Further Concepts and Methods in Survival Analysis
Author Information
Author(s): Clark T G, Bradburn M J, Love S B, Altman D G
Primary Institution: Cancer Research UK/NHS Centre for Statistics in Medicine
Conclusion
The paper discusses various advanced methods and considerations in survival analysis, particularly addressing issues like missing data and informative censoring.
Supporting Evidence
- Categorizing continuous variables can lead to biased estimates and reduced ability to detect real relationships.
- Missing data can significantly reduce the power of survival analyses.
- Informative censoring can bias standard survival analysis methods.
Takeaway
This paper talks about how to analyze survival data better, especially when some information is missing or when patients drop out of studies.
Methodology
The paper reviews various survival analysis methods, including handling missing data and informative censoring, and discusses advanced techniques like time-dependent covariate methods.
Potential Biases
Informative censoring can introduce bias into survival analysis results.
Limitations
The paper notes that many advanced methods are rarely used due to the complexity and data requirements.
Statistical Information
P-Value
0.002
Statistical Significance
p=0.002
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website