Molecular point-of-care testing versus conventional RT-PCR for influenza A: Diagnostic performance, clinical decision-making, and costs in a tertiary hospital in Shenzhen, China.

BACKGROUND

Rapid diagnosis of influenza A is important for timely treatment and clinical management. However, evidence on the real-world clinical and economic utility of molecular point-of-care testing (POCT) remains limited, particularly in China.

METHODS

This retrospective study was conducted at a tertiary hospital in Shenzhen, China. A performance validation study included 300 fever-clinic patients tested by both molecular POCT and conventional real-time reverse transcription polymerase chain reaction (RT-PCR) from April to May 2024. An application study compared 1021 patients tested by conventional RT-PCR from January to March 2024 with 4210 patients tested by molecular POCT from January to March 2025. Outcomes included diagnostic concordance, turnaround time (TAT), medication use, direct medical costs, and user satisfaction.

RESULTS

Molecular POCT showed 100% concordance with conventional RT-PCR in the validation cohort, with 100% positive, negative, and overall agreement (Cohen's κ = 1.00). In the application study, median TAT was markedly shorter with POCT than with conventional RT-PCR, at 51 (IQR 44-60) min vs. 555 (IQR 370-1328) min (p < 0.001). Among patients receiving antivirals, etiological confirmation before prescription was more frequent in the POCT group than in the conventional RT-PCR group (58.0% vs. 31.7%; p < 0.001). Overall antiviral use was lower in the POCT group (44.4% vs. 55.2%; p < 0.001), whereas antibiotic use was higher (47.6% vs. 36.4%; p < 0.001), mainly among influenza A-negative patients. Total medical costs were lower in the POCT group, at 440.12 (IQR 347.35-606.01) vs. 613.44 (IQR 480.70-748.93) Chinese yuan (CNY) (p < 0.001), mainly because of reduced laboratory fees. Satisfaction with POCT was high among both patients and healthcare providers.

CONCLUSIONS

Molecular POCT for influenza A showed diagnostic accuracy comparable to conventional RT-PCR while substantially shortening TAT, supporting more etiologically informed antiviral prescribing, and reducing direct medical costs in routine clinical practice. Its effect on antibiotic prescribing, however, was less predictable.