Fatality and hospitalization in hemodialysis patients in a health plan

Abstract

OBJECTIVE: To analyze clinical outcomes of patients on hemodialysis linked to health care plan provider. METHODS: Cohort study of hemodialysis events in Belo Horizonte, MG, Southeastern Brazil, between 2004 and 2008, based on records from health care plan provider databases. The independent variables were: sex, age, time between first appointment with nephrologist and starting hemodialysis, type of first vascular access, diabetes mellitus, length of time spent in hospital/year of treatment and death. Dependent variables: time between starting hemodialysis and death and length of time spent in hospital/year of treatment >; 7.5 days. Statistical analysis was carried out using Pearson's Chi-squared test in the univariate analysis for the outcomes 'death' and 'length of time spent in hospital/year of treatment'; the Kaplan-Meier method was used to analyze survival; the Cox model and Poisson regression were used for risk of death and chance of length of time spent in hospital/year of treatment >; 7.5 days. The Business Intelligence tool and Stata(r) 10.0 software were used to extract data. RESULTS: There were 311 patients on hemodialysis included in the study, with a mean age of 62 (sd 16.6 years), of whom 55.5% were male. Prevalence increased 160% during the period in question. Survival analysis showed a higher mortality among older patients, patients that did not consult a nephrologists, those whose first vascular access was using a temporary catheter, those with diabetes mellitus, those admitted to hospital within a month of beginning hemodialysis. The Cox model showed that a higher risk of death was associated with age, diabetes mellitus, not consulting a nephrologists and those that were hospitalized within a month of beginning hemodialysis. Greater length of time spent in hospital/year of treatment was not associated with sex or diabetes. According to Poisson regression, the variables were not significant. CONCLUSIONS: Assessment by a specialist before starting hemodialysis decreases the risk of death in cases of chronic kidney disease, whereas the presence of diabetes and being hospitalized within a month of beginning hemodialysis are markers of risk of death.