3 Mar 2016
March 4, 2016. Results from a new study show that people with psychiatric disorders pair with partners who also have a psychiatric disorder more frequently than would be expected by chance. The study, led by Ashley Nordsletten of the Karolinska Institute in Sweden and published online on February 24 in JAMA Psychiatry, may help solve some mysteries about the role of genetics in these disorders, and it may also require some recalibration of genetic analyses.
"I think this is the first time this has been found," said Matthew Keller of the University of Colorado, Boulder, who was not involved with the study, "and certainly it's the first time that somebody's been able to do this with sample sizes that allow us to have confidence that the correlations that are reported are probably pretty close to being true."
In an editorial accompanying the paper, Robert Plomin, Eva Krapohl, and Paul O'Reilly of King's College London wrote, "Nordsletten and colleagues report the first general population study to date of assortative mating for psychiatric disorders, which may help to solve three puzzles in psychiatric genetics: Why are psychiatric disorders so highly heritable when they are associated with reduced fecundity? Why are some psychiatric disorders so much more highly heritable than others? Why is there so much genetic comorbidity across psychiatric disorders?"
The researchers were able to conduct such a large study—which included hospital and outpatient psychiatric data from more than 700,000 people, including more than 70,000 diagnosed with schizophrenia—thanks to Sweden's comprehensive population registers. The study included people who had been diagnosed with at least one of 11 psychiatric disorders (schizophrenia, bipolar disorder, autism spectrum disorder, anorexia nervosa, substance abuse disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, major depressive disorder, social phobia, agoraphobia, and generalized anxiety disorder).
The study also included people with select non-psychiatric diseases (Crohn's disease, type 1 and type 2 diabetes, multiple sclerosis, and rheumatoid arthritis). For each subject, the researchers matched five control individuals based on age, sex, and location.
To identify the opposite-sex partners of these subjects, Nordsletten and colleagues used marriage records and birth certificates. They found that people with schizophrenia were much less likely to have married or mated compared to their matched controls: 26 percent of men with schizophrenia had mated compared to 80 percent of the matched controls; 51 percent of women with schizophrenia had mated compared with 84 percent of the matched controls. This is in line with previous studies that have reported decreased fecundity in people schizophrenia (Power et al., 2012).
The researchers calculated the correlations between people with a given psychiatric diagnosis and the corresponding diagnostic status of their mates, compared to the matched controls. For almost every psychiatric disorder, people with that disorder were significantly more likely to have a partner with the same diagnosis compared to the matched controls (i.e., a person with schizophrenia was more likely to have a partner with schizophrenia than a control person was to have a partner with schizophrenia). Spousal correlations for the same diagnostic status ranged from 0.11-0.48 and were especially high—about 0.4—for people with schizophrenia, ASD (0.48 for men and 0.45 for women), and ADHD (0.45 for men and women). The spousal correlations for mood disorders were considerably lower—between 0.15 and 0.19. For people with non-psychiatric disorders, meaningful spousal correlations were rare. The spousal correlations for Crohn's disease were -0.02 and -0.03, for example.
"I found the magnitude of the correlations within [psychiatric] disorders very interesting," Keller told SRF. "It's a lot bigger than what we've seen with a lot of other traits that are the poster children for assortative mating—like height or IQ—which are in the 0.25 range."
These results may also shed light on why some psychiatric disorders—like schizophrenia—appear to be more heritable than others; namely, these disorders have more assortative mating. However, as Plomin and colleagues note, "From the findings by Nordsletten and colleagues alone, one would predict that DNA-based heritability estimates should be greater for ADHD, ASD, and schizophrenia than for affective disorders, but the results so far suggest otherwise." It will be up to follow-up studies to unravel this mystery.
Nonrandom mating across disorders
Intriguingly, there were also strong correlations between different disorders, including the thought disorders. In particular, people with schizophrenia were more likely to mate with people with ASD compared to their matched controls (spousal correlations were 0.26 and 0.42).
This finding of cross-associative mating may help explain a puzzle in psychiatric genetics: Why there is so much genetic comorbidity across disorders? For example, genetic correlations between schizophrenia and ASD are about 0.2 (see SRF related news report). However, one wrinkle in this idea, as Plomin et al. point out, is the lower cross-assortative mating between people with schizophrenia and bipolar disorder (0.15) reported in this study, despite a high genetic correlation of 0.6 (see SRF related news report).
Assortative mating might also help explain why psychiatric disorders appear to be increasing in the population. "Now that we live in a huge modern industrialized society in which people can move about and choose mates a lot more freely than they could have several hundred years ago, it does raise the possibility that the rates of disorders are actually increasing over the last five or 10 generations due to assortative mating," Keller told SRF.
Nordsletten wrote in an email to SRF that systemic nonrandom mating "could impact the genetic variance of offspring at the population level in a manner that would heighten risk for these conditions," although she also cautioned: "It is important to note that any such risk increase would not imply a determinant risk for any child, as psychiatric disorders arise from a complex interplay of genes and environment which vary not only by disorder, but also gender and even time."
Implications for future studies
This study may have profound implications for future psychiatric genetics research—especially for studies that use genetic models that assume random mating. "Our findings suggest that this is not an accurate assumption in many cases, and thus models should allow for the correlation of spouses to avoid biases in heritability estimates (for instance, in twin studies, where neglect of these spousal correlations may result in an underestimation of disease heritability)," Nordsletten wrote to SRF.
Keller agreed with this sentiment: "Assortative mating mucks up our estimates of heritability; it can lead to biased estimates.… One of the important things that this paper is going to do is make investigators realize that they've got to account for the spousal correlation when they're designing genetic studies of psychiatric disorders."
There may also be clinical implications of this work. As the Plomin editorial pointed out, "Beyond genetics and genomics, assortative mating matters because it means that the person closest to an individual with a psychiatric disorder is also likely to have psychiatric problems, which could exacerbate problems for both spouses and their offspring."—Summer E. Allen.
Nordsletten AE, Larsson H, Crowley JJ, Almqvist C, Lichtenstein P, Mataix-Cols D. Patterns of Nonrandom Mating Within and Across 11 Major Psychiatric Disorders. JAMA Psychiatry. Published online February 24, 2016. Abstract
Plomin R, Krapohl E, O'Reilly PF. Assortative Mating—A Missing Piece in the Jigsaw of Psychiatric Genetics. JAMA Psychiatry. Published online February 24, 2016. Abstract