Selecting a database for literature searches in nursing: MEDLINE or CINAHL?
Hugh Brazier, Cecily Begley
Abstract
Hugh Brazier, Cecily Begley
Abstract
This study compares the usefulness of the MEDLINE and CINAHL databases for students on post-registration nursing courses. We searched for nine topics, using title words only. Identical searches of the two databases retrieved 1162 references, of which 88% were in MEDLINE, 33% in CINAHL and 20% in both sources. The relevance of the references was assessed by student reviewers. The positive predictive value of CINAHL (70%) was higher than that of MEDLINE (54%), but MEDLINE produced more than twice as many relevant references as CINAHL. The sensitivity of MEDLINE was 85% (95% CI 82-88%), and that of CINAHL was 41% (95% CI 37-45%). To assess the ease of obtaining the references, we developed an index of accessibility, based on the holdings of a number of Irish and British libraries. Overall, 47% of relevant references were available in the students' own library, and 64% could be obtained within 48 hours. There was no difference between the two databases overall, but when two topics relating specifically to the organization of nursing were excluded, references found in MEDLINE were significantly more accessible. We recommend that MEDLINE should be regarded as the first choice of bibliographic database for any subject other than one related strictly to the organization of nursing.
OpenAlex reports 46 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This study compares the usefulness of the MEDLINE and CINAHL databases for students on post-registration nursing courses. We searched for nine topics, using title words only. Identical searches of the two databases retrieved 1162 references, of which 88% were in MEDLINE, 33% in CINAHL and 20% in both sources. The relevance of the references was assessed by student reviewers. The positive predictive value of CINAHL (70%) was higher than that of MEDLINE (54%), but MEDLINE produced more than twice as many relevant references as CINAHL. The sensitivity of MEDLINE was 85% (95% CI 82-88%), and that of CINAHL was 41% (95% CI 37-45%). To assess the ease of obtaining the references, we developed an index of accessibility, based on the holdings of a number of Irish and British libraries. Overall, 47% of relevant references were available in the students' own library, and 64% could be obtained within 48 hours. There was no difference between the two databases overall, but when two topics relating specifically to the organization of nursing were excluded, references found in MEDLINE were significantly more accessible. We recommend that MEDLINE should be regarded as the first choice of bibliographic database for any subject other than one related strictly to the organization of nursing.
Key concepts: CINAHL, MEDLINE, Medicine, Cochrane Library, Nursing, Meta-analysis, Internal medicine, Psychological intervention