Small area analysis of lumbar spine surgery in South Australia

The hospital separation records for 1987 in the health planning regions of South Australia were reviewed using a selection algorithm to identify all hospitalizations involving a lumbar spine surgery (LSS) for low back or leg pain. Among 16 health planning regions (two additional regions were excluded from the analysis because of the low number of observations) the LSS rate varied almost four-fold, from 25 to 92/100,000, with a mean of 55/100,000. The effect of 24 socioeconomic and health care supply characteristic variables upon observed differences in rates were tested. The unemployment rate was the only significant variable in the analysis, explaining 11% of the variation in the surgery rates for the 16 regions. This finding is in agreement with studies from other countries that suggest that characteristics of small areas do not substantially predict the rates of elective surgical procedures. The decision-making processes of surgeons and their patients remain poorly defined; the contributions to the rate of lumbar spine surgery by the health care delivery system, physician behaviours or patient expectations are not yet identified.