Marriage and Mortality

Publisher Website
Google Scholar
Abstract
The 1950 census data and the 1949-1951 mortality data for the United States were reviewed in an effort to analyze differences in mortality rates observed among married, single, widowed and divorced persons. As an alternative to the commonly used mortality ratios, the "increased risk of death" observed in the nonmarried groups was estimated by an index related to the "population at risk." By this index, the recorded excess mortality among nonmarried persons increased with age and was greater for males than for females and for nonwhites than for whites. There was no consistent disadvantage for any particular marital class among the 3 nonmarried groups: single, widowed or divorced. A large part of this excess mortality is probably a result of the fact that healthy persons are more likely to marry or to remarry than are sick persons. There is strong evidence of unequal errors in the census classification and of possible misclassification in death certificates which may also be responsible in part for the data. The possible role of other factors cannot be conclusively demonstrated from population mortality data of this type. It is emphasized that mortality ratios do not necessarily go up or down in a simple, regular fashion, but are influenced by many factors. Therefore, it may be fallacious to base conclusions on trends in these ratios.

This publication has 6 references indexed in Scilit: