The Life Course Epidemiology of Psychotic Symptoms in Schizophrenia and Other Psychoses: from environment to psychosis

John Grace QC PhD Scholarship 2015 -  University College London, Division of Psychiatry,

Supervisor: Dr James B Kirkbride 


Schizophrenia risk is strongly linked to urban living and ethnic minority status. We don’t know, however, whether these factors act on specific psychotic symptoms which underpin schizophrenia (i.e. hallucinations, delusions, paranoia, changes in mood, impaired thinking or social withdrawal). In this PhD research project we will use statistical data from two population-based studies to (a) investigate which psychotic symptoms are most strongly linked to detailed social & environmental factors in childhood and adolescence, (b) establish the ages when such associations are strongest, and (c) examine the pathways through which these factors increase risk of symptoms. 

Research Student: Jennifer Dykshoorn

I am a PhD student, working with Dr. James Kirkbride and Dr. Glyn Lewis in the Division of Psychiatry at UCL. My current research focuses on social and spatial determinants of psychotic disorders, including schizophrenia. For part of my research, I am collaborating closely with researchers at the Karolinska Institutet. 

My research investigates the timing of exposure to adverse social environments in order to explore how these experiences influence the risk of developing psychotic disorders. I will be using longitudinal data available through record linkages in Sweden as well as the ALSPAC birth cohort from Bristol in order to investigate these questions.           

My research is made possible through the generous support of Mental Health Research UK and UCL Overseas Research Scholarship. 

Start Date: September 2015
Scientific goal: 

The scientific goal of this proposed PhD project is to (a) test which types of psychotic symptoms are most strongly linked to socio-environmental risk factors, (b) establish 
when such associations are strongest over the life course, and (c) examine the direct and mediated pathways through which such risk factors contribute to psychosis aetiology. 

2017 Report 

It’s been another exciting, research-intensive year at UCL. I am continuing my research on psychotic disorders including schizophrenia and bipolar disorder with psychotic symptoms, with a particular focus on the social environmental factors that have been linked to the development of these disorders.

In my first year, I initiated a project investigating the role that age-at-migration and region of origin have on the development of psychotic and non-psychotic disorders utilizing data from the Swedish population. Over the past year, I have finalized the results from this investigation (see Figure 1) and have had the opportunity to present these findings at the Epidemiology & Social Psychiatry section meeting of the European Psychiatry Association as well as at the International Congress on Schizophrenia Research. I am in the process of submitting a manuscript based on this research for publication.

Figure 1: Hazard ratios by age-at-migration (to be added later)

Building from this research, which demonstrated increased risk of psychotic disorders in migrant groups, I have started to investigate other factors in the social environment that could explain the increased risk of schizophrenia and other psychotic disorders in migrants. In my current work, I have developed a novel measure of family-based social networks to investigate if migrating with family or joining family who already reside in Sweden mitigates schizophrenia risk.
I am using the Swedish population registers, with 3.3 million individuals born between 1968 and 1997. I have conducted survival analysis to determine if family status alters the risk of psychosis. In order to conduct this analysis, I collaborated closely with researchers at the Karolinska Institutet and have greatly benefited from two extended research visits to Sweden. This has allowed me to work closely with experts in the registers and has greatly enhanced my understanding of these data. I look forward to strengthening this research relationship over the coming years of research.

I am excited for the next phase of research, where I will finalize these results and initiate research in factors affecting psychosis risk among children of migrants. In addition to making progress on my research questions, I successfully defended my research in the “upgrade” process at UCL in October, contributed to a book chapter on the epidemiology of public mental health, and continued to enhance my teaching skills through lecturing and assisting the MSc courses in statistics and epidemiology. I have also had the opportunity to attend several workshops and conferences to increase my epidemiological skills as well as build my network to other researchers in mental health and epidemiology.

2016 Report

Psychotic disorders, including schizophrenia and bipolar disorder with psychotic symptoms, are serious mental illnesses that have a debilitating impact on individuals and society. While both genetic and environmental markers have been linked to the development of psychotic disorders, it is unclear whether there are sensitive windows during which exposure to adverse social environments, including migration and social exclusion, increases risk.
My research investigates the timing of exposure to such factors in relation to the development of psychotic disorders, with the hope that by better understanding the timing of exposures can provide important information to what underlying mechanisms are affected.

I am finishing the first analysis for this, where I was able to use Swedish registry data to investigate how timing of migration affects the risk of developing affective and non-affective psychosis. My analysis looked at over 2 million people, and I conducted a survival analysis which revealed an increased risk of psychosis for both first and second-generation migrants. Figures 1 and 2 show the age- and sex- adjusted hazard ratios ((which compare the rate at which diagnoses of psychotic disorders happen in each group (e.g. migrants, second-generation migrants) to the reference group (e.g. Swedish born)) for non-affective and affective psychosis by migrant status. These figures show a significantly increased risk for both diagnostic categories in both first and second-generation migrants, when compared to the Swedish-born (with Swedish-born parents) population.

Figure 1: Age- and sex-adjusted hazard ratios                                 
Figure 2: Age- and sex-adjusted hazard 
for non-affective psychosis by migrant status                                   ratios for affective psychosis by migrant status

* denote significantly increased hazard ratio in that group compared to the
reference category, in this case Swedish born individuals.

I also demonstrated the increased risk of psychosis appears to vary age at which migration occurred. I have been accepted to present these results as part of a symposium at the European Psychiatric Association’s Epidemiology and Social Psychiatry in November. In the next phase of research, where I will be using these data to investigate how timing of maternal migration affects risk among second- generation migrants. I will also examine if the increased risk of psychotic disorders varies according to country of origin. I hypothesize that the adverse stressors may be exaggerated among those who have moved from countries that are very culturally different from Sweden, which in turn may have an increased impact on mental health, including psychotic disorders.

In addition to making progress on my research questions, I have had the opportunity to co-author a book chapter, participate in teaching and lecturing the MSc students, and attending training courses and conferences to increase my knowledge and skills. I was also awarded the UCL Overseas Research Scholarship, which covers the difference between domestic and international fees. This award, in combination with the support from MHRUK, allows me to continue this fascinating research.