Work-Package I:
With data from the Icelandic register data on 47 thousand parents who lost a child during the past 200 years, we found that mothers who lost a child were compared to their full siblings at increased risk of premature mortality (death before the age of 50) across the 200 years of follow-up (Valdimarsdóttir et al., eLife 2019). In the Swedish registers, we further identified 415 thousand individuals who lost a first-degree family member (parents, siblings, partners, children) with dramatically increased risks of all psychiatric disorders (except schizophrenia and eating disorders) compared to their full siblings and population controls (Yufeng et al., manuscript is publication process). The sibling design inherently adjusts for familial factors, including genetic factors; thus, the limited differences in results obtained in the population vs. the sibling comparison do not suggest a considerable familial contribution to major morbidities or mortality after the loss of a relative.
Using Swedish nationwide registry resources, we aimed to discover new disease-related phenotypes associated with stress and trauma. We identified all individuals in Sweden diagnosed with trauma/stress-related disorders (N>100.000) and found that they were, compared to their siblings and population-controls, at increased subsequent risk of several autoimmune diseases (Song et al., JAMA 2018), most cardiovascular diseases (Song et al., BMJ 2019), neurodegenerative disease (Song et al., JAMA Neurol 2020), all tested life-threatening infections (Song et al., BMJ 2019) as well as overall mortality (Tian et al., Lancet Regional Health 2022). Again, contrary to our hypothesis, the limited differences in results between sibling- and population comparisons do not suggest a notable familial component in risks of somatic diseases or mortality after trauma/diagnosis of stress-related disorders.
Finally, we have used data from Icelandic registers and UK Biobank to understand the genetic contribution to risks of psychiatric disorders and cardiovascular disease after cancer diagnosis. We found that the risks of cardiovascular disease (Chen et al., BMC Med 2022) as well as depression or anxiety (Ge et al., manuscript in preparation) during the first year after cancer diagnosis was not modified by PRS for these disorders. These results indicate that a cancer diagnosis is a robust stressor influencing the risks of these adverse outcomes independent of genetic propensity for these diseases/disorders.
Work-Package II:
We established cohorts of 31,795 women in Iceland (The SAGA cohort) and 5,000 individuals exposed to the SA-Asian Tsunami in 2004 (The Swedish Tsunami Cohort). These data combined with the Swedish Register data (e.g. on Swedish Twins) have been used identify predictors of posttraumatic symptomology and evaluate the genetic contribution to posttraumatic stress disorder (PTSD).
Through the nationwide representative SAGA cohort, we have demonstrated that the high lifetime prevalence of interpersonal violence is a major contributor to gender disparities in mental health (Jónsdóttir et al., Lancet Publ H, 2022; Unnarsdóttir et al., Sleep 2022; Thordardóttir et al., manuscript in publication process) and development of cardiometabolic disease (Lynch et al., Gísladóttir et al., manuscripts in publication process). We have further reported on adverse childhood experiences and their role in adult mental health and resilience (Daníelsdóttir et al., eLife 2022), premenstrual disorders (Yang et al., BMC Medicine 2022), and risk of obesity (Kristjánsdóttir et al., manuscript in preparation).
With respect to the Swedish tsunami cohort, we identified early evacuation from disaster areas as a risk factor of long-term posttraumatic symptoms (Gudmundsdóttir et al, Scand J Publ Health 2019) and in an extended follow-up, we found Swedish tsunami survivors to be at increased risk of stress-related disorders up to six months after the traumatic event (Thordardottir et al., Epidemiology 2022).
In modern genetics it is generally recognized that large sample sizes are required to obtain novel findings. Thus, with respect to the genetics of PTSD, we are contributing our data to an ongoing global effort through the Psychiatric Genomic Consortium. The preliminary findings from 1.2 million individuals were presented at the World Congress of Psychiatric Genetics identifying 82 genome-wide significant loci (Maihofer et al., 2022). The final results will be published in 2023. Furthermore, in ancillary studies to the StressGene program, we demonstrated through a sibling comparison a strong familial component in stress-related disorders (including PTSD) (Li et al., Molecular Psychiatry, 2022) and identified two novel loci associated with long-term PTSD in childhood cancer survivors (Lu et al., Transl Psychiatry 2022)
Our StressGene program has already generated more than 25 publications, many of which are published in leading scientific journals with many more pending.
