Cost-effectiveness of a lifestyle intervention in high-risk individuals for diabetes in a low- and middle-income setting: Trial-based analysis of the Kerala Diabetes Prevention Program.

Background: Data on the cost-effectiveness of lifestyle-based diabetes prevention programs are mostly from high-income countries, which cannot be extrapolated to low- and middle-income countries. We performed a trial-based cost-effectiveness analysis of a lifestyle intervention targeted at preventing diabetes in India.

Methods: The Kerala Diabetes Prevention Program was a cluster-randomized controlled trial of 1007 individuals conducted in 60 polling areas (electoral divisions) in Kerala state. Participants (30-60 years) were those with a high diabetes risk score and without diabetes on an oral glucose tolerance test. The intervention group received a 12-month peer-support lifestyle intervention involving 15 group sessions delivered in community settings by trained lay peer leaders. There were also linked community activities to sustain behavior change. The control group received a booklet on lifestyle change. Costs were estimated from the health system and societal perspectives, with 2018 as the reference year. Effectiveness was measured in terms of the number of diabetes cases prevented and quality-adjusted life years (QALYs). Three times India's gross domestic product per capita (US$6108) was used as the cost-effectiveness threshold. The analyses were conducted with a 2-year time horizon. Costs and effects were discounted at 3% per annum. One-way and multi-way sensitivity analyses were performed.

Results: Baseline characteristics were similar in the two study groups. Over 2 years, the intervention resulted in an incremental health system cost of US$2.0 (intervention group: US$303.6; control group: US$301.6), incremental societal cost of US$6.2 (intervention group: US$367.8; control group: US$361.5), absolute risk reduction of 2.1%, and incremental QALYs of 0.04 per person. From a health system perspective, the cost per diabetes case prevented was US$95.2, and the cost per QALY gained was US$50.0. From a societal perspective, the corresponding figures were US$295.1 and US$155.0. For the number of diabetes cases prevented, the probability for the intervention to be cost-effective was 84.0% and 83.1% from the health system and societal perspectives, respectively. The corresponding figures for QALY gained were 99.1% and 97.8%. The results were robust to discounting and sensitivity analyses.

Conclusions: A community-based peer-support lifestyle intervention was cost-effective in individuals at high risk of developing diabetes in India over 2 years.

Trial Registration: The trial was registered with Australia and New Zealand Clinical Trials Registry ( ACTRN12611000262909 ). Registered 10 March 2011.