Trauma associated with the loss of a child might contribute to the onset of psychosis, according to an analysis of data from the National Comorbidity Survey Replication.
Findings from this cross-sectional survey that suggest the trauma of losing a child could be an etiologic factor in the onset of psychosis are a significantly later age of onset of psychosis among those experiencing the loss of a child, onset within a year of the trauma in many subjects, and the specificity of the association with psychosis compared with the specificity of the association for other psychiatric conditions. These findings build on the increasing evidence of an association between various types of trauma exposure and psychosis, Jordan E. DeVylder of Columbia University, New York, and his colleagues reported online in the Aug. 29 issue of Psychiatry Research.
For example, in the National Comorbidity Survey, childhood physical abuse significantly predicted psychosis, and cumulative trauma exposure was found to increase the likelihood of psychosis in a dose-response fashion – a finding that was replicated in subsequent studies.
In the current study, 409 of 6,458 participants in the National Comorbidity Survey Replication (NCS-R) reported the loss of a child, and 251 reported a lifetime psychotic experience.
There was a statistical trend toward a higher incidence of psychosis among those who had lost a child, with 18/409 of those who had lost a child also having a psychotic experience (odds ratio = 1.52), the investigators said (Psychiatry Res. 2012 Aug. 29 [doi:10/1016/j.psychres.2012.08.008]).
Child loss significantly predicted the age of onset of psychosis (age of onset was 36.4 and 27.2 years for those experiencing and not experiencing child loss). After demographic factors, number of children, and co-occurring diagnoses were controlled for, a main effect of child loss was seen, as was an interaction between child loss and gender; the effect of bereavement on age of psychosis onset was greater for men than for women, the investigators noted.
Also, psychosis onset followed child loss in nearly 60% of participants (4 of 4 males, and 6 of 13 females), and in 60% of those subjects, psychosis onset was within 1 year of child loss.
The specificity of the association between child loss and psychosis was evaluated using additional regression models. After investigators controlled for confounding variables such as number of living children, age, female gender, nonwhite race, foreign birth, and other psychiatric conditions, child loss was not shown to have a significant effect on age of onset for six other conditions, including posttraumatic stress disorder, major depression, substance abuse, generalized anxiety disorder, mania, and social phobia.
"The presented results together demonstrate that, in some individuals, child loss may contribute specifically to the onset of psychotic experience, but not other psychiatric conditions, even when controlling for confounding factors," the investigators wrote.
Participants in the NCS-R were U.S. adults aged 18 years or older who were surveyed between February 2001 and April 2003. The investigators "examined the relationship between child loss and age of psychosis onset, building on prior population-based studies by using a measure of trauma exposure that is less likely to be underreported or subject to recall bias, by examining psychosis onset rather than prevalence, and by controlling for co-occurring conditions," they said.
The main finding of this study is that child loss is associated with a later age of onset of psychosis among a large sample of people with a lifetime psychotic experience. However, it is of particular interest that psychosis onset occurred within 1 year of the loss of a child in a large proportion of those who experienced child loss (35.3% of the total sample), the investigators said.
This demonstrates a very close temporal relationship between child loss and psychosis in some individuals, confirming findings from a prior study of patients undergoing psychiatric hospitalization in the year after the loss of a child and extending those findings to nonhospitalized parents who experienced psychosis.
Furthermore, the lack of an effect of child loss on six other psychiatric conditions suggests that the relationship may be specific to psychosis and "not reflective of a broader influence of child loss on global mental health," although it is possible that child loss influences these conditions through mechanisms other than those represented by the models used in this study, they noted.
The fact that the overall increase in the odds of developing psychosis among those who experienced child loss did not reach statistical significance was not surprising, given the unlikelihood that most people with psychosis experience child loss – and that most people experiencing child loss develop psychosis.