ORIGINAL ARTICLE Development and Validation of Measures to Assess Prevention and Control of AMR in Hospitals Mindy Flanagan, PhD,*† Rangaraj Ramanujam, PhD,*†‡§ Jason Sutherland, PhD,*†¶ Thomas Vaughn, PhD,** Daniel Diekema, MD, MS,†† and Bradley N. Doebbeling, MD, MSc*†§†† Background: The rapid spread of antimicrobial resistance (AMR) in the US hospitals poses serious quality and safety problems. Expert panels, identifying strategies for optimizing antibiotic use and pre- venting AMR spread, have recommended hospitals undertake efforts to implement specific evidence-based practices. Objective: To develop and validate a measurement scale for assess- ing hospitals’ efforts to implement recommended AMR prevention and control measures. Study Design: Surveys were mailed to infection control profession- als in a national sample of 670 US hospitals stratified by geographic region, bedsize, teaching status, and VA affiliation. Subjects: Four hundred forty-eight infection control professionals participated (67% response rate). Methods: Survey items measured implementation of guideline rec- ommendations, practices for AMR monitoring and feedback, AMR- related outcomes (methicillin-resistant Staphylococcus aureus prev- alence and outbreaks MRSA), and organizational features. “Derivation” and “validation” samples were randomly selected. Exploratory factor analysis was performed to identify factors under- lying AMR prevention and control efforts. Multiple methods were used for validation. Results: We identified 4 empirically distinct factors in AMR pre- vention and control: (1) practices for antimicrobial prescription/use, (2) information/resources for AMR control, (3) practices for isolat- ing infected patients, and (4) organizational support for infection control policies. The Prevention and Control of Antimicrobial Resis- tance scale was reliable and had content and construct validity. MRSA prevalence was significantly lower in hospitals with higher resource/ information availability and broader organizational support. Conclusions: The Prevention and Control of Antimicrobial Resis- tance scale offers a simple yet discriminating assessment of AMR prevention and control efforts. Use should complement assessment methods based exclusively on AMR outcomes. Key Words: AMR, factor analysis, measurement development, MRSA, organizational (Med Care 2007;45: 537–544) U S hospitals continue to experience a rapid increase in the incidence of infection and colonization with antibiotic- resistant organisms. The emergence of antimicrobial resis- tance (AMR) is highly correlated with inappropriate or ex- cessive use of antimicrobial agents. 1–3 Concerns that the development of new antimicrobial drugs is not keeping pace with the accelerating emergence and spread of resistant mi- croorganisms have led to dire predictions about the “postan- tibiotic era” and a “shadow epidemic” of AMR. 4–6 Although the factors driving the emergence of AMR are multiple and far-ranging, including societal and technological changes that are outside the control of any 1 organization, 7 there is general agreement that healthcare organizations should be doing more locally to prevent and control AMR (eg, implement evidence- based infection control measures and interventions, reduce the excessive use of broad spectrum antimicrobials). 1,4,8 –16 To provide guidance to healthcare organizations in their efforts to manage AMR, the healthcare community convened over 25 expert panels since 1987 to identify and recommend specific goals and actions for AMR prevention and control. 17 The recommendations of 1 such expert panel, convened by the National Foundation for Infectious Diseases (NFID) and Centers for Disease Control and Prevention (CDC), have been especially influential. 8 Based on an extensive review of the evidence-base and consultations with infection control experts, the panel identified 10 strategies or goals for AMR prevention and control, which were grouped under 2 catego- From the *VA HSR&D Center for Implementing Evidence-based Practice, Richard L. Roudebush VA Medical Center, Indianapolis, Indiana; †IU Center for Health Services & Outcomes Research, Regenstrief Institute, Inc., Indianapolis, Indiana; ‡Krannert School of Management and §Re- genstrief Center for Healthcare Engineering, Purdue University, West Lafayette, Indiana; ¶Division of Biostatistics and Division of General Medicine & Geriatrics, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana; **Department of Health Management and Policy, College of Public Health, University of Iowa, Iowa City, Iowa; ††Division of Infectious Diseases, Department of Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, Iowa. Supported by the Department of Veterans Affairs, Veterans Health Admin- istration, Medical Research Service VA Merit Epidemiology Grant (Doebbeling, Principal investigator). The work was also partially sup- ported by HSRD Center grant HFP 04-148. Dr. Flanagan was supported by a VA postdoctoral fellowship in medical informatics. This report presents the findings and conclusions of the authors; it does not necessarily represent the Department of Veterans Affairs (VA) or HSR&D. Reprints: Mindy Flanagan, PhD, HSR&D Center for Implementing Evi- dence-Based Practice, Roudebush VAMC, 1481 W. 10th Street (11-H), Indianapolis, IN 46202. E-mail: meflanag@iupui.edu. Copyright © 2007 by Lippincott Williams & Wilkins ISSN: 0025-7079/07/4506-0537 Medical Care • Volume 45, Number 6, June 2007 537