Cancer Deaths and the Lognormal Model
Author Information
Author(s): J. W. Berg
Primary Institution: Memorial Hospital for Cancer and Allied Diseases
Hypothesis
The lognormal model can accurately describe the distribution of cancer death times.
Conclusion
The lognormal model often fits cancer death time distributions well, particularly in treated patient populations.
Supporting Evidence
- The lognormal model was found to fit the data well in many cases.
- Only half of the groups studied showed uniform lognormality.
- Groups with more than fifty cancer deaths tended to fit the lognormal model better.
- Treatment type significantly influenced the applicability of the lognormal model.
- Non-conformity to the model often revealed important patient or cancer differences.
Takeaway
This study looked at how cancer deaths happen over time and found that a specific math model can help us understand it better.
Methodology
The study analyzed cancer death data from various patient groups, focusing on treatment types and survival rates.
Potential Biases
Potential bias due to selection of patient groups and treatment types.
Limitations
The model may not apply to all cancer types or patient groups, particularly those with mixed treatment outcomes.
Participant Demographics
Included various cancer types and treatment histories, totaling about 5000 patients.
Want to read the original?
Access the complete publication on the publisher's website