Handling Missing Data in Breast Cancer Survival Analysis
Author Information
Author(s): Ali A M G, Dawson S-J, Blows F M, Provenzano E, Ellis I O, Baglietto L, Huntsman D, Caldas C, Pharoah P D
Primary Institution: University of Cambridge
Hypothesis
The study investigates the effectiveness of different methods for handling missing data in survival analysis of breast cancer.
Conclusion
The study found that multiple imputation with the outcome included generally provided the most accurate estimates, while complete case analysis was less precise.
Supporting Evidence
- Over half the cases had missing data on at least one of the seven variables.
- 11 percent had missing data on 4 or more variables.
- Multiple imputation with the outcome included was found to be the least biased and most accurate method.
Takeaway
When researchers have missing information about patients, they can use different methods to fill in the gaps. This study shows that one method usually works best for breast cancer data.
Methodology
The study pooled data from over 11,000 breast cancer cases and compared four methods for handling missing data using Cox regression analysis.
Potential Biases
The complete case analysis may introduce bias if the missing data are not missing completely at random.
Limitations
The study's findings may not be generalizable to all types of missing data scenarios.
Participant Demographics
The study included early-stage breast cancer patients from five different studies.
Statistical Information
P-Value
p<0.0001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website