2005 HSR&D National Meeting Abstract
3039 — Nephrologist Care Reduced One-Year Mortality for Diabetic Patients with Chronic Kidney Diseases
Tseng C (New Jersey Health Care System)
Tiwari A (New Jersey Health Care System)
Kern E (Cleveland VAMC)
Miller D (Bedford Center for Health Quality, Outcomes and Economic)
Pogach L (New Jersey Health Care System)
Our objective was to evaluate the provision of nephrologist care among diabetic veterans with chronic kidney diseases (CKD) and its association with mortality when controlling for age, gender, race, medical and psychiatric comorbidities, and ACEI/ARB/statin usage.
This was a retrospective study design. We first identified 561,088 veterans with diabetes in Fiscal year (FY) 1999 who had no end stage renal disease or dialysis in FY 1997 or 1998 and were alive as of 9/30/1998. We used the 4-variable MDRD formula to determine eGFR(adjusted for age, sex, race, and creatinine levels) The study population was a total of 261,507 diabetic veterans that could definitely be classified into one of the five CKD stages according to K/DOQI criteria and had no dialysis prior to death. Death outcome was obtained within 365 days from the initial eGFR during the study period. Multiple logistic regression was used.
There were 69.0% diabetic veterans in the CKD 1 and 2 stages, 28.8% in CKD 3, and 2.4% in CKD 4 and 5 stages. The overall one year mortality rate was 3.6%. about 6.4% of the population had any nephrologist visit after their initial eGFR. Nephrologist care was more prevalent in more severe CKD groups: <10% for CKD 1 and 2, 13.5% for CKD 3, and ~55% for CKD 4 and 5. Our multiple logistic regression models showed that older age, men, non-white racial groups, and more comorbidities were all significantly (p<0.001) associated with increased mortality. Having nephrologist care reduced mortality, but only for patients in the later CKD stages (CKD 3: Odds ratio (OR)=0.62, 95% confidence interval (CI)=(0.56, 0.69); CKD 4 and 5: OR=0.39, CI=(0.33, 0.47)). Results were similar when medication was also entered into the model.
A low percentage of diabetic veterans with CKD had any nephrologist visit after their initial eGFR. Nephrologist care was associated with reduced one year mortality among diabetic veterans in devastating CKD stages.
Our data suggest compliance with proposed “quality of care measures” according to the K/DOQI and VA-DOD Guidelines needs to be improved for diabetic veterans with chronic kidney diseases.