Exploring the genetic etiology across the continuum of the general psychopathology factor: a Swedish population-based family and twin study.

Abstract

Psychiatric comorbidity can be accounted for by a latent general psychopathology factor (p factor), which quantifies the variance that is shared to varying degrees by every dimension of psychopathology. It is unclear whether the entire continuum of the p factor shares the same genetic origin. We investigated whether mild, moderate, and extreme elevations on the p factor shared the same genetic etiology by, first, examining the linearity of the association between p factors across siblings (N = 580,891 pairs). Second, we estimated the group heritability in a twin sample (N = 17,170 pairs), which involves testing whether the same genetic variants influence both extreme and normal variations in the p factor. In both samples, the p factor was based on 10 register-based psychiatric diagnoses. Results showed that the association between siblings' p factors appeared linear, even into the extreme range. Likewise, the twin group heritabilities ranged from 0.42 to 0.45 (95% CI: 0.33-0.57) depending on the thresholds defining the probands (2-3.33 SD beyond the mean; >2 SD beyond the mean; >4.33 SD beyond the mean; and >5.33 SD beyond the mean), and these estimates were highly similar to the estimated individual differences heritability (0.41, 95% CI: 0.39-0.43), indicating that scores above and below these thresholds shared a common genetic origin. Together, these results suggest that the entire continuum of the p factor shares the same genetic origin, with common genetic variants likely playing an important role. This implies, first, genetic risk factors for the aspect that is shared between all forms of psychopathology (i.e., genetic risk factors for the p factor) might be generalizable between population-based cohorts with a higher prevalence of milder cases, and clinical samples with a preponderance of more severe cases. Second, prioritizing low-cost genome-wide association studies capable of identifying common genetic variants, rather than expensive whole genome sequencing that can identify rare variants, may increase the efficiency when studying the genetic architecture of the p factor.