Supply sensitive services in Swiss ambulatory care: An analysis of basic health insurance records for 2003-2007

Abstract

Abstract Background Swiss ambulatory care is characterized by independent, and primarily practice-based, physicians, receiving fee for service reimbursement. This study analyses supply sensitive services using ambulatory care claims data from mandatory health insurance. A first research question was aimed at the hypothesis that physicians with large patient lists decrease their intensity of services and bill less per patient to health insurance, and vice versa: physicians with smaller patient lists compensate for the lack of patients with additional visits and services. A second research question relates to the fact that several cantons are allowing physicians to directly dispense drugs to patients ('self-dispensation') whereas other cantons restrict such direct sales to emergencies only. This second question was based on the assumption that patterns of rescheduling patients for consultations may differ across channels of dispensing prescription drugs and therefore the hypothesis of different consultation costs in this context was investigated. Methods Complete claims data paid for by mandatory health insurance of all Swiss physicians in own practices were analyzed for the years 2003-2007. Medical specialties were pooled into six main provider types in ambulatory care: primary care, pediatrics, gynecology & obstetrics, psychiatrists, invasive and non-invasive specialists. For each provider type, regression models at the physician level were used to analyze the relationship between the number of patients treated and the total sum of treatment cost reimbursed by mandatory health insurance. Results The results show non-proportional relationships between patient numbers and total sum of treatment cost for all provider types involved implying that treatment costs per patient increase with higher practice size. The related additional costs to the health system are substantial. Regions with self-dispensation had lowest treatment cost for primary care, gynecology, pediatrics and for psychiatrists whereas "prescription only" areas had lowest cost for specialists with non-invasive and invasive activities. Conclusions The results indicate that payment methods for services and for prescription drugs are associated with variations in treatment cost that are unlikely warranted by different medical needs of patients alone. Promoting physician accountability of care by linking reimbursements to quality, not quantity, of services are important policy measures to be considered for health care in Switzerland.