Abstract
A sample of 1200 breast cancer patients with complete follow up from 1978 to 2002 was used to assess the evolution survival of patients over time using several survival estimators: Kaplan and Meier´s, Cox´s proportional hazards regression, multinomial logit regression model, median-quartiles residual life time. Patients were classified into groups according to risk factors and results for the groups were compared through standard ad-hoc tests and confidence bands, obtained using bootstrap re-sampling methods. The analysis of effect of risk factors considered only non metastatic patients since the presence of metastasis overcome all possible factors affecting survival. The results showed that patients at stage three of the disease present a differential result in survival when considering factors such as age at detection and age at first child birth, which are considered also as risk factors.
| Original language | English |
|---|---|
| Pages | 1-15 |
| Number of pages | 15 |
| State | Published - 01 Dec 2015 |
| Event | Joint Statistical Meetings 2025 - Seattle, United States Duration: 08 Aug 2015 → 13 Aug 2015 https://ww2.amstat.org/meetings/jsm/2015/ |
Conference
| Conference | Joint Statistical Meetings 2025 |
|---|---|
| Abbreviated title | JSM2025 |
| Country/Territory | United States |
| City | Seattle |
| Period | 08/08/15 → 13/08/15 |
| Internet address |
Keywords
- Breast cancer
- survival estimation
- product limit estimator
- proportional hazard models
- residual median lifetime
- bootstrap
