Racial Differences in the Effectiveness of a Multifactorial Telehealth Intervention to Slow Diabetic Kidney Disease

Abstract

Background: African Americans are significantly more likely than non-African Americans to have diabetes, chronic kidney disease, and uncontrolled hypertension, increasing their risk for kidney function decline.

Objective: The objective of this study was to compare how African Americans and non-African Americans with diabetes responded to a multifactorial telehealth intervention designed to slow kidney function decline.

Research design: Secondary analysis of a randomized trial. Primary care patients (N=281, 56% African American) were allocated to either: (1) a multifactorial, pharmacist-delivered phone-based telehealth intervention focused on behavioral and medication management of diabetic kidney disease; or (2) an education control.

Measures: The primary study outcome was change in estimated glomerular filtration rate (eGFR). Linear mixed models were used to explore the moderating effect of race on the relationship between study arm and eGFR decline over time; the mean annual rate of eGFR decline was estimated by race and study arm.

Results: Findings demonstrated a differential intervention effect on kidney function over time by race (Pinteraction=0.005). Among African Americans, the intervention arm had significantly greater preservation of eGFR over time than the control arm (difference in the annual rate of eGFR decline=1.5 mL/min/1.73 m; 95% confidence interval: 0.04, 3.02). For non-African Americans, the intervention arm had a faster decline in eGFR over time than the control arm (difference in the annual rate of eGFR decline=-1.7 mL/min/1.73 m; 95% confidence interval: -3.3, -0.02).

Conclusion: A multifactorial, pharmacist-delivered telehealth intervention for diabetic kidney disease may be more effective for slowing eGFR decline among African Americans than non-African Americans.

Figure 1.

Observed and predicted mean eGFR over time. Abbreviation: eGFR: Estimated Glomerular Filtration Rate (ml/min/1.73m²). a) Observed mean eGFR over time, b) Primary Analysis: Model Predicted eGFR over Time (Continuous), c) Secondary Analysis: Model Predicted eGFR over Time (Categorical).