Use the Table of Summaries to link to the summaries either by author or by risk factor.
Abstract | Bibliography | Introduction | Table of
Summaries
Mortality Rates By Marital Status: A Discussion
Brown, Robert L., and John Di Meo. Research Report 95–07, Institute of Insurance and
Pension Research, University of Waterloo, Ontario, Canada. †
This paper was written in response to one by C. Trowbridge (T.S.A. XLVI, pp 321–
345) in which it was shown that marital status mortality differentials exist. Married people have lower death
rates than those who are not married and the differentials are increasing. Those who are widowed or divorced
tend to experience the highest death rates. It has also be found that the disparity in marital status
mortality is usually more pronounced for males.
This paper also comprises a literature review of some thirty related articles. This
review was included as an appendix to the paper. The review touches on socio–economic status, gender,
education and income as factors that are correlated with both marital status and mortality. The main paper
highlights the pertinent issues from the review to provide a useful overview of the marital status mortality
differentials and the mechanisms involved.
The paper aims to evaluate if the support provided by marriage is the reason for the
reduced mortality experienced by married individuals. It appears that the protection offered by marriage to
males is in it social support, but for females it operates mainly through economic factors. The author
focuses on this 'protection' aspect of marriage. It has been suggested that in addition to the presence of
social ties, the perception of having social support is also important. In addition, both income and marital
status have been shown to be correlated to mortality and the author queries, if they are both indicators of
well–being and security, would the provision of social security also enhance life expectancy? That is,
could the provision of Social Security impact mortality in a manner similar to marriage?
Generally, there are two reasons proposed for these mortality differentials: the
selection of marriage theory and the theory of social support. The selection of marriage theory suggests that
the physically and psychologically unhealthy are predisposed to higher mortality and are also less likely to
be chosen into marriage. The theory of social support suggests that the obligation, expectations and
responsibilities of marriage lead to a certain level of protection. This implies that marriage restricts
the extent of unhealthy behaviors such as smoking and alcohol consumption, and that those who are married
generally take better care of themselves and they have a care–giver if needed.
The author proffers that the selection theory is likely to have some effect, but that it
would only be small and that it would be difficult to quantify. Evaluation of this theory would require
further consideration to be given to the marriage process. Indeed, if the effect was significant, it should
extend to the widowed, persist throughout all causes of death and apply equally to males and females. It has
been shown that none of these apply.
It has been suggested that marriage leads to a different lifestyle due to the reduction
in unhealthy behaviors and the increased means for social integration that it provides. In turn, this
improves sources of information and advice, emotional bonding and reduces stress. This proposed lifestyle
change is supported by the fact that deaths based on social and psychological factors are lower in married
people while there is no significant difference between marital states with regard to mortality which is
largely unaffected by social factors. Marriage generally provides less protection to females. It is suggested
that the lifestyle change is less significant for females as they are more likely to maintain social
networks and generally lead less risky lives, regardless of marital status. This theory agrees with evidence
that mortality rates increase for those leaving the married state.
From the literature review, it was stated that the social network provides information on
good health care, it constrains its members to healthy activity and provides support and care in and of
itself. Although lack of a social network and its intimacy was deemed stressful, it was also argued that
social support must be of high quality to be helpful. Therefore situations where couples can offer mutual
and (relatively) equal support seem to be positive.
In order to determine if the provision of social security reduces mortality in a similar
manner to marital status, the US data on mortality ratios by marital status from the Trowbridge report is
compared to corresponding Canadian data for 1960 and 1980. Within this period, universal Medicare, the
Canada/Quebec Pension Plan, the Guaranteed Income Supplement and the Spousal Pension Allowance were all
established in Canada. Therefore, in 1960 Canada was less protective than the US and was much more protective
than the US by 1980. Between 1971 and 1986 the disparity in mortality rates between the top and bottom
income quintiles in Canada decreased for both males and females.
It would be expected that the importance of being married in Canada in 1980 would be
significantly less important than in 1960, that is there would be less improvement amongst married persons
versus the unmarried in that period. The mortality ratios by marital status in 1960 in Canada would be
expected to be similar to or greater than those in the US and that they would be less in 1980. However, none
of these effects were observed, in fact the opposite was observed.
Therefore, it was concluded, there is no evidence that the provision of additional social
security in Canada has enhanced life expectancy. This would suggest that state support is not an adequate
proxy for the benefits received from marriage.
† This study is based on data from the and Canada.