Publications

    Canadian Forces Cancer and Mortality Study

    Boxes

    Warning View the most recent version.

    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.

    Box 1 – Released Definition

    In this study "released" personnel are individuals with some military service in Canada's Regular Force between 1972 and 2006, but not in the Canadian Forces on December 31, 2006.

    Box 2 – Out-of-Country Deaths

    Statistics Canada does not systematically collect information on deaths of Canadian citizens that occur outside of Canada; therefore there is no baseline to compare out-of-country military deaths. Consequently, this study does not analyse CF deaths that occurred out of country such as combat related deaths.

    Box 3 – Methods and Concepts

    Record linkage:

    The final linked database containing both human resources information on CF personnel and causes of death was obtained through probabilistic record linkage of the combined DND historical (1972 to 1997) and current (1998 to 2006) administrative databases, to the Canadian Mortality Data Base (CMDB) records of death that occurred between January 1, 1972 and December 31, 2006. Prior to the CMDB linkage, the DND administrative records were also linked to the Historical Tax Summary File. This file contains information about the years individuals have filed an income tax report, whether there was notification of the death of the filer or whether the filer moved out of the country. The pre-match to tax filer information provided some indication as to the vital status of the individuals at the end of the follow-up period and aided in the identification and confirmation of deaths.

    Standardized Mortality Ratios (SMR):

    Standardized Mortality Ratios (SMR) were used to compare the mortality of the CF CAMS cohort with that of the general Canadian population. All-cause and cause-specific mortality rates were calculated separately for males and females, each year of follow-up, and each age group in the CF CAMS cohort and compared with the corresponding group in the general population using indirect standardization methods. In compliance with the requirements of the Statistics Act, all counts of deaths < 10 were suppressed. Therefore, SMRs were reported only for causes (by ICD chapter) if the sex specific counts of deaths across the 35 year study were greater than ten. Deaths were reported using three different versions of the World Health Organization's International Classification of Diseases (ICD) throughout the 35 year study. For consistency, all deaths classified using ICD-8 and ICD-9 where mapped onto ICD-10 categories prior to SMR analysis.

    An SMR value of 1.0 indicates that the observed mortality in the cohort was the same as that observed in the Canadian population. Values less than 1.0 suggest lower mortality in the cohort while values greater than 1.0 suggest higher than expected mortality in the cohort. In addition to the point estimate, 95% confidence intervals were calculated using either normal approximation (count of deaths 100) or the exact Poisson method (count of deaths < 100).

    Proportional hazards model:

    Cox's semiparametric proportional hazards model was used to identify risk factors associated with suicide in the released CF CAMS cohort. The exponentiated coefficients of a Cox regression are hazard ratios and can be interpreted in terms of relative risk compared to a reference category. If, for instance, women were the reference category, a hazard ratio of 2.0 for men would imply that the relative risk of death was twice as high among men as it was among women. Conversely, a hazard ratio of 0.5 for men would mean that men's risk of death was half that of women. A hazard ratio of 1.0 would indicate that the risks were identical for both men and women.

    Reasons for release:

    Reasons for release were classified into three categories (voluntary, involuntary, medical). Voluntary reasons included service completed, immediate annuity, fixed service duration, retirement age. Involuntary reasons included not advantageously employed, unsuitable for further service, unsatisfactory conduct or performance, irregular or fraudulent enrolment, illegally absent, misconduct, dismissal, and reduction in strength. Medical reasons are the result of an administrative decision to release an individual owing to medical employment limitations.

    Date modified: