P. Diehr et al., SMALL AREA VARIATION ANALYSIS - METHODS FOR COMPARING SEVERAL DIAGNOSIS-RELATED GROUPS, Medical care, 31(5), 1993, pp. 190000045-190000053
In small-area variation analysis, the variation of health care utiliza
tion rates, e.g., admission rates, among small areas is calculated. Fr
equently, the variation of one diagnosis, diagnosis-related group (DRG
), or procedure is compared with the variation of another. Unfortunate
ly, the methods generally used to make these comparisons are not consi
stent. They diff er on whether they 1) adjust for the prevalence of th
e DRGs, 2) distinguish between variation among areas and variation wit
hin areas, 3) weight all areas equally, and 4) adjust for multiple adm
issions per person. None has an associated confidence interval. These
discrepancies occur in part because there is no statistical model of s
mall area variation. Without such a model, it is not known how to meas
ure variation, and thus, it is not known how to compare different DRGs
. Here, the authors use data on 473 DRGs from 28 counties in Washingto
n state to study the nature of variability. The variation was higher f
or the more prevalent DRGs, suggesting that adjusting for prevalence m
ay be reasonable. The true coefficient of variation appears to be a ''
natural'' measure of variation, but the usual small area variation sta
tistics do not provide good estimates of the true coefficient of varia
tion. A new estimate is proposed that can be used to compare and test
the variability of several DRGs.