Health and parenting
I was intrigued by this story in the Washington Post on Monday, reporting on a study that examined the cost of childbearing on parental health. The researchers took advantage of the huge amount of geneological data collected by the Mormon church, and studied the effects of family size on both parental and child health.
As you’d expect, the odds of dying in childbirth or immediately thereafter rose for women the more children they had borne. But the odds of dying in the next year rose significantly for women even after the first few months, and for men as well. In an online Q and A, the reporter said that the findings held across imputed socioeconomic status, which suggests that it’s not just a matter of having too many mouths to feed. The article suggests that the findings may be a sign of the health impacts of stress. Children in large families were also more likely to die than those in small families, possibly due to inadequate supervision.
I wasn’t surprised to see that children were more likely to die in childhood if one of their parents died before they reached age 5. I was surprised that this finding was so much stronger for maternal death than paternal death. I can see how maternal death would be a disaster for an infant, but my stereotypical image of pioneer families makes me think that loss of a father would be a greater disaster for older children. But there may have been more social support for widows and their children than I imagine. (I also think there may be some bias introduced by the sample design, which is limited to couples who were each married only once; my impression is that both widows and widowers tended to remarry out of simple economic necessity.)
Given both smaller family sizes today and better medical care, I’m not sure if this study has any practical implications today, but I thought it was interesting.
January 18th, 2007 at 10:54 pm
Having looked at lots and lots of LDS genealogical data, I can theorize about the effects of maternal death:
1. Fathers often farmed out very young children to other relatives. All other things being equal (i.e., the other relatives also having large families), these children would now be least likely to win the competition for care. They would be least likely to get the resources of food, shelter, and attention necessary for very young children to thrive. I bet this goes a long way toward explaining the finding that ‘The later-born children in very large families had less chance than their older brothers and sisters of surviving into adulthood and having children themselves.’ Even within the original household, later-born children would have been out-competed for resources, and been born to more worn-down parents.
2. Mothers of very young children were more likely also to be mothers of older children. Those older children could support the family if the father died. The family unit could also support itself with the assistance of the kinship groups among whom almost all western settlers lived. (If this was an LDS-predominant data set, community support was a decisive factor in everyone’s lives.) Mothers were more likely to farm out older children after the death of a husband, and to keep the younger children close by. But the older children were in a better position to defend themselves in the competition for scarce resources.
3. Genealogical data is itself somewhat worrisome for historical purposes, because in every generation there is, almost by definition, at least one demographically successful individual. Although LDS genealogists are more likely to have at least rudimentary information about demographically unsuccessful sibling branches, the bias still concerns me from time to time. (Demographica success is defined as reproduction in the second generation: in other words, having grandchildren.) To say it with less jargon, the family information collected in genealogical records is biased towards those families strong enough to reproduce over multiple generations. There’s no immediate reason why that would effect THIS particular study, it’s just a problem with genealogical data generally. (Before the 1850s, anyone who didn’t marry and have children is almost impossible to trace reliably into adulthood. Even those who married are hard to track down. Genealogical records are compiled through a combination of hard data — which is scarce outside the Nordic countries, or closed by legislative protection — and connections made by contact with other branches of the tree. Branches that didn’t produce descendants disappear from the pool.)
4. I’m especially wary of theorizing about the general population based on LDS family histories. (Calder’s family — the ones bearing his last name — came to this country as LDS immigrants.) There are unique features of their lives that made them unlike other groups of Western migrants. Although honestly, as a group they tended to be more demographically successful and more healthy, so I’m not sure how that affects this study data.
5. A last point: 1.5 percent of women who had 1-3 children died within a year of the last birth. 6 percent of women with 12 or more children died within a year of the last birth. But how much older were the women giving birth to the last in a long line of children? Some of the difference in those statistics does relate to bodily stress after multiple pregnancies. But some of it just reflects pure and simple demographics: women in their forties more more likely to die than women in their twenties. [The Q&A indicates that the researches just dismissed this problem, focusing instead on the fact that after age 50, women who’s had more children died sooner than women who’d had fewer. That’s a good, intuitive finding: but it leaves the question of maternal age at last pregnancy unaddressed. Even today, a woman who has a baby in her forties is at greater risk for post-partum complications than a women having a baby in her twenties, regardless of how many babies she had before.]
January 19th, 2007 at 9:39 am
Thanks for pointing out the article…I had missed it. In anecdotal evidence that makes me wonder if there is, indeed, some sort of link, my then-39-year-old husband had a heart attack when our daughter was three months old. It was partly due to his physiology, but to this day I still wonder if the stress of infant care sped up the process. Needless to say, he’s a little wary about having a second child.
January 19th, 2007 at 11:44 am
Fascinating stuff (and Jody’s comments, too), but it’s of mostly historical interest, right? I would imagine that things have changed enormously in a 100 years. I do think such historical reports are vital, though, because they help us put our current views of family in context. I’ve always found it deeply disturbing when people take a distorted view of a decade in time, and hold it up as an example of an ideal.
The info on tracking people who don’t leave grandchildren is fascinating, too. I’ve always wondered why we hear so little of those who died in WWII (as opposed to those who served). I’d guess the answer is that they were brothers and children, not fathers and grandfathers.
bj
January 19th, 2007 at 5:19 pm
Thanks for the info – this was a great post and comments.