IS HIGHER-PRICED INSULIN ASSOCIATED WITH BETTER ECONOMIC AND HEALTH OUTCOMES? A STUDY FROM A VALUE-BASED PAYMENT PERSPECTIVE.
Author(s)
Shao H1, Michael L2, Thomas N3, Benoit S2, Cheng Y2, Gregg E2, Zhang P2
1CDC, Dunwoody, GA, USA, 2CDC, Atlanta, GA, USA, 3USF Center of Excellence in Maternal and Child Health, Tampa, FL, USA
OBJECTIVES The overall Insulin prices have been increasing recently, accompanied by emergence of newer types of insulin with vastly different costs. The health and economic impact of these cost differences are unknown. We examined if the highest and lowest tiers of insulin costs are associated with better or worse health and cost outcomes among adults with type 2 diabetes. METHODS The MarketScan (2006-2016) commercial claims database was used to identify a population with type 2 diabetes enrolled in fee-for-service plans, who newly initiated insulin and stayed continuously on that product at least two years. The study sample was grouped into three categories based on payment level for a one-year insulin supply (high>$8000, middle=$4000 -$8000, and low=<$4000), decided based on the clustering pattern of the users. Twenty-one variables covering health service utilization, medical expenditure, macrovascular/microvascular complications, and adverse events were used to measure the effects of insulin products from different cost tier. Generalized linear models with gamma/binomial distribution were used, controlling for demographics, disease history, individual-level fixed effect, propensity-score weights, and Simes correction for multiple hypothesis testing. RESULTS We identified 2,922 individuals (mean age: 50.3, 47% female). Compared with the middle-cost group, the low-cost group had a lower annual drug cost (-$1482) and out-of-pocket payment (-$264), but a higher risk of hypoglycemia (OR=1.61), Ketoacidosis (OR=1.68), and stroke (OR=5.5). In contrast, the high-cost group had a higher annual drug cost ($900), but lower risk of insulin discontinuation (OR=0.72), and annual physician visits (-0.6), compared with the middle-cost group (all p<0.05). Differences in other outcomes between groups were not statistically significant. CONCLUSIONS Tradeoffs between drug spending and clinical benefits were observed when comparing low-cost insulin to middle-cost insulin. However, no extra benefit on improving clinical outcomes were observed among users under high-cost tier, when compares to insulin users with middle-cost tier.
Conference/Value in Health Info
2019-05, ISPOR 2019, New Orleans, LA, USA
Value in Health, Volume 22, Issue S1 (2019 May)
Code
PDB53
Topic
Clinical Outcomes, Economic Evaluation
Topic Subcategory
Comparative Effectiveness or Efficacy, Cost-comparison, Effectiveness, Utility, Benefit Analysis, Performance-based Outcomes
Disease
Diabetes/Endocrine/Metabolic Disorders