Abstract: TH-PO546

Using Kidney Failure Risk Scores to Identify Veterans Needing CKD Care

Session Information

Category: Chronic Kidney Disease (Non-Dialysis)

  • 307 CKD: Health Services, Disparities, Prevention

Authors

  • Cameron, C. Blake, Duke University Medical Center, Durham, North Carolina, United States
  • Boggan, Joel, Durham VA Health Care System, Durham, North Carolina, United States
  • Gurley, Susan B., Duke University Medical Center, Durham, North Carolina, United States
  • Atkins, Richard Matthew, Durham VA Health Care System, Durham, North Carolina, United States
Background

Both over- and under-referral to nephrology threaten the quality and efficiency of CKD care. Optimizing veterans' nephrology referrals may improve their outcomes. Few studies have evaluated patterns of nephrology referral among veterans as a function of CKD progression risk.

Methods

Using the Veterans Health Administration Clinical Data Warehouse, we identified all non-ESRD individuals who received primary care at the Durham VA Health Care System between Dec 2014 and Jun 2017 and had >1 outpatient serum creatinine measurement during that time. For each individual, we identified the assigned primary care provider (PCP); tabulated nephrology/CKD clinic visits; and calculated the Kidney Failure Risk Equation (KFRE), an internationally validated predictor of 5-year ESRD risk utilizing age, sex, CKD-EPI eGFR, and optionally, urine albumin-to-creatinine ratio. We stratified the population by KFRE risk (low <5%, intermediate 5-15%, and high >15%), by nephrology referral status and by assigned PCP. We performed descriptive analyses.

Results

Overall, 48,700 unique, non-ESRD individuals with at least one creatinine measurement received care from 139 PCPs. Only 32% (n=359/1,116) and 58% (n=503/865) of individuals at intermediate and high risk for CKD progression respectively had been seen in nephrology clinic. Conversely, among the 1,816 individuals seen in nephrology clinic, 53% (n=957) were at low risk [Figure 1A]. Nephrology referral rates for high-risk patients varied widely across PCPs (mean 58% [s.d. 20%]) [Figure 1B].

Conclusion

Within a single integrated medical center, nephrology referral rates were not aligned with clinical risk. More than 40% of individuals with high-risk CKD had not received nephrology care. Conversely, approximately half of individuals seen in nephrology/CKD clinic were at low risk of progression to ESRD and potentially could have avoided referral. Substantial provider-to-provider variation in nephrology referral rates exists. Identifying the sources of variation will be critical to developing decision support tools and models of care that better align the provision of CKD care with clinical risk.

Funding

  • Veterans Affairs Support