, Track incidence in different facility locations, Identify trends, Recognize deviations from the norm, and Compare rates with other facilities.The NHSN also collects data on antimicrobial use and resistance in a separate module. CDC’s Targeted Assessment for Prevention (TAP) provides infection prevention resources and guidance on how to interpret surveillance data and report feedback to stakeholders such as facility leaders and administrators. 11 Links to this and other resources are available later in this section of the CDI chapter.
Facility surveillance practices include using alerts for positive CDI cultures and tracking the movement of CDI patients within a facility or health system. 12 , 13 It is recommended that facilities have procedures for investigating outbreaks, protocols to guide referrals for strain typing, and processes to communicate with associated healthcare facilities and relevant jurisdictional bodies, as required. 14
The epidemiology of CDI has been evolving, with particular increases in CO CDI and hypervirulent strains. 15 Regional and national surveillance provide information on CDI epidemiology and help to identify clusters, outbreaks, and emerging ribotypes. Analyses of these data inform policy and public health programs. 16
At the facility level, CDI surveillance is used to identify transmission pathways and CDI clusters, evaluate safety improvement initiatives, and signal when facilities must enhance measures to prevent further transmission. 13 , 16 Monitoring HO-CDI incidence is a first step in identifying and controlling outbreaks at facilities. In one example, an outbreak on a vascular surgery unit was identified by an increase in the number of cases within 30 days and a change in the pattern of new cases. Samples were sent to a regional lab for PCR testing and results revealed that outbreak cases were caused by C. difficile ribotype 106, a clindamycin-resistant strain. Based on these findings, the facility implemented restrictions on the prescribing of clindamycin. Controlling the outbreak was attributed to this measure. 17 Root cause analysis of HO-CDI cases, another surveillance practice, helps facilities understand the reasons for hospital transmission and make workflow improvements, such as reducing testing delays. 18
In 2007, the CDC adopted standardized case definitions to track disease trends, detect outbreaks, facilitate comparison of CDI rates among similar institutions, and incorporate previous healthcare facility exposure information. 19 These definitions have been updated. For example, the 2007 case definition for healthcare facility onset was defined as a patient with CDAD symptom onset more than 48 hours after admission to a healthcare facility. Now, the definition for healthcare facility onset is defined as LabID events collected >3 days after admission to an inpatient facility. 4
CDI case identification and classification were traditionally conducted by individual case review; however, manual data abstraction is labor intensive, burdensome, and costly. 20 As technology evolves and reporting mandates increase, more facilities are using commercial infection control systems that process electronic health data to identify and classify cases. 12 , 20 Swift and automated identification of patients with C. difficile helps expedite contact precautions and reduce the potential for additional healthcare transmissions. 12 Research using genotyping technology (described below) supports rapid identification of CDI isolates and helps track transmission and identify virulent strains both within a facility and regionally. 21 Ribotyping (described below) during periods of increased CDI incidence can help identify CDI clusters and outbreaks. 22
Currently, there are a lack of studies that compare or evaluate facility-level CDI surveillance strategies.
The question of interest for this review is: What are the most recommended and promising institutional surveillance practices for C. difficile?
To answer this question, we searched the databases CINAHL and MEDLINE from 2008 to 2018 for “Clostridium difficile” and related MeSH terms and synonyms, as well as “Surveillance” OR “monitoring and surveillance” OR “epidemiologic surveillance” OR “infectious diseases surveillance” and synonyms. The search string also included a variety of healthcare settings, such as “hospitals,” “inpatient,” “long-term care,” “transitional care,” and “home health.” After duplicates were removed, the initial search yielded 503 results, all of which were screened for inclusion, and 42 full-text articles were retrieved.
Reference lists of included articles were also screened to ensure thoroughness and 14 additional studies were identified and retrieved. Articles were excluded if the intervention or outcomes were not relevant or precisely reported or if the study design was insufficient (e.g., opinion pieces, nonsystematic reviews). Studies in which surveillance was followed by other significant infection control practices (e.g., changes in environmental cleaning) were ruled out for this section and are considered in Section 4.6, Multicomponent CDI Prevention Interventions. Of the total retrieved articles, 16 studies and 2 systematic reviews were selected for inclusion in this review.
General methods for this report are described in the Methods section of the full report.
For this patient safety practice, a PRISMA flow diagram and evidence table, along with literature-search strategy and search-term details, are included in the report A through C appendixes.
We found 16 studies and 2 systematic reviews that examined facility C. difficile surveillance practices. These practices include the use of different statistical analyses, automated surveillance alerts, CDI case identification and classification, genotyping practices, and use of biomarkers to track CDI virulence. Most of these studies are descriptive case studies with no comparison group. Several studies examined the utility and accuracy of International Classification of Diseases (ICD) code data alone or in combination with medication data to conduct HO-CDI surveillance. Overall, there is a gap in the literature with regard to facility practices for implementing surveillance to reduce CDI.
One study from the United Kingdom demonstrates how surveillance can be used to identify CDI clusters and trigger implementation of enhanced infection prevention practices. In this study, Hardy et al. (2010) described the use of an HO-CDI case threshold to identify CDI clusters at a 1,800-bed teaching hospital. The case threshold was two or more HO-CDI cases within a 28-day period. Two or more HO-CDI cases was considered a period of increased incidence. The studied intervention was implemented upon identification of a period of increased CDI incidence. It included a standardized set of interventions, including notifying staff of the increased incidence and auditing compliance with hand hygiene, using environmental decontamination practices, isolating patients, and providing clinical management of patients with confirmed or suspected CDI.
If the audit identified any shortcomings in these prevention practices, steps were taken to make improvements. Additional enhanced cleaning was also implemented upon identification of the period of increased incidence (PII). If there were postaudit incident HO-CDI cases, a more detailed environmental audit was conducted by one of the head nurses. In the first 9 months of the study, isolates were ribotyped on PIIs with more than 10 cases; for the last 8 months of the study, isolates were ribotyped for all PIIs. In this case, an outbreak was defined as two or more cases of the same PCR ribotype within a 28-day period.
While less common in the United States outside of research contexts, ribotyping of C. difficile isolates helps determine transmission pathways and confirm presence of an outbreak. During roughly 1.5 years of the intervention, the number of PIIs investigated per month decreased from a peak of 14 per month in February 2008 to 1 in June 2009. For the first 9 months, five of seven periods with more than 10 cases were confirmed as outbreaks. In the final 8 months, ribotyping of the isolates confirmed nine (32%) of these periods to be outbreaks, with three being due to ribotype 027, two ribotype 078, and all the others distinct ribotypes. 22
Two of the included studies examined different statistical methods for CDI surveillance. 23 , 24 Lavan et al. (2012) compared the value and efficiency associated with manual tracking and calculating the incidence and prevalence of CDI in two wards in an acute 751-bed hospital in Ireland that were experiencing an increase in the number of severe CDI cases. For 6 weeks, the researchers measured the prevalence of CDI, antibiotic use, and associated comorbidity, and then for 13 weeks identified all new CDI cases, all using manual data collection. CDI cases were assessed for CDI risk factors, disease severity, response to treatment, and outcomes at 6 months.
The researchers found that manual data collection and analysis took less time in their prevalence study than the incidence study. The prevalence study provided useful information about differences between the two wards in CDI prevalence and CDIs with MRSA colonization, the extent of multiple antibiotic prescriptions in CDIs, and areas that required more indepth surveillance. The incidence study permitted a more detailed evaluation of CDI risk factors, origin and severity of disease, and patient outcomes. Overall, researchers found that incidence analysis was more useful for their institution for planning preventive initiatives and focusing antibiotic stewardship efforts. 23
Screening for outbreaks is often based on a relative increase in incidence or when incidence reaches an absolute threshold. 16 A temporal scan statistic approach examines new cases within a particular window of time and can be used prospectively or retrospectively. Faires et al. (2014) applied a retrospective scan statistic to identify several CDI clusters and potential outbreaks in a hospital based on 5 years of laboratory results and bacteriology reports. PCR was used to identify C. difficile isolates for the most recent year of data. CDI clusters were identified using the temporal scan statistic, and statistically significant clusters were compared with CDI outbreaks that had been identified using standard hospital surveillance. A negative binomial regression model identified associations between year, season, and month rate of CDI cases.
Results of the statistical analyses indicated that the incidence rate for CDI was significantly higher in the spring than in the fall and winter seasons. Overall, 86 CDI cases were identified, 18 specimens were analyzed, and 9 ribotypes were classified. The temporal scan statistic identified three significant clusters (p≤0.05), including potential outbreaks, not previously identified by hospital personnel using standard surveillance analyses. One outbreak was identified as starting a month before it had been recognized by the hospital. The researchers note that temporal analyses, applied prospectively and in tandem with other methods, could be useful in identifying clusters and outbreaks in a timely manner. 24
Over the last 10 years, CDI surveillance has become increasingly automated. 25 Automated and consistent measurement of CDI is preferable to disparate systems for surveillance of CDI. 21 Several studies in this review examined the feasibility and efficacy of electronic surveillance systems. Studies have found that the use of automated systems and EHR data assist in the rapid detection of cases and outbreaks, 12 , 13 , 26 and electronic strategies can provide timely alerts and help expedite contact precautions. Zilberberg et al. (2011) demonstrate that electronic patient data can be used to calculate risk-stratified HO-CDI rates to help inform practice. 27 Dubberke at al. (2012) and Benoit et al. (2011) found that automated surveillance using electronically available data (e.g., admission date) was accurate and more efficient than manual case review. 28 , 29
The results for CO, other HCFA were less sensitive (57%), were highly specific (99%), and had a kappa value of 0.65. In discussing the lower sensitivity for CO, other HCFA infections, they note the challenges of accurately capturing previous healthcare episodes using the available data. Several hundred discordant cases (out of 1,767 patients with a positive CDI test) required review and correction due to misclassifications in the data. Overall, the researchers reported that automated surveillance reduces staff time and may help facilities better track CO CDI. 28
While Dubberke et al. (2012) found that sensitivity and specificity for automated surveillance using EHR data was adequate, other researchers have found that, in practice, automated surveillance may overestimate the rate of HO CDI. 30 , 31 For example, Durkin et al. (2015) compared LabID reporting (for the NHSN) with traditional surveillance in 29 community hospitals in the southeastern United States. LabID is designed to use electronically captured laboratory data and hospital admission dates to determine HO versus CO surveillance CDI categories.
Diagnostic testing delay >3 days despite the presence of symptoms of CDI in the first 2 days of admission triggering an HO-CDI LabID categorization,
Misclassification of recurrent or continuation episodes as incident events by LabID, and Lack of an indeterminate category in LabID definitions.The differences based on surveillance method may affect hospital quality rankings. 31 Several hospitals in the study showed significantly lower rankings based on LabID surveillance (versus traditional surveillance). Once the coding was corrected, hospital rankings based on LabID HO rates were similar to rankings based on traditional surveillance.
In a recent study, Albert et al. (2018) examined the misclassification of HO CDIs reported to the NHSN by a large urban medical center. Using retrospective chart review of 212 HO-CDI cases, they found that only 62.2 percent of the cases reported to NHSN actually met the clinical definition of probable or possible HO CDI. The researchers estimate that the remaining cases may have been misclassified due to delays in testing, inappropriate testing, or use of stool softeners and laxatives. The researchers cite prior evidence that PCR testing is less able to distinguish between infection and colonization cases and that testing patients for CDI either too late or without clinically significant diarrhea contributes to overdiagnosis of HO CDI. 32 Truong et al. (2017) suggest real-time electronic tracking of diarrheal episodes and laxative therapy, to verify C. difficile testing criteria. 33
Quan et al. (2015) explored the accuracy and efficiency of a system for five MDROs and C. difficile tracking in a 410-bed tertiary care center that automated the following: monitoring microbiology results and initiating chart-based flags, ordering contact precautions on admission, and ensuring appropriate removal of precautions. The system was initiated as an alternative to manual case review, which required the assessment of laboratory results and tracking prior history of MDRO carriage and C. difficile infection. The system automatically reviewed daily positive laboratory results for 110,212 patient days and identified 1,543 results representing either new incident CDI cases or cases not previously known to the system, which triggered organism-specific flags. The automated ordering of precautions for inpatients occurred immediately after laboratory results were finalized, without a delay for manual order submission.
To test the accuracy of the system, the researchers conducted a point-prevalence assessment and found that all precautions were appropriate. The advantages of the automated system included preventing missed precautions and timelier weekend and after-hours isolation precautions. The researchers estimated that the automated alerts could save 850 annual hours of staff time. 12 Automated alerts have also been shown to expedite contact precautions and significantly increase the rate of appropriately isolated patients for other HAIs. 26
Automated surveillance of CDI can be conducted using clinical data (e.g., the LabID system) or administrative code data. 34 We found three studies and a systematic review that examined the accuracy of using ICD code data for the identification of CDI. 20 , 35 – 37 There are advantages to using ICD data since these codes are used by all facilities for insurance billing purposes and are stored in electronic formats. 20 , 35 One disadvantage is that the ICD coding rules may not match the standard surveillance definitions or account for testing sensitivity 35 or clinical context. 38 While useful for tracking overall CDI burden, some research shows that ICD-9 codes are not adequately accurate in identifying the place of onset (i.e., HO CDI vs. CO-HCFA infection).
Use of present-on-admission (POA) criteria, which CMS required to better distinguish CO versus HO-CDI cases began on October 1, 2008. In a review of overall cases of CDI, ICD coding may be useful, as evidenced in a recent national report using Healthcare Cost and Utilization Project data that focused on the burden of CDI for hospitals (using ICD-9 codes) and provided quarterly and annual estimates of CDI hospitalization rates from 2011 through 2015. 6 The POA indicator in ICD codes can be used to help distinguish which cases originated in the facility. This report shows how the POA-CDI rate is associated with the HO-CDI rate. However, the numbers do not account for CDI infections that resolved without an inpatient stay and cases that originated in a different health facility. Another challenge when working with these data is that coding practices may differ across hospitals and States. 6
To improve the accuracy of ICD data, Schmiedeskamp et al. (2009) examined the use of ICD-9 Clinical Modification code CDI data combined with medication treatment data, in an automated HO-CDI case identification system. The researchers examined a year of discharge data (23,920 adult patients) for over 300 hospitals. They identified adults discharged with an ICD-9-CM code for CDI and documentation of CDI therapy with oral vancomycin or metronidazole compared with ICD-9 code only. Case review was used to determine true cases. The sensitivity of the ICD-9-CM code alone for identifying nosocomial CDI was 96.8 percent, the specificity was 99.6 percent, the positive predictive value was 40.8 percent, and the negative predictive value was 100 percent. When CDI drug therapy was included with the ICD-9-CM code, the sensitivity ranged from 58.1 percent to 85.5 percent, specificity was virtually unchanged, and the range in positive predictive value was 37.9 percent to 80.0 percent, depending on the parameters of number of days of therapy and when therapy started. 36
One U.K. study explored how PCR ribotyping can be used to help identify local/facility outbreaks and virulent strains and inform infection prevention initiatives. 39 Wilcox et al. (2012) evaluated England’s Clostridium difficile Ribotyping Network and changes in CDI rates in the country. From 2007 to 2010, the network received samples from facilities for 10.8 percent of all CDI patients in the country (12,603 fecal specimens), along with demographic information, the name of the requesting hospital, and antibiotic history in the 30 days before the onset of CDI symptoms.
Hospitals were notified of the ribotyping results with a targeted turnaround time of less than 2 weeks. Ribotype 027, a ribotype associated with increased complications and mortality, was the most frequently detected in all 3 years but decreased over the 3 years. After 3 years, there was a 61 percent reduction in reported C. difficile in England. The researchers believe that the Clostridium difficile Ribotyping Network helped facilities get control of ribotype 027 by providing timely data on ribotypes, enabling targeted interventions for ribotype 027. 39
Compared with PCR ribotyping, whole genome sequencing offers greater detail about diversity within genotypes. Next-generation sequencing is a rapid form of whole genome sequencing. These technologies identify differences between isolates usually using single nucleotide variants. 16 , 40 With PCR ribotyping only, there is a greater likelihood of cases being flagged as sharing the same genotype, simply by chance. 16
Moloney et al. (2016) used next-generation sequencing to enhance epidemiological information and identify and resolve a C. difficile outbreak at an Irish hospital. Seven patients with CDI were all found to have ribotype 020 and C. difficile with a particular classification of bacterial isolates (sequence type 295). Using this information, the researchers were able to link the patients and track transmission back to a community hostel for homeless adults. Infection prevention and control measures were taken in the hostel under the guidance of public health personnel, and the outbreak was resolved. Of note, the standard surveillance definitions inaccurately classified three of the cases as HO CDI when in fact they were exposed in the hostel. For most patients in the study, the researchers suspected several weeks between ST-295 exposure and symptoms. 40
Monitoring patient biomarkers is a potential research strategy for early detection of increasing C. difficile strain virulence. Schlackow et al. (2012) used an automated monitoring system to examine routinely collected laboratory hospital data at a group of U.K. hospitals. In particular, they used iterative sequential regression and monitored biomarkers of inflammation and neutrophil counts upon CDI diagnosis, because these measures are taken frequently prior to therapy and are associated with mortality in C. difficile colitis.
Contextual factors include the type of setting in which C. difficile surveillance is conducted as participation in the NHSN expands beyond acute care facilities. In addition, the sensitivity and specificity of different testing methods impact surveillance rates. There is debate about the role of asymptomatic colonized C. difficile carriers—how they impact surveillance data and whether they should be actively surveilled.
In addition to acute care hospitals, current participants in NHSN C. difficile reporting include skilled nursing facilities, LTCFs, long-term acute care hospitals, inpatient rehabilitation facilities, and inpatient psychiatric units (NHSN, n.d.). Some argue that surveillance case definitions may overestimate LTCF-associated CDI. For example, current surveillance case classifications may overestimate the incidence of nursing home-associated CDI. Mylotte et al. (2012) found that of 75 incident CDI cases, 52 (69%) developed within 30 days of admission to an LTCF and 23 (31%) developed more than 30 days after admission.
Of the 52 cases that developed within 30 days, 68 percent were in residents admitted for subacute care. The mean number of days ± SD to develop CDI was 10.5 ± 2.5 in those who developed infection within 30 days, and 75 percent of these cases developed within 15 days of admission. 42 Jump and Donskey (2015) proposed surveillance definitions for LTCFs in which a case would not be considered as originating in the LTCF if a patient had been discharged from a hospital in the last 30 days; such a case would be considered LTCF onset, hospital acquired. 43
CDI testing methods have different sensitivities and specificities, which impact CDI rates. Therefore, the CDC adjusts for the different tests in NHSN reporting. A number of recent studies have shown that more sensitive molecular testing methods result in higher CDI surveillance rates. For example, Moehring et al. (2013) studied a change in testing from nonmolecular to molecular testing using PCR at 10 hospitals. The mean incidence rate of CO-HCFA CDI (using the 2007 case definitions) before the switch was 6.0 CDIs per 10,000 patient days compared with 9.6 CDIs per 10,000 patient days 18 months after the switch. The researchers stated that the improved sensitivity of molecular tests allows infected and colonized patients to be rapidly and reliably identified but can be “too good” at identifying patients who are colonized but not truly infected with C. difficile. 44 We explore the impact of testing type on CDI rates in more detail in Section 4.5, Testing (Indepth).
CDC Targeted Assessment for Prevention (TAP) Strategy: https://www.cdc.gov/hai/prevent/tap.html
CDC Updated Guidelines for Evaluating Public Health Surveillance Systems: https://www.cdc.gov/mmwr/preview/mmwrhtml/rr5013a1.htm
Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA): https://www.idsociety.org/globalassets/idsa/practice-guidelines/clinical-practice-guidelines-for-clostridium-difficile.pdf
How to: Surveillance of Clostridium difficile infections: https://www.ncbi.nlm.nih.gov/pubmed/29274463
NHSN Surveillance for C. difficile (CDI) and Multidrug Resistant Organisms (MRDO): https://www.cdc.gov/nhsn/ltc/cdiff-mrsa/index.html
While there are numerous case studies on CDI surveillance and how surveillance practices may overestimate HO CDI, there is limited research on CDI surveillance implementation, best practices, and challenges. In addition, while several studies pointed to the cost-effectiveness of automated surveillance systems 12 , 29 a more robust economic analysis of CDI surveillance programs could be beneficial. As with other PSPs, most of the CDI surveillance studies are in the context of hospitals, and other settings are poorly represented. The IDSA/SHEA 2017 C. difficile guidelines 4 identified additional gaps in understanding the epidemiology of C. difficile, including the need to better understand sources for C. difficile transmission in the community and the incubation period for C. difficile. Finally, some researchers have called for a standardized surveillance classification to define an “outbreak” of CDI. 16
The implementation and capabilities of automated surveillance will continue to grow 25 and global strategies may be implemented. In the future, integrated healthcare databases to track CDI patients across health systems could help track transmission outside a particular facility, ward, or healthcare system. 17 Increased research and tracking of CO CDI and CO-HCFA CDI will help to better understand CDI epidemiology outside of the healthcare setting. Although tracking CO-HCFA CDI is not mandated and requires the collection/evaluation of patients’ prior healthcare facility admissions, it is useful in order to better understand the epidemiology of CDI. 45
Strains of C. difficile have shown resistance to certain antimicrobials, and resistance plays a role in occurrence and recurrence of CDI. 46 According to Peng et al. (2017), with technological advances in the future, clinical microbiology laboratories could rapidly perform antimicrobial susceptibility testing to determine antimicrobial resistance and report the information to clinicians in real time. 46 Similarly, more rapid and affordable genotyping and molecular typing has the potential to identify cases that are part of an outbreak and improve response times. 4 , 16
Efforts in Europe have shown the potential for more standardized C. difficile PCR ribotyping. 47 After examining C. difficile ribotypes from six locations across the United States, Waslawski et al. (2013) called for greater C. difficile ribotype data in order to better understand the impact of ribotype on sensitivity and specificity of testing and clinical treatment for CDI. They also recommend the establishment of an internationally recognized C. difficile ribotype reference collection. 48
Participation in surveillance reporting will increase and include a broader spectrum of settings. For example, data from a larger group of LTCFs will be used to establish national benchmarks and track achievement of prevention goals. 49 A number of studies found discrepancies between surveillance definitions and clinical incidence. 40 , 50 , 51 Review and refining of surveillance definitions may be warranted as we continue to better understand CDI incubation periods. Finally, in the future, there is likely to be continued debate about “active surveillance” for C. difficile, i.e., the identification and isolation of asymptomatic carriers at hospital admission. 52 , 53 We explore this issue in more detail in Section 4.5, Testing (In-Depth).
Hall HI, Correa A, Yoon PW, et al. Lexicon, definitions, and conceptual framework for public health surveillance. MMWR Suppl. 2012;61(3):10–4. pmid: 22832991. [PubMed : 22832991 ]
Thacker S, Berkelman R. History of public health surveillance. In: W H, Baker E, Monson R, editors. Public health surveillance. New York, NY: Van Nostrand Reinhold; 1992. p. 1–15.
Association for Professionals in Infection Control and Epidemiology. Guide to preventing Clostridium difficile infections. Association for Professionals in Infection Control and Epidemiology; 2013. https://apic .org/Resource_ /EliminationGuideForm /59397fc6-3f90-43d1-9325-e8be75d86888 /File/2013CDiffFinal.pdf.
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Centers for Disease Control and Prevention. About NHSN. https://www .cdc.gov/nhsn /about-nhsn/index.html. Accessed December 16, 2019
Barrett M, Owens P. Clostridium difficile Hospitalizations, 2011–2015. Rockville, MD: Agency for Healthcare Research and Quality 2018. https://www .hcup-us.ahrq .gov/reports/HCUPCDiffHosp2011-2015Rpt081618.pdf.
Agency for Healthcare Research and Quality. AHRQ National Scorecard on Hospital-Acquired Conditions. Rockville, MD: Agency for Healthcare Research and Quality; 2019. https://www .ahrq.gov/hai/pfp/index.html.
Reagan J, Herzig CT, Pogorzelska-Maziarz M, et al. State law mandates for reporting of healthcare-associated Clostridium difficile infections in hospitals. Infect Control Hosp Epidemiol. 2015;36(3):350–2. doi: 10.1017/ice.2014.61. [PMC free article : PMC4514439 ] [PubMed : 25695178 ] [CrossRef]
Herzig CT, Reagan J, Pogorzelska-Maziarz M, et al. State-mandated reporting of health care-associated infections in the United States: Trends over time. Am J Med Qual. 2015;30(5):417–24. doi: 10.1177/1062860614540200. [PMC free article : PMC4272669 ] [PubMed : 24951104 ] [CrossRef]
Centers for Disease Control and Prevention. Tap Strategy ‘How To’ Guide for the Individual Facility User. Atlanta, GA: Centers for Disease Control and Prevention; 2018. https://www .cdc.gov/hai /pdfs/prevent/TAP-Guide-for-Individual-Facility-User.pdf.
Quan KA, Cousins SM, Porter DD, et al. Automated tracking and ordering of precautions for multidrug-resistant organisms. Am J Infect Control. 2015;43(6):577–80. doi: 10.1016/j.ajic.2014.12.019. [PubMed : 25681303 ] [CrossRef]
Hebden JN, Wright M-O, Fuss EP, et al. Leveraging surveillance technology to benefit the practice and profession of infection control. Am J Infect Control. 2008;36(3):S7–S11. doi: 10.1016/j.ajic.2007.10.011. [CrossRef]
Van Gessel H, McCann R, Peterson A, et al. Implementation Guide for Surveillance of Clostridium difficile Infection. Australian Commission on Safety and Quality in Health Care; 2011. https://www .safetyandquality .gov.au/sites /default/files/migrated /1303-CDI-Implementation_Guide-_V10 .pdf.
Argamany JR, Aitken SL, Lee GC, et al. Regional and seasonal variation in Clostridium difficile infections among hospitalized patients in the United States, 2001–2010. Am J Infect Control. 2015;43(5):435–40. doi: 10.1016/j.ajic.2014.11.018. [PubMed : 25952045 ] [CrossRef]
Eyre DW, Walker AS. Clostridium difficile surveillance: Harnessing new technologies to control transmission. Expert Rev Anti Infect Ther. 2013;11(11):1193–205. doi: 10.1586/14787210.2013.845987. [PubMed : 24151834 ] [CrossRef]
Ratnayake L, McEwen J, Henderson N, et al. Control of an outbreak of diarrhoea in a vascular surgery unit caused by a high-level clindamycin-resistant Clostridium difficile pcr ribotype 106. J Hosp Infect. 2011;79(3):242–7. doi: 10.1016/j.jhin.2011.06.013. [PubMed : 21864938 ] [CrossRef]
Saeed K, Petridou C, Gray H, et al. Clostridium difficile multidisciplinary team root cause analysis: Impact on clinical care and circumvention of financial penalties posed by clinical commissioning groups, but at what cost? J Infect Prev. 2018;19(2):54–61. doi: 10.1177/1757177417726155. [PMC free article : PMC5846979 ] [PubMed : 29552095 ] [CrossRef]
McDonald LC, Coignard B, Dubberke E, et al. Recommendations for surveillance of Clostridium difficile-associated disease. Infect Control Hosp Epidemiol. 2007;28(2):140–5. doi: 10.1086/511798. [PubMed : 17265394 ] [CrossRef]
Goto M, Ohl ME, Schweizer ML, et al. Accuracy of administrative code data for the surveillance of healthcare-associated infections: A systematic review and meta-analysis. Clin Infect Dis. 2014;58(5):688–96. doi: 10.1093/cid/cit737. [PubMed : 24218103 ] [CrossRef]
Krutova M, Kinross P, Barbut F, et al. How to: Surveillance of Clostridium difficile infections. Clin Microbiol Infect. 2018;24(5):469–75. doi: 10.1016/j.cmi.2017.12.008. [PubMed : 29274463 ] [CrossRef]
Hardy KJ, Gossain S, Thomlinson D, et al. Reducing Clostridium difficile through early identification of clusters and the use of a standardised set of interventions. J Hosp Infect. 2010;75(4):277–81. doi: 10.1016/j.jhin.2009.12.004. [PubMed : 20227140 ] [CrossRef]
Lavan AH, McCartan DP, Downes MM, et al. Monitoring Clostridium difficile infection in an acute hospital: Prevalence or incidence studies? IR J Med Sci. 2012;181(3):315–20. doi: 10.1007/s11845-011-0798-x. [PubMed : 22350440 ] [CrossRef]
Faires MC, Pearl DL, Ciccotelli WA, et al. Detection of Clostridium difficile infection clusters, using the temporal scan statistic, in a community hospital in southern Ontario, Canada, 2006–2011. BMC Infect Dis. 2014;14:254. doi: 10.1186/1471-2334-14-254. [PMC free article : PMC4030047 ] [PubMed : 24885351 ] [CrossRef]
Russo PL, Shaban RZ, Macbeth D, et al. Impact of electronic healthcare-associated infection surveillance software on infection prevention resources: A systematic review of the literature. J Hosp Infect. 2018;99(1):1–7. doi: 10.1016/j.jhin.2017.09.002. [PubMed : 28893614 ] [CrossRef]
Kho AN, Dexter PR, Warvel JS, et al. An effective computerized reminder for contact isolation of patients colonized or infected with resistant organisms. Int J Med Inform. 2008;77(3):194–8. doi: 10.1016/j.ijmedinf.2007.02.005. [PMC free article : PMC2974622 ] [PubMed : 17398145 ] [CrossRef]
Zilberberg MD, Tabak YP, Sievert DM, et al. Using electronic health information to risk-stratify rates of Clostridium difficile infection in U.S. hospitals. Infect Control Hosp Epidemiol. 2011;32(7):649–55. doi: 10.1086/660360. [PubMed : 21666394 ] [CrossRef]
Dubberke ER, Nyazee HA, Yokoe DS, et al. Implementing automated surveillance for tracking Clostridium difficile infection at multiple healthcare facilities. Infect Control Hosp Epidemiol. 2012;33(3):305–8. doi: 10.1086/664052. [PMC free article : PMC3649760 ] [PubMed : 22314071 ] [CrossRef]
Benoit SR, McDonald LC, English R, et al. Automated surveillance of Clostridium difficile infections using biosense. Infect Control Hosp Epidemiol. 2011;32(1):26–33. doi: 10.1086/657633. [PubMed : 21128815 ] [CrossRef]
Gase KA, Haley VB, Xiong K, et al. Comparison of 2 Clostridium difficile surveillance methods: National Healthcare Safety Network’s laboratory-identified event reporting module versus clinical infection surveillance. Infect Control Hosp Epidemiol. 2013;34(3):284–90. doi: 10.1086/669509. [PubMed : 23388363 ] [CrossRef]
Durkin MJ, Baker AW, Dicks KV, et al. A comparison between National Healthcare Safety Network laboratory-identified event reporting versus traditional surveillance for Clostridium difficile infection. Infect Control Hosp Epidemiol. 2015;36(2):125–31. doi: 10.1017/ice.2014.42. [PMC free article : PMC4405780 ] [PubMed : 25632994 ] [CrossRef]
Albert K, Ross B, Calfee DP, et al. Overreporting healthcare-associated C. difficile: A comparison of NHSN labID with clinical surveillance definitions in the era of molecular testing. Am J Infect Control. 2018;46(9):998–1002. doi: 10.1016/j.ajic.2018.03.001. [PubMed : 29655671 ] [CrossRef]
Truong CY, Gombar S, Wilson R, et al. Real-time electronic tracking of diarrheal episodes and laxative therapy enables verification of Clostridium difficile clinical testing criteria and reduction of Clostridium difficile infection rates. J Clin Microbiol. 2017;55(5):1276–84. doi: 10.1128/jcm.02319-16. [PMC free article : PMC5405247 ] [PubMed : 28250001 ] [CrossRef]
van Mourik MSM, Perencevich EN, Gastmeier P, et al. Designing surveillance of healthcare-associated infections in the era of automation and reporting mandates. Clin Infect Dis. 2018;66(6):970–6. doi: 10.1093/cid/cix835. [PubMed : 29514241 ] [CrossRef]
Dubberke ER, Butler AM, Yokoe DS, et al. Multicenter study of surveillance for hospital-onset Clostridium difficile infection by the use of ICD-9-CM diagnosis codes. Infect Control Hosp Epidemiol. 2010;31(3):262–8. doi: 10.1086/650447. [PMC free article : PMC3640806 ] [PubMed : 20100085 ] [CrossRef]
Schmiedeskamp M, Harpe S, Polk R, et al. Use of international classification of diseases, ninth revision, clinical modification codes and medication use data to identify nosocomial Clostridium difficile infection. Infect Control Hosp Epidemiol. 2009;30(11):1070–6. doi: 10.1086/606164. [PubMed : 19803724 ] [CrossRef]
Jones G, Taright N, Boelle PY, et al. Accuracy of ICD-10 codes for surveillance of Clostridium difficile infections, france. Emerg Infect Dis. 2012;18(6):979–81. doi: 10.3201/eid1806.111188. [PMC free article : PMC3358151 ] [PubMed : 22607707 ] [CrossRef]
Sips ME, Bonten MJM, van Mourik MSM. Automated surveillance of healthcare-associated infections: state of the art. Curr Opin Infect Dis. 2017;30(4):425–31. doi: 10.1097/qco.0000000000000376. [PubMed : 28505027 ] [CrossRef]
Wilcox MH, Shetty N, Fawley WN, et al. Changing epidemiology of Clostridium difficile infection following the introduction of a national ribotyping-based surveillance scheme in England. Clin Infect Dis. 2012;55(8):1056–63. doi: 10.1093/cid/cis614. [PubMed : 22784871 ] [CrossRef]
Moloney G, Mac Aogain M, Kelleghan M, et al. Possible interplay between hospital and community transmission of a novel Clostridium difficile sequence type 295 recognized by next-generation sequencing. Infect Control Hosp Epidemiol. 2016;37(6):680–4. doi: 10.1017/ice.2016.52. [PubMed : 27074865 ] [CrossRef]
Schlackow I, Walker AS, Dingle K, et al. Surveillance of infection severity: A registry study of laboratory diagnosed Clostridium difficile. PLoS Med. 2012;9(7):e1001279. doi: 10.1371/journal.pmed.1001279. [PMC free article : PMC3409138 ] [PubMed : 22859914 ] [CrossRef]
Mylotte JM, Russell S, Sackett B, et al. Surveillance for Clostridium difficile infection in nursing homes. J Am Geriatr Soc. 2013;61(1):122–5. doi: 10.1111/jgs.12041. [PubMed : 23253029 ] [CrossRef]
Jump RL, Donskey CJ. Clostridium difficile in the long-term care facility: Prevention and management. Curr Geriatr Rep. 2015;4(1):60–9. doi: 10.1007/s13670-014-0108-3. [PMC free article : PMC4322371 ] [PubMed : 25685657 ] [CrossRef]
Moehring RW, Lofgren ET, Anderson DJ. Impact of change to molecular testing for Clostridium difficile infection on healthcare facility-associated incidence rates. Infect Control Hosp Epidemiol. 2013;34(10):1055–61. doi: 10.1086/673144. [PMC free article : PMC3967243 ] [PubMed : 24018922 ] [CrossRef]
Dubberke ER, Butler AM, Hota B, et al. Multicenter study of the impact of community-onset Clostridium difficile infection on surveillance for c. Difficile infection. Infect Control Hosp Epidemiol. 2009;30(6):518–25. doi: 10.1086/597380. [PMC free article : PMC4156290 ] [PubMed : 19419269 ] [CrossRef]
Peng Z, Jin D, Kim HB, et al. Update on antimicrobial resistance in Clostridium difficile: Resistance mechanisms and antimicrobial susceptibility testing. J Clin Microbiol. 2017;55(7):1998–2008. doi: 10.1128/jcm.02250-16. [PMC free article : PMC5483901 ] [PubMed : 28404671 ] [CrossRef]
van Dorp SM, Kinross P, Gastmeier P, et al. Standardised surveillance of Clostridium difficile infection in European acute care hospitals: A pilot study, 2013. Euro Surveill. 2016;21(29). doi: 10.2807/1560-7917.Es.2016.21.29.30293. [PubMed : 27472820 ] [CrossRef]
Waslawski S, Lo ES, Ewing SA, et al. Clostridium difficile ribotype diversity at six health care institutions in the United States. J Clin Microbiol. 2013;51(6):1938–41. doi: 10.1128/jcm.00056-13. [PMC free article : PMC3716112 ] [PubMed : 23554188 ] [CrossRef]
Palms DL, Mungai E, Eure T, et al. The national healthcare safety network long-term care facility component early reporting experience: January 2013–December 2015. Am J Infect Control. 2018;46(6):637–42. doi: 10.1016/j.ajic.2018.01.003. [PubMed : 29478758 ] [CrossRef]
Kuntz JL, Polgreen PM. The importance of considering different healthcare settings when estimating the burden of Clostridium difficile. Clin Infect Dis. 2014;60(6):831–6. doi: 10.1093/cid/ciu955 [PubMed : 25477426 ] [CrossRef]
Han A, Jump RL. Discrepancies between surveillance definition and the clinical incidence of Clostridium difficile infection in a veterans affairs long-term care facility. Infect Control Hosp Epidemiol. 2014;35(11):1435–6. doi: 10.1086/678434. [PMC free article : PMC4306334 ] [PubMed : 25333446 ] [CrossRef]
Longtin Y, Paquet-Bolduc B, Gilca R, et al. Effect of detecting and isolating Clostridium difficile carriers at hospital admission on the incidence of C difficile infections: A quasi-experimental controlled study. JAMA Intern Med. 2016;176(6):796–804. doi: 10.1001/jamainternmed.2016.0177. [PubMed : 27111806 ] [CrossRef]
McDonald LC, Diekema DJ. Point-counterpoint: Active surveillance for carriers of toxigenic Clostridium difficile should be performed to guide prevention efforts. J Clin Microbiol. 2018;56(8). doi: 10.1128/jcm.00782-18. [PMC free article : PMC6062805 ] [PubMed : 29769275 ] [CrossRef]
Screening and isolating asymptomatic carriers can prevent CDI transmission but is resource intensive.
NAATs of unformed stool have relatively accurate sensitivity and specificity.Concerns with NAATs include that they detect toxigenic C. difficile genes, not the actual damaging toxins and may capture colonized patients in addition to those infected with C. difficile.
Certain multistep test algorithms (that include a test for C. difficile and for CDI toxins) perform as well as or better than NAATs but take longer.
Tools that identify patient risk for CDI could be useful in preventing CDI.Tools that identify a high risk of severe CDI or mortality show promise for prevention of severe CDI outcomes.
Future directions include improved diagnostic technology for increased efficiency and accuracy of diagnosis.
This section includes a summary of evidence published from 2008 to 2018 on diagnostic testing as a safety practice for CDI. After providing a brief practice description and testing recommendations by the IDSA/SHEA and others, we review how testing works as a safety practice for preventing CDI. In the evidence summary, we discuss testing criteria and whether to test asymptomatic patients, a summary of systematic reviews and meta-analyses on the accuracy of different testing methods, and studies on tools to predict CDI and CDI severity. Finally, we discuss gaps and future directions for CDI testing. Key findings are located in the box on the right.
Use patients with unexplained and new-onset ≥3 unformed stools in 24 hours as the preferred target population for testing for CDI (weak recommendation, very low quality of evidence).
Use a stool toxin test as part of a multistep algorithm (i.e., glutamate dehydrogenase [GDH] plus toxin; GDH plus toxin, arbitrated by NAAT; or NAAT plus toxin) rather than NAAT alone for all specimens received in the clinical laboratory when there are no preagreed institutional criteria for patient stool submission (weak recommendation, low quality of evidence).
Use NAAT alone or a multistep algorithm for testing (i.e., GDH plus toxin; GDH plus toxin, arbitrated by NAAT; or NAAT plus toxin) rather than a toxin test alone when there are preagreed institutional criteria for patient stool submission (weak recommendation, low quality of evidence).
Do not perform repeat testing (within 7 days) during the same episode of diarrhea and do not test stool from asymptomatic patients, except for epidemiological studies (strong recommendation, moderate quality of evidence).
Recent published guidelines and systematic reviews recommend only testing symptomatic patients for C. difficile, except for the purpose of epidemiological studies. 1 , 2 The recommendations are somewhat flexible with regard to the number of episodes of diarrhea that justify the need for CDI testing, noting that providers should take into account whether the patient has risk factors for CDI, most notable of which is antimicrobial use. 3 Before testing, physicians should attempt to rule out other causes of diarrhea. 4 Considerations with regard to repeat testing include the background prevalence of CDI at the facility. 1 , 4 SHEA/IDSA provide no recommendations for the use of biologic markers as an adjunct to diagnosis and do not recommend testing to determine if CDI has been cured. 1
The guidelines also recommended that, while laboratory diagnosis is pending, treatment should be initiated empirically for patients who present with fulminant CDI or if obtaining the test results takes more than 48 hours. If test results cannot be obtained on the same day, patients with suspected CDI should be placed on preemptive contact precautions pending the C. difficile test results. As treatment recommendations differ, it is important to know the severity of the infection and whether it is an initial or recurrent episode. 1
An abdominal CT scan may be used to differentiate between CDI and other causes of colitis and to determine the extent of the disease. However, to diagnose regular CDI (e.g., while test results are pending), when an abdominal CT has poor sensitivity, endoscopy can be used in certain urgent situations. The American College of Gastroenterology guidelines recommend endoscopy when a rapid diagnosis is needed or an initial negative toxin assay when CDI is strongly suspected, when there is an ileus and stool is not available, or when other colonic diseases are in the differential diagnosis. 5
Patients with C. difficile shed C. difficile spores, which contaminate the environment and may infect other patients. 6 , 7 Rapid identification of patients with CDI helps expedite contact precautions and isolation of these patients and prevent transmission to other patients. 8 The symptoms of CDI often match those of other causes of diarrhea 9 , 10 ; therefore, early and rapid diagnosis is important to start the appropriate treatment and improve patient outcomes. 11 Starting treatment and infection protocols sooner may ultimately reduce hospital length of stay, thereby reducing healthcare costs. 12 Rapid diagnosis also ensures that providers modify any existing therapies, such as discontinuing antimicrobial agents, which could worsen a patient’s condition. 13
While testing accuracy and speed have improved in the last 10 years, there is currently no consensus on the best testing method. 1 , 14 It is helpful for clinicians to understand the strengths and limitations of the testing methods when interpreting test results. The testing methods have varying sensitivities and specificities, due to each test’s detection ability and the tests’ different detection targets.
Each class of test targets one of the following: C. difficile toxin, genes that produce toxin, or identification of toxigenic C. difficile in the stool. Detection of genes that produce toxins and toxigenic C. difficile indicates a patient may be colonized or infected with C. difficile. Detection of C. difficile toxin indicates infection. Each of the targets can indicate different stages in the progression of the disease. 9 Some patients may remain colonized and acquire protection from disease while others progress to the disease. Some with symptoms may be treated and become asymptomatic carriers. 15
While the guidelines support accounting for C. difficile risk factors, Marra and Ng (2015) point out that the common risk factors for HA CDI are not as prevalent in CA CDI. 16 The criteria for whom to test for CDI such as the number and frequency of diarrheal stools that should trigger testing have decreased in the last few decades. 1 Whole genome sequencing and molecular typing indicate that most CDI is acquired from sources other than symptomatic cases. 17 , 18
Asymptomatic colonized patients do not shed as many C. difficile spores as CDI patients; however, they still contaminate the environment. 7 Evidence supports identifying asymptomatic colonized C. difficile patients for the purpose of isolation and contact precautions. 19 – 21 One study found that 29 percent of CDI cases were linked to transmission from colonized patients. 22
In the last decade, the most commonly used standalone test method has shifted from enzyme immunoassays to tests that detect DNA. Known as nucleic acid amplification testing, or NAAT, these tests generally have better detection abilities than enzyme immunoassays. 3 A shift to more rapid and accurate testing results in less use of unnecessary CDI-targeted antimicrobials 23 and a decrease in laboratory testing volume. 24
NAAT detects toxigenic C. difficile genes, not the damaging toxins, and may identify asymptomatic carriers as well as those with C. difficile disease; also, there is debate about whether the presence of toxigenic C. difficile alone is sufficient to diagnosis CDI. Guidelines therefore suggest that only symptomatic (i.e., those with diarrhea) patients should be tested. 25
To improve accuracy, combinations of tests are being used. Particularly if laboratories lack clinical input on specimen criteria and accept any unformed stool for testing, it may be most appropriate to use a combination of tests such as a test for organism combined with a relatively sensitive test for toxin in the stool. 3 These combinations test for the toxigenic organism and test for the actual toxin. Some guidelines do not promote the use of NAATs as a singular method even when patients are symptomatic. 4 , 9
We discuss the testing methods in more detail in the evidence summary. Some evidence from European studies shows that CDIs are being underdiagnosed due to lack of clinical suspicion or inaccurate testing. 26 , 27 It is likely that continued research will lead to improved testing methods and protocols.
The question of interest for this review is: What are the best testing methods and protocols for identifying and preventing CDI?
To answer this question, we searched the databases CINAHL and MEDLINE from 2008 to 2018 for ““Clostridioides difficile” and related MeSH terms and synonyms, as well as terms such as “diagnostic test,” “testing algorithms,” “rapid identification,” “stool sampling,” and “screening.” The search string also included a variety of healthcare settings, including “hospitals,” “inpatient,” “long-term care,” “transitional care,” and “home health.” The search yielded 732 results. After duplicates were removed, there were 710 papers, all of which were screened for inclusion. Articles were excluded if they were out of scope or were not primary studies, meta-analyses, or systematic reviews, leaving 78 full-text articles that were retrieved.
Reference lists of included articles were also screened to ensure thoroughness and seven additional studies were retrieved via this method. An additional systematic review was identified and retrieved when we researched background information on C. difficile testing. 28 Of the retrieved articles, 26 studies, 3 systematic reviews, and 4 meta-analyses were selected for inclusion in this review. Articles were excluded at each step if the outcomes were not relevant or precisely reported or if the study design was insufficient.
Due to the large number of search results for certain topics, we include a sample of studies rather than all results. Similarly, for the performance of individual test types, we chose to include a summary of published meta-analyses instead of reviewing individual studies.
General methods for this report are described in the Methods section of the full report.
For this patient safety practice, a PRISMA flow diagram and evidence table, along with literature-search strategy and search-term details, are included in the report appendixes A through C.
This review includes 26 studies, 3 systematic reviews, and 4 meta-analyses that address key issues in diagnostic testing for C. difficile. Four studies examined CDI testing criteria, including whether to systematically test hospitalized patients with diarrhea and whether to conduct repeat testing for CDI. Four studies and one review examined the question of whether to screen for and isolate asymptomatic C. difficile carriers.
We summarize the CDI testing methods and implications outlined by several reviews and studies. The performance of the tests is summarized by five recent meta-analyses. We also review five studies that evaluated tools for measuring patient risk of CDI and five studies that evaluated tools for measuring risk of CDI severity, including mortality.
While the guidelines promote testing of patients with three unformed stools in a 24-hour period, some researchers advocate for a more systematic process for C. difficile identification. Reigadas et al. (2015) tested all diarrheal stool for 6 months at a 1,550-bed hospital in Spain, regardless of clinician request. They found that 45 (18.1%) positive CDIs would have been excluded from testing because they did not meet the testing criterion (three unformed stools in 24 hours). Community-acquired cases and young age were risk factors for underdiagnoses.
Reigadas et al. (2015) recommend that all patients hospitalized with diarrhea be tested for CDI. 27 The European Society of Clinical Microbiology and Infectious Diseases suggests that all submitted unformed stool samples (whether they are submitted for testing for other conditions or for CDI) from patients 3 years or older should be tested for CDI. 4
Several studies evaluated the use of repeat testing. To better understand the factors that might contribute to a negative test followed by a positive test, Mostafa et al. (2018) examined 2 years of hospital laboratory test orders for C. difficile PCR, for which the test result, clinic-pathologic patient features, and previous test results were recorded. In a retrospective chart review, they found that 1,637 of 20,866 lab orders were repeat tests within the first 7 days of initial diagnosis. Out of 554 patients who first tested positive, 2.3 percent (13) of patients were retested as negative within 7 days. Of the patients who first tested negative (970), 4.5 percent (44) were positive on the repeat test. Prior C. difficile infection was the only factor significantly correlated with change from negative to positive C. difficile test result within 7 days. 29
The likelihood of a change in test result after a repeat test within 7 days appears to be somewhat linked to the test type and whether the initial test was positive or negative. Aichinger et al. (2008) conducted an observational study and examined the results of patients who had been retested within 7 days of the initial test result. There were 792 patients tested twice by enzyme immunoassay samples and 351 patients tested twice by PCR samples. The patients were all retested within 7 days of the initial diagnosis.
The authors found that retesting patients who were initially negative by enzyme immunoassay and PCR tests resulted in positive tests in 1.9 percent and 1.7 percent of cases, respectively. Patients with positive enzyme immunoassays and PCR results retested as negative in 4.8 percent and 2.9 percent of cases, respectively. 30 The findings about retesting negative results are consistent with the findings of others; it is generally noted that that negative CDI tests are very unlikely to change within 7 days. 1 Repeat testing on negative tests may be helpful in an endemic or outbreak setting. 4
Studies on MRSA found that active surveillance was not more effective than enhanced infection control policies,
Isolating asymptomatic CDI carriers requires additional hospital resources (e.g., single rooms), andOther interventions, such as hand hygiene, are effective at reducing multiple HAIs and are a better use of resources. 31
In addition, cohorting symptomatic patients with colonized but asymptomatic patients increases risk of infection of the latter. 32
Several published studies found public health benefits from screening asymptomatic carriers. One quasi-experimental study and three simulations found that detecting and isolating asymptomatic carriers was associated with prevention of future cases. 19 – 21 , 33 In the quasi-experimental study, Longtin et al. (2016) examined the impact of testing all patients admitted through the emergency room at a 354-bed Canadian acute care facility. Patients with a positive test were put into isolation (excluding patients who stayed less than 24 hours). Roughly 92.5 percent of eligible patients were screened over 17 months and 368 (4.8%) were identified as asymptomatic C. difficile carriers. During the intervention, 38 patients (3.0 per 10,000 patient days) developed an HA CDI compared with 416 patients (6.9 per 10,000 patient days) during the pre-intervention baseline period (p <0.001).19
In their simulation, Lanzas and Dubberke (2014) also found that testing asymptomatic carriers reduced the number of new colonizations and HO-CDI cases by 40 percent to 50 percent and 10 percent to 25 percent, respectively, compared with the baseline scenario. 20 In the simulations, factors that impacted the percentage of reduced cases include test sensitivity, test turnaround time (as it relates to delaying isolation), colonization prevalence at admission, strain, and effectiveness of patient isolation. 20 , 21
Screening and treating high-risk populations (regardless of CDI symptomology) is also explored in the literature. Saab et al. (2015), for example, conducted a simulation model with cirrhosis patients to compare costs and outcomes of two strategies for screening CDI. The first strategy consisted of screening all cirrhosis patients (regardless of symptoms) for CDI and treating if C. difficile was detected. In the second strategy, only patients with symptomatic CDI were treated.
The results showed that screening all cirrhosis patients for CDI was consistently associated with improved healthcare outcomes and decreased healthcare utilization across all variables in the one- and two-way sensitivity analyses. Using baseline assumptions, the authors found the costs associated with only screening symptomatic patients for CDI were 3.54 times greater than the costs to screen all cirrhosis patients. 33
Another approach, outlined by Furuya-Kanamori et al. (2015) in their review, suggests that patients at high likelihood of being asymptomatic carriers are not tested but medical staff should use enhanced infection control practices such as the use of gloves. In addition, units or facilities with high likelihood of asymptomatic carriers should carry out CDI cleaning protocols. 15
In this segment, we start by providing an overview of the distinctions between “reference standard” tests and tests most commonly used in clinical practices. We then summarize recent meta-analyses on commercial diagnostic testing methods. These meta-analyses are highlighted in Table 4.5.
The two most common reference standards for identifying C. difficile are toxigenic culture (TC) and cell cytotoxicity assay (CCTA). These are the “gold standards” against which commercial tests are compared. 3 , 9 , 10 Neither test is useful in a clinical setting as they take several days to complete and require specific expertise and equipment. 2 , 25
TC is intended to detect whether C. difficile is present and whether it can produce toxins. This test takes between 4 and 7 days. 16 Typically, toxigenic strains of C. difficile cause symptoms and the disease of C. difficile; however, the presence of toxigenic strains may not always result in active infection. 9 Therefore, a positive test result is not entirely indicative of a CDI.
The other common reference standard, the CCTA, measures the presence of free toxin in feces. The detection of free toxin with CCTA indicates that the patient has diarrhea caused by C. difficile. This test takes about 2 to 4 days for results and has a higher specificity than TC. 16 Planche et al. (2013) sought to validate the reference methods according to clinical outcomes using test results, length of hospital stay, and 30-day mortality. In a study of 12,420 fecal samples from four U.K. laboratories, the researchers found no increase in mortality when toxigenic C. difficile was present (as indicated by a positive TC test). CCTA was positivity correlated with clinical outcomes, making this a better reference method to define CDI and C. difficile-associated disease. 14
TC is useful for identifying patients who may be asymptomatic and capable of transmitting the organism to others. Culture for the organism of C. difficile (regardless of the potential for toxin production) was rarely mentioned in the reviewed studies and meta-analyses, except by Crobach et al. (2016) as a reference test for GDH immunoassays. 4
Many studies compare and measure the performance of individual tests. We report here on systematic reviews and meta-analyses to summarize the accuracy of different diagnostic testing methods. 4 , 16 , 28 , 34 We focus on the testing methods and not distinctions between the brands of tests available for each method. However, performance of tests does vary across manufacturers. 2 , 10 Table 4.5 outlines the detection targets and drawbacks of common reference and commercial C. difficile testing methods.
C. difficile Testing Methods.
TC and CCTA were standard diagnostic practice when C. difficile was first discovered, but now faster and less expensive tests are widespread. 16 , 35 The first alternatives to TC and CCTA to be used widely were toxin enzyme immunoassays. 9 Studies and meta-analyses group the immunoassays generally into those that test for toxins A and B and those that test for GDH. Crobach et al. (2016) further characterized the immunoassays into well-type and membrane-type; well-type tests are used for testing samples in batches, and membrane-type tests are used for testing solitary samples. 4
The enzyme immunoassays for C. difficile toxins A and B cost $5 to $15 per test 10 and take a few hours to complete. 16 It is most appropriate to compare toxins A and B tests against CCTA since these tests detect C. difficile toxins. 9 The immunoassays for toxins A and B were widely used as standalone tests until about 10 years ago. Because of very poor sensitivity, and moderately poor specificity, they are now primarily recommended as part of a two-step or three-step testing algorithm. 9 , 16 , 25 , 36
GDH is a common C. difficile enzyme antigen produced in large amounts by all strains of C. difficile, independent of toxigenicity. 2 Like TC, the GDH test indicates the presence of the organism in feces and does not indicate toxin production. Although the GDH immunoassay is sensitive, it is not as specific for CDI since both toxigenic and nontoxigenic organisms produce GDH. 16 The cost per test is $5 to $15 10 and test time is 15 to 45 minutes. 16 Because the GDH immunoassay does not detect toxin-producing C. difficile, it is not recommended as a standalone test and should be paired with a test that detects toxin. 25
After FDA approval in 2009, NAATs became available. 2 NAATs include rapid testing PCR and LAMP. NAATs test for the genes of C. difficile that produce toxins and identify the presence of toxigenic C. difficile. 25 NAATs are more expensive than the enzyme immunoassays for toxins A and B and GDH at about $30 to $50 a test. 10 NAAT testing is estimated to take about 1 to 2 hours. 9
Due to the limitations of these individual tests, combinations of tests can be used to improve specificity and positive predictive value of diagnosis. 16 While the SHEA/IDSA guidelines support the use of NAATs as a single step, Crobach et al. (2016) found that none of the individual commercial methods was satisfactory as a single test to diagnose CDI.
Several strategies can be used for multi-step testing. 4 One is to do two simultaneous rapid tests and then retest concordant results. Another strategy involves testing for GDH and toxins A and B, then further testing concordant positive results with PCR. 25 In their prospective study of 12,420 fecal samples, Planche et al. (2013) found that the optimal algorithm when TC was the reference was a combination of GDH and NAAT. For CCTA as the reference, the best algorithms were toxins A and B/NAAT and GDH/toxins A and B. 14
Table 4.6 presents a summary of sensitivities and specificities from six studies. Butler et al. (2016) reviewed and pooled results from 37 studies from 2011 to 2014. 28 For studies that used multiple reference standards, such as culture, TC, and cell cytotoxicity neutralization assay (CCNA), Crobach et al. (2016) conducted a meta-analysis of immunoassay tests, including those for toxins A and B and GDH, as well as NAATs. They found 56 studies that included sensitivity and specificity for toxins A and B, 31 studies with sensitivities and specificities for GDH tests, and 14 studies on NAATs. 4
O’Horo et al. (2012) reviewed 11 databases and found 25 PCR studies going back to the mid-1990s and 6 LAMP studies going back to 2005. Heterogeneity in the LAMP studies did not allow meta-analysis. 34 Wei et al. (2015) conducted a meta-analysis of nine LAMP studies published before February 2014 and concluded that LAMPs were suitable as standalone tests for CDI. 37
Bagdasarian et al. (2015) reviewed 13 studies on testing algorithms. In general, multistep algorithms using NAAT had good sensitivity (0.68–1.0) and specificity (0.92–1.0), but algorithms using only GDH or toxin enzyme immunoassay testing performed worse and had greater variability. 25 Four of the studies analyzed by Butler et al. (2016) involved multistep algorithms.
Meta-Analyses of CDI Diagnostic Tests.
Because PCRs are highly sensitive, they may detect asymptomatic colonized patients as well as symptomatic infected patients. 38 , 39 Koo et al., for example, found that universal PCR testing of all 101 adult hospitalized patients resulted in 18 positive tests, and of these, 72 percent were for patients with asymptomatic C. difficile colonization, which, from a treatment perspective is a false positive. 38 Therefore, many experts recommend only testing symptomatic patients with PCR. 1 , 25
Some researchers have pointed out that more sensitive testing methods result in an increase in reported HO CDI. Moehring et al. (2013) studied 10 hospitals (and 22 controls) that switched to PCR from immunoassays. The mean incidence rate of HCFA CDI before the switch was 6.0 CDIs per 10,000 patient days compared with 9.6 CDIs per 10,000 patient days a year and a half after the switch. After adjustment in the mixed-effects model, the overall IRR comparing CDI incidence after the switch to before the switch was 1.56 (95% CI, 1.28 to 1.90). 40 There is concern about lack of standardization in testing and higher HO-CDI reporting rates for those facilities using more sensitive methods. 41
Other researchers found decreased or stable CDI rates after switching from enzyme immunoassays to NAATs and a decrease in laboratory testing volume. Casari et al. (2018) found that more sensitive testing methods had beneficial results in terms of reductions in the number of samples tested and minor reductions in positive CDI tests at a 750-bed hospital. In 2011, the hospital tested 2,746 samples and the following year, after switching from toxin A and B immunoassay to NAAT with sampling criteria, 677 samples. The rate of healthcare-acquired CDI infections decreased from 3.74 per 1,000 admissions to 2.92 per 1,000 admissions a year after the switch in testing method. Other hospitals in the region saw steady CDI rates. 42
Napierala et al. (2013) found that 20 months after a switch from toxin A and B immunoassay to PCR for diagnosis of CDI at three hospitals, there was a significant decrease in laboratory testing volume (and decreased associated workload). Site-specific C. difficile testing volume decreased by 32.5 to 53.9 percent following implementation of PCR. C. difficile toxin detection rates were largely unchanged across the three hospitals. 24
Both GDH and C. difficile toxin A and B with concordant positives treated, concordant negatives not treated, and discordant results confirmed by PCR.
For the model, the researchers assessed cost and effectiveness from the hospital/healthcare perspective (e.g., laboratory testing, isolation protocol, treatment, prolonged hospitalization, and transmission of disease). For traditional algorithms, in which the test results were available after 4 hours, the assumption was that patients would be placed in isolation and initiated on CDI treatment while awaiting CDI test results. For the rapid testing algorithms, the assumption was no presumptive isolation or treatment.
A missed CDI case resulted in less than $5,000 of extended hospital stay costs and <2 transmissions, GDH diagnostic sensitivity was >93 percent, or The symptomatic carrier proportion among the TC-positive cases was >80 percent.The number of missed CDI cases was minimized by standalone PCR, whereas the number of false-positive diagnoses was minimized by GDH/PCR. 43
It is theorized that identifying patients at risk of CDI could help guide preemptive testing, infection prevention measures, and treatment. 44 , 45 As shown in Table 4.7, five studies developed or validated tools for predicting patients’ risk of developing CDI. 44 – 48 In one study, researchers measured patient outcomes associated with a screening tool that identified high-risk patients and implemented enhanced infection control policies for these patients. 45 The screening tool was informed by literature on CDI risk factors and a retrospective examination of 1 year of data on healthcare-acquired CDI at a 20-bed vascular-thoracic ICU.
Patients who met certain criteria (e.g., over 55 years old, prescribed a fluoroquinolone agent for any duration or prescribed any other antimicrobial agents for ≥5 days, history of immunosuppression) were identified as high risk for CDI. Measures were taken to reduce risk, such as a review of medication, hand hygiene audits and enhanced environmental cleaning measures for the patients’ rooms, and education for patients and families. During the first year, 1,066 patients were screened, and 157 patients were placed in the preventive model. During the pre-intervention phase, 10 cases of healthcare-acquired CDI occurred (overall incidence rate, 14.7) and during the 12-month study period, two cases of healthcare-acquired CDI were identified (incidence rate, 3.12) (p=0.025).
Other tools for predicting risk of CDI were validated retrospectively but were not implemented as a preventive measure. For example, Cooper et al. (2013) developed a tool that weighted certain EHR variables, such as admission from another facility, to provide a patient risk score. The variables were selected based on review of hospital data and previously published data on CDI risk factors. When a patient’s score met the tool criteria, the risk factors and score, along with the patient’s basic demographic data, appeared on a daily review report. The tool was validated over the course of a year and the final model resulted in an area under the curve (AUC) of 0.929 (95% CI, 0.926 to 0.932).
AUC is a measure of how well a tool can distinguish between two diagnostic groups. The AUC is calculated from a graph of the true positive rate (sensitivity) with the false positive rate for different cutoff points of the parameter. A perfect tool would result in an AUC of 1.0. The optimal cutoff score was 0.636, where both sensitivity and specificity were at 91.61 and 86.96, respectively. Of 4,927 patients identified as at risk for CDI, 254 (92.7% of total CDI cases in the study period) developed the disease. 44
Predictive Tools for CDI Incidence.
We found several other studies that validated tools to predict CDI severity or mortality; five of these studies are highlighted in Table 4.8. Van der Wilden et al. (2014), for example, studied and validated a risk scoring system to identify patients at risk for developing fulminant C. difficile colitis, which carries a high risk of mortality. Patients with fulminant colitis may have frequent bloody stools, abdominal pain, distension, and acute, severe toxic symptoms, including fever. It is possible that early surgical intervention may help improve outcomes for patients at risk of developing severe C. difficile colitis. 49
The researchers sought to develop a simplified scoring system based on four weighted factors: age >70, white blood cell count ≥20.000 or ≤2.000/µL, cardiorespiratory failure (the need for mechanical ventilation or vasopressor support), and diffuse abdominal tenderness. Over the course of 2 years, all patients with fulminant C. difficile colitis (746) were prospectively enrolled in the study; 48 (6.4%) of them progressed to fulminant C. difficile colitis. The risk scoring system (RSS) successfully distinguished patients with CDI from those who went on to have fulminant C. difficile colitis (AUC, 0.98). The researchers found that the system performed as well as a more complex system based on 12 variables and suggested that it could be useful as a bedside tool for clinicians to identify patients at risk of fulminant C. difficile colitis. 49
Predictive Tools for CDI Severity and Mortality.
APIC Implementation Guide: Guide to Preventing Clostridium difficile Infections: Includes section on C. difficile diagnosis: https://apic.org/wp-content/uploads/2019/07/2013CDiffFinal.pdf
CDC: FAQs for clinicians about C. difficile: Which laboratory tests are commonly used for diagnosis? https://www.cdc.gov/cdiff/clinicians/faq.html#anchor_1529601768432
SHEA/IDSA Clinical Practice Guidelines for C. difficile: 2017 Update: These guidelines provide updated recommendations regarding C. difficile epidemiology, diagnosis, treatment, infection prevention, and environmental management. Each recommendation includes a brief summary of the literature on the practice: https://www.idsociety.org/globalassets/idsa/practice-guidelines/clinical-practice-guidelines-for-clostridium-difficile.pdf
It may be beneficial for further exploration into the range of factors that impact the speed and accuracy of testing. For example, Kundrapu et al. (2013) found that delays included not providing stool collection supplies to patients in a timely fashion, rejecting specimens due to incorrect labeling or leaking from the container, and holding samples in the laboratory for batch processing. A corrective intervention consisted of easier-to-use containers, prioritization of CDI testing at the laboratory, on-demand specimen pickup and delivery (rather than at scheduled pickup times), and clinician education. The intervention was associated with reduced average time from CDI test order to result from 1.8 to 0.8 days. Additional studies that help inform systems processes would help expedite CDI testing. 50
Obtaining stool specimens may delay testing since it is not always possible to obtain specimens on demand, if a patient is not able to produce stool. Another study examined the use of rectal swabs (rectal swabs with liquid transport medium and nylon flocked dry swabs) for diagnosing CDI, with mixed results. The authors concluded that rectal swabs could not replace stool samples in the two-step laboratory diagnosis of CDI, as the sensitivities were too low, probably due to diluting effects of the fecal sample in the liquid medium. For simple PCR-based detection of C. difficile, however, dry swabs were a suitable alternative to stool samples. 51
Factors that lead to case misclassification will continue to be studied, especially given financial penalties for HO CDI. One study addressed concern about overreporting of HO-CDI rates and examined the role of laxatives. As diarrhea in the hospital can have many causes, including the use of laxatives, Truong et al. (2017) evaluated a system in which lab testing criteria combined the presence of diarrhea (≥3 unformed stools in 24 hours) and absence of laxative intake in the prior 48 hours. The researchers found that 7.1 percent (164) and 9.1 percent (211) of 2,321 C. difficile test orders were canceled due to absence of diarrhea and receipt of laxative therapy, respectively. HO-CDI incidence rate decreased from an average of 13.0 cases to 9.7 cases per 10,000 patient days (p=0.008). Oral vancomycin days of therapy decreased from an average of 13.8 days to 9.4 days per 1,000 patient days (p=0.009). 52
In the future, it is likely that the speed, accuracy, and convenience of CDI testing will continue to improve. One weakness of NAAT testing is that it does not detect C. difficile toxin. Some have proposed tests for toxin that are as accurate as CCTAs but fast and more practical for the clinical setting. 53 Other researchers examined lightweight, rapid, and portable CDI testing systems that could expedite and simplify the diagnostic process. 54 , 55
Yet another rapid CDI identification strategy explored in the literature is the use of dogs to scent-detect patients with C. difficile. Bomers et al. (2014) conducted a study in which a trained 5-year-old dog was presented with patients and asked to identify those with CDI. During a total of nine hospital visits, the dog performed 651 screenings involving 371 patients and correctly identified 12 of 14 CDI cases (sensitivity 86 percent [95% CI, 56% to 97%]) and 346 of 357 CDI-negative participants (specificity of 97% [95% CI, 94% to 98%]). Of the 11 CDI-negative participants that were “falsely” indicated by the dog as positive, 2 (18%) developed CDI during the 3 months of followup after the detection period, compared with only 12 of the 346 participants (3.5%) that the dog identified as C. difficile negative (p=0.06). 56 More research on this technique with larger samples would be useful.
Currently, genotyping is used for CDI surveillance and understanding transmission pathways, but the technology also has potential diagnostic value. Identifying a patient’s particular strain of CDI could help inform antimicrobial treatment decisions. 57 Whole genome sequencing has shown promise in identifying whether recurrent infection is due to relapse or reinfection with CDI. 58 Durovic et al. (2017) used genotyping to determine whether CDIs were due to recurrent infection or reinfection. Among 750 patients with CDI, 130 (17.3%) were diagnosed with recurrence or reinfection and strains were available from 106 patients. The period that showed the best indication of when an infection might actually be a reinfection was 20 weeks. None of the independent clinical characteristics was statistically sufficient to indicate whether infection was due to relapse or recurrence. 59
If C. difficile continues to be a common cause of infection and mortality, risk identification tools could be implemented for clinical use. In addition, understanding of differences in the symptomology of CA CDI may help improve diagnostic accuracy. Finally, the role of asymptomatic carriers as a source of CDI transmission will continue to be discussed and potentially addressed by actively screening for colonized carriers. More real-world research is needed to explore the potential of this practice.
McDonald LC, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–e48. doi: 10.1093/cid/cix1085. [PMC free article : PMC6018983 ] [PubMed : 29462280 ] [CrossRef]
Burnham CA, Carroll KC. Diagnosis of Clostridium difficile infection: An ongoing conundrum for clinicians and for clinical laboratories. Clin Microbiol Rev. 2013;26(3):604–30. doi: 10.1128/cmr.00016-13. [PMC free article : PMC3719497 ] [PubMed : 23824374 ] [CrossRef]
Gerding DN, File TM, Jr., McDonald LC. Diagnosis and treatment of Clostridium difficile infection (CDI). Infect Dis Clin Pract (Baltim Md). 2016;24(1):3–10. doi: 10.1097/IPC.0000000000000350. [PMC free article : PMC5769958 ] [PubMed : 29348706 ] [CrossRef]
Crobach MJ, Planche T, Eckert C, et al. European Society of Clinical Microbiology and Infectious Diseases: Update of the diagnostic guidance document for Clostridium difficile infection. Clin Microbiol Infect. 2016;22: Suppl 4:S63–81. doi: 10.1016/j.cmi.2016.03.010. [PubMed : 27460910 ] [CrossRef]
Oldfield EI, Oldfield EC, III, Johnson DA. Clinical update for the diagnosis and treatment of Clostridium difficile infection. World J Gastrointest Pharmacol Ther. 2014;5(1):1–26. doi: 10.4292/wjgpt.v5.i1.1. [PMC free article : PMC3951810 ] [PubMed : 24729930 ] [CrossRef]
Best EL, Fawley WN, Parnell P, et al. The potential for airborne dispersal of Clostridium difficile from symptomatic patients. Clin Infect Dis. 2010;50(11):1450–7. doi: 10.1086/652648. [PubMed : 20415567 ] [CrossRef]
Sethi AK, Al-Nassir WN, Nerandzic MM, et al. Persistence of skin contamination and environmental shedding of Clostridium difficile during and after treatment of D. difficile infection. Infect Control Hosp Epidemiol. 2010;31(1):21–7. doi: 10.1086/649016. [PubMed : 19929371 ] [CrossRef]
Donskey CJ. Preventing transmission of Clostridium difficile: Is the answer blowing in the wind? Clin Infect Dis. 2010;50(11):1458–61. doi: 10.1086/652649. [PubMed : 20415566 ] [CrossRef]
Planche T, Wilcox MH. Diagnostic pitfalls in Clostridium difficile infection. Infect Dis Clin North Am. 2015;29(1):63–82. doi: 10.1016/j.idc.2014.11.008. [PubMed : 25595842 ] [CrossRef]
Chen S, Gu H, Sun C, et al. Rapid detection of Clostridium difficile toxins and laboratory diagnosis of Clostridium difficile infections. Infection. 2017;45(3):255–62. doi: 10.1007/s15010-016-0940-9. [PubMed : 27601055 ] [CrossRef]
Barbut F, Surgers L, Eckert C, et al. Does a rapid diagnosis of Clostridium difficile infection impact on quality of patient management? Clin Microbiol Infect. 2014;20(2):136–44. doi: 10.1111/1469-0691.12221. [PubMed : 23565919 ] [CrossRef]
Brecher SM, Novak-Weekley SM, Nagy E. Laboratory diagnosis of Clostridium difficile infections: There is light at the end of the colon. Clin Infect Dis. 2013;57(8):1175–81. doi: 10.1093/cid/cit424. [PubMed : 23788237 ] [CrossRef]
Manek K, Williams V, Callery S, et al. Reducing the risk of severe complications among patients with Clostridium difficile infection. Can J Gastroenterol. 2011;25(7):368–72. doi: 10.1155/2011/153020. [PMC free article : PMC3174077 ] [PubMed : 21876858 ] [CrossRef]
Planche TD, Davies KA, Coen PG, et al. Differences in outcome according to Clostridium difficile testing method: A prospective multicentre diagnostic validation study of C difficile infection. Lancet Infect Dis. 2013;13(11):936–45. doi: 10.1016/s1473-3099(13)70200-7. [PMC free article : PMC3822406 ] [PubMed : 24007915 ] [CrossRef]
Furuya-Kanamori L, Marquess J, Yakob L, et al. Asymptomatic Clostridium difficile colonization: Epidemiology and clinical implications. BMC Infect Dis. 2015;15:516. doi: 10.1186/s12879-015-1258-4. [PMC free article : PMC4647607 ] [PubMed : 26573915 ] [CrossRef]
Marra F, Ng K. Controversies around epidemiology, diagnosis and treatment of Clostridium difficile infection. Drugs. 2015;75(10):1095–118. doi: 10.1007/s40265-015-0422-x. [PubMed : 26113167 ] [CrossRef]
Eyre DW, Cule ML, Wilson DJ, et al. Diverse sources of C. difficile infection identified on whole-genome sequencing. N Engl J Med. 2013;369(13):1195–205. doi: 10.1056/NEJMoa1216064. [PMC free article : PMC3868928 ] [PubMed : 24066741 ] [CrossRef]
Walker AS, Eyre DW, Wyllie DH, et al. Characterisation of Clostridium difficile hospital ward-based transmission using extensive epidemiological data and molecular typing. PLoS Med. 2012;9(2):e1001172. doi: 10.1371/journal.pmed.1001172. [PMC free article : PMC3274560 ] [PubMed : 22346738 ] [CrossRef]
Longtin Y, Paquet-Bolduc B, Gilca R, et al. Effect of detecting and isolating Clostridium difficile carriers at hospital admission on the incidence of C difficile infections: A quasi-experimental controlled study. JAMA Intern Med. 2016;176(6):796–804. doi: 10.1001/jamainternmed.2016.0177. [PubMed : 27111806 ] [CrossRef]
Lanzas C, Dubberke ER. Effectiveness of screening hospital admissions to detect asymptomatic carriers of Clostridium difficile: A modeling evaluation. Infect Control Hosp Epidemiol. 2014;35(8):1043–50. doi: 10.1086/677162. [PubMed : 25026622 ] [CrossRef]
Maghdoori S, Moghadas SM. Assessing the effect of patient screening and isolation on curtailing Clostridium difficile infection in hospital settings. BMC Infect Dis. 2017;17(1):384. doi: 10.1186/s12879-017-2494-6. [PMC free article : PMC5455129 ] [PubMed : 28577357 ] [CrossRef]
Curry SR, Muto CA, Schlackman JL, et al. Use of multilocus variable number of tandem repeats analysis genotyping to determine the role of asymptomatic carriers in Clostridium difficile transmission. Clin Infect Dis. 2013;57(8):1094–102. doi: 10.1093/cid/cit475. [PMC free article : PMC3783061 ] [PubMed : 23881150 ] [CrossRef]
Sydnor ER, Lenhart A, Trollinger B, et al. Antimicrobial prescribing practices in response to different Clostridium difficile diagnostic methodologies. Infect Control Hosp Epidemiol. 2011;32(11):1133–6. doi: 10.1086/662381. [PubMed : 22011545 ] [CrossRef]
Napierala M, Munson E, Skonieczny P, et al. Impact of toxigenic Clostridium difficile polymerase chain reaction testing on the clinical microbiology laboratory and inpatient epidemiology. Diagn Microbiol Infect Dis. 2013;76(4):534–8. doi: 10.1016/j.diagmicrobio.2013.04.020. [PubMed : 23731555 ] [CrossRef]
Bagdasarian N, Rao K, Malani PN. Diagnosis and treatment of Clostridium difficile in adults: A systematic review. Jama. 2015;313(4):398–408. doi: 10.1001/jama.2014.17103. [PMC free article : PMC6561347 ] [PubMed : 25626036 ] [CrossRef]
Alcala L, Reigadas E, Marin M, et al. Impact of clinical awareness and diagnostic tests on the underdiagnosis of Clostridium difficile infection. Eur J Clin Microbiol Infect Dis. 2015;34(8):1515–25. doi: 10.1007/s10096-015-2380-3. [PubMed : 25904126 ] [CrossRef]
Reigadas E, Alcala L, Marin M, et al. Missed diagnosis of Clostridium difficile infection; a prospective evaluation of unselected stool samples. J Infect. 2015;70(3):264–72. doi: 10.1016/j.jinf.2014.10.013. [PubMed : 25452039 ] [CrossRef]
Butler M, Olson A, Drekonja D, et al. Early Diagnosis, Prevention, and Treatment of Clostridium difficile: Update. Comparative Effectiveness Review No. 172. (Prepared by the Minnesota Evidence-based Practice Center under Contract No. 290-2012-00016-I.) AHRQ Publication No. 16-EHC012-EF. Rockville, MD: Agency for Healthcare Research and Quality; 2016. [PubMed : 27148613 ]
Mostafa ME, Flynn T, Hartley CP, et al. Effective utilization of C. difficile PCR and identification of clinicopathologic factors associated with conversion to a positive result in symptomatic patients. Diagn Microbiol Infect Dis. 2018;90(4):307–10. doi: 10.1016/j.diagmicrobio.2017.11.021. [PubMed : 29290473 ] [CrossRef]
Aichinger E, Schleck CD, Harmsen WS, et al. Nonutility of repeat laboratory testing for detection of Clostridium difficile by use of PCR or enzyme immunoassay. J Clin Microbiol. 2008;46(11):3795–7. doi: 10.1128/jcm.00684-08. [PMC free article : PMC2576580 ] [PubMed : 18784320 ] [CrossRef]
McDonald LC, Diekema DJ. Point-counterpoint: Active surveillance for carriers of toxigenic Clostridium difficile should be performed to guide prevention efforts. J Clin Microbiol. 2018;56(8). doi: 10.1128/jcm.00782-18. [PMC free article : PMC6062805 ] [PubMed : 29769275 ] [CrossRef]
Islam J, Cheek E, Navani V, et al. Influence of cohorting patients with Clostridium difficile infection on risk of symptomatic recurrence. J Hosp Infect. 2013;85(1):17–21. doi: 10.1016/j.jhin.2013.06.009. [PubMed : 23910403 ] [CrossRef]
Saab S, Alper T, Sernas E, et al. Hospitalized patients with cirrhosis should be screened for Clostridium difficile colitis. Dig Dis Sci. 2015;60(10):3124–9. doi: 10.1007/s10620-015-3707-8. [PubMed : 25986524 ] [CrossRef]
O’Horo JC, Jones A, Sternke M, et al. Molecular techniques for diagnosis of Clostridium difficile infection: Systematic review and meta-analysis. Mayo Clin Proc. 2012;87(7):643–51. doi: 10.1016/j.mayocp.2012.02.024. [PMC free article : PMC3538482 ] [PubMed : 22766084 ] [CrossRef]
Crobach MJ, Dekkers OM, Wilcox MH, et al. European Society of Clinical Microbiology and Infectious Diseases (ESCMID): Data review and recommendations for diagnosing Clostridium difficile-infection (CDI). Clin Microbiol Infect. 2009;15(12):1053–66. doi: 10.1111/j.1469-0691.2009.03098.x. [PubMed : 19929972 ] [CrossRef]
Wilcox MH, Shetty N, Fawley WN, et al. Changing epidemiology of Clostridium difficile infection following the introduction of a national ribotyping-based surveillance scheme in England. Clin Infect Dis. 2012;55(8):1056–63. doi: 10.1093/cid/cis614. [PubMed : 22784871 ] [CrossRef]
Wei C, Wen-En L, Yang-Ming L, et al. Diagnostic accuracy of loop-mediated isothermal amplification in detection of Clostridium difficile in stool samples: A meta-analysis. Arch Med Sci. 2015;11(5):927–36. doi: 10.5114/aoms.2015.54846. [PMC free article : PMC4624739 ] [PubMed : 26528332 ] [CrossRef]
Koo HL, Van JN, Zhao M, et al. Real-time polymerase chain reaction detection of asymptomatic Clostridium difficile colonization and rising C. difficile-associated disease rates. Infect Control Hosp Epidemiol. 2014;35(6):667–73. doi: 10.1086/676433. [PubMed : 24799643 ] [CrossRef]
Silva ALO, Marra AR, Martino MDV, et al. Identification of Clostridium difficile asymptomatic carriers in a tertiary care hospital. Biomed Res Int. 2017;2017:5450829. doi: 10.1155/2017/5450829. [PMC free article : PMC5643058 ] [PubMed : 29098156 ] [CrossRef]
Moehring RW, Lofgren ET, Anderson DJ. Impact of change to molecular testing for Clostridium difficile infection on healthcare facility-associated incidence rates. Infect Control Hosp Epidemiol. 2013;34(10):1055–61. doi: 10.1086/673144. [PMC free article : PMC3967243 ] [PubMed : 24018922 ] [CrossRef]
Bogaty C, Levesque S, Garenc C, et al. Trends in the use of laboratory tests for the diagnosis of Clostridium difficile infection and association with incidence rates in Quebec, Canada, 2010–2014. Am J Infect Control. 2017;45(9):964–8. doi: 10.1016/j.ajic.2017.04.002. [PubMed : 28549882 ] [CrossRef]
Casari E, De Luca C, Calabro M, et al. Reducing rates of Clostridium difficile infection by switching to a stand-alone NAAT with clear sampling criteria. Antimicrob Resist Infect Control. 2018;7:40. doi: 10.1186/s13756-018-0332-2. [PMC free article : PMC5848539 ] [PubMed : 29564088 ] [CrossRef]
Schroeder LF, Robilotti E, Peterson LR, et al. Economic evaluation of laboratory testing strategies for hospital-associated Clostridium difficile infection. J Clin Microbiol. 2014;52(2):489–96. doi: 10.1128/JCM.02777-13. [PMC free article : PMC3911327 ] [PubMed : 24478478 ] [CrossRef]
Cooper PB, Heuer AJ, Warren CA. Electronic screening of patients for predisposition to Clostridium difficile infection in a community hospital. Am J Infect Control. 2013;41(3):232–5. doi: 10.1016/j.ajic.2012.03.022. [PubMed : 22990298 ] [CrossRef]
Cruz-Betancourt A, Cooper CD, Sposato K, et al. Effects of a predictive preventive model for prevention of Clostridium difficile infection in patients in intensive care units. Am J Infect Control. 2016;44(4):421–4. doi: 10.1016/j.ajic.2015.11.010. [PubMed : 26775936 ] [CrossRef]
Stites SD, Cooblall CA, Aronovitz J, et al. The tipping point: Patients predisposed to Clostridium difficile infection and a hospital antimicrobial stewardship programme. J Hosp Infect. 2016;94(3):242–8. doi: 10.1016/j.jhin.2016.07.022. [PubMed : 27742079 ] [CrossRef]
Tabak YP, Johannes RS, Sun X, et al. Predicting the risk for hospital-onset Clostridium difficile infection (HO-CDI) at the time of inpatient admission: HO-CDI risk score. Infect Control Hosp Epidemiol. 2015;36(6):695–701. doi: 10.1017/ice.2015.37. [PMC free article : PMC5768429 ] [PubMed : 25753106 ] [CrossRef]
Kuntz JL, Polgreen PM. The importance of considering different healthcare settings when estimating the burden of Clostridium difficile. Clin Infect Dis. 2014;60(6):831–6. doi: 10.1093/cid/ciu955 [PubMed : 25477426 ] [CrossRef]
van der Wilden GM, Chang Y, Cropano C, et al. Fulminant Clostridium difficile colitis: Prospective development of a risk scoring system. J Trauma Acute Care Surg. 2014;76(2):424–30. doi: 10.1097/ta.0000000000000105. [PubMed : 24458048 ] [CrossRef]
Kundrapu S, Jury LA, Sitzlar B, et al. Easily modified factors contribute to delays in diagnosis of Clostridium difficile infection: A cohort study and intervention. J Clin Microbiol. 2013;51(7):2365–70. doi: 10.1128/jcm.03142-12. [PMC free article : PMC3697705 ] [PubMed : 23678072 ] [CrossRef]
Jazmati N, Kirpal E, Piepenbrock E, et al. Evaluation of the use of rectal swabs for laboratory diagnosis of Clostridium difficile infection. J Clin Microbiol. 2018;56(7):e00426–18. doi: 10.1128/JCM.00426-18. [PMC free article : PMC6018322 ] [PubMed : 29720433 ] [CrossRef]
Truong CY, Gombar S, Wilson R, et al. Real-time electronic tracking of diarrheal episodes and laxative therapy enables verification of Clostridium difficile clinical testing criteria and reduction of Clostridium difficile infection rates. J Clin Microbiol. 2017;55(5):1276–84. doi: 10.1128/jcm.02319-16. [PMC free article : PMC5405247 ] [PubMed : 28250001 ] [CrossRef]
Pollock NR. Ultrasensitive detection and quantification of toxins for optimized diagnosis of Clostridium difficile infection. J Clin Microbiol. 2016;54(2):259–64. doi: 10.1128/jcm.02419-15. [PMC free article : PMC4733189 ] [PubMed : 26659205 ] [CrossRef]
Hirvonen JJ, Matero P, Siebert C, et al. Novel portable platform for molecular detection of toxigenic Clostridium difficile in faeces: A diagnostic accuracy study. Eur J Clin Microbiol Infect Dis. 2017;36(5):783–9. doi: 10.1007/s10096-016-2860-0. [PubMed : 27988815 ] [CrossRef]
Liu C, Jiang DN, Xiang GM, et al. DNA detection of Clostridium difficile infection based on real-time resistance measurement. Genet Mol Res. 2013;12(3):3296–304. doi: 10.4238/2013.September.3.6. [PubMed : 24065671 ] [CrossRef]
Bomers MK, van Agtmael MA, Luik H, et al. A detection dog to identify patients with Clostridium difficile infection during a hospital outbreak. J Infect. 2014;69(5):456–61. doi: 10.1016/j.jinf.2014.05.017. [PubMed : 24973552 ] [CrossRef]
Kociolek LK, Gerding DN. Clinical utility of laboratory detection of Clostridium difficile strain bi/nap1/027. J Clin Microbiol. 2016;54(1):19–24. doi: 10.1128/jcm.02340-15. [PMC free article : PMC4702757 ] [PubMed : 26511742 ] [CrossRef]
Mac Aogain M, Moloney G, Kilkenny S, et al. Whole-genome sequencing improves discrimination of relapse from reinfection and identifies transmission events among patients with recurrent Clostridium difficile infections. J Hosp Infect. 2015;90(2):108–16. doi: 10.1016/j.jhin.2015.01.021. [PubMed : 25935700 ] [CrossRef]
Durovic A, Widmer AF, Frei R, et al. Distinguishing Clostridium difficile recurrence from reinfection: Independent validation of current recommendations. Infect Control Hosp Epidemiol. 2017;38(8):891–6. doi: 10.1017/ice.2017.119. [PubMed : 28592348 ] [CrossRef]
Archbald-Pannone LR, McMurry TL, Guerrant RL, et al. Delirium and other factors with Clostridium difficile infectiion that predeic mortality in hospitalized patients. Am J Infect Control 2015;43(7):690–693. doi:10.1016/j.ajic.2015.03.017 [PMC free article : PMC4490980 ] [PubMed : 25920706 ] [CrossRef]
Kassam Z, Fabersunne CC, Smith MB, et al. Clostridium difficile Associated Risk of Death Score (CARDS): A novel score to predict mortality among hospitalized patients with Clostridium difficile infection. Aliment Pharmacol Ther. 2016;43(6):725–733. doi: 10.1111/apt.13546 [PMC free article : PMC5094350 ] [PubMed : 26849527 ] [CrossRef]
van Beurden YH, Bomers MK, van der Werff SD, et al. Cost analysis of an outbreak of Clostridium difficile infection ribotype 027 in a Dutch tertiary care centre. J Hosp Infect. 2017;95(4):421–425. doi: 10.1016/j.jhin.2016.12.019. [PubMed : 28169013 ] [CrossRef]
Figh ML, Zoog ESL, Moore R, et al. External validation of Velazquez-Gomez Severity Score Index and ATLAS scores and the identification of risk factors associated with morality in Clostridium difficile infections. Am Surg. 2017;83(12):1347–1351. [PubMed : 29336752 ]
The most common component was environmental cleaning, followed by hand hygiene and patient isolation practices; antimicrobial stewardship and contact precautions; and CDI testing and surveillance.
No single CDI prevention resource was used across studies. Information was limited on staff compliance and financial costs of interventions.Collaborations and teamwork were reported to be facilitators of implementation of multicomponent interventions.
Additional facilitators of staff compliance included adequate supplies (e.g., gowns, soap), communication, signage, and institutional support. Barriers included time it takes to perform prevention practices (e.g., wash hands, put on gowns), inadequate staff education, inconsistency in testing criteria and unclear roles for ordering CDI tests, visitors not practicing contact precautions, and lack of isolation rooms.
Real-world studies on the implications of different practice combinations, as well as studies on regional prevention efforts and nonhospital settings, will help improve understanding.
Our search for articles on individual CDI PSPs published from 2008 to 2018 uncovered studies that looked at patient outcomes associated with the combination of two or more CDI PSPs. To accurately reflect the number of articles on multicomponent CDI prevention interventions, we decided to include a review of these studies.
In this addendum to the CDI patient safety chapter, we provide a practice description and evidence summary of the research published from 2008 to 2018 on multicomponent CDI prevention interventions. We then discuss qualitative research on implementation barriers and facilitators, as well as gaps and future directions.
Most of the included articles were identified in the searches for the five other PSPs (hand hygiene, antimicrobial stewardship, environmental cleaning and decontamination, surveillance, and testing) or from reference lists of articles identified in these searches. To ensure thoroughness, we conducted a brief additional search for multicomponent interventions and identified three additional sources. 1 – 3
For all searches, we excluded articles without clearly stated methodology or a methods section. We also excluded studies that did not quantify or clearly report CDI outcomes, did not clearly explain the interventions, did not describe baseline prevention practices, or did not measure statistical associations with more than one of the interventions. The remaining eight studies and three reviews addressed multicomponent prevention interventions and CDI patient outcomes. Key findings are located in the box above.
Barker et al. (2017) describe a CDI bundle as any set of multiple (>1) interventions focused on reducing CDI in the inpatient setting. To guide their decision about which set of practices to implement, the researchers whose studies we reviewed cited different influences. Several cited prior research and recent IDSA/SHEA recommendations as guiding the decision. 4 , 5 In one study, a team of experts (assembled by the facility) reviewed facility epidemiological data and determined which practices to implement. 6 Two articles stated that the practices in their respective facilities were guided by government mandates or recommendations. 7 , 8
Some studies and resources recommend that facilities assess their current practices to identify gaps and targets for improvement. Facilities should use multidisciplinary teams to oversee cross-cutting efforts and set achievable goals. 9 , 10 There are different contextual recommendations within the 2017 IDSA/SHEA guidelines. 11 Several of the guidelines are framed as minimum recommendations and some are tailored for outbreak or endemic situations. 11 Resources are available to assist facilities in identifying targets for a multicomponent intervention, for example, CDC’s CDI Targeted Assessment for Prevention (TAP) tool, which helps facilities use surveillance data to inform prevention efforts. 12
Three reviews and eight studies found reductions in CDI rates following implementation of multicomponent CDI prevention interventions. In this evidence summary, we first provide an overview of the reviews and then examine the studies in depth and present the primary outcomes, different intervention components, cross-cutting factors, process measures, and economic outcomes. We then present two simulation studies that attempt to measure the impact of different combinations of prevention components.
Three systematic reviews address multicomponent interventions and had sufficient methodologic quality for inclusion in this report. 1 , 2 , 13 The reviews found that studies on multicomponent interventions showed reductions in CDI, although Barker et al. (2017) 14 noted that p-values were not provided in 11 of the studies they reviewed. Barker et al. (2017) 14 reviewed 26 studies on multicomponent interventions published from database inception up to April 30, 2016. Seven of the studies they found are included in this review (many of the studies they included were published prior to 2008 and thus were not within the parameters of our searches). We include one study by Koll et al. (2014) 9 that was not included in the review by Barker et al. (2017). 14
In another review, Louh et al. (2017) examined studies published from January 1, 2009, to August 1, 2015, on CDI prevention practices in acute care hospitals. They identified 14 studies on “bundled” interventions, 13 5 of which we include in this review. Yakob et al. (2014) 2 conducted a meta-analysis of studies published up until March 2014 that measured CDI rates before and after implementation of multicomponent prevention interventions. Six studies were included, four of which are included in this review. 4 , 7 , 9 , 15 The six studies showed reductions in CDI from 33 to 61 percent. In addition to the review, they conducted simulations to assess the impact of different combinations of multicomponent interventions. These findings are described later in this section.
We found eight studies that measured CDI rates before and after implementation of a multicomponent CDI prevention intervention 3 – 9 , 15 and two simulation studies that explored different combinations of prevention components. 2 , 14 The eight real-world studies were observational or quasi-experimental with an interrupted time series or pre/post design. These studies are presented in Table 4.9.
Across studies there was a range in the number of implemented components; the multicomponent intervention studied by Price et al. (2009) included two components (a dedicated CDI isolation ward and antimicrobial stewardship), 8 while the remaining studies we reviewed all included more than three components. Studies outside of the United States are noted as such in the “Setting” column of Table 4.9.
Studies on Multicomponent CDI Prevention Interventions 2008–2018.
As shown in Table 4.10 below, in the reviewed studies, the most common component of the multicomponent interventions was environmental cleaning and decontamination, which was included in seven of the eight studies. Isolation of CDI patients and hand hygiene practices were the next most common components—each was included in five studies. Antimicrobial stewardship practices and contact precautions were each included in four studies. Testing and surveillance practices were included in three studies. In their review, Barker et al. (2017) found that in 26 studies, hand hygiene and environmental cleaning were the most common components (each in 23/26 studies) followed by patient isolation/cohorting (20/26) and contact precautions (19/26) and antimicrobial stewardship (19/26). 1 Louh et al. (2017) did not quantify the individual components across studies. 13
Components in Multicomponent CDI Prevention Interventions.
The individual practices deemed crucial to the multicomponent interventions varied across studies in this review. Some researchers felt that inclusion of antimicrobial stewardship as part of a multicomponent intervention was the primary factor in reducing CDI. 6 , 8 Conversely, Salgado et al. (2009), 15 Weiss et al. (2009), 7 Koll et al. (2014), 9 and Cheng et al. (2015) 3 all emphasized that they saw CDI reductions by focusing on C. difficile transmission prevention, without the inclusion of antimicrobial stewardship or reductions in antimicrobial use. Across the studies, the most common transmission prevention practices were use of gloves/handwashing with soap and water, 3 , 5 , 7 , 15 new training and protocols for environmental cleaning staff training, 3 , 5 , 7 , 9 and CDI patient isolation/cohorting. 3 , 8 , 9
Notably, Louh et al. (2017) found that multicomponent interventions that included environmental cleaning and decontamination were more effective than multicomponent interventions that did not include a focus on environmental cleaning. 13 However, Brakovich et al. (2013) called out the importance of surveillance as part of a multicomponent intervention in a long-term acute care hospital. 5
When discussing which cross-cutting practices facilitated the success of a multicomponent intervention, researchers highlighted several practices. The use of checklists and assigned roles was noted 4 , 5 , 9 (as well as staff education). 3 , 5 , 6 , 9 , 15 Barker et al. (2017) and Abbett et al. (2009) stated the importance of improved workflow systems and Barker et al. (2017) also pointed out that staff compliance with bundle practices is highly important and rarely adequately measured. 4 , 14 Communicating laboratory results 4 and communicating CDI patient status through door signs 4 , 9 were also highlighted. Two studies spoke to the benefits of teams, inter- and intrafacility collaborations, data collection and feedback, and collaborative learning. 6 , 9
In the study by Power et al. (2010), 6 an 850-bed hospital implemented a multicomponent intervention that included antimicrobial stewardship, hand hygiene, environmental cleaning and decontamination, and education about CDI. In five wards with higher baseline CDI rates, there was an implementation of an “improvement collaborative,” in which staff were broken into teams who planned, implemented, and measured the impact of selected PSPs as outlined by a systems improvement toolkit. 16 The five selected collaborative wards saw a 73 percent reduction in HA-CDI cases per 1,000 patient bed days after 3 months, and the rest of the hospital saw a 56 percent reduction in CDI cases per 1,000 patient bed days after 6 months (see Table 4.10). 6
Process measures included antimicrobial use, CDI tests ordered, and staff compliance with intervention components. Although not all interventions included antimicrobial stewardship, antimicrobial use was a common process measure. 3 , 7 , 8 , 15 For example, following a multicomponent intervention that included antimicrobial stewardship (in addition to a new isolation ward), Price et al. (2010) found decreases in antimicrobial use. The multicomponent intervention took place in an 820-bed hospital in the United Kingdom. After 15 months, the level of cephalosporin and quinolone use declined (22.0% and 38.7%, respectively, p <0.001), and antipseudomonal penicillin use increased by 20.7 DDD per month (p=0.011).8
Abbett et al. (2009) measured number of CDI tests as a process measure. The multicomponent intervention was in a 750-bed hospital and included the promotion of testing of suspected CDI patients (in addition to several other practices). After 2 years, Abbett et al. (2009) found a 15 percent increase in the rate (tests per 1,000 patient days) of C. difficile testing (testing rate ratio, 1.15 [95% CI, 1.12 to 1.17]; p <0.001). Koll et al. (2014) collected data on compliance from 35 acute care hospitals participating in a regional CDI prevention effort. For the submitted data (based on staff observations), the mean reported compliance with a prevention bundle was 95 percent and the mean reported compliance reported for an environmental cleaning protocol was 96 percent.4
Brakovich et al. (2013) and Weiss et al. (2009) provided financial information on the cost to implement the respective prevention interventions. Brackovich et al. (2013) reported that the cost of HPD equipment and contracted services was $1,800 per month. The cost of new microfiber mops and environmental services staff training was approximately $650. 5 While exact figures were not provided, Weiss et al. (2009) reported that costs of the intervention they studied included paying salary for four new infection preventionists and a 26.2 percent increase in staffing costs for environmental services personnel. They also reported an increase of 89.6 percent in cost of cleaning supplies, although this amount represented less than 0.03 percent of the total hospital budget. 7
In addition, Koll et al. (2014) reported savings in healthcare costs associated with a regional multicomponent intervention. They noted that 35 hospitals prevented approximately 1,084 cases of HO CDI, resulting in cost savings of $2.7 million to $6.8 million on healthcare costs. 9
To determine what combination of CDI prevention practices are most effective as a multicomponent intervention, Barker et al. (2017) conducted a simulation using a model of C. difficile transmission. The model was based on prior data to construct potential C. difficile transmissions by patients, visitors, nurses, and physicians and includes parameters such as patient antimicrobial use and length of stay. The interventions were “implemented” in a theoretical 200-bed hospital for 1 year.
After analyzing nine multicomponent intervention strategies, the researchers found that daily cleaning with sporicidal disinfectant and screening and isolating asymptomatic C. difficile carriers reduced CDI by 68.9 percent and 35.7 percent, respectively (both p <0.001). Combining these interventions into a two-intervention bundle reduced hospital-onset CDI by 82.3 percent and asymptomatic HO colonization by 90.6 percent (both, p<0.001). Adding patient hand hygiene to HCW hand hygiene reduced hospital-onset CDI rates an additional 7.9 percent (p<0.001).14
Yakob et al. (2014) conducted a series of simulations of different combinations of prevention methods based on their model of C. difficile transmission. The prevention methods included antimicrobial stewardship; administration of probiotics/intestinal microbiota transplantation; and improved hygiene and sanitation. They also examined the impact of reduced length of stay for inpatients. The researchers examined the impact of the prevention interventions on both colonization and CDI rates and found that, for infection control, the combined benefit of reducing length of stay and improving sanitation and hand hygiene significantly exceeds that achieved with either method alone. Antimicrobial stewardship showed greater efficacy in colonization control than it did in disease control. In terms of symptomatic disease incidence reduction, antimicrobials, probiotics, and intestinal microbiota transplantation proved substantially less effective than reducing length of stay and improving hygiene. 2
Two studies used a systems engineering framework to examine barriers and facilitators to prevention practices. 17 , 18 A systems engineering framework is one that examines workflow systems in relation to tasks, tools, and technologies, the physical environment, and the organization. 18 Yanke et al. (2018) conducted a qualitative analysis on barriers and facilitators of implementation of the VA C. difficile prevention bundle. The study consisted of four focus groups of healthcare staff in a variety of roles (e.g., physicians, nurses, and health technicians) at an 87-bed VA hospital. Bundle components included rapid PCR testing and diagnosis, hand hygiene promotion, and contact isolation precautions; facilitators and barriers were identified for each component.
