Variation in Michigan hospital use rates: do physician and hospital characteristics provide the explanation?

Previous small area analysis studies have shown that hospital admission rates (total, medical and surgical) vary among hospital service areas. Using 1983 Michigan hospital inpatient data from 53 nonmetropolitan Detroit lower peninsula hospital service areas, one physician characteristic and 13 hospital characteristics (in the categories of resource supply, services offered and organization) were tested for their association with and explanation of 14 hospital use rates. Registered nurses per bed and the weighted proportion of board certified physicians to total physicians were inversely related to and offered significant contribution to the explanation of the variation in total use rates and in four medical causes for admission rates (circulatory, respiratory, digestive and genito-urinary). Physician and hospital variables provided significant explanation for six of the seven surgical procedure rates tested (appendectomy, hemorrhoidectomy, cholecystectomy, inguinal hernia repair, prostatectomy and hysterectomy). Four causative factors derived from the characteristics studied were postulated to influence the hospital use rates. The first factor was the small rural nature of the average high use hospital service area. High use areas had a lower proportion of board certified physicians and fewer RNs per bed, beds per hospital, and house staff per 10,000 population than did low use areas. Another factor was the inequality in the distribution of high technology diagnostic services. High use hospital service areas had fewer diagnostic services than did low use areas. The third factor was the inequality in the rural hospital environment produced by the presence or absence of medical education programs. The fourth factor was the impact of the definition and size of a hospital service area. Current small area analysis methodology assigns every small area to a hospital service area, no matter what the probability of the population using the hospital(s) within the service area. This research questions that methodology, suggests the need for hospital service area definitions based upon the specific diagnosis or procedure being studied and postulates that some rural hospital distance decay curves may turn upward at farther distances when the immediate availability of treatment is too critical to allow patients to return to distant residences.