Childhood trauma and premorbid adjustment among individuals at clinical high risk for psychosis and normal control subjects

Abstract

Aim

Traumatic childhood experiences are associated with psychotic illness and are frequently reported in patients at clinical high risk (CHR) for psychosis. Moreover, deteriorating premorbid functioning from childhood, and through adolescence, is related to greater severity of overall symptomatology and poorer outcomes in patients with psychosis. We studied the prevalence of traumatic childhood experiences and premorbid adjustment and their association with each other in patients at CHR for psychosis and normal control subjects (NCSs).

Methods

A total of 20 CHR patients for psychosis and 30 NCSs aged 14 to 35 participated in the present study. The CHR patients were identified as prodromal to psychosis using the Structured Interview for Prodromal Syndromes/Scale of Prodromal Symptoms. Premorbid adjustment was assessed by using the premorbid adjustment scale (PAS), and self-reported childhood trauma was assessed with the Trauma and Distress Scale (TADS).

Results

In CHR patients, TADS and PAS scores were higher than in NCSs. In CHR patients, TADS correlated significantly with the PAS general section and observably, but not significantly, with adolescence and adulthood sections.

Conclusion

CHR patients reported more childhood trauma experiences and poorer premorbid adjustment than NCSs. In CHR patients, traumatic childhood experiences are associated with poor general premorbid adjustment.

Introduction

Several studies point to a significant association between traumatic childhood experiences (e.g. sexual, physical, and/or psychic abuse or neglect) and the later development of psychotic illness. Population-based studies suggest that childhood trauma is associated with later psychosis.[1-4] Self-reported childhood trauma is prevalent in individuals with psychotic disorders[4, 5] and reveals higher rates than are found in non-clinical groups.[4-7] Traumatic childhood experiences are frequently reported in patients identified as ‘prodromal’ to, or at heightened clinical risk for, psychosis.[8, 9] Additionally, trauma experiences, especially previous sexual trauma, may be a predictor of onset of psychotic disorder in patients at clinical high risk (CHR) for psychosis.[9]

Premorbid functioning has received much attention in the field of psychosis research. The terms premorbid functioning and premorbid adjustment refer to an individual's social, interpersonal, academic and occupational functioning in the period before the onset of psychotic symptoms.[10] Multiple lines of research show that poorer premorbid functioning is related to greater severity of overall symptomatology, cognitive impairment, negative symptom and social functioning.[11-13] Deteriorating premorbid functioning from childhood, and through adolescence, has been linked to earlier onset of schizophrenia[14, 15] and long duration of untreated psychosis.[16, 17] Moreover, a recent study suggests that premorbid functioning is predictive of transition to a psychosis in subjects at high risk for psychosis.[18] However, little is known about premorbid functioning and its relationship to childhood trauma experiences in clinical samples of adolescents and young adults at high risk for psychosis.

The present study aimed to examine an association between premorbid adjustment and childhood trauma in patients at CHR.

We aim to address the following questions:

1. Do the risk patients have more traumatic childhood experiences than the normal control subjects?
2. Do the risk patients have poorer premorbid functioning than the normal control subjects?
3. Is there an association between premorbid adjustment and childhood trauma in patients at CHR?

Methods

The study is a part of the Detection of Early Psychosis Project which was carried out in the Turku area. The research plan was approved by the Ethical Committee of Turku University and Turku University Hospital.

Subjects

A total of 20 CHR patients for psychosis and 30 normal control subjects (NCSs) aged 14 to 35 (mean age = 23.1, SD = 5.2) participated in the present study. The risk patients were identified as prodromal to psychosis using the Stuctured Interview for Prodromal Syndromes/Scale of Prodromal Symptoms (SIPS/SOPS[19]). Patients at high risk for psychosis had to meet the criteria for one of three definitions of the prodromal state. Inclusion criteria included a new treatment period (either the first or at least 6 months after the end of the last treatment). The patient sample was recruited from in- and outpatient clinics for young and adult psychiatric patients in The Hospital District of Southwest Finland. Initially, the control sample was recruited from the population register of a previous early psychosis study, the EPOS study (for a short description of the recruitment of controls, see Klosterkötter et al.[20] (Finnish control sample)), but, because the subjects did not significantly differ from the patient sample in terms of age or education years, the same control sample was used in the present study. All subjects provided either informed consent (for those aged 18 and older) or assent with parental consent (for those under age 18).

Table 1 shows the demographic information about the study groups. Patients at CHR for psychosis did not significantly differ from the control subjects in terms of age, gender, education years, work situation or marital status. The groups differed in living situation. The CHR patients lived with parents more often than the NCSs, and less frequently alone or with spouse and/or own family.

Table 1. Demographic characteristics across comparison groups (mean ± SD or n (%))

	CHR (n = 20a)	NCS (n = 30)	Statistic
^a Marital status, living situation and work situation missing for one CHR patient. CHR, clinical high risk of psychosis; NCS, normal control subject; ns, non-significant.
Age (years)	22.2 ± 5.4	23.6 ± 5.1	ns
Gender			ns
Female	15 (75)	19 (63.3)
Male	5 (25)	11 (36.7)
Marital status			ns
Single	13 (68.4)	19 (63.3)
Married	0	4 (21.1)
Cohabiting	6 (31.6)	7 (23.3)
Living situation			χ^2(3) = 13.0, P < 0.01
Lives alone	5 (26.3)	14 (46.7)
With parents	6 (31.6)	0
With spouse and/or own family	7 (36.8)	16 (53.3)
Other	1 (5.3)	0
Years of education	12.6 ± 3.4	13.6 ± 2.2	ns
Work situation			ns
Full-time work	2 (10.5)	12 (40)
Part-time work	1 (5.3)	1 (3.3)
In full-time education	9 (47.4)	13 (43.3)
Unable to work	4 (21.1)	1 (3.3)
Unemployed	2 (10.5)	0
In military/civilian service	0	2 (6.7)
Other	1 (5.3)	1 (3.3)

Measures

Interviews

The SIPS/SOPS[19] was used to determine candidacy for the study. The SIPS/SOPS is a semi-structured clinical interview that identifies CHR status on the basis of the presence of attenuated psychotic symptoms, brief intermittent psychotic symptoms and/or genetic risk with recent functional deterioration.

The premorbid adjustment scale (PAS[10]) was used to determine the premorbid development of the subjects. PAS is one of the most widely used retrospective rating scales which assesses the ‘degree of achievement of developmental goals’ over the course of childhood, adolescence and, where applicable, adulthood. The scale is divided into a general scale and four distinct developmental age periods: childhood to age 11, early adolescence to age 15, late adolescence to age 18 and adulthood. Individual items in the childhood and adolescence categories assess premorbid adjustment by asking about sociability and social withdrawal, peer relationships, scholastic performance, adaptation to school and ability to form sociosexual relationships. Ratings in the adult period focus on social relationships, while the General Scale ratings are broader, including educational achievement, social relationships, and level of interest in and enjoyment of major life activities (work, family, etc.). Higher scores indicate poorer adjustment. All interviews were administered by trained interviewers.

Questionnaires

Self-reported childhood trauma was retrospectively assessed with the Trauma and Distress Scale (TADS[21]), a 43-item self-rating scale that probes for traumatic experiences that occurred in childhood and adolescence. The respondent estimates the frequency of traumatic experiences using a five-point scale (0 = never, 4 = almost always). Specific experiences assessed by the TADS are grouped into five domains: sexual abuse, physical abuse, emotional abuse, physical neglect and emotional neglect. Trauma in each domain was defined as existing if the subject obtained a value from 2 (sometimes) to 4 (almost always) on any item of the domain. The total trauma score was calculated by summing the scores of the five domains.

Data analysis

All statistical analyses were performed using PASW 18 statistical software (SPSS Inc., Chicago, IL, USA). The statistical significance level for testing the hypothesis was set at P < 0.05. Non-parametric tests were used because sample sizes were small. Comparative analyses were performed with the help of the Mann–Whitney U-test for continuous variables and with chi-square test for categorical variables. Associations between two continuous variables were analysed with the Spearman correlation coefficient.

Results

Age did not correlate significantly with the total trauma score (r = −0.04, P = 0.811) or domains (r = −0.18–0.14, P > 0.05), but did correlate with general premorbid adjustment (r = −0.36, P = 0.006). Gender had no significant effects on total trauma score (U = 229, P = 0.455) or on general premorbid adjustment (U = 300, P = 0.455).

All patients met SIPS/SOPS criteria for the attenuated positive symptom prodromal syndrome, and four (20%) also met the criteria for genetic risk (with a first-degree relative with psychotic illness) with recent functional decline. No one met the criteria for brief intermittent psychotic symptoms.

A history of traumatic experiences and premorbid adjustment

All patients and 18 (60%) control subjects endorsed at least one trauma experience, mainly general emotional or physical neglect (CHR = 20 (100%), NCS = 10 (33.3%)). Seven patients (35%) and three (10%) controls reported a history of physical abuse. Fifteen patients (75%) and four (13.3%) controls endorsed emotional abuse. Four patients (20%) and one control (3.3%) reported sexual abuse. The patients at high risk for psychosis differed significantly from control subjects in their history of traumatic experiences and premorbid adjustment (see Table 2).

Table 2. Trauma and premorbid adjustment scores across comparison groups

	CHR (n = 20a)	NCS (n = 30)	Statistic
^a Adolescence section missing for two subjects and adulthood section missing for four subjects due to young age of the subjects. General section missing for one subject. CHR, clinical high risk of psychosis; NCS, normal control subject.
Trauma and Distress Scale (TADS)
Emotional abuse	5.9 ± 4.4	1.4 ± 1.6	U = 500.5, P < 0.001
Physical abuse	3.0 ± 4.5	0.4 ± 0.6	U = 416.5, P < 0.01
Sexual abuse	1.9 ± 4.4	0.3 ± 1.5	U = 371.5, P < 0.05
Emotional neglect	9.0 ± 4.5	3.0 ± 2.3	U = 523, P < 0.001
Physical neglect	4.5 ± 2.8	2.2 ± 1.9	U = 655.5, P < 0.001
TADS sum	24.1 ± 16.3	7.1 ± 5.3	U = 529, P < 0.001
Premorbid adjustment scale
Childhood	0.3 ± 0.2	0.1 ± 0.1	U = 725, P < 0.001
Adolescence (early)	0.4 ± 0.2	0.1 ± 0.1	U = 852, P < 0.001
Adolescence (late)	0.5 ± 0.3	0.1 ± 0.1	U = 707.5 P < 0.001
Adulthood	0.5 ± 0.3	0.1 ± 0.1	U = 499, P < 0.001
General	1.1 ± 0.4	0.7 ± 0.1	U = 612, P < 0.001

Associations between childhood trauma and premorbid adjustment

Traumatic childhood experiences were not significantly associated with childhood and early adolescence premorbid adjustment. However, there was an observable association between sexual abuse and childhood premorbid adjustment. Late adolescence premorbid adjustment was significantly and highly associated with emotional abuse, and strongly but not significantly associated with physical neglect. There was also a strong but not significant association between adulthood and emotional abuse. In addition, general premorbid adjustment was significantly and highly related to traumatic childhood experiences: emotional abuse, emotional neglect and psychical neglect (see Table 3).

Table 3. Associations of trauma scores with the Premorbid Adjustment Scale (PAS) for the patients at high risk for psychosis

PAS (n = 19)	Trauma score total	Emotional abuse	Emotional neglect	Physical abuse	Physical neglect	Sexual abuse
*P ≤ 0.05; **P ≤ 0.01. Note: Adolescence section missing for two subjects and adulthood section missing for four subjects due to young age of the subjects. General section missing for one subject.
Childhood	0.180	0.060	0.249	−0.020	0.209	0.399
Adolescence (early)	0.307	0.346	0.251	0.155	0.253	0.278
Adolescence (late)	0.466	0.649**	0.367	0.259	0.480	0.298
Adulthood	0.431	0.526	0.340	0.146	0.380	0.194
General	0.593**	0.580*	0.487*	0.071	0.582*	0.139

Table 4 shows associations between the trauma scores and the subsectors of the PAS general section in the CHR patients. Both occupational performance and social adjustment were related to the total trauma score. Specifically, emotional abuse, emotional neglect and physical neglect were associated with general adjustment, most notably social–personal adjustment.

Table 4. Associations of trauma scores with the subsectors of the Premorbid Adjustment Scale (PAS) general section in the clinical high-risk patients

(n = 18) PAS general	Trauma score total	Emotional abuse	Emotional neglect	Physical abuse	Physical neglect	Sexual abuse
*P ≤ 0.05; **P ≤ 0.01.
Education	0.036	0.151	−0.045	−0.232	−0.070	−0.147
In paid employment or functioning in school	0.568*	0.518*	0.431	0.500*	0.533*	0.296
Change in work or school performance occurred	0.044	0.081	0.067	0.116	0.057	0.199
Frequency of job change or interruption of school attendance	0.530*	0.571*	0.368	0.300	0.513*	0.214
Establishment of independence	0.304	0.186	0.412	0.119	0.100	0.089
Global assessment of highest level of functioning	0.517*	0.311	0.554*	0.195	0.528*	0.412
Social–personal adjustment	0.597**	0.605**	0.464	0.002	0.661**	0.481*
Degree of interest in life	0.318	0.340	0.309	−0.064	0.363	0.134
Energy level	0.380	0.408	0.363	0.041	0.351	0.135
General total	0.593**	0.580*	0.487*	0.071	0.582*	0.139

Discussion

In line with previous research,[8, 9] our results show that the patients at CHR for psychosis generally reported traumatic childhood experiences. Additionally, our findings show that the CHR patients are more likely to have a history of traumatic childhood experiences, including emotional, physical and sexual trauma, than the NCSs. In our data, all patients endorsed at least one trauma experience which is consistent with the previous study of Thompson et al.[8] (97%). In a study by Bechdolf et al.,[9] 70% of the CHR patients reported trauma experiences. However, we found that 60% of healthy control subjects also reported trauma experiences, suggesting that some kinds of childhood trauma experiences are highly prevalent. We showed that a fifth of the patients have experienced sexual abuse. This figure is quite congruent with the findings of Thompson et al. (27%) and Bechdolf et al. (28%). Additionally, our finding any incidence of emotional abuse (75%) among risk patients is congruent with the finding of Thompson et al. (67%). We found that 35% of the risk patients reported physical abuse, which is less than reported in the previous study of Thompson et al. (83%), but more than reported in the study of Bechdolf et al. (26%). That difference is explained by the inclusion of a highly traumatized ethnic minority in the research of Thompson et al. which decreases the comparability of the results. Moreover, there were differences between the studies in how trauma was categorized and measured, so the prevalence of other types of trauma cannot be compared.

To our knowledge, this was the first investigation which studied an association between traumatic childhood experiences and premorbid adjustment in patients at CHR for psychosis. The patients at CHR had poorer premorbid adjustment than the controls, and it was retrospectively reported as impaired throughout their development. As hypothesized, premorbid adjustment was associated with self-reported exposure to trauma. In particular, emotional abuse and physical neglect were associated with late adolescence and general premorbid adjustment. There was also an observable, but not significant, relationship between emotional abuse and adulthood or early adolescence premorbid adjustment. According to Dragt et al.,[18] of the general section of the PAS, five items are associated with developing psychosis in CHR patients. We found three of them (frequency of job change or interruption of school attendance, global assessment of highest level of functioning, social–personal adjustment) to be strongly related to traumatic experiences, particularly, physical neglect, emotional abuse and emotional neglect. The social–personal adjustment item was also associated with sexual abuse. Bechdolf et al. demonstrated that previous sexual trauma may be a predictor of onset of psychotic disorder in patients at CHR.[9] Altogether, it seems that traumatic experiences may be predictive of transition from CHR to psychosis, but further research is needed. Traumatic childhood experiences were related to both occupational performance and social adjustment. The findings are consistent with previous studies suggesting that childhood maltreatment is associated with the development of interpersonal and productivity problems among young adults.[22, 23] According to the prospective study of Currie and Widom,[22] childhood abuse is related to later unemployment, lower education level, and problems with occupation in early and middle adulthood. Although most of our findings are congruent with these, we did not find an association between childhood maltreatment and education. However, this difference is explained by the young age of our subjects.

There are some limitations to consider, the first being that our study reports cross-sectional clinical data, and thus we do not know how many patients assessed to be at CHR for psychosis will go on to develop manifest psychosis. According to follow-up studies of the CHR patients fulfilling the SIPS/SOPS criteria, the conversion rate to schizophrenic psychosis was 46–54% at 12 months.[19] Thus, there is reason to suppose that adolescents and adults assessed to be at CHR for psychosis are at real risk of psychosis, though not necessarily of schizophrenia. Second, the PAS and the TADS are retrospective measures and thus might be weakened by recall bias. Both the PAS and the TADS were based on self-report recall without regular corroboration from family members or other sources. However, a recent study[24] found only small and non-significant differences between retrospective and contemporaneous ratings of premorbid functioning, thus supporting the value of the PAS based on self-report data. Additionally, reports of abuse by psychiatric patients are surprisingly reliable and congruent with contemporaneous evidence.[25-28] Third, the sample sizes of the groups were small and included more females than males. The control subjects were not exactly matched with patients at CHR and a history of previous psychosis of patients was not controlled for.

Acknowledgements

The authors are indebted to the psychiatric outpatient teams of the study districts for the invaluable help they gave us in the data collection of the screening phase. The Turku University Central Hospital has supported the study financially.