Risk for Incident Renal Dialysis in a Managed Care Population

Background The influence of poorly controlled blood pressure (BP) on progression to end-stage renal disease typically has been studied in qualified populations. Few observational studies examined the influence of change in BP. Methods: We conducted a retrospective observational study of a cohort of adults 18 years or older with glomerular filtration rates of 15 to 89 mL/min/1.73 m(2) (0.25 to 1.48 mL/s) in 2001 (N = 16,299) in a managed care organization. The cohort generally was representative of many urban areas in the United States. The principal outcome was incident renal dialysis in 2002 (N = 73). BP was measured in 1999 and 2001. Risk for incident dialysis therapy was estimated in the entire cohort by using Cox proportional hazards regression and in a subset of 4 randomly selected controls (n = 292) matched (for 2001 glomerular filtration rate, age, and sex) with incident dialysis cases by using conditional logistic regression. Results: Within this cohort, dialysis incidence was 4.5 cases/1,000 adults. For incident dialysis cases, mean time to dialysis therapy in 2002 (from December 31, 2001) was 167 days. Greater levels of systolic or diastolic BP in 1999 were associated significantly with greater risk for incident dialysis treatment in 2002 (controlling for age, sex, coronary artery disease, diabetes, and tobacco use). Decreased systolic or diastolic BP from 1999 to 2001 was associated with significantly reduced risk for incident dialysis treatment in 2002. Conclusion: Improving BP control in adults with early or advanced kidney disease and moderately to severely elevated BP should continue to be a focus of care management programs seeking to decrease the risk for incident dialysis treatment.