Study: Conditional survival analyses by cancer type

View the most recent version.

The predicted prognosis for people diagnosed with cancer between 2004 and 2006 tended to improve, sometimes substantially, after one or more years' survival.

For the first time in Canada, five-year survival ratios for a large number of cancers were estimated for people who had already survived from one to five years since diagnosis. These estimates were based on records from the Canadian Cancer Registry linked to the Canadian Vital Statistics Death Database.

The estimated five-year relative survival ratio for cancer of the pancreas is 6% at diagnosis, but among one-year survivors is 28%, and among five-year survivors, is 88%. Similarly, among those surviving esophageal cancer for at least five years, the estimated relative survival ratio is 83%, although the initial prognosis is 13%.

For five-year survivors from colon cancer, a cancer with an initial estimated five-year relative survival ratio of 63%, the updated prognosis is 97%.

There was no apparent improvement in survival prospects during the first five years after diagnosis for chronic lymphocytic leukemia, the estimated five-year relative survival ratio remained just under 80%.

Note: Relative survival is a measure of survival which compares the observed survival of those with cancer to the expected survival for comparable people in the general population.

Conditional survival is a measure of survival based solely on cases that have met a certain condition such as survival for one year after diagnosis.

Definitions, data sources and methods: survey number 3207.

The article, "Conditional survival analyses across cancer sites," which is part of today's online release of Health Reports, Vol. 22, no. 2 (82-003-X, free), is now available from the Key resource module of our website under Publications. For information about this article, contact Larry Ellison (613-951-5244; larry.ellison@statcan.gc.ca), Health Statistics Division.

Today's release includes one other article.

"Breast cancer incidence and neighbourhood income" uses data from the Canadian Cancer Registry to calculate national age-specific and age-standardized incidence rates of breast cancer from 1992 through 2004 by neighbourhood income quintile and region. For more information about the article, contact Marilyn J. Borugian (604-675-8058; mborugian@bccrc.ca), British Columbia Cancer Agency.

For information about Health Reports, contact Janice Felman (613-951-6446; janice.felman@statcan.gc.ca), Health Analysis Division.