Background
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.
The prognosis of a cancer patient is influenced by host factors (e.g., age, sex, ethnicity, risk of death from other diseases), tumour-related factors (e.g., extent of disease, histological subtype) and factors related to the cancer control program for that population (e.g., availability and quality of diagnosis and treatment services, quality and coverage of screening services). Because population-based survival estimates are based on the experiences of a highly heterogeneous group of people, they are useful "average" outcome indicators of the efficiency of health services and can be used for comparative purposes between populations or over time. 1 Comparison of survival estimates can help to identify priorities and suggest measures to improve patients' survival. 2
Observed survival is defined as the proportion of patients alive after a given length of follow-up. A disadvantage of using this measure is that deaths from causes other than the disease of interest lower the observed survival proportion and preclude comparison of results between groups experiencing different general mortality. 3 For example, it is difficult to compare observed survival proportions between age groups, since an older person is more likely to die of other causes than a younger person. What is desired is an objective measure of the proportion of patients dying from the direct or indirect consequences of disease in a given population. That is, a measure of patient survival corrected for the effect of other independent causes of death. 4 , 5 This measure has been deemed net survival and has been defined as the proportion of patients who would have survived a given period of time or longer following diagnosis in the hypothetical situation where the disease of interest was the only possible cause of death. 3 , 6 Net survival can be estimated by using either the cause-specific survival proportion or the relative survival ratio.
Cause-specific survival considers only those deaths that can be attributed to the cancer in question as events, while all other deaths are considered as censored observations. It is practical to use in clinical studies where more effort is used to distinguish between deaths due to the disease under study and deaths due to competing causes. However, there is no fully satisfactory method of deciding whether a given death should be classified as a cancer or as a "non-cancer" death. Even if such a method existed, the necessary information, if available, would rarely be sufficiently accurate to make such a classification possible when dealing with population data. 5 Given these limitations, relative survival was proposed. 4 , 7 A major advantage of this measure is that information on cause of death is not required.
In the context of cancer, relative survival is defined as the ratio of the observed survival for a group of individuals diagnosed with cancer to the survival that would have been expected for members of the general population who have the same main factors affecting survival (e.g., sex, age, area of residence) as those with cancer. 7 It can be interpreted as the proportion of patients alive after a given length of follow-up in the hypothetical situation where the cancer in question is the only possible cause of death. The validity of this interpretation is dependent on the accurate and appropriate estimation of expected survival and the assumption that non-cancer mortality is independent of cancer mortality.
- Date modified: