Use of quasi-experimental studies to evaluate causal effects of public health interventions in Portugal: a scoping review
More details
Hide details
Departamento de Epidemiologia, Instituto Nacional de Saúde Doutor Ricardo Jorge, Lisbon, Portugal, NOVA National School of Public Health, Public Health Research Center, Universidade NOVA de Lisboa, Portugal, Comprehensive Health Research Center, Portugal
Publication date: 2023-04-26
Popul. Med. 2023;5(Supplement):A1862
Background and Objective:
Evaluating causal effects of public health interventions using traditional randomized controlled trials might not be feasible. Quasi-experimental designs are a valid option but still not widely used in Portugal. Knowing their application in real public health problems will support the development of this research area. We thus performed a scoping review aiming at identifying and characterising the use of quasi-experimental studies to evaluate causal effects of public health interventions in Portugal.

We included studies that used a quasi-experimental design to assess causal effects of one or more public health intervention in Portugal. PubMed, Scopus, Web of Science and CINHAL were searched from inception, combining free text and controlled vocabulary terms. Grey literature was identified through screening of tables of contents of non-indexed publications and institutional repositories of national Public Health PhD and MSc programmes theses. Studies were selected after title and abstract followed by full-text, double-screening. Searches were supplemented by reference mining and contact with authors of eligible studies. We extracted information on the intervention assessed, study design, statistical analysis approach and reporting guidelines followed using a standardised extraction form.

This study is ongoing, we present preliminary results from the databases search. After deduplication we identified 500 studies, 38 were included for full-text screening, and 25 were eligible. Studies assessed interventions in various areas, namely healthcare services (40.0%), tobacco and drugs control policy (20.0%), and pharmaceutics policy (16.0%), among others. Study designs were mainly interrupted time series (44.0%), followed by difference-in-differences approaches (40.0%).

There is a paucity of studies in this area with interrupted time series and difference-in-differences approaches being the most often used study designs. Training in this area might promote the use and dissemination of quasi-experimental studies.