Death and Marital Status: The Link Is Not What You Think

Staying single and staying alive

Posted Jul 22, 2010

Quick - who lives the longest? I'm talking about marital status here. Unless you've read some myth-busting book, you probably think the answer is married people. Some in the Marriage Mafia even push the misleading heuristic, "get married, live longer."

Previously (in this post and in Singled Out), I've made fun of one of the tricks the pro-marriage people use to claim that getting married makes you live longer, and that's to pretend that people who are divorced or widowed never did get married. So if married people seem to live longer than, say, divorced people, they use that as support for their argument that getting married makes you live longer. Some even use that as "evidence" that people shouldn't divorce - even though we can never know whether those people who divorced would have died even sooner if they had stayed together. Another Mafioso trick is to ignore the longest running study (started in 1921) in which the people who stayed single lived just as long as the people who married and stayed married.

When I (or others) report comparisons in which the divorced people have outcomes that are not as good as those of other groups, some people object. They are right that these differences should not be exaggerated. When the data show that the differences are small, or that they decrease over time, I note that. I also note the studies showing advantages of the divorced over other groups. Now I'm going to add another qualification: How do you define who counts as divorced?

I was alerted to this by a study published last year in which the authors set out to look more closely at the link between divorce and death. Participants were followed for about 40 years, from the early 1960s (when they were all older than 35) until the end of that century.

The first way the authors examined the relationship between marital status and death was to separate the participants by their marital status when the study first began - married, divorced/separated, widowed, always-single - and then chart the proportion who were still living over the course of the 40 years. (That's Figure 1A, if you can access the article.)

Here's the first finding I noticed (not highlighted anywhere in the article): When you compare the line showing the proportion of married people who are still alive at each point in time, to the same line for single people, they are so close together that you can't tell them apart. So, those people (all over age 35) who had always been single at the start of the study lived just as long as the people who started the study married.

The next comparison looks like the usual bad news for the divorced. The people who were divorced at the start of the study did not live as long as the people in all of the other marital status groups.

Some researchers would, upon discovering that, close their laptops and send their paper off to the journal. Not these authors.

Next, they looked at those people who started the study divorced and stayed divorced throughout. Compared to all of the other people in the study, they still had shorter lives.

Now comes still another comparison. Let's take all of the people who ever got divorced at any point during the study (not just those who were divorced at the very beginning, or those who started out divorced and stayed that way) and compare them to everyone else in the study.

Guess what? There's no difference whatsoever in how long the people in the two groups lived. To quote the authors, "the mere experience of a marital breakup produced no elevation in mortality risk." (That's Figure 1C - the two lines are right on top of each other.)

If you wanted to defend the oft-made claim that getting married makes you live longer, could you still do it, based on the results of this study? You'd probably try to make something of the fact that those who started out divorced and stayed that way had shorter lives than married people did. You could suggest that all those years without a spouse must have done them in. (The ones who got divorced later in the study, you might hypothesize, had more total years of marriage, and those extra years of marriage gave them extra years of life.)

The authors actually considered that argument and discarded it. Do you see why? Those who had always been single lived just as long as the married people. And in this study, those who started out widowed lived just as long as the married people, too. The always-single people and the people who started out widowed had just as many spouse-less years as those who started out divorced and stayed that way, but they lived just as long as the married people.

There was no way of explaining definitively (from the data that were collected) why one subset of divorced people (those who were divorced from the beginning and stayed that way) had shorter lives when others who divorced did not. The authors could only offer speculations. The one I found especially intriguing did not even make it into the abstract: The meaning of divorce has changed over time. It was much more stigmatized a half-century ago than it became in successive decades.

Suppose we start studying marital status now, and look at the longevity results 40 years from now. What will we find? Maybe that marital status just doesn't matter.

Oh, one last thing. Did you hear all about this study in the media? Neither did I.