Question: Almost everybody believes that teenagers who experience an early sexual debut (the phrase is redolent of the cotillion) are more likely to get into trouble than their peers who don’t. But is that the case?
New research: Two sets of researchers recently addressed this question and came to very different conclusions. Stacey Armour and Dana Haynie, sociologists at Ohio State University, used data from the National Longitudinal Study of Adolescent Health. They focused on 7,000 or so high-school and middle-school students who reported, via questionnaire, that they were virgins in 1994-95 and were then resurveyed twice over the following six or seven years. They divided the teenagers into three groups—“early,” “on time,” and “late” sexual initiates. Since the average age at which the kids first had sex varied a lot from school to school (from 11¼ to 17½), the groupings were calibrated accordingly. Early means first experience of sex before 84 percent of a teen’s schoolmates; late means sex after 84 percent of peers lost their virginity; and on time means everything in between.
Findings: As we might expect, Armour and Haynie found a higher incidence of delinquency, of about 20 percent, in the early-debut group and a 20 percent lower delinquency rate for the late bloomers. These results troubled K. Paige Harden and her colleagues in the department of psychology at the University of Virginia. They felt that the study fell into a common trap: the attribution of a causal relationship to an observed association. So, they re-examined a much smaller number of adolescents drawn from the same data set: about 500 same-sex twin pairs.
Findings Part 2: In one way, the second study confirmed Armour and Haynie’s results. For instance, Harden’s team found that twins who both had a relatively early sexual debut were more likely to engage in delinquent behavior than twins whose first sexual intercourse was later—an association they attributed to family influences. However, many of the twins varied from each other in the age at which they each first had sex, allowing for comparisons between the two. Here, the results were very different: The twin with earlier first sexual intercourse was less likely to become delinquent.
Conclusion: If this relationship is causal (which is more likely for twins, because of the genes and environment they share), then the assumption that early sex is psychologically risky for adolescents may be wrong. Harden and her co-authors also cite a number of studies in which short-term and long-term benefits are associated with early sex. They’re careful to remind us once again, though, association and causation are not at all the same thing. More research is needed to answer the question, but Harden’s work is a beautiful example of a more subtle interpretation by a second researcher who dives deeper into a pool of data.
Assumption: Somewhere along my medical training, I developed a strong sense that what you eat plays no role in whether you develop acne. But maybe I’m wrong.
Research: A new study examined the effect of changing diet on acne in a group of 54 male patients between the ages of 15 and 25. Half the group was assigned a diet rich in high-protein foods, like fish and lean meat, and complex carbohydrates, like whole grains (what most of us would call healthy). The other half was encouraged to eat sweets and highly processed carbohydrates (what most of us eat). The hypothesis was that patients with high levels of blood sugar, produced by the unhealthy foods, would also have high levels of insulin, which have already been implicated as an acne factor.
Finding: A dermatologist who didn’t know which diet the patients were on examined them at regular intervals. After 12 weeks, the results were clear, though to my mind the interpretation is not. The healthy diet led to less acne—fewer pimples and less inflammation. It also produced weight loss.
Conclusion: So—what led to improvement in the acne? The weight loss? The lower levels of insulin? Less saturated fat? More fiber? Beats me. But it is clear that diet does affect acne, and now we are obliged to figure out just how and why.
Question: The risks of drinking during pregnancy—developmental delay and birth defects—are well-known. But how much drinking is too much, and is ongoing drinking more or less dangerous than binge drinking? Different countries give differing advice to pregnant women. In the United States, complete abstinence is recommended; in Australia and New Zealand, moderation. In Great Britain, the guidelines are specific: Pregnant women should avoid intoxication and limit themselves to half to one glass of wine or 6 to 12 ounces of beer (equal to one-quarter to half of a martini) once or twice a week.
New research: A recent publication by Jane Henderson of Oxford University combed through 3,500 possibly relevant papers, and found 14 that rigorously looked at different effects of binge drinking. Based on these studies, Henderson’s team concluded that binge drinking might result in a small degree of decreased birth weight, perhaps a slight excess of birth defects, and significantly more learning and developmental problems.
Caveat: Their summary makes it sound as if binge drinking in pregnancy is no big deal, but I am not a bit convinced. The studies were so different from each other—in their definitions of binge drinking, when during pregnancy women indulged in it, and the nature of potentially confounding variables—that all the studies together don’t provide a strong basis for making recommendations. Nor do we know how many pregnancies were evaluated in the studies pooled for this analysis. Finally, I wouldn’t be so cavalier about the increased learning and developmental problems.
New research, Part 2: Another new study about the effects of smoking during pregnancy is much more productive in what it tells us. The research began as an attempt to better understand why smoking by pregnant women has been associated with genital abnormalities in boys born to them. The authors examined almost 70 aborted male fetuses, looking for genital differences between fetuses whose mothers had smoked or not smoked. They found one key difference: Fetuses exposed in utero to the products of cigarette smoking showed a depressed function of a gene named desert hedgehog.
Conclusion: Previous research had shown decreased fertility and abnormal testicular development in men with DHH mutations. So, the new study may explain what’s going on: some material from the smoke that inhibits the function of the DHH gene leading to a mutationlike effect that disables the gene. Unlike the binge drinking study, the implications of this research are clear—one more reason, if they needed one, that women should not smoke during pregnancy.