• Kenneth J Smith, Shihchen Kuo, Janice C Zgibor, Kathleen M McTigue, Rachel Hess, Tina Bhargava, and Cindy L Bryce.
• Section of Decision Sciences, School of Medicine, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: smithkj2@upmc.edu.
• Prev Med. 2016 Jun 1; 87: 103-109.

ObjectiveTo assess the cost-effectiveness of an online adaptation of the diabetes prevention program (ODPP) lifestyle intervention.MethodsODPP was a before-after evaluation of a weight loss intervention comprising 16 weekly and 8 monthly lessons, incorporating behavioral tools and regular, brief, web-based individualized counseling in an overweight/obese cohort (mean age 52, 76% female, 92% white, 28% with diabetes). A Markov model was developed to estimate ODPP cost effectiveness compared with usual care (UC) to reduce metabolic risk over 10years. Intervention costs and weight change outcomes were obtained from the study; other model parameters were based on published reports. In the model, diabetes risk was a function of weight change with and without the intervention.ResultsCompared to UC, the ODPP in our cohort cost $14,351 and $29,331 per quality-adjusted life-year (QALY) gained from the health care system and societal perspectives, respectively. In a hypothetical cohort without diabetes, the ODPP cost $7777 and $18,263 per QALY gained, respectively. Results were robust in sensitivity analyses, but enrolling cohorts with lower annual risk of developing diabetes (≤1.8%), enrolling fewer participants (≤15), or increasing the hourly cost (≥$91.20) or annual per-participant time (≥1.45h) required for technical support could increase ODPP cost to >$20,000 per QALY gained. In probabilistic sensitivity analyses, ODPP was cost-effective in 20-58% of model iterations using an acceptability threshold of $20,000, 73-92% at $50,000, and 95-99% at $100,000 per QALY gained.ConclusionsThe ODPP may offer an economical approach to combating overweight and obesity.Copyright © 2016 Elsevier Inc. All rights reserved.

20 keywords...

hide…