Sex differences in clinical presentation in youth at high risk for psychosis who transition to psychosis

have been observed in individuals with schizophrenia and for those at clinical high risk (CHR) for psychosis. However, specific differences in CHR individuals who transition to psychosis remain inconsistent and understudied. This study aimed to investigate sex differences in 156 CHR individuals who made the transition to psychosis. A wide range of demographics, positive and negative symptoms, depression, anxiety, social and role functioning, trauma, and substance use were assessed at baseline and symptoms and diagnoses at the time of transition. Fluctuations in positive and negative symptoms and different medications were also assessed. No sex differences were observed at baseline for those who later transitioned to psychosis. At transition, males were significantly more likely to be diagnosed as having schizophrenia or schizophreniform disorder and through the course of the study, males were more likely to be taking stimulants. Limitations in this study was the lack of longitudinal follow-up post transition. The study highlights the need for further research on sex differences in individuals who transition to psychosis. Understanding these differences can have implications for treatment and monitoring of CHR individuals.


Introduction
There is a long history of reports on sex and gender differences in the clinical expression and outcome of schizophrenia spectrum disorders (SSDs) (Castle et al., 1995;Seeman, 1982Seeman, , 2019;;Sommer et al., 2020).In terms of definitions, sex relates to the biological and physiological characteristics (e.g., chromosomes, genes, physiology, genetics, hormone levels), and physical factors (e.g., reproductive anatomy) categorized as female or male.Whereas, gender is used to describe the socially constructed characteristics associated with being feminine or masculine, that is, how one perceives themselves in the context of society's traditional roles of girls/boys and women/men (Riecher-Rössler, 2017).The current paper focuses on sex differences.Although gender was not studied here, it may play an important role and deserves attention in future research.
Commonly cited sex and gender clinical differences are that males have more severe negative symptoms, poorer premorbid and current social functioning, (Cotton et al., 2009;Thorup et al., 2014) increased substance use, including cannabis, (Cotton et al., 2009;Thorup et al., 2014) a younger age of onset (Faraone et al., 1994) and overall, a more chronic course with poorer outcomes.Females are more likely to present with mood disturbance and depressive symptoms (Cotton et al., 2009).Thus, with a major focus in the current literature on young people who are at clinical high-risk (CHR) of developing psychosis, we can understand prospectively how sex differences present for CHR individuals who make the transition to psychosis and whether they fit with those reported in the literature for individuals with an SSD.
Examination of sex differences in CHR has focused on individuals at their initial study assessments.Little has been reported on sex differences among CHR individuals who later transition to psychosis.However, Rosen et al. (2020) reported that in their sample of 242 referred CHR participants there was a sub-sample of 84 individuals who did transition.Rosen et al. (2020) noted some small sex effects in those who transitioned, in that females had more pronounced positive symptoms (perceptual abnormalities, bizarre thinking and odd behaviours), whereas males had more pronounced ratings on some negative symptoms.
Thus, there are some clear sex differences for those who suffer from SSDs.However, to the best of our knowledge, little is known about sex differences in those who have been prospectively followed and made the transition to a full-blown psychotic disorder.Thus, the goal of this project is to investigate sex differences in specific demographic, clinical, and functional characteristics for those who made the transition to psychosis and to determine how they compare to those reported in the literature for individuals with psychosis.A key aim for this project is that sex differences will be examined at a specific time, which is at the point of transition, before major influences such as substance abuse, stigma, difficulties with social functioning and with employment may have an impact.Thus, offering a unique picture of sex differences at this time of transition.

Participants
All participants described in this paper were recruited as part of two of the projects of the multisite North American Prodrome Longitudinal Study (NAPLS-2 and NAPLS-3).NAPLS is a consortium of nine CHR research programs across North America (Yale University, University of California at Los Angeles, San Francisco, and San Diego, Harvard University, University of Calgary, University of North Carolina, Emory University and Zucker Hillside Hospital).The core methods, including study design, study population, ascertainment, and description of assessments for NAPLS-2 and -3 have previously been published in detail (Addington et al., 2012;Addington et al., 2022).CHR participants had to be between 12 and 30 years old and meet diagnostic criteria for a psychosis-risk syndrome as per the Criteria of Psychosis-Risk Syndromes (COPS) using the Structured Interview for Psychosis-Risk Syndromes (SIPS) (McGlashan et al., 2010).Exclusion criteria included (i) meeting criteria for current or lifetime Axis I psychotic disorder, including affective psychoses, (ii) IQ <70, (iii) history of a central nervous system disorder, or (iv) diagnostic psychosis-risk symptoms that were clearly caused by an Axis 1 disorder.Other non-psychotic DSM-IV or DSM-5 disorders were not exclusionary (e.g., substance-related disorders, major depression, anxiety disorders, personality disorders), provided the disorder did not account for the individual's psychosis-risk symptoms.
NAPLS-2 consisted of 764 CHR participants and NAPLS-3 of 710 CHR participants.Combining the samples there are 1318 participants who did not transition to psychosis and 156 who did within the 2-year duration of the studies.Thus, in the combined sample approximately 11 % transitioned to psychosis.Transition to psychosis was defined as a) the Presence of Psychosis Syndrome (POPS) criteria (i.e., any of the five Scale of Psychosis-Risk Symptoms [SOPS] where attenuated positive symptoms (APS) reached a 6 in intensity (where 0 indicates absence of symptom, and 6 is maximum severity) for a minimum of one hour per day, four days per week, for a minimum of 1 month; or b) one's functioning is severely impacted by APS, e.g.severe disorganization and/or presenting a danger to oneself or others (McGlashan et al., 2010).This paper will focus on the 156 CHR individuals (66 females, 90 males) that transitioned to psychosis.Females accounted for 42.3 % of those who did transition.

Measures
A range of demographic details were collected at baseline that included sex at birth (i.e., female, or male), age, race, immigration status, marital status, living arrangements, level of education for participant, mother and father, and income source and category.
The SIPS was used to determine if participants met criteria for being at CHR for psychosis, and the Scale of Psychosis-risk Symptoms (SOPS) was used to measure attenuated psychotic and negative symptoms at three timepoints -baseline, over time until point of transition, and at transition (McGlashan et al., 2010).Depression was measured using the Calgary Depression Scale for Schizophrenia (CDSS) (Addington et al., 1993;Addington et al., 2014).Suicide attempts were counted using ratings on the CDSS suicide item 8.These were calculated as a suicide attempt between entering the study and transition.Functioning was assessed with Global Functioning: Social and Role (Cornblatt et al., 2007) and the Premorbid Adjustment Scale (Cannon-Spoor et al., 1982).
Traumatic experiences before the age of 18 years were assessed at baseline using an adapted version of the Childhood Trauma and Abuse Scale (Janssen et al., 2004).Trauma was categorized into the following four domains: psychological abuse, physical abuse, sexual abuse, and emotional neglect.Each trauma type was scored as reported or not reported.Bullying included psychological and/or physical bullying.
Current substance use was determined with the Alcohol/Drug Use Scale (AUS/DUS) (Drake et al., 1996).Since cannabis, alcohol, and tobacco are reported as the most used substances in both NAPLS sampleswhile other illicit substance use, including hallucinogens, opioids, sedatives, stimulants, amphetamines, and ecstasy are negligible (Buchy et al., 2015;Santesteban-Echarri et al., 2022), only the use of alcohol, tobacco and cannabis were examined in this paper.
Use of stimulants (for attention deficit hyperactivity disorder (ADHD)), antidepressants and antipsychotics was determined by whether they were taking the medication (i) at baseline or prior to baseline; (ii) at baseline and at subsequent assessments prior to transition or (iii) sometime after baseline until transition.
In NAPLS-2 the structured clinical interview for DSM-IV (First et al., 1998) and in NAPLS-3 the Structured Clinical Interview for DSM-5 (SCID-5) were conducted at baseline, and at the transition assessment.Diagnoses that were assessed included schizophrenia spectrum and other disorders, substance use disorders, mood disorders, bipolar disorder, and anxiety disorders.A composite anxiety disorder variable was created including any SCID-rated diagnosis of panic disorder, agoraphobia, social anxiety disorder, specific phobia, or generalized anxiety disorder because of the small sample sizes within each individual anxiety disorder.

Procedures
The NAPLS-2 and -3 studies were approved by the Institutional Review Boards of all nine participating sites.Informed consent was obtained from all participants, including parental consent for all minors.
For all CHR participants, symptom ratings and psychosis-risk diagnoses were compiled with study eligibility to construct a detailed vignette for consensus diagnosis.Weekly consensus calls, chaired by J. Addington and T. McGlashan, were conducted with clinical raters at all nine sites to determine diagnosis.This process was repeated should a participant be considered to have made the transition to psychosis.A similar vignette outlining the increase in symptoms was submitted and reviewed on the consensus call to determine a consensus on transition to psychosis.

Data analysis
Statistical analyses were performed using the IBM Statistical Package for the Social Sciences (SPSS) version 25 and Statistical Analysis Software (SAS) version 9.4.
Descriptive statistics were used to report baseline demographics, clinical and functioning characteristics, and medications.Continuous data are presented as mean (M) and standard deviation (SD), and categorical data as counts and percentages.Sex differences in demographic and clinical characteristics were compared using Chi-square (χ2) tests or Fisher.
Exact tests for categorical data, and t-test or Mann-Whitney for continuous data, were used as relevant.Since multiple comparisons were conducted to determine which variables are likely to have a significant effect in detail, we applied a Bonferroni correction.
Time to psychosis transition was evaluated using the Kaplan Meier method and compared using the Log-rank test.Multivariable Cox proportional hazards models were used to compare time to transition between the females and males.The proportional hazards assumption was checked using Schoenfeld residuals.
To examine if the males and females differed over time on SOPS positive and negative symptoms, we used a generalized linear model (GLM) to accommodate missing data and account for intra-participant correlation over time.

Results
The sample consisted of 156 CHR individuals (66 females, 90 males) that transitioned to psychosis.Females accounted for 42.3 % of those who did transition.There were no differences in the percentage of females relative to males.
Table 1 presents baseline demographics.There were no significant sex differences for any demographic variable.
Table 2 presents clinical and functioning variables.There were no significant differences between males and females on any of these variables.In terms of suicide attempts, two females each made one attempt during the study.
Baseline DSM diagnoses are presented in Table 3.There were no sex differences in baseline diagnoses.
Medication usage is presented in Table 4. Participants who had been prescribed stimulants were on these medications for the duration of the study.Significantly more males were using stimulants for ADHD.Antidepressants and antipsychotics were often used at various times in the study.There were no significant sex differences in use of antidepressant and/or antipsychotics.
Fluctuations in symptoms were examined.Table 5 presents the changes over time in positive and negative symptoms for males and females.Interestingly, males showed improvement in positive symptoms over time.There were no significant changes over time for males or females in negative symptoms and no differences between males and females for changes in positive or negative symptoms.
Males and females were compared on clinical variables at the time of their transition to psychosis.There were no differences in attenuated psychotic symptoms, negative symptoms, or social or role functioning at the time of transition.See Table 6.
Table 7 presents DSM-diagnoses at the time of transition.Specific diagnoses are presented first.Males were significantly more often give a  diagnosis of schizophrenia or schizophreniform disorder (χ2 = 8.97, p < 0.01).The other diagnostic groups were too small to compare.The group was divided into affective and non-affective psychotic diagnoses.The majority presented with a non-affective psychosis and there were no sex differences for either transition diagnosis.There were no sex differences in comorbid diagnoses at the time of transition.Significant sex differences were observed for days from baseline assessment to time of transition to psychosis.Overall, the mean number of days from baseline assessment to psychosis transition was 228.8 days (SD = 198.5].Females transitioned to psychosis later than males (M days ± SD, females vs. males, 267.6 ± 222.3 vs. 200.3± 175.0, respectively; p = 0.036).Unadjusted Kaplan Meier survival curves for transition to psychosis are shown in Fig. 1 (log-rank test p = 0.034).In the model, female CHR participants had significantly slower transition to psychosis compared to males (model, hazard ratio [HR] 0.70 [95%CI, 0.51-0.97],p = 0.034).

Discussion
This study investigated sex differences in a large cohort of CHR individuals from two studies combined, in which 156 (approximately 11 %) developed a psychotic illness within two years.Overall, few clinically relevant sex differences were observed.Furthermore, there were no sex differences at the time of transition that might have been expected based on knowledge of sex differences in the SSD population.Expectations might have been that females would have been older and that there were clinical differences favouring females.
An examination of baseline demographics, clinical ratings, and diagnoses for those who did develop psychosis revealed no significant differences between males and females.Females were, on average, one year younger than males at baseline but this was not significant.The only clinical difference at the time of transition was that males were significantly more likely to have a diagnosis of schizophrenia or schizophreniform compared to females.The majority of those who made the transition to psychosis had a diagnosis of non-affective psychosis with again there being no sex differences.Thus, we are not seeing evidence of increased affective disorders for females even when examining comorbid diagnoses.Finally, it was observed that females did have a significantly longer time until transition.
In the adult schizophrenia population, women are known to develop psychosis later, with peak incidence in their late twenties (Seeman, 1982).In this CHR cohort, the similarity in ages at time of transition gives rise to questions about course and trajectory of illness.With early age of onset for psychosis being linked to poorer outcome (Clemmensen et al., 2012;Iasevoli et al., 2022;Lally et al., 2016;Meltzer et al., 1997), the comparable ages and the lack of symptomatic differences observed here between females and males raises the possibility that CHR females participating in NAPLS and who have presented for help at the prepsychotic stage may be at risk for poorer outcome.
However, there were no significant sex differences in comorbid diagnosis of depression or anxiety at the time of transition with rates of depression and anxiety remaining high for both males and females.Surprisingly, there were no sex differences in affective versus nonaffective psychosis.Although it is possible with longer term follow-up the large group of individuals, approximately one-third, with a psychosis NOS disorder may evolve to diagnoses that differ between the  sexes.In one first episode study, a small group presented with another psychotic disorder (OPD) versus an SSD (Chaves et al., 2006).In this sub-group it was observed that one year after the initial presentation men who initially presented with an OPD mainly shifted towards a schizophrenia diagnosis while women shifted away from a OPD towards a mood disorder (Chaves et al., 2006).In summary, the only notable sex differences for those who transitioned was that males were more likely to have a schizophrenia or schizophreniform diagnosis at the time of transition, and that there was a slightly increased time to transition for females.This trend does fit with what is observed in the literature for those with psychosis that females develop psychosis later than males.It may be expected that males who are receiving a diagnosis of schizophrenia/schizophreniform at the time of transition are more likely to develop poorer functioning.However, this difference was not significant in this study after applying the Bonferroni correction.On the other hand, if these results are not a trend of what has been noted for psychotic patients in the literature it may be that the lack of differences suggests that CHR clinics and research studies are recruiting females who have an early onset and may have a poorer outcome that does not really differ from age-matched males.
The strengths of this project are that first by combining NAPLS-2 and -3 we were able to examine sex differences in a large sample of those that did transition to psychosis.Secondly, we examined sex differences at a specific time period, that is at the point of transition, before major influences such as substance abuse, stigma, difficulties with social functioning and with employment may have an impact.Thus, offering a unique picture of sex differences at this time of transition.
There are several limitations.Compared to other studies we do report on a considerable number of transitions, but it is still a small sample.In CHR samples where a small proportion of participants make the transition to psychosis, it is possible that the transition sample is underpowered.Thus, results in this paper need to be interpreted with caution.Secondly, time to transition is calculated from the baseline assessment  until the time of transition.However, baseline does not mark the onset of the psychosis-risk symptoms, only that they began or worsened in the past year prior to ascertainment.Thus, it is difficult to truly assess if females transition more slowly after onset of symptoms.A third limitation is that there was no significant follow-up of participants once they made the transition to psychosis.In NAPLS-2 and -3, transition to psychosis meant that their study participation ended, and although there was an intent to follow those who transitioned for one-year post-transition the number followed was insignificant.Many were lost to followup and since NAPLS-2 and -3 were 2-year studies, the studies ended before many who transitioned were eligible for a one-year post transition follow-up.Since one-third of those transitioning had a psychosis NOS diagnosis at transition it would be important to know what a subsequent diagnosis might be.For example, as described above, in first episode psychosis studies we see noticeable changes in diagnoses from time of onset to one or two years later and often with relevant sex differences (Addington et al., 2006;Chaves et al., 2006).Finally, there are limitations in considering differences in the literature between CHR youth who transition to psychosis with individuals who already have a psychotic diagnosis.CHR participants who transition to psychosis are being assessed when they have a duration of untreated psychosis (DUP) as essentially zero whereas even with first episode participants the mean time of DUP is typically around one year (Addington et al., 2015;Addington et al., 2004).Furthermore, of patients in, for example, first episode clinics only a small proportion have been assessed in a clinic or research study for CHR individuals (Ajnakina et al., 2017).Even when they are seen in a CHR clinic or research program only a small proportion of CHR individuals actually develop psychosis (Addington et al., 2011;Addington et al., 2019).
It is possible there are some unique features of youth who present for care or participation in CHR research programs.Indeed, although there are limitations, it was suggested in the UK study that first episode patients who had initially presented at CHR services had a more insidious onset of illness (Ajnakina et al., 2017).
In conclusion, the aim of this study was to investigate sex differences in demographics, diagnoses, symptoms and functioning in CHR individuals at the point of transition to psychosis and compare to differences reported in the literature for individuals who have an SSD.Future research may require larger samples of those who transition to determine if there are more meaningful sex differences.Furthermore, it is imperative that future research focuses on continued follow-up of those who make the transition to psychosis to determine how initial diagnoses at the point of transition evolve.This would have important implications for treatment at the first episode.Finally, it should be noted that this paper focused only on reporting sex differences in clinical presentation.Gender was not studied in this paper thus future work should attempt to understand potentially different pathways by which sex and gender may influence psychosis.This would include examining the impact of different biomarkers such as brain structure, immune processes, HPA-Axis, and hormones as well as gender socialization, roles and identity (Christiansen et al., 2022).

CRediT authorship contribution statement
The study was designed, and monitored by Addington, Cannon, Cornblatt, Cadenhead, Woods, McGlashan, Perkins, Tsuang, Bearden, Walker, Mathalon, Keshevan, Stone & Seidman.Liu and Akseer undertook the data analysis.Chintoh wrote the first draft of the manuscript.Addington worked on later drafts.Braun helped with paper drafts and later data analyses.All authors contributed to and have approved the final manuscript.

Declaration of competing interest
Authors declare they have no conflict of interest with respect to this study.

Table 1
Sex comparison of baseline demographics of transitioning CHR participants.

Table 2
Sex comparison of baseline clinical and functioning characteristics of transitioning CHR participants.

Table 3
Sex comparisons of baseline diagnoses for transitioning CHR participants.a,b,c a Data missing for 25 participants (12 males, 13 females).b major depressive disorder, n = 424; persistent depressive disorder, n = 48; unspecified, n = 106.c Data missing for 2 more males, bipolar disorder, n = 7; cyclothymia, n = 4; bipolar NOS, n = 70.d Includes any one of panic disorder, agoraphobia, social anxiety disorder, specific phobia, generalized anxiety disorder or anxiety NOS NOS, Not Otherwise Specified.e data missing for 3 participants; 2 male, 1 female.f dysthymia or other, n = 17; major depressive disorder, n = 71.g bipolar NOS, n = 13; In partial remission, n = 1; cyclothymia, n = 1.No participant had bipolar disorder with psychotic features or bipolar at a severe level.

Table 4
Sex comparison of medications for transitioning CHR participants.

Table 5
Change over time in Positive and Negative Symptoms for Males and Females.

Table 6
Sex comparisons of clinical and functioning characteristics of transitioning CHR participants at the time of transition.
Abbreviations: SD, Standard Deviation; SOPS, Scale of Psychosis Risk Symptoms.

Table 7
Sex comparisons of diagnoses at time of transition for CHR.