Relationship between the description of primary outcomes and significance of the results in trials with diet and lifestyle in pregnancy

ID: 

98

Session: 

Poster session 1

Date: 

Monday 24 October 2016 - 10:30 to 11:00

Location: 

All authors in correct order:

Rogozinska E1, Zamora J2, Thangaratinam S1
1 Women's Health Research Unit, Queen Mary University of London, UK
2 Clinical Biostatistics Unit, Hospital Ramon y Cajal and CIBER Epidemiology and Public Health, Spain
Presenting author and contact person

Presenting author:

Ewelina Rogozinska

Contact person:

Abstract text
Background: The quality of outcome reporting in primary studies can significantly hamper the findings of a systematic review. The CONSORT statement specifies that outcomes should be described to allow their reproducibility, and that treatment effects should be reported for all predefined outcomes. Nevertheless, a recent assessment of Cochrane Reviews showed ongoing problems with outcome reporting. Empirical evidence shows a strong association between statistical significance of outcomes and the likelihood of publication.
Diet and physical activity-based interventions are extensively evaluated in randomised controlled trials (RCTs) for their effect on pregnancy outcomes (Figure 1). Gestational weight gain was reported in 80% of these trials; in one-third it was the primary outcome. However, data from less than three-quarters of the trials could be meta-analysed.

Objectives: The aim of our work was to investigate if there is an association between the quality of reporting of the primary outcomes, the statistical significance of the treatment effect and its magnitude in RCTs investigating diet and physical activity-based interventions in pregnancy.

Methods: We collected information on outcomes clearly defined as primary outcomes or used for power estimation in included trials. Outcome description, magnitude, and significance of the treatment effect (P value) and statistical method used were recorded. The precision of outcome description was assessed according to CONSORT requirements. We used multivariate analysis to evaluate the association between the covariates accounting for clustering of outcomes at the paper level, where more than one primary outcome was used.

Results and discussion: Systematic reviews with meta-analysis of RCTs are considered to be the highest level of evidence synthesis when assessing the effectiveness of interventions. Therefore, it is important to identify any trends that are shaping outcome reporting. We will provide a detailed description of the associations and discuss the implication of our findings.

Attachments: