The organisational value of diagnostic strategies using high-sensitivity troponin for patients with possible acute coronary syndromes: a trial-based cost-effectiveness analysis

OBJECTIVES: To evaluate hospital-specific health economic implications of different protocols using high-sensitivity troponin I for the assessment of patients with chest pain. DESIGN: A cost prediction model and an economic microsimulation were developed using a cohort from a single centre recruited as part of the (ADAPT) trial, a prospective observational trial conducted from 2008 to 2011. The model was populated with 40 000 bootstrapped samples in five high-sensitivity troponin I-enabled algorithms versus standard care. SETTING: Adult emergency department (ED) of a tertiary referral hospital. PARTICIPANTS: Data were available for 938 patients who presented to the ED with at least 5 min of symptoms suggestive of acute coronary syndrome. The analyses included 719 patients with complete data. MAIN OUTCOME(S)/MEASURE(S): This study examined direct hospital costs, number of false-negative and false-positive cases in the assessment of acute coronary syndrome. RESULTS: High-sensitivity troponin I-supported algorithms increased diagnostic accuracy from 90.0% to 94.0% with an average cost reduction per patient compared with standard care of $490. The inclusion of additional criteria for accelerated rule-out (limit of detection and the modified 2-hour ADAPT trial rules) avoided 7.5% of short-stay unit admissions or 25% of admissions to a cardiac ward. Protocols using high-sensitivity troponin I alone or high-sensitivity troponin I within accelerated diagnostic algorithms reduced length of stay by 6.2 and 13.6 hours, respectively. Overnight stays decreased up to 43%. Results were seen for patients with non-acute coronary syndrome; no difference was found for patients with acute coronary syndrome. CONCLUSIONS: High-sensitivity troponin I algorithms are likely to be cost-effective on a hospital level compared with sensitive troponin protocols. The positive effect is conferred by patients not diagnosed with acute coronary syndrome. Implementation could improve referral accuracy or facilitate safe discharge. It would decrease costs and provide significant hospital benefits. TRIAL REGISTRATION: The original ADAPT trial was registered with the Australia-New Zealand Clinical trials Registry, ACTRN12611001069943.