Cost-effectiveness of prevention program for type 2 diabetes mellitus in high risk patients in the Republic of Srpska, Bosnia and Herzegovina

Abstract

The Republic of Srpska (RS), as a part of the Western Balkans (WB) region, has a higher diabetes prevalence than the EU. This study aims to assess the cost-effectiveness of early treatment of high-risk patients with pre-diabetes and undiagnosed diabetes in our setting. We designed a Markov chain Monte Carlo (MCMC) model which reflects the current International Diabetes Federation (IDF) three-step plan for the prevention of T2DM in those at increased risk. The model captures the evolution of the disease in FINDRISC high-risk patients from normal glucose tolerance (NGT) to impaired fasting glucose (IFG) or impaired glucose tolerance (IGT) and then to T2DM and its complications. We developed two MCMC models, in order to follow the progression of the disease in high-risk cases, ie, when early treatment is undertaken or when it is not undertaken. The health costs and quality adjusted life years (QALY) were discounted at an annual rate of 3%. The key model parameters were varied in one-way and probabilistic sensitivity analysis. Early treatment resulted in increased life expectancy, postponement of the onset of diabetes and increased QALY for all patients. The discounted incremental cost-effectiveness-ratios (ICER) in NGT, IFG, IGT, and T2DM patients were -289.9, 9724.03, -1478.59 and 4084.67 €. In high-risk IGT patients, ICER was the most favorable, being both a cost saving and QALY gaining, with the consistent results confirmed by the sensitivity analysis. The results recommend the acceptance of a new health policy of identifying IGT patients with the use of FINDRISC questionnaire and plasma glucose measurements; providing them with a lifestyle change program; and implementing intensive diabetes treatment, as their disease progresses. Our results are especially significant for the Western Balkan countries, since this was the first cost-effectiveness study of T2DM prevention in this region.

Keywords: Cost-effectiveness; FINDRISC; Markov chain model; impaired glucose tolerance; prevention; type 2 diabetes mellitus.

Figure 1.

End of year transitions of health states of the MCMC model (Abbreviations: DR – diabetic retinopathy, mNPDR – mild/moderate non-proliferative diabetic retinopathy, sNPDR – severe non-proliferative diabetic retinopathy, PDR – proliferative diabetic retinopathy, DKD – diabetic kidney disease, DN – diabetic neuropathy, DFU – diabetic foot ulcer, AMP – amputation, MI – myocardial infarction, CVI – cerebrovascular insult, MULT – multiple complications, nrec – non-recurrent, rec – recurrent).

Figure 2.

Histogram of the distribution of the mean discounted costs gained for NGT, IFG, IGT and T2DM patients for 1000 random values of parameters.

Figure 3.

Histogram of the distribution of the mean discounted QALY gained for NGT, IFG, IGT and T2DM patients for 1000 random values of parameters.