No abstract

Background: The COVID-19 pandemic is a novel population-level stressor. As such, it is important to examine pandemic-related changes in mental health and to identify which individuals are at greatest risk of worsening symptoms.Methods: Online questionnaires were administered to 34,465 individuals in the UK, recruited from existing cohorts or via social media. Around one third (n = 12,718) with prior diagnoses of depression or anxiety completed pre-pandemic mental health assessments, allowing prospective investigation of symptom change. We examined changes in depression, anxiety and PTSD symptoms using prospective, retrospective and global ratings of change assessments. We also examined the effect of key risk factors on changes in symptoms. Research in contextEvidence before this study We conducted a literature search (PubMed, Scopus) with the terms "mental*" or "psychiatr*" and "covid*" or "coronavirus" published before 8th February 2021. This resulted in 4,573 unique references, but only 15 longitudinal studies examining changes in symptoms of mental health conditions from before to during the COVID-19 pandemic. Results to date are mixed. Some studies found increases in mental distress, some found increases in either depression or anxiety, and others saw no observable change in symptoms.Examining individual-level risk factors, heightened vulnerability to worsening mental health during the pandemic has been demonstrated among young people, females, individuals with lower incomes/financial problems and among health care or key workers. Only one previous study used a large sample with prior mental health diagnoses to examine changes in anxiety and depression. This study showed that having a prior mental health diagnosis was associated with higher levels of perceived worsening of mental health but, when examining actual prospectively measured symptoms, a prior mental health diagnosis was actually associated with a lower likelihood of symptom worsening, compared to no prior diagnosis. This discrepancy across measures requires further investigation in order to understand the nature of changing mental health during the COVID-19 pandemic. Added value of this studyThis study prospectively examined changes in symptoms of depression, anxiety and PTSD in a large UK-based sample of individuals with prior depression or anxiety. Analyses were supplemented with data from additional cohorts to examine individual difference risk factors with greater statistical power. Inclusion of both prospectively measured and retrospectively estimated changes in symptoms, as well as ratings of perceived change in mental health, allowed closer examination of discrepancies in subjective experience versus actual objective change in symptoms.people who are students or are unemployed. Additionally, discrepancies in estimated symptom change across prospective and retrospective measures highlight the importance of using prospectively collected data to examine longitudinal changes.

Background Retrospective self-reports of childhood trauma are associated with a greater risk of psychopathology in adulthood than prospective measures of trauma. Heritable reporter characteristics are anticipated to account for part of this association, whereby genetic predisposition to certain traits influences both the likelihood of self-reporting trauma and of developing psychopathology. However, previous research has not considered how gene–environment correlation influences these associations. Aims To investigate reporter characteristics associated with retrospective self-reports of childhood trauma and whether these associations are accounted for by gene–environment correlation. Method In 3963 unrelated individuals from the Twins Early Development Study, we tested whether polygenic scores for 21 psychiatric, cognitive, anthropometric and personality traits were associated with retrospectively self-reported childhood emotional and physical abuse. To assess the presence of gene–environment correlation, we investigated whether these associations remained after controlling for composite scores of environmental adversity across development. Results Retrospectively self-reported childhood trauma was associated with polygenic scores for autism spectrum disorder (ASD), body mass index (BMI), post-traumatic stress disorder (PTSD) and risky behaviours. When composite scores of environmental adversity were controlled for, only associations with the polygenic scores for ASD and PTSD remained significant. Conclusions Genetic predisposition to ASD and PTSD may increase liability to experiencing or interpreting events as traumatic. Associations between genetic predisposition for risky behaviour and BMI with self-reported childhood trauma may reflect gene–environment correlation. Studies of the association between retrospectively self-reported childhood trauma and later-life outcomes should consider that genetically influenced reporter characteristics may confound associations, both directly and through gene–environment correlation.