Advice
Search strategy and evidence selection
Search strategy
A strategy was designed in Medline and adapted for the following databases:
-
Medline In Process
-
Embase
-
Cochrane library (CDSR, CENTRAL, DARE, HTA, NHS EED and HEED).
Additional searches were also conducted of the following:
-
ClinicalTrials.gov
-
ICTRP
-
FDA MAUDE
-
MHRA.
Evidence selection
References were imported into Reference Manager and screened independently in duplicate by title and abstract. The full publications, where available, of selected references were also screened independently in duplicate. Data from each of the studies were extracted by 1 reviewer and independently checked for accuracy by a second reviewer. Disagreements were resolved through discussion. Studies were assessed for quality by using an adapted version of the QUADAS‑2 tool as directed by the NICE interim process and methods statement. Assessment was performed by 1 reviewer and independently checked by a second reviewer and disagreements resolved by discussion.
Three studies were excluded:
-
Villar et al. 2012 is a poster abstract. It may also contain some patient data that crosses over with the Scalioni et al. 2014 study.
-
Lee et al. 2012 is a conference abstract only, and therefore does not contain much methodological detail.
-
Stockman et al. 2014 is a screening study using OraQuick in four HCV community outreach sites in Wisconsin, USA. Only positive OraQuick samples had follow‑up EIA and PCR tests, and therefore no diagnostic performance data could be calculated for OraQuick from the results.