There is considerable evidence that exposure to stressful life events can increase the risk that a vulnerable individual will go on to have a first psychotic episode. It has also been established that a substantial proportion of those who have a first psychotic episode have another episode in their lifetime, typically leading to poor clinical and psychosocial outcomes. After an initial episode, up to one in two individuals will suffer a relapse within 2 years severe enough to require hospital admission.

For this reason, say the authors of a new study appearing in Lancet Psychiatry, “it is important to identify risk factors for relapse that individuals could recognize and monitor themselves.” Knowledge of risk factors, they say, could also help researchers develop targeted interventions to prevent relapse.

A team led by Sagnik Bhattacharyya, M.D., Ph.D., of King’s College London, set out to clarify what is known about the association between stressful life events and psychosis relapse, focusing in particular on causality. Can a causal relation be established between relapse and exposure to (and number of) stressful life events following a first episode of psychosis? Alexis E. Cullen. Ph.D., a 2019 BBRF Young Investigator, and Dr. Robin Murray, Fellow of the Royal Society and 1999 BBRF Distinguished Investigator, were also members of the research team.

In prior research, the team had reviewed evidence from 23 studies investigating the relationship between adult stressful life events and psychosis relapse. While 18 of these studies suggested that life events occurring after illness onset are associated with higher likelihood of relapse, the team judged them to be non-comparable, due to a variety of factors. Specifically, they studied patients at different stages of psychosis, with different forms of the disorder, and who varied markedly in terms of sociodemographic and clinical factors. More important, the studies were not designed to establish whether stressful life events played a causal role in relapse.

For its new study, the team recruited 256 hospitalized individuals treated for onset of psychosis in London, average age 28, 61% of whom were male. This cohort was assessed carefully at illness onset and again 2 years later, in interviews that assessed exposure to, and emotional and stressful impact of, 12 severe or major life events, dates of which were recorded. Participants were at the same stage of illness, and all were documented and could be followed in the electronic patient records of the U.K.’s National Health Service. Thus, there was a record of any subsequent hospitalizations related to psychosis. Inpatient admission to hospital because of symptom exacerbation was the criterion used for “relapse.”

The team looked specifically for a “dose-response” relationship, i.e., whether the number of exposures to stressful life events proportionately raised the risk of relapse. They were also careful to account for factors that can skew the results if not accounted for statistically, for instance, genetic and environmental factors that do not change over time. In addition, they were alert to “reverse causation”—the possibility that psychosis relapse itself precipitates stressful life events.

After analyzing the data they collected, the researchers concluded that reverse causation was not, in fact, relevant. Rather, they found “converging evidence of a causal effect of stressful life events on the risk of relapse in psychosis.” Specifically, they found higher risk of relapse in individuals exposed to any post psychosis-onset stressful life events, compared with those unexposed; they found greater risk of relapse with each additional stressful exposure; higher incidence and longer duration of psychosis relapse in those exposed to any post-onset stressful life event; and an increase in the incidence and duration of relapse for each additional exposure.

The magnitude of the impact of stressful life events in the 2 years following psychosis onset upon the likelihood of relapse was likened by the team to the impact of a patient failing to adhere to antipsychotic medicines in the same interval.

“Our results have important practice and policy implications,” the researchers wrote. “They suggest the need to focus on development of interventions that bring together conventional therapies (e.g., psychotherapy) and psychoeducation, as well as sensitive awareness campaigns and complementary approaches (e.g., lifestyle changes) to help mitigate the harmful effects of stressful life events” in order to prevent psychosis relapse.