Cost-effectiveness of screening strategies to detect heart failure in patients with type 2 diabetes

BACKGROUND Heart failure (HF), especially with preserved ejection fraction (HFpEF) is common in older patients with type 2 diabetes (T2DM), but often not recognized. Early HF detection in older T2DM patients may be worthwhile because treatment may be initiated in an early stage, with clear beneficial treatment in those with reduced ejection fraction (HFrEF), but without clear prognostic beneficial treatment in those with HFpEF. Because both types of HF may be uncovered in older T2DM, screening may improve health outcomes at acceptable costs. We assessed the cost-effectiveness of five screening strategies in patients with T2DM aged 60 years or over. METHODS We built a Markov model with a lifetime horizon based on the prognostic results from our screening study of 581 patients with T2DM, extended with evidence from literature. Cost-effectiveness was calculated from a Dutch healthcare perspective as additional costs (Euros) per additional quality-adjusted life-year (QALY) gained. We performed probabilistic sensitivity analysis to assess robustness of these outcomes. Scenario analyses were performed to assess the influence of the availability of effective treatment of heart failure with preserved ejection fraction. RESULTS For willingness to pay values in the range of €6050/QALY-€31,000/QALY for men and €6300/QALY-€42,000/QALY for women, screening-based checking the electronic medical record for patient characteristics and medical history plus the assessment of symptoms had the highest probability of being cost-effective. For higher willingness-to-pay values, direct echocardiography was the preferred screening strategy. Cost-effectiveness of all screening strategies improved with the increase in effectiveness of treatment for HFpEF. CONCLUSIONS Screening for HF in older community-dwelling patients with T2DM is cost-effective at the commonly used willingness-to-pay threshold of €20.000/QALY by checking the electronic medical record for patient characteristics and medical history plus the assessment of symptoms. The simplicity of such a strategy makes it feasible for implementation in existing primary care diabetes management programs.