Impact of physician characteristics and specialization on resource use within diagnostic related groups

Financial pressures generated under the Medicare Prospective Payment System and more recently managed care, continue to provide incentives for hospitals to use resources efficiently. Physicians determine majority of the resources in the hospital. However, research on the impact of physician factors on resource use has been limited. This study uses a 5 percent national sample of Medicare inpatient and physician claims to examine the effect of physician characteristics and physician specialty on resource use within DRGs. Hospital and physician resource use measures include length of stay, hospital adjusted charges, physician Relative Value Units and physician allowed charges. The study sample consists of 3158 patients admitted between March and October 1992 in DRGs 320 and 321 for kidney and urinary tract infections. The attending physician for each admission was identified using an algorithm based on the maximum number evaluation and management services provided during the hospital stay. Claims data were linked to Medicare's physician database to determine the number of years since graduation from medical school, whether the physician belonged to a group practice, whether the physician accepted Medicare assignment and the physician's specialty. Specialty included general practice, family practice, internal medicine and urology. Severity and comorbidity were assessed using the All-Patient Refined DRGs and the Charlson Index respectively. Patient demographics, hospital characteristics and geographic region were used as control variables in the analyses.;There are three major findings: (1) Physician characteristics explain residual variation in resource use within DRGs after controlling for patient health status. The contribution is greater for physician resource use than hospital resource use. (2) Physician specialty has an independent effect on resource use. General practitioners use more hospital resources and less physician resources, family practitioners use fewer resources than internists, and there are no significant differences in inpatient resource use between primary care physicians and urologists. (3) Specialty did not have an effect on use of physician services within 30 days of discharge, but patients of urologists had a greater probability of readmission within 60 days.;This study provides evidence that physician have discernible practice styles by specialty that are not explained by patients' clinical needs.