The authors evaluated the retrieval power of PubMed “Clinical Queries,” narrow search string,about therapy in comparison with a modiﬁed search string to avoid possible retrieval bias. PubMed search strategy was compared to a slightly modiﬁed string that included the Britannic English term “randomised.” The authors tested the two strings joined onto each of four terms concerning topics of broad interest: hypertension, hepatitis, diabetes, and heart failure. In particular, precision was computed for not-indexed citations. The added word “randomised” improved total citation retrieval in any case. Total retrieval gain for not-indexed citations ranged from 11.1% to 21.4%. A signiﬁcant number of Randomized Controlled Trial(s) (RCT)s (9.1-18.2%) was retrieved for each of the selected topics. They were often recently published RCTs. The authors think that correction of the Clinical Queries ﬁlter (when they focus on therapy and use narrow searches) is necessary to avoid biased search results with loss of relevant and up-todate scientiﬁcally sound information.
|Numero di pagine||3|
|Rivista||JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION|
|Stato di pubblicazione||Published - 2006|
All Science Journal Classification (ASJC) codes
- Health Informatics