Meta-analytic methods for health services research--an example from geriatrics

Area of Science:

• Gerontology
• Health Services Research
• Evidence-Based Medicine

Background:

• A prior meta-analysis examined controlled trials of comprehensive geriatric assessment (CGA).
• Findings indicated CGA efficacy is linked to patient characteristics, program objectives, and design.
• Specific program models and features correlate with improved health outcomes.

Purpose of the Study:

• To detail the methods used in the CGA meta-analysis.
• To explain method development considering trial data and scientific context.
• To highlight the utility of specific methodological approaches.

Main Methods:

• Surveyed primary trialists to retrieve unpublished data and standardize information.
• Developed a program typology to structure the main analysis.
• Integrated program design features as covariates to address statistical heterogeneity.

Main Results:

• The chosen methods, including data recovery and typology development, were highly effective.
• Incorporating design features as covariates successfully managed heterogeneity.
• These approaches offer valuable insights for future systematic reviews.

Conclusions:

• Methodological rigor is crucial for systematic reviews of complex interventions.
• The developed methods enhance the reliability of findings for CGA and similar health services research.
• This work provides a framework for evaluating multifaceted healthcare interventions.