Distinguishing recurrent Clostridium difficile infection (CDI), defined as CDI caused by the same genotype, from reinfection with a different genotype, has important implications for surveillance and clinical trials investigating treatment effectiveness. We validated the proposed 8-week period for distinguishing “same genotype CDI” from “different genotype CDI,” and we aimed to identify clinical variables with distinctiveness to propose an improved definition.

From January 2004 to December 2013, a cohort of all inpatients with CDI at the University Hospital Basel, Switzerland, was established, and respective strains were collected. In patients with a second episode of CDI, both strains were compared using polymerase chain reaction (PCR) ribotyping. The standard definition of recurrence (within 8 weeks after initial diagnosis) was evaluated for its performance to predict CDI caused by the same genotype.

Among 750 patients with CDI, 130 (17.3%) were diagnosed with recurrence or reinfection. Strains from both episodes were available from 106 patients. Identical strains were identified in 36 patients with recurrence (36 of 47) and 27 patients with reinfection (27 of 59). Sensitivity, specificity, and negative and positive predictive values of the standard definition were 56%, 74%, 53%, and 76%, respectively. An extended period of 20 weeks resulted in the best match for both sensitivity and specificity (83% and 58%, respectively), while none of the clinical characteristics revealed independent distinctive power.

Our results challenge the utility of the 8-week cutoff for distinguishing recurrent CDI from reinfection. An extended period of 20 weeks may result in improved overall performance characteristics, but this finding requires external validation.

Infect Control Hosp Epidemiol 2017;38:891–896

Estimating the risk of a complicated course of Clostridium difficile infection (CDI) might help doctors guide treatment. We aimed to validate 3 published prediction models: Hensgens (2014), Na (2015), and Welfare (2011).

The validation cohort comprised 148 patients diagnosed with CDI between May 2013 and March 2014. During this period, 70 endemic cases of CDI occurred as well as 78 cases of CDI related to an outbreak of C. difficile ribotype 027. Model calibration and discrimination were assessed for the 3 prediction rules.

A complicated course (ie, death, colectomy, or ICU admission due to CDI) was observed in 31 patients (21%), and 23 patients (16%) died within 30 days of CDI diagnosis. The performance of all 3 prediction models was poor when applied to the total validation cohort with an estimated area under the curve (AUC) of 0.68 for the Hensgens model, 0.54 for the Na model, and 0.61 for the Welfare model. For those patients diagnosed with CDI due to non-outbreak strains, the prediction model developed by Hensgens performed the best, with an AUC of 0.78.

All 3 prediction models performed poorly when using our total cohort, which included CDI cases from an outbreak as well as endemic cases. The prediction model of Hensgens performed relatively well for patients diagnosed with CDI due to non-outbreak strains, and this model may be useful in endemic settings.

Infect Control Hosp Epidemiol 2017;38:897–905

To describe the hospital stays of patients with Clostridium difficile infection (CDI) and to measure the hospitalization costs of CDI (as primary and secondary diagnoses) from the French national health insurance perspective

Burden of illness study

All acute-care hospitals in France

Data were extracted from the French national hospitalization database (PMSI) for patients covered by the national health insurance scheme in 2014. Hospitalizations were selected using the International Classification of Diseases, 10threvision (ICD-10) code for CDI. Hospital stays with CDI as the primary diagnosis or the secondary diagnosis (comorbidity) were studied for the following parameters: patient sociodemographic characteristics, mortality, length of stay (LOS), and related costs. A retrospective case-control analysis was performed on stays with CDI as the secondary diagnosis to assess the impact of CDI on the LOS and costs.

Overall, 5,834 hospital stays with CDI as the primary diagnosis were included in this study. The total national insurance costs were €30.7 million (US $33,677,439), and the mean cost per hospital stay was €5,267±€3,645 (US $5,777±$3,998). In total, 10,265 stays were reported with CDI as the secondary diagnosis. The total national insurance additional costs attributable to CDI were estimated to be €85 million (US $93,243,725), and the mean additional cost attributable to CDI per hospital stay was €8,295±€17,163, median, €4,797 (US $9,099±$8,827; median, $5,262).

CDI has a high clinical and economic burden in the hospital, and it represents a major cost for national health insurance. When detected as a comorbidity, CDI was significantly associated with increased LOS and economic burden. Preventive approaches should be implemented to avoid CDIs.

Infect Control Hosp Epidemiol 2017;38:906–911

To estimate the incidence and identified risk factors for community-acquired (CA) and hospital-acquired (HA) Clostridium difficile infection (CDI)

We conducted 2 parallel case-control studies at Tel Aviv Sourasky Medical Center from January 1, 2011, to December 31, 2014. We identified persons with CDI, determined whether infection was community or hospital acquired, and calculated incidence rates from 2007 to 2014. We collected demographic, clinical, and epidemiological information for CDI cases and hospitalized control cases and estimated the odds ratio with 95% confidence interval using conditional logistic regression.

In total, 1,563 CDI cases were identified in the study. The incidence rate of CA-CDI and HA-CDI increased by 1.6-fold and 1.2-fold, respectively, during 2012–2014. However, the incidence rate of CA-CDI was 0.84 per 100,000 (95% CI, 0.52–1.30), the rate for HA-CDI was 4.7 per 10,000 patient days (95% CI, 4.08–5.38), respectively, in 2014. We identified several factors as independent variables significantly associated with HA-CDI: functional disability, presence of nasogastric tube, antibiotic use, chemotherapy, infection by extended-spectrum β-lactamases, and mean of albumin values. Risk factors independently associated with CA-CDI were close contact with a family member who had been hospitalized in the previous 6 months, inflammatory bowel disease, and home density index (adjusted odds ratio, 25.7; 95% confidence interval, 3.99–165.54; P=.001).

The identification of the main modifiable risk factors for HA-CDI (antibiotic exposure and hypoalbuminemia) and for CA-CDI (close contact with a family member who had been hospitalized in the previous 6 months) is likely to optimize prevention efforts; these factors are critical in preventing the spread of CDI.

Infect Control Hosp Epidemiol 2017;38:912–920

Carbapenem-resistant Enterobacteriaceae (CRE) are a significant clinical and public health concern. Understanding the distribution of CRE colonization and developing a coordinated approach are key components of control efforts. The prevalence of CRE in the District of Columbia is unknown. We sought to determine the CRE colonization prevalence within healthcare facilities (HCFs) in the District of Columbia using a collaborative, regional approach.

Point-prevalence study.

This study included 16 HCFs in the District of Columbia: all 8 acute-care hospitals (ACHs), 5 of 19 skilled nursing facilities, 2 (both) long-term acute-care facilities, and 1 (the sole) inpatient rehabilitation facility.

Inpatients on all units excluding psychiatry and obstetrics-gynecology.

CRE identification was performed on perianal swab samples using real-time polymerase chain reaction, culture, and antimicrobial susceptibility testing (AST). Prevalence was calculated by facility and unit type as the number of patients with a positive result divided by the total number tested. Prevalence ratios were compared using the Poisson distribution.

Of 1,022 completed tests, 53 samples tested positive for CRE, yielding a prevalence of 5.2% (95% CI, 3.9%–6.8%). Of 726 tests from ACHs, 36 (5.0%; 95% CI, 3.5%–6.9%) were positive. Of 244 tests from long-term-care facilities, 17 (7.0%; 95% CI, 4.1%–11.2%) were positive. The relative prevalence ratios by facility type were 0.9 (95% CI, 0.5–1.5) and 1.5 (95% CI, 0.9–2.6), respectively. No CRE were identified from the inpatient rehabilitation facility.

A baseline CRE prevalence was established, revealing endemicity across healthcare settings in the District of Columbia. Our study establishes a framework for interfacility collaboration to reduce CRE transmission and infection.

Infect Control Hosp Epidemiol 2017;38:921–929

To characterize the risk of infection after MRSA decolonization with intranasal mupirocin.

Multicenter, retrospective cohort study.

Tertiary care neonatal intensive care units (NICUs) from 3 urban hospitals in the United States ranging in size from 45 to 100 beds.

MRSA-colonized neonates were identified from NICU admissions occurring from January 2007 to December 2014, during which a targeted decolonization strategy was used for MRSA control. In 2 time-to-event analyses, MRSA-colonized neonates were observed from the date of the first MRSA-positive surveillance screen until (1) the first occurrence of novel gram-positive cocci in sterile culture or discharge or (2) the first occurrence of novel gram-negative bacilli in sterile culture or discharge. Mupirocin exposure was treated as time varying.

A total of 522 MRSA-colonized neonates were identified from 16,144 neonates admitted to site NICUs. Of the MRSA-colonized neonates, 384 (74%) received mupirocin. Average time from positive culture to mupirocin treatment was 3.5 days (standard deviation, 7.2 days). The adjusted hazard of gram-positive cocci infection was 64% lower among mupirocin-exposed versus mupirocin-unexposed neonates (hazard ratio, 0.36; 95% confidence interval [CI], 0.17–0.76), whereas the adjusted hazard ratio of gram-negative bacilli infection comparing mupirocin-exposed and -unexposed neonates was 1.05 (95% CI, 0.42–2.62).

In this multicentered cohort of MRSA-colonized neonates, mupirocin-based decolonization treatment appeared to decrease the risk of infection with select gram-positive organisms as intended, and the treatment was not significantly associated with risk of subsequent infections with organisms not covered by mupirocin’s spectrum of activity.

Infect Control Hosp Epidemiol 2017;38:930–936

To examine variation in antibiotic coverage and detection of resistant pathogens in community-onset pneumonia.

Cross-sectional study.

A total of 128 hospitals in the Veterans Affairs health system.

Hospitalizations with a principal diagnosis of pneumonia from 2009 through 2010.

We examined proportions of hospitalizations with empiric antibiotic coverage for methicillin-resistant Staphylococcus aureus (MRSA) and Pseudomonas aeruginosa (PAER) and with initial detection in blood or respiratory cultures. We compared lowest- versus highest-decile hospitals, and we estimated adjusted probabilities (AP) for patient- and hospital-level factors predicting coverage and detection using hierarchical regression modeling.

Among 38,473 hospitalizations, empiric coverage varied widely across hospitals (MRSA lowest vs highest, 8.2% vs 42.0%; PAER lowest vs highest, 13.9% vs 44.4%). Detection rates also varied (MRSA lowest vs highest, 0.5% vs 3.6%; PAER lowest vs highest, 0.6% vs 3.7%). Whereas coverage was greatest among patients with recent hospitalizations (AP for anti-MRSA, 54%; AP for anti-PAER, 59%) and long-term care (AP for anti-MRSA, 60%; AP for anti-PAER, 66%), detection was greatest in patients with a previous history of a positive culture (AP for MRSA, 7.9%; AP for PAER, 11.9%) and in hospitals with a high prevalence of the organism in pneumonia (AP for MRSA, 3.9%; AP for PAER, 3.2%). Low hospital complexity and rural setting were strong negative predictors of coverage but not of detection.

Hospitals demonstrated widespread variation in both coverage and detection of MRSA and PAER, but probability of coverage correlated poorly with probability of detection. Factors associated with empiric coverage (eg, healthcare exposure) were different from those associated with detection (eg, microbiology history). Providing microbiology data during empiric antibiotic decision making could better align coverage to risk for resistant pathogens and could promote more judicious use of broad-spectrum antibiotics.

Infect Control Hosp Epidemiol 2017;38:937–944

Simulation models have been used to investigate the impact of hand hygiene on methicillin-resistant Staphylococcus aureus (MRSA) transmission within the healthcare setting, but they have been limited by their ability to accurately model complex patient–provider interactions.

Using a network-based modeling approach, we created a simulated neonatal intensive care unit (NICU) representing the potential for per-hour infant–infant MRSA transmission via the healthcare worker resulting in subsequent colonization. The starting prevalence of MRSA colonized infants varied from 2% to 8%. Hand hygiene ranged from 0% (none) to 100% (theoretical maximum), with an expected effectiveness of 88% inferred from literature.

Based on empiric care provided within a 1-hour period, the mean number of infant–infant MRSA transmissible opportunities per hour was 1.3. Compared to no hand hygiene and averaged across all initial colonization states, colonization was reduced by approximately 29%, 51%, 67%, 80%, and 86% for the respective levels of hygiene: 24%, 48%, 68%, 88%, and 100%. Preterm infants had a 61% increase in MRSA colonization, and mechanically ventilated infants had a 27% increase.

Even under optimal hygiene conditions, horizontal transmission of MRSA is possible. Additional prevention paradigms should focus on the most acute patients because they are at greatest risk.

Infect Control Hosp Epidemiol 2017;38:945–952

Multistate and competing risks models have become an established and adequate tool with which to quantify determinants and consequences of nosocomial infections. In this tutorial article, we explain and demonstrate the basics of these models to a broader audience of professionals in health care, infection control, and hospital epidemiology.

Using a publicly available data set from a cohort study of intensive care unit patients, we show how hospital infection data can be displayed and explored graphically and how simple formulas are derived under some simplified assumptions for illustrating the basic ideas behind multistate models. Only a few simply accessible values (event counts and patient days) and a pocket calculator are needed to reveal basic insights into cumulative risk and clinical outcomes of nosocomial infection in terms of mortality and length of stay.

We show how to use these values to perform basic multistate analyses in own data or to correct biased estimates in published data, as these values are often reported. We also show relationships between multistate-based hazard ratios and odds ratios, which are derived from the popular logistic regression model.

No sophisticated statistical software is required to apply a basic multistate model and to avoid typical pitfalls such as time-dependent or competing-risks bias.

Infect Control Hosp Epidemiol 2017;38:953–959

A diverse group of hospitals in Iowa implemented a program to objectively evaluate and improve the thoroughness of disinfection cleaning of near-patient surfaces. Administrative benefits of, challenges of, and impediments to the program were also evaluated.

We conducted a prospective, quasi-experimental pre-/postintervention trial to improve the thoroughness of terminal room disinfection cleaning. Infection preventionists utilized an objective cleaning performance monitoring system (DAZO) to evaluate the thoroughness of disinfection cleaning (TDC) expressed as a proportion of objects confirmed to have been cleaned (numerator) to objects to be cleaned per hospital policy (denominator)×100. Data analysis, educational interventions, and objective performance feedback were modeled on previously published studies using the same monitoring tool. Programmatic analysis utilized unstructured and structured information from participants irrespective of whether they participated in the process improvement aspects to the program.

Initially, the overall TDC was 61% in 56 hospitals. Hospitals completing 1 or 2 feedback cycles improved their TDC percentages significantly (P<.0001; P<.005). Overall, 22 hospitals (39.3%) completed all 3 study phases and significantly increased their TDC percentages to a mean of 89%. Moreover, 6 hospitals maintained the program beyond the planned study period and sustained TDC percentages >90% for at least 38 months. A survey of infection preventionists found that lack of time and staff turnover were the most common reasons for terminating the study early.

The study confirmed that hospitals using this program can improve their TDC percentages significantly. Hospitals must invest resources to improve cleaning and to sustain their gains.

Infect Control Hosp Epidemiol 2017;38:960–965

To compare the rates of and reasons for presenteeism associated with influenza-like illness (ILI) among healthcare professionals (HCPs) caring for hospitalized inpatient transplant recipients and internal medicine patients.

We designed a 10-question anonymous survey in which ILI was defined as fever (>37.8°C) and cough and/or sore throat and ILI B was defined as fever (>37.8°C) or cough or sore throat; both definitions were considered in the absence of another known cause.

Tertiary-care center.

Physicians, advanced practice providers (APPs) and nurses.

Survey deployed at peak of influenza activity in 2016.

Rates of ILI, presenteeism, wearing masks, and time away due to ILI.

Of 707 HCPs surveyed, 286 (40%) responded; 15 (5.2%) reported having ILI, and 73 (25.5%) reported having ILI B in the preceding 2 weeks. Presenteeism rates were 92% in both groups of HCPs and were higher among women (adjusted odds ratio [AOR], 2.64; 95% CI, 1.23–5.71; P=.01) and those ≤40 years old (AOR, 1.92; 95% CI, 1.03–3.68; P=.04). Healthcare professionals who cared for transplant recipients and female HCPs were more likely to wear masks (AOR, 2.13; 95% CI, 1.05–3.40; P=.04 for transplant recipients and AOR, 3.96; 95% CI, 1.35–11.63; P=.01 for female HCPs). Nurses were more likely than physicians and APPs to take time off (AOR, 3.33; 95% CI, 1.10–10.09; P=.03.)

Presenteeism among HCPs with ILI is common, including among those caring for transplant recipients. Nonpunitive systems should encourage HCPs not to work with ILI and to wear masks to prevent spread of infections.

Infect Control Hosp Epidemiol 2017;38:966–969

To identify predictors of influenza vaccine acceptance among VHA healthcare workers (HCWs), with emphasis on modifiable factors related to promotion campaigns.

Survey.

National single-payer healthcare system with 140 hospitals and 321,000 HCWs.

National voluntary sample of HCWs in the Veterans Health Administration (VHA) system.

We invited a random sample of 5% of all VHA HCWs to participate. An 18-item intranet-based survey inquired about occupation, vaccination status, employer policy, and local campaign efforts.

The response rate was 17.4%. Of 2,502 initial respondents, 2,406 (96.2%) provided usable data. This sample includes respondents from all 140 VA hospitals. Self-reported influenza vaccination rates were highest among physicians (95.6%) and licensed independent providers (88.3%). Nonclinical staff (80.7%) reported vaccine uptake similar to other certified but nonlicensed providers (81.2%). The strongest predictor of vaccine acceptance among VHA HCWs was individual awareness of organizational policy. Vaccine acceptance was also higher among HCWs who reported more options for access to vaccination and among those in facilities with more education activities.

Influenza vaccine acceptance varied significantly by employee awareness of employer policy and on-site access to vaccine. Employer-sponsored activities to increase access continue to show positive returns across occupations. Local influenza campaign efforts to educate HCWs may have reached saturation in this target group. These results suggest that focused communications to increase HCW awareness and understanding of employer policy can drive further increase in influenza vaccination acceptance.

Infect Control Hosp Epidemiol 2017;38:970–975

To determine the number and proportion of healthcare worker (HCW) tuberculosis (TB) cases infected while working in healthcare institutions in the Netherlands and to learn from circumstances that led to these infections.

Cohort analysis.

We included all HCW TB patients reported to the Netherlands TB Register from 2000 to 2015. Using data from this register, including DNA fingerprints of the bacteria profile and additional information from public health clinics, HCW TB cases were classified into 4 categories: (1) infected during work in the Netherlands, (2) infected in the community, (3) infected outside the Netherlands, or (4) outside these 3 categories. An in-depth analysis of category 1 cases was performed to identify factors contributing to patient-to-HCW transmission.

In total, 131 HCW TB cases were identified: 32 cases (24%) in category 1; 13 cases (10%) in category 2; 42 cases (32%) in category 3; and 44 cases (34%) in category 4. The annual number of HCW TB cases (P<.05), the proportion among reported cases (P<.01), and the number of category 1 HCW TB cases (P=.12) all declined over the study period. Delayed diagnosis in a TB patient was the predominant underlying factor of nosocomial transmission in 47% of category 1 HCW TB patients, most of whom were subsequently identified in a contact investigation. Performing high-risk procedures was the main contributing factor in the other 53% of cases.

In low-incidence countries, every HCW TB case should warrant timely and thorough investigation to help further define and fine-tune the HCW screening policy and to monitor its proper implementation.

Infect Control Hosp Epidemiol 2017;38:976–982

In 2013, the Centers for Disease Control and Prevention released a novel surveillance concept called the “ventilator-associated event,” which focused surveillance on objective measures of complications among patients that underwent invasive ventilations.

To evaluate the concordance and possible differences in efficacy (ie, disease severity and outcomes) between 2 surveillance paradigms: (1) infection-related ventilator-associated complications (iVAC) and (2) on conventional ventilator-associated pneumonia (VAP).

Prospective, observational, single-center cohort study.

This study included 85 adult patients that received invasive ventilation for at least 2 consecutive calendar days in a 22-bed, adult, mixed medical-surgical intensive care unit in Finland between October 2014 and June 2015.

Among these patients, 9 (10.1 per 1,000 days of mechanical ventilation) developed iVAC (10.6%) and 20 (22.4 per 1,000 days of mechanical ventilation) developed conventional VAP (23.5%). The iVAC indicators were most often caused by atelectasis and fluid overload. Compared with patients with conventional VAP, patients with iVAC had significantly worse respiratory status but no other differences in disease severity or outcomes.

The incidence of conventional VAP was >2-fold that of iVAC, and the surveillance paradigms for VAP and iVAC capture different patterns of disease. Our results suggest that this novel surveillance concept, although based on objective measures of declining oxygenation, actually identified deteriorations of oxygenation due to noninfectious causes.

Infect Control Hosp Epidemiol 2017;38:983–988