Small area variation analysis: a tool for primary care research.

Small area variation analysis is a research tool used by health services researchers to describe how rates of health care use and events vary over well-defined geographic areas. Significant variation has been shown to exist in the rates of hospitalization for chronic obstructive lung disease, pneumonia, hypertension, and in surgical procedures, such as hysterectomy, cholecystectomy, and tonsillectomy. Potential sources of variation include differences in underlying morbidity, access to care, physician judgment, quality of care delivered, patient demand for services, and random variation. Small area variation studies have been used to determine if significant variation exists across geographic areas and to describe relationships between the observed variation and potential causal factors. Methodologic concerns include the definition of small areas, defining the at-risk population within each small area, sample size, case mix adjustments, and stability of rates over time. The use of small area analysis in primary care will require definition of appropriate small areas for ambulatory care, description of the variation in ambulatory events across small areas, development of appropriate measures for ambulatory case mix, and development of appropriate tools to measure the outcomes of ambulatory care.