Epidemiology

Epidemiology and Control of Antimicrobial Resistance in Hospitals

Principal Investigator: Brad Doebbeling, MD, MSc

Project Number MRS EPID funded by VA Health Services Research & Development

October 1, 1999-October 31, 2006

RESEARCH OBJECTIVES:  There is an unprecedented worldwide crisis related to the rapid emergence and dissemination of antimicrobial resistant organisms.  Important pathogens are increasingly resistant to previously effective agents.  With excessive use of broad spectrum antimicrobials, resistance to previously effective agents has become common among microorganisms infecting hospitalized patients.  Furthermore, inconsistent use of basic infection control techniques by hospital personnel has contributed importantly to the dissemination of resistant strains within hospitals.  Concern about the transfer of resistance to organisms with greater pathogenic potential than currently resistant microbes, raises the specter of frequent infections for which no effective antimicrobial regimen is known.  Previous efforts to control resistance have not worked.  Perhaps this is partially due to the complex series of factors within hospitals that act locally to influence institutional leadership, management and the development and implementation of policies.  National consensus strategic goals or actions designed to prevent the emergence and control the spread of antimicrobial resistance have been published, including specific process and outcome measures to evaluate progress towards these goals.  However, the impact of these strategic goals on the prevention or control of antimicrobial resistance within hospitals has not been evaluated.

RESEARCH PLAN:  The proposed observational or quasi-experimental project will describe the prevalence and secular trends of antimicrobial resistance among nosocomial pathogens over time in a national sample of VA and non-VA acute care hospitals.  The study will assess the degree of implementation of specific components of recommended antimicrobial control measures at the institutional level, and the relationship between implementation of these recommended measures and the prevention and control of emergence and spread of antimicrobial resistance.  Subsequently, the specific control measures identified as important in preventing antibiotic resistance will be validated in a sample of physicians, nurses and administrators at a sample of facilities with high and low rates of antimicrobial resistance.  The institutional factors associated with effective prevention and control of antimicrobial resistance will be validated through medical record review at a stratified subsample of facilities.  The sources of data for the study will include: provider and administrator surveys, review of laboratory, infection control and pharmacy data, review of institutional policies and reports, and review of medical records.  Data analysis is planned using multiple linear, hierarchical linear and logistic regression modeling and validation.  The study methodology and instruments were previously approved by the University of Iowa IRB.

METHODOLOGY:  We plan to conduct this research using three consecutive phases of data collection.  Phase I of the proposed study will describe the prevalence of antimicrobial resistance among nosocomial pathogens through mail-out surveys of a national sample of VA and non-VA acute care hospitals, proportionately sampled on the basis of geographic region, bedsize and teaching affiliation.  Data has been obtained about each hospital’s antimicrobial guidelines and programs, as well as three-year historical data on prevalence, clusters and outbreaks of infections due to epidemiologically-important resistant nosocomial pathogens. The survey describes the implementation of specific recommended antimicrobial resistance prevention and control measures at the institutional level, and begin to assess the effectiveness of specific approaches to prevent emergence and spread of antimicrobial resistance among nosocomial pathogens.  Furthermore, the relationships between organizational factors, processes and outcomes of care, and patient infections and outbreaks due to resistant organisms will be examined in separate multivariate models.

Phase II will validate and expand the study of differences between facilities with and without problems with antimicrobial resistance, and with high and low levels of antimicrobial resistance prevention and control guideline implementation.  A validation study to confirm the survey data is planned through primary review of institutional policies, reports, and medical records in a subsample of facilities reporting high and low rates of antimicrobial resistance. 

Phase III has involved a repeat survey of facilities participating in the Phase I survey to determine changes in antimicrobial control programs over time since the original survey. The Phase III survey has expanded on previous findings to include factors identified in the initial and more focused studies as potentially important in antimicrobial resistance control.  In addition, the previously developed multivariate models will be validated by comparison of predicted and observed levels of antimicrobial resistance.  All data collection has been completed.

FINDINGS:  Antibiotic resistance varies by hospital type, bed size, geographic location and teaching status.  Levels of resistance in the surveys have been validated with antibiogram reports.  Levels of prevention and control measures vary by hospital.  Further analyses are planned.