Variations in surgical rates in Quebec: does access to teaching hospitals make a difference?

Abstract

Objective: To document the geographic variations in the rates of 10 common surgical procedures in Quebec and to examine the relation between surgical rates and level of access to teaching hospitals.

Design: Population-based rates standardized for age and sex were calculated from 1985-88 data from MED-ECHO (the provincial hospital discharge database) for each of the 32 community health districts (départements de santé communautaire [DSCs]) in Quebec. Variation across DSCs was analysed with the use of the ratio of the highest to the lowest rate, the coefficient of variation and the systematic component of variation. On the basis of an urbanization index designed by Statistics Canada, DSCs were classified as having a low, medium, high or very high level of access to teaching hospitals.

Patients: All Quebec residents except those whose DSC of residence could not be traced (accounting for no more than 1.7% of patients for any study procedure) and aboriginal people from northern Quebec. SURGICAL PROCEDURES STUDIED: Appendectomy, cesarean section, cholecystectomy, coronary artery bypass grafting (CABG), hysterectomy, inguinal hernia repair, prostatectomy, tonsillectomy with or without adenoidectomy, total hip replacement and varicose vein stripping.

Results: There was considerable systematic variation in the surgical rates for all 10 procedures. Cesarean section, the rates of which varied the least, still exhibited almost a twofold variation between the highest and lowest rates. The rates of tonsillectomy varied the most, by a factor of more than five. With a few exceptions there was relatively little relation between the rates of the procedures within the DSCs. The rates of appendectomy, cholecystectomy, hysterectomy and tonsillectomy tended to be lower in the DSCs containing teaching hospitals, whereas the reverse was true for CABG. For the other procedures the relation between the rates and the level of access to teaching hospitals was nonsignificant.

Conclusions: The overall consistency of the observed variations with those of previous studies invites a closer look at the practice patterns and the scientific basis of the clinical decisions associated with procedures showing high variations in rates. The lack of a systematic relation between the surgical rates and the level of access to teaching hospitals challenges the belief that remote regions are underserved with respect to the procedures studied.