Table 1 Study characteristics

Ayton et al. (2008) [37], England

2003-06

Cohort study

All (N = 50) children and adolescents admitted to a specialist hospital in England during the study period

Eating disorder

32% of all patients were detained. Significantly more detained patients had depression at admission, reported self-harm or suicidal behaviour, and had more than one hospital admission in the past. Detained patients had younger age of onset of eating disorder, been ill for longer, higher weight (for height ratio) due to being transferred from another hospital following treatment to restore weight, and more required Naso gastric feeding. Overall, detained patients had better outcomes upon discharge.

Chaplin et al. (2015) [38], England

Not stated

Cohort study

151 children and adolescents (6–17 years) admitted to general adult mental health units and intellectual disability specialist units in England.

Intellectual disability

17% of patients were detained. A higher proportion of patients with intellectual disability (21%) were detained compared to those without an intellectual disability (16%) (not significant).

Christy et al. (2006) [50], USA

2000-03

Cohort study

36,551 children and young (2–17 years) who were examined at any Baker Act Receiving Facilities in Florida.

All patients

Mean age was 14.4 years; 48.3% were girls and 47.7% boys; 62.6% were Caucasian, 20.0% Black/African-American, 6.6% Hispanic and 0.3% Asian. Most examinations were initiated due to harm (85.9%), followed by 7.3% for self-neglect and 4.1% for both neglect and harm. Perceived risk of harm was primarily towards self (48.9%).

Clausen et al. (2018) [34], Denmark

2000-13

Cohort study

All 1,953 patients with anorexia nervosa aged 10–17 years admitted for treatment in Denmark (all ages N = 5,767)

Eating disorder

Of all patients aged 10–17 years, 36.2% had an episode of involuntary treatment (admission or detention). Of those who were detained, 90.4% of patients aged 10–14 years and 71.7% of patients age 15–17 years had a registered current registered eating disorder. Analyses of predictors did not stratify by age groups.

Corrigall & Bhugra (2010) [39], England

2001-10

Cohort study

435 adolescents (12–17 years) admitted to an adolescent inpatient psychiatric service in south London

All patients

Mean age of patients was 16.3 years; 49% were Black, 32% White, 3% Asian, and 15% other ethnicity; 53% were girls. Overall, 36% of the study sample had been subject to the Mental Health Act at some point during the admission. Black adolescents with psychosis had higher odds of being detained (OR = 3.0, 95% CI: 1.3–6.7) and other ethnicities (OR = 3.1, 95% CI: 3.1–1.1), compared to their white counterparts.

Deolmi et al. (2021) [48], Italy

2013-15

Cohort study

51 adolescents admitted to Parma Local Health Unit general psychiatric wards

All patients

A total of 21.6% (11 patients) were detained, which varied across the years. The highest number of cases (n = 6) were for conduct disorder, though no statistical analysis was conducted comparing detained patients to voluntary patients.

Ellila et al. (2008) [19], Finland

2000

Cross-sectional

278 children and adolescents (12–17 years) admitted to psychiatric inpatient wards in Finland on a given day in January

All patients

29.5% patients were detained. The largest age group of detained patients was 16–17 years (48%) and 59% were girls. Significantly more detained patients had psychotic disorder (62% vs. 15%, p < 0.001) and substance use disorder (9% vs. 1%, p = 0.005) than voluntarily admitted patients.

Eswaravel and O’Brien (2018) [40], England

2011-16

Cohort study

85 children and adolescents under the age of 18 years to a mental health trust in south-west London

All patients

Mean age was 15.7 years; 60% were female; 78.7% White, 8.2% were Black, 8.2% Asian and 4.6% Mixed ethnic groups; 16.5% had more than one admission over the period. The most common reason for detention was attempted suicide or deliberate self-harm (56.7%), 62.4% had a recorded history of self-harm, 64.4% were already known to CAMHS. Of those admitted and assessed, most were discharged home (67.3%) while 20.2% were detained under s.2 or s.3 of the Mental Health Act and 12.5% were voluntarily admitted.

Geng et al. (2020) [59], China

2019

Cohort study

196 adolescents discharged from 41 tertiary psychiatric hospitals in 29 provinces of mainland China between 19–31 March 2019

All patients

32.1% were detentions. Detained patients were slightly older, had significantly lower global assessment of functioning (GAF) scores, admitted with psychotic symptoms or aggressive behaviour but were less likely to have depressive symptoms. There was no significant difference in suicidal or self-harming behaviour during the admission. Detained patients had longer length of stay and were more likely to be diagnosed with schizophrenia but less likely to be diagnosed with depressive disorder.

Greenham and Persi (2013) [55], Canada

2009-10

Cross-sectional

Information about admissions received from 25 out of 27 inpatient services in Ontario, Canada.

NA

Detention for psychiatric assessment was reported by 21 units, on average 40% were involuntary (ranging from 0–80%). Admission for treatment was reported by 15 units, on average 5% were involuntary (range: 0–40%).

Hanssen-Bauer et al. (2011) [32], Norway

2005

Cohort study

192 children and adolescents (10–18 years) with admitted to 4 units in Norway (out of a total of 16 units) with a first episode of care starting in 2005. Only included patients admitted within 7 days of referral

All patients

Admission status for compulsory vs. voluntary status was only relevant for patients aged ≥ 16 years^{Footnote 1}. Of these 33.3% were involuntary, ranging from 7–67% between units (p < 0.001). Those detained had a higher Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA) score (mean = 21.0, SD = 6.1 vs. mean = 16.8, SD = 6.3, p < 0.001).

Jaworowski & Zabow (1995) [58], Israel

1991-92

Cohort study

78 children and adolescents (< 18 years) admitted for psychiatric care in Beersheva, Israel

All patients

18% of patients were detained. The most common diagnoses related to detention were conduct disorder (38%) and personality disorder (31%), of which the majority were borderline personality disorder (62.5%).

Jendreyshak et al. (2013) [46], Germany

2004-09

Cohort study

10,547 minors (< 18 years) treated in inpatient psychiatric services across 27 administrative districts in Germany.

All patients

29.2% of admissions were involuntary, most of whom were aged 14–15 years (39.7%) and 15–16 (37.3%). The proportion of involuntary admissions decreased over time, from 32.4% in 2004 to 25.7% in 2009, while the length of stay of detentions increased. The highest odds for being detained was for suffering from “mental retardation” (OR = 15.74), other predictors included being adolescent, having substance abuse problems, psychotic disorders, and an admission during duty hours (odds ratios > 3).

Kaltiala-Heino (2010) [20], Finland

2004-06

Cohort study

187 adolescent aged 11–17 years admitted to Tampere University Hospital in Southern Finland

All patients

49.7% of young people were involuntarily referred and 22.5% were involuntarily treated. Mean age was 15 years, 64.2% were girls, 17.1% lived in child welfare institutions and 5.7% lived in foster care. Those admitted involuntarily were more likely to be referred from primary care, from non-psychiatric specialties to lesser extent had received community-based treatment in the past than those admitted voluntarily. Patients who were both involuntarily referred and treated were more likely to have psychotic symptoms, temper tantrums, and were breaking property compared to voluntary patients. Those involuntarily referred were more likely to have violent behaviour, but this was not the case for involuntarily treated. Involuntarily treated, but not referred, were less likely to have depression compared to voluntarily treated patients.

Kaltiala-Heino and Frojd (2007) [21], Finland

2003

Qualitative interview study

44 child and adolescent psychiatrists and psychiatrists in training across 8/21 child and adolescent psychiatric departments in Finland

NA

Psychiatrists did not believe that difficulties in defining severity of mental disorder could be justified using ICD or DSM diagnoses. Acute severity (deemed as presenting as a risk of harm to self or others or loss of life) was differentiated from chronic (leading to regression or impact on development. The criteria for severe mental disorder must be justified alongside the risk that the minor deteriorates unless committed involuntarily, a risk of harm to self or others, and that voluntary treatment is inadequate.

Kaltiala-Heino et al. (2004) [22], Finland

1996–2000

Cohort study

Involuntary admissions of children aged 12–17 in all healthcare settings providing inpatient treatment in Finland

All patients

4.8% of admissions in children (< 12 years) and 22% of adolescent admissions (12–17 years) were involuntary. There was no difference between girls and boys. Involuntary admissions were more likely to be for substance use-related disorders, schizophrenia spectrum disorders, or mood disorders. Involuntary psychiatric admissions increased over time from 2.4 per 100,000 in 1995 to 7.2 in 2000.

Khenessi et al. (2004) [23], Finland

1994–2002

Cohort study

106 adolescents (13–18 years) sent for involuntary treatment at an adolescent psychiatric unit in south-west Finland

All patients

88% were admitted for psychiatric observation while 11% were released from involuntary treatment at admission. Among patients who were admitted for involuntary treatment after observation, more had psychotic symptoms compared to those who were released (41.0% vs. 19.4%, p = 0.016).

Kilgus et al. (1995) [51], USA

1988

Cohort study

All 352 psychiatric inpatient admissions (12–18 years) to a state hospital facility in South Carolina, US

All patients

Most admissions were White (71%) and the remaining 29% were African American. Mean age of all admissions was 15.4 years and 55.4% were male. Overall, 78.1% of admissions were detentions, which was significantly higher among African Americans (87.1%) than Whites (74.5%) (p = 0.01) with the odds of detention among African American being 2.05 (p = 0.043).

Kronström et al. (2016) [24], Finland

2000,2011

Cross-sectional

916 children and adolescents (< 1 years) admitted on a given day to 64/69 wards (2000) and 74/79 inpatient psychiatric wards in Finland.

All patients

Data also reported on in [19]. The proportion of admissions that were involuntary was 18% in 2000 and 19% in 2011. The proportion among admissions in children’s wards was 1% in 2000 and 4% in 2011 while the proportion in adolescent wards was 34% in 2000 and 31% in 2011. None of these differences were statistically significant. Due to the aim of the study, no exploration of characteristics were made within this group of patients.

Kronström et al. (2021) [31], Finland

2000, 2011, 2018

Cross-sectional

1276 inpatients in Finland (93–95% response rate)

All patients

Data also reported on in [24]. The proportion of detained patients remained stable - in 2000 a total of 18% were detained, in 2011 19% and in 2018 22% of patients were detained.

Lindsey et al. (2010) [52], USA

2001-02

Cohort study

1,450 African American minors (< 18 years, total sample also including 18–22 years) presenting at one of the five crisis response centres in Philadelphia, USA.

African American patients

Among those aged under 18 years, 25.6% of those arriving at the psychiatric emergency services (PES) had involuntary status, which was higher among those aged 13–17 years than those 12 years or younger (32.4% vs. 16.3%). The study explored a subsample of minors for which there was an official petition of civil commitment; 445 of these 501 minors were under 18 years. Within this group the commitment decision was involuntary for 59.3%, voluntary for 16.2% while 24.5% had their case dismissed. Subsequent analyses on predictors combined all age groups, including those over 18.

Mears et al. (2003) [41], England and Wales

Cross-sectional

51 of the 76 inpatient CAMHS consultants in England and Wales

NA

63% of respondents had undertaken at least one day’s training in mental health law in the last two years. 37% felt fully up-to-date with law changes relating to children and adolescents and 57% were partially up-to-date. 90% felt that their access to legal advice was at least adequate. 74% either agreed or strongly agreed that guidance on when to use which act was needed, 88% felt that more training on legal issues was needed. Correct responses for criteria for using the Mental Health Act was 68% and 45% for the Children Act. The mean correct response rate around consent relating to children and young people was 77%. Consultants who used the Mental Health Act at least once every six months had significantly higher correct response rate than those who used it less often 93.1 vs. 2.4, p < 0.05).

Mears and Worral (2001) [43], England and Wales

2001

Cross-sectional

258 (54% response rate) psychiatrists working in England and Wales

NA

The most common theme was choosing between the Mental Health Act and the Children Act when detaining an adolescent. Other themes related to issues around consent for treatment, social services, and stigma associated with being detained under the Mental Health Act.

Mears et al. (2003) [42], England and Wales

1999

Cross-sectional

663 children and young people (age not clearly stated) inpatients across 71 units in CAMHS units in England and Wales on a given census day

All patients

19% were formally admitted and almost all of these were under sections of the Mental Health Act (n = 119). The proportion of detained patients varied by type of unit (100% in forensic and secure units, 85% in learning disability, 9% in general psychiatry). Detentions were significantly higher among > 16s than < 16s (35% vs. 8%, p < 0.01), among boys than girls (23% vs. 16%, p < 0.05), among patients with schizophrenia (45% vs. 10%) and personality disorder (16% vs. 3%). Detained patients had significantly higher levels of reported youth had a history of sexual abuse, physical abuse, emotional abuse, multiple self-harming, and requiring one-to-one observation.

Nicholls et al. (1996) [44], England

1983-94

Cohort study

492 young people aged 12–17 years admitted to an inpatient unit for young people in the West Midlands, England

All patients

Among detained patients, 63.6% were male, mode age was 16 years. 6.7% of admissions were detentions at some point during the hospital stay, of these 42.2% were admitted under the Mental Health Act. There was a higher proportion of detentions in later years; 0–9% in 1983-90 compared to 11–27% in 1991-94.

Nyttingenes et al. (2018) [33], Norway

2015

Cross-sectional

96 admitted patients (13–17 years) to 10 (out 16 wards) acute and combined psychiatric wards in Norway.

All patients

18.8% of the sample were detained, which was higher among those aged 16–17 years (22.2%) compared to 13–15 years (12.1%). For voluntary admitted patients, informal pressure from parents was associated with higher perceived coercion whereas for detained patients more informal pressure from their parents was associated with a lower perception of coercion.

Park et al. (2011) [60], New Zealand

2002-07

Cohort study

332 children and adolescents < 18 years admitted to general inpatient psychiatric unit in Hamilton, Auckland

All patients

61.4% were detained (“involved the Mental Health Act”). Significantly higher proportion detained patients were boys (74.2% vs. 49.7%, p = 0.000), Maori were (compared to Caucasian youth; 68.2% vs. 57.1%, p = 0.04), and due to “deterioration of mental state” (79.7%) or aggression (75.0%). Those admitted under the Mental Health Act had longer length of stay than voluntary admissions (11.23 days vs. 3.75 days).

Pelto-Piri et al. (2016) [35], Sweden

2002-03

Cohort study

142 young people aged 10–18 years across all 16 child and adolescent clinics in Sweden who used coercive care

All patients

Median age was 16 years, 64.1% were girls, 9.2% were asylum seekers. Most common diagnoses were eating disorders, psychosis, depression, and neuropsychiatric disorders. 21.1% also had substance abuse. The most recorded reason for coercive care was the ‘protection’ argument^{Footnote 2} in 96% of Psychiatric Care Certificates and 99% of complete medical records, followed by ‘treatment requirement’ in 69% of complete medical records and 56% of Psychiatric Care Certificates, and ‘parental support’ on 48% of medical records and 24% of Psychiatric Care Certificates.

Persi et al. (2016) [55], Canada

2007-08

Cohort study

225 discharges of children and adolescents (5–17 years) in 26 acute hospitals in Ontario, Canada.

All patients

80% of admissions were detentions. A higher proportion of detained patients were adolescents (89% vs. 73%, p < 0.05), not living with family (27% vs. 4%, p < 0.05), and at risk of suicide (89% vs. 71%, p < 0.05). Of detained patients considered at risk of suicide at referral, 45% were considered risk of suicide at the psychiatric assessment (p < 0.05) compared to 66% of voluntary patients (p < 0.05). Length of stay was shorter for detained compared to voluntary patients (Mdn = 6 vs. Mdn = 6, p < 0.05). Among those referred on detained status, only 13% remained detained after psychiatric inpatient assessment.

Ramel et al. (2015) [36], Sweden

2011

Cohort study

261 children and adolescents receiving psychiatric care from Child and Adolescent Psychiatry emergency unit in Malmö, Sweden.

Unaccompanied minors

10.7% were detentions, which was significantly higher among unaccompanied refugees than other patients (19.6% vs. 8.3%, p = 0.024). There was a higher proportion of detentions of boys (71.4%), and among unaccompanied refugee minors all detentions were of boys compared to 56.2% of accompanied minors.

Rice et al. (2021) [53], USA

2017

Qualitative interview study

25 children and adolescents (13–17 years) admitted for suicidality to a CAMHS unit in a psychiatric hospital in a Southeastern State, USA.

Suicidal patients

The young people felt stigmatised both before and after arriving at the hospital for involuntary inpatient treatment. Many felt disregarded and dehumanised during the admission, leading to increased sense of stigma. Receiving and providing support from other young people in the hospital was an important part of the admission experience to not feel alone. Time away from things like social media, friends and family gave the young people an opportunity to engage with practices to cope with stressors which had positive outcomes such as reduction in stress levels.

Siponen et al. (2007) [25], Finland

1996–2003

Cohort study

9865 admissions of adolescents (12–17 years) to psychiatric hospitals across Finland.

All patients

23.6% were detained, which increased from 16.2% in 1996 to 26.3% in 2003 (a 1.6-fold increase). Between regions the proportion ranged from 4–32% of all admissions. Across all years, the rate of detention was 22.4 per 100,000 (95% CI: 20.94–23.92), which ranged between regions from as low as 5.06 (95% CI: 1.63–15.70) to 36.67 (95% CI: 28.70–46.84) per 100,000. There was a positive correlation between standardised rate of detentions and child welfare placements (correlation coefficient = 0.44, p = 0.048).

Siponen et al. (2011) [26], Finland

1996–2003

Cohort study

520 adolescents aged 13–17 in two hospital districts in Finland (above and below average rate of involuntary admission and detention)

All patients

Data also reported on in [25]. In the area with above average rate of involuntary admissions was 8.6 per 1,000 and detention (see Supplementary Table 2 for definition) was 5.8 per 1,000. For the below average area the rate was 3.9 and 1.9 per 1,000, respectively. Overall use of compulsory care differed significantly between the two regions (8.8 vs. 3.9, p < 0.0001). The above average district had a significantly higher prevalence of diagnoses of schizophrenia and personality disorders among detained patients (personality disorder was also higher among voluntary patients), significantly lower employment rate, rate of further education, migration to and from the area, higher number of divorces and single parent families, higher youth and overall crime rate, exclusion, individuals in detoxification treatment, patients in A-clinics, and mental health service use. The above average district had more outpatient service positions, adolescent psychiatric positions, more private and public welfare institutions for children, non-institutional welfare support, but significantly less adolescent psychiatry outpatient visits (119.2 per 1,000 below average district vs. 30.2 in above average district, p < 0.0001).

Siponen et al. (2012) [27], Finland

1996–2003

Cohort study

9,865 admissions to inpatient psychiatric treatment aged 12–17 years across Finland

All patients

Involuntary treatment increased from 14.4% in 1996 to 21.4% in 2003. Among all involuntary treatments during the study period, coercive measures (“seclusion, restraint, involuntary i.m. medication and physical holding”, p. 1403) were used in 27%. The most common diagnoses for involuntary treatment episodes were mood disorders (28.6%), conduct disorders (26.8%), and schizophrenia group disorders (20.1%). Coercive treatment within those treated involuntarily was higher among girls than boys (29.5% vs. 23.6%, p = 0.005) but there was no difference between younger (12–14 years) and older (15–17 years) adolescents.

Smith et al. (2004) [56], Canada

1998–2003

Cohort study

All patients < 16 years (total number not reported) admitted to a paediatric ward to one regional hospital in Ontario, Canada.

Paediatric patients

8.9% of admissions were involuntary; in 1998-99, 0 of the 25 admissions were done using involuntary measure compared to 11 out of 45 (24.4%) in 2002-03. Of the total 15 involuntary admissions across all years, 40% were for suicidal behaviour, 33% for behaviour disturbance, 13.3% for mood disorder, 7% (1 admission) for mood disorder and suicidal behaviour, and 7% (1 admission) for psychosis.

So et al. (2021) [47], Netherlands

2008-17

Cohort study

All 227 emergency admissions of children and young people (6–18 years) following outpatient emergencies in Amsterdam and Greater Rotterdam, the Netherlands

All patients

39.6% of admissions were compulsory. Regression analyses found significant association between being compulsory admitted and prior compulsory emergency admission (OR = 10.48, 95% CI: 2.44–45.09), severe or moderate suicide risk score (OR = 4.10, 95% CI: 1.63–10.30), being a danger to others (OR = 2.82, 95% CI: 1.00-7.96), lack of motivation for treatment (OR = 22.77, 95% CI: 8.48–61.14), lack of compliance with medication (OR = 4.31, 95% CI: 1.75–10.61), and all DSM disorders aside from relational and adjustment disorders (OR = 40.41, 95% CI: 1.12-1,458.79).

Sourander & Turunen (1999) [28], Finland

1990 and 1993

Cohort study

All 1,776 adolescents aged < 18 years discharged from discharged from psychiatric treatment in Finland

All patients

Data also reported on in [29]. 8.3% of patients were treated under compulsory measures in 1990, compared to 6.5% in 1993. The prevalence and incidence of compulsory care per 10,000 population increased from 7.2 to 8.2 (prevalence) and from 4.7 to 5.9 (incidence) from 1990 to 1993. The rate was highest in the 12-17-year group and among boys.

Sourander et al. (1998) [29], Finland

1990 and 1993

Cohort study

All 1,014 children and adolescents aged 12–17 years discharged from psychiatric treatment in Finland

All patients

The number and proportion of compulsory treatment of minors was 65/462 (14%) in 1990 and 62/552 (11%) in 1993. Most (> 90%) were aged 16–17 years, 51.2% were male (57% in 1990 and 45% in 1993), and most were admitted to adult wards (71.1%). The most common diagnosis was for psychotic disorder, which was a significant predictor for involuntary treatment (OR = 4.68, 95% CI: 2.69–6.81). Other predictors included being 15–17 years (OR = 1.48, 95% CI: 1.24–1.76) and being admitted to an adult ward (OR = 3.14, 95% CI: 1.99–4.94). The prevalence of compulsory care ranged from 0 to 3.4 per 10,000 minors. The proportion of compulsory treated minors, of all admissions, ranged from 0–21%.

Stein et al. (1988) [57], Canada

1977-84

Cohort study

All 294 discharged patients (no age provided) from an adolescent psychiatric unit it Ontario, Canada.

Suicidal patients

Among all discharges, 25 patients (8.5%) were detained at some point during the admission of whom 23 could be followed up. Of those detained, 52% were boys, mean age was 16.5 years for boys and 17.0 years for girls. Five (20%) of the detained sample had died by suicide during the follow up period (approximately five years). The diagnoses of those who were detained were (n = 11), personality disorder (n = 5), major affective disorder (n = 4), and other diagnosis (n = 3).

Tolmac & Hodes (2004) [45], England

2001

Cross sectional

113 adolescents (13–17 years) across adolescent psychiatric units and in-patient psychiatric wards in Greater London, focusing on a subsample of 55 adolescent with psychotic disorders

All patients

The mean age among all admitted young people was 16.2 years. Among adolescents with psychotic disorder 70% were male, 45% were White, 35% were Black, 13% were Asian, and 7% were of Other ethnicity. Overall, 62% of adolescents were subject to the Mental Health Act 1983 at some point during admission. There was no significant difference in detention at any point, but Black adolescents were more likely to be subject to the Act upon admission (63% vs. 16% White, p < 0.03).

Turunen et al. (2010) [30], Finland

2003

Qualitative interview study

44 child and adolescent psychiatrists and psychiatrists in training across 8 of the 21 child and adolescent psychiatric departments in Finland

NA

Data also reported on in [21]. Psychiatrists in general believed that detaining minors under different criteria to adults was appropriate. Detaining minors, who due to their age may not be able to weigh up the benefits of treatment, was seen as a possible early intervention to prevent future mental ill health and the criteria was seen as appropriately broad for this age group, but possibly too narrow for adults. Lack of definition of severe mental disorder was seen as potentially leading to different application of the law across the country.

Voultous et al. (2020) [49], Greece

2005-14

Cohort study

131 involuntary admitted minors in Thessaloniki, Greece.

All patients

Mean age was 14.19 years and 61% were boys. Of all involuntary admitted minors, 69.7% were discharged to go to their home while the remaining patients primarily were transferred to an institution. Among all patients, 48.9% had behavioural disorder and impulsive behaviour, 21% had pervasive developmental disorder, 14.5% had no diagnosis and 10.7% had intellectual disability. A smaller proportion had schizophrenia (4.6%), drug or alcohol abuse (3.1%) and personality disorder (2.3%). 7.6% of patients were re-hospitalised (follow-up period unclear)