576. Continuing Disparities in Virologic Control for People Living with HIV (PLWH) Receiving Care at a Large, Urban, Safety-Net Clinic

Background

The National HIV/AIDS Strategy highlights reduction of HIV-related disparities as a key goal. Despite universal access to therapy in the United States, the CDC estimates that only 58% of PLWH have achieved virologic suppression. We carried out a recent analysis of virologic suppression, examining for associated factors for PLWH receiving care at one of the nation’s largest, urban, safety-net clinics in order to identify ongoing outcome disparities.

Methods

Ruth M. Rothstein CORE Center, Cook County Health and Hospital System’s large, urban, safety-net HIV clinic cares for nearly 5,000 PLHW in the Chicago area. We report rates of virologic suppression for PLWH who attended at least one primary care visit between March 31, 2017 and April 1, 2018. We assessed for associations between key demographic characteristics, inclusive of zip code of residence, and virologic suppression (VL < 200 copies/mL³).

Results

A total of 4,660 patients attended at least one visit primary care visit at CORE between March 31, 2017 and April 1, 2018, of whom 84% were virologically suppressed. Sixty-six percent of our patients were African-American (AA), and 25% identified as Hispanic; 74% were male; patients’ median age was 49. On multivariate analysis, AA race (OR 1.54, P = 0.006) correlated with ongoing viremia (VL > 200 copies/mL³), while older age (age group 30 – 49, OR 0.62, P < 0.001; age group > 50 OR 0.27, P < 0.001) and identification as Hispanic (OR 0.63, P = 0.011) associated with virologic suppression. Other HIV transmission categories and demographic characteristics, inclusive of a health literacy measure, did not associate with virologic control. Of the Top 10 most populated zip codes of residence for our patients, three had a significantly higher proportion of viremic patients; while one had significantly more suppressed patients.

Conclusion

Disparities in virologic suppression persist in younger and African-American PLWH who attended care at Chicago’s largest, safety-net HIV clinic, with our data highlighting particular geographic areas of need. Structural interventions and quality improvement initiatives, at the health system and regional level, must continue to focus on improving outcomes for PLWH who fall into these demographic categories.

Disclosures

R. Lubelchek, Viiv: Scientific Advisor, Salary.

2167. Predicting Carbapenem-Resistant Enterobacteriaceae (CRE) Carriage on Admission using Updated Statewide Hospital Discharge Data

Background

We previously built a patient-level prediction model to assess an individual’s risk of Carbapenem-resistant Enterobacteriaceae (CRE) carriage upon hospital admission based on the following factors: past hospital visits (short- and long-term acute care (STACHs and LTACHs)), endoscopic procedures, infection-related diagnosis codes, and patient age and sex. Our model discriminated CRE cases relatively well (c-statistic = 0.86). In the hopes of operationalizing our results, we evaluated the distribution of predicted probabilities on an updated dataset using existing model parameters.

Methods

We used Illinois Hospital discharge data (CYs 2015–2016) with ICD-10 diagnosis and procedure codes to establish baseline exposure history (2015) and to generate predicted probabilities (2016). We calculated the number of hospital visits and the average number of hospital days in the past year (STACH and LTACH). We identified infection-related diagnosis codes using prior knowledge, and included procedure codes for endoscopic retrograde cholangiopancreatography (ERCP). We then used the model parameters from our previous work to generate predicted probabilities corresponding to each hospital visit.

Results

Our study year (2016) included 1,229,158 visits by 816,500 unique adult patients. Sixty-two percent of patients had no inpatient visits in the previous year. Among those with a prior hospitalization, the median STACH length of stay was 4 days (IQR: 2–6). Three thousand five hundred and sixty-six patients (0.4%) had previous LTACH exposure upon admission, with a median length of stay of 25 days (IQR: 13–40). Thirty-two percent of hospital visits had an infection-related diagnosis code, and 0.5% had an ERCP procedure code. Of the more than 1.2 million visits, our model predicted 10,614 visits associated with a CRE risk of over 1%, 946 visits of over 10%, and 96 visits by 63 unique patients with over a 50% risk. On average, highest risk patients were exposed to (median) 15 (7–97) STACH, 104 LTACH (37–174) days; 83% had infection codes.

Conclusion

Using a large, de-identified statewide dataset, we were able to identify a small number of extremely high-risk individuals. Selective screening of these individuals upon admission could prove to be a valuable way to identify CRE-colonized patients in order to take proper precautions.

Disclosures

All authors: No reported disclosures.

Electronically Available Comorbidities Should Be Used in Surgical Site Infection Risk Adjustment

Background

Healthcare-associated infections such as surgical site infections (SSIs) are used by the Centers for Medicare and Medicaid Services (CMS) as pay-for-performance metrics. Risk adjustment allows a fairer comparison of SSI rates across hospitals. Until 2016, Centers for Disease Control and Prevention (CDC) risk adjustment models for pay-for-performance SSI did not adjust for patient comorbidities. New 2016 CDC models only adjust for body mass index and diabetes.

Methods

We performed a multicenter retrospective cohort study of patients undergoing surgical procedures at 28 US hospitals. Demographic data and International Classification of Diseases, Ninth Revision codes were obtained on patients undergoing colectomy, hysterectomy, and knee and hip replacement procedures. Complex SSIs were identified by infection preventionists at each hospital using CDC criteria. Model performance was evaluated using measures of discrimination and calibration. Hospitals were ranked by SSI proportion and risk-adjusted standardized infection ratios (SIR) to assess the impact of comorbidity adjustment on public reporting.

Results

Of 45394 patients at 28 hospitals, 573 (1.3%) developed a complex SSI. A model containing procedure type, age, race, smoking, diabetes, liver disease, obesity, renal failure, and malnutrition showed good discrimination (C-statistic, 0.73) and calibration. When comparing hospital rankings by crude proportion to risk-adjusted ranks, 24 of 28 (86%) hospitals changed ranks, 16 (57%) changed by ≥2 ranks, and 4 (14%) changed by >10 ranks.

Conclusions

We developed a well-performing risk adjustment model for SSI using electronically available comorbidities. Comorbidity-based risk adjustment should be strongly considered by the CDC and CMS to adequately compare SSI rates across hospitals.

Regional Epidemiology of Methicillin-Resistant Staphylococcus aureus Among Adult Intensive Care Unit Patients Following State-Mandated Active Surveillance

Background

In 2007, Illinois became the first state in the United States to mandate active surveillance of methicillin-resistant Staphylococcus aureus (MRSA). The Illinois law applies to intensive care unit (ICU) patients; contact precautions are required for patients found to be MRSA colonized. However, the effectiveness of a legislated "search and isolate" approach to reduce MRSA burden among critically ill patients is uncertain. We evaluated whether the prevalence of MRSA colonization declined in the 5 years after the start of mandatory active surveillance.

Methods

All hospitals with an ICU having ≥10 beds in Chicago, Illinois, were eligible to participate in single-day serial point prevalence surveys. We assessed MRSA colonization among adult ICU patients present at time of survey using nasal and inguinal swab cultures. The primary outcome was region-wide MRSA colonization prevalence over time.

Results

All 25 eligible hospitals (51 ICUs) participated in serial point prevalence surveys over 8 survey periods (2008-2013). A total of 3909 adult ICU patients participated in the point prevalence surveys, with 432 (11.1%) found to be colonized with MRSA (95% confidence interval [CI], 10.1%-12.0%). The MRSA colonization prevalence among patients was unchanged during the study period; year-over-year relative risk for MRSA colonization was 0.97 (95% CI, .89-1.05; P = .48).

Conclusions

MRSA colonization prevalence among critically ill adult patients did not decline during the time period following legislatively mandated MRSA active surveillance. Our findings highlight the limits of legislated MRSA active surveillance as a strategy to reduce MRSA colonization burden among ICU patients.

Assessing the Ability of Hospital Diagnosis Codes to Detect Inpatient Exposure to Antibacterial Agents

OBJECTIVE Because antibacterial history is difficult to obtain, especially when the exposure occurred at an outside hospital, we assessed whether infection-related diagnostic billing codes, which are more readily available through hospital discharge databases, could infer prior antibacterial receipt. DESIGN Retrospective cohort study. PARTICIPANTS This study included 121,916 hospitalizations representing 78,094 patients across the 3 hospitals. METHODS We obtained hospital inpatient data from 3 Chicago-area hospitals. Encounters were categorized as "infection" if at least 1 International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) code indicated a bacterial infection. From medication administration records, we categorized antibacterial agents and calculated total therapy days using Centers for Disease Control and Prevention (CDC) definitions. We evaluated bivariate associations between infection encounters and 3 categories of antibacterial exposure: any, broad spectrum, or surgical prophylaxis. We constructed multivariable models to evaluate adjusted risk ratios for antibacterial receipt. RESULTS Of the 121,916 inpatient encounters (78,094 patients) across the 3 hospitals, 24% had an associated infection code, 47% received an antibacterial, and 13% received a broad-spectrum antibacterial. Infection-related ICD-9-CM codes were associated with a 2-fold increase in antibacterial administration compared to those lacking such codes (RR, 2.29; 95% confidence interval [CI], 2.27-2.31) and a 5-fold increased risk for broad-spectrum antibacterial administration (RR, 5.52; 95% CI, 5.37-5.67). Encounters with infection codes had 3 times the number of antibacterial days. CONCLUSIONS Infection diagnostic billing codes are strong surrogate markers for prior antibacterial exposure, especially to broad-spectrum antibacterial agents; such an association can be used to enhance early identification of patients at risk of multidrug-resistant organism (MDRO) carriage at the time of admission. Infect Control Hosp Epidemiol 2018;39:377-382.

A Novel Patient Recruitment Strategy: Patient Selection Directly from the Community through Linkage to Clinical Data

OBJECTIVE

This article presents and describes our methods in developing a novel strategy for recruitment of underrepresented, community-based participants, for pragmatic research studies leveraging routinely collected electronic health record (EHR) data.

METHODS

We designed a new approach for recruiting eligible patients from the community, while also leveraging affiliated health systems to extract clinical data for community participants. The strategy involves methods for data collection, linkage, and tracking. In this workflow, potential participants are identified in the community and surveyed regarding eligibility. These data are then encrypted and deidentified via a hashing algorithm for linkage of the community participant back to a record at a clinical site. The linkage allows for eligibility verification and automated follow-up. Longitudinal data are collected by querying the EHR data and surveying the community participant directly. We discuss this strategy within the context of two national research projects, a clinical trial and an observational cohort study.

CONCLUSION

The community-based recruitment strategy is a novel, low-touch, clinical trial enrollment method to engage a diverse set of participants. Direct outreach to community participants, while utilizing EHR data for clinical information and follow-up, allows for efficient recruitment and follow-up strategies. This new strategy for recruitment links data reported from community participants to clinical data in the EHR and allows for eligibility verification and automated follow-up. The workflow has the potential to improve recruitment efficiency and engage traditionally underrepresented individuals in research.

Patient-level exclusions from mHealth in a safety-net health system

Excitement about mobile health (mHealth) for improving care transitions is fueled by widespread adoption of smartphones across all social segments, but new disparities can emerge around nonadopters of technology-based communications. We conducted a cross-sectional survey of urban low-income adults to assess inadequate reading health literacy and limited English proficiency as factors affecting access to and engagement with mHealth. Although the proportion owning smartphones were comparable to national figures, adjusted analysis showed fewer patients with inadequate reading health literacy having Internet access (odds ratio [95% confidence interval]: 0.50 [0.26-0.95]), e-mail (0.43 [0.24-0.79]), and interest in using e-mail (0.34 [0.18-0.65]) for healthcare communications. Fewer patients with limited English proficiency were interested in using mobile apps (0.2 [0.09-0.46]). Inpatient status was independently associated with less interest in text messaging (0.46 [0.25-0.87]). mHealth exclusions around literacy and language proficiency threaten equity, and innovative solutions are needed to realize mHealth's potential for reducing disparities. Journal of Hospital Medicine 2017;12:90-93.

Importance of Molecular Methods to Determine Whether a Probiotic is the Source of Lactobacillus Bacteremia

There has been an increasing interest in the use of probiotic products for the prevention of Clostridium difficile infection (CDI). Bio-K+(®) is a commercial probiotic product comprising three strains of lactobacilli--Lactobacillus acidophilus CL1285(®), Lact. casei LBC80R(®) and Lact. rhamnosus CLR2(®)--that have been applied to prevent CDI. Generally considered as safe, lactobacilli have potential to cause bacteremia, endocarditis and other infections. The source of Lactobacillus bacteremia can be normal human flora or lactobacilli-containing probiotic. The aim of this study was to assess whether probiotic lactobacilli caused bacteremia and to show the value of molecular identification and typing techniques to determine probiotic and patient strain relatedness. We report an episode of Lactobacillus bacteremia in a 69-year-old man admitted to a hospital with severe congestive heart failure. During his hospitalization, he required long-term antibiotic therapy. Additionally, the patient received Bio-K+(®) probiotic as part of a quality improvement project to prevent CDI. Subsequently, Lactobacillus bacteremia occurred. Two independent blinded laboratory evaluations, using pulse field gel electrophoresis, 16S rRNA gene sequencing and DNA fingerprint analysis (rep-PCR), were performed to determine whether the recovered Lact. acidophilus originated from the probiotic product. Ultimately, the patient strain was identified as Lact. casei and both laboratories found no genetic relation between the patient's strain and any of the probiotic lactobacilli. This clinical case of lactobacillus bacteremia in the setting of probiotic exposure demonstrates the value of using discriminatory molecular methods to clearly determine whether there were a link between the patient's isolate and the probiotic strains.

Regional Epidemiology of Methicillin-Resistant Staphylococcus aureus Among Critically Ill Children in a State With Mandated Active Surveillance

BACKGROUND

In theory, active surveillance of methicillin-resistant Staphylococcus aureus (MRSA) reduces MRSA spread by identifying all MRSA-colonized patients and placing them under contact precautions. In October 2007, Illinois mandated active MRSA surveillance in all intensive care units, including neonatal intensive care units (NICUs) and pediatric intensive care units (PICUs). We evaluated MRSA trends in a large metropolitan region in the wake of this law.

METHODS

Chicago hospitals with a NICU or PICU were recruited for 8 single-day point prevalence surveys that occurred twice-yearly between June 2008 and July 2011 and then yearly in 2012 to 2013. Samples from all patients were cultured for MRSA (nose and umbilicus for neonates, nose and groin for pediatric patients). Hospital-reported admission MRSA-screening results also were obtained. Point prevalence cultures were screened for MRSA by using broth enrichment, chromogenic agar, and standard confirmatory methods.

RESULTS

All eligible hospitals (N = 10) participated (10 NICUs, 6 PICUs). Hospital-reported adherence to state-mandated MRSA screening at admission was high (95% for NICUs, 94% for PICUs). From serial point prevalence surveys, overall MRSA prevalences in the NICUs and PICUs were 4.2% (89 of 2101) and 5.7% (36 of 632), respectively. MRSA colonization prevalences were unchanged in the NICUs (year-over-year risk ratio [RR], 0.93 [95% confidence interval (CI), 0.78-1.12]; P = .45) and trended toward an increase in the PICUs (RR, 1.25 [95% CI, 0.72-2.12]; P = .053). We estimated that 81% and 40% of MRSA-positive patients in the NICUs and PICUs, respectively, had newly acquired MRSA.

CONCLUSIONS

In a region with mandated active MRSA surveillance, we found ongoing unchanged rates of MRSA colonization and acquisition among NICU and PICU patients.

Itch in the General Internal Medicine Setting: A Cross-Sectional Study of Prevalence and Quality-of-Life Effects

BACKGROUND

Itch is a well-established symptom in cutaneous disease. However, little is known about the burden of itch outside the dermatology setting.

PURPOSE

To determine the prevalence and impact of itch on quality of life (QOL) in the general internal medicine setting.

METHODS

We performed a cross-sectional study of 2076 adults from an outpatient general internal medicine clinic, using an audio computer-assisted self-administered interview. A history of itch (acute or chronic) and other physical symptoms in the past week, Patient-Reported Outcomes Measurement Information System (PROMIS) 10-item Global Health Questionnaire scores, and Patient Health Questionnaire-2 scores were assessed.

RESULTS

The prevalence of itch was 39.9 % and increased with age from 33.1 % at age 19-39 years to 45.9 % at age ≥80 years. In multivariable models controlled for socio-demographics, even feeling "a little" or "some" distress from itch was significantly associated with lower PROMIS global physical and mental health T-scores and estimated health utility scores (P ≤ 0.01). Further, feeling "quite a lot" of distress or "very much" distress from itch was associated with higher adjusted odds ratios for depressed mood (4.91 [95 % confidence interval (CI) 3.36-7.18]) and anhedonia (4.46 [95 % CI 3.07-6.47]). The patient burden of itch was similar to those of pain, constipation, sexual dysfunction, cough, and weight loss.

CONCLUSIONS

Itch occurs commonly in the primary care setting and is associated with poor QOL. Physicians should inquire about itch and its associations during review of systems. Future studies are needed to distinguish between the effects of acute and chronic itch.

Spread of Carbapenem-Resistant Enterobacteriaceae Among Illinois Healthcare Facilities: The Role of Patient Sharing

BACKGROUND

Carbapenem-resistant Enterobacteriaceae (CRE) spread regionally throughout healthcare facilities through patient transfer and cause difficult-to-treat infections. We developed a state-wide patient-sharing matrix and applied social network analyses to determine whether greater connectedness (centrality) to other healthcare facilities and greater patient sharing with long-term acute care hospitals (LTACHs) predicted higher facility CRE rates.

METHODS

We combined CRE case information from the Illinois extensively drug-resistant organism registry with measures of centrality calculated from a state-wide hospital discharge dataset to predict facility-level CRE rates, adjusting for hospital size and geographic characteristics.

RESULTS

Higher CRE rates were observed among facilities with greater patient sharing, as measured by degree centrality. Each additional hospital connection (unit of degree) conferred a 6% increase in CRE rate in rural facilities (relative risk [RR] = 1.056; 95% confidence interval [CI], 1.030-1.082) and a 3% increase among Chicagoland and non-Chicago urban facilities (RR = 1.027; 95% CI, 1.002-1.052 and RR = 1.025; 95% CI, 1.002-1.048, respectively). Sharing 4 or more patients with LTACHs was associated with higher CRE rates, but this association may have been due to chance (RR = 2.08; 95% CI, .85-5.08; P = .11).

CONCLUSIONS

Hospitals with greater connectedness to other hospitals in a statewide patient-sharing network had higher CRE burden. Centrality had a greater effect on CRE rates in rural counties, which do not have LTACHs. Social network analysis likely identifies hospitals at higher risk of CRE exposure, enabling focused clinical and public health interventions.

Electronic Public Health Registry of Extensively Drug-Resistant Organisms, Illinois, USA

In response to clusters of carbapenem-resistant Enterobacteriaceae (CRE) in Illinois, USA, the Illinois Department of Public Health and the Centers for Disease Control and Prevention Chicago Prevention Epicenter launched a statewide Web-based registry designed for bidirectional data exchange among health care facilities. CRE occurrences are entered and searchable in the system, enabling interfacility communication of patient information. For rapid notification of facilities, admission feeds are automated. During the first 12 months of implementation (November 1, 2013-October 31, 2014), 1,557 CRE reports (≈4.3/day) were submitted from 115 acute care hospitals, 5 long-term acute care hospitals, 46 long-term care facilities, and 7 reference laboratories. Guided by a state and local public health task force of infection prevention specialists and microbiologists and a nonprofit informatics entity, Illinois Department of Public Health deployed a statewide registry of extensively drug-resistant organisms. The legal, technical, and collaborative underpinnings of the system enable rapid incorporation of other emerging organisms.

Data requirements for electronic surveillance of healthcare-associated infections

Electronic surveillance for healthcare-associated infections (HAIs) is increasingly widespread. This is driven by multiple factors: a greater burden on hospitals to provide surveillance data to state and national agencies, financial pressures to be more efficient with HAI surveillance, the desire for more objective comparisons between healthcare facilities, and the increasing amount of patient data available electronically. Optimal implementation of electronic surveillance requires that specific information be available to the surveillance systems. This white paper reviews different approaches to electronic surveillance, discusses the specific data elements required for performing surveillance, and considers important issues of data validation.

When do patient-reported outcome measures inform readmission risk?

OBJECTIVE

To characterize changes in patient-reported outcome measures from hospital discharge to assess when they best inform risk of utilization as defined by readmissions or emergency department use.

PARTICIPANTS

Patients discharged from an urban safety-net hospital.

DESIGN

Longitudinal cohort study.

MAIN MEASURES

We serially administered the Memorial Symptom Assessment Scale (MSAS) and the PROMIS Global Health short form assessing General Self-Rated Health (GSRH), Global Physical (GPH), and Mental (GMH) Health at 0, 30, 90, and 180 days from hospital discharge. Time to first utilization from each survey was plotted by dichotomizing our sample on each patient-reported measure, and equivalence of the time-to-event curves was assessed using the log-rank test. Cox proportional hazard models were used to control for available covariates including prior utilization during the study, Charlson score, age, gender, and race/ethnicity. We assessed each measure's effect on the fit of the predictive models using the likelihood ratio test.

KEY RESULTS

We recruited 196 patients, of whom 100%, 98%, 90%, and 88% completed each respective survey wave. Participants' mean age was 52 years, 51% were women, 60% were non-Hispanic black, and 21% completed the questionnaires in Spanish. In-hospital assessments revealed high symptom burden and poor health status. In-hospital assessments of GMH and GSRH predicted 14-day reutilization, whereas posthospitalization assessments of MSAS and GPH predicted subsequent utilizations. Each measure selectively improved predictive model fit.

CONCLUSIONS

Routine measurement of patient-reported outcomes can help identify patients at higher risk for utilizations. At different time points, MSAS, GPH, GMH, and GSRH all informed utilization risk.

Implementation of an audio computer-assisted self-interview (ACASI) system in a general medicine clinic: patient response burden

BACKGROUND

Routine implementation of instruments to capture patient-reported outcomes could guide clinical practice and facilitate health services research. Audio interviews facilitate self-interviews across literacy levels.

OBJECTIVES

To evaluate time burden for patients, and factors associated with response times for an audio computer-assisted self interview (ACASI) system integrated into the clinical workflow.

METHODS

We developed an ACASI system, integrated with a research data warehouse. Instruments for symptom burden, self-reported health, depression screening, tobacco use, and patient satisfaction were administered through touch-screen monitors in the general medicine clinic at the Cook County Health & Hospitals System during April 8, 2011-July 27, 2012. We performed a cross-sectional study to evaluate the mean time burden per item and for each module of instruments; we evaluated factors associated with longer response latency.

RESULTS

Among 1,670 interviews, the mean per-question response time was 18.4 [SD, 6.1] seconds. By multivariable analysis, age was most strongly associated with prolonged response time and increased per decade compared to < 50 years as follows (additional seconds per question; 95% CI): 50-59 years (1.4; 0.7 to 2.1 seconds); 60-69 (3.4; 2.6 to 4.1); 70-79 (5.1; 4.0 to 6.1); and 80-89 (5.5; 4.1 to 7.0). Response times also were longer for Spanish language (3.9; 2.9 to 4.9); no home computer use (3.3; 2.8 to 3.9); and, low mental self-reported health (0.6; 0.0 to 1.1). However, most interviews were completed within 10 minutes.

CONCLUSIONS

An ACASI software system can be included in a patient visit and adds minimal time burden. The burden was greatest for older patients, interviews in Spanish, and for those with less computer exposure. A patient's self-reported health had minimal impact on response times.

Multicenter Intervention Program to Increase Adherence to Hand Hygiene Recommendations and Glove Use and to Reduce the Incidence of Antimicrobial Resistance

Objective.

To determine whether a multimodal intervention could improve adherence to hand hygiene and glove use recommendations and decrease the incidence of antimicrobial resistance in different types of healthcare facilities.

Design.

Prospective, observational study performed from October 1, 1999, through December 31, 2002. We monitored adherence to hand hygiene and glove use recommendations and the incidence of antimicrobial-resistant bacteria among isolates from clinical cultures. We evaluated trends in and predictors for adherence and preferential use of alcohol-based hand rubs, using multivariable analyses.

Setting.

Three intervention hospitals (a 660-bed acute and long-term care hospital, a 120-bed community hospital, and a 600-bed public teaching hospital) and a control hospital (a 700-bed university teaching hospital).

Intervention.

At the intervention hospitals, we introduced or increased the availability of alcohol-based hand rub, initiated an interactive education program, and developed a poster campaign; at the control hospital, we only increased the availability of alcohol-based hand rub.

Results.

We observed 6,948 hand hygiene opportunities. The frequency of hand hygiene performance or glove use significantly increased during the study period at the intervention hospitals but not at the control hospital; the maximum quarterly frequency of hand hygiene performance or glove use at intervention hospitals (74%, 80%, and 77%) was higher than that at the control hospital (59%). By multivariable analysis, preferential use of alcohol-based hand rubs rather than soap and water for hand hygiene was more likely among workers at intervention hospitals compared with nonintervention hospitals (adjusted odds ratio, 4.6 [95% confidence interval, 3.3-6.4]) and more likely among physicians (adjusted odds ratio, 1.4 [95% confidence interval, 1.2-1.8]) than among nurses at intervention hospitals. A significantly reduced incidence of antimicrobial-resistant bacteria among isolates from clinical culture was found at a single intervention hospital, which had the greatest increase in the frequency of hand hygiene performance.

Conclusions.

During a 3-year period, a multimodal intervention program increased adherence to hand hygiene recommendations, especially to the use of alcohol-based hand rubs. In one hospital, a concomitant reduction was found in the incidence of antimicrobial-resistant bacteria among isolates from clinical cultures.