Open Access Research article

An economic evaluation for prevention of diabetes mellitus in a developing country: a modelling study

Xiaoqian Liu1, Changping Li1, Hui Gong1, Zhuang Cui1, Linlin Fan1, Wenhua Yu1, Cui Zhang2 and Jun Ma1*

Author Affiliations

1 Department of Health Statistics, College of Public Health, Tianjin Medical University, No. 22 Qixiangtai Road, Heping District, Tianjin 300070, P.R. China

2 Institute of New Catalytic Materials Science, College of Chemistry, Nankai University, Tianjin 300071, P.R. China

For all author emails, please log on.

BMC Public Health 2013, 13:729  doi:10.1186/1471-2458-13-729

Published: 7 August 2013



The serious consequences of diabetes mellitus, and the subsequent economic burden, call for urgent preventative action in developing countries. This study explores the clinical and economic outcomes of strategies that could potentially prevent diabetes based on Chinese circumstances. It aims to provide indicators for the long-term allocation of healthcare resources for authorities in developing countries.


A representative sample of Chinese adults was used to create a simulated population of 20,000 people aged 25 years and above. The hybrid decision tree Markov model was developed to compare the long-term clinical and economic outcomes of four simulated diabetes prevention strategies with a control group, where no prevention applied. These preventive strategies were the following: (i) one-off screening for undiagnosed diabetes and impaired glucose tolerance (IGT), with lifestyle interventions on diet, (ii) on exercise, (iii) on diet combined exercise (duo-intervention) respectively in those with IGT, and (iv) one-off screening alone. Independent age-specific models were simulated based on diverse incidences of diabetes, mortalities and health utilities. The reported outcomes were the following: the remaining survival years, the quality-adjusted life years (QALYs) per diabetes or IGT subjects, societal costs per simulated subject and the comparisons between preventions and control over 40 years. Sensitivity analyses were performed based on variations of all assumptions, in addition to the performance and the compliance of screening.


Compared with the control group, all simulated screening programmes prolonged life expectancy at the initiation ages of 25 and 40 years, postponed the onset of diabetes and increased QALYs at every initiation age. Along with an assumption of six years intervention, prevention programmes were associated with cost-saving compared with the control group, especially in the population aged 25 years. The savings were at least US$2017 per subject, but no statistically significant difference was observed among the intervention strategies within each age groups. The cost savings were reduced when screening was affected by poor performance and noncompliance.


Developing countries have few effective strategies to manage the prevention of diabetes. One-off screening for undiagnosed diabetes and IGT, with appropriate lifestyle interventions for those with IGT are cost saving in China, especially in young adults.