Cross-sectional and longitudinal associations among healthcare costs and deficit accumulation

Abstract

Background

Type 2 diabetes mellitus and overweight/obesity increase healthcare costs. Both are also associated with accelerated aging. However, the contributions of this accelerated aging to increased healthcare costs are unknown.

Methods

We use data from a 8-year longitudinal cohort followed at 16 U.S. clinical research sites. Participants were adults aged 45–76 years with established type 2 diabetes and overweight or obesity who had enrolled in the Action for Health in Diabetes clinical trial. They were randomly (1:1) assigned to either an intensive lifestyle intervention focused on weight loss versus a comparator of diabetes support and education. A validated deficit accumulation frailty index (FI) was used to characterize biological aging. Discounted annual healthcare costs were estimated using national databases in 2012 dollars. Descriptive characteristics were collected by trained and certified staff.

Results

Compared with participants in the lowest tertile (least frail) of baseline FI, those in the highest tertile (most frail) at Year 1 averaged $714 (42%) higher medication costs, $244 (22%) higher outpatient costs, and $800 (134%) higher hospitalization costs (p < 0.001). At Years 4 and 8, relatively greater increases in FI (third vs. first tertile) were associated with an approximate doubling of total healthcare costs (p < 0.001). Mean (95% confidence interval) relative annual savings in healthcare costs associated with randomization to the intensive lifestyle intervention were $437 ($195, $579) per year during Years 1–4 and $461 ($232, $690) per year during Years 1–8. These were attenuated and the 95% confidence interval no longer excluded $0 after adjustment for the annual FI differences from baseline.

Conclusions

Deficit accumulation frailty tracks well with healthcare costs among adults with type 2 diabetes and overweight or obesity. It may serve as a useful marker to project healthcare needs and as an intermediate outcome in clinical trials.