Heterogeneity in application, design, and analysis characteristics was found for controlled before-after and interrupted time series studies included in Cochrane reviews

Polus, Stephanie; Pieper, Dawid; Burns, Jacob; Fretheim, Atle; Ramsay, Craig; Higgins, Julian P. T.; Mathes, Tim; Pfadenhauer, Lisa M. und Rehfuess, Eva A. (2017): Heterogeneity in application, design, and analysis characteristics was found for controlled before-after and interrupted time series studies included in Cochrane reviews. In: Journal of Clinical Epidemiology

Abstract

OBJECTIVES The aim of the study was to examine the application, design, and analysis characteristics of controlled before-after (CBA) and interrupted time series (ITS) studies and their use in Cochrane reviews. STUDY DESIGN AND SETTING We searched the Cochrane library for reviews including these study designs from May 2012 to March 2015 and purposively selected, where available, two reviews each across 10 prespecified intervention types. We randomly selected two CBA and two ITS studies from each review. Two researchers independently extracted information from the studies and the respective reviews. RESULTS Sixty-nine reviews considered CBA and ITS studies for inclusion. We analyzed 21 CBA and 16 ITS studies from 11 to 8 reviews, respectively. Cochrane reviews inconsistently defined and labeled CBA and ITS studies. Many studies did not meet the Cochrane definition or the minimum criteria provided by Cochrane Effective Practice and Organisation of Care. The studies present a heterogeneous set of study features and applied a large variety of analyses. CONCLUSION While CBA and ITS studies represent important study designs to evaluate the effects of interventions, especially on a population or organizational level, unclear study design features challenge unequivocal classification and appropriate use. We discuss options for more specific definitions and explicit criteria for CBA and ITS studies.

• Dokumententyp: Zeitschriftenartikel
• Keywords: Controlled before-after studies; Interrupted time series analysis; Methods; Nonrandomised study designs; Public health; Review
• Fakultät: Medizin > Institut für Medizinische Informationsverarbeitung, Biometrie und Epidemiologie
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• ISSN: 0895-4356; 1878-5921
• Sprache: Englisch