Validity of HealthImpact Diabetes Risk Model Among VA Patients

A recent study published online in the Journal of Managed Care & Specialty Pharmacy confirmed the validity of HealthImpact in stratifying patients according to diabetes risk, however, its practical utility to enhance diabetes screening in a real-world clinical environment may depend on the pattern of existing screening practices.

“HealthImpact is a novel algorithm using administrative health care data to stratify patients according to risk for incident diabetes,” Jonathan Linder, PharmD, of the Department of Pharmacy Service, VA Health Care System, and colleagues explained.

In order to assess the predictive validity of HealthImapct and explore its utility in diabetes screening within a nationally integrate health care system, Dr Linder and colleagues used National Veterans Health Administration data to create two cohorts—replication and utility. Within the replication cohort, patients without diagnosed diabetes as of October 1, 2012, were examined to determine if HealthImpact scores were significantly associated with both type 1 and 2 diabetes incidences within the subsequent 3 years. The utility cohort included patients without diagnosed diabetes, as of August 1, 2015, and assessed diabetes screening rates in the 2 years surrounding the study index date, stratified by HealthImpact scores.

According to the findings, the 3-year incidence of diabetes in the replication cohort (n = 3,287,240) was 9.1%. Of 100,617 patients with HealthImpact scores >90, a total of 30,028 developed diabetes, yielding a positive predictive value of 29.8%. Notably, these patients accounted for 9.9% of all incident diabetes cases. Further, sensitivity and negative predictive value improved with descending HealthImpact threshold scores, whereas specificity and positive predictive value declined. 

Within the utility cohort, of the 3,499,406 patients, 85.3% received either a blood glucose or hemoglobin A1c test during the 2-year observation period. Dr Linder and his colleagues also found that among 101,355 patients with a HealthImpact score >90, nearly all (98.3%) were screened, and 86.3% had an A1c test.

“Our independent analysis corroborates the validity of HealthImpact in stratifying patients according to diabetes risk,” Dr Linder and colleagues concluded. “However, its practical utility to enhance diabetes screening in a real-world clinical environment will be strongly dependent on the pattern and frequency of existing screening practices.”

—Julie Gould