Performing meta-analysis with incomplete statistical information in clinical trials
2008

New Methods for Meta-Analysis with Missing Data

publication Evidence: moderate

Author Information

Author(s): Ma Jianbing, Liu Weiru, Hunter Anthony, Zhang Weiya

Primary Institution: Queen's University Belfast

Hypothesis

Can new methods effectively estimate missing standard errors in clinical trial meta-analysis?

Conclusion

The prognostic and interval methods are effective alternatives for handling missing data in meta-analysis.

Supporting Evidence

  • The prognostic method predicts missing standard errors from known data.
  • The interval method provides a range for the missing standard errors.
  • Both methods were tested on clinical trials for Type-2 diabetes and intraocular pressure reduction.

Takeaway

This study shows how to fill in missing information in clinical trials so that we can still analyze the results. It's like guessing the missing pieces of a puzzle to see the whole picture.

Methodology

The study developed and tested two methods to estimate missing standard errors from clinical trial data.

Potential Biases

Potential bias if the missing data are not missing at random.

Limitations

The methods may not perform well if the input data are imprecise.

Digital Object Identifier (DOI)

10.1186/1471-2288-8-56

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication