Systematic review and meta-analysis of prevalence and single group continuous data: methodological and practical challenges and benefits

ID: 

18133

Session: 

Long oral session 10: Meta-analysis methods A

Date: 

Thursday 14 September 2017 - 11:00 to 12:30

Location: 

Meeting Room 1.43

All authors in correct order:

Jadotte Y1, Holly C1, Salmond S1, Benenson I1, Paplanus L2, Thompson-Stewart P3
1 Northeast Institute for Evidence Synthesis and Translation at Rutgers University SN, USA
2 NYU Langone Medical Center, USA
3 Rutgers University SN, USA
Presenting author and contact person

Presenting author:

Yuri Jadotte

Contact person:

Yuri Jadotte
Cheryl Holly
Abstract text
Background: Systematic reviews and meta-analyses of evidence from studies other than RCTs are becoming more prevalent. In particular, meta-analysis of prevalence and single-group continuous data is an important new application of evidence synthesis for decision making. While there are published methodology papers that provide guidance on the appropriate strategies for performing these newer reviews, no study has yet explored the utility of these evidence-synthesis methods and the resulting methodological and practical challenges and benefits using real-world cases.

Objectives: This project examines the methodological and practical challenges and benefits of the synthesis of prevalence and single-group continuous data, using four real-world systematic reviews as cases.

Methods: Two pairs of systematic reviews were conducted on prevalence and on single-group continuous outcomes. Comprehensive multi-step searches were undertaken for published and unpublished studies. Only English language studies were included. Retrieved papers were assessed for methodological quality using standardised critical-appraisal instruments from the Joanna Briggs Institute. Separate meta-analyses of prevalence and single-group continuous outcomes were performed for each review.

Results: Seventeen studies were included in the meta-analyses for the two prevalence reviews, while fifteen studies were included in the meta-analyses for the two single-group continuous data reviews. The key methodological issues were the definition of the null value and the utility of statistical significance. The central practical challenge was the conceptualisation of generalisability for these types of synthesised evidence. The key benefit was the ability to explore geographic and other types of population heterogeneity using subgroup analysis.

Conclusions: The systematic review and meta-analysis of prevalence and single-group continuous outcomes present different methodological and practical challenges and benefits that must inform the careful application of this type of synthesised evidence to clinical decision making. Additional research is needed to advance and further validate these methods.