Cost-effectiveness of rapid laboratory-based mass-spectrometry diagnosis of bloodstream infection: evidence from the RAPIDO randomised controlled trial.

Abstract

Bloodstream infection, the presence of viable micro-organisms in the blood, is a prevalent clinical event associated with substantial mortality. Patient outcomes may be improved when the causative micro-organism is identified quickly. We assessed the cost-effectiveness of rapid microbial identification by matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry. Economic evaluation alongside a randomised multicentre trial (RAPIDO: RAPId Diagnosis on Outcome) assessing the impact of rapid identification by MALDI-TOF spectrometry. Adult inpatients with bloodstream infections at seven National Health Service hospital trusts in England and Wales. Net monetary benefit, estimated as incremental costs compared with incremental 28-day survival, of rapid identification by MALDI-TOF spectrometry compared with conventional identification. Patients were randomised (1:1) to receive diagnosis by conventional methods of microbial identification (conventional arm) only or by MALDI-TOF spectrometry in addition to conventional identification (RAPIDO arm). Data from 5550 patients were included in primary analysis. Mean imputed costs in 2018/2019 prices per patient were lower by £126 in the RAPIDO arm (95% CI -£784 to £532) but the proportion of patients alive at day 28 was lower (81.4% vs 82.3%). The probability of cost-effectiveness of MALDI-TOF was <0.5 at cost-effectiveness thresholds between £20 000 and £50 000. Adjunctive MALDI-TOF diagnosis was unlikely to be cost-effective when measured as cost per death avoided at 28 days. However, the differences between arms in cost and effect were modest, associated with uncertainty and may not accurately reflect 'real-world' routine use of MALDI-TOF technology in this patient group. ISRCTN97107018/UKCRN 11978.