Table 2.

Best multiple-term strategies maximising sensitivity and minimising the difference between sensitivity and specificity. Values are percentages (95% confidence intervals)

Search strategy in Ovid format	Sensitivity*	Specificity**	Precision†
Top sensitivity strategies‡
search:.tw. or meta-analysis.mp,pt. or review.pt. or di.xs. or associated.tw.
Development	100 (97.3 to 100)	63.5 (62.5 to 64.4)	3.41 (2.86 to 4.03)
Validation without CDSR	99.7 (99.1 to 100)	51.1 (50.7 to 51.6)	1.4 (1.2 to 1.5)
Validation	99.9 (99.6 to 100)	52.0 (51.6 to 52.5)	3.14 (2.92 to 3.37)
Top strategy minimising the difference between sensitivity and specificity§
meta-analysis.mp,pt. or review.pt or search:.tw.
Development	92.5 (86.6 to 96.3)	93.0 (92.5 to 93.5)	14.6 (12.3 to 17.2)
Validation without CDSR	95.5 (93.3 to 97.7)	89.9 (89.7 to 90.2)	6.1 (5.5 to 6.8)
Validation	98.0 (97.0 to 99.0)	90.8 (90.5 to 91.1)	14.2 (13.3 to 15.2)

CDSR=Cochrane Database of Systematic Reviews.

^*Development dataset (n=133); validation dataset without CDSR (n=332); validation dataset (n=753).

^**Development dataset (n=10 313); validation dataset without CDSR (n=48 258); validation dataset (n=48 275).

^†Numbers vary by row.

^‡Keeping specificity ≥50%; adding the Cochrane Database of Systematic Reviews (using Boolean OR) did not improve performance.

^§Keeping sensitivity ≥90%.