Abstract
Individuals with psychosis-spectrum conditions have strikingly high rates of suicidal thoughts and behaviors, especially in the early stages of illness. Given increasing rates of suicide among adolescents, and given that psychosis symptoms often emerge during this developmental period, it is important that practitioners working with adolescents are attuned to the intersection of psychotic experiences and suicide risk. Furthermore, youth with emerging signs of psychosis often struggle with other mental health concerns that are linked to suicidal thoughts and behaviors, including depression, anxiety, mania, trauma, and substance use. Taken together, these factors indicate that identifying early signs of psychosis may be critical for addressing suicide risk, and psychosis-spectrum symptoms may be important experiences to include in safety planning for youth. Herein we discuss considerations for safety planning for adolescents experiencing psychosis-spectrum symptoms, drawn from clinical observations and pilot data collected from adolescents on a psychiatric inpatient unit and their families.
Keywords: adolescence, psychosis, suicidality, safety planning
Suicide is a major public health concern for youth, as it is the second leading cause of death for those aged 10–24 years.1 Individuals with psychosis-spectrum conditions have a markedly elevated risk for suicide in the early stage of illness, with even higher risk for those under the age of 18 and in the months before and after first presentation for psychiatric care.2,3 The time surrounding an individual’s first episode of psychosis, typically occurring in late adolescence or young adulthood, is linked to increased suicidal ideation, attempts, and death by suicide.3,4 Suicide risk is also elevated during the clinical high risk (CHR) phase of illness,5 when individuals begin to experience lower-intensity psychosis-spectrum experiences that put them at high risk for developing full-threshold psychosis in the future. The CHR phase of psychosis often emerges in adolescence and is characterized by the persistence of distressing and/or impairing positive symptoms (e.g., hallucinations or delusions) which are attenuated or below the threshold for a psychotic disorder diagnosis (e.g., experiences are less intense, more fleeting, and lacking full conviction).6 Examples of attenuated experiences include hearing indistinct whispers or intermittent voices that are recognized as not being real, or recurrent thoughts of being targeted by others, accompanied by doubt as to whether this is truly occurring. Approximately two-thirds of individuals at CHR endorse suicidal ideation, and an estimated 18% of people at CHR attempt suicide.5 Given that suicide risk may be particularly high for those with untreated psychosis-spectrum symptoms, under the age of 18, and within the first year of psychosis onset,4,5,7 health professionals working with adolescents need to be attuned to psychosis-spectrum symptoms and the intersection of psychosis and suicidal thoughts and behaviors (STBs).
Psychosis-spectrum conditions
Although psychotic disorders are relatively rare, with less than 1% estimated prevalence of psychotic diagnoses among adults, psychotic experiences are far more common, especially among adolescents. Research indicates that 10–25% of adolescents report some level of psychosis-spectrum experiences, however, the clinical significance of these experiences varies greatly across individuals.8 Notably, among adults with psychotic disorders, 20% report symptom onset prior to the age of 18, and distressing symptoms, compared to benign experiences, are linked to higher risk for future psychosis and comorbid concerns including suicidality. Taken together, these findings indicate that psychosis-spectrum experiences are relatively common and individuals’ reactions to these experiences may influence their clinical relevance and associated risk.
Individuals with full-threshold psychosis are often identified by providers due to outwardly observable disorganized and/or alarming thoughts, behavior, and communication. For those at CHR, however, attenuated psychosis symptoms may not be readily disclosed and/or perceived by providers due to lower-level symptom presentation (e.g., less severe, distressing, and impairing symptoms) that may be masked because of stigma, embarrassment, confusion, lack of insight, or relational factors that impede disclosure (e.g., lack of trust, rapport, or space to open up about experiences). Thus, providers may not be aware of important psychosis-spectrum symptoms associated with suicide risk without screening for them specifically and probing the impact of these experiences on STBs and other risk factors for STBs (e.g., stress, hopelessness, etc.). Notably, adolescents at CHR commonly experience comorbid mental health concerns including depression, anxiety, PTSD, mania, and substance abuse.9,10 These comorbid difficulties often motivate help-seeking behavior and may predominate over attenuated psychosis, and they are also associated with increased suicide risk. While co-occurring mental health concerns such as depression may account for some STBs for youth with psychosis-spectrum conditions, research indicates that positive symptoms of psychosis such as hallucinations and delusions may be uniquely associated with STBs (beyond the effects of other mental health diagnoses and symptoms).11–15
Psychosis-spectrum experiences and suicidality
Although more research is needed to be better understand the co-occurrence of psychosis and STBs, it is evident that identifying early signs of psychosis may be critical for reducing suicide risk for some youth, and psychosis-spectrum symptoms may be important factors to include in safety planning for these individuals. Pilot data collected by the authors from hospitalized adolescents in a psychiatric inpatient setting suggest that for some youth with psychosis-spectrum symptoms, unsafe behaviors labelled as “suicidal” are a function of positive symptoms (e.g., hallucinations or delusions), and subsequent life-threatening behaviors may lack intent or desire to die.16 For example, unsafe behaviors may stem from experiences such as command hallucinations (e.g., voices encouraging suicide), delusions of grandeur (e.g., believing oneself is invincible), impaired reality testing (e.g., believing that the world, oneself, or threats are not real), or other psychotic processes (e.g., injuring oneself to stop or silence voices or thoughts perceived to be externally controlled). These experiences are distinct from symptoms that typically drive suicidal intent (e.g., hopelessness, a desire to escape), and strategies to mitigate risk related to intent to die may not be helpful for these types of psychosis-driven experiences. As such, these contributing factors have important treatment implications and may require unique consideration in the context of safety planning.
Safety planning considerations
The goal of safety planning is to mitigate future risk by helping youth identify vulnerabilities, triggers, and warning signs, and problem-solve future occurrences of suicidal ideation by planning coping skills, help-seeking strategies, and ways to keep one’s environment safe (e.g., by limited access to lethal means).17 Although safety planning has been shown to reduce occurrences of suicidal behaviors, some evidence suggests that safety planning may not be as effective for individuals with severe mental illness.18 This may be particularly true for certain psychiatric conditions, such as psychotic disorders, that impede insight and distort perceptions of reality, necessary faculties for effective problem-solving. Furthermore, elements of traditional safety planning that focus on hopelessness and intent to die may not be as applicable to unsafe behaviors linked to hallucinations, delusions, and distorted reality. For these reasons, safety planning for youth with psychosis-spectrum conditions warrants special considerations aimed at improving the applicability and helpfulness for these adolescents and their caregivers. Outlined below are several steps that providers can take to help reduce safety-related risk for this population.
1. Screen for psychosis-spectrum experiences in mental health settings serving adolescents
There are several validated screening tools that can be used across settings to screen for potential psychosis-spectrum experiences in youth.19 Screening tools that are designed to measure attenuated psychosis are helpful tools to capture the full spectrum of experiences, rather than focusing on full-threshold psychotic experiences. These measures include more broad, tentative, and perhaps gentler language (e.g., “do you ever think that you may hear voices or other sounds that others don’t seem to hear”). This approach allows for endorsement of lower-level experiences, as compared to probing overt psychotic experiences that may be met with trepidation or symptom denial due to perceived connotation of saying “yes” to the experience (e.g., “do you hear voices that are not really there”). Some validated, open-source options for screening include the PRIME Screen-Revised,20 the Prodromal Questionnaire-Brief (PQ-B),21 and the Youth Psychosis At-Risk Questionnaire (YPARQ).22
2. Explore the clinical characteristics of potential psychosis-spectrum experiences
Follow up on screening questions to explore why adolescents have endorsed specific items. If adolescents are not asked explicitly about symptoms, they may be less likely to spontaneously disclose or open up about their experiences. Ask follow-up questions in an open and non-judgmental manner, remaining calm and supportive throughout the conversation. Empathize with the individual’s emotional experiences in reaction to symptoms, and refrain from challenging the veracity of experiences such as hallucinations or delusions, as this may impede rapport and/or rupture trust. As you seek to understand the adolescent’s experience, it may be helpful to probe specific characteristics that influence the intensity or severity of symptoms: duration (how long the experience lasts), frequency (how often it happens), conviction (how real it feels versus self-generated doubt), distress (how it makes them feel), impairment (how it interferes with functioning), meaning (how does the adolescent understand the experience and what it means), reactions (how the adolescent behaves in response to symptoms), and patterns (co-occurrence with other mental health or behavioral experiences; are symptoms linked to specific triggers or contexts).23 Having a full understanding of the adolescent’s experiences will help you to probe how psychosis-spectrum symptoms may intersect with STBs. Keep in mind that for some individuals, psychosis-spectrum experiences may not be considered clinically significant (i.e. they do not cause significant distress or impairment), and they may not need to be included in a safety plan, underscoring the need for conversation around the impact of these experiences.
3. Discuss safety in relation to psychosis-spectrum experiences
Once you have a better understanding of an individual’s psychosis-spectrum experiences, ask about how these experiences might impact safety, mood, anxiety, stress, substance use, and other mental health experiences that may influence STBs. Psychosis symptoms may directly impact safety and/or indirectly influence STBs via other factors (e.g., depression or hopelessness).24 Alternatively psychosis-spectrum experiences and STBs may not be related for some individuals. Attend to how psychosis-spectrum experiences may exacerbate stress and coincide with other risk factors that are relevant for the individual. Try to avoid making assumptions about the impact of symptoms by using Socratic questioning and open-ended prompts to explore how the individual understands their experiences. Another important thing to keep in mind is that among individuals with early psychosis, higher levels of insight and beliefs about negative outcomes for psychosis have been uniquely linked to suicidality, beyond the effects of depression.25 Insight and negative beliefs may also directly impact depression.3,26 These findings indicate that a person’s understanding of their illness and their beliefs about what psychosis might mean for their future may play an important role in their experience of STBs. Although more research is needed to better understand how to address psychosis-related stigma in youth,27 normalizing adolescents’ experiences, instilling hope, and taking a recovery-based approach to care may help individuals to be more future-oriented in their goals and perspectives.28
4. Tailor the safety plan components specifically to psychosis-spectrum experiences
As you start to create a safety plan, include as many psychosis-spectrum experiences related to safety as possible. It may even be helpful to generate a separate safety plan for psychosis-specific experiences, in addition to one created for general mood-related suicidality. Discuss ways to reduce exposure to triggers for hallucinations or delusions, identifying early warning signs of these experiences, listing unique coping or grounding skills effective for psychotic experiences, and naming trusted people to go to for distraction or adults to go to for help when specific experiences occur. It is particularly important to discuss ways to cope with each psychosis experience, and it could be helpful to provide a battery of coping options, including a variety of ideas for what someone can do on their own and what others may be able to do to offer support or direction when they notice signs of distress. For example, some coping skills for hearing voices might include listening to music, wearing an earplug in one ear, practicing vocal activities (singing, humming, reading or counting aloud, talking, or whispering), or using general grounding skills to orient oneself to the present reality. Remember that finding the right coping skills may be a process of trial and error, and more options will increase the likelihood of finding something that works.
5. Include caregivers in the process of safety planning
It is important to include caregivers in safety planning as much as possible, as they may be able to offer unique insight into the adolescent’s behavior patterns, especially in situations where the individual may have limited insight. Caregivers may also benefit from hearing about and understanding their adolescents’ experiences, as this insight may help them support the individual in different ways. Caregivers may need support in coping with their adolescent’s mental health struggles, and they will likely benefit from efforts to enhance their own self-efficacy in coping with a crisis. It may be helpful to ask the caregiver to create a safety plan for the adolescent from their perspective, outlining their observations of triggers and warning signs, ways that they can support coping, ways that they can avoid contributing to distress and minimize triggers, considerations for keeping the environment safe (e.g., locking up sharps and medications), and options for help to utilize in times of crisis. The method of safety planning alone with the adolescent, alone with the parent, and then together in a joint session has been shown to be helpful and well-received by families coping with adolescent suicidality.29 It may also be helpful to tailor safety plans to various contexts such as school or multiple households to help support the adolescent across settings. It might take some additional time to discuss the adolescent’s willingness to share their plan with supportive adults and problem solve barriers that may get in the way of discussing or using their plan in different contexts.
6. Incorporate a crisis plan
Include in the safety plan signs of mental health deterioration and stress that may indicate a need for hospitalization (for psychosis or safety concerns). This might include increasing signs of disorganization, such as difficulty communicating one’s thoughts, expressing bizarre ideas or not making sense, believing ideas that may influence unsafe behaviors, having difficulty participating in daily routines such as hygiene or work/school responsibilities, or demonstrating odd behaviors or changes in appearance that may indicate disconnection from reality. Other signs of crisis might include an individual skipping or refusing medication, social withdrawal or isolation, impulsive or risky behaviors, and expressing a desire or plan to end one’s life or leave the world. Collaborative crisis planning should include a discussion of the adolescent’s preferences for care in crisis situations, and a clear plan for seeking appropriate treatment in the event of crisis (who to include in the conversation and where to go for help). Crisis planning has been shown to reduce involuntary admissions among adults with psychosis,30 and this type of planning may reduce the likelihood of the adolescent experiencing traumatic psychiatric admissions.
7. Discuss strategies to maintain a healthy lifestyle and manage stress
While some individuals may report a direct influence of psychotic experience on STBs (e.g., command hallucination encouraging suicide), other individuals with psychosis-spectrum experiences report that STBs seem to be driven primarily by other mental health concerns such as depression and anxiety. For these youth, psychosis-spectrum symptoms may indirectly influence STBs by exacerbating stress, mood, anxiety, and other risk factors. In either case, it is important to discuss the impact of lifestyle choices and stress on mood, psychosis, STBs, and other risk factors relevant for the individual. Early psychosis is linked to high rates of substance misuse,31 and misuse of substances, particularly marijuana, is linked to higher suicide risk among those experiencing their first episode of psychosis.2 Youth with psychosis symptoms may be especially vulnerable to negative effects from marijuana use,32 including exacerbation of mental health distress, so it is important to discuss this openly and brainstorm alternatives to using.
8. Consider and frame safety planning as an ongoing, iterative process
Lastly, take your time with safety planning. It may take several sessions to complete a comprehensive draft and ensure that the individual and caregiver(s) understand all of the components and are willing and able to use it effectively. As families try to implement their safety plans, they may face barriers to address via problem solving and role playing.19 Consider the safety plan a living document that may need to evolve over time and with experience. It is helpful to complete a safety plan in a format that is editable (e.g., via an app, word doc, cloud or Google drive, etc.) and accessible across settings (e.g., on an app or saved as a picture on one’s phone or other devices). Families should be encouraged to add to and edit their safety plan as needed, or create a new one altogether, especially as strategies are determined to be helpful (or not). Remember that an individual’s safety planning needs may change over the course of treatment, and it is helpful to check in periodically to ensure that the safety plan stays relevant.
Conclusion
Assessing suicide risk and safety planning are important components of treatment for many adolescents experiencing mental health difficulties. Given the high risk for suicide associated with psychosis, practitioners working with adolescents may benefit from implementing procedures to screen for psychosis-spectrum symptoms and assess how these experiences may influence STBs. Helpful safety planning considerations for youth at CHR or with early psychosis might include reducing exposure to triggers for hallucinations or delusions, identifying early warning signs of these experiences, listing unique coping or grounding skills effective for psychotic experiences, and incorporating parent input to provide insight that may be unrecognized by the teen. A careful process of safety planning may be a critical preventative measure for the unique population of adolescents experiencing psychosis-spectrum symptoms who are at markedly high risk for suicide.
Acknowledgments
Dr. Thompson is supported by grant funding from the National Institute of Mental Health (K23 MH119211) and the American Foundation for Suicide Prevention (YIG-1-167-19).
Footnotes
Disclaimer
The views expressed herein are those of the authors and do not necessarily reflect the views of the National Institute of Mental Health or the American Foundation for Suicide Prevention.
Contributor Information
Elizabeth C. Thompson, Rhode Island Hospital and the Alpert Medical School of Brown University, Providence, RI.
Margaret Nail, Rhode Island Hospital and the Alpert Medical School of Brown University, Providence, RI.
Shirley Yen, Massachusetts Mental Health Center (MMHC)/Harvard Medical School, Boston, MA, and the Alpert Medical School of Brown University, Providence, RI.
References
- 1.Heron M Deaths: Leading Causes for 2017. Natl Vital Stat Rep. 2019. Jun;68(6):1–77. [PubMed] [Google Scholar]
- 2.Ayesa-Arriola R, Alcaraz E, Hernández B, Pérez-Iglesias R, Moríñigo J, Duta R, David A, Tabares-Seisdedos R, Crespo-Facorro B. Suicidal behaviour in first-episode non-affective psychosis: Specific risk periods and stage-related factors. Eur Neuropsychopharmacol. 2015. Dec;25(12):2278–88. [DOI] [PubMed] [Google Scholar]
- 3.Pompili M, Serafini G, Innamorati M, Lester D, Shrivastava A, Girardi P, Nordentoft M. Suicide risk in first episode psychosis: a selective review of the current literature. Schizophr Res. 2011. Jun;129(1):1–11. [DOI] [PubMed] [Google Scholar]
- 4.Palmer B, Pankratz V, Bostwick J. The Lifetime Risk of Suicide in Schizophrenia: a Reexamination. Arch Gen Psychiatry. 2005. Mar;62(3):247–53. [DOI] [PubMed] [Google Scholar]
- 5.Taylor P, Hutton P, Wood L. Are people at risk of psychosis also at risk of suicide and self-harm? A systematic review and meta-analysis. Psychol Med. 2015. Apr;45(5):911–26. [DOI] [PubMed] [Google Scholar]
- 6.Miller T, McGlashan T, Rosen J, Cadenhead K, Ventura J, McFarlane W, Perkins D, Godfrey P, Woods S. Prodromal assessment with the structured interview for prodromal syndromes and the scale of prodromal symptoms: predictive validity, interrater reliability, and training to reliability. Schizophr Bull. 2003;29(4):703–15. [DOI] [PubMed] [Google Scholar]
- 7.Nordentoft M, Madsen T, Fedyszyn I. Suicidal behavior and mortality in first-episode psychosis. J Nerv Ment Dis. 2015. May;203(5):387–92. [DOI] [PubMed] [Google Scholar]
- 8.Stevens J, Prince J, Prager L, Stern T. Psychotic disorders in children and adolescents: a primer on contemporary evaluation and management. Prim Care Companion CNS Disord. 2014;16(2):PCC.13f01514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Thompson E, Kline E, Ellman L, Mittal V, Reeves G, Schiffman J. Emotional and behavioral symptomatology reported by help-seeking youth at clinical high-risk for psychosis. Schizophr Res. 2015. Mar;162(1–3):79–85. [DOI] [PubMed] [Google Scholar]
- 10.Addington J, Piskulic D, Liu L, Lockwood J, Cadenhead K, Cannon T, Cornblatt B, McGlashan T, Perkins D, Seidman L, Tsuang M, Walker E, Bearden C, Mathalon D, Woods S. Comorbid diagnoses for youth at clinical high risk of psychosis. Schizophr Res. 2017. Dec;190:90–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.DeVylder J, Thompson E, Reeves G, Schiffman J. Psychotic experiences as indicators of suicidal ideation in a non-clinical college sample. Psychiatry Res. 2015. Apr 30;226(2–3):489–93. [DOI] [PubMed] [Google Scholar]
- 12.Hielscher E, DeVylder J, Saha S, Connell M, Scott J. Why are psychotic experiences associated with self-injurious thoughts and behaviours? A systematic review and critical appraisal of potential confounding and mediating factors. Psychol Med. 2018. Jul;48(9):1410–1426. [DOI] [PubMed] [Google Scholar]
- 13.Thompson E, Spirito A, Frazier E, Thompson A, Hunt J, Wolff J. Suicidal thoughts and behavior (STB) and psychosis-risk symptoms among psychiatrically hospitalized adolescents. Schizophr Res. 2020. Apr;218:240–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Capra C, Kavanagh D, Hides L, Scott J. Subtypes of psychotic-like experiences are differentially associated with suicidal ideation, plans and attempts in young adults. Psychiatry Res. 2015. Aug 30;228(3):894–8. [DOI] [PubMed] [Google Scholar]
- 15.Granö N, Salmijärvi L, Karjalainen M, Kallionpää S, Roine M, Taylor P. Early signs of worry: Psychosis risk symptom visual distortions are independently associated with suicidal ideation. Psychiatry Res. 2015. Feb 28;225(3):263–7. [DOI] [PubMed] [Google Scholar]
- 16.Thompson E, Yen S, Nail M, Spirito A, Wolff J. Exploring Links between Psychosis Spectrum Experiences and Suicidal Thoughts and Behaviors: Quantitative Patterns and Qualitative Themes from an Adolescent Inpatient Setting. Oral session at the 2022 Congress of the Schizophrenia International Society (SIRS), Florence, Italy. 2022. [Google Scholar]
- 17.Stanley B, Brown G. Safety planning intervention: a brief intervention to mitigate suicide risk. Cogn Behav Pract. 2012;19(2):256–264. [Google Scholar]
- 18.Stanley B, Brown G, Brenner L, Galfalvy H, Currier G, Knox K, Chaudhury S, Bush A, Green KL. Comparison of the safety planning intervention with follow-up vs usual care of suicidal patients treated in the emergency department. JAMA Psychiatry. 2018. Sep 1;75(9):894–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kline E, Schiffman J. Psychosis risk screening: a systematic review. Schizophr Res. 2018. Sep 1;75(9):894–900. [DOI] [PubMed] [Google Scholar]
- 20.Miller T The SIPS-Screen: a brief self-report screen to detect the schizophrenia prodrome. Schizophr Res. 2004;78. [Google Scholar]
- 21.Loewy R, Pearson R, Vinogradov S, Bearden C, Cannon T. Psychosis risk screening with the Prodromal Questionnaire—brief version (PQ-B). Schizophr Res. 2011. Jun;129(1):42–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ord L, Myles-Worsley M, Blailes F, Ngiralmau H. Screening for prodromal adolescents in an isolated high-risk population. Schizophrenia Res. 2004. Dec 1;71(2–3):507–8. [DOI] [PubMed] [Google Scholar]
- 23.McGlashan T, Walsh B, Woods S. The psychosis-risk syndrome: handbook for diagnosis and follow-up. Oxford University Press. 2010. [Google Scholar]
- 24.Huang X, Fox K, Ribeiro J, Franklin J. Psychosis as a risk factor for suicidal thoughts and behaviors: a meta-analysis of longitudinal studies. Psychol Med. 2018. Apr;48(5):765–776. [DOI] [PubMed] [Google Scholar]
- 25.Barrett E, Sundet K, Faerden A, Agartz I, Bratlien U, Romm K, Mork E, Steen N, Andreassen O, Melle I. Suicidality in first episode psychosis is associated with insight and negative beliefs about psychosis. Schizophr Res. 2010. Nov;123(2–3):257–62. [DOI] [PubMed] [Google Scholar]
- 26.Mintz A, Dobson K, Romney D. Insight in schizophrenia: a meta-analysis. Schizophr Res. 2003. May 1;61(1):75–88. [DOI] [PubMed] [Google Scholar]
- 27.DeLuca J, Yang L, Lucksted A, Yanos P, DeVylder J, Anglin D, Landa Y, Corcoran C. Reducing Stigma Among Youth at Risk for Psychosis: A Call to Action. Schizophr Bull. 2021. Oct 21;47(6):1512–1514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Williams A, Fossey E, Farhall J, Foley F, Thomas N. Recovery After Psychosis: Qualitative Study of Service User Experiences of Lived Experience Videos on a Recovery-Oriented Website. JMIR Ment Health. 2018. May 8;5(2):e37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.O’Brien K, Almeida J, Schofield M, Hall W, Aguinaldo L, Ryan C, Maneta E. A Safety and Coping Planning Intervention for Suicidal Adolescents in Acute Psychiatric Care. Cogn Behav Pract. 2021;28(1):22–39. [Google Scholar]
- 30.Molyneaux E, Turner A, Candy B, Landau S, Johnson S, Lloyd-Evans B. Crisis-planning interventions for people with psychotic illness or bipolar disorder: systematic review and meta-analyses. BJPsych Open. 2019. Jun 13;5(4):e53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Addington J, Addington D. Patterns, predictors and impact of substance use in early psychosis: a longitudinal study. Acta Psychiatr Scand. 2007. Apr;115(4):304–9. [DOI] [PubMed] [Google Scholar]
- 32.Bagot K, Milin R, Kaminer Y. Adolescent initiation of cannabis use and early-onset psychosis. Substance Abus. 2015;36(4):524–33. [DOI] [PubMed] [Google Scholar]
