Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

IIR 09-099 – HSR Study

IIR 09-099
Improving Practice: Automated Compliance Monitoring in Infection Control
Eli N. Perencevich, MD MS BS
Iowa City VA Health Care System, Iowa City, IA
Iowa City, IA
Funding Period: January 2011 - December 2013
The VA MRSA Initiative's potential to reduce MRSA infections will only be realized if healthcare workers are compliant with hand hygiene and contact precautions (wearing gown and glove by healthcare workers each time they enter a patient's room). Thus, if healthcare workers do not wash their hands and wear gloves/gowns when they see patients, the very expensive MRSA Initiative may be ineffective. Two factors are known to improve compliance with infection control practice: 1) continuous monitoring of compliance and 2) direct feedback of compliance rates to healthcare workers with a goal to get compliance up to 100%. However, direct observation of hand hygiene and glove wearing is labor intensive, expensive and rarely occurs during night shifts. Additionally, isolated patients may be subject to better or poorer hand hygiene compliance.

The objectives of this study were to: (1) measure, using direct observations, monthly compliance with hand-hygiene, glove use and gown use and correlate this measure with compliance as estimated through monthly supply-chain delivery of gowns, gloves and bottles of alcohol-based hand rub at an intensive care unit and hospital ward level; (2) complete both a cross-sectional study and a case-crossover study to assess risk-factors (e.g. nursing-staff workload, time of day) for poor compliance with hand-hygiene, gown and glove use and measure and compare the independent associations of these risk-factors with directly observed compliance with hand-hygiene and contact precautions (gown and glove use); and (3) provide sustained feedback of estimated compliance calculated through the use of automated monthly supply delivery counts of alcohol-based hand rub, gowns and gloves to specific hospital wards and statistically link estimated improved compliance with actual directly observed compliance using an upper-level quasi-experimental design with cross-over and non-equivalent control groups within 5 ICUs and 6 wards at 3 VA acute care hospitals. Additionally, we sought to determine if contact isolation modifies or improves healthcare worker behavior including hand hygiene.

This project contained two distinct studies. The first study, using an observational cohort study, developed and validated supply chain delivery of alcohol-based hand rub and glove use as a viable estimate of the related actual infection control compliance measure. The second study tested the efficacy of sustained feedback of automated monthly supply delivery counts of alcohol-based hand rub and gloves in improving compliance and reducing MRSA infections in 6 hospital intensive care units in 3 VA acute care hospitals using an interrupted time-series analysis quasi-experimental design with non-equivalent control groups.

We have completed and published five articles. The first was published in January 2013. In this article, observers performed "secret shopper" monitoring of healthcare worker (HCW) activities during routine care, using a standardized collection tool and fixed 1-hour observation periods. The goal was to identify risk factors for poor hand hygiene compliance. (Objective #2) A total of 7,743 patient visits by the healthcare worker were observed over 1,989 hours. Patients isolated had 36.4% fewer hourly HCW visits than patients not on contact precautions (2.78 vs 4.37 visits per hour); as well as 17.7% less direct patient contact time with HCWs (13.98 vs 16.98 minutes per hour). Patients on contact precautions tended to have fewer visitors (23.6% fewer). Healthcare workers were more likely to perform hand hygiene on exiting the room of a patient on contact precautions (63.2% vs 47.4%) in rooms of patients not on contact precautions; Thus, independent of the direct effect of contact isolation, isolation was also associated with activities likely to reduce transmission of resistant pathogens, such as fewer visits and better hand hygiene at exit, while exposing patients on contact precautions to less HCW contact, less visitor contact, and potentially other unintended outcomes. For the second article, we performed weighted linear regression to assess the ability of soap and alcohol usage to predict observed hand-hygiene compliance proportions using the entire data set (Objective #1). The models were weighted according to the number of observations used to compute the hand-hygiene compliance proportion. We initially fit three models: soap (mL) as the predictor (Model 1), alcohol (mL) as the predictor (Model 2), and soap and alcohol both as the predictors (Model 3). Next, we computed soap usage rate (mL per patient per day) and alcohol usage rate (mL per patient per day) and fit three more models: soap usage rate as the predictor (Model 4), alcohol usage rate as the predictor (Model 5), and soap and alcohol usage rates both as the predictors (Model 6). Lastly, we refit the same six models separately in ICUs and non-ICUs. We found that the model using soap and alcohol usage rates explained the most variability in hand-hygiene compliance (R2 =16.1%). When considering only ICUs, the model with raw soap usage (R2=11.5%) and the model with both soap and alcohol usage rates (R2=11.6%) explained the most variability in hand-hygiene compliance. In contrast, when considering only non-ICUs, the model using raw soap and alcohol usage explained the most variability in hand-hygiene compliance (R2 =32.4%). Thus, soap and alcohol usage rates were poor surrogates for tracking hand hygiene compliance using VA data. The third article accepted for publication in Infect Control Hosp Epidemiol, Reisinger et al. described the results of a clinical trial of point-of-use reminder signs for improving hand hygiene compliance (Objective #3). In Feb 2012, one of 4 signs were placed next to hand rub dispensers on 5 randomly chosen VA wards and ICUs. The remaining 6 control wards/ICUs did not have signs. 13,221 hand hygiene opportunities were observed pre-intervention and 915 opportunities were observed in the intervention period. The sign with patient-focused and gain-framed language had the highest entry compliance 53% vs 29% for the other signs (p=0.042) and the highest exit compliance 79% vs 52% for the other signs (p=0.36). The fourth article (accepted for publication, Infect Control Hosp Epidemiol) used 11,444 observations collected during 3,432 hours of hand hygiene compliance to measure the timing of the emergence of the Hawthorne Effect. Exit compliance increased after 14 minutes of observation (p < 0.001) and increased further after 50 minutes (p < 0.001). Entry compliance increased after 38 minutes (p = 0.005).

This project will have several impacts on improving the safety and quality of care for our Veterans. First, it showed that soap and alcohol usage is a poor surrogate for directly observed hand hygiene compliance. Second, it showed that the Hawthorne effect causes directly observed hand hygiene compliance surveillance to be less accurate if observation extends beyond 15 minutes. These two findings have directly influenced an information letter that is to be distributed in VA-wide via the Office of Public Health and National Infectious Diseases Service that affirms the role of directly observed compliance of hand hygiene. We also identified variation of practice in hand hygiene surveillance in the three hospitals under study. This finding informed a survey that was distributed administratively to all 141 medical centers via the VHA Deputy Under Secretary of Healthcare Operations and Management Office in March 2012. A memo from the Deputy Undersecretary requested the person most familiar with hand hygiene practices at each of the medical centers to complete the survey. All medical centers completed the survey. This survey was published in Am J Infect Control.

External Links for this Project

NIH Reporter

Grant Number: I01HX000107-01

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


None at this time.

DRA: Health Systems, Infectious Diseases
DRE: Epidemiology, Prevention, Technology Development and Assessment
Keywords: Implementation, Infectious disease, Quality assurance, improvement
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.