Skip to main content

Advertisement

Table 2 Search permutations and performance (recall) in the Filter Development Set

From: Development of a heart failure filter for Medline: an objective approach using evidence-based clinical practice guidelines as an alternative to hand searching

Baseline search Representation No. records retrieved in Filter Development Set (n = 394) Recall (%)
Heart failure.mp. T1 365 92.6
  T1 OR left ventricular ejection fraction.mp. 372 94.4
  T1 OR cardiomyopathy.mp. 373 94.7
  T1 OR Ventricular dysfunction, Left.sh. 373 94.7
Heart failure.mp. OR cardiomyopathy.mp. T1 OR T2(a)* 373 94.7
  T1 OR T2(a) OR left ventricular ejection fraction.mp. 378 95.9
  T1 OR T2(a) OR Ventricular dysfunction, Left.sh. 380 96.4
Heart failure.mp. OR Ventricular dysfunction, Left.sh. T1 OR T2(b)* 373 94.7
  T1 OR T2(b) OR left ventricular ejection fraction.mp. 376 95.4
  T1 OR T2(b) OR cardiomyopathy.mp. 380 96.4
Heart failure.mp. OR Ventricular dysfunction, Left.sh. OR cardiomyopathy.mp. T1 OR T2(b) OR T2(a) 380 96.4
  T1 OR T2(b) OR T2(a) OR left ventricular ejection fraction.mp. 382 96.9
  1. *The two terms cardiomyopathy.mp. and Ventricular dysfunction, Left both qualified for term 2 (T2), having retrieved the same number of references in combination with T1. They have been labelled T2(a) and T2(b) in order to assess their relative performance as part of the baseline search.
  2. Note: The OvidSP Medline suffix '.mp.' forces a search on the title, abstract, subject headings, and chemical registry fields while the '.sh.' suffix forces a search on the subject headings field alone.