16 décembre 2004

Study: Married Adults Healthier Than Most

By RANDOLPH E. SCHMID, Associated Press Writer

WASHINGTON - Confirming what many couples already knew, a government study concludes it's healthy to be married.
Sure, the majority of husbands pack on some extra pounds. But overall, married people are sick less often and more active. They smoke and drink less and in general feel better than single, divorced, never married or even folks just living together. The National Center for Health Statistics report didn't name the reasons why married people are healthier. But health statistician Charlotte Schoenborn in an interview described two major theories.
One is that marriage may be protective of health. For example married couples may have advantages in terms of economic resources, social and psychological support and encouragement of healthy lifestyles.
A second possibility, she said, is marital selection — "the theory that healthy people get married and stay married, whereas less healthy people either do not marry or are more likely to become separated, divorced or widowed."
"Overall, this association between marital status and health persists regardless of socio-economic status, education and poverty, where people were born or their ethnicity," Schoenborn said. Among adults 18 and over, 11.9 percent said they were in fair or poor health, the study found. By comparison, some 10.5 percent of married people reported being in poor or fair health, while all other groups were higher. At 19.6 percent, widows and widowers were the most likely to be in these less healthy categories.
"In general, married adults were the least likely to experience health problems and the least likely to engage in risky health behaviors, with the notable exception of being overweight," Schoenborn wrote. The report was based on a survey of 127,545 people in 1999-2002 conducted by the center, part of the federal Centers for Disease Control and Prevention.
In addition to reporting better health overall, the study found that married people said they had less low back pain, fewer headaches and less psychological stress. They also were less likely to drink and smoke and were more physically active than people in general.
However, they're not immune to the weight problems plaguing America. Currently more than half of all adults are overweight or obese — 56.7 percent — the center said. Some 70.6 percent of husbands were overweight or obese compared with 65.1 percent of all men. For females, 48.6 percent of married women were overweight or obese, virtually the same as the 48.5 percent of women in general. Widows made up the largest share of overweight women at 53.2 percent.
An association between marriage and health was first reported in the 1970s. The relationship persists although much has changed since then. People are waiting longer to marry now, and living with a domestic partner outside of marriage has become more common, Schoenborn noted.
"For most negative health indicators, adults living with a partner had higher rates than married adults: they were more likely to be in fair or poor health, to have some type of limitation of activity for health reasons and to have experienced low back pain and headaches ... and serious psychological distress," Schoenborn reported.
The report found that married people were least likely to light up a smoke, at 18.8 percent, compared with 22.9 percent for all adults. The most likely to smoke were those living with an unmarried partner, 38.4 percent, and divorced and separated people, 34.7 percent. Some 4.7 percent of adults reported they had become heavier drinkers than previously, with the lowest rate among married at 3.7 percent.
Again, those living with an unmarried partner had the largest share reporting more drinking, 8.2 percent, followed by the divorced and separated, 6.4 percent.
Overall the study found that 58.2 percent of adults are married, 10.4 percent are separated or divorced, 6.6 percent are widowed, 19 percent are never married and 5.7 percent are living with a partner.
On the Net: National Center for Health Statistics: http://www.cdc.gov/nchs

Contrairement au résumé fait en première ligne du corps de l’article, il est difficile de conclure sur une causalité à partir d’une étude de corrélation. Et les deux hypothèses exprimées par la statisticienne dans cet article sont loin d’être les seules.
De façon très générale, l’existence d’une corrélation entre A et B peut signifier que A est la cause de B, que B est la cause de A ou que A et B, bien que ‘sans rapport’, ont une cause commune. Appliqué à cet article : les personnes mariées sont-elles en meilleure santé parce qu’elles sont mariées, ou sont-elles mariées parce qu’elles sont en meilleure santé, ou y a-t-il une raison qui fait que les personnes soient en meilleure santé et soient plus souvent mariées ? Rien dans cette étude ne permet de trancher.