Translation and validation of a geographic search filter to identify studies about Germany in Embase (Ovid) and MEDLINE(R) ALL (Ovid)

Background: In healthcare decision-making, systematic reviews (SRs) and other types of evidence syntheses are essential to guide practitioners and policy makers. A key step is the development of carefully designed search strategies and systematically developed and validated search filters can promote this process. Geographic search filters, designed to retrieve records with a common geographic focus, are helpful in conducting reviews aimed at understanding and addressing health and healthcare challenges related to local context factors. Objective: To translate and validate a formerly developed geographic search filter for retrieving studies about Germany from PubMed [1] for its use in Embase (Ovid) and MEDLINE(R) ALL (Ovid). Methods: We modified the PubMed field tags to align with Ovid’s syntax. Additional adjustments included incorporating a keyword heading field for both databases and a correspondence address field specifically for Embase. To validate the filters, we conducted case studies (CSs), using 17 SRs that included studies about Germany but did not restrict their search strategies geographically. The filters’ performances were assessed by calculating sensitivity, precision, and number-needed-to-read (NNR). These metrics were then compared to the corresponding metrics of the original search strategies of the 17 SRs. Results: The translated filters achieved a median sensitivity of 100% across 17 CSs in both Embase and MEDLINE(R) ALL (Ovid), with interquartile ranges (IQR) of 100–100% for Embase and 93.75–100% for MEDLINE(R) ALL. In Embase, applying the new filter increased the median precision of the original search strategies from 0.11% (IQR: 0.05–0.30%) to 1.65% (IQR: 0.78–3.06%) while reducing the NNR from 839.40 (IQR: 354.81–2219.58) to 60.44 (IQR: 33.94–128.97). Similarly, in MEDLINE(R) ALL, precision improved from 0.19% (IQR: 0.11–0.60%) to 5.13% (IQR: 1.77–6.85%), with the NNR decreased from 513.29 (IQR: 167.35–930.99) to 19.50 (IQR: 14.66–59.35). Implication for research and/or (healthcare) practice: The validated Embase and MEDLINE(R) ALL filters for Ovid reliably retrieve studies about Germany. Additionally, applying the filters substantially improves search precision compared to not using them, reducing the workload associated with screening and subsequent review steps. Our filters can be particularly useful for researchers aiming to comprehensively identify primary studies about Germany, facilitating the conduct of high-quality SRs and other review types crucial for evidence-informed decision-making. Moreover, the filters can also be applied to exploratory or rapid searches where efficient retrieval of relevant studies is needed. Finally, the methodological insights provided in this study can guide search filter developers in translating and validating search filters for other contexts.