The Canadian census mortality follow-up study, 1991 through 2001

Abstract

Background

An important step in monitoring progress toward reducing or eliminating inequalities in health is to determine the distribution of mortality rates across various groups defined by education, occupation, income, language, ethnicity, and Aboriginal, visible minority and disability status. This article describes the methods used to link census data from the long-form questionnaire to mortality data, and reports simple findings for the major groups. 

Data and methods

Mortality from June 4, 1991 to December 31, 2001 was tracked among a 15% sample of the adult population of Canada, who completed the 1991 census long-form questionnaire (about 2.7 million, including 260,000 deaths). Age-specific and age-standardized mortality rates were calculated across the various groups, as were hazard ratios and period life tables.   

Results

Compared with people of higher socio-economic status, mortality rates were elevated among those of lower socio-economic status, regardless of whether status was determined by education, occupation or income. The findings reveal a stair-stepped gradient, with bigger steps near the bottom of the socio-economic hierarchy.

Keywords

age-standardized mortality rates, longitudinal, non-institutional, proportional hazards, record linkage, socio-economic, survival analysis

Authors

Russell Wilkins (1-613-951-5305; Russell.Wilkins@statcan.gc.ca) is with the Health Information and Research Division at Statistics Canada, Ottawa, Ontario K1A 0T6; Michael Tjepkema (1-416-952-4620; Michael. Tjepkema@statcan.gc.ca) is with the Health Information and Research Division at Statistics Canada in Toronto, Ontario; Cameron Mustard is with the Institute for Work and Health and the University of Toronto, Toronto, Ontario; and Robert Choinière is with the Institut national de santé publique du Québec, Montreal, Quebec.

Findings

A goal of Canadian health policy is to reduce or eliminate socio-economic inequalities in health. An important step in achieving this goal is to determine the distribution of health status across groups defined by income, education, occupation, language and ethnicity, Aboriginal or visible minority status, and disability status. Each of these characteristics must be directly addressed in terms of the most fundamental aspects of health: life or death, and relative risks of premature death from various causes. [Full text]

Acknowledgements

We are grateful to the former Occupational and Environmental Health Research Section of the Health Statistics Division for carrying out the record linkages, thanks to the work of Bryan LaFrance, Lana Marjama, Peggy Cyr, Pierre Lalonde, Marie Beaudet, Christel LePetit, and Martha Fair. We thank Evelyne Perkins of the Record Linkage Resource Centre (for linkage software assistance), Paul Hartung of the Data Access and Retrieval System (for census long-form data), and the following members of the Health Analysis and Measurement Group (now Health Information and Research Division): François Gendron, Julie Bernier, Jean-Marie Berthelot, and Jillian Oderkirk. Michael Wolfson, Assistant Chief Statistician, provided moral and material support for this work since its inception. This study benefited from earlier work on the design of census-mortality linkages by Howard Newcombe and Martha Smith, Elizabeth Coppock, Loraine Marrett, Ruhee Chaudhry, and Kristan Aronson. We are especially grateful to the Canadian Population Health Initiative, which funded the creation of the census mortality follow-up study, for their patience and perseverance in the face of numerous delays. The Canadian Institutes of Health Research, the Ontario Workplace Safety and Insurance Board, and the Institut national de santé publique du Québec contributed to other analyses of the linked data. Finally, we would like to acknowledge Canada's provincial and territorial registrars of vital statistics, who furnish the death data for the Canadian Mortality Data Base.