Background Systematic reviews (SRs) are considered one of the most reliable types of studies in evidence-based medicine. SRs rely on a comprehensive and systematic data gathering, including the search of academic literature databases. This study aimed to investigate which combination of databases would result in the highest overall recall rate of references when conducting SRs of qualitative research regarding diabetes mellitus. Furthermore, we aimed to investigate the current use of databases and other sources for data collection. Methods Twenty-six SRs (published between 2010 and 2020) of qualitative research regarding diabetes mellitus, located through PubMed, met the inclusion criteria. References of the SRs were systematically hand searched in the six academic literature databases CINAHL, MEDLINE/PubMed, PsycINFO, Embase, Web of Science, and Scopus and the academic search engine Google Scholar. Recall rates were calculated using the total number of included references retrieved by the database or database combination divided by the total number of included references, given in percentage. Results The SRs searched five databases on average (range two to nine). MEDLINE/PubMed was the most commonly searched database (100% of SRs). In addition to academic databases, 18 of the 26 (69%) SRs hand searched the reference lists of included articles. This technique resulted in a median (IQR) of 2.5 (one to six) more references being included per SR than by database searches alone. 27 (5.4%) references were found only in one of six databases (when Google Scholar was excluded), with CINAHL retrieving the highest number of unique references (n = 15). The combinations of MEDLINE/PubMed and CINAHL (96.4%) and MEDLINE/PubMed, CINAHL, and Embase (98.8%) yielded the highest overall recall rates, with Google Scholar excluded. Conclusions We found that the combinations of MEDLINE/PubMed and CINAHL and MEDLINE/PubMed, CINAHL, and Embase yielded the highest overall recall rates of references included in SRs of qualitative research regarding diabetes mellitus. However, other combinations of databases yielded corresponding recall rates and are expected to perform comparably. Google Scholar can be a useful supplement to traditional scientific databases to ensure an optimal and comprehensive retrieval of relevant references.

Background Systematic reviews (SRs) are considered one of the most reliable types of studies in evidence-based medicine. This relies on a comprehensive and systematic search by choosing ideal data gathering methods including selection of databases. The aim of this study was to investigate which combination of databases results in the highest recall of references when conducting SRs on qualitative research regarding diabetes mellitus. Furthermore, we aimed to investigate the current use of databases and the importance of other sources for data collection. Methods 23 SRs (published between the year 2010 and 2019) on qualitative research regarding diabetes mellitus were located through searching PubMed and met the inclusion criteria. Data, including number of databases searched, names of databases, use of additional data sources and use of information specialists were collected for each SR. The SRs concluded a total of 459 unique, qualitative references on diabetes mellitus. These references were systematically hand searched in the five most searched databases: Cumulative Index to Nursing and Allied Health Literature (CINAHL), MEDLINE, PsycINFO, Embase and PubMed. Results The SRs searched four databases on average (range two to eight). CINAHL was the most searched database, (19 of the 23 SRs). Two SRs mentioned the involvement of an information specialist and 16 of 23 (70%) SRs searched reference lists of included references, which on average resulted in 16% more references being included. A total of 36 (8%) references were found only in one of the five databases, with CINAHL retrieving the highest number of unique references. Searching the combination of the three databases PubMed, Embase and CINAHL resulted in an overall recall rate of 99.3%, while adding PsycINFO increased overall recall to 99.8%. Conclusions We recommend combining the searches of CINAHL, PubMed, Embase and PsychINFO, and to involve an information specialist to ensure high recall rates, when conducting SRs on qualitative research regarding diabetes mellitus. Especially CINAHL is highly relevant and important to use when searching for qualitative research.