Introduction

Detaining an individual against their will for assessment and/or treatment for mental illness (hereafter ‘detention’) raises important human rights concerns [1,2,3]. Detention under mental health legislation requires certain criteria need to be fulfilled, which vary between countries [4]. For children and adolescents this is even more complex, as parents may be able to consent to detention on behalf of the child, depending on factors which may include the child’s age or level of understanding. This could result in children being effectively detained without this being recorded as a mental health detention. In recent years, the use of detention has received increased attention from clinicians, policy makers, and academics due to legislative reform. In the UK, the Mental Health Act in England and Wales is undergoing reform, with conclusions of extensive consultation published in 2018 [5] and a draft Mental Health Bill published in 2022 [6]. In Scotland, a final report of the review of mental health and incapacity legislation was published in 2022 [7]. Reform of frameworks outlining criteria for detention is important, as the number of detentions has increased over time [8] within a wider context of increased burden of disease from mental and substance use disorders globally [9].

Domestic legislation is best viewed in the light of international human rights instruments. The UN Convention on the Rights of the Child (UNCRC) [10] and the UN Convention on the Rights of Persons with Disabilities (CRPD) [11] require strong justification for detention, which should take account of the Convention principles. This includes respect for the evolving capacities of children, that detention cannot be arbitrary, must be objectively justified, and that the basis of detention should be non-discriminatory. The UN Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest Attainable Standard of Physical and Mental Health issued a report in 2017 calling for an end to the institutionalisation of children, and to “take targeted, concrete measures to radically reduce medical coercion and facilitate the move towards an end to all forced psychiatric treatment and confinement” (p.21) [12]. At the very least, any evidence of increased detention and involuntary treatment of children would need extremely strong justification, in the light of this demand. The UN Committee on the Rights of Persons with Disabilities has repeatedly stated that it views current forms of detention as breaching the CRPD, stating that “involuntary commitment of persons with disabilities on health-care grounds contradicts the absolute ban on deprivation of liberty on the basis of impairment (art. 14 [1] (b))” and “the principle of free and informed consent of the person concerned for health care (art. 25)” [13].

In adults, a global systematic review found that the risk of being detained is more than double for those with experience of previous episodes of detention and those with a psychotic disorder [14]. Similarly, a separate review found that individuals from ethnic minority groups are at increased risk of mental health detention [15], compared to their white counterparts, which raises questions on how mental health care can adequately support individuals from all ethnic backgrounds, without discrimination. A systematic review which assessed factors associated with detention of children and adolescents found that also in this group psychotic disorder was associated with higher odds of detention [16]. Other factors included substance misuse, having an intellectual disability, being at risk of harming oneself or others, and being older than 12 years of age. This systematic review also found differences between ethnic groups, but this was only true for Black adolescents compared to their white counterparts. The authors however noted that only a small number of included studies reported on ethnicity [16].

While a previous review have addressed factors influencing the likelihood of children and adolescents of being detained, no review to date has synthesised how legislation is used and any views from children and adolescents themselves, or relevant practitioners, on detention of these age groups.

We undertook a scoping review of the evidence relating to mental health detentions of children and adolescents (aged < 18 years). We aimed to examine/document:

  • the incidence of detentions in relation to overall admissions, including changes in detention rates over time;

  • the clinical, sociodemographic, and behavioural factors associated with detention (compared to voluntary admission/treatment);

  • the views of professionals regarding detaining children and adolescents under mental health legislation; and

  • the views of children or adolescents on being detained.

Methods

The review followed the guidance on scoping reviews set out by the Joanna Briggs Institute (JBI) [9], and the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist [17]. No protocol was registered.

Inclusion and exclusion criteria

Primary research studies published in peer-reviewed journals in English language that included children and adolescents aged < 18 years (up until, but not including, their 18th birthday), health professionals, parents, or carers were eligible for inclusion. It is worth noting that definitions of adolescence is an ongoing debate [18], however this review focused on < 18 years as the UNCRC definition of a child. Qualitative, quantitative, or mixed methods study designs were included, regardless of the type of order (emergency, short-term etc.). Any study of psychiatric care of children and adolescents where compulsory care was part of the study, and for which data could be extracted, was eligible for inclusion. Systematic reviews, conference abstracts, editorials, book chapters and sources published in languages other than English were excluded. Studies involving detention of adults or studies including adults, and children and adolescents where results for those aged < 18 years could not be distinguished from those of adults, and studies that exclusively focused on voluntary psychiatric care, were not eligible for inclusion. Detentions in the criminal justice system, detentions relating to immigration status, detentions within general child protection law, and placements in educational institutions for children with special needs education were also excluded.

Search strategy

We searched PsycINFO, MEDLINE, and Embase for peer-reviewed journal articles published from database inception to 21 September 2022 that included empirical data and were published in English. We used a combination of search terms related to children and adolescents, mental illness and treatment, mental health legislation, and detention (Supplementary Table 1). The search strategy was developed with assistance from a subject librarian at The University of Edinburgh. Database searches were undertaken by ZT, and titles and abstracts were screened by ZT and LS using Rayyan [14]. Any disagreements were resolved through discussion. Following title and abstract screening, full text screening was undertaken by ZT and checked by LS. Forward citation searches and manual reference searches were undertaken by LS.

Data extraction and synthesis

Data extraction was performed by ZT and LS and LS subsequently summarised the findings and drafted the manuscript. The key characteristics of studies were extracted into an excel spreadsheet using a combination of a pre-determined proforma (country, data collection period, study design, sample, patient group, and key findings) and iterative extracting additional information (e.g. predictors and type of detention). Criteria for detention, although not an original aim of the review, was extracted from the paper or, where unavailable, from papers that described the legal criteria in that country and denoted as ‘not defined’ where no information could be found. Once we started extracting data we decided to extract information about the type of detention as well, and add it to the original focus of the review, as we deemed it relevant to comment on how studies were not necessarily reporting on the same type of detention. This was relevant to studies that described proportion of detentions, rates, and experience of detention. Descriptive statistics were calculated for quantitative studies that reported on proportion of detentions to all psychiatric admissions and all key findings were summarised using a thematic approach.

Results

Study characteristics

Following title and abstract screening and citation linking a total of 165 full text articles were reviewed resulting in 42 articles included in the review (Fig. 1). Notably, none of the articles excluded at full-text due to foreign language or being unavailable were from countries outside Europe or the USA. Most of the included studies were quantitative (n = 39; 93%) and three were qualitative. Study characteristics are presented in Table 1. Notably, 18/39 studies were from Scandinavian countries (13 from Finland) [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36] with the remaining studies from other European countries [37,38,39,40,41,42,43,44,45,46,47,48,49], North America [50,51,52,53,54,55,56,57], Israel [58], China [59], and New Zealand [60].

Fig. 1
figure 1

Study selection flow chart

Full size image
Table 1 Study characteristics
Full size table

In 36 studies the study samples, or part of an overall sample, were children and adolescents aged 5–17 years, in four the study sample were psychiatrists, and in one study the data related to services. Data were collected over a period ranging from one day to 11 years. In the 39 quantitative studies, 13 covered a national sample, 14 service level, nine regional level, and three city level.

Legal frameworks for detaining children and adolescents

The legal criteria and types of detentions were mapped out to better understand differences between studies and countries (Supplementary Table 2). Criteria consistently included the presence of mental disorder and risk of harm, with differing qualifiers as to the severity of mental disorder or level of risk. The English and Finnish criteria also included the availability of suitable treatment. Only the criteria cited in the Greek study [49] appear explicitly to include lack of competence to reach a decision about one’s own treatment.

Included studies concerned different frameworks for detention, with some variation between them and a wide range of periods where a child or adolescent was deprived of liberty for assessment or treatment of mental illness (Supplementary Table 2). Twenty studies (47.6%) did not define the detention period, and Philipps et al. [61] noted that there is no limit to the period in which someone can be detained in China. In five studies detention lengths ranged from 24 h in a place of safety (i.e., not an admission to a psychiatric ward) to 72 h [40, 50, 53, 55, 56], while three studies included both shorter periods of detention and longer periods from 14 days up to 6 months [42, 44, 54]. Pelto-Piri et al. [35] focused on analysis of reasons for compulsory care and the period from when a Care Certificate is made to decision on compulsory care is taken (24 h after Care Certificate is created). Studies relating to the Finnish Mental Health Act 1990 [19,20,21,22,23,24,25,26,27, 30, 31] appeared to refer to any assessment or treatment period.

Prevalence, incidence, and trends in detention

The proportion of detentions in relation to overall inpatient admissions, where relevant, could be calculated for 30 studies and ranged from 7% in a nation-wide sample in Finland [29] to 80% among discharges of 225 minors in Canada [55]. The mean across all studies was 30% (median = 24%), with a mean of 21% in the eight studies that used nationwide samples (median = 20%). Ten studies reported on changes in detentions over time, most being from Finland where the rate of detention increased from 7.2 to 8.2 per 10,000 population and incidence rate from 2.7 to 5.9 between 1990 and 1993 [28]. Prevalence of detentions was 14% in 1990 and 11% in 1993 [29], while in later years higher prevalence was reported but with little change across the years (18% in 2000, 19% in 2011, and 22% in 2018) [31]. Two additional studies, from Canada [56] and England [44], reported on increases in detentions but both using service-level data making it difficult to comment on whether it was part of an overall trend. Only one study, by Jendreyshak et al. [46], found a decrease in detentions from 32% in 2004 to 26% in 2009 across 27 districts in Germany.

Factors associated with detention

Clinical and behavioural factors

In 22 studies factors associated with detention were explored either univariately or in multivariate models. Compared to children and adolescents who had voluntary status, those detained had a higher prevalence of psychosis, or psychotic symptoms [19, 20, 29, 39, 46, 59], conduct disorder [27, 48, 58], substance use disorder [19, 22, 46], schizophrenia [42, 59], and personality disorder [42, 58]. So et al. [47] found that detention was significantly associated with wider diagnostic categories (‘internalising’, ‘externalising’ and ‘other’) derived from the Diagnostic and Statistical Manual of Mental Disorders (DSM) other than a category for ‘relational and adjustment disorder’. In children and adolescents with an eating disorder, Ayton et al. [37] found a higher prevalence of depression and Jendreyshak et al. [46] found that ‘mental retardation’ was the strongest predictor for detention.

Other, less frequently factors reported included: self-harm and/or suicidal behaviour at admission [37] requiring one-to-one observation [42], deterioration of mental state [60], longer length of stay [60], referral from non-psychiatric specialty [20], temper tantrums and violent behaviour [20], aggressive behaviour [59, 60], learning disability [46], being treated in an adult ward [29], admission during out of hours [46], experience of abuse [42], being an unaccompanied refugee [36], positive correlation between detentions and child welfare placements [25], higher Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA) score (a general health and social functioning tool) [32], prior emergency admission [47], danger to others [47], lack of motivation [47], and lack of compliance with medication [47].

Sociodemographic and socioeconomic factors

Most studies that compared non-detained individuals found no difference in the proportion of detentions by sex [19, 20, 22, 23, 47, 55], two studies found a higher proportion of boys [42, 60], and one study found a significantly higher proportion among girls [46]. In studies that did not compare with non-detained individuals, seven found that detentions were predominantly girls [19, 23, 25, 26, 47, 55, 59], and in three predominantly boys [36, 45, 60]. Three studies reported differences between ethnic groups. Tolmac and Hodes [45] found that Black adolescents were more likely to be detained on admission than their White counterparts, but there was no significant difference in being detained during the admission. In contrast, one study from England found that psychosis patients from Black or ‘Other’ ethnic backgrounds were more likely to be detained during the admission [39], a US study found higher proportion of detentions among African American compared to White patients [51], and a New Zealand study found higher proportion of detentions among Māori compared to Caucasian adolescents [60]. It is noteworthy that, in many studies, ethnicity was not included as a variable. Finally, detained patients were reported to be older than voluntary patients in four studies [29, 33, 42, 46].

Siponen et al. [26] explored wider socioeconomic and service-related factors and their impact on detentions. They found that the areas with above average rate of detentions typically fared poorer on factors related to social environment, such as a higher divorce rate, lower employment rate, and rate of clients in substance misuse treatment. However, there were more outpatient service (staff) positions, adolescent psychiatric positions, more private and public welfare institutions for children, non-institutional welfare support, but fewer adolescent psychiatric outpatient visits in the above average detention area. While this ecological study could not prove causation, the authors noted that there may be an association between socio-economic disadvantages and detentions and “to reduce the use of involuntary care in adolescent psychiatry and child welfare, approaches focusing on the well-being of families may be indicated” (p.660) [26].

Views on detention of children and adolescents

Five studies included views on detaining children and adolescents; two were cross-sectional surveys of English psychiatrists [41, 43], two were qualitative studies of Finnish psychiatrists [21, 30], and one study included the views of children and adolescents [53].

A survey from England in the early 2000s showed that psychiatrists felt up to date with legislative changes, but the majority agreed that more guidance and training was needed. Importantly, 18% provided an incorrect answer to the question whether parents’ consent to treatment overrides a child’s refusal [48]. A subsequent survey by Mears and Worral [43] found that the main issues for using the Children Act or the Mental Health Act were (i) choosing which Act to use; (ii) general issues around consent for treatment and social services; (iii) stigma; and (iv) conflict between the child’s wishes and parental consent [43]. Studies from Finland were more in-depth and specifically focused on views on the criteria for detention. Kaltiala-Heino and Fröjd [21] interviewed 44 Finnish child and adolescent psychiatrists who all believed that severity of mental disorder could not be arrived at by using ICD or DSM diagnoses (Supplementary Table 2). They differentiated between acute and chronic severity and felt that the criteria for ‘severe mental disorder’ must be justified alongside risk of deterioration, a risk of harm to self or others, and that voluntary treatment is inadequate [21]. Turunen et al. [30] found that arguments supporting a broader criteria for minors included the need for paternalistic intervention due to inability of minors to weigh up need for treatment, detention as an early intervention to prevent future deterioration, and difficulty in diagnosing minors with a mental illness. Argument against a broad criteria for minors was that lack of a definition of severe mental disorder could lead to differences in the interpretation and application of the law, and overall psychiatrists argued that the criteria is too narrow for adults rather than too broad for minors [30].

Finally, Rice and colleagues [53] interviewed 25 children and adolescents staying in a crisis stabilisation unit in USA. Participants felt stigmatised before and after arriving at the hospital for involuntary treatment; some arriving in handcuffs and escorted by police officers leading many to feel disregarded and dehumanised during the detention process. However, receiving and providing peer support was important to feel that they were not alone. By discharge, interviewees reported improvements included ‘opening up’ in group therapy, supporting others, and receiving support from clinical staff. Time away from social media, friends and family gave the young people an opportunity to engage with practices to cope with stressors with positive outcomes including reduced stress levels [53].

Discussion

About one fifth of psychiatric child and adolescent inpatients are treated while detained under mental health legislation, but evidence on how this might be changing is limited. This review has demonstrated that the evidence base relates to data mainly from a few Western high-income countries (HICs) and great variations in the type of detention order studied. Any deprivation of liberty and treatment against one’s will is important and relevant, but despite many included study indicating a significant proportion of inpatients are treated while detained there is little exploration of patient or practitioner views.

Most included studies were conducted in HICs, primarily in Europe, where financing of mental health services is higher. In 2020, expenditure for mental health services in low- and middle-income countries (LMICs) was lower than in upper-middle and HICs, as were the number of health workers in Child and Adolescent Mental Health Services (CAMHS), and insurance arrangements means patients often pay out of pocket for mental health services and/or psychotropic medicines [62]. Lack of funding for mental health services likely deter individuals to seek care, alongside stigma which has been shown to differ between Eastern (Asian) and Western countries [63]. The current review leaves many questions of how children and adolescents are detained in LMICs when voluntary psychiatric care is no longer an option. Coverage of data on major mental disorders is poor in LMICs [64] and future work should focus on collecting data on involuntary psychiatric care of children and adolescents in countries where this is currently lacking. This also includes research into detention of children and adolescents in HICs, as we note a lack of studies from Australia and the USA. Australian research, such as a recent descriptive study of administrative data relating to CAMHS, did not report on legal status of admissions [65]. The reason for the lack of studies noting the legal status is unclear, as data from the Australian Institute of Health and Welfare breaks down involuntary treatment by age [66]. In the USA, many states do not report on detentions and of those who do, only six states separate data on adult and minors (though this article does not define minor and definitions varies in different states) [67].

There is a notable lack of qualitative research on detention of children and adolescents. The only study we found involving lived experiences included short detention in crisis management settings due to suicidality [53]. While these views might differ from those detained for other reasons, some experiences resonate with findings from a review of qualitative studies including adults. A review found that individuals experienced an unnecessary loss of freedom as they felt other alternatives were available, but also that detention was a sanctuary and way to recover away from life problems [68]. Such views may however change over the course of the detention [53, 68]. A study including young people (16–27 years) found that the majority felt the experience of detention had significant impact on trust to disclose their feelings, which in turn impacted on their post-discharge help seeking. While other participants reported some positive outcomes, those who lacked trust ended up actively withholding how they felt in fear that they would get detained again [69]. Given the impact detention may have on future treatment and a potential cycle of inequality [16], along with calls from international human rights bodies to reduce or eliminate medical coercion [12], more research on children’s and adolescents’ experiences is urgently needed.

Views of parents, caregivers, and care providers are also important. We found no study involving parents’ views and only few that described the experiences of psychiatrists [21, 30, 41, 43]. A major gap remains in the literature on studies exploring how psychiatrists’ perceptions and practices of detaining the youngest patients, their views on addressing inequalities [4], and views on trends or changes in presentations of detained patients in these age groups. In studies on voluntary inpatient treatment parents have reported feeling unprepared, struggling to get access to CAMHS services, and the wider family impact [37,38]. Considering that more older adolescents were detained, it may suggest that parental authority plays a part in the decision to admit a child without their consent. This requires studies to understand how parental views play a part in voluntary or involuntary status of admissions.

While mental health problems among young people are increasing in many countries [70,71,72,73], this review could not draw conclusions of trends in detention internationally. Finnish studies showed increases in detention rates [22, 25, 27, 28, 31], while German study showed a decrease in proportion of admissions that were detentions [46]. Changes over time in sub-national samples are less informative, especially as evidence from Finland identified significant variations across regions [29]. Research from Finland has indicated a drastic increase in first treatment of adolescents in psychiatric inpatient care between 1980 and 2010, alongside a decrease in length of stay and global assessment scores [74]. The latter suggests increases in admissions are not related to changes in clinician perceptions of admission thresholds and questions remain regarding higher readmission rates in girls, but possibly due to “the shift from socio-ecological social policies in earlier decades to individual risk and psychopathology-oriented health and social policies” (p.7) [74]. In all age groups, research from England has indicated that reasons for increases in detentions may include the impact from austerity measures, financial crises, and legislative changes [75]. Variation in detention rates between countries is largely unexplained with weak associations between higher incidence of detentions and higher gross domestic product (GDP) and health care expenditure, lower rates of poverty, higher number of inpatient beds, and proportion of foreign born individuals [4]. Kaltiala-Heino [22] also noted that “concern about legal and civil rights of minors may paradoxically increase commitments through more awareness of the obligation to act legally instead of simply deciding over minors without formally recording coercion” (p.57) [22]. In addition, it is difficult to report on the role that diagnoses might play in both the understanding of who gets detained and whether there are differences globally and the appropriateness of detention as there are differences between jurisdictions. The adoption of the ICD-11 in many countries might form a basis to include diagnostic frameworks, wherever appropriate, in recording diagnoses at the point where detention is associated with treatment for a specific condition(s). Changes in detention rates are, however, likely related to a multitude of factors and much of it remains unexplained. Longitudinal nationwide register studies that allow for cross-country comparisons among children and adolescents are lacking and should be a priority for future studies, including comparing detention rates and characteristics with adult populations to better characterise how detention of children and adolescents is used.

Finally, this review has demonstrated the differences in types of detention in studies, which further limits cross-country comparisons. A previous study, which aimed to compare rates of detention and legislative frameworks, highlighted differences in how orders are used and counted in different countries, as no association has been found between detention rates and characteristics of legal criteria [4]. The authors suggested that factors such as detention based on perception of risk vs. urgent need for treatment and coercion within voluntary admissions, where patients are informed they will get detained if they don’t consent to treatment, could impact on detention rates [4]. Differentiating between types of detention, when exploring trends, might be important as in England, for example, there has been a 13% increase in short-term detention over the last five years [76]. The trend in Sect. 5 [2] (assessment up to 72 h) and 5 [4] (nurses’ power to detain) detentions has declining while place of safety orders (s.135 and s.136) have increased in the last two years [76], which may be under circumstances related to lockdowns during Covid-19. Disentangling assessment, treatment, and crisis interventions involving police (in the case of s.135 and s.136 in the UK) might provide information on upstream interventions or resources needed to prevent detention in children and adolescents.

Limitations

We developed this scoping review with input from a subject librarian to ensure the search strategy was comprehensively designed. However, no protocol was registered prior which is acknowledged as a limitation. This review specifically focused on children and adolescents (aged < 18 years) due to UNCRC’s definition of a child and our interest was also based on the service provision in the UK with specialised child and adolescent mental health services. Within current debates about how adolescence now may span a longer period [18], this restriction may have excluded valuable findings. The review focused on peer reviewed articles, meaning information about trends in detention of children and young people published in statistical reports published by governments or other organisations have been missed, which may give an indication of international trends. As we only reviewed articles published in English, we may have missed information relating to detentions in other contexts than described here.

Conclusion

Detentions account for about one fifth of psychiatric admissions among children and adolescents, but evidence on trends based on national register data has only been published for a few Western HICs. The circumstances justifying detention and the criteria authorising detention varied between studies, with a mix of clinical factors and observed behaviours. There is some evidence to suggest minority ethnic children and adolescents and those with a history of abuse are disproportionately affected by detention. From a human rights perspective, psychiatric detention based on observed behaviours may be unjustified because of the lack of a ‘true’ mental disorder which benefits from specialist treatment. It may be more justified in emergency situations for urgent and short-term interventions, so understanding what kind of order is being imposed is important. Future research should look in greater detail into different kinds of detention and how they are applied using human rights frameworks. More qualitative studies on the experiences of detention of children and adolescents are urgently needed.