Older fathers and the risk of psychiatric disorders in their offspring

For many decades, the most well-known association between parental age and offspring outcome was the relationship between advanced maternal age and the risk of Down syndrome in the youth. Over the last decade, since the pioneer work of Dolores Malaspina in the early 2000’s, there has been an increasing awareness of the relationship of advanced paternal age and the risk for a variety of mental health diagnoses, most commonly schizophrenia.

This fact should not be surprising as in mammals the vast majority of new mutations into the gene pool are actually introduced by the males, due to their constantly dividing spermatogonia accumulating mutations as we age. In an average male, by the age of 40 the germ cell precursors will have undergone 660 divisions, while the number of such divisions in females is usually 24, all but the last occurring during fetal period. These new mutations, introduced by males as they age, has been associated with a number of diagnoses, such as achondroplasia, Apert syndrome, and progeria. With schizophrenia being considered a complex brain disorder, how could we study such an association?

It turns out that a number of small epidemiological studies, in the 1950s and ’60s, had called attention to the fact that patients with schizophrenia seemed to have later paternal age. That finding was then forgotten until the late 1990s when Malaspina and her collaborators decided to take a second look at that association in a large population-based birth cohort in Israel, the Jerusalem Perinatal Study. They linked the data from that study to the national registry of psychiatric illness maintained by the State of Israel. Looking at this detailed, comprehensive, and reliable source they calculated the incidence of schizophrenia stratified by parental age. They found that fathers in the youngest category (those under 25 years of age) had the lowest incidence of schizophrenia diagnosis in their offspring (2.5/1000) and that that incidence increased progressively as they looked at older fathers by 5-year increments, to 3.5 per 1000, 3.7 per 1000, 4.4 per 1000, 4.6 per 1000, and 5.0 per 1000, reaching 11.4 per 1000 in the offspring of fathers 50 years or older. They also looked at maternal age and did not find any such risk.

Since that initial report, published in 2001, there have been over 20 studies looking at the relationship between advanced paternal age and schizophrenia. In 2011, we were able to summarize data from 12 such epidemiological studies – both cohorts and case-controls – into a meta-analysis examining schizophrenia risk and paternal age. Looking at all these studies combined, we found a consistent increased risk for offspring schizophrenia as paternal age increased until the maximum risk category which included fathers fifty years of age or older. The increased risk appeared significant as early as 30 years of age with the highest risk being for those with fathers over 50 years of age. Also in 2011, another meta-analysis was published supporting the hypothesis that advanced paternal age is also a risk factor for autism in the offspring. In reviewing all the autism data the authors estimated that offspring of men over the age of 50 had more than double the risk of autism (a relative risk of 2.2 in statistical/epidemiological terms) compared to children of men younger than 30; these estimates were controlled for maternal age and other know autism risk factors.

All these findings lead to a more recent, and thought provoking, study recently published in JAMA Psychiatry. In that study, John McGrath – an Australian epidemiologist expert in schizophrenia risk factors – and a team from Denmark used the magnificent national registry of that Scandinavian country. Data on over 2.8 million people, born in Denmark between 1955 and 2006, were included. Since they had complete data on parents age as well the medical history of each individual, the effects of parental age – both parents – was analyzed and the risk estimated, not only for schizophrenia but for a broad range of psychiatric diagnosis. At first look it seemed that those born from both younger and older parents, compared to parents between ages 25 and 29, were at higher risk for a psychiatric diagnosis. However, at closer examination, a more nuanced, and interesting, picture emerged. First, not all psychiatric disorders are linked to early or advanced parental age as diagnoses such as schizoaffective disorders, eating disorders, and bipolar disorder, showed negligible or no such association. Second, fathers with advanced age seemed in deed to have a higher risk of having offspring with schizophrenia or autism – in their study advanced paternal age was defined as 45 years of age. Third, children of younger mothers seemed to have an increased risk of hyperkinetic disorders, like attention-deficit disorder, as well as behavioral and emotional disorders with onset in childhood and adolescence. Fourth, children of younger parents – both father and mother – are at increased risk of psychiatric problems associated with the use of alcohol and other psychoactive substances.

These observations coming from large epidemiological studies confirm the impression that we have a lot to learn by observing populations as they live their lives. Results like the ones described above can link to genetic mechanism – like the hypothesis that de novo mutations that may be emerging in the male germline – as well as the difficulties and tribulation of teen pregnancy, with all its social, cultural, and economic determinants. We have to be aware of these associations, think about the complexity of our human existence, and hope we can help each other and become a better society for all.

Erick Messias, MD, PhD, is the medical director of the Walker Family Clinic in the University of Arkansas for Medical Sciences’ Psychiatric Research Institute and an associate professor in the UAMS Department of Psychiatry.