Because marriage is likely to be both a cause and a consequence of health outcomes, research must disentangle the influence of selection from the true causal influence of marriage. Distinguishing between these two factors requires careful analysis and advanced statistical methods that have been absent from many studies. This review focuses on studies that provide the most reliable evidence on whether marriage has a causal influence on health outcomes.
The studies providing the strongest evidence use longitudinal data and examine the association between changes in health outcomes and transitions into and out of marriage. Studies of this type provide more convincing evidence of a causal relationship between marriage and health because sample members serve as their own control group, and the effect of marriage is measured by comparing their outcomes before and after marriage. This method avoids comparing two groups that may have different background characteristics — in particular, people who marry and people who do not — which may lead to misleading and inaccurate results.
Some health outcomes are not well suited for this type of analysis, however. For example, many physical health outcomes cannot be examined in this way, because changes can unfold over a long time and may not be apparent immediately after a marital transition. For this reason, the evidence on the effects of marriage on physical health is more limited and somewhat more speculative than evidence on the effects of marriage on other health outcomes examined in this review.