A description of the primary studies of diagnostic test accuracy indexed on the DiTA database

Abstract

PEDro (the Physiotherapy Evidence Database) is a widely used, comprehensive, freely available, online database that indexes studies of the effectiveness of physiotherapy interventions. We have recently built another database, called DiTA, on the same platform as PEDro. DiTA provides a comprehensive index of studies of the accuracy of diagnostic tests used by physiotherapists. This study aims to describe the number and scope of such studies. A comprehensive search was conducted for studies of the accuracy of diagnostic tests. The search was conducted on the MEDLINE, EMBASE and CINAHL databases from their inceptions to November 2018. Subsequently, monthly searches have updated the database. To be included on DiTA, studies need to investigate (a) both a pathology and patients that a physiotherapist could assess in clinical practice, and (b) an index test that a physiotherapist would perform themselves rather than one which they would request. To date, the searches have yielded 44,884 titles. Screening has identified 1,419 reports that meet the inclusion criteria. The most frequently studied subdisciplines are "musculoskeletal" (1,050/1,419; 74.0%) and "cardiothoracics" (241; 17.0%); the most frequently studied categories of pathologies are joint pathologies (463; 32.6%) and nervous system pathologies (175; 12.3%); and the most frequently studied body part is the "lower leg or knee" (232; 16.3%). Most studies investigate index tests which are "physical examination" procedures (851; 60.0%); fewer investigate "questions or questionnaires" (420; 29.6%) and "health technologies" (351; 24.7%). There is a rapidly growing body of evidence on the accuracy of diagnostic tests relevant to most physiotherapy subdisciplines. While the volume of evidence is substantial, it is not yet clear how much of the evidence is of good enough quality to support clinical decision-making.