Statistical aspects of prognostic factor studies in oncology
1994
Statistical Issues in Cancer Prognostic Factor Studies
Editorial
Evidence: low
Author Information
Author(s): R. Simon, D.G. Altman
Conclusion
Prognostic factor evaluations are complex and often unreliable, requiring careful planning and analysis.
Supporting Evidence
- Prognostic factor studies can be exploratory or confirmatory, with different phases indicating their purpose.
- Many studies do not state their hypotheses in advance, leading to unreliable results.
- Statistical power calculations are often neglected in prognostic studies, affecting their reliability.
Takeaway
This article talks about how studies that try to predict cancer outcomes can be tricky and need to be done carefully to be useful.
Methodology
The article discusses various types of prognostic studies and their design, focusing on statistical issues and guidelines for conducting these studies.
Potential Biases
Retrospective studies may have bias due to missing data and selection of patients based on outcomes.
Limitations
Many prognostic studies lack clearly stated hypotheses and often suffer from issues related to sample size and data completeness.
Want to read the original?
Access the complete publication on the publisher's website