Researchers used an observational cohort study to investigate a potential causal association.
“Ms Taylor” is a 24-year-old African-American female with a history of schizophrenia. She experienced her first episode of psychosis at age 22 years. She was stabilized on antipsychotic medication and has been stable for the past 2 years. She is currently living with her family, taking classes part-time at the local community college and volunteering at a local animal shelter. Four months ago, her elderly dog had to be euthanized. Then, 2 months ago, a beloved uncle died unexpectedly following complications of an infectious illness.
Ms Taylor is currently in the local psychiatric emergency department. She is actively attending to internal stimuli and is endorsing persecutory beliefs and passive suicidal ideation. Her urine drug screen is negative, and there is no history of alcohol or illicit drug use. Collateral history from her parents indicates good adherence with psychotropic medications, for which they supervise administration. Her parents inquire whether these recent stressful life events could have contributed to her illness relapse. How would you advise them?
Illness relapse following a first episode of psychosis is the rule rather than the exception.1 Identification of factors associated with illness exacerbation could help identify individuals at heightened risk for relapse and develop targeted interventions. Stressful life events are associated with psychosis risk2 and potentially with illness relapse,3-5 although findings for the latter are difficult to interpret.
It is possible that the association between stressful life events and relapse of psychosis could reflect shared genetic and environmental factors, increasing the risk of both. Reverse causation is also possible (eg, illness relapse might increase risk of loss of employment).
The Current Study
Bhattacharyya and colleagues6 investigated the association between stressful life events and relapse of psychosis by combining multiple inferential approaches. In order to address issues of temporality, the authors used a prospective design to investigate the effects of stressful life events on risk within 2 years after onset for psychosis, controlling for potential confounding factors, including sociodemographic and clinical factors and medication adherence.
They investigated the effects of stressful life events across a range of outcomes, tested for a dose-response relationship, and applied a fixed-effects analyses of longitudinal data. They also used a cross-lagged path analysis approach to investigate the directionality of the association.
The authors recruited individuals with first-episode psychosis (ICD-10) presenting to psychiatric services in the catchment area of South London, United Kingdom. Participants were aged 18 to 65 years, and there were no exclusion criteria for comorbid conditions. Individuals were followed prospectively for 2 years. They were assessed via face-to-face interview at baseline and either face-to-face or via telephone at follow-up, with review of electronic clinical records.
Stressful life events since the onset of psychosis were assessed via interview using the List of Threatening Experiences questionnaire7 at the follow-up interview, with dates of occurrence of the events to determine temporality. Psychosis relapse was defined as inpatient admission due to symptom exacerbation within 2 years of onset. Potential confounders included sex, age of onset, relationship status, ethnicity, alcohol and substance use, smoking, intensity, medication adherence, and onset diagnosis.
The primary outcome was relapse of psychosis during the 2-year follow-up. Other outcomes included number of relapses, length of relapse, and time to first relapse. The effect of any stressful life events on relapse were analyzed using Cox proportional hazards regression in a multivariate model, controlling for potential confounding factors.
Multiple negative binominal regression analyses were performed for number and duration of relapses. Fixed-effects models were used to analyze whether the effect of stressful life events on psychosis relapse was explained by unmeasured confounders varying across individuals but not over time. Fixed-effect regression models were also used to investigate reverse causation.
The study sample consisted of 253 individuals with first-episode psychosis recruited between 2002 and 2013. Approximately 61% of individuals were male, 55% were Black African or Caribbean, and 34% were white. The mean age of onset of psychosis was 28 years. Within 2 years of psychosis onset, 36% of the 253 individuals experienced at least 1 relapse of psychosis requiring hospitalization.
Individuals experiencing at least 1 stressful life event after psychosis onset (n=47, 19%) had a significantly increased risk of relapse (Hazard ratio [HR]=2.60, 95% confidence interval [CI] 1.63-4.13), after controlling for potential confounders. The inclusion of medication adherence in the model did not attenuate the association. Furthermore, there was a significant dose-dependent association between the number of stressful life events and risk of relapse (HR=1.36, 95% CI 1.09-1.69).
In fixed-effects analyses, the risk of relapse was significantly higher during the year in which stressful life events preceded relapse of psychosis after controlling for time-varying factors (Odds Ratio [OR]=4.33, 95% CI 2·00–9·37). Again, there was a dose-response relationship between the number of stressful life events and odds of relapse (OR=1.62, 95% CI 1.18-2.21) in the adjusted analyses.
The number of relapses within the first year of follow-up did not predict the number of stressful life events occurring over the second year of follow-up, suggesting a unidirectional effect of exposure to stressful life events on relapse of psychosis.
The authors used causal inference methods to investigate relationships between stressful life events and psychosis, with findings providing converging evidence of a causal effect.
They found a higher adjusted risk of relapse in individuals exposed to any post-onset stressful life events compared with those unexposed; a greater risk of relapse for each additional exposure to post-onset stressful life events (a dose-response effect); a higher incidence and longer duration of psychosis relapse in individuals exposed to any post-onset stressful life events compared with those unexposed; and an increase in the incidence and duration of relapse for each additional exposure to post-onset stressful life events (dose response).
Furthermore, the authors found a higher risk and greater number of relapses during the year in which stressful life events preceded relapse of psychosis. However, there was no effect of relapse of psychosis or number of relapses on the risk of exposure to or number of stressful life events (ie, no evidence of reverse causation).
Study limitations include the retrospective, self-reported assessment of post-onset stressful life events and confounders (eg, medication adherence and substance use), a lack of consideration of the emotional impact of stressful life events, the use of observational data, and potential effects of cognitive impairment on recall bias.
The Bottom Line
Collectively, evidence from this study is consistent with a causal interpretation of the association between stressful life events and relapse of psychosis, and not the reverse. Targeted biopsychosocial interventions are needed to help mitigate the harmful effects of stressful life events and prevent psychosis relapse.
Dr Miller is a professor in the Department of Psychiatry and Health Behavior at Augusta University in Augusta, Georgia. He is on the Editorial Board and serves as the schizophrenia section chief for Psychiatric Times®. The author reports that he receives research support from Augusta University, the National Institute of Mental Health, and the Stanley Medical Research Institute.
1. Alvarez-Jimenez M, Priede A, Hetrick SE, et al. Risk factors for relapse following treatment for first episode psychosis: a systematic review and meta-analysis of longitudinal studies. Schizophr Res. 2012;139(1-3):116-128.
2. Beards S, Gayer-Anderson C, Borges S, et al. Life events and psychosis: a review and meta-analysis. Schizophr Bull. 2013;39(4):740-747.
3. Leff J, Kuipers L, Berkowitz R, et al. Life events, relatives' expressed emotion and maintenance neuroleptics in schizophrenic relapse. Psychol Med. 1983;13(4):799-806.
4. Birley JL, Brown GW. Crises and life changes preceding the onset or relapse of acute schizophrenia: clinical aspects. Br J Psychiatry. 1970;116(532):327-333.
5. Martland N, Martland R, Cullen AE, Bhattacharyya S. Are adult stressful life events associated with psychotic relapse? a systematic review of 23 studies. Psychol Med. 2020;50(14):2302-2316.
6. Bhattacharyya S, Schoeler T, Di Forti M, et al. Stressful life events and relapse of psychosis: analysis of causal association in a 2-year prospective observational cohort of individuals with first-episode psychosis in the UK. Lancet Psychiatry. 2023;10(6):414-425.
7. Brugha T, Bebbington P, Tennant C, Hurry J. The List of Threatening Experiences: a subset of 12 life event categories with considerable long-term contextual threat. Psychol Med. 1985;15(1):189-194.