Evaluating an integrated approach to clinical quality improvement - Clinical guidelines, quality measurement, and supportive system design

BACKGROUND. Implementing clinical practice guidelines to change patient out comes presents a challenge. Studies of single interventions focused on chan ging provider behavior demonstrate modest effects, suggesting that effectiv e guideline implementation requires a multifaceted approach. Traditional bi omedical research designs are not well suited to evaluating systems interve ntions. OBJECTIVES. RAND and the Army Medical Department collaborated to develop an d evaluate a system for implementing guidelines and documenting their effec ts on patient care. RESEARCH DESIGN. The evaluation design blended quality improvement, case st udy, and epidemiologic methods. A formative evaluation of implementation pr ocess and an outcome evaluation of patient impact were combined. SUBJECTS. Guidelines were implemented in 3 successive demonstrations targeting low ba ck pain, asthma, and diabetes. This paper reports on the first wave of 4 fa cilities implementing a low back pain guideline. METHODS. Organizational climate and culture, motivation, leadership commitm ent, and resources were assessed. Selected indicators of processes and outc omes of care were compared before, during, and after guideline implementati on at the demonstration facilities and at comparison facilities. Logistic r egression analysis was used to test for guideline effects on patient care. RESULTS. Process evaluation documented varied approaches to quality improve ment across sites. Outcome evaluation revealed a significant downward trend in the percentage of acute low back pain patients referred to physical the rapy or chiropractic care (10.7% to 7.2%) at demonstration sites and no suc h trend at control sites. CONCLUSIONS. Preliminary results suggest the power of this design to stimul ate improvements in guideline implementation while retaining the power to e valuate rigorously effects on patient care.