Income gradient for children’s mental health

Lining them up (by income) and knocking them down.

I didn’t realize how strong the income gradient is for children’s emotional and behavioral problems. This new graph from the CDC combines data from 6 years of the National Health Interview Survey, and shows a steep relationship at all ages:

Percentage of Children with Serious Emotional or Behavioral Difficulties, by Age Group and Family: U.S., 2004-2009

The question asked was, “Overall, do you think that [child] has any difficulties in one or more of the following areas: emotions, concentration, behavior, or being able to get along with other people?” Children are included here if the parent said “yes, definite difficulties” or, “yes, severe difficulties.”

As background: I’ve posted before on the income gradient for asthma, overall health, diagnosis timing, mammography, pregnancy, and women generally. That makes me curious, but not an expert. That is probably a good description for the authors of this recent review article, Janet Currie and Wanchuan Lin, who conclude:

Low-income children are in worse health than other children are. This paper explores the extent to which insults to health and activity limitations are responsible. In the most recent National Health Interview Survey (NHIS) data, low-income children are more likely than other children to have virtually every measured chronic or acute condition and are more likely to be limited by these conditions. Mental health conditions are particularly common and limiting. But the higher incidence of measured conditions and limits does not explain all of the relationships between income and overall health status, which suggests that unmeasured illnesses and injuries are also involved.

And finally, this reminds me of a good research tip. To get started on your subject, find a review article that’s a few years old or older, and then see which articles cite it — that should help bring you up to date. In this case, you could get these, which look highly relevant:

• Impact of Income and Income Inequality on Infant Health Outcomes in the United States. Pediatrics December 2010 126:61165-1173.
• Child Health and Young Adult Outcomes. J. Human Resources July 2010 45:3517-548.
• Policy Interventions to Address Child Health Disparities: Moving Beyond Health Insurance. Pediatrics November 2009 124:Supplement_3S246-S254. [Are we really ready to move "beyond health insurance"?]
• Early Childhood Poverty and Adult Body Mass Index. AJPH March 2009 99:3527-532.
• Access to and Use of Paid Sick Leave Among Low-Income Families With Children. Pediatrics August 2008 122:2e480-486.

Putting teen birth rates on the maps

The latest Morbidity and Mortality Weekly Report is out, with a report on teen birth rates in the U.S. The press release announces, “U.S. Teen Birth Rate Fell to Record Low in 2009.” (The report has information about birth control, virginity, and sex education as well.)

The CDC’s vital signs pamphlet still calls the rates “unacceptably high,” and notes they are “up to 9 times higher than in most other developed countries.”* Within the U.S. we have about a 4-to-1 ratio in teen birth rates between the states with highest and lowest rates, as you can see from this map:

Teen birth rates are the number of births per 1,000 women ages 15-19.

For comparison, using the U.N. Demographic Yearbook, I made a map of Europe using the same color scale as the CDC’s state map, though I had to add a few categories. (If you don’t know which countries are which, why not take a little time to learn them?)

Light blue, 31-39; White, 16-29; Pink, 10-16; Red, 0-10; Black, unavailable.

You can see the high rates in the Eastern European countries of Russia, Ukraine, Romania and Bulgaria, with teen birth rates in the range of our l0w-middle states (like California). The high-middle European countries — including Britain, Ireland, the Baltics and some Central European countries — are comparable to our lowest-rates states (New England, etc.). Then the rest of Europe is off the U.S. chart, down to 4.3 in Switzerland.

Addendum: The UN has a longer list of country teen birth rates here. The US ranks 95th out of 207 countries for 2007 on that list.

*It’s a little strange that teen births are considered a problem by definition, even though some of these teens are married, which should make their births officially not a problem.

Risks women share, more and less

What is the basis for women’s global unity?

The other day I discounted the idea — expressed in the myth of women owning less than one percent of all property in the world — that women share a universal propertylessness. “If global feminist unity is to be had,” I said. “It won’t be built on a shared poverty experience.” One person commenting on the Huffington Post retorted: “The shared experience of women is patriarchy.” And she challenged me to produce a “gift-wrapped statistic that might make people think twice about gender inequality.”

I don’t have it. But for discussion, consider maternal mortality. I have previously shared the worldwide trend (except in the U.S.) toward reducing maternal mortality — the deaths of women related to pregnancy and childbirth. For every 10,000 live births in the world, 260 mothers still die.

This isn’t a risk all women face, since many have no pregnancies or births, but it’s something that is unique to women (more so even than rape). It is at least a potential risk women have in common.

In reality, however, the risk is so unevenly distributed as to virtually undermine its universality. In Sub-Saharan Africa, among all women, one out of every 31 women is estimated to die from maternal causes; in Western Europe that number is one-in-8,800. That is partly because African women have more children, and partly because they are more likely to die during each pregnancy or birth.

Those numbers are from a new data sheet published by the Population Reference Bureau. They estimated the lifetime chance that a given woman would die from maternal causes (factoring in both birth rates and risks of death). I’ve converted those to deaths per 10,000 women, by world region:

As is the case with wealth, statistically anyway, the women of the richest countries have more in common with their male peers than they do with the women at the bottom of the scale. So this isn’t the gift-wrapped statistic for global feminist unity based on shared personal risks. But do people need to experience the same hardships in order to unite against them?

Getting the story straight on working mothers and children’s risks

When the result is not the news. Or, woe is the status of most social science reporting.

The news release was titled, “Children Of Working Moms Face More Health Problems.”

The news headlines, repeated around the world on the Internet’s instant, editor-free news rebroadcasting systems, were predictable:

• Kansas City Star: Study: Working moms have sicker kids
• Sify: Kids of working mums more likely to be ill
• Top News Arab Emirates: Kids of moms who work outside get hurt more
• Food Consumer: Does a child’s health pay the price for working moms?
• Parent Dish: Study: Working Mothers Have One More Reason to Feel Guilty

Corbis helpfully sold clip art of a working mother putting her child at risk (note the ancient computer monitor and elephant-sized phone, showing the venerability of this story…):

In fact, we had the whole panoply of clip art on display, helpfully collected by Google news:

A complicated story

The research, by North Carolina State University economist Melinda Morrill, is in press at the Journal of Health Economics (the abstract is here, and Morrill has an earlier version for free on her website here).

Her conclusion, reasonably justified by the analysis was this: “I identify the effects on overnight hospitalizations, asthma episodes, and injuries/poisonings for children ages 7–17. Maternal employment increases the probability of each adverse health event by nearly 200 percent.”

Wow. Sounds awful. But understand one thing. The risk of all these events is very low, whether the kids mothers are employed or not. Doubling or tripling these rates still means that the vast majority of children are unaffected (triple-emphasis added). Using her data, a few hundred thousand National Health Interview Survey respondents from 1985 to 2004, the overall rates of each outcome look like this:

At a glance, it doesn’t match the headlines. Children whose mothers worked are less likely to be hospitalized or have asthma attacks (but more likely to have injury or poisoning). That’s probably just because healthy, rich mothers are more likely to work and have healthy, rich, safer kids. (It’s OK, we can control for that.) More importantly, the rates are low and the differences are small. Nevertheless…

The real contribution of this article is a clever application of what Belinda Luscombe at Time helpfully (I mean that) called “difficult-to-explain statistical techniques.” Specifically, Morrill used — there is no way to sugar-coat this — an instrumental variable method with two-stage least-squares regression, otherwise known as IV-2SLS.

Can this be explained to a non-expert audience? As a non-practitioner who has read a bunch of these papers, let’s see if I can do it.

• Whereas, for many questions in social science, we would like to investigate a causal relationship, but the complexity of such relationships makes that difficult to establish; and,
• Whereas, from a scientific point of view, the ideal study design is a true experiment, in which we randomly assign people to different conditions and trace their effects, thereby removing contaminating factors such as past experience, personal decisions and preferences, strengths and weaknesses, etc.; and,
• Whereas, social scientists often can’t do true experiments because of ethics (and other reasons), and when we do experiments (like laboratory simulations), they differ substantially from real-life situations;
• Therefore, some social scientists (usually called “economists”) use instrumental variable (IV) analysis, in which the trick is to find something (an “instrument”) that acts like an experiment, (more or less) randomly assigning people to different conditions, so that their true effects can be identified.

That’s the gist of it: has something (more or less) random occurred which (a) causes the independent variable to change (e.g., driving mothers into the labor force) while (b) neither causing, nor resulting from, the dependent variable (hospitalization, injury or asthma attack).

In this case, Morrill cleverly split children into two groups: those who had younger siblings who were just old enough to start kindergarten, and those who had younger siblings that were just too young to start kindergarten. Because mothers have a tendency to start work when the younger child goes off to kindergarten, but children reaching kindergarten age is neither cause nor effect of older children’s health outcomes, this acts like an experiment — some moms are assigned to the go-to-work group and some aren’t, and membership in the two groups is more or less random.

The method is called 2SLS because, using a complex prediction model, the economist first identifies those mothers whose employment was likely the result of the the younger child reaching kindergarten age, and then (in the second stage) uses a complex prediction model to determine whether those mothers’ older children were more likely to end up sick or injured than the children of those who did not start work.

Bottom lines

This method creates something close to an experiment, close enough that it is sometimes called a “natural experiment,” since the scientist didn’t engineer it. However, it also analyzes events that are extremely narrowly construed. It really is only a test of what happens when American mothers of two or more children started work after the younger child reached kindergarten age (over a 20-year period) — holding constant a wide array of social and demographic variables. Since the randomness of the school-age “instrument” can, in a practical sense, be confirmed statistically, the effect is reasonably called “causal,” but caution in the interpretation is wise.

And Morrill was cautious. Although she found that older children of mothers who went off to work in these conditions were indeed more likely to suffer these ill effects, she did a number of other checks to make sure things were as they appeared.

In fact, one of the mostly-overlooked aspects of the paper was a section on “heterogeneous effects.” Here, she tested whether the overall effect she found actually resulted from some subset of the families experiencing large effects while others experienced none. In my interpretation, this is where the real story is.

The effect of mothers going off to work on children’s hospitalization was three-times greater for Blacks than for Whites (and non-existent for Hispanics). The effect was only significant for mothers who had no more than high school education (unlike most or all of the women in the clip art above!). And the effect was three-times larger for single mothers than married mothers.

With no measures of child care availability or any details about the care arrangements of the families’ children, I’m left to conclude that the results probably reflect the simple fact that poorer women have fewer good options for childcare, so that when they enter the labor force, their children experience some increased risk of accident or illness.

Stop the presses

I see this result as a confirmation of common sense, not shocking or disturbing, or in any substantive way altering my understanding of the work-family-children situation: mothers working for pay increases the risks of illness or injury associated with non-supervision, or supervision by others. That this seems obvious does not detract from the value of the study, just from the breathlessness of its news coverage.

What are the implications of this? I can think of two. First, mothers (or, obviously, any caretakers) who are considering entering the labor force need to consider the availability and quality of alternative care arrangements for the children they will no longer be caring for during their working hours. Hopefully, they already knew this. Second, for public policy, we need to consider the availability and affordability of care arrangements for children whose parents are employed.

As for the bigger question, the one about mothers’ guilt and hard choices, Belinda Luscomb was good enough to link to a recent meta-analysis — a study of studies — that analyzed 69 different studies of the effect of mothers’ early employment on their children’s school achievement and psychological health, published by the American Psychological Association. That study concluded:

The small effect size and primarily nonsignificant results for main effects of early maternal employment should allay concerns about mothers working when children are young. However, negative findings associated with employment during the child’s first year are compatible with calls for more generous maternal leave policies. Results highlight the importance of social context for identifying under which conditions and for which subgroups early maternal employment is associated with positive or negative child outcomes.

Now we can confirm that another risk — small and manageable in the vast majority of cases — is illness or injury associated with loss of parental supervision. Something to watch out for. But didn’t we already know that?

There, I said it. Sorry it took so long.

If we could teach kids one thing

Is “self control” that thing?

The parenting advice pile in my blog reader is brimming over again. It’s a frustrating pile, which includes everything from marketing hucksters to well-intentioned ignorance and naive extension of reasonable ideas to unsupported generalizations. One recent article, however — which didn’t come through the parenting channels — offers a model of scientific method. It also reinforces some basic facts about inequality, and shows the limits of what we know.

Researchers Terrie Moffitt and colleagues, writing in the Proceedings of the National Academy of Sciences, traced a sample of children born in New Zealand in the early 1970s through age 32. Their study used a measure of “self control” from the first 10 years of life to see whether it was associated with health, wealth, and criminality by age 32.

By “self control” — the key concept in the study — they mean:

nine measures of childhood self-control [including] observational ratings of children’s lack of control, parent and teacher reports of impulsive aggression, and parent, teacher, and self reports of hyperactivity, lack of persistence, inattention, and impulsivity.

The study is observational, rather than experimental, in that they didn’t assign children to a self-control condition, but rather just observed how they turned out in relation to the self control they displayed. That means we can’t conclude the relationship is causal. There are lots of things about these kids and their lives that we don’t know, which could be hiding behind that self-control “effect.” (If we could get this idea alone to catch on with the parenting-advice-reading public, the social world would be a more relaxed place.)

Anyway, to me, three things stand out in their results:

1. Self control does successfully predict health, wealth and criminality in the ways they expect. Kids with higher levels of self control do better on these measures later in life. And that holds with simple statistical controls for family socioeconomic status (low versus not low), and childhood IQ score (low versus not low).
2. Family socioeconomic status (SES) is even more important. We already knew that, but it’s nice to be able to see that, even controlling for IQ and self control, SES is a key determinant of well being later in life.
3. IQ scores in childhood are the least important, compared with SES and self control.

The authors are focused on self control, and their correlational evidence is quite strong, as seen in this key figure:

One more empirical point to reinforce: even though the science news was headlined “Don’t Take that Cookie!“, this article does not show that efforts to change children’s self control have beneficial effects. Although they do find that children whose self control improves over time are headed in a good direction, that improvement is not from the result of a measured intervention. So we really can’t say that working to improve self control makes a difference. Not that there’s anything wrong with it.

Finally, let me add one point on the philosophy of social science regarding studies like this. Neither this nor any other study of what makes children “turn out” a certain way speaks to absolute principles of well being — they are all socially situated in space and time. That is, there may be social contexts in which self control matters more, or less, than it did among New Zealanders born in 1972-73; the same holds for IQ scores and socioeconomic status.

As we should expect, today’s parents are concerned with what they can do to help their kids in the social here and now sweepstakes. But from a social point of view, we might just as concerned with how to reduce the well-being gaps between those with more versus less self control, IQ points and socioeconomic status as we are with how to get some kids more of these assets in order to help them get ahead. That’s our choice.

Health inequality compendium

The CDC releases a slew of inequality trends.

Many people will find the new report from the Centers from Disease Control very helpful. It’s called CDC Health Disparities and Inequalities Report — United States, 2011, and it covers everything from inadequate and unhealthy housing to preterm birth by race/ethnicity:

I previously reported national comparisons showing the U.S. bringing up the rear on this health indicator, and discussed the evidence for the role of obesity. This table was nice because it broke out the Latino groups, which we often don’t get (next step, Asians).

Anyway, very nice to see CDC putting resources into the collection and dissemination of inequality indicators. This report should be especially useful to teachers who want to include health in their discussion of inequality, but aren’t specialists in health outcomes (like me).

A pause to appreciate the social science of inequality

Sometimes I just pause to marvel at the power and, yes, beauty of social science.

From long experience and careful evaluation, we have come to learn that one simple, completely subjective question produces answers with amazing predictive power.

The question is:

Would you say your health in general is excellent, very good, good, fair, or poor?

There may be one, but I’m not aware of a more accurate or reliable predictor of mortality that can be ascertained in so few words. This “self-rated health” question is one of the most widely used survey items when the goal is to minimize cost and invasiveness. It doesn’t work with every individual, of course, but if you get 1,000 people to answer this question, and then check up on them five years later, you’ll see what I mean.

One review article on the effect of self-rated health on mortality, by Ellen Idler and Yael Benyamini, is the fifth-most cited article in the history of the Journal of Health and Social Behavior (1,740 citations as of yesterday). They showed that, in most studies, the question (or something close to it)  predicts mortality quite well even when other basic demographic factors are controlled statistically.

For those interested in inequality, this has become a useful way of assessing health inequality in various contexts and an indicator of population health across groups. Today’s case in point is from the 2009 National Health Interview Survey, which features a beautiful example of the health-social class gradient (which I’ve written about before).

(In this case, the question was asked about family members age>24 with whom the respondent lived, and adjusted for age differences across the education groups.)

Those gradients are very steep: from 38% of the high school dropouts with “excellent or very good” health to 74% of those with BAs or more. At the other end the ratio of dropout-to-BA percentages with “fair or poor” health is 4-to-1.

Part of what I love about this is the use of subjective assessments to measure a hard reality. We could draw blood, take blood pressure measurements or DNA, do CT scans and fitness tests. Or we can just ask, effectively, “how’s your health?”

Social scientists do a lot of fancy science (not that there’s anything wrong with that). This isn’t fancy science, just a simple, useful question based on a lot of previous research. And in this case it wasn’t cheap – this was a sample of about 34,000 households, reflecting a response rate of 82%. But it is good science, and that’s a sight to see — even when the news isn’t good.

Weathering health inequality

A hypothesis with legs.

In the early 1990s, Arline Geronimus proposed a simple yet profound explanation for why Black women on average were having children at younger ages than White women, which she called the “weathering hypothesis.”

It goes like this: Racial inequality takes a cumulative toll on Black women, increasing the chance they will have health problems at younger ages. So, early childbearing might pose health risks for White women, but for Black women it makes more sense to start earlier — before their health declines. Although it’s hard to measure the motivations of people having children, her suggestion was that early childbearing reflected a combination of cumulative cultural wisdom and individual adaptation (for example, reacting to the health problems experienced by their 40-something mothers).

She showed the pattern nicely with data from Michigan in 1989, in which the percentage of first births that were “very low birthweight,” increased with the age of Black women, but decreased for White women, through their twenties:

Source: My graph from Geronimus (1996).

If the hypothesis is correct, she reasoned, the pattern would be stronger among poor women, who experience more health problems, which is also what she found.

The most recent national data, for 2007, continue to show Black women have their first children, on average, younger than White women: age 22.7 versus 26.0. And the infant mortality rates, by mothers’ age, also show the lowest risk for White women at older ages than for Black women:

Source: My graph from CDC data.

Note that, for White women, mothers have children in the early thirties face less than half the infant-mortality risk of those having children as teenagers. For Black women, waiting till their lowest-risk age — the late 20s — yields only a 14% reduction in infant mortality risk. So it looks like waiting is much more important for White women, at least as far as health conditions are concerned.

The implications are profound. If you base your perceptions on the White pattern, it makes sense to discourage early childbearing for health reasons. But if you look at the Black pattern, it becomes more important to try to improve health problems at early ages — and all the things that contribute to them — rather than (or in addition to) trying to delay first births.

Mammography’s income gradient

Rich women get more regular cancer screening.

Women with more money are healthier, generally, and new data from the CDC shows a strong relationship between income and regular mammography for women over age 49. (The pattern for education is about the same.)

Because Black women have more breast cancer — with worse outcomes on average — than White women, I was expecting a stronger race/ethnicity pattern. Instead, the data show that Black, White, and Asian/Pacific Islander women all have rates over 80%. The exception is American Indian women, only 70% of whom have had the recommended screening.

The big difference is in having health insurance, not surprisingly. Among those with health insurance, screening is 84%; among those without it’s just 56%. And to make matters more rational — and less humane — being married helps, too.

The report concludes:

Health-care reform is likely to increase access by increasing insurance coverage and by reducing out-of-pocket costs for mammography screening. Widespread implementation of evidence-based interventions also will be needed to increase screening rates. These include patient and provider reminders to schedule a mammogram, use of small media (e.g., videos, letters, brochures, and flyers), one-on-one education of women, and reduction of structural barriers (e.g., more convenient hours and attention to language, health literacy, and cultural factors).

Lifestyles of the 4x-poverty set

Rich Americans drink more, smoke less, exercise more, weigh less, and sleep more.

Of course, some of those poor people could be “physically active” at work…

…and some of those rich people might not be counting all their sleep: