Impact of income on mortality in urban Canada

The gap in life expectancy at birth between the poorest and richest neighbourhoods in Canada's urban areas narrowed substantially from 1971 to 1996, according to a new study that examines the impact of income on mortality.

In 1971, the disparity in life expectancy between the 20% of people in urban neighbourhoods with the lowest incomes and the 20% in those with the highest incomes was more than 6 years for men, and nearly 3 years for women.

During the next quarter-century, life expectancy improved substantially for all income levels. However, the gains were larger for those in the lowest income neighbourhoods than for those in the highest.

Consequently, by 1996, the gap in life expectancy between the lowest and highest income neighbourhoods was down to 5 years for men, and considerably less than 2 years for women.

This study also found similar improvements in levels of infant mortality and a considerable improvement in the probability of survival to age 75.

In addition, socio-economic disparities fell markedly over time for most causes of death. However, there was little improvement for some causes, while a few, such as lung cancer for females, showed clearly widening disparities.

This study found that socio-economic differentials in mortality are still of major importance in Canada, despite considerable progress in many areas.

Gains in life expectancy greatest in poorest neighbourhoods

For the purposes of this study, the population in census tracts, or neighbourhoods, of Canada's major urban areas was ranked by income and categorized into five groups based on income levels. Each group, therefore, represented 20%, or one-fifth, of the urban population.

This study, in line with many previous studies, observed that the poorer the neighbourhood, the shorter the life expectancy of its residents at birth. This applied to both sexes together and for men in all years, as well as for women in 1971.

For both men and women in all years, the poorest neighbourhood income group was particularly disadvantaged. Nevertheless, life expectancy rose substantially for all income groups from 1971 to 1996.

However, these gains were greater among the poorest neighbourhood income groups than they were among the richest. For men in the poorest neighbourhoods, life expectancy at birth rose 6.0 years from 1971 to 1996, compared with 4.7 years for men in the richest neighbourhoods.

For women in the poorest neighbourhoods, life expectancy at birth increased 3.8 years in the same time period, compared with only 2.6 years for those in the richest neighbourhoods.

The difference in life expectancy between the richest neighbourhoods and the entire population reveals how much the population as a whole would gain if the mortality rates of the richest neighbourhoods applied to everyone.

In 1971, that difference was nearly 3 years for men, and almost 17 months for women. By 1996, the difference had narrowed to 2 years for men, and just 6 months for women.

Infant mortality: Gap between poorest and richest neighbourhoods narrows markedly

Infant mortality rates - deaths before the age of one - declined during the 25-year study period in each of the five income groups. Once more, gains were greater among the poorest neighbourhood income groups.

In 1971, the gap between the richest and poorest neighbourhoods was 9.8 infant deaths for every 1,000 live births. By 1996, the gap had declined to only 2.4 deaths.

In 1996, infant mortality in Canada's poorest neighbourhoods, 6.4 deaths for every 1,000 live births, was considerably lower than the national rate for the United States (7.8). However, the rate in Canada's richest neighbourhoods was no better than Sweden's national rate (4.0).

If the rate in Canada among the richest neighbourhoods had applied to all urban neighbourhoods, as well as non-metropolitan areas, Canada would have had about 2,000 fewer infant deaths in 1971, and about 500 fewer in 1996.

Probability of survival to age 75: Considerable improvement across all income groups

In 1996, 53% of men in the poorest neighbourhood income group were expected to survive to the age of 75, compared with 69% of men in the richest group. Among women, 73% in the poorest group were expected to survive to 75, as opposed to 80% in the richest.

During this 25-year period, there was about the same improvement in the probability of surviving to the age of 75 across all income groups. Consequently the gaps between rich and poor neighbourhoods persisted at about the same magnitude.

Premature mortality related to income still of major concern in Canada

A useful way of examining premature death from various causes is to calculate potential years of life lost due to deaths before age 75. Excess potential years of life lost before age 75 show the public health impact of income-related disparities in premature mortality.

If all income groups had experienced the mortality rates of the richest group, and the same rates had also applied to rural and small town Canada, then 13,000 fewer men and 5,000 fewer women would have died before the age of 75 in 1996.

In 1971, young people under the age of 15 accounted for 39% of such excess deaths. By 1996, mortality at younger ages had declined to such an extent that this age group accounted for only 12% of those excess deaths.

In terms of potential years of life lost before age 75, in 1996 the burden of income-related excess mortality was greater than that due to all injuries or circulatory diseases.

These findings show that though all Canadians have shown significant improvement in life expectancy over time, with the greatest gains among those in the poorest neighbourhoods, there is still some way to go in terms of income-related disparities.

Causes of death: General pattern of lower mortality for all income groups

In terms of specific causes of death, trends from 1971 to 1996 show a general pattern of lower mortality for all income groups, for both sexes and for most causes of death.

Throughout this period, the most common pattern was a gradient in which the richest neighbourhood income groups had the lowest mortality, and the poorest had the highest. The gradients generally persisted over time, although they tended to be less steep in recent years, particularly for women.

For most causes of death, socio-economic disparities in mortality diminished markedly over time. These causes included ischemic heart disease, most injuries, cirrhosis of the liver and perinatal conditions. For those causes, mortality rates declined over the 25-year study period, and differences among income groups narrowed.

Some causes of death showed little change in socio-economic disparities: lung cancer and prostate cancer for men, and breast cancer for women.

Finally, a few other causes showed clearly widening disparities: lung cancer among women, mental disorders and diabetes for both sexes. Mortality rates for lung cancer increased rapidly for women in all income groups, and the gap between rich and poor widened, reflecting previous increases in rates of smoking among women. From 1986 on, the rates in the poorest neighbourhoods were much higher than those in other income groups.

Information on methods and data quality available in the Integrated Meta Data Base: survey number 3233.

The study Trends in mortality by neighbourhood income in urban Canada, 1971 to 1996 (82-003-SIE, free) is now available on Statistics Canada's Web site (http://www.statcan.gc.ca). From the Our products and services page, choose Free publications, then Health.

For more information, or to enquire about the concepts, methods or data quality of this release, contact Russell Wilkins (613-951-5305; wilkrus@statcan.gc.ca) or Jean-Marie Berthelot (613-951-3760; berthel@statcan.gc.ca), Health Analysis and Measurement Group.