Development of a search filter to retrieve reports of interrupted time series studies from MEDLINE and PubMed

Abstract

Background: Interrupted time series (ITS) studies contribute importantly to systematic reviews of population-level interventions. However, there is no search filter designed to identify only ITS studies from bibliographic databases. We aimed to develop and validate search filters to retrieve ITS studies in MEDLINE and PubMed. Methods: A set of 1,017 known ITS studies (published 2013-2017) were analysed using text mining to generate candidate terms. A population set of 1,398 time-series studies were used to select discriminating terms. Various combinations of candidate terms were iteratively tested to generate three search filters. An independent set of 700 ITS studies were used to validate the filters' sensitivities. The filters were test-run in Ovid MEDLINE; their retrieved records were randomly screened for ITS studies to determine their precision. Finally, all three MEDLINE filters were translated to PubMed format and their sensitivities in PubMed were estimated. Results: Three search filters were created: a precision-maximising filter with high precision (78%; 95% CI 74%-82%) but moderate sensitivity (63%; 59%-66%), useful for rapid reviews; a sensitivity-and-precision-maximising filter with high sensitivity (81%; 77%-83%) but lower precision (32%; 28%-36%), useful when a balance of expediency and comprehensiveness is required; and a sensitivity-maximising filter with high sensitivity (88%; 85%-90%) but likely very low precision, useful for search strategies accompanying specific content terms. Conclusion: Our filters strike different balances between comprehensiveness and screening workload and suit different research needs. To improve retrieval of ITS studies, authors need to identify ITS designs in titles more frequently.