Improving the quality of quality improvement in health care: case studies of provider office practices in the use of assisted office systems analysis within a quality improvement project

Quality improvement (QI) is an important activity expected of the U.S. health care system. A method of performing QI that meets the needs of small health care organizations has not been well defined. Office practices have limited time to invest in QI. This multiple case study explored a method of QI based on workflow analysis, AA3, and measured its acceptance by staff and its effect both on office processes and clinical outcomes. The study used a prospective, mixed methods approach to explore three Vermont primary care practices engaged in different QI projects: improving pediatric lead screen rates, integrating behavioral health services, and increasing the percentage of "After Visit Summaries" provided to patients. Data were collected from qualitative and quantitative sources. Each case was analyzed using non-equivalent dependent variable pattern matching over time along with observations about dependent and independent variables. The results of cross case analysis revealed that median survey scores show higher than neutral responses for all sites regarding acceptance and clinical outcomes; two of three sites also produced high median scores for process outcomes. Team members rated AA3 method very highly as a method of QI. However, the time series studies, measuring changes in survey scores from pre- to post-periods, were not significant. Qualitative analysis revealed other insights: the importance of "organizational slack" to enable staff to engage in QI, the impact of the "big picture" in translating change to the work place, and the value of reinforcing changes that benefit the staff as well as the patient. All of these features are available via AA3, combining process and clinical outcomes improvement in a single QI method. These findings are consistent with management theory regarding change. The results of this study indicated that AA3 may have a positive effect on the outcomes of QI and is worth further study. It provides support for grant-based research, both to improve the AA3 method and to explore how it may help translate policy into practice effectively. It can also provide the basis for future education of clinicians and health care leaders in the use of QI as part of clinical practice