You are here

  • Home
  • Online First
  • Meta-analysis of continuous outcomes: a user’s guide for analysis and interpretation

Article Text

Article menu
Download PDFPDF
Education
Meta-analysis of continuous outcomes: a user’s guide for analysis and interpretation
  1. Madelin R Siedler1,2,
  2. Reem A Mustafa1,3,4,
  3. Lifeng Lin5,
  4. Rebecca L Morgan1,3,6,
  5. Yngve Falck-Ytter1,6,
  6. Philipp Dahm1,7,8,
  7. Shahnaz Sultan1,9,
  8. Mohammad Hassan Murad1,10
  1. 1Evidence Foundation, Cleveland, OH, USA
  2. 2Kinesiology and Sport Management, Texas Tech University, Lubbock, TX, USA
  3. 3Department of Health Research Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, Canada
  4. 4Department of Internal Medicine, The University of Kansas Medical Center, Kansas City, Kansas, USA
  5. 5Department of Epidemiology and Biostatistics, The University of Arizona Mel and Enid Zuckerman College of Public Health, Tucson, Arizona, USA
  6. 6School of Medicine, Case Western Reserve University, Cleveland, Ohio, USA
  7. 7Urology Section, Minneapolis VA Health Care System, Minneapolis, Minnesota, USA
  8. 8Department of Urology, University of Minnesota Twin Cities, Minneapolis, Minnesota, USA
  9. 9Division of Gastroenterology, University of Minnesota, Minneapolis, Minnesota, USA
  10. 10Evidence-based Practice Center, Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Madelin R Siedler; madelinsiedler{at}gmail.com

https://doi.org/10.1136/bmjebm-2024-113031

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Binary outcomes, such as mortality and myocardial infarction, are often viewed as most important to patients. However, outcomes that are expressed on a non-dichotomous scale such as quality of life, severity of depression or length of hospitalization, can also be critical to patients. For the purposes of this paper, all of the above examples will be referred to as continuous outcomes. While many binary outcomes are dichotomised from a continuous scale and in some cases can accommodate non-linear associations, such dichotomisation can also lead to loss of information.1

    Measures of dispersion (eg, SD or SE) required for conducting a meta-analysis of continuous outcomes are often unavailable in published manuscripts and continuous outcomes are frequently under-reported in published randomised controlled trials, presented with limited details and are at higher likelihood of selective reporting bias.2 Because of these challenges and the unique characteristics of continuous outcomes, we present approaches to meta-analysis of continuous outcomes including the different effect measures, dealing with imputations and missing data, analysis methods and interpretation of the findings. While available textbooks such as the Cochrane Handbook for Systematic Reviews of Interventions3 provide broad overviews of a wide range of topics related to meta-analysis, the purpose of this article is to outline the step-by-step approaches for identifying and pooling continuous outcomes in particular, all in one place and in simple language.

    Identifying continuous variables

    A continuous variable is defined as a variable that assumes one of infinite values within a range. Thus, it does not assume discrete values. Continuous variables are often called ‘numerical variables’ and can be further categorised as either interval or ratio variables (box 1). Both types are usually treated similarly during statistical analysis and as outcome measures in meta-analysis.4

    Box 1

    Types of continuous variables

    Interval variables: a continuous or numerical variable where the difference between two values is meaningful, …

    View Full Text

    Footnotes

    • X @EBMUrology

    • Contributors Writing—original draft: MRS and MHM. Writing—review and editing: RAM, LL, RLM, YF-Y, PD, SS and MHM.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests The authors declare no financial conflicts of interest. MRS, RAM, RLM, YF-Y, PD, SS and MHM are members of the US GRADE Network. MRS is a fellow of the Evidence Foundation and receives a direct stipend. All other authors declare no conflicts of interest.

    • Provenance and peer review Not commissioned; externally peer reviewed.