Alicia outlines findings from our recent paper investigating gene-environment correlation in the association between heritable traits and retrospective self-reports of childhood trauma.


Alicia Peel, EDIT Lab PhD student

When childhood trauma is assessed using retrospective self-reports it is associated with a greater risk of poor mental and physical health in adulthood than prospective accounts, such as court records. But why are these retrospective self-reports of childhood trauma associated with a greater risk of poor outcomes than more objective records?

One proposed explanation for this is genetic confounding, whereby genetic predisposition to certain traits influences both the reporting of traumatic events and of developing a mental health disorder. Genetic confounding can make it seem that two concepts, such as self-reporting trauma and mental health, are more strongly associated than they actually are, as the same genetic factors increase the likelihood of both.

We can estimate an individual’s genetic predisposition for any heritable trait using polygenic scores. Simply, polygenic scores are the sum of all the variants related to a trait that a person carries, weighted by how strongly that variant is associated with the trait. Using this approach, previous research has shown that individuals with a high genetic predisposition to autism, attention-deficit hyperactivity disorder, schizophrenia, depression, and neuroticism were more likely to retrospectively self-report childhood trauma in adulthood. This finding provides support for the genetic confounding hypothesis, indicating that genetic predisposition may partly account for the association between self-reported childhood trauma and poor mental health.

However, this research often does not take into account the different mechanisms through which genetics could influence self-reports. As well as acting through the individual directly, genetics also impact a person’s environment. This is called gene-environment correlation, and it occurs when genetic factors influence both an individual’s liability for a trait and the corresponding environments that they are exposed to.

For example, parents with high levels of anxiety will pass on some of this genetic predisposition to their child, and their behaviours will also shape the child’s home environment.

Our study aimed to investigate whether the associations between genetically-influenced reporter characteristics and retrospective self-reports of childhood trauma are accounted for by gene-environment correlation. We used information from nearly 4,000 individuals from the Twins Early Development Study, TEDS, a cohort of twins born in England and Wales in the mid 1990s. One twin from each pair was randomly selected from TEDs to create a sample of unrelated adults with genetic data.

First, we assessed whether polygenic scores capturing genetic liability for a range of demographic, psychiatric, personality, and cognitive traits were associated with retrospective self-reports of childhood trauma. To assess the presence of gene-environment correlation, we then investigated whether these associations remain after controlling for environmental adversity across development.

We found that polygenic scores for autism spectrum disorder, body mass index (BMI), post-traumatic stress disorder (PTSD) and risky behaviours were independently associated with a greater likelihood of reporting childhood trauma. When we controlled for environmental adversity, the associations between the polygenic scores for risky behaviour and high BMI were no longer present. This indicates that gene-environment correlation may be occurring, whereby genetic liability for these traits could increase the likelihood of experiencing adverse environments. On the other hand, the polygenic scores for autism and PTSD remained significant even after controlling for environmental adversity. This suggests that genetic liability to these traits may directly influence reporting of trauma, possibly by increasing an individual’s sensitivity to experiencing, interpreting, or recalling events as being traumatic.

These results emphasise the importance of considering how the subjective experience relates to the outcomes of self-reported trauma. For research, they reinforce the importance of using genetically sensitive designs when exploring associations between self-reported experiences and later life outcomes, as genetic predisposition may confound these associations, both directly and through gene–environment correlation.

Read the full paper:

Peel, A.J., Purves, K.L., Baldwin, J.R., Breen, G., Coleman, J.R., Pingault, J.B., Skelton, M., ter Kuile, A.R., Danese, A. & Eley, T.C. (2022). Genetic and early environmental predictors of adulthood self-reports of trauma. The British Journal of Psychiatry, 1-8.


We gratefully acknowledge the ongoing contribution of the participants in the Twins Early Development Study (TEDS) and their families.

Alicia Peel

Author Alicia Peel

More posts by Alicia Peel