Article Figures & Tables
Figures
- Figure 1:
Flow chart showing the Canadian Forces Cancer and Mortality Study II (CFCAMS II) cohort-building process. Note: CCPS = Central Computerized Pay System, HRMS = Human Resources Management System, MOSID = Military Occupation Structure Identification code.
- Figure 2:
Expected timelines for the Canadian Forces Cancer and Mortality Study II (CFCAMS II).
Tables
Variable Description Demographic information These data elements will be used by Statistics Canada to link the study cohort file to the Vital Statistics – Death Database and the Canadian Cancer Registry. Once they are linked, only date of birth and sex will be retained; name and social insurance number will be truncated, and a randomly generated unique identifier will be issued for each unique individual. Name Sex Date of birth Social insurance number Occupational information Rank(s) Initial rank and subsequent rank changes, as well as related date(s) of rank changes (promotions and demotions) will be included in the linked cohort file. Each rank change and corresponding date will be captured as their own variables. Command(s) Initial command (Regular Force or Class C Reserve Force) and subsequent command change(s), as well as related date(s) of command changes will be included in the linked cohort file. Each command change and corresponding date will be captured as their own variables. Element(s) Initial element (Army, Navy or Air Force) and subsequent element changes, as well as related date(s) of element changes will be included in the linked cohort file. Each element change and corresponding date will be captured as their own variables. Enrolment date(s) Start of a person’s employment as a military member. The first iteration of this variable will be used to ascertain whether a possible participant’s first enrolment date corresponds with the cohort’s date-related inclusion criteria. As personnel can leave and subsequently reenrol, as well as change between Regular and Reservist service, participants may have > 1 enrolment date. Each enrolment date will be captured as its own variable. Release date(s) End of person’s employment as a military member. As personnel can leave and subsequently reenrol, as well as change between Regular and Reservist service, participants may have > 1 enrolment date. However, it is possible that a person may have no release date; this means he/she was still in service at the time of study end. Each release date will be captured as its own variable. Reason(s) for release For each release date captured in the cohort file, there should be a corresponding reason for release (voluntary, involuntary, medical or unknown). In the event of multiple releases, each reason for release and its corresponding date will be captured as their own variables. Occupational history Information on participants’ military occupation (Military occupation code or Military Occupation Structure Identification code). Foreign posting and deployment history Start and stop date(s) of any international military operation. The information captured as part of this includes the operation name (if applicable), the start and stop date(s) and the location(s) of the posting/deployment. Derived information Person-years of service Variable derived with the use of enrolment and release dates to calculate cumulative person-years of service for all participants. Will also be used in analyses to control for effect of length of service on outcomes of interest. Time since release Variable derived only for participants with a terminal release date (i.e., no subsequent reenrolment date). Quantifies the time between release and death or censorship. Era of service Variable derived to control for era effects within the data. Era of service categories used elsewhere, as well as other stratifications based on salient evidence, will be evaluated for suitability and optimal statistical power. Deployment — dichotomous Dichotomizes participants into “have deployed” v. “never deployed” on international deployment. Deployment — cumulative time Variable derived by calculating a cumulative amount of time on international deployment until release, death or censorship. Deployment — frequency Variable derived by counting the number of discrete international deployments a participant will have been on until release, death or censorship. Deployment — specific deployment flag For specific deployments of interest (e.g., Afghanistan, Gulf War, Rwanda), a flag may be created to identify those who participated in specific foreign military operations. Variable Description Patient variables Patient record type Type of record (new, update, delete) Sex Date of birth Province or country of birth Province of diagnosis Date of death Underlying cause of death Autopsy confirming cause of death Code indicating whether cause of death from official death certificate takes into account autopsy findings Derived patient variables Vital status No. of tumours No. of tumour records belonging to patient record Tumour variables Tumour reference number Unique tumour identification number Method of diagnosis Date of diagnosis ICD-9 cancer code Source flag classification Indicates classification system used to code topography, histology and behaviour of tumour ICD-O-2/3 topography Site of origin of neoplasm – ICD-O-2/3 coding ICD-O-2 histology Histological description of neoplasm – ICD-O-2 coding ICD-O-2 behaviour Behaviour associated with histological description of neoplasm – ICD-O-2 coding Laterality Site-specific localization of tumour in paired organs or on side of body on which tumour originated (right, left, bilateral) ICD-O-3 topography Site of origin of neoplasm – ICD-O-3 coding ICD-O-3 histology Histological description of neoplasm – ICD-O-3 coding ICD-O-3 behaviour Behaviour associated with histological description of neoplasm – ICD-O-3 coding Grade, differentiation or cell indicator Describes system used to identify type of grade/differentiation/cell indicator Method used to establish date of diagnosis Code that specifies method by which date of diagnosis of tumour was established Diagnostic confirmation Most accurate diagnostic confirmation CS tumour size Largest dimension/diameter of primary tumour in millimetres CS extension Primary tumour growth within organ of origin or its direct extension into neighbouring organs CS tumour size/ext-eval Code indicating how “CS tumour size” and “CS extension” were determined and diagnostic methods used CS lymph nodes Site-specific code identifying regional lymph nodes involved with cancer at time of diagnosis CS reg nodes eval Code indicating how “CS lymph nodes” code was determined and diagnostic methods used Regional nodes examined Total no. of regional lymph nodes that were removed/examined by pathologist Regional nodes positive Exact no. of regional nodes examined by pathologist and found to contain metastases CS mets at dx Code identifying distant site(s) of metastatic involvement at time of diagnosis CS mets eval Code indicating how “CS mets at dx” code was determined and diagnostic methods used AJCC clinical T Site-specific code evaluating primary tumour clinically (T) and reflecting tumour size and/or extension AJCC clinical N Site-specific code identifying absence/presence of clinical regional lymph node (N) metastasis; describes extent of regional lymph node metastasis as recorded AJCC clinical M Site-specific code identifying absence/presence of clinical distant metastasis (M) AJCC pathologic T Site-specific code evaluating primary tumour pathologically (T) and reflecting tumour size and/or extension AJCC pathologic N Site-specific code identifying absence/presence of pathological regional lymph node (N) metastasis; describes extent of regional lymph node metastasis as recorded AJCC pathologic M Site-specific code identifying absence/presence of clinical pathological metastasis (M) AJCC clinical TNM stage group Site-specific code identifying anatomic extent of disease based on clinical T, N and M elements as recorded in TNM clinical T, N and M fields AJCC pathological TNM stage group Site-specific code identifying anatomic extent of disease based on pathologic T, N and M elements as recorded in TNM pathologic T, N and M fields AJCC TNM stage group Site-specific code identifying stage group when clinical/pathologic T, N, M values are incomplete and do not lead to a clinical/pathologic T, N, M group AJCC edition number Identified edition of cancer staging manual used to stage case Note: AJCC = American Joint Committee on Cancer, ICD-9 = International Classification of Diseases, 9th Revision, ICD-O = International Classification of Diseases for Oncology.
Measure Mortality data Cancer data Univariate/bivariate analyses Frequencies
Measures of central tendency (where appropriate)
Cross-tabulationsIncidence Age- and sex-adjusted rates per 100 000 person-years of observation Age- and sex-adjusted rates per 100 000 person-years of observation, new cases Standardized mortality ratio
Lexis diagramsStandardized incidence ratio
Standardized mortality ratio*
Lexis diagramsPrevalence NA Person-based prevalence, per 100 000 person-years of observation†
Tumour-based prevalence, per 100 000 person-years of observation†Survival Kaplan–Meier survival analysis
Cox proportional hazards regressionNA Pohar-Perme estimator of net survival Regression Poisson regression
Joinpoint models
Age–period–cohort models
In this issue
Article tools
Respond to this article
