Comparing the effectiveness and cost-effectiveness of alternative type 2 diabetes monitoring intervals in resource limited settings

Abstract

Type 2 diabetes (T2D) represents a growing disease burden in South Africa. While glycated haemoglobin (HbA1c) testing is the gold standard for long-term blood glucose management, recommendations for HbA1c monitoring frequency are based on expert opinion. This study investigates the effectiveness and cost-effectiveness of alternative HbA1c monitoring intervals in the management of T2D. A Markov model with three health states (HbA1c <7%, HbA1c ≥ 7%, Dead) was used to estimate lifetime costs and quality-adjusted life years (QALYs) of alternative HbA1c monitoring intervals among patients with T2D, using a provider’s perspective and a 3% discount rate. HbA1c monitoring strategies (three-monthly, four-monthly, six-monthly and annual tests) were evaluated with respect to the incremental cost-effectiveness ratio (ICER) assessing each comparator against a less costly, undominated alternative. The scope of costs included the direct medical costs of managing diabetes. Transition probabilities were obtained from routinely collected public sector HbA1c data, while health service utilization and health-related-quality-of-life (HRQoL) data were obtained from a local cluster randomized controlled trial. Other parameters were obtained from published studies. Robustness of findings was evaluated using one-way and probabilistic sensitivity analyses. A South African indicative cost-effectiveness threshold of USD2665 was adopted. Annual and lifetime costs of managing diabetes increased with HbA1c monitoring, while increased monitoring provides higher QALYs and life years. For the overall cohort, the ICER for six-monthly vs annual monitoring was cost-effective (USD23 22.37 per QALY gained), whereas the ICER of moving from six-monthly to three-monthly monitoring was not cost effective (USD6437.79 per QALY gained). The ICER for four-monthly vs six-monthly monitoring was extended dominated. The sensitivity analysis showed that the ICERs were most sensitive to health service utilization rates. While the factors influencing glycaemic control are multifactorial, six-monthly monitoring is potentially cost-effective while more frequent monitoring could further improve patient HrQoL.