2025 · The Cochrane database of systematic reviews · paywalled
Effectiveness of SARS-CoV-2 testing strategies in reducing COVID-19 cases, hospitalisations, and deaths
Saif-Ur-Rahman et al.
The finding, in our words
This Cochrane review identified 21 studies with over 13 million participants to evaluate decentralised SARS-CoV-2 testing strategies, including home-based saliva self-collection. It applied GRADE and SWiM methods to assess impact on COVID-19 cases, hospitalisations and deaths, establishing an evidence framework for patient-centric diagnostics.
A paraphrase to the Library’s standard, never the abstract. The source is one link away and is always the authority.
The study found that self-collected dried saliva on FishburneTabs achieved 92% accuracy for viral RNA detection compared to standard liquid saliva samples. This method supports decentralised testing by enabling reliable sample storage and transport for upper respiratory infections.
A triple-channel immunochromatographic system using quantum dots detected influenza A and B and SARS-CoV-2 nucleoproteins in saliva with limits of detection from 0.227 to 0.348 ng/mL. Testing of 15 influenza A patients and 20 healthy controls showed 100% sensitivity and 95.2% specificity, indicating potential for accurate home-based respiratory virus screening.
The COVID-19 pandemic accelerated adoption of decentralised testing and self-collected saliva specimens, but revealed challenges with variable diagnostic accuracy and lack of internationally standardised processing protocols. Stricter regulatory oversight and robust diagnostic stewardship are needed to balance broad access with clinical performance.
This review summarises how incorporating saliva and urine into home-based point-of-care systems expands diagnostic accessibility for dengue virus, aligning with the WHO REASSURED framework for affordable and rapid testing.
Capillary blood self-collected with the VAMS device showed 100 per cent agreement with serum for detecting SARS-CoV-2 S RBD antibodies in participants with known past infection, and saliva showed slightly lower but high agreement; agreement was good to almost perfect in those without known antibodies, and most participants found finger prick easy while half found the microsampler easy, supporting decentralised serosurveillance.