One-stage parametric meta-analysis of time-to-event outcomes
2010

Meta-Analysis of Time-to-Event Outcomes

Sample size: 12 publication 10 minutes Evidence: moderate

Author Information

Author(s): Siannis F, Barrett J K, Farewell V T, Tierney J F

Primary Institution: Department of Mathematics, University of Athens

Hypothesis

Can percentile ratios provide a better alternative to hazard ratios in meta-analysis of time-to-event outcomes?

Conclusion

The study suggests that percentile ratios can effectively summarize treatment effects in time-to-event outcomes.

Supporting Evidence

  • The overall hazard ratio for time to death was found to be 1.18.
  • The proposed methodology was illustrated using glioma data.
  • Likelihood inference for percentile ratios was outlined.

Takeaway

This study looks at how we can compare survival times from different studies using a new method called percentile ratios instead of the usual hazard ratios.

Methodology

The study proposes a parametric approach for meta-analysis of individual patient data with survival endpoints, using percentile ratios as an alternative to hazard ratios.

Potential Biases

The study acknowledges potential biases in the selection of trials and data extraction.

Limitations

The study could not obtain permission to use data from 3 of the 12 trials.

Participant Demographics

Patients with high-grade glioma from 12 randomized controlled trials.

Statistical Information

P-Value

0.18

Confidence Interval

95% CI 1.09–1.28

Statistical Significance

p=0.18

Digital Object Identifier (DOI)

10.1002/sim.4086

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication