Introduction

Clinical High Risk for Psychosis (CHR), also known as prodromal, At-Risk Mental State (ARMS), and Ultra High Risk (UHR), denotes an elevated risk of developing psychosis1. CHR individuals are assessed using the Comprehensive Assessment of At-Risk Mental States (CAARMS) assessment tool2 to determine one of three syndromes: attenuated positive symptoms (APS) (sub-threshold psychotic symptoms), brief limited intermittent psychotic symptoms (BLIPS) (brief psychosis lasting less than one week), or genetic risk and/or deterioration (GRD) (Positive family history of psychosis plus a decline in functioning)2. The CHR concept was devised in order to identify at-risk individuals thereby affording the opportunity for preventative strategies3.

Psychosis has been deemed to be a multifactorial polygenic disorder with heritability estimates ranging between 31% and 44%4,5,6. While this finding suggests that the etiology of psychosis involves a significant genetic contribution, environmental insults are also thought to play a critical role. Adverse life events have repeatedly been implicated in the development of First Episode Psychosis (FEP)7 with 89% of FEP patients reporting one or more adversities compared to 37% of controls8. Specifically, childhood/adolescent sexual, physical, and emotional abuse, physical/emotional neglect, separation, and institutionalization were 4–17 times higher for the FEP group. Moreover, for each additional adversity, the risk of psychosis increased 2.5 times8. Similarly, CHR individuals were found to have experienced significantly more severe adverse events than controls, regardless of trauma subtype9. Specifically, CHR individuals were 5.5, 2.5, and 3.1 times as likely to report emotional abuse, physical abuse, and bullying victimization, respectively9. Yet, it is clear that not everyone who develops psychosis has experienced severe adversity, such as abuse, neglect, or separation, and that other environmental factors may also play a significant role. Specifically, psychosocial stress is emerging as a possible contributory factor in the onset of psychotic-like experiences in CHR individuals10. Psychosocial stress has been defined as any social or cultural situation that causes physical, emotional, or psychological strain on an individual11. The physiological effects of psychosocial stress include increased heart rate and variability, skin conductance, decreased brain volumes, inflammation, alteration in hypothalamic–pituitary adrenal axis function, and increased cortisol secretion12. Psychological consequences of psychosocial stress include reduced self-esteem and motivation, increased negative affect, aggression, and withdrawal from social situations12. These negative effects can increase the risk of psychopathology, which is consistent with the Stress-Vulnerability Model13. This model posits that an individual’s predisposing bio-psychosocial vulnerability (biological, psychological, and social risk factors) interacts with stress caused by various life experiences leading to the manifest illness such as depression, anxiety, as well as psychosis13. Therefore, an individual with a high bio-psychosocial vulnerability will only need to experience a low level of (internal or external) stress in order to develop psychosis; while in contrast, an individual with a low overall level of bio-psychosocial vulnerability will need to experience a high level of stress in order to manifest the illness. Behavioral Sensitization has been proposed as a possible mechanism to account for the relationship between stress and psychosis symptoms14. This notion suggests that cumulative exposures to environmental insults produces an increased sensitivity to stress and elevated emotional responses to similar stressors subsequently experienced14. Indeed, early experiences of trauma and life events have been found to contribute to increased stress sensitivity in adulthood15,16,17 and patients with psychosis have been found to react with more intense emotions to perceived stress in daily life compared to controls18,19, giving credence to the behavioral sensitization concept. Therefore, the stress-vulnerability model denotes the effects of cumulative stress on a pre-existing trait vulnerability, while the concept of stress sensitivity refers to the magnitude of affective arousal in response to repeated stressors.

Exposure to psychosocial stress has been found to be higher in the CHR individuals compared to the general population20. Contrarily, another study found that the exposure to psychosocial stress may actually be comparable between these groups but was found to have a greater negative impact in the CHR group21. Indeed, a greater perception of psychosocial stress was associated with more severe positive symptoms in a CHR group compared to help-seeking controls22. Greater subjective distress in response to psychosocial stressors in CHR individuals may in part be accounted for by individual constitutional characteristics such as personality and temperament22, in addition to their level of sensitization, whereby repeated exposure to stress leads to an elevated affective response to subsequent stressors14,23. It is important to note that most measures of stress do not differentiate general stressors from psychosocial stressors perhaps because most, if not all, stressful events would contain a social component, highlighting the difficulty in ascertaining the differences between these two types of stressors. However, it is possible that they may exert similar effects on the stress response system if they are both appraised as challenging, threatening, or harmful24. It has therefore been suggested that the appraisal of an event as stressful, rather than the type of event, may be important in understanding the relationship between stress and the onset of psychosis25.

In regards to possible constitutional factors, interpersonal sensitivity refers to the undue and excessive awareness of, and sensitivity to, the behavior and feelings of others. This concept comprises of interpersonal awareness, a fragile inner self, need for approval, separation anxiety, and timidity26. High levels of interpersonal sensitivity have been characterized by avoidant behaviors such as social withdrawal and appeasement behaviors so as to avoid conflict or rejection by complying with the expectations of others27. The aforementioned coping strategies employed by individuals with high interpersonal sensitivity inadvertently affects social performance and functioning28.

Alongside personality traits, behaviors such as social withdrawal have also been investigated in psychosis. Social withdrawal can be defined as retreat from interpersonal relationships usually accompanied by an attitude of indifference and detachment29. Social withdrawal often leads to social isolation, loneliness, disturbed sleep hygiene, loss of support, and the development of psychiatric conditions30. Substance misuse, mood disorders, and psychotic disorders are one of many psychopathologies that can be associated with avoidant behaviors such as social withdrawal. In relation to psychosis, some have suggested that it precedes its onset31, while others argue it is a consequence of the disorder32. Compared to controls, CHR individuals exhibit greater levels of social withdrawal33, which is associated with increased symptomatology, such as positive and negative symptoms, reduced psychosocial and occupational functioning, and increased suicidal thoughts and substance misuse30. These behaviors in turn may contribute to the formation and persistence of psychosis symptomatology. Indeed, higher levels of social withdrawal have been associated with an increased likelihood of transition to psychosis34.

Psychosis symptomatology therefore appears to be influenced by a plethora of social factors such as psychosocial stress, interpersonal sensitivity, and social withdrawal. The aforementioned social factors have been observed to influence the development of symptoms and the progression of the illness, as well as impacting on long-term outcomes. Nevertheless, the reliability of these associations within the literature remains unclear. Furthermore, it remains unclear whether social factors are precipitating factors, perpetuating factors, or both. Therefore, a synthesis of the existing literature may help to elucidate the influence of psychosocial stress in individuals at CHR and its role in the transition to psychosis. To date, no systematic review has been conducted on the impact of different types of psychosocial stressors in CHR individuals incorporating the role of social behaviors and personality characteristics.

Therefore, the aim of this review was to summarize the existing evidence regarding the relationship between psychosocial stress, interpersonal sensitivity, and social withdrawal on transition to psychosis in CHR individuals. The following outcomes will be considered in the included papers: CHR status vs. controls, psychosis/affective symptomatology, and rate of transition to psychotic disorders.

Method

Protocol and registration

This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines35. Methods and inclusion criteria were specified in advance and documented in a protocol registered with the International Prospective Register of Systematic Reviews (PROSPERO; PROSPERO registration: CRD42021264478).

Search strategy and selection criteria

A systematic search of Ovid (PsychINFO, EMBASE, MEDLINE, and GLOBAL HEALTH) was conducted, including studies from database conception to February 28, 2022. The following search strings were used: (at-risk mental state OR ultra-high risk OR clinical high risk OR attenuated psycho* OR prodrom* OR transition or conver* or psycho*) AND (interpersonal sensitiv* OR interpersonal awareness OR relational sensitiv* OR social withdrawal OR social avoidance OR social network OR social stress* OR social advers* OR psychosocial stress*).

Any length of follow-up and any date of publication were included. Eligible studies measured psychosocial stress, interpersonal sensitivity, or social withdrawal and CHR status. Studies written in languages other than English and conference abstracts were excluded from the review. The study selection process is summarized in the PRISMA flow diagram (see Fig. 1).

Fig. 1
figure 1

PRISMA flow diagram.

Full size image

Data extraction process

A data extraction form was developed in Microsoft Excel. One review author (PR) extracted the following data from the included studies: (1) participant characteristics (CHR status) and the included paper’s inclusion and exclusion criteria; (2) outcome measures (psychosocial stress, interpersonal sensitivity, social withdrawal); (3) additional outcomes (including symptom severity, transition to psychosis, early/recent trauma); (4) statistical analysis used; (5) risk of bias assessment outcome; and (6) main findings (including means, standard deviations, effect sizes, and confidence intervals, where available). Study screening was performed independently by one reviewer and was subsequently cross-checked by a second reviewer (KM). Disagreements were resolved by consensus. All study designs were included in the review, except case studies.

The Grading of Recommendations Assessment, Development and Evaluation working group methodology36 was employed to assess the quality of evidence by examining the following domains: risk of bias, consistency, directness, precision, and publication bias. The quality of evidence was therefore rated as high, moderate, low, or very low. The risk of bias and certainty assessment was also performed.

Analysis

Due to the expected low power and sparsity of literature, a descriptive summary of findings was planned. Summaries of each study were written in a Microsoft Word document, which, combined with the data extraction form and the Grading of Recommendations Assessment, were used to draw out analytical themes. No additional analyses were conducted.

Results

A total of 5222 articles were retrieved during the initial search and 4684 articles remained following de-duplication. Through the initial screening involving title and abstract review 4518 articles were removed, thus 166 full-text evaluations took place. The full-text evaluations resulted in 29 articles matching the inclusion criteria for this review. Articles were excluded if they did not examine CHR status and at least one of the following variables: psychosocial stress, interpersonal sensitivity, or social withdrawal. Studies were also excluded if no full-text was available, if it was classified as a conference abstract, was not written in English, and if the study examined virtually constructed social situations rather than real-life occurrences.

Study characteristics

This review included 29 studies published from 1999 to 2021, with most of the studies published from 2011 onwards. Out of 29 studies, 25 of them were carried out in Western countries (USA, Canada, UK, Netherlands, Italy, and Australia), while only 4 studies were carried out in non-Western countries (China, Seoul, and Brazil). The total sample size of included studies was n = 3143. The sample size ranged from n = 25–764, with the average sample size consisting of 108 participants. The participants were mainly recruited through Universities or specialized clinics and GP referrals, while all the assessors possessed a relevant postgraduate or doctoral degree. Out of all the included studies, 16 were cross-sectional in design, whereas 13 were longitudinal (1 to 5-year follow-up). In addition, one study employed a combined cross-sectional and longitudinal design.

Population characteristics

Participants’ age ranged from 16 to 29 (average = 20). In terms of gender, males comprised more than 50% of the sample in all of the studies. The measurement of socio-economic status was not reported in 6 studies, while other studies measured educational achievement, employment status, or social class. The average years spent in education for CHR subjects was ~13.69, which was reported in 4 studies, while 6 studies looked at education level through the highest level of attainment. Three studies reported that the highest level of education was high school, while 3 studies reported it as university completion. Eight studies examined employment status of the clinical sample with 4 studies reporting higher rates of unemployment amongst CHR population and 4 studies reporting the opposite. The ethnicity of study participants was only reported in 13 studies. The samples were predominantly Caucasian, ranging from 50–80% of the sample. The exception was two studies37,38, which included a higher proportion of ethnic minority participants. Lastly, control groups were demographically matched to the clinical sample.

Clinical and psychosocial stress measures

The following measures were employed to determine CHR status and transition to psychosis: Structured Interview for Prodromal Symptoms (SIPS)39; Comprehensive Assessment of At-Risk Mental States (CAARMS)2; Prodromal Questionnaire (PQ)40; Schizophrenia Proneness Inventory-Adult41; Present State Examination (PSE)42; Scale of Prodromal Symptoms (SOPS)43; Diagnostic Statistical Manual-Fourth Edition44; Diagnostic Interview for Psychosis45; and the Structured Clinical Interview for DSM Disorders (SCID)46.

Furthermore, the frequency and severity of psychosocial stress was measured using the following outcome measures: Childhood Trauma and Abuse scale47; Behavior Assessment for Children-Second Edition (BACS-2)48; The Schedler-Westen Assessment Procedure-200049; The Daily Stress Inventory50; Childhood Trauma Questionnaire51; Coddington Life Events Record52; Life Events Scale53; Kiddie Schedule for Affective Disorders and Schizophrenia-Present and Lifetime version54; Schedule of Recent Experiences55; Trier Inventory for Chronic Stress (TICS)56; Psychiatric Epidemiology Research Interview Life Events Scale57; and Individual and Structural Exposure to Stress in Psychosis-risk-states Interview58.

Lastly, the concepts of social withdrawal and interpersonal sensitivity were measured using the following scales: Premorbid Adjustment Scale59; Social Interaction Scale60; Social Functioning Scale61; and the Interpersonal Sensitivity Measure26.

Risk of bias and certainty assessment

The risk of bias assessment is summarized in Table 1 for cross-sectional studies and Table 2 for longitudinal studies. The certainty assessment was conducted using the GRADE criteria and deemed the effects and conclusions of this systematic review as moderate. The moderate score was given due to the high consistency, precision, and directness found in the included studies. The potential for risk of bias was due to lack of blinding and the possible effects of confounding variables, which prohibited granting a higher-ranking score.

Table 1 The NOS assessment—cross-sectional studies.
Full size table
Table 2 The NOS assessment—longitudinal studies.
Full size table

Effects of psychosocial stress, social withdrawal, and interpersonal sensitivity on psychosis risk

From the included studies, 16, 8, and 5 studies examined the effect of psychosocial stress, social withdrawal, and interpersonal sensitivity on risk of psychosis in individuals at CHR, respectively (see Tables 3–8).

Table 3 Psychosocial stress: characteristics of studies meeting inclusion criteria (n = 16).
Full size table
Table 4 Social withdrawal: characteristics of studies meeting inclusion criteria (n = 8).
Full size table
Table 5 Interpersonal sensitivity: characteristics of studies meeting inclusion criteria (n = 5).
Full size table
Table 6 Findings of individual studies—psychosocial stress.
Full size table
Table 7 Findings of individual studies—social withdrawal.
Full size table
Table 8 Findings of individual studies—interpersonal sensitivity.
Full size table

Psychosocial stress and psychosis risk

From the total of 16 studies, 13 studies reported higher levels of psychosocial stress in the CHR group compared to controls, while 3 studies found no difference between these groups (see Tables 3 and 6). The significant association was mainly driven by the presence of trauma (10 out of 13 studies), while the non-significant associations defined psychosocial stress as significant life events (2 studies) or as daily hassles (1 study).

All ten studies examining trauma noted significantly higher levels in CHR individuals compared to controls. Similarly, the two studies examining general psychosocial stress also demonstrated significant associations with CHR individuals compared to controls. However, the two studies investigating significant life events found no difference between CHR and control groups.

Nine studies measured the association between the psychosocial stress and symptom severity. Five studies found increased symptomatology (positive and/or negative) in CHR individuals who experienced greater rather than lower levels of psychosocial stress, and 4 studies found no relationship between psychosocial stress and symptom severity. Seven studies examined risk of transition, and 3 studies found an association between elevated psychosocial stress and increased risk of transition to psychosis21,62,63. Greater exposure to life events and distress associated with these events were found in CHR individuals who transitioned to psychosis compared to those that did not21, while emotional abuse increased the risk of transition to psychosis 3.8 fold in CHR individuals63. Perceived discrimination also increased risk of transition in CHR individuals by 52.4% for every unit increase in scores on lifetime perceived discrimination62.

Social withdrawal and psychosis risk

Eight studies examined social withdrawal in CHR individuals (see Tables 4 and 7). All but one study64 reported higher levels of social withdrawal in individuals at CHR as compared to controls. One of the studies examined social withdrawal prior to CHR identification65 while the others measured its presence in individuals deemed to meet CHR criteria. Four studies examined the association between social withdrawal and psychosis symptomatology, two of which found a positive correlation, while the other two found no association between the levels of social withdrawal and the worsening of psychosis symptoms in CHR individuals. Two studies found higher levels of social withdrawal and subsequent conversion to psychosis.

Interpersonal sensitivity and psychosis risk

Five studies examined interpersonal sensitivity in CHR individuals (see Tables 5 and 8). All five studies found significantly higher levels of interpersonal sensitivity in CHR individuals compared to controls. Only two studies examined interpersonal sensitivity in relation to psychosis symptomatology and a positive correlation was found for both positive and negative symptoms. Lastly, no study included in this review measured the effect of interpersonal sensitivity on transition risk.

Bio-psychosocial model of transition to psychosis

The biological and neurodevelopmental vulnerabilities that play a key role in the stress-vulnerability model include hyperactivation of the hypothalamic–pituitary–adrenal (HPA) axis, dysregulation of neurotransmitters such as dopamine, GABA, and glutamate, aberrant salience, increased stress sensitivity and emotional reactivity to stressors, epigenetic effects (modification to the genome that affect gene expression without altering the DNA sequence), and gene-environment interactions (genetic factors influence the impact of an environmental exposure on an individual)66,67,68. Psychological factors include core beliefs and appraisals, emotions, attachment style, social cognition (set of neurocognitive processes related to understanding, recognizing, processing, and appropriately using social stimuli in one’s environment), and theory of mind deficits (the capacity to infer one’s own and other persons’ mental states)69,70,71,72,73; while social factors include the experience of trauma (physical, sexual, and emotional abuse, physical and emotional neglect, and parental loss), life events, and daily hassles74,75,76.

Based on the findings of the present paper, we propose the following Bio-psychosocial Model of Transition to Psychosis (see Fig. 2). This is a conceptual framework for integrating the influence of biological, psychological, and social factors in the transition to psychosis. Bio-psychosocial vulnerabilities, coupled with trauma/significant life events, lead to the formation of core beliefs, which are re-activated in response to psychosocial stressors. This core belief reactivation, in addition to the influence of interpersonal sensitivity, appraisals of stress, and social withdrawal, may give rise to increased affective arousal, which may be further amplified by the influence of cognitive biases and maladaptive coping. This affective pathway contributes to the generation of anomalous experiences/sub-threshold psychosis symptoms and the subsequent search for meaning for these unusual experiences, which may then lead to the formation of manifest psychosis. This model suggests that an underlying bio-psychosocial vulnerability coupled with exposure to trauma/significant life events during one’s early life leads to the formation of core beliefs about oneself, others, and the world (e.g., I’m worthless, others are untrustworthy, world is dangerous)71. Indeed, compared to healthy controls, CHR individuals reported significantly more negative beliefs about self and others and significantly less positive beliefs about self and others77. Moreover, negative core beliefs have been found to partially mediate the relationship between childhood trauma and persecutory beliefs and have been shown to be characteristic of patients with psychosis78,79,80,81. These authors concluded that these findings provide preliminary evidence about the cognitive mechanisms that may underlie the association between childhood trauma and later risk for psychosis.

Fig. 2: Bio-psychosocial model of transition to psychosis.
figure 2

A conceptual framework for integrating the influence of biological, psychological, and social factors in the transition to psychosis.

Full size image

Early experiences shape our attachment style and can lead to an interpersonally sensitive style of relating, which would make the individual hypervigilant to threat and hypersensitive to ambiguity and perceived rejection/harm from others71,82,83. Social withdrawal would remove opportunities to receive positive reinforcement/gathering corrective information, which would in turn foster ruminative/erroneous thinking, which would negatively impact one’s mood84. Psychosocial stress (trauma/significant life events/daily hassles) would drive subsequent appraisals of such events, such as “I am a failure/inferior/unlovable/vulnerable/weak” inducing dysregulation of the HPA axis and the dopaminergic system. Appraisals of stress may also influence social withdrawal and interpersonal sensitivity, which both may in turn further reinforce the appraisal of stress. For example, an interpersonal sensitive style of relating may misinterpret the actions of others as threatening/rejecting while ensuing social avoidance would preclude the receipt of disconfirmatory evidence, thereby maintaining the appraisal of stress.

In addition, if the psychosocial stress experienced later in life is thematically similar to the trauma/significant life events during one’s early life, it may serve to reactivate pre-existing core beliefs, further inducing a HPA stress response, elevated affective reactivity, and heightened stress sensitivity. Increased affective arousal/emotional dysregulation, in addition to the influence of cognitive biases, and maladaptive coping (e.g., rumination, substance misuse, emotion-oriented coping) may subsequently contribute to the formation of anomalous experiences as typified in the CHR state. One’s attempt to make sense of their experiences, otherwise known as their search for meaning, may then lead to the formation of manifest psychosis, such as delusions and hallucinations85. The search for meaning refers to the search for an explanation for anomalous experiences, recent events, or arousal and in doing so; pre-existing beliefs about self, world, and others are drawn upon85. These positive symptoms of psychosis in turn may reflect exaggerated themes of pre-existing insecurities and/or interpersonal concerns (e.g., repeated historical experiences of childhood physical abuse is likely to lead to the development of a vulnerability core belief, which would be reactivated by a critical incident concerning physical threat from others such as a mugging). This may give rise to persecutory delusions of threat of harm from others and congruent auditory hallucinations (e.g., voice saying “you’re in danger”, “don’t go outside”). Therefore, comprehensive assessments exploring the client’s early life experiences/significant life events, core beliefs, psychosocial stressors and associated theme (e.g., intrusion, harm, loss of control, helplessness, unlovability, worthlessness, weakness, inferiority), appraisals of stress, interpersonal sensitivity, social withdrawal, cognitive biases, maladaptive coping, affective responses, and emerging anomalous experiences/psychosis presentation (delusional and/or hallucinatory subtype) would aid in the development of personalised formulations and targeted interventions.

Discussion

This systematic review sought to synthesize the literature examining the relationship between psychosocial stress and onset of psychosis in CHR individuals. Psychosocial stress, interpersonal sensitivity, and social withdrawal were higher in CHR individuals compared to healthy controls and there was some evidence of their association with positive symptoms of psychosis. There was also some evidence to support the role of psychosocial stress and social withdrawal in the transition to psychosis and no studies to date have examined the role of interpersonal sensitivity on transition to psychosis in CHR, which proposes avenues for future research.

While the present review focuses on the existing evidence of the relationship between psychosocial stress and risk of psychosis in individuals at Clinical High Risk (CHR) for psychosis, our companion review (Almuqrin et al. submitted) considers such evidence in individuals with a First Episode of Psychosis (FEP).

Psychosocial stress

A repeated finding of the present review was for the association between higher levels of psychosocial stress in CHR individuals compared to controls. It is important to note that most studies driving this association examined trauma rather than significant life events or daily hassles, each of which are conceptually distinct constructs that need to be differentially defined and studied in relation to CHR symptomatology and conversion to psychosis. Psychosocial stress has been associated with increased positive and negative symptoms in individuals at CHR and appears to occur in a dose-response manner, as the greater the level of psychosocial stress experienced, the greater its impact on symptom severity22. This is in line with the dysregulation of the HPA Axis and the stress-vulnerability model regarding psychosis symptomatology arising from the cumulative effects of stress on a pre-existing bio-psychosocial vulnerability13. Interestingly, a lab-based study examining the effects of an experimentally induced psychosocial stressor found that CHR individuals produced higher overall cortisol levels from the pre-anticipation period through to the recovery period of the Trier Social Stress Test and exhibited higher levels of subjective stress prior to the stressor compared to controls86. These findings provide further support for the stress-vulnerability model and highlight the importance of examining the biological and subjective impact of psychosocial stress in CHR to further elucidate possible mechanisms of transition to psychosis. Environmental risks have also been found to act additively and synergistically with childhood trauma87,88,89 and stressful life events90,91, contributing to the persistence of sub-threshold psychosis symptoms in the general population. Psychosocial stressors such as childhood trauma and stressful life events have further been associated with higher levels of positive, negative, and depressive symptoms in the general population92,93. In addition, these psychosocial stressors, along with polygenic risk scores were found to exhibit independent additive effects on these three dimensions of subclinical psychosis92,93 further supporting the stress-vulnerability hypothesis. However, it is important to note that no gene-environment interaction was found93. These general population studies appear to parallel the CHR findings22 and are consistent with the stress-vulnerability and stress sensitivity models, whereby the cumulative effects of stress increase the likelihood of psychosis expression and increased affective arousal in response to such stressors, respectively. However, the finding of a positive correlation between psychosocial stress and positive psychosis symptoms in both CHR and non-CHR help-seeking controls22 suggests that the effects of psychosocial stress on psychosis symptomatology appears to be independent of clinical vulnerability to psychosis, thereby only providing partial support for the stress-vulnerability model. A bi-directional relationship may also occur in that psychosis symptomatology may also elicit unwanted interpersonal responses leading to further experiences of psychosocial stress22. It has also been found that compared to healthy controls, individuals at CHR are significantly more distressed by stressful events and that the appraisal of these events differentiated CHR individuals from controls21,25. However, it is important to note that to date, the majority of case-control studies examining CHR have utilized healthy non-clinical controls, as opposed to help-seeking (i.e., psychiatric) controls, which risks attributing group differences to psychosis-risk status rather than to non-specific psychopathology and comorbidities occurring in the CHR group94. The inclusion of both healthy controls and help-seeking samples in CHR would aid in elucidating a psychosis specific vulnerability, as opposed to a more general mental illness vulnerability94.

Psychosocial stress and psychosis onset

Only three out of seven studies found a significant association between elevated psychosocial stress and increased risk of transition to psychosis21,62,63. Greater exposure to psychosocial stress, emotional abuse, and perceived discrimination were found to significantly increase the risk of transition to psychosis in CHR individuals compared to controls. It is important to note that the three studies that demonstrated significant associations with transition to psychosis had a 24-month follow-up period and sample sizes ranging from 259 to 764 CHR individuals and 162 to 280 controls, which contrasts the 35–105 CHR individuals and 24–28 controls in the non-significant studies and a shorter 12-month follow-up period. Therefore, the lack of significant findings regarding psychosocial stress and transition to psychosis in CHR may be due to existing studies being statistically underpowered to detect such effects and not having a long enough follow-up period to capture cases of transition.

Interpersonal sensitivity

All five studies found significantly higher levels of interpersonal sensitivity in CHR individuals compared to controls. Higher levels of interpersonal sensitivity exhibited by CHR individuals compared to controls is congruent with the theory that certain personality characteristics may predispose individuals to mental illness, such as psychosis95. Indeed, interpersonal sensitivity has been associated with a greater severity of psychosis symptomatology. This finding implicates the important role of social interactions as a factor in influencing one’s well-being. Moreover, this relationship may indeed be bi-directional as individuals interact with their environment, which in turn impacts on the individual66. High interpersonal sensitivity can give rise to rumination about social performance and speech, preoccupation with the emotions displayed, and excessive focus on other people’s opinions, which may hinder social performance. Feelings of exclusion and perceptions concerning a lack of understanding from others can negatively impact on self-esteem, confidence, and motivation, which could exacerbate negative symptoms such as negative self-image, asociality, and avolition96.

In respect to positive symptoms of psychosis, increased levels of distress and subsequent avoidance of social situations could possibly account for their association with interpersonal sensitivity. The increased levels of distress could be attributed to one’s preoccupation with receiving negative social feedback and perceptions of a lack of approval from others. Preoccupation with social feedback would elicit both physiological and psychological reactions, thereby increasing levels of stress97. In regard to coping with stress, it has been found that CHR individuals reported feeling significantly more distressed by events, felt that they coped more poorly, and employed more emotion-oriented coping as opposed to task-focused coping25. These authors also found that compared to controls, CHR individuals were less likely to employ social diversion as a means of coping, which involves engaging with others to divert attention from stressors.

Even though the evidence from this review supports the concept of higher levels of interpersonal sensitivity leading to worsening symptomatology in CHR, it is useful to note that only two studies examined this association and further investigations are needed to examine the strength of this association.

No study to date has examined the relationship between interpersonal sensitivity and psychosis transition. The findings of increased severity of psychosis symptomatology in relation to interpersonal sensitivity might suggest the possibility of a direct link with conversion to psychosis. Nevertheless, it could also be possible that symptoms are affected by this personality trait but it may not extend to conversion to psychosis, as was the case for social withdrawal. Henceforth, without a number of studies investigating this aforementioned relationship, no conclusion can be drawn at this time.

Social withdrawal

Compared to controls, social withdrawal was frequently observed in CHR individuals, who might avoid social situations due to emerging suspiciousness98. Moreover, auditory and/or visual hallucinations may directly impact the individual’s capacity to engage and follow a conversation, which may be distressing and may negatively impact self-esteem. Individuals may thus choose to avoid social situations to circumvent anticipated embarrassment and rejection from others98. Social avoidance/withdrawal may ensue failed attempts at social engagement or may be present pre-morbidly99. Unfavorable feedback from social interactions can also lead to social avoidance so as to avoid unpleasant feelings100. Social withdrawal results in a lack of social support, which in turn minimizes potential sources of external support that could challenge delusions and hallucinations, thus indirectly causing a greater reality mismatch101.

The mixed findings regarding the association between social withdrawal and increased symptomatology might be accounted for by the inclusion of varied outcome measures for social functioning. Interestingly, the two negative findings employed the Social Functioning Scale (SFS)61, which was validated on an outpatient schizophrenia sample with a mean illness duration of 8.8 years and assesses seven areas including social engagement/withdrawal, interpersonal behavior, prosocial activities, recreation, independence-competence, independence–performance, and employment/occupation. Of the two studies that found a positive association between social withdrawal and symptoms, one employed the Premorbid Adjustment Scale (PAS)59, which measures the level of functioning in four major areas: social accessibility-isolation, peer relationships, ability to function outside the nuclear family, and capacity to form intimate socio-sexual ties. The PAS may therefore be more nuanced and applicable to the CHR population compared to the SFS, possibly accounting for the significant positive finding. The other significant finding employed the Experience Sampling Method (ESM)102, which captures daily life data regarding social context and frequency, as well as emotional reactivity throughout the day for seven consecutive days. This is advantageous to questionnaire measures, as it allows for the investigation of experiences and interactions within a real-world context103. The mixed findings regarding the association between social withdrawal and increased symptoms of psychosis might also be accounted for by the included studies tending to examine positive rather than negative symptoms, highlighting the presence of a possible publication bias. It has been suggested that social withdrawal is more closely related to the latter rather than the former. Indeed significant associations have been found for social withdrawal and negative symptoms in CHR individuals104. The closer association between social withdrawal and negative symptoms may be due to the similarities in its characteristics. Social withdrawal is presumed to result from the lack of motivation to engage in social interactions, asociality, which is one of the primary negative symptoms105. Henceforth, the results might have been different if the association between social withdrawal and negative symptoms was assessed in the included studies.

The perception of social support may benefit reality testing, as it can weaken the intensity of delusions and hallucinations, subsequently improving insight in CHR101. If delusions and hallucinations remain unchallenged, they may intensify and exacerbate distress. It has also been posited that the lack of sensory stimulation derived from social interactions may contribute to an increase in hallucinatory experiences owing to the over-compensatory mechanisms of the nervous system106. In addition, socially withdrawn CHR individuals may not recognize their emerging illness and need for treatment, which would adversely affect their recovery107.

Two studies found higher levels of social withdrawal and subsequent transition to psychosis. The higher transition rates in CHR individuals who exhibit higher levels of social withdrawal can possibly be explained by a social network approach. Having social support can greatly impact one’s well-being in number of positive ways such as increasing self-esteem, improving confidence, providing an opportunity for new experiences, easing stress, and preventing loneliness. Alternatively, the lack of social networks can lead to social isolation, poor psychosocial functioning, and an increase in negative thoughts and feelings108. In addition, it has been suggested that higher transition rates are linked to a longer duration of untreated psychosis, as a result of a lack of social support. Having a close friend or supportive family environment can facilitate the more timely identification of behavioral change and mental health deterioration, which could then precipitate prompt engagement with services and the initiation of treatment109.

Strengths and limitations

This is the first review to synthesize the literature regarding psychosocial stress, social withdrawal, and interpersonal sensitivity in CHR and highlights the importance of social factors in CHR symptomatology and their possible involvement in conversion to psychosis. However, this review has some limitations. The majority of studies included in the present review had a higher number of male than female participants. Poorer premorbid and psychosocial functioning, and increased substance misuse has been found in males compared to females, which would negatively impact on their ability to form social connections and to seek help from others110. Substance misuse may also negatively impact symptom severity, functioning, engagement with services, and recovery111,112. The included sample was predominantly Caucasian, which limits the representativeness and generalizability of the sample to Black and Minority groups, which have a higher incidence of psychosis113. Alongside the characteristics of the sample, the size of it could also have an impact on the results. Even though the average sample size was 100, 9 out of 27 studies had sample sizes under 100 participants. Small sample sizes can affect the reliability and representativeness of results. Additionally, the majority of the studies included in this review originated from Western countries, while a small number of studies were conducted in developing countries. This therefore limits the generalizability of the present findings114.

Furthermore, the present review examined three different types of psychosocial stress (trauma, significant life events, and daily hassles), however some studies did not specify the social stressor but referred to it by the collective name of social/psychosocial stress. Using this umbrella term limits the possibility of examining which factors influence CHR status and transition outcomes. Trauma, significant life events, and daily hassles represent conceptually distinct constructs that need to be differentially defined and studied in relation to CHR symptomatology and their role in transition. Therefore, specifying the type of stress would enhance the specificity of future investigations. In addition, it would be important to examine the effect of more recent and historical significant life events, as well as the appraisal of those events, as they may exert a differential influence on symptom formation and expression. Indeed, appraisals of stress have been found to differ between CHR individuals and controls and may impact on psychosis transition25.

Clinical implications and future directions

This review synthesized the literature examining psychosocial stress, social withdrawal, and interpersonal sensitivity in individuals at CHR for psychosis. The negative impact of psychosocial stress on CHR individuals in terms of psychosis symptomatology and transition to psychosis emphasizes the importance of social factors in the CHR state. The proposed Bio-Psychosocial Model of Transition to Psychosis offers an explanatory framework for devising personalized, idiosyncratic, and symptom-specific formulations accounting for psychosis emergence in CHR and FEP. It highlights the possible mediating role of core beliefs in explaining the relationship between trauma and psychosis symptomatology, which would benefit from further investigation of core beliefs in CHR, FEP, and controls and their association with hallucinatory and delusion subtypes (e.g. persecutory, grandiose, religious, somatic, bizarre). This would subsequently inform psychosocial treatments, allowing for targeted interventions. For example, psychosocial treatments, such as Cognitive Behavior Therapy for Psychosis (CBTp) could employ this Bio-Psychosocial Model of Transition to Psychosis within collaborative case formulations with clients and could draw upon social aspects of treatment, such as social skills training, encouraging help-seeking behaviors, utilizing social support, and enhancing communication skills. This would also help to avoid the negative impact of social withdrawal. Additionally, Group CBTp could be employed to target social isolation and withdrawal, while also enhancing social skills in a group setting. Current interventions advocated for CHR include CBTp for alleviating symptoms and preventing transition to psychosis, as well as Cognitive Behavioral Family Interventions for Psychosis (CBFIP) in order to reduce stress, improve problem-solving, and enhance interpersonal communication skills115. However, neither intervention explicitly targets the appraisal of psychosocial stress and associated coping, which could improve social functioning in CHR individuals. This therefore affords an opportunity to devise and develop targeted CBTp interventions with an explicit focus on minimizing the effects psychosocial stress and improving social functioning. Additionally, as psychosocial stress is often unavoidable, learning to identify and challenge unhelpful appraisals and maladaptive coping leading to increased distress may enhance well-being. To date, there has been a lack of emphasis on the appraisals of stressful events, therefore their exploration in CHR, FEP, and controls in future studies are warranted. Interpersonal sensitivity could also be a therapeutic target, with the aim of elucidating vicious cycles and enhancing more adaptive behavioral coping methods of social integration rather than social avoidance. Therefore, identifying potential cognitive and affective mediators accounting for the relationship between psychosocial stress and psychosis could hold significant implications for the identification, prevention, and treatment of CHR individuals.