This project proposes to study prospectively the determinants of referral among adult patients with one of three common chronic conditions-- congestive heart failure (CHF), benign prostatic hyperplasia (BPH), and peptic ulcer disease (PUD). Patients will be selected from the enrollment files of ten California physician groups which have a variety of incentive arrangements. The principal objective is to study the effects of financial arrangements and utilization management techniques used by the ten physician groups on referral decisions to specialists. Three hundred primary care physicians working in the ten physician groups will participate in the study. Groups will be selected to ensure they represent a gradient of incentive strategies. A secondary objective of this project is to evaluate the possible association between referral patterns and quality of care. Study investigators will examine each physician group to determine the nature of its capitation arrangements with insurers, physician reimbursement arrangements within the group, group composition (primary care versus multispecialty), and nonfinancial utilization management techniques within the group. They will also survey physicians to assess their demographic characteristics, specialty training, patient panel size, knowledge of and comfort with the three conditions of interest, opinions about the availability and quality of specialists in the three disease areas in their practice setting, and awareness of practice and referral guidelines as well as reimbursement arrangements related to referrals. Patients of these physicians with one of the three target diseases will be identified, and 2,100 patients with each disease will be randomly selected. They will be surveyed at baseline and again at one year. Data to be collected include: quality of care, including change in global and disease-specific functional status; receipt of various recommended process measures; and patient satisfaction. The specific questions to be addressed are: 1) Which patient, physician, and practice characteristics affect the likelihood of referral to specialists for each condition in managed care settings? Do these factors vary across conditions? 2) After adjusting for these characteristics, does variation in managed care physician reimbursement arrangements affect the likelihood of referral? 3) Do innovations such as referral guidelines or structured, immediate telephone consultations affect referral rates for these conditions? 4) Are referrals related to more frequent performance of appropriate diagnostic and treatment interventions, better functional status, or greater patient satisfaction? If so, are the associations independent of the factors that lead to or predict referral?
