Rapid diagnostic testing for influenza: An evidence-based review and comparison with unaided clinical diagnosis
Session III: Petrozzino: Poster Only
Title of Contribution: Rapid diagnostic testing for influenza: An evidence-based review and comparison with unaided clinical diagnosis
Author(s): Jeffrey J. Petrozzino, M.D., Ph.D.1; Cynthia Smith, R.N.1; Mark J. Atkinson, M.Ed., Ph.D.1,2
Affiliation(s):
1 The Aequitas Group, Inc.
12730 High Bluff Drive, Suite 160
San Diego, CA 92130
2 UCSD Department of Family & Preventive Medicine, San Diego, CA
Abstract:
Introduction: Worldwide, influenza imposes a considerable health-economic burden. Clinical diagnosis of influenza-like illness (ILI) is complicated by non-specific symptomatology. Rapid flu tests (RFTs) impact treatment decisions and may improve patient care; yet recommendations for RFT use are broad, and the performance of unaided clinical diagnosis relative to RFTs is unclear. Methods: A systematic literature review was conducted using article selection criteria identifying studies published in the following databases: PubMed/MEDLINE, the Cochrane Library, and other pertinent sources of studies reporting sensitivity, specificity, and effects of RFTs and clinical diagnosis on decision-making for patients with ILI. Age-stratified, overall sensitivities and specificities of the widely-used RFT, QuickVue®, and clinical diagnosis of ILI were calculated by metaanalysis and factors associated with poorer clinical diagnostic discrimination were sought. Results: QuickVue’s® diagnostic specificity exceeds that of unaided clinical diagnosis by 31-36%. False positive results occur approximately 6.4 times more frequently by unaided clinical diagnosis than by the RFT alone. These findings were unaffected by seasonal variations in disease prevalence. RFTs reduce diagnostic testing, antibiotic use, and ED utilization, while increasing antiviral prescription rates. No systematic relationship between the broadness of clinical diagnostic criteria for influenza and diagnostic performance was observed across studies included in this review. Conclusions: Use of RFTs improves influenza diagnostic accuracy above that based on unaided clinical diagnosis irrespective of the broadness of clinical diagnostic criteria, and affects clinical decision-making. These results provide an improved framework upon which to diagnose influenza, design future RFT studies, and modify existing recommendations for improved ILI patient management.
