Bita Mesgarpour1, Markus Müller, Harald Herkner. 1. Department of Clinical Pharmacology, Medical University Vienna, General Hospital, Währinger Gürtel 18-20, 1090 Vienna, Austria.
Abstract
OBJECTIVE: We developed search strategies that facilitate the identification of studies on "off-label" drug use in the bibliographic database OvidSP MEDLINE. STUDY DESIGN AND SETTING: We compiled a gold standard reference set of reports classified as relevant or not relevant to off-label drug use. We conceived search queries, including search words and word strings. We searched MEDLINE via OvidSP from 1948 to 2011. In comparison with the gold standard, we determined sensitivity and precision of search queries and their combinations. We attempted to achieve the highest possible sensitive search strategy and an optimal balance of sensitivity and precision. RESULTS: Our gold standard set contained 4,067 relevant documents overall of 6,785 records, among those 2,177 could be retrieved from MEDLINE. The most sensitive single term was "off label∗.af." (overall sensitivity 40.9%, sensitivity within MEDLINE 76.4%, and precision 84.4%). A combination of 31 search queries had the maximum overall sensitivity of 53.3% (sensitivity within MEDLINE 99.5%) at a precision of 60.3%. A search strategy with the maximum precision (84.0%) yielded a sensitivity of 49.0% (sensitivity within MEDLINE 91.5%). CONCLUSION: We empirically developed two versions of optimized sensitive search strategies, which can achieve reasonable performance for retrieving off-label drug use documents in OvidSP MEDLINE.
OBJECTIVE: We developed search strategies that facilitate the identification of studies on "off-label" drug use in the bibliographic database OvidSP MEDLINE. STUDY DESIGN AND SETTING: We compiled a gold standard reference set of reports classified as relevant or not relevant to off-label drug use. We conceived search queries, including search words and word strings. We searched MEDLINE via OvidSP from 1948 to 2011. In comparison with the gold standard, we determined sensitivity and precision of search queries and their combinations. We attempted to achieve the highest possible sensitive search strategy and an optimal balance of sensitivity and precision. RESULTS: Our gold standard set contained 4,067 relevant documents overall of 6,785 records, among those 2,177 could be retrieved from MEDLINE. The most sensitive single term was "off label∗.af." (overall sensitivity 40.9%, sensitivity within MEDLINE 76.4%, and precision 84.4%). A combination of 31 search queries had the maximum overall sensitivity of 53.3% (sensitivity within MEDLINE 99.5%) at a precision of 60.3%. A search strategy with the maximum precision (84.0%) yielded a sensitivity of 49.0% (sensitivity within MEDLINE 91.5%). CONCLUSION: We empirically developed two versions of optimized sensitive search strategies, which can achieve reasonable performance for retrieving off-label drug use documents in OvidSP MEDLINE.