Related Experiment Videos
A method for identifying abrupt changes in U.S. cancer mortality trends.
K C Chu1, S G Baker, R E Tarone
1Office of Special Populations Research, National Cancer Institute, Bethesda, Maryland 20892-7161, USA.
Cancer
|July 3, 1999
Summary
Cancer mortality rates declined for males in the 1990s, but trends varied by race and sex. Gaps in breast and colorectal cancer rates between Black and White individuals are widening.
Area of Science:
- Epidemiology
- Biostatistics
- Public Health
Background:
- Evaluating cancer trends over extended periods is crucial for understanding changes.
- Identifying the timing and magnitude of these changes offers valuable insights beyond short-term analyses.
Purpose of the Study:
- To analyze long-term cancer mortality trends from 1973-1995 for various anatomic sites.
- To identify significant change points and assess trends across different racial and gender groups.
Main Methods:
- Weighted piecewise linear regression analysis was applied to cancer mortality data.
- The natural logarithm of annual mortality rates was the dependent variable, weighted by annual cancer deaths.
- Bootstrapping residuals was used to assess the variability of estimated change points.
Main Results:
- For males, cancer mortality declined in the 1990s, driven by lung, prostate, and colorectal cancers (white males) and lung, esophageal, oral cavity, and prostate cancers (Black males).
- Total cancer mortality declined significantly for white females but not significantly for Black females.
- Significant declines in the 1990s were observed for white females in breast, colon, and rectal cancers, but not for Black females.
Conclusions:
- A novel method for detecting major shifts in cancer trends has been established.
- Widening disparities in breast and colorectal cancer rates between Black and White populations were identified.