Abstract
Objective:
The aim was to describe treatment history including medications, psychosocial therapy and hospital visits of participants in the Canadian Psychiatric Risk and Outcomes Study (PROCAN).
Methods:
PROCAN is a 2-site study of 243 youth/young adults aged 12 to 25 y, categorized into 4 groups: healthy controls (n = 42), stage 0 (non-help seeking, asymptomatic with risk mainly family history of serious mental illness (SMI); n = 41), stage 1a (distress disorders; n = 52) and stage 1b (attenuated syndromes; n = 108). Participants were interviewed regarding lifetime and current treatments, including medications, psychosocial therapies and hospital visits.
Results:
The number receiving baseline medications differed significantly across groups (P < 0.001): 0% healthy controls, 14.6% stage 0, 32.7% stage 1a and 34.3% stage 1b. Further, 26.9% and 49.1% of stage 1a and stage 1b participants received psychosocial therapy at baseline, indicative of statistically significant differences among the groups (P < 0.001). Similar results were observed for lifetime treatment history; stage 1b participants had the highest frequency of lifetime treatment. Medications started in adulthood (>18 y of age) were the most common for initiation of treatment compared to childhood (0 to 12 y) and adolescence (13 to 17 y) for stage 1a and 1b participants. Lifetime mental health hospital visits differed significantly across groups (P < 0.001) and were most common in stage 1b participants (29.6%) followed by stage 1a (13.5%), stage 0 (4.9%) and healthy controls (2.4%).
Conclusion:
We found that treatment history for participants in the PROCAN study differed among the at-risk groups. Future initiatives focused on determining the effects of treatment history on SMI are warranted.
Keywords: clinically high-risk, psychosis, serious mental illness, treatment history, medications, psychosocial therapy, hospital visits
Abstract
Objectif :
Le but était de décrire les antécédents de traitements, y compris les médicaments, la thérapie psychosociale et les visites à l’hôpital, des participants à l’Étude canadienne sur les risques et les résultats en psychiatrie (PROCAN).
Méthodes :
PROCAN est une étude en deux endroits de 243 adolescents/jeunes adultes de 12 à 25 ans, catégorisés en quatre groupes: témoins en santé (n = 42), stade 0 (asymptomatique sans recherche d’aide, dont le risque est principalement les antécédents familiaux de maladie mentale grave [MMG]; n = 41), stade 1a (troubles de détresse; n = 52) et stade 1b (syndromes atténués; n = 108). Les participants ont été interviewés concernant les traitements de duré de vie et actuels, y compris les médicaments, la thérapie psychosociale et les visites à l’hôpital.
Résultats :
Le nombre de sujets qui recevaient des médicaments au départ différait significativement entre les groupes (p < 0,001): 0% des témoins en santé, 14,6% du stade 0, 32,7% du stade 1a, et 34,3% du stade 1b. En outre, 26,9% du stade 1a et 49,1% du stade 1b recevaient une thérapie psychosociale au départ, indicatif de différences statistiquement significatives entre les groupes (p < 0,001). Des résultats semblables ont été observés pour les antécédents de traitements de durée de vie; les participants du stade 1b ayant la fréquence la plus élevée de traitements de durée de vie. Les médicaments commencés à l’âge adulte (> 18 ans) étaient les plus communs pour l’initiation d’un traitement comparativement à l’enfance (de 0 à 12 ans) et à l’adolescence (13 à 17 ans) pour les participants des stades 1a et 1b. Les visites à l’hôpital de durée de vie pour la santé mentale différaient significativement entre les groupes (p < 0,001) et étaient plus communes chez les participants du stade 1b (29,6%) suivis du stade 1a (13,5%), du stade 0 (4,9%) et des témoins en santé (2,4%).
Conclusion :
Nous avons constaté que les antécédents des traitements pour les participants à l’étude PROCAN différaient entre les groupes à risque. De futures initiatives visant à déterminer les effets des antécédents des traitements sur la MMG sont justifiées.
Introduction
Most symptoms related to mental health disorders first emerge during adolescence, often with a negative impact that can continue or even worsen into adulthood.1–3 Early intervention in psychiatry and youth mental health are now areas of focused research and suggested to be the number one priority for health services development in Canada.4,5 The emergence of symptoms, accompanied by various risk factors, suggests that youth may be at risk for a diagnosis of a serious mental illness (SMI), such as psychosis, bipolar disorder, or recurrent depression.6 McGorry and colleagues7,8 have proposed a clinical staging model, which incorporates risks and symptoms to define stages of severity along a continuum of progression from healthy to full illness expression. This clinical model stages individuals based on known risk factors, such as family history, history of trauma or obstetric complications, presentation of symptoms, experience of attenuated syndromes/disorders, formal discrete mental health diagnoses and progressive disease. The lower stages (0 to 1b) are indicative of individuals categorized as at-risk for SMI, whereas the higher stages (2 to 4), categorize individuals with a formal diagnosis of SMI where the impact and progression of disease are objectively present. Early intervention and youth mental health research specifically focusing on at-risk or subthreshold populations may provide insights to later SMI development.9
The Canadian Psychiatric Risk and Outcome Study (PROCAN) is a 2-site longitudinal study of 243 youth, ages 12 to 25 y, who are at different stages of risk, as suggested by the clinical model described above. Our overall aim is the longitudinal assessment of several clinical and psychosocial factors to improve our knowledge of youth at risk of SMI, and to develop and validate a prediction algorithm of clinical outcome that includes demographic, clinical, and psychosocial predictors. Our first goals are to characterize these young people who may be at various stages of risk of developing SMI. Initially, to conceptualize stages of risk, we utilized the clinical staging model suggested by McGorry and colleagues.7,8
To date, the major focus in the literature for subthreshold diagnoses is in psychosis. Young people may present with attenuated psychotic symptoms (i.e., mild psychotic-like symptoms) that have not yet manifested into full-blown psychosis and are said to be at clinical high-risk (CHR) for psychosis.10 CHR youth may receive various types of treatment;11 however, to date, little is known about the treatment history of individuals who are, according to the clinical staging model, at-risk for SMI. Similarly, little is known about how participants in the different stages of risk compare to one another in terms of treatment or mental health service utilization.
It is estimated that, in Canada, about 12% of individuals aged 12 to 24 y interact with mental health professionals regarding their emotions, mental health, or drug use.12 Another report observed that approximately 10% to 20% of youth are experiencing mental health illnesses or disorders, which also represents a need for treatment and services.13 Therefore, understanding the lifetime treatment history and access to healthcare resources of CHR and at-risk of SMI groups may inform healthcare professionals regarding the risk profiles of youth who seek and receive service.
Our objective for this paper was a focus on both current and past treatments of youth at various stages of risk, including lifetime medication use, psychosocial therapies, and hospital visits. We hypothesized that those with subthreshold diagnoses (stage 1b) would have received a greater proportion of psychotropic medications, psychosocial therapy and hospital visits compared with those who are “help-seeking” and have mild symptoms (stages 1a). We also hypothesized that those with mild symptoms (stage 1a) accessed more services compared to non-help seeking individuals with risk factors (stage 0) and healthy controls.
Methods
Study Sample
Youth and young adults aged 12 to 25 y, male and female, were recruited to participate in the Canadian Psychiatric Risk and Outcome Study (PROCAN) via various recruitment strategies in Calgary, Alberta and Toronto, Ontario. A more complete description of the study methods has been published elsewhere.14 Participants were recruited through general health professionals, mental health clinicians, school personnel, and community strategies including in-service presentations, public health outreach and advertisements in local newspapers and other public places (e.g., buses, trains). In addition, self- and family-referrals were accepted for recruitment. Individuals experiencing a mental health episode were not recruited into the PROCAN study. Sixty-five percent of all referrals were self/family referrals from advertising and 18.9% from a triage service, most of whom were in stage 1b. Details of recruitment are presented in Supplementary Table 1.
Individuals were screened over the phone and if they did not meet the exclusion criteria, were invited for an in-person screening interview. Inclusion criteria included youth and young adults who were identified as one of the following: a healthy control (no symptoms and no family history of mental health diagnosis), stage 0 (no symptoms and presenting with risk factors, such as a family member with a formal mental health diagnosis, evidence of obstetric complications or a history of severe trauma; thus considered high-risk for SMI), stage 1a (experiencing low mood or distress), or stage 1b (those experiencing subthreshold symptoms for a SMI). See details of risk stage model classification in Supplementary Table 2. Further exclusion criteria included: 1) a current or lifetime Axis I disorder of psychosis, bipolar disorder, or recurrent major depression, 2) an IQ of less than 70, and 3) a past or current history of a significant central nervous system disorder.
We enrolled 243 youth in the PROCAN study, including 42 healthy controls, 41 stage 0, 52 stage 1a, and 108 stage 1b. This study was approved by the University of Calgary Conjoint Health Research Ethics Board (REB 14-1710) and the Sunnybrook Research Ethics Board (100-2015). All individuals provided written consent or assent (for minors) accompanied by parental consent.
Mental Health Treatment and Resource Utilization
Participants and parents (for younger participants) were interviewed and asked to recall 3 types of treatment history: 1) psychotropic medications, 2) psychosocial therapy, and 3) hospital visits related to mental health, including both emergency visits and inpatient stays. Lifetime treatment history was documented from birth up to and including the date of the initial assessment. Therefore, we collected a full treatment history, including time periods of no treatment use, over the life-course of participants.
Medication start and stop dates, daily dose if routine, average frequency of use, and medication compliance were all documented. Changes in dose were captured as a separate prescription of medication. Type of medications for which information was collected included anti-depressants, antipsychotics, mood stabilizers, anxiolytics, stimulants and non-stimulants for attention defect hyperactivity disorder (ADHD). For psychosocial therapy, the type of therapy recorded included: case management, supportive therapy, cognitive-behavioural therapy (CBT), psychodynamic therapy, interpersonal therapy, stress management therapy, school counselling, family therapy, cognitive remediation, and multiple types combined. Participants were asked about the number of sessions attended, and the start and stop dates of those sessions. Finally, types of hospital visits were indicated as either visits to the emergency department or inpatient stays for mental health/psychiatric concerns, including start and stop dates of each hospital visit.
Current treatments (medication and psychosocial therapy) were derived as yes or no as to whether a treatment was being received on recruitment into the PROCAN study. Participants may have been on more than one medication, e.g., an antipsychotic and an antidepressant, in which case both medications would be counted as “yes”. Duration of current treatment was calculated only for those on a treatment and included the start time of the treatment up to the PROCAN study recruitment date. Lifetime treatments (medication and psychosocial therapy) were captured as yes or no depending on whether participants had ever received treatment in their lifetime (lifetime use included current use).
Statistical Analysis
Descriptive statistics were used to describe participants separately for all 4 groups (healthy controls, stage 0, stage 1a, and stage 1b). Frequencies outlining current (at baseline assessment) and lifetime (from birth up to and including baseline assessment) use of psychotropic medications, psychosocial therapy, and mental health-related hospital visits were calculated separately for all groups. Chi-squared tests were used to compare the dichotomous measures (yes v. no) whereas ANOVAs were used to compare continuous measures to determine differences between groups. In addition, for participants who had a treatment history, frequencies of different types of medications, psychosocial therapy, and hospital visits were calculated. Duration of medication use and inpatient hospital stays, as well as the number of psychosocial therapy sessions were computed. Lifetime treatment history for medications and psychosocial therapies were categorized into life-periods including childhood (ages 0 to 12 y), adolescence (13 to 17 y), and adulthood (18 y and over). All statistical analyses were performed using Stata v13 (Stata-Corp LP, College Station, TX) with an α < 0.05.
Results
Demographics
The subject demographics are presented in Table 1. There was at least one difference between the mean baseline age and years of education among the study groups (age, P = 0.002; years of education, P < 0.001). Further, the proportion of participants living with family was different among healthy controls (69.1%), stage 0 (65.9%), stage 1a (76.9%) and stage 1b (88.0%), with P = 0.008. Because youth in stage 1b were significantly younger than those in the other 3 groups, this age difference most likely accounts for the observed significant differences in years of education, likelihood of living at home with family, and employment status.
Table 1.
Participant demographics.
| Baseline Characteristics | Healthy Controls (n = 42) |
Stage 0 (n = 41) |
Stage 1a (n = 52) |
Stage 1b (n = 108) |
P valuea |
|---|---|---|---|---|---|
| Age | 19.1 (3.8) | 18.4 (4.3) | 18.4 (3.5) | 16.8 (3.3) | 0.002 |
| Years of education | 12.4 (3.3) | 11.5 (3.4) | 11.4 (2.6) | 10.3 (2.6) | <0.001 |
| Sex | |||||
| Male | 22 (52.4%) | 18 (43.9%) | 22 (42.3%) | 47 (43.5%) | 0.755 |
| Female | 20 (47.6%) | 23 (56.1%) | 30 (57.7%) | 61 (56.5%) | |
| Race | |||||
| White (European) | 21 (50.0%) | 29 (70.8%) | 33 (63.5%) | 68 (63.0%) | 0.159 |
| Interracial | 4 (9.5%) | 6 (14.6%) | 9 (17.3%) | 14 (13.0%) | |
| Other | 17 (40.5%) | 6 (14.6%) | 10 (19.2%) | 26 (24.0%) | |
| Marital status | |||||
| Single | 42 (100.0%) | 37 (90.2%) | 52 (100.0%) | 104 (96.3%) | 0.035 |
| Other | 0 | 5 (9.8%) | 0 | 4 (3.7%) | |
| Current living arrangement | |||||
| Living with family | 29 (69.1%) | 27 (65.9%) | 40 (76.9%) | 95 (88.0%) | 0.008 |
| Other | 13 (30.9%) | 15 (34.1%) | 12 (23.1%) | 13 (22.0%) | |
| Currently employment status | |||||
| Fulltime/part time | 20 (47.6%) | 16 (39.0%) | 30 (57.7%) | 30 (27.8%) | 0.002 |
| Unemployed | 22 (52.4%) | 25 (61.0%) | 22 (42.3%) | 78 (72.2%) | |
| Current student | |||||
| No | 9 (21.4%) | 5 (12.2%) | 14 (26.9%) | 14 (13.0%) | 0.110 |
| Yes | 33 (78.6%) | 36 (87.8%) | 38 (73.1%) | 94 (87.0%) |
Values presented as: means (SD) or n (%).
aANOVA for continuous and χ2 test for categorical baseline characteristics.
Current Treatment History
There was a statistically significant difference in medication use at baseline among the study groups (P < 0.001) and when excluding healthy controls (P = 0.021). No healthy controls were on medications, whereas 32.7% and 34.3% of stage 1a and stage 1b participants, respectively, were prescribed medications. The most common type of medication reported in stages 1a and 1b were antidepressants followed by stimulants; antipsychotics were less often prescribed. The mean duration of medication use was highest in the stage 0 group (47.1 [56.3]); however, there were no statistically significant differences among the groups (P = 0.807).
A significant difference was apparent for current psychosocial therapy among the study groups (P < 0.001) and when excluding healthy controls (P < 0.001). Almost half of the stage 1b participants (49.1%) reported currently being involved in psychosocial therapy, whereas 26.9% of stage 1a participants were involved in psychosocial therapy. Case management and supportive therapy were the most commonly reported psychosocial therapies amongst stage 1a and stage 1b participants. The mean number of psychosocial therapy sessions was the highest in the stage 1b group (40.3 [112.0]) but not statistically significantly different from the other groups (P = 0.495). These results are presented in Table 2.
Table 2.
Current use of medications and psychosocial therapy for mental health.
| Treatment use | Healthy Controls (n = 42) |
Stage 0b
(n = 41) |
Stage 1a (n = 52) |
Stage 1b (n = 108) |
P value (including all 4 groups)c | P value (including 3 at-risk groups)c |
|---|---|---|---|---|---|---|
| Current medication use | ||||||
| No | 42 (100.0%) | 35 (85.4%) | 35 (67.3%) | 71 (65.7%) | <0.001 | 0.021 |
| Yes | 0 | 6 (14.6%) | 17 (32.7%) | 37 (34.3%) | ||
| Types of medicationsa | ||||||
| Antidepressants | – | 1 (11.1%) | 14 (58.3%) | 23 (41.8%) | ||
| Antipsychotics | – | 2 (22.2%) | 3 (12.5%) | 10 (18.2%) | ||
| Anxiolytics | – | 0 | 1 (4.2%) | 4 (7.3%) | ||
| Stimulants (ADHD) | – | 3 (33.3%) | 6 (25.0%) | 14 (25.5%) | ||
| Non-stimulants (ADHD) | – | 3 (33.3%) | 0 | 4 (7.3%) | ||
| Mean duration of current medication use (months) | – | 47.1 (56.3) | 36.1 (39.9) | 35.8 (50.9) | – | 0.807 |
| Current Psychosocial therapy use | ||||||
| No | 41 (97.6%) | 37 (90.2%) | 38 (73.1%) | 55 (50.9%) | <0.001 | <0.001 |
| Yes | 1 (2.4%) | 4 (9.8%) | 14 (26.9%) | 53 (49.1%) | ||
| Types of psychosocial therapya | ||||||
| Case management | 0 | 2 (40.0%) | 4 (25.0%) | 23 (31.5%) | ||
| CBT | 0 | 0 | 0 | 7 (9.6%) | ||
| Family therapy | 0 | 2 (40.0%) | 2 (12.6%) | 3 (4.1%) | ||
| School counselling | 0 | 0 | 3 (18.7%) | 16 (21.9%) | ||
| Stress management | 0 | 0 | 3 (18.7%) | 1 (1.4%) | ||
| Supportive | 1 (100.0%) | 1 (20.0%) | 4 (25.0%) | 21 (28.8%) | ||
| Multiple/combined | 0 | 0 | 0 | 2 (7.4%) | ||
| Mean number of sessions | – | 15.0 (15.0) | 12.4 (11.6) | 40.3 (112.0)d | – | 0.495 |
Values presented as: n (%) or mean (SD).
a Specific types of medications and psychosocial therapies are not mutually exclusive; therefore, some participants may have multiple medications and/or doses of medications, or multiple psychosocial therapies.
b Participants are in one of the following situations: For antipsychotics, 3 participants are prescribed low-dose risperidone (2 are adopted twins with ADHD whose biological mother has schizophrenia; one takes risperidone for sleep while the other has schizophrenia); one participant is prescribed an anti-depressant due to one past episode of major depressive disorder that has been in remission for some time; 2 participants have ADHD (one prescribed a stimulant and the second prescribed one stimulant and one non-stimulant medication).
c X2 tests for dichotomous outcomes and ANOVA tests for continuous outcomes.
d This mean (SD) appears to be quite large due to 3 participants receiving many sessions of school counselling on an ongoing and weekly basis for several years. If these outliers are dropped from the analysis, the mean (SD) = 14.1 (17.1), with a P = 0.904.
Lifetime Treatment History
With respect to lifetime medication use, there was a significant difference among all groups (P < 0.001) and when excluding healthy controls (P = 0.001); see Table 3. Over half of stage 1b participants were prescribed medications at some point in their lifetime (53.7%) compared with stage 1a (42.3%), stage 0 (19.5%), and healthy controls (7.1%). The most commonly reported type of medication in stage 1a and 1b groups was antidepressants followed by stimulants.
Table 3.
Lifetime treatment history of medication use for mental health.
| Medication | Healthy Controls | Stage 0 | Stage 1a | Stage 1b | P value (including all 4 groups)b | P value (including 3 at-risk groups)b |
|---|---|---|---|---|---|---|
| Lifetime medication use | n = 42 | n = 41 | n = 52 | n = 108 | ||
| No | 39 (92.9%) | 33 (80.5%) | 30 (57.7%) | 50 (46.3%) | <0.001 | 0.001 |
| Yes | 3 (7.1%) | 8 (19.5%) | 22 (42.3%) | 58 (53.7%) | ||
| Types of medicationsa | ||||||
| Antidepressants | 1 (33.3%) | 3 (16.7%) | 15 (44.1%) | 40 (40.4%) | ||
| Antipsychotics | 0 | 3 (16.7%) | 4 (11.8%) | 15 (15.2%) | ||
| Anxiolytics | 1 (33.3%) | 0 | 1 (2.9%) | 7 (7.1%) | ||
| Mood stabilizers | 0 | 2 (11.1%) | 0 | 3 (3.0%) | ||
| Stimulants (ADHD) | 1 (33.3%) | 5 (27.8%) | 11 (32.4%) | 26 (26.3%) | ||
| Non-stimulants (ADHD) | 0 | 5 (27.8%) | 3 (8.8%) | 8 (8.1%) | ||
| Medication use in childhood (medication start date: 0 to 12 y) | n = 42 | n = 41 | n = 51 | n = 107 | ||
| No | 41 (97.6%) | 35 (85.4%) | 47 (90.4%) | 95 (88.8%) | 0.273 | 0.734 |
| Yes | 1 (2.4%) | 6 (14.6%) | 5 (9.6%) | 12 (11.2%) | ||
| Type of medicationsa | ||||||
| Antidepressants | 0 | 2 (18.2%) | 1 (20.0%) | 0 | ||
| Antipsychotics | 0 | 1 (9.1%) | 0 | 1 (9.1%) | ||
| Anxiolytics | 0 | 0 | 0 | 0 | ||
| Mood stabilizers | 0 | 1 (9.1%) | 0 | 0 | ||
| Stimulants (ADHD) | 1 100.0%) | 4 (36.4%) | 4 (80.0%) | 9 (81.8%) | ||
| Non-stimulants (ADHD) | 0 | 3 (27.3%) | 0 | 1 (9.1%) | ||
| Medication use in adolescence (medication start date: 13 to 17 y) | n = 42 | n = 41 | n = 52 | n = 108 | ||
| No | 41 (97.6%) | 36 (87.8%) | 38 (73.1%) | 68 (63.0%) | <0.001 | 0.019 |
| Yes | 1 (2.4%) | 5 (12.2%) | 14 (26.9%) | 40 (37.0%) | ||
| Type of medicationsa | ||||||
| Antidepressants | 1 (100.0%) | 1 (16.7%) | 8 (40.0%) | 27 (47.4%) | ||
| Antipsychotics | 0 | 1 (16.7%) | 3 (15.0%) | 7 (12.3%) | ||
| Anxiolytics | 0 | 0 | 1 (5.0%) | 2 (3.5%) | ||
| Mood stabilizers | 0 | 1 (16.7%) | 0 | 2 (3.5%) | ||
| Stimulants (ADHD) | 0 | 1 (16.7%) | 6 (30.0%) | 13 (22.8%) | ||
| Non-stimulants (ADHD) | 0 | 2 (33.3%) | 2 (10.0%) | 6 (10.5%) | ||
| Medication use in adulthood (medication start date: 18+ y) | n = 28 | n = 21 | n = 28 | n = 36 | ||
| No | 27 (96.4%) | 19 (90.5%) | 19 (67.9%) | 18 (50.0%) | <0.001 | 0.018 |
| Yes | 1 (3.6%) | 2 (9.5%) | 9 (32.1%) | 18 (50.0%) | ||
| Type of medicationsa | ||||||
| Antidepressants | 0 | 0 | 6 (66.7%) | 13 (43.3%) | ||
| Antipsychotics | 0 | 2 (100.0%) | 1 (11.1%) | 6 (20.0%) | ||
| Anxiolytics | 1 (100.0%) | 0 | 0 | 5 (16.7%) | ||
| Mood stabilizers | 0 | 0 | 0 | 1 (3.3%) | ||
| Stimulants (ADHD) | 0 | 0 | 1 (11.1%) | 4 (13.3%) | ||
| Non-stimulants (ADHD) | 0 | 0 | 1 (11.1%) | 1 (3.3%) |
Values presented as: n (%).
a Specific types of medications are not mutually exclusive, therefore, some participants may have multiple medications.
bχ2 tests for dichotomous outcomes.
When examining medication use in different life periods, the prevalence of starting medications in childhood (0 to 12 y) was not statistically significantly different among the groups (P = 0.273) nor when excluding healthy controls (P = 0.734). Overall, there were only 24 participants across all groups who reported starting a medication as a child. The highest recollection of medication use was reported in stage 1b participants (11.2%) and the most commonly recalled medications were stimulants. However, when examining those who started medications either in adolescence or in adulthood, there were significant differences among the groups (P < 0.001), and when excluding healthy controls (adolescence: P = 0.019; adulthood: P = 0.018). Adolescence was the life-period during which youth were most likely to start medications; although, this may be attributable to the age of our participants at enrollment, given that only 46.5% of participants were over 18 y of age at the baseline assessment. The percentage of participants starting medication in stage 1a and 1b in adulthood (50.0% and 32.1%) was numerically higher than that during adolescence (37.0% and 26.9%). Similar to current medication use, the most commonly reported medication class across groups was antidepressants; although, stimulants prescribed for ADHD were more commonly recalled in adolescence (stage 1a: 6 [30.0%]; stage 1b: 13 [22.8%]) than in adulthood (stage 1a: 1 [11.1%]; stage 1b: 4 [13.3%]).
Table 4 presents the differences in the prevalence of lifetime psychosocial therapy between all groups and when excluding healthy controls (P < 0.001). Specifically, 86.1% of stage 1b and 82.7% of stage 1a participants reported having had psychosocial therapy, compared with only 48.8% of stage 0 and 26.2% of healthy controls. The most commonly recalled psychosocial therapy in stage 0, stage 1a, and stage 1b groups was supportive therapy, whereas school counselling was most commonly recalled for healthy controls.
Table 4.
Lifetime treatment history of psychosocial therapy for mental health.
| Psychosocial therapy | Healthy Controls | Stage 0 | Stage 1a | Stage 1b | P value (including all 4 groups)b | P value (including 3 at-risk groups b |
|---|---|---|---|---|---|---|
| Lifetime psychosocial therapy | n = 42 | n = 41 | n = 52 | n = 108 | ||
| No | 31 (73.8%) | 21 (51.2%) | 9 (17.3%) | 15 (13.9%) | <0.001 | <0.001 |
| Yes | 11 (26.2%) | 20 (48.8%) | 43 (82.7%) | 93 (86.1%) | ||
| Types of psychosocial therapya | ||||||
| Blinded in a study | 0 | 0 | 0 | 2 (1.1%) | ||
| Case management | 1 (5.9%) | 3 (10.0%) | 12 (16.2%) | 45 (25.0%) | ||
| Cognitive behavioural therapy | 1 (5.9%) | 2 (6.7%) | 5 (6.8%) | 14 (7.8%) | ||
| Family therapy | 2 (11.8%) | 7 (23.3%) | 6 (8.1%) | 12 (6.7%) | ||
| Interpersonal | 0 | 0 | 1 (1.4%) | 0 | ||
| Psychodynamic | 0 | 0 | 1 (1.4%) | 1 (1.1%) | ||
| School counselling | 7 (41.2%) | 3 (10.0%) | 12 (16.2%) | 42 (23.3%) | ||
| Stress management | 0 | 2 (6.7%) | 7 (9.5%) | 6 (3.3%) | ||
| Supportive | 5 (29.4%) | 10 (33.3%) | 28 (37.8%) | 56 (31.1%) | ||
| Multiple/combined | 1 (5.9%) | 3 (10.0%) | 2 (2.7%) | 2 (1.1%) | ||
| Psychosocial therapy in childhood (therapy start date: 0 to 12 y) | n = 42 | n = 41 | n = 52 | n = 108 | ||
| No | 39 (92.9%) | 36 (87.8%) | 42 (80.8%) | 76 (70.4%) | 0.008 | 0.057 |
| Yes | 3 (7.1%) | 5 (12.2%) | 10 (19.2%) | 32 (29.6%) | ||
| Types of psychosocial therapya | ||||||
| Blinded in a study | 0 | 0 | 0 | 0 | ||
| Case management | 0 | 2 (33.3%) | 1 (11.1%) | 8 (21.6%) | ||
| Cognitive behavioural therapy | 0 | 1 (16.7%) | 0 | 1 (2.7%) | ||
| Family therapy | 1 (25.0%) | 0 | 0 | 3 (8.1%) | ||
| Interpersonal | 0 | 0 | 1 (11.1%) | 0 | ||
| Psychodynamic | 0 | 0 | 1 (11.1%) | 0 | ||
| School counselling | 1 (25.0%) | 0 | 4 (44.4%) | 9 (24.3%) | ||
| Stress management | 0 | 0 | 1 (11.1%) | 3 (8.1%) | ||
| Supportive | 2 (50.0%) | 2 (33.3%) | 1 (11.1%) | 12 (32.4%) | ||
| Multiple/combined | 0 | 1 (16.7%) | 0 | 1 (2.7%) | ||
| Psychosocial therapy in adolescence (therapy start date: 13 to 17 y) | n = 42 | n = 41 | n = 52 | n = 108 | ||
| No | 35 (83.3%) | 29 (70.7%) | 19 (36.5%) | 41 (38.0%) | <0.001 | 0.001 |
| Yes | 7 (16.7%) | 12 (29.3%) | 33 (63.5%) | 67 (62.0%) | ||
| Types of psychosocial therapya | ||||||
| Blinded in a study | 0 | 0 | 0 | 2 (1.8%) | ||
| Case management | 1 (9.1%) | 0 | 9 (20.5%) | 29 (26.6%) | ||
| Cognitive behavioural therapy | 1 (9.1%) | 0 | 1 (2.3%) | 5 (4.6%) | ||
| Family therapy | 1 (9.1%) | 5 (41.7%) | 5 (11.4%) | 9 (8.3%) | ||
| Interpersonal | 0 | 0 | 0 | 0 | ||
| Psychodynamic | 0 | 0 | 0 | 1 (0.9%) | ||
| School counselling | 4 (36.4%) | 3 (25.0%) | 5 (11.4%) | 29 (26.6%) | ||
| Stress management | 0 | 0 | 3 (6.8%) | 3 (2.8%) | ||
| Supportive | 3 (27.3%) | 4 (33.3%) | 19 (43.2%) | 30 (27.5%) | ||
| Multiple/combined | 1 (9.1%) | 0 | 2 (4.6%) | 1 (0.9%) | ||
| Psychosocial therapy in adulthood (therapy start date: 18+ y) | n = 28 | n = 21 | n = 28 | n = 36 | ||
| No | 25 (89.3%) | 12 (57.1%) | 14 (50.0%) | 12 (33.3%) | <0.001 | 0.172 |
| Yes | 3 (10.7%) | 9 (42.9%) | 14 (50.0%) | 25 (66.7%) | ||
| Types of psychosocial therapya | ||||||
| Blinded in a study | 0 | 0 | 0 | 0 | ||
| Case management | 0 | 2 (13.3%) | 2 (10.5%) | 11 (30.6%) | ||
| Cognitive behavioural therapy | 0 | 1 (6.7%) | 3 (15.8%) | 7 (19.4%) | ||
| Family therapy | 0 | 2 (13.3%) | 1 (5.3%) | 0 | ||
| Interpersonal | 0 | 0 | 0 | 0 | ||
| Psychodynamic | 0 | 0 | 0 | 0 | ||
| School counselling | 2 (66.7%) | 0 | 3 (15.8%) | 5 (13.9%) | ||
| Stress management | 0 | 2 (13.3%) | 3 (15.8%) | 0 | ||
| Supportive | 1 (33.3%) | 6 (40.0%) | 6 (31.6%) | 13 (36.1%) | ||
| Multiple/combined | 0 | 2 (13.3%) | 1 (5.3%) | 0 |
Values presented as: n (%).
a Specific types of medications are not mutually exclusive, therefore, some participants may have multiple psychosocial therapies.
bχ2 tests for dichotomous outcomes and ANOVA tests for continuous outcomes.
Unlike childhood medication use, childhood psychosocial therapy was different among the groups (P = 0.008) but no longer significant when excluding healthy controls (P = 0.057). The percentage of participants receiving psychosocial therapy in childhood was much lower than the lifetime percentages in all groups. Stage 1b participants had the highest prevalence (29.6%) of receiving psychosocial therapies, with supportive therapy (n = 12 [32.4%]) being the most common, followed by school counselling and case management. Receiving psychosocial therapy in adolescence and adulthood was more commonly reported for all participants, compared to childhood, except for healthy controls in adulthood (7.1% v. 10.7%). Over half of the participants in stage 1a and stage 1b received psychosocial therapy in adolescence (63.5% and 62.0%) and/or adulthood (50.0% and 66.7%). Comparable to childhood psychosocial therapy, the most common psychosocial therapies received in adolescence and adulthood were supportive therapy and case management. Not surprisingly, school counselling was reported more commonly in adolescence than adulthood.
Lifetime hospital visits for mental health were captured for all participants; see Table 5 for details. A total of 42 participants had visited a hospital for mental health concerns at some point in their lifetime, with statistically significant differences among the groups (P < 0.001) and when excluding healthy controls (P = 0.001). Less than 10% of healthy controls and stage 0 participants had a lifetime hospital visit related to mental health. In contrast, 13.5% of stage 1a and 29.6% of stage 1b participants visited the hospital for mental health concerns. Emergency room visits (n = 5 [71.4%] and n = 24 [75.0%] for stage 1a and stage 1b participants) were more common than inpatient stays (n = 2 [28.6%] and n = 8 [25.0%]). The mean duration of inpatient stay was similar between stage 1a participants (26.0 [7.1]) and stage 1b participants (24.6 [20.1]), with P = 0.677.
Table 5.
Lifetime hospital visits for mental health.
| Hospital visits | Healthy Controls (n = 42) |
Stage 0 (n = 41) |
Stage 1a (n = 52) |
Stage 1b (n = 108) |
P value (including all 4 groups)b | P value (including 3 at-risk groups)b |
|---|---|---|---|---|---|---|
| Mental health hospital visits | ||||||
| No | 41 (97.6%) | 39 (95.1%) | 45 (86.5%) | 76 (70.4%) | <0.001 | 0.001 |
| Yes | 1 (2.4%) | 2 (4.9%) | 7 (13.5%) | 32 (29.6%) | ||
| Types of hospital visitsa | ||||||
| Emergency room visits | 1 (100.0%) | 1 (50.0%) | 5 (71.4%) | 24 (75.0%) | ||
| Inpatient stays | 0 | 1 (50.0%) | 2 (28.6%) | 8 (25.0%) | ||
| Mean duration of inpatient stay (days) | – | – | 26.0 (7.1) | 24.6 (20.1) | – | 0.677 |
Values presented as: n (%) or means (SD).
a Specific types of healthcare resource use are not mutually exclusive; therefore, some participants may have multiple healthcare resource uses.
bχ2 tests for dichotomous outcomes and ANOVA tests for continuous outcomes.
Discussion
In the PROCAN study, we have utilized the clinical staging model of McGorry and colleagues7,8 to examine current and past treatment history of a sample of youth who may be considered at risk for SMI. Based on this model, we had 3 “at-risk- groups” and a group of healthy controls. There were differences among the groups for all types of mental health treatment histories. Less than one-third of participants had been prescribed medications at the time of assessment, with higher percentages in the symptom-related, at-risk groups (stage 1a and 1b) compared with individuals in stage 0. The most common medications prescribed were antidepressants and stimulants, suggesting that mood symptoms may be most common and that several of the participants may have a diagnosis of ADHD in addition to their presenting concerns. Even though no participant was experiencing psychotic symptoms—only subthreshold or attenuated psychotic symptoms—18.2% of stage 1b participants were prescribed antipsychotic medications at assessment.
With respect to psychosocial therapies, there were significant differences among the groups. Half of the participants in stage 1b were currently (at baseline) receiving psychosocial therapies; this was substantially lower in stage 1a, and even lower in stage 0 and healthy controls. The most common therapies were case management and supportive therapy and, for those in group 1b, school counselling. Interestingly, there was limited use of CBT. This fits with the notion that for groups 1a and 1b, these young people are help-seeking and not necessarily receiving formal treatment (i.e., CBT) that might best help them and their symptoms. Even though group 1b is more symptomatic, they had not—at the time of recruitment—received more sessions than any other group, possibly suggesting a need for or gap in treatment.
An examination of lifetime medication treatment history revealed that proportions of lifetime treatment were higher than current (at baseline) treatment, with almost half of the participants in stages 1a and 1b being prescribed medications. Antidepressants were the most common medication, with stimulants second. Notably, over 80% of stage 1a and 1b participants had received psychosocial therapies at one point in their lifetimes, compared to less than half (48.8%) in stage 0 and 26.2% in healthy controls. Again, the most common therapies were school counselling, supportive counselling and case management, with CBT being much less frequent.
Because we had comprehensive logs of all past lifetime treatment we could examine treatments that started in childhood v. adolescence and adulthood. Adulthood and adolescence were the most common times to begin treatment both for medications and for psychosocial therapies. Very few participants had been prescribed medication treatment in childhood and, for those who had, the most common medication was stimulants (most likely for ADHD). However, psychosocial therapies began more often in childhood compared to medications. But as with current therapy, psychosocial therapies over the lifetime mainly consisted of supportive therapy or school counselling.
Stage 1b participants had the highest rates of lifetime hospital visits compared with all other groups. This was not surprising given their treatment history relative to the other groups. As hypothesized, stage 1b participants were consistently more likely to have received treatment compared with all other groups, particularly with psychotropic medications. Stage 1a participants had less than half of the amount of lifetime hospital visits compared with stage 1b participants but much higher compared with stage 0 and healthy controls. Additionally, as hypothesized, stage 1a participants had consistently higher rates of treatment compared with stage 0 and healthy controls.
To our knowledge, no other study has examined the natural history of exposure to treatment in youth at specified stages of risk for SMI.7,8 Other work has focused on broad populations of youth12,15,16 or those specifically at CHR.11,17–19 The North American Prodrome Longitudinal Study-2 (NAPLS-2)17 reported that 82% of CHR participants had received some type of psychosocial treatment, with supportive therapy being the most common. These results are similar to the PROCAN study, as we found that 86% of stage 1b participants had received psychosocial therapy. It was reported that, when combining the NAPLS-1 and NAPLS-2 samples, the use of psychotropic medications was common in CHR youth, with antidepressants followed by antipsychotics most frequently prescribed.18 One other study investigated different stages of risk (stage 1a and stage 1b) in youth similar to our study but they did not report treatment history.19
Although debatable, it is possible that early intervention may reduce the risk of SMI.20 Less than half of our study participants categorized as being at-risk for SMI (stages 1a and 1b) were ever prescribed medication. Although higher percentages received psychosocial therapies, there are individuals at-risk who did not receive treatment. However, further research to explore the efficacy of treatment for at-risk populations is needed.
Strengths and limitations should be noted when interpreting our study results. This study includes a large sample of youth who may to varying degrees present with some risk for developing mental illness. We allocated them to 1 of 3 “at-risk” groups based on a current clinical model of staging. However, the validity of clinical stage models has not been established. There is no evidence that individuals who may develop SMI progress through the different stages. At best, it may be a useful way to group youth at-risk based on what may appear to be severity of symptoms. Second, participants in group 1b were younger than those in the other 3 groups; however, they had still received more treatment over time. Third, although data collection included lifetime treatment history, recalled information is subject to misreporting, specifically under-reporting, of treatments. Our results might underestimate the actual treatment history details for these participants. Fourth, the categorization of medications was broad. For example, there are multiple types of antidepressants and antipsychotics that could be categorized more specifically. Finally, because of the various methods of recruitment in the PROCAN study, the generalizability of the groups may be limited. Given the willingness of our participants to engage in a longitudinal research study, they might represent a more help-seeking, proactive group than is typical of youth with similar risks and symptoms. Alternatively, they might have represented a portion of youth who felt that they were not adequately accessing mental health services. These limitations may only be mitigated through a better understanding of youth and young adults at-risk for SMI.
This study provided descriptive information about mental health service utilization in youth at-risk for SMI. Those with attenuated syndromes had the highest prescribed medication and psychosocial therapy rates compared with those who were asymptomatic, those with distress disorders, and healthy controls. Consistent across all study groups was a large proportion of participants who did not seek healthcare resources. Whether receiving early treatment, in whatever form, attenuates the risk of illness progression remains to be confirmed in longitudinal studies.
Supplemental Material
Supplemental Material, Supplemental_Tables_792195 for Treatment History of Youth At-Risk for Serious Mental Illness by Megan S. Farris, Glenda MacQueen, Benjamin I. Goldstein, JianLi Wang, Sidney H. Kennedy, Signe Bray, Catherine Lebel, and Jean Addington in The Canadian Journal of Psychiatry
Acknowledgments
We acknowledge Catherine Marshall for her role in helping set up the study and collection of data for this paper.
Data Access
Data are not available at this time due to ongoing data collection.
Footnotes
Declaration of Conflicting Interests: The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: JA has received research support from National Institute of Mental Health, Brain Canada and the Mathison Centre at the University of Calgary. SHK has received funding or honoraria from the following sources: Abbott, Allergan, AstraZeneca, BMS, Brain Cells Inc., Brain Canada, Clera, CIHR, Eli Lilly, Janssen, Lundbeck, Lundbeck Institute, OMHF, Ontario Brain Institute, Otsuka, Pfizer, Servier, St. Jude Medical, Sunovion and Xian-Janssen. GM has been on advisory board or speaker for Allergen, Lundbeck, Lilly, Pfizer, Janssen. BG, JW, SB, CL, SH, and MF list no competing interests.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the Brain Canada Foundation and the Mathison Centre for Research & Education at the University of Calgary. The opinions, results and conclusions are those of the authors and no endorsement by either funding source is intended or should be inferred.
Supplemental Material: Supplemental material for this article is available online.
References
- 1. Kessler R, Amminger G, Aguilar-Gaxiola S, et al. Age of onset of mental disorders: a review of recent literature. Curr Opin Psychiatry. 2007;20(4):359–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Statistics Canada. Table 105-1101 Mental health profile, Canadian Community Health Survey - mental health (CCHS), by age group and sex, Canada and provinces Statistics Canada; 2013. [Google Scholar]
- 3. Patton G, Hetrick S, McGorry P. Service responses for youth onset mental disorders. Curr Opin Psychiatry. 2007;20(4):319–324. [DOI] [PubMed] [Google Scholar]
- 4. National Research Council (US) and Institute of Medicine (US) Committee on the Prevention of Mental Disorders and Substance Abuse Among Children, Youth, and Young Adults. Research advances and promising interventions. In O'Connell ME, Boat T, Warner KE, editors. Preventing mental, emotional, and behavioral disorders among young people: progress and possibilities Washington (DC): National Academies Press (US); 2009. [PubMed] [Google Scholar]
- 5. Malla A, Shah J, Iyer S, et al. Youth Mental Health Should Be a Top Priority for Health Care in Canada. Can J Psychiatry. 2018;63(4):216–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Costello E. Early detection and prevention of mental health problems: Developmental epidemiology and systems of support. J Clin Child Adolesc Psychol. 2016;45(6):710–717. [DOI] [PubMed] [Google Scholar]
- 7. McGorry P. Early clinical phenotypes, clinical staging, and strategic biomarker research: Building blocks for personalized psychiatry. Biol Psychiatry. 2013;74(6):394–395. [DOI] [PubMed] [Google Scholar]
- 8. Hickie I, Scott E, Hermens D, et al. Applying clinical staging to young people who present for mental health care. Early Interv Psychiatry. 2013;7(1):31–43. [DOI] [PubMed] [Google Scholar]
- 9. Min J, Lee C, Lee C. Mental health promotion and illness prevention: a challenge for psychiatrists. Psychiatry Investig. 2013;10(4):307–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Woods S, Carlson J, McGlashan T. DSM-5 and the ‘Psychosis Risk Syndrome’: The DSM-5 proposal is better than DSM-IV. Psychosis. 2010;2(3):187–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Cadenhead K, Addington J, Cannon T, et al. Treatment history in the psychosis prodrome: characteristics of the North American prodrome longitudinal study cohort. Early Interv Psychiatry. 2010;4(3):220–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Findlay L, Sunderland A. Professional and informal mental health support reported by Canadians aged 15 to 24. Health Rep. 2014;25(12):3–11. [PubMed] [Google Scholar]
- 13. Mental Health Commission of Canada. Making the Case for Investing in Mental Health in Canada Ottawa, ON: Mental Health Commission of Canada; 2013. [Google Scholar]
- 14. Addington J, Goldstein B, Wang JL, et al. Youth at-risk for serious mental illness: Methods of the PROCAN study. BMC Psychiatry. 2018;18(1):219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Catania L, Hetrick S, Newman L, et al. Prevention and early intervention for mental health problems in 0–25 year olds: are there evidence-based models of care? Adv Mental Health. 2011;10(1):6–19. [Google Scholar]
- 16. Beck C, Williams J, Wang J, et al. Psychotropic medication use in Canada. Can J Psychiatry. 2005;50(10):605–613. [DOI] [PubMed] [Google Scholar]
- 17. Woodberry K, Seidman L, Bryant C, et al. Treatment precedes positive symptoms in North American adolescent and young adult clinical high risk cohort. J Clin Child Adolesc Psychol. 2016;47(1):69–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Woods S, Addington J, Bearden C, et al. Psychotropic medication use in youth at high risk for psychosis: comparison of baseline data from two research cohorts 1998-2005 and 2008-2011. Schizophr Res. 2013;148(1-3):99–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Purcell R, Jorm A, Hickie I, et al. Demographic and clinical characteristics of young people seeking help at youth mental health services: Baseline findings of the Transitions Study. Early Interv Psychiatry. 2015;9(6):487–497. [DOI] [PubMed] [Google Scholar]
- 20. Watsford C, Rickwood D, Vanags T. Exploring young people’s expectations of a youth mental health care service. Early Interv Psychiatry. 2013;7(2):131–137. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Material, Supplemental_Tables_792195 for Treatment History of Youth At-Risk for Serious Mental Illness by Megan S. Farris, Glenda MacQueen, Benjamin I. Goldstein, JianLi Wang, Sidney H. Kennedy, Signe Bray, Catherine Lebel, and Jean Addington in The Canadian Journal of Psychiatry