RESUMO
BACKGROUND: The recent need for information has prompted this collaboration between health system epidemiologists (Basilicata) and clinicians to compare models of 'local' epidemiology in the management of diseases. The referral of patients to a nephrologist represents a working hypothesis of research- intervention. METHODS: Analysis of renal registry (RR) and administrative databases (hospital discharge abstracts/HDA, ambulatory);ad hoc surveys. RESULTS: Patients on dialysis between 1994 and 1998 are 594, cumulative deaths are 190 (32%). Males and the elderly (age = 65 years) are associated with more than 50% and threefold increase in relative risk of death, and with a diabetic nephropathy of 60% vs other renal diseases. Of 570 patients alive in 1996, 442 are linked with 2,628 HAD. Comorbid conditions are underreported in the RR (the Charlson index has been computed using HDA). Of 66 new dialysis cases, 31 are referred to a nephrologist only 6 months before the start of dialysis (47%) (22% diabetics). Patients discharged with chronic nephropathies (CN) and diabetes are 21% of CN patients (5% of diabetics). Of 100 patients with pre-end stage renal disease and diabetes, only 11-14 are discharged from the nephrology ward. At the local level, 3 out of 4 patients with serum creatinine higher than 1.5 mg/dl are not referred to a nephrologist. The prevalence of CN may vary from 0.4% to more than 1%. CONCLUSIONS: While an improvement in health databases in the regions is underway, collaboration studies are essential for planning specific interventions for prevention and management of diabetic nephropathy to improve the use of resources in nephrology.