When Opportunity Is the Best Birth Control

I was talking with a friend this morning who is a social worker at a large Brooklyn high school. She told me that of the 12 girls she’s seen regularly this year for counseling and group sessions, four of them are currently pregnant. Some of the other ones already have babies or toddlers; others have had abortions in the past. Pretty much everyone at this high school knows someone who has been pregnant or already has a child.

This was especially surprising to me because I read this week that the nation’s teen birthrate actually dropped by 17% between 2007 and 2010 to 34.3 births per 1,000, the lowest rate ever recorded. How could there be such great dissonance between the fact that in New York State, teen birthrates have dropped 13% over that period and my friend’s first-hand experience with her students?

Of course I’ve been around health statistics long enough to know that anecdotal evidence often has little to do with larger-scale trends or findings. Take the case of mammography and its contribution to saving women’s lives: when I wrote about the wealth of evidence pointing out that yearly mammograms for women under 50 actually has little impact on reducing breast cancer mortality but does increase the rate of false positives and over-treatment, I received phone calls and emails from plenty of women whose own experience told them otherwise. “My life was saved by a mammogram that caught my cancer before it could spread,” a close family friend argued. “I don’t believe these studies and I think what you’re writing is dangerous.”

But back to teen pregnancy—after all, May is National Teen Pregnancy Month. This dissonance between the encouraging large data picture and the local experience got me thinking about what is becoming an important part of all public health and disease interventions. The imminent expansion and also specificity of the data pool as more and more information is recorded and dumped into large computerized repositories begs for a new kind of analysis that identifies “hot-spots” and outliers that resist the trends; local areas, single hospitals, particular members of ethnic or racial populations and other demographic subsets that defy the odds when it comes to national health outcomes or usage of medical services.

We’re already seeing this in the realm of Medicare payment data; the Dartmouth Atlas has for years been identifying regional variations in the cost of care in different parts of the country and even between neighboring towns. Now, as the first step toward linking reimbursement to efficient care, the Center for Medicare and Medicaid Services has released new findings from its own study of hospital and post-discharge cost variation. According to coverage in Kaiser Health News, “the figures show wide variance among hospitals around the country, even ones just a few miles apart. In Los Angeles, for example, the average patient admitted to Los Angeles Community Hospital cost Medicare nearly $24,644 during the stay and in the month afterward, 37 percent above the national median. Across town, according to the data, an essentially similar patient admitted to Ronald Reagan UCLA Medical Center cost Medicare $17,628, or 2 percent below the median.”

There is still much controversy over what leads to these cost variations; hospitals in particular have a lot at stake in 2014 when CMS begins using this kind of data to determine which providers receive financial bonuses for efficient care and which are penalized for being high spenders. But the data do spur further examination and changes that might focus on better discharge coordination in one hospital for example, or reducing excess testing in another.

The power of this new focus on what I’ll call “micro-data” is that it naturally follows that interventions should be more effective because they can be focused on unique problems. Last year I wrote about a laudable effortin San Francisco to identify the highest users of the city’s safety-net services; ambulances, emergency rooms, homeless shelters and detox centers. In order to do this, all the safety net providers had to agree to take their own electronic patient data and dump it into a central repository for analysis. San Francisco still struggles with the high cost of caring for its almost 230,000 “high utilizers across multiple systems” or HUMS; many are homeless, alcohol and/or drug abusers, mentally ill and suffering from chronic disease. But once a sub-population like HUMS is identified, tailored interventions can be rolled out, for example establishing a medical home at a community health center or safety-net hospital that helps coordinate all the services accessed by a given individual—substance abuse treatment, mental health care, medical care and housing.

To deal with teen pregnancy and birthrate there needs to a similar emphasis on micro-data and, if you will, micro-interventions. It’s great that the teen birthrate is going down in the country as a whole, but the fact is that American teens are still two and a half times as likely to give birth as compared to teens in Canada, around four times as likely as teens in Germany or Norway, and almost ten times as likely as teens in Switzerland. There has been progress, yes, but in South Georgia, for example, the pregnancy rate among 15-17 year-olds is still 40 per 1,000l (the national average in this age group is about 25 per 1,000). In Yuma, Arizona, a county along the Mexican border, the teen pregnancy rate (15-19 year-olds) is 66.6 per 1,000 females  , almost twice the national average. And in New York City, my social worker friend is seeing girls from high-poverty neighborhoods who, according to the latest report from the NYC Department of Health, “are three times more likely to become pregnant than teens in low-poverty neighborhoods.”

According to the Centers for Disease Control, the declines we’ve seen in teen pregnancy and birthrates are primarily due to the effect of stronger pregnancy prevention messages in schools and more teens using effective forms of birth control like oral contraceptives. That means that comprehensive sex ed and policies that promote easier, cheaper access to birth control pills and other contraceptives have had the desired effect. For those who continue to champion abstinence-only education, there is no evidence that this has any effect on reducing teen pregnancy rates—and in fact, may actually may deter contraceptive use among sexually active teens.

So why does teen pregnancy still continue to be so high in the U.S., and particularly in certain “hot-spots?” In a paper that is set to appear in the May 22 issue of the Journal of Economic Perspectives, two economists suggest that it’s actually a symptom of underlying social and economic problems. The root problem, they determined, is that the U.S., unlike its European counterparts, has a very high level of income inequality. This in turn, creates a sense of hopelessness and despair in teens that the researchers found is closely related to higher rates of teen pregnancy.

According to Phillip B. Levine, an economics professor at Wellesley College, “If a young woman sees little chance of improving her life by investing in her education and career skills, or by marriage, she is more likely to choose the security, immediate gratification and happiness of parenthood. Our work captures this idea in a standard economics model of decision-making.”

I don’t know if this theory of opportunity applies to other areas around the country where teen pregnancy rates are far higher than the norm—for example, the entire state of Mississippi where the rate is a nation high of 55 per 1000—but my social worker friend told me it rings true to her. The girls she sees have had pregnancy prevention education, they have easy access to contraceptives and are pretty sophisticated in their knowledge about sex. But they are also far behind in the credits they need to graduate high school, many have difficult home lives, some abuse alcohol and drugs, and they live in poverty in a city that also lays claim to some of the wealthiest people in the world.

So what kinds of social policies might help lower the teen pregnancy and birthrate further? The researchers conclude; “opportunity is the most effective birth control for teens.” A summary of the journal study concludes: “If the problem is perceived lack of economic opportunity, then policy interventions need to attack that. Access to early childhood education programs and college financial aid, for instance, have proven to be successful in improving the earnings–and sense of hope–of participants. Our findings show that these programs may also have the added benefit of lowering teen pregnancy rates. Giving teens a sense of opportunity and hope may be a much more powerful prescription than abstinence-only, sex education, or birth control combined.”

This is not a simple solution. It certainly is a lot easier for the federal government to throw $50 million at abstinence education programs; cross our fingers and hope that kids learn to “Just Say No.” But in specific areas that continue to buck the trend when it comes to teen pregnancy and birthrates, it might be necessary to put policies in place that do address the deeper socio-economic issues that make having a baby the only rewarding choice for teens facing an otherwise bleak future.