Friday, December 07, 2012

No children? You're Gonna Die Young!


Here's an interesting Danish study and its popularizations for you!  When I was going through several of the popular media summaries I had to go and bang my head against the garage door, to get some relief from the inanity of much of it.  To see why:

Let's start with some of those popularizations:

First:

By Maggie Fox, NBC News
They may feel like they are going to be the death of you, but having kids may not take years off your life. In fact, a study published Thursday shows that people who tried but failed to have children are two to four times more likely to die young than parents.
The startling findings support other research that shows childless couples don’t live as long as parents. They also add another piece to the puzzle: This one looks at people who either adopted children, or had them through fertility treatments.

Second:

People who have never had children are more likely to die prematurely and develop mental problems than those who have had children, researchers from Aarhus University, Denmark, reported in the Journal of Epidemiology and Community Health.

The authors added that the link between shorter lifespans, mental health and childlessness is particularly noticeable among women.

The belief that having children may shorten our lives is a myth, the researchers explained, as the reverse seems to be the case.

Third:

Dec. 5, 2012 -- Many parents might good-naturedly scoff at the notion, but a new study shows that being a parent may help you live longer.
Danish researchers compared men and women who had children with those who did not to see if the childless were more likely to die early.
They were. "Childless couples are at increased risk of dying early of all causes," says researcher Esben Agerbo, PhD, associate professor at Aarhus University in Aarhus, Denmark.
The benefit of parenthood on longevity was stronger for women than for men, says Agerbo.
The study findings echo those of previous research. It is published in the Journal of Epidemiology and Community Health.

Look at those url codes in more detail!  The-secret-to-a-long-life-children!  Having-kids-doesn't-kill-you-after-all!   And note how we are offered  a causality going like this:

having children > you live longer

Not all the popularizations I read were that bad.  (That second one above is a real humdinger, making false claims, as is the whole piece it comes from).  But some that stuck closer to the abstract about the study included interesting little jabs and pushes which may be even worse.  Here's the now-with-more conservatism BBC website on the study:

Involuntary childlessness may increase the likelihood of early death, the Journal of Epidemiology and Community Health reports.
The Danish study looked at more than 21,000 couples seeking IVF treatment.
They found women who were unsuccessful in having a child were four times more likely to die prematurely than women who had been mothers.
Critics stress that the risk of early death was low - with just 316 people in total dying over the 11 year study.
The authors of the paper also point out that their research suggests a link between childlessness and premature death and not a cause. They wrote: "Mindful that association is not the same thing as causation, our results suggest that the mortality rates are higher in the childless."
The researchers based their findings on data obtained from various population registers in Denmark on births, deaths and IVF procedures from 1994 to 2005.
During this time 21,276 childless couples registered for IVF treatment, 15,149 children were born and a total of 96 women and 220 men died.
Four times more likely
After analysis, the results suggested having a child cut the risk of early death, particularly among women.

That's the beginning of their piece.  A little bit later down it notes that mortality rate among the men was twice as high in the group which failed to have a baby.   This makes it certain that we see the problem as one of being a mother or not.   Then note this sentence in the above quote:

They found women who were unsuccessful in having a child were four times more likely to die prematurely than women who had been mothers.

Very sloppy writing, given that the whole sample in this study consisted of couples who sought fertility treatments.  A better way of writing it would be:

They found women who were unsuccessful in having a child were four times more likely to die prematurely than women who were unsuccessful in that endeavor.

If there is a propaganda story here it is one which suggests that women would live longer if only they had children. 

That's enough about the popularizations, though I would like to point out that even though the BBC asked other experts for their opinions, those experts pretty clearly were not aware of the actual piece of research they were asked to comment.  I base this on the fluffiness of their comments.

Then to the study itself. 

 Or to its abstract because I haven't obtained access to the study:

Background Childlessness increases the risk of premature mortality and psychiatric illness. These results might, however, be confounded by factors that affect both the chance of parenthood as well as the risk of premature death and psychiatric illness. 
Methods Using population-based health and social registers, we conducted a follow-up study of 21 276 childless couples in in vitro fertility treatment. 
Results The crude death rate ratio in women who become mothers to a biological child is 0.25 (95% CI 0.16 to 0.39). In other words, childless women experience a fourfold higher rate of death, that is, 4.02 (2.56 to 6.31). The analogous death rate in fathers is approximately halved: 0.51 (0.39 to 0.68) and 0.55 (0.32 to 0.96) associated with having a biological child and an adopted child, respectively. With substance use disorders being the exception, none of the crude rates of psychiatric illness in parents of a biological child were statistically distinguishable from the rates in the childless. These findings were slightly confounded by age, calendar year, income, education, somatic comorbidity and marital breakup. 
Conclusions Mindful that association is not causation, our results suggest that the mortality rates are higher in the childless. Rates of psychiatric illness do not appear to vary with childlessness, but the rate of psychiatric illness in parents who adopt is decreased.

With all due respect and understanding for those who write in a foreign language, that is an awfully poorly written abstract.   The first three sentences of the Results section must have something missing (that "in other words" bit)  and the "analogous" term doesn't refer to anything in what goes before.

I couldn't get very far with the abstract alone.  But what I have been able to gather from various sources is that the study looked at the death rates among 21 276 couples who had resorted to in vitro fertility treatments.   Here's one popularization on the data and study design:

In the study, which was published in the Journal of Epidemiology and Community Health, the scientists from Aarhus University looked at data from more than 21,000 couples who underwent IVF treatment between 1994 and 2005. They saw that 15,000 children were born and almost 1,600 were adopted during that time period. Unfortunately, 96 women and 200 men who were included in the study died.
Statistically, the death rate for childless women was four times as high as those who gave birth to their own child, while childless men were twice as likely to meet an early death. These findings caused the researchers to conclude that a couple who cannot conceive is at increased risk of early death.
Study co-author Esben Agerbo, an associate professor at Aarhus University in Aarhus, Denmark, pointed out that his team simply found a correlation and not a cause-and-effect relationship.
“My best guess is health behaviors,” he told WebMD. “When people have kids, they tend to live healthier.”
He suggested that having children promotes healthier and more responsible living as parents feel the need to both provide and set an example for their children.
The study also examined the mental health aspects of failing to conceive and found that childless couples were twice as likely to suffer mental illness as those who adopted children.

Can we trust that popularization?  Note that the abstract states that with the exception of substance abuse,  there was no difference between the mental health indicators of couples who succeeded in having a biological child and couples who did not succeed.   Yet the popularization tells us in that last sentence that there was!

Never mind.  Let's assume that the study was carried out properly, that it did, indeed find four times higher crude death rates among the women who failed to conceive after the treatment and twice as high crude death rates among the men whose partners failed to conceive after the treatment. 

What does this all mean?  

1.  Can we generalize from this study to everyone in the world?  Is a sample of people who are all seeking in vitro fertilization treatment just the same as a random sample of from the general population?   Is it OK for those popularizations (far above in my post)  to say what they did say : that children help us live longer, given that the sample in this study consisted of only people for whom fertility was large enough a problem that they resorted to in vitro fertilization?

I can think of a whole bunch of reasons why that would not be the case.   

First,  the infertility of some of the couples, at least, could have been caused by pre-existing ill health.  This ill health could make conceiving via in vitro fertilization less likely and it also could make premature death more likely.

Second, the infertility treatments themselves may have health risks for women.  Those who conceive stop any hormone treatments, for example, but those who have failed to conceive may continue the treatments for several more rounds.  The overall impact of the treatments themselves could cause fatal illness in some women among the failed-to-conceive group.

Third, wanting a child when one cannot have it may be a major stress factor in a person's life.  Major stress factors are a known health risk, and the longer they last the more likely it is that they affect either behavior or the health of the body itself.  Couples who failed to conceive live in that stress longer than couples who conceived early in the treatment.


2.  You may have noticed that my three points above all suggest reverse causality to explain the findings.  I'm not arguing that my suggestions are the correct ones, just that we cannot run quickly to the other extreme and argue that having children protects us from dying young. 

Esben Agerbo, one of the study co-authors, does warn us about correlation not necessarily being causation, and so does the study abstract.  Yet in that last quote where he is interviewed he offers a causal explanation for lower death rates of those who succeeded in the fertilization treatments:

Children promote healthier and more responsible living.

And that is probably true.  But is it the major causal arrow which explains these findings?  Let's put this into a diagram.  We have two simple options for the main arrow directions:

childlessness  >>>  ill health

ill health >>> childlessness

These simple options are not meant to be the only two possibilities.  My guess is that the lower one is the explanation which accounts for more involuntary childlessness, because most individuals born with severe health conditions never have children, for instance.  But the general explanation allows for a smaller arrow going in the direction, from having children to better health behavior, say.

You may have noticed that my three points about why the Danish sample cannot be seen as a perfect microcosm of the general population are all about the possibility that the causality mostly runs from ill health to childlessness or that at least the meaning of "childlessness" is closer to something of a major life trauma than the absence of children as such.

Conclusions

My apologies for not covering the actual study.  I couldn't get hold of it without paying thirty dollars.  But I suspect that it does what I've been able to glean from the various (though rather imprecise) popularizations.  What would have been interesting is to see what role suicides played in those death rates, but one study author notes that the information is absent from the official records.

If you care to read this post again (as if!), note how many errors the popularizations make.  The abstract tells us that the mental health indicators, with the exception of substance abuse, did not differ between those who conceived in the in vitro fertilization and those who did not.  Yet two of the popularizations I quote do state that they differed, and one states that they did so especially so for women.

One final comment:  I have never really thought about the possible correlation between having children and longevity, but had you asked me about it before I started waded in this mud puddle I'd have supposed that those without children might live longer.  That's because of anecdotal evidence from my own life and should be discounted.  Still, having children has real health risks for women, and a proper study of the relationship between having children and health should take those into account. 

It should also compare like with like.   If we wish to measure any impact having children has one one's health, then the pre-existing health states of the individuals in the study should be the same.  If they are not the same, the study tells us nothing much, because those with chronic conditions may simply be less likely to partner with someone and also less likely to have the physical health needed to conceive and, for women, to give birth.