Data quality, concepts and methodology: Interpretation
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Population-based cancer survival estimates are based on the experiences of a heterogeneous group of people and as such are useful general outcome indicators. They do not necessarily reflect a person's chances of surviving five years after diagnosis nor do the accompanying confidence intervals represent the range of possible prognoses for these individuals. The results presented here reflect the survival experience of those diagnosed from 1992 to 2003.
The observed five-year survival proportions should be interpreted as the percentage of eligible cases, diagnosed in a given time period, that were still alive five years after being diagnosed with a particular cancer. Relative survival compares the observed survival for a group of individuals diagnosed with cancer to the expected survival for members of the general population of the same age, sex, province of residence, and time period. A five-year relative survival ratio of 80% for a particular cancer means that people diagnosed with that cancer had 80% of the likelihood of living for 5 years after diagnosis compared to similar people in the general population.
In theory, relative survival ratios greater than 100% indicate that the observed survival of people with cancer is better than that expected from the general population. In these instances it could be that the persons diagnosed with cancer experienced lower mortality from other causes of death because of a greater than usual amount of interaction with the health care system. However, estimates of relative survival over 100% should be interpreted with caution as several other factors may be at play including random variation in the observed number of deaths, failure to register some cancer patient deaths, and imprecision in the estimation of expected survival.
Ideally, lung cancer cases would also be matched by smoking status to members of the general population, because most people diagnosed with lung cancer are smokers or ex-smokers and smoking is known to reduce life expectancy. However, life tables by smoking status were not available. While lung cancer relative survival ratios would likely have been higher if life tables by smoking status were available, a previous study found that adjusting the expected survival for the excess mortality related to smoking increased estimates of relative survival by 1% or less. 1
In the years under study, there was a known under reporting of cancer cases in Newfoundland and Labrador as a consequence of the cancer registry not receiving death certificate information from the provincial vital statistics office. There is likely to be some overestimation of survival for this province as the survival of such "missed" cases is generally less favourable than that of cases in the registry population. 2 As such, survival estimates from the province of Newfoundland and Labrador should be treated with caution.
Age-specific and 'all ages' (i.e., 15 to 99) survival estimates provide information on the actual survival experience, observed or relative, of those diagnosed with cancer. However, because survival estimates vary with age and the age distribution of cancer cases can vary over time and between geographic areas, it is usually preferable to use age-standardized survival estimates to compare survival across time, across provinces, or between a province and Canada as a whole. Age-standardized survival estimates have been provided and are interpretable as the overall survival estimate that would have occurred, if the age distribution of the cancer group under study had been the same as that of the standard population.
The number of cases provides an indication of the precision of the survival estimate. In general, the greater the number of cases, the more precise the survival estimate. When the estimates are based on a small number of cases, it is more likely that observed differences are due to random, rather than systematic, influences. 3
Age-specific and 'all ages' (i.e., 15 to 99 years) estimates based on fewer than 10 cases were suppressed. Age-standardized estimates were suppressed if two or more of the age-specific estimates used in their calculation were based on fewer than five cases or one of the age-specific estimates was based on fewer than five cases and the total number of cases for all age groups combined was less than 100.
- Date modified: