The Coronavirus Disease 2019(COVID-19) virus testing can be done using multiple specimen types, mainly nasopharyngeal, saliva, and serum. The nasopharyngeal swab (NPS) is a gold standard in COVID-19 testing and diagnosis but is often uncomfortable for the patient and requires professional expertise in sample collection. The rationale of this study was to evaluate saliva, nasopharyngeal, and serum detection of COVID-19 and compare saliva with other specimens in COVID-19 testing. Using PRISMA 2020 guidelines, a data search was performed in the PubMed, Saudi Digital Library, and Cochrane COVID-19 study register. QUADAS 2 tool was applied to assess the quality of the studies included. The efficacy of saliva, serum, and nasopharyngeal specimens was the primary outcome measured in terms of sensitivity and specificity, and the secondary outcome was the comparison of saliva with NPS and serum for COVID-19 detection. Data were extracted from 39 studies-20 countries- 20,024 patients and 22123 samples. QUADAS-2 tool was applied. Meta-analysis showed significant differences in sensitivity between all specimens and when NPS is compared to saliva. Within the limitations, despite a significant heterogeneity (P < 0.001), the efficacy in the detection of COVID-19 is more in a balance between saliva and NPS. Saliva-Area under the curve (AUC) = 0.97, nasopharyngeal specimen (NPS): AUC= 0.94, AUC=l.00, suggestive of an excellent performance of serum (active infection)>saliva>NPS specimen in SARS- CoV-2 detection. The study's outcomes suggested that saliva specimens can be used as a non-invasive diagnostic method in COVID-19 testing.
