Menu
  • Join
  • Login
  • Contact
 

Search abstracts


Meta-analysis of pharmacy services: an overview of their quality, data transparency, and replicability

  • At: PPR SIG 2021 (2021)
  • Type: Digital
  • By: BONETTI, Aline de Fátima (Federal University of Paraná, Brazil)
  • Co-author(s): Aline F. Bonetti, Fernanda S. Tonin, Rosa Lucchetta, Ana Maria Della Roca, Fernando Fernandez-Llimos F, Roberto Pontarolo.
  • Abstract:

    Introduction

    Suboptimal, and conflicting meta-analyses are not rare in biomedical disciplines. Recently, several studies from the pharmacy practice field have criticized for their limited data transparency and reproducibility.

    Objectives

    To evaluate the methodological and reporting quality, the risk of bias, and the replicability of meta-analyses of pharmacy services.

    Methods

    A systematic search was performed in PubMed, Scopus, and Web of Science databases to identify meta-analyses of pharmacy services. As part of a PhD thesis, several steps were performed to: (i) map the characteristics of published meta-analyses of pharmacy services; (ii) assess the methodological quality and risk of bias of these meta-analyses (iii) identify the elements that limit the replicability and robustness of meta-analysis of pharmacist-led medication review; (iv) propose a minimum set of standards for the conduction and reporting of meta-analyses in the biomedical area.

    Results

    The quality of conduction and reporting meta-analyses of pharmacy services is sub-optimal, especially in studies that favor the pharmacist’s interventions. Of the 109 studies included in our systematic review, 83 (76.1%) were judged to have a high risk of bias; median PRISMA and R-AMSTAR scores were of 24 (IQR 21.75-25) and 30 (IQR 27-32.5), respectively. Additionally, different results were obtained after the replication of the meta-analyses assessing pharmacist-led medication review for the outcomes of mortality and hospital admission (n = 11) when compared to the original studies. The original studies also failed to report most meta-analytical data, which may prevent research reproducibility. Nine recommendations, including providing information about imputed data and reporting additional parameters such as prediction intervals were proposed to the authors of meta-analyses from the biomedical field to increase studies’ transparency.

    Conclusion

    Editors and reviewers of journals should encourage researchers to comply with the recommended guidelines of systematic reviews and meta-analyses to enhance the research reproducibility, quality, and reliability.

Last update 4 October 2019

FIP Congresses