Absolute risk estimation in a case cohort study of prostate cancer
MetadataShow full item record
Background: Life expectancy calculations play a major role in the clinical management of prostate cancer with harms generally outweighing the benefits of screening and treatment if life expectancy is less than 10 years. Comorbidities also affect treatment choice because prostate cancer has a long latency period and depending on their comorbidities, a patient may die with prostate cancer and not from it. Clinicians currently calculate life expectancy using population-based life table probabilities, which are based on age alone. The primary objective of this project was to develop a probability model based on both age and comorbidity in order to obtain absolute risk estimates of death from prostate cancer. Methods: Data was obtained from a population-based case cohort study of 2,740 patients diagnosed with prostate cancer between 1990 and 1998 in Ontario. Using Cox proportional hazards model, we used age and comorbidity status to predict survival probabilities. The median of the survival curve was used as a measure for the expected remaining lifetime. Results: We calculated the relative and absolute risk estimates of death from prostate cancer, death from other causes and death from all causes, using case cohort design methodology. Based on these results, we developed a clinically usable table for age and comorbidity score. Conclusion: We have developed an interactive web tool, for use by clinicians with prostate cancer patients , that computes survival probabilities and life expectancy.