Children born at short and long interpregnancy intervals face heightened risk of autism, according to a Columbia University study.

The researchers said that the risk is higher in children born either less than 1 year or more than 5 years of the birth of their prior sibling. Meanwhile, children delivered between 2 and 5 years are not in danger.

"It was intriguing to see that the risk of ASD diagnosis was higher in both closely and distantly spaced pregnancies. It is important to realize that we can't say from this study that spacing of pregnancies per se is a cause of ASD-this is most likely a proxy of other factors that are more directly related to the chance of the child's developing ASD. In other words, the importance of this finding lies in the clues that it can provide in terms of understanding how the prenatal environment is related to outcomes after birth," Dr. Cheslack-Postava said in a press release.

For the study, the researchers analysed data of 7371 children, born between 1987 and 2005 in Finland, from the Finnish Prenatal Study of Autism (FIPS-A). About a third of the children in the study were diagnosed with autism. The researchers also used information from several national registries to compare the interval of pregnancies between autistic and normal children.

The researchers found that autism risk is one and half time higher in children conceived less than a year following a sibling's birth than those conceived following an interval of 24-59 months. Meanwhile, children conceived following an interval of 60-120 months were 30 percent more likely to be diagnosed with autism. The risk was 40 percent for intervals of over 120 months.

The factors that are most responsible for the risk are parents' age, prior number of children, and parental history of psychiatric disorders.

"This study provides further evidence that environmental factors occurring during or near the prenatal period play a role in autism, a serious and disabling condition that afflicts millions of individuals and that is increasing in prevalence. This work also exemplifies the importance of large samples of pregnancies with data acquired during pregnancy and their linkage to comprehensive, national databases of reproductive factors and psychiatric diagnoses," the senior author of the study, Dr. Alan Brown of Columbia University, said.

The finding is published in the Journal of the American Academy of Child and Adolescent Psychiatry