PREGNANCIES HAMPERED BY COKE, SMOKE

Pregnant women who smoke cigarettes or use cocaine have a higher risk of
miscarriage, the most common problem of pregnancy, a study has found.

While other studies have suggested a link, this was the first to use hair
and urine testing to determine women's drug use and smoking, instead of
relying on their own reports.

"One of the things we have learned from this study is that self-reporting
is far from perfect," said Dr. Roberta Ness, of the University of
Pittsburgh School of Public Health, lead author of the study reported in
Thursday's New England Journal of Medicine.

Smokers are almost twice as likely to miscarry as nonsmokers, and cocaine
users are nearly one-and-a-half times as likely to miscarry as nonusers,
according to the study.

About 15 percent of pregnancies end in spontaneous abortions treated by
doctors, but many women do not seek medical care for early miscarriage.

The researchers looked at 970 women who sought emergency room treatment for
miscarriage or other problems at the Hospital of the University of
Pennsylvania in Philadelphia while they were less than 22 weeks pregnant.
Most of the women were black, poor, single and had not received prenatal
care.

Four hundred suffered a miscarriage, while 570 carried their pregnancies
beyond 22 weeks.

Hair sample tests showed nearly 29 percent of the women who miscarried had
used cocaine, compared to 20 percent of women with continuing pregnancies.
Urine tests showed about 35 percent of the women who miscarried were
smokers, compared to 22 percent of the nonsmokers.

In interviews, only 14 percent of the women who miscarried admitted using
cocaine, and only 30 percent admitted smoking cigarettes.

The researchers did an impressive job of isolating the effects of tobacco
and cocaine on fetal health, said Dr. James Mills, of the National
Institute of Child Health and Human Development.

Still, the link between cocaine use and miscarriage was not entirely
persuasive, Mills wrote in an accompanying editorial. The increased risk
was small and other factors could have skewed the results, he said.

The study found no link between drinking or marijuana use and spontaneous
abortion.