Diagnostic accuracy of multiplex respiratory pathogen panels for influenza or respiratory syncytial virus infections: systematic review and meta-analysis

Table 2 Summary of findings for diagnostic accuracy of Luminex NxTAG RPP™ for the diagnosis of RSV

Outcome	Effect per 1000 patients tested		№ of studies (№ of samples)	Test accuracy certainty of evidence
Outcome	Pre-test probability of 5%	Pre-test probability of 20%	№ of studies (№ of samples)	Test accuracy certainty of evidence
Index and comparator tests positive (I + /C +) (patients with RSV infection)	50 (48 to 50)	198 (192 to 200)	8 studies 527 samples	⊕ ⊕ ⊝ ⊝^a LOW
Index test negative, comparator positive (I−/C +) (patients incorrectly classified as not having RSV infection)	0 (0 to 2)	2 (0 to 8)		⊕ ⊕ ⊝ ⊝^a LOW
Index and comparator tests negative (I−/C−) (patients without RSV infection)	950 (931 to 950)	800 (784 to 800)	8 studies 5601 samples	⊕ ⊕ ⊝ ⊝^a LOW
Index test positive, comparator negative (I + /C−) (patients incorrectly classified as having RSV infection)	0 (0 to 19)	0 (0 to 16)		⊕ ⊕ ⊝ ⊝^a LOW

C comparator, CrI credible interval, I index test, RPP respiratory pathogen panel
Patient or population: adults and children with symptoms of acute lower respiratory infection
Setting: worldwide
Index test: Luminex NxTAG RPP™
Comparator tests: other RPP
Predicted sensitivity: 0.99 (95% CrI: 0.96 to 1.00) | Predicted specificity: 1.00 (95% CrI: 0.98 to 1.00)
^aDowngraded two levels for risk of bias: there is high or unclear risk of bias on the patient selection and flow and timing domains for all included studies. Six studies were planned after the test was performed on patients selected by physician (e.g. respiratory symptoms) with no clear inclusion and exclusion criteria, which is at high risk of introducing bias for evaluating diagnostic test accuracy.

ISSN: 1471-2334