Introduction
Psychotic disorders can significantly affect the individual, their family, as well as society on a broader scale (Whiteford et al., Reference Whiteford, Degenhardt, Rehm, Baxter, Ferrari, Erskine, Charlson, Norman, Flaxman, Johns, Burstein, Murray and Vos2013). Frequently, these disorders have their onset in late adolescence and early adulthood (Simon et al., Reference Simon, Coleman, Yarborough, Operskalski, Stewart, Hunkeler, Lynch, Carrell and Beck2017) and if the affected individual does not receive the appropriate holistic treatment, it may result in higher rates of disability and loss of productivity. Furthermore, there is a high early mortality associated with psychotic disorders (Olfson et al., Reference Olfson, Gerhard, Huang, Crystal and Stroup2015; Simon et al., Reference Simon, Stewart, Yarborough, Lynch, Coleman, Beck, Operskalski, Penfold and Hunkeler2018). Research conducted in the United States has demonstrated overall mortality after the initial diagnosis of a psychotic disorder being roughly 8 times that of general outpatients with deaths due to injury or poisoning (including self-harm, misadventure and assaults) accounting for the majority of that difference. This study also showed deaths due to self-inflicted injury or poisoning were more than 30 times as likely in the group with new psychotic disorder diagnoses as opposed to general outpatients (Simon et al., Reference Simon, Stewart, Yarborough, Lynch, Coleman, Beck, Operskalski, Penfold and Hunkeler2018).
Longitudinal studies have also echoed this high rate of suicide in early stages of psychosis with rates of up to eighteen times more than comparable general populations (Nordentoft et al., Reference Nordentoft, Jeppesen, Abel, Kassow, Petersen, Thorup, Krarup, Hemmingsen and Jørgensen2002; Dutta et al., Reference Dutta, Murray, Hotopf, Allardyce, Jones and Boydell2010; Yuen et al., Reference Yuen, Harrigan, Mackinnon, Harris, Yuen, Henry, Jackson, Herrman and McGorry2014; Simon et al., Reference Simon, Stewart, Yarborough, Lynch, Coleman, Beck, Operskalski, Penfold and Hunkeler2018).
Systematic reviews have reported the mortality rate from suicide in first-episode psychosis (FEP) to be between 1% and 4% (Sicotte et al., Reference Sicotte, Iyer, Kiepura and Abdel-Baki2021).
Depressive symptoms commonly occur in individuals with FEP and have been linked to poorer long-term outcomes. (Upthegrove et al., Reference Upthegrove, Birchwood, Ross, Brunett, McCollum and Jones2010; Gardsjord et al., Reference Gardsjord, Romm, Friis, Barder, Evensen, Haahr, ten Velden Hegelstad, Joa, Johannessen, Langeveld, Larsen, Opjordsmoen, Rund, Simonsen, Vaglum, McGlashan, Melle and Røssberg2016). Meta-analysis suggest depressive symptoms during FEP convey a greater risk of suicidal behaviours later in life (McGinty, et al., Reference McGinty, Sayeed Haque and Upthegrove2018) One quarter of individuals presenting with a first episode of a schizophrenia spectrum disorder also meet criteria for a full threshold depressive episode and a half have depressive symptoms that warrant further investigation and possibly treatment (Herniman et al., Reference Herniman, Allott, Phillips, Wood, Uren, Mallawaarachchi and Cotton2019). Alongside this, a meta-analysis found that nearly one in five individuals with a FEP engage in self-harm before starting treatment (Herniman et al., Reference Herniman, Allott, Phillips, Wood, Uren, Mallawaarachchi and Cotton2019). An Irish Early Intervention service reported 47% of FEP patients having suicidal ideation with 38% attesting to this in the month prior to diagnosis. (Foley et al., Reference Foley, Jackson, McWilliams, Renwick, Sutton, Turner, Kinsella and O’Callaghan2008). This study also reported that a higher degree of insight into mental illness is associated with a higher risk of suicide in those with a FEP. This supports the finding that greater insight is associated with deliberate self harm in those with a first episode of psychosis (Challis et al., Reference Challis, Nielssen, Harris and Large2013).
Interestingly, demographic risk factors associated with self-harm in FEP cohorts differ from other clinical cohorts. Younger females from lower social classes are more likely to present with self-harm and suicidal behaviours, however in a FEP cohort as per the AESOP study, it was found that males and those from higher social classes were more likely to engage in self-harm (Harvey et al., Reference Harvey, Dean, Morgan, Walsh, Demjaha, Dazzan, Morgan, Lloyd, Fearon, Jones and Murray2008). In addition, the presence of depressive symptoms at the time of a first episode is strongly associated with self-harm, as opposed to any specific psychotic symptom (Challis et al., Reference Challis, Nielssen, Harris and Large2013).
While the risk of suicide is present throughout the lifespan of an individual affected by a chronic and enduring psychotic illness, the early phase of the disorder represents a particularly high-risk period (Upthegrove et al., Reference Upthegrove, Birchwood, Ross, Brunett, McCollum and Jones2010; Sicotte et al., Reference Sicotte, Iyer, Kiepura and Abdel-Baki2021). Tragically, it is estimated that up to 5% of those with a diagnosis of schizophrenia will die by suicide (Hor and Taylor, Reference Hor and Taylor2010). This risk appears to be heightened during the early stages of the illness in comparison to later stages (Dutta et al., Reference Dutta, Murray, Hotopf, Allardyce, Jones and Boydell2010).
To further the existing research in this area this study aimed to determine: (i) the prevalence of depressive symptoms and any associated demographic or clinical factors in young people with a FEP and (ii) the prevalence of self-harm and suicidal behaviours in young people presenting with a FEP and any associated demographic or clinical factors.
Methods
Participants
This study included young people diagnosed with a FEP between 1st March 2019 and 31st October 2020. The Early Psychosis Prevention and Intervention Centre (EPPIC) service provides care to young people aged 15–24, diagnosed with a FEP, operationalised as experiencing full threshold psychotic symptoms daily for at least one week. The service includes all individuals with a FEP and this includes individuals with a diagnosis of drug induced psychosis, concurrent substance abuse or personality disorders.
Setting
The EPPIC is an early intervention for psychosis service within Orygen, a state-funded youth mental health service covering a defined catchment area of over one million people within Northern and Western Melbourne.
Study design
This was a secondary analysis of an existing dataset of a cohort study examining incidence and admission rates of first episode of psychosis. The study consisted of consecutive presentations of young people with a FEP over a twenty-month period between March 2019 and October 2020.
Instruments & sources of data
The Patient Health Questionnaire (PHQ-9) is an instrument that can be used to screen, diagnose and measure the severity of depression (Kroenke et al., Reference Kroenke, Spitzer and Williams2001). This instrument can be self-administered, and it scores each of the diagnostic and statistical manual of mental disorders criteria for major depressive disorder from 0 (not at all) to 3 (nearly every day), providing a total score ranging from 0 to 27. A PHQ-9 scores of 0–4 indicate no depressive symptoms, 5–9 mild, 10–14 moderate, 15–19 moderately severe and 20–27 severe depressive symptoms (Kroenke et al., Reference Kroenke, Spitzer and Williams2001). Scores below 5 typically indicate the absence of a depressive disorder. Scores ranging from 5 to 9 generally indicate either the absence of depression or the presence of subthreshold depression. Scores between 10 and 14 encompass a spectrum of patients. Scores of 15 or higher typically suggest the presence of major depression (Kroenke et al., Reference Kroenke, Spitzer and Williams2001).
The PHQ-9 has been shown to be able to accurately identify those at increased risk of suicide amongst outpatients with psychotic disorders (Simon et al., Reference Simon, Yarborough, Rossom, Lawrence, Lynch, Waitzfelder, Ahmedani and Shortreed2019).
Other research looking at the PHQ-9’s validity in a wider psychiatric patient sample which includes psychosis supported its use as a severity measure and as a measure of treatment outcome (Beard et al., Reference Beard, Hsu, Rifkin, Busch and Björgvinsson2016).
The categories of moderately severe and severe were grouped and categorised as ‘severe depression’. A clinician-rated standardised risk assessment ‘Clinical Risk Assessment and Management in the Community’ completed for all participants at the time of presentation. This is available in Supplementary Material attached.
Statistical analysis
SPSS version 27 was used for statistical analysis. The data were determined to be either normally or not normally distributed and the appropriate parametric or non-parametric statistical test was then conducted. We used the mean and standard deviation as descriptive statistics for parametric data and the median and intra-quartile ranges for non-parametric data. Binary logistic regression was performed to determine odds ratios for the predictor variables for the outcomes of severe depression and also suicidal intentions & behaviours. Bivariate analysis was conducted initially, and then multivariate regression was used to analyse any significant predictor variables at a p value < 0.05 in the bivariate analysis. T-tests, Mann Whitney U test and chi square tests were used for sociodemographic factors. Cases with missing data were excluded from the analysis.
Ethical approval
Ethical approval was received from the Melbourne Health Human Research Ethics Committee (HREC). The study included routinely collected clinical data and a waiver for individual consent was granted by the HREC.
Results
Eligibility of participants
A total of 480 young people presented with a FEP during the study period and of these 74.0% (n = 355) had an assessment of suicidal behaviours at the time of presentation and were included in this study. A comparison of those who were included and not included in the study was undertaken and it was found that migrants were less likely to have an assessment of suicidal risk at presentation (66.0% v. 78.3% of Australian-born, χ 2 = 8.03, df = 1, p = .005) and young people who were not in employment or education were also less likely to have a formal assessment of suicidal behaviours (69.0% v. 78.6%, χ 2 = 5.69, df = 1, p = .017). In addition, young people who were admitted at the time of presentation were also less likely to be included in this study, as the formal risk assessment was not completed (67.8% v. 78.9%, χ 2 = 7.58, df = 1, p = .006). There were no differences in regards to sex or marital status in those who were eligible to be included and those who were actually included.
Description of participants
Of the 355 young people included in the study, the median age of participants was 20 years (I.Q.R.: 17–22). Of these, 57.2% (n = 203) were male, 42.8% (n = 152) were female, 95.4% (n = 331) were single and over one quarter (28.3%, n = 97) were migrants. A total of 58.4% (n = 206) were employed or studying at the time of presentation. Concurrent substance abuse was present in 42.8% (n = 148). A total of 40.8% (n = 145) of young people were admitted to hospital at the time of presentation and 69.0% (n = 100) of these admissions were involuntary. The demographic and clinical characteristics of the included cohort are presented in Table 1.
Table 1. Demographic and clinical characteristics of total cohort
Prevalence of suicidal behaviours and associated demographic or clinical factors
At the time of presentation, 34.6% (n = 123) of young people had suicidal ideation, while 6.2% (n = 22) had suicidal intent with a plan. A total of 21.4% (n = 76) of young people had engaged in self-harm before their presentation. Combined, 39.7% (n = 141) of the young people presented with suicidal ideation, planning or self-harm (collectively termed suicidal ideation or behaviours). On univariate analysis, migrants were less likely than Australian-born individuals to present with suicidal ideation or behaviours (23.7% v. 45.9%, OR = 0.37, 95% C.I. 0.22–0.62). Additionally, 26.9% of young people who were admitted to hospital had suicidal ideation or behaviours compared to 48.6% of those who treated as an outpatient (OR = 0.39, 95% C.I. 0.25–0.62) and this was similar for those admitted involuntarily (22.0% v. 47.6%, OR = 0.32, 95% C.I. 0.19–0.55). To further understand this finding, the individual components of the group variable of suicidal ideation or behaviours were examined individually. It was found that young people admitted to hospital were less likely to present with suicidal ideation (21.4% v. 44.0%, OR = 0.35, 95% C.I. 0.21–0.56), but there was no difference in the proportion with suicidal plan or intent (6.9% v. 5.7%, OR = 1.22, 95% C.I. 0.51–2.90) or self-harm (17.9% v. 23.9%, OR = 0.70, 95% C.I. 0.41–1.18). In interpreting these findings, it needs to be considered that those admitted to hospital were more likely to have missing data relating to risk. The association between demographic and clinical factors and suicidal ideation or suicidal behaviours are presented in Table 2.
Prevalence of depressive symptoms and associated demographic or clinical factors
The PHQ-9 was only completed in 46.5% (n = 165) of the sample. The mean level of depressive symptoms on the PHQ-9 was 17.8 (s.d. ± 7.2) and this represented moderately severe depressive symptoms. At the time of presentation, 23.6% (n = 39) of young people with a FEP had moderately severe depressive symptoms and a further 47.9% (n = 79) had severe depressive symptoms. Combined, this resulted in 71.5% (n = 118) having moderately severe or severe depressive symptoms at the time of presentation. Only 6.1% (n = 10) of the total cohort had no depressive symptoms at the time of presentation. The associations between demographic and clinical factors associated with the presence of moderately severe or severe depressive symptoms are presented in Table 3. Individuals who were involuntarily admitted were less likely to have moderate severe or severe depressive symptoms (52.5% v. 77.6%, OR = 0.32, 95% C.I. 0.15–0.69).
Table 2. Prevalence of suicidal ideation or behaviours (intent or self-harm)
Variables that had a p < 0.10 on univariate analysis were entered into a multivariate analysis model to determine the adjusted odds ratios.
Association between depressive symptoms and suicidal ideation or behaviours
A total of 64.4% (n = 76) of young people with moderately severe or severe depressive symptoms had suicidal ideation or behaviours compared to 25.5% (n = 12) with lower levels of depressive symptoms (OR = 4.60, 95% C.I. 2.38–8.90). When controlled for potential confounds, the association between moderately severe or severe depressive symptoms and suicidal ideation or behaviours remained significant (OR = 4.21, 95% C.I. 2.10–8.44), as did the association with suicidal ideation or behaviours with migrant status (OR = 0.35, 95% C.I. 0.14–0.87) and these results are presented in Table 2.
Discussion
Summary of findings
Depressive symptoms and suicidal behaviours are highly prevalent in young people presenting with a first episode of psychosis, with nearly four out of ten young people presenting with a FEP having suicidal ideation, intent or having engaged in self-harm. Depressive symptoms were also highly prevalent in this clinical population and the presence of depressive symptoms was the strongest predictor of suicidal ideation and behaviours.
Limitations
There are important limitations in this study. There are high rates of missing data, which affected the representativeness of the cohort. It was found that young people admitted to hospital were less likely to attest to suicidal ideation or behaviours. This may be attributed to a bias from missing data as those admitted to hospital were more likely to have missing data in relation to risk or it could be due to the intensive community services that have been established within this service, including home based care teams.
Secondly, the information utilised for this study was obtained in the initial assessment at the time of presentation and it is possible that an individual may reveal more information in subsequent assessments in regards to suicidal ideation, intent and behaviours, particularly as trust within the therapeutic relationship is established.
Furthermore, the specific first-episode psychotic disorder diagnoses were not available for this study, as these diagnoses are typically confirmed after a longitudinal assessment. It would have been informative to have been able to examine the prevalence of depressive symptoms, suicidal ideation and behaviours across the different diagnostic groups. Additionally, we did not have information pertaining to psychotic symptoms at the time of presentation and while this information was elicited in later clinical assessments, the severity of symptoms was not assessed using a structured instrument.
Comparison to previous literature
Depression
This study showed a prevalence rate of moderately severe/severe depression in the sample at 71.5%. Another study using the CDSS recorded the prevalence rate of depression in FEP at 59% (Upthegrove et al., Reference Upthegrove, Ross, Brunet, McCollum and Jones2014). Older literature using both Becks Depression Inventory and the CDSS recorded the prevalence of depression in acute psychosis at 70% (Birchwood et al., Reference Birchwood, Iqbal, Chadwick and Trower2000), broadly in line with this study’s findings. As only 46.5% of the cohort were assessed in respect of depressive symptoms there may be some concerns with regard to the potential of selection bias with the possibility of assessors evaluating only those they suspected of having depression.
Across previous literature there is a wide prevalence rates cited – a previous review of 36 studies cited the depression rate in patients with schizophrenia between 6% and 65%, modal rate being 25% (Siris and Bench Reference Siris and Bench2003) This wide range of prevalence owes to differences in how depressive symptoms are defined and measured across various studies (Hou et al., Reference Hou, Ma, Cai, Li, Zang, Jia, Lin, Chiu, Ungvari, Hall, Zhong, Cao and Xiang2016). Ideally all participants in this sample would have been assessed for depression symptoms at presentation in order to allow for a more representative result. Whilst this study utilised the PHQ-9 to measure depressive symptoms it would be beneficial to have this supported with a concurrently completed CDSS.
Suicidal behaviours
This study reported 34.6% of the sample had suicidal ideation at the time of presentation to services, compare to 47% reported in an Irish study (Foley et al., Reference Foley, Jackson, McWilliams, Renwick, Sutton, Turner, Kinsella and O’Callaghan2008) and 27% in a meta-analysis (Sicotte et al., Reference Sicotte, Iyer, Kiepura and Abdel-Baki2021).
This study showed a total of 21.6% had engaged in self-harm prior to presentation to the service. As previously stated a meta-analysis found that nearly one in five individuals with a FEP engage in self-harm before starting treatment (Herniman et al., Reference Herniman, Allott, Phillips, Wood, Uren, Mallawaarachchi and Cotton2019) which is quite consistent with our findings.
Clinical implications
There are important clinical implications that arise from the finding of this study. First, it highlights the need to assess for depressive symptoms alongside psychotic symptoms. Typically, depressive symptoms would be managed using a combination of pharmacological, psychological and social interventions, depending upon the severity of symptoms and the individual’s choice (Barnes et al., Reference Barnes, Drake, Paton, Cooper, Deakin, Ferrier, Gregory, Haddad, Howes, Jones, Joyce, Lewis, Lingford-Hughes, MacCabe, Owens, Patel, Sinclair, Stone, Talbot, Upthegrove, Wieck and Yung2020). However, there is a lack of evidence for the effectiveness of specific interventions for depressive symptoms in individuals with a psychotic disorder, which leads to a lack of clinical guidelines in this area (Barnes et al., Reference Barnes, Drake, Paton, Cooper, Deakin, Ferrier, Gregory, Haddad, Howes, Jones, Joyce, Lewis, Lingford-Hughes, MacCabe, Owens, Patel, Sinclair, Stone, Talbot, Upthegrove, Wieck and Yung2020). A trial of antidepressant medication in individuals with a diagnosis of schizophrenia with concurrent depressive symptoms has been recommended by the British Association for Psychopharmacology, however it emphasises that more research is needed in this area (Barnes et al., Reference Barnes, Drake, Paton, Cooper, Deakin, Ferrier, Gregory, Haddad, Howes, Jones, Joyce, Lewis, Lingford-Hughes, MacCabe, Owens, Patel, Sinclair, Stone, Talbot, Upthegrove, Wieck and Yung2020). The majority of prescribing clinicians appear to adhere to this recommendation, with nearly 70% of prescribers stating that they prescribe antidepressant medication as a first line treatment for these individuals, while nearly one third recommend cognitive behavioural therapy and 12% do not recommend an active intervention but continue to monitor the individuals mental state (Bashir et al., Reference Bashir, Griffiths and Upthegrove2022).
The ‘watch and wait’ approach could also be considered, as previous research has indicated that the majority of depressive symptoms will resolve after the initiation of antipsychotic medication and that levels of depression are relatively low when positive psychotic symptoms are in remission (Oosthuizen et al., Reference Oosthuizen, Emsley, Niehaus, Koen and Chiliza2006). However, it also needs to be considered that this is a high-risk period for further self-harm or suicide attempts and the findings from this study demonstrate that depressive symptoms are strongly associated with suicidal ideation, intent and self-harm, therefore, the more conservative ‘watch and wait’ approach would need to include frequent monitoring of the severity of depressive symptoms and risks.
A further challenge in this clinical population is that differentiating depressive symptoms from negative symptoms can be challenging, especially when using rating scales that may exhibit symptom overlap, such as anhedonia, anergia, and amotivation. In this study, negative symptoms were not measured separately, so it is possible that these symptoms may have been recorded as depressive symptoms and this is an important limitation to consider when interpreting the findings of this study. However, there is some evidence that negative symptoms may respond to antidepressant medication (Jockers-Scherubl et al., Reference Jockers-Scherübl, Bauer, Godemann, Reischies, Selig and Schlattmann2005; Singh et al., Reference Singh, Singh, Kar and Chan2010).
This study highlights the established importance of the need to assess young people presenting with a FEP for depressive symptoms, suicidal ideation, intent and self-harm but also conversely highlights the need to screen young people presenting with suicidal behaviours for any signs of psychosis. Self-harm and suicidal behaviour tend to present out of hours and to emergency departments (McNicholas et al., Reference McNicholas, O’Sullivan, Lennon, Doherty and Adamson2010) or other on-call settings, whilst this can be busy environment screening for psychotic symptoms should be a pertinent part of assessment. There is also some emerging evidence that self-harm and suicidal behaviours can precede the onset of a psychotic disorder, emphasising the importance of early identification and intervention. as it has been found that 12.8% of young people who presented with self-harm subsequently were diagnosed with a psychotic disorder by the age of 28 (Bolhuis et al., Reference Bolhuis, Lång, Gyllenberg, Kääriälä, Veijola, Gissler and Kelleher2021).
Conclusions
Even when the limitations of this study are considered, the main finding of the study still stand that the prevalence of depressive symptoms, suicidal ideation and behaviours are high in young people with a first episode of psychosis and these factors need to be considered routinely in the assessment and management decisions of this clinical population.
Table 3. Prevalence of severe depression
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/ipm.2024.26.
Competing interests
None.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Open access
