Premature birth may increase the risk for serious mental illness in adolescence and young adulthood, a recent study reports.
Researchers reviewed birth and hospital admissions records of more than 1.3 million Swedes born from 1973 to 1985. They found that compared with those born at term, young adults born very premature — at less than 32 weeks’ gestation — were more than twice as likely to be hospitalized for schizophrenia or delusional disorders, almost three times as likely for major depression, and more than seven times as likely for bipolar illness.
The lead author, Chiara Nosarti, a senior lecturer in neuroimaging at Kings College London, emphasized that while the increase in relative risk is substantial, the absolute increase in numbers of people with the illnesses is not.
“Despite these findings,” she said, “the majority of people born preterm have no psychiatric problems, and the number of people hospitalized with psychiatric disease is very low.”
Still, she added, “routine screening may help to detect early signs of illness.”
The risk also increased for people born late preterm, or 32 to 36 weeks’ gestation, but not as sharply. They were 60 percent more likely to be admitted for schizophrenia or delusional disorders, 34 percent more likely for depressive disorder, and about twice as likely to be hospitalized for bipolar illness.
The researchers acknowledge that the findings could be affected by factors they could not control for, including unmeasured sociodemographic factors, a family history of preterm delivery, maternal substance abuse or smoking, and bacterial or viral infections.
But experts not involved in the study were impressed with the work. “This is an important, expertly executed epidemiological study,” said Catherine Monk, an associate professor of clinical psychology at Columbia who conducts research in perinatal psychiatry and neuroscience. “It provides compelling evidence that our mental health trajectories are initiated early in development.”
Dr. Monk, too, noted that the study did not mean that being born prematurely inevitably leads to mental illness.
“The risk found is still low in an absolute sense, and finally it, too, can be modified by other factors in the child’s development,” she said.
Dr. Bradley Peterson, director of the Center for Developmental Neuropsychiatry at the New York State Psychiatric Institute, said the size of the sample was “stunning” and “allows very strong levels of confidence that the findings are not spurious.”
The study’s authors said there was a plausible biological explanation for their finding. The preterm brain is particularly vulnerable to injury, and functional M.R.I. imaging of young adults born very preterm has found disruptions in brain networks similar to those found in psychiatric patients. In addition, genetic factors that would not by themselves lead to illness could be activated by preterm birth.
Dr. Peterson agreed that the study offered strong evidence “for a causative relationship rather than simply an association.”
The study, published online last month in Archives of General Psychiatry, used obstetric data from the Swedish Medical Birth Register, which includes information on more than 99 percent of all hospital births in the country since 1973 and contains prospectively collected information on the 95 percent of Swedish women who attend antenatal clinics.
The data on hospitalization came from the National Hospital Discharge Register, which contains records on all hospital diagnoses. The researchers connected the birth records to hospitalization data using the personal identification numbers assigned to all Swedish citizens.
The scientists considered two other pregnancy outcomes in addition to preterm birth: birth weight and Apgar score — a general measure of a newborn’s health — at five minutes. Being small for gestational age was significantly associated only with hospitalization for drug or alcohol dependency, and Apgar score only with depressive disorder.
The association of hospitalization for psychiatric illness with preterm birth persisted after controlling for Apgar score, poor fetal growth, maternal sociodemographics and maternal psychiatric history.