Medical practice patterns and socio-economic factors may explain geographical variation of end-stage renal disease incidence

Background. Incidence rates of renal replacement therapy (RRT) for end-stage renal disease (ESRD) vary geographically not only between but within countries. This study uses data from the French REIN registry to quantify the extent to which socio-economic environment, health care supply and medical practice patterns such as early dialysis initiation or greater propensity to accept frail or elderly patients for dialysis, may explain spatial patterns of ESRD incidence in 85 French districts. Methods. The association between age-and sex-adjusted incidence rates of RRT in 2008-09 and 17 indicators was explored at the district level with geographically appropriate methods, before and after controlling for the effects of diabetes and the other significant indicators. Rate ratios (RR) and credible intervals (CI) were estimated for a 1-SD increase of each covariate. Results. Crude RRT incidence by district ranged from 85.8 to 225.5 per million inhabitants. The age-and sex-adjusted RRT incidence increased with the proportion of people unemployed (RR: 1.05, 95% CI 1.01-1.09), the population density (RR: 1.07, 95% CI 1.02-1.12) and the prevalence of diabetes (RR: 1.08, 95% CI 1.03-1.12). It also increased with the proportions of incident ESRD patients >85 years (RR: 1.02, 95% CI 0.99-1.06), of deaths within the first 3 months of RRT (RR: 1.03, 95% CI 1.0-1.06) and of nephrologists in private practice (RR: 1.05, 95% CI 1.01-1.08) and with the median estimated glomerular filtration rate (eGFR) at dialysis initiation (RR: 1.06, 95% CI 1.02-1.09). Conclusion. This study confirms that socio-economic factors and diabetes explain substantial between-area variations in RRT incidence and highlights the variability of practice patterns, especially decisions about RRT and their potential impact on incidence.