Related Experiment Videos
Combining studies using effect sizes and quality scores: application to bone loss in postmenopausal women
1McGill University, Department of Epidemiology and Biostatistics, and Centre for Clinical Epidemiology and Community Studies, Jewish General Hospital, Montreal, Quebec, Canada.
Journal of Clinical Epidemiology
|October 8, 1998
Summary
This study introduces a new random effects model for meta-analysis, incorporating study quality scores to improve effect size estimation. The model effectively integrates physical activity research for postmenopausal women, yielding reliable results.
Area of Science:
- Biostatistics
- Epidemiology
- Gerontology
Background:
- Meta-analysis is crucial for synthesizing research findings.
- Heterogeneity in study quality can impact meta-analysis results.
- Physical activity's role in preventing bone loss in postmenopausal women requires robust evidence synthesis.
Purpose of the Study:
- To present a novel random effects meta-analysis model.
- To integrate study quality scores into effect size estimation.
- To evaluate the effectiveness of physical activity in preventing bone loss in postmenopausal women.
Main Methods:
- Developed a random effects model incorporating effect sizes and quality scores.
- Conducted a systematic literature search for relevant studies (1966-1996).
- Utilized modified Chalmers' scale for quality assessment and Hedges and Olkin for effect sizes.
Main Results:
- The new model estimated the effect of physical activity on spinal bone mineral density loss as ESoverall = 0.4263 (SE = 1.1361).
- The model incorporating quality scores produced narrower confidence intervals compared to the DerSimonian and Laird model alone.
- Random effects model was deemed more appropriate than fixed effects due to study heterogeneity.
Conclusions:
- Incorporating quality scores into meta-analysis enhances the quantification of between-study variation.
- The proposed model offers a more refined approach to synthesizing evidence, particularly in the presence of heterogeneity.
- This method provides a valuable tool for researchers assessing interventions in fields like osteoporosis prevention.