"Stay single, wind up dead” is a headline to file under clickbait if ever there was one. The TimesLive article in question cited a recently released report by Statistics South Africa, which asserted that single people - men in particular - are more likely to end up dead than their married, widowed or divorced counterparts.
The Stats SA report stated that only 24.4% of deaths in 2015 were of married people, whereas 47% of the people were filed under the “never-married” category. At first glance, this is a massively disproportionate number of deaths and likely to send us running to the first person that will have us.
The statistician-general casually speculated that without a spouse to hold single men accountable when they get home, they’re significantly more likely to go out and engage in risky, life-threatening behaviour.
This would seem like a fairly straight-forward case. However, possible causes for this effect aren’t investigated nor is the data used to make the claim appropriately presented.
Is correlation, in fact, being mistaken for causation?
Marital status unspecified for 16.2% of deaths
The first concern to raise is the way the comparison is being carried out. No attempt is made to compare the actual lifespans of married people with those of unmarried people.
In addition, it is up to the clinician’s best judgement whether the deceased was single, married, widowed or divorced. Blank entries on the South African notice of death form are captured as unknown/unspecified. Since the marital status of 16.2% of the 2015 deceased fell into this category, Stats SA itself warns that the results “need to be carefully interpreted”.
Furthermore, a number of minors will end up in the single category, somewhat skewing the data. A way to limit your study to a more eligible sample would be to focus on the adult population.
Researchers can only point to possible connection
Luckily for us, it's unnecessary to re-invent the wheel and repeat a whopping amount of analysis. The effect of marriage on longevity is an entire field of research on its own, spanning decades and yielding volumes of research.
The typical approach entails recruiting a few hundred to thousand participants and observing their mortality trends over many decades. The same is also often done for existing governmental statistics, which may see sample sizes enter into the millions. The roaring spoiler to all of this is that despite everything I’ve said so far, marriage does indeed seem to influence our mortality.
Many international studies, published in respected journals such as the Journal of Health Psychology, or journals under the banner of the British Medical Journal publishing group, do indeed support the conclusion that consistently married people - which is to say, those that haven’t divorced and remarried - are less likely to die in any given year. But these studies can only point to a possible connection as researchers have been unable to conclusively demonstrate how the two are linked.
The effect is more nuanced than one might assume. If marriage absolutely increases your lifespan, surely all marriages would yield a similar effect?
In a review of multiple studies, totalling a sample of 250,000 people, this has been shown not to be the case. Widowed people are expected to live as long as those that never married while divorced people have the greatest mortality risk of all.
So now we’re faced with a gamble: getting married may prolong your life, have no effect on mortality, or increase the likelihood of dying.
It may all come down to stress
Marriage seems to affect mortality statistics from two sides. On the one hand, you have selection pressure determining eligibility for marriage. On the other, marriage itself has a tangible effect.
Getting married is a competitive process. We need to outcompete other potentially suitable people and ultimately prove that we might be the best possible life-long mate for our partners, and them for us.
Health and personality factors are both important. Single people have been shown to be 5 times more likely to die from infectious diseases than their married counterparts.
This statistic has led researchers to wonder whether is it a case of never-married people being more prone to death. Or rather, is it harder to find a mate when you’re suffering from disease?
Those better adjusted to modern life and with more agreeable personalitiesare more likely to find a mate and enter the married pool. So in effect “marriage” is selecting for healthier, more stable people rather than necessarily creating them. Then again, that might exactly be what marriage is doing.
Ultimately, it may all come down to stress. Consistent emotional and physical support might be at the root of the benefits marriage may provide, whilst equally explaining why bad marriages have a negative effect on mortality.
When you have a supportive partner, one that is able to help you deal with the curveballs life throws at you, research shows us that you’re objectively able to cope in a superior manner to someone that doesn’t. The reverse holds true for unhappy marriages, which are ultimately a source of stress. Improved stress coping behaviours promotes longevity.
Taking care of other half’s health
Which leaves us with one final loose end to tie up, the idea that men in particular stand to gain a greater increase in longevity from being married. Although the verdict is still out, the statistician-general's theory doesn’t seem to be too far removed from the current thinking.
The idea is that married people not only feel motivated to maintain their own well-being for their partners but also play a role in ensuring that their partners take care of their own health. Women, in particular, seem to be more prone to actively exert an influence on the health of their partners.
Despite seeming outrageous at first glance, there does seem to be some truth in the claim that marriage prolongs life. The data in the Stats SA report is insufficient to gain significant insight into the local magnitude of this effect, but international research shows that marriage quality seems to be the true determinant of the effect, if any, on mortality.
Petrie Jansen van Vuuren is a medical student at the University of the Witwatersrand, with a postgraduate background in neuroscience research at the University of Pretoria.
This article was originally published on Africa Check. Follow them on Twitter: @AfricaCheck
Have something to say? Tweet or Facebook us on @Bhekisisa_MG
Work at a non-profit media house? Then you know your job is not just reporting anymore.
As deaths mount, take an inside look at the detectives working around the clock to solve the country’s medical mystery.
Until now, the national and provincial health departments have not been able to say where services are provided. Here's how we found them.
Bhekisisa means "to scrutinise" in Zulu
In South Africa, Zulu patients who would like to be thoroughly assessed by a doctor, would ask the physician to "bhekisisa" them.