Open AccessPoint-of-care diagnostic tests for influenza in the emergency department: A cost-effectiveness analysis in a high-risk population from a Canadian perspective
Author(s) -
Stephen Mac,
Ryan O’Reilly,
Neill K. J. Adhikari,
Robert Fowler,
Beate Sander
Publication year - 2020
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0242255
Subject(s) - medicine , cost effectiveness , diagnostic test , emergency department , quality adjusted life year , economic evaluation , emergency medicine , health care , population , intensive care medicine , pre and post test probability , pathology , environmental health , risk analysis (engineering) , psychiatry , economics , economic growth
Background Our objective was to assess the cost-effectiveness of novel rapid diagnostic tests: rapid influenza diagnostic tests (RIDT), digital immunoassays (DIA), rapid nucleic acid amplification tests (NAAT), and other treatment algorithms for influenza in high-risk patients presenting to hospital with influenza-like illness (ILI). Methods We developed a decision-analytic model to assess the cost-effectiveness of diagnostic test strategies (RIDT, DIA, NAAT, clinical judgement, batch polymerase chain reaction) preceding treatment; no diagnostic testing and treating everyone; and not treating anyone. We modeled high-risk 65-year old patients from a health payer perspective and accrued outcomes over a patient’s lifetime. We reported health outcomes, quality-adjusted life years (QALYs), healthcare costs, and net health benefit (NHB) to measure cost-effectiveness per cohort of 100,000 patients. Results Treating everyone with no prior testing was the most cost-effective strategy, at a cost-effectiveness threshold of $50,000/QALY, in over 85% of simulations. This strategy yielded the highest NHB of 15.0344 QALYs, but inappropriately treats all patients without influenza. Of the novel rapid diagnostics, NAAT resulted in the highest NHB (15.0277 QALYs), and the least number of deaths (1,571 per 100,000). Sensitivity analyses determined that results were most impacted by the pretest probability of ILI being influenza, diagnostic test sensitivity, and treatment effectiveness. Conclusions Based on our model, treating high-risk patients presenting to hospital with influenza-like illness, without performing a novel rapid diagnostic test, resulted in the highest NHB and was most cost-effective. However, consideration of whether treatment is appropriate in the absence of diagnostic confirmation should be taken into account for decision-making by clinicians and policymakers.