Adolescent admissions to adult psychiatric units: patterns and implications for service provision

http://apy.sagepub.com/ Australasian Psychiatry http://apy.sagepub.com/content/19/4/345 The online version of this article can be found at: DOI: 10.3109/10398562.2011.601311 2011 19: 345Australas Psychiatry Chohye Park, Brett McDermott, Jik Loy and Peter Dean Adolescent Admissions to Adult Psychiatric Units: Patterns and Implications for Service Provision Published by: http://www.sagepublications.com On behalf of: The Royal Australian and New Zealand College of Psychiatrists can be found at:Australasian PsychiatryAdditional services and information for http://apy.sagepub.com/cgi/alertsEmail Alerts: http://apy.sagepub.com/subscriptionsSubscriptions: http://www.sagepub.com/journalsReprints.navReprints: http://www.sagepub.com/journalsPermissions.navPermissions: What is This? - Aug 1, 2011Version of Record >> at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/ http://apy.sagepub.com/content/19/4/345 http://www.sagepublications.com http://www.ranzcp.org http://apy.sagepub.com/cgi/alerts http://apy.sagepub.com/subscriptions http://www.sagepub.com/journalsReprints.nav http://www.sagepub.com/journalsPermissions.nav http://apy.sagepub.com/content/19/4/345.full.pdf http://online.sagepub.com/site/sphelp/vorhelp.xhtml http://apy.sagepub.com/ http://apy.sagepub.com/ CHILD AND ADOLESCENT PSYCHIATRY A ustralasian Psychiatry • Vol 19, N o 4 • A ugust 2011 345 Adolescent admissions to adult psychiatric units: patterns and implications for service provision Chohye Park, Brett McDermott , Jik Loy and Peter Dean Objective : The aim of this study was to investigate patterns of child and adolescent admissions to an acute adult psychiatric unit in a rural city. Correlates of admissions were then considered in terms of service reform for this vulnerable, under-resourced group. Method : The study reviewed consecutive clinical records of children and adolescents who were admitted to an acute general psychiatric inpatient unit over a 6 year period (N � 332). Results : Patients generally experienced numerous pre-admission psychosocial stressors; there were many abuse histories and/or juvenile justice involvement. The principal diagnosis was varied and comorbidity was common. Maori patients were over-represented. The majority of admissions occurred out of working hours and more than half came from rural areas. There was high usage of the Mental Health Act on admission. Common causes of admission were self- harm and suicidal behaviour. The majority of the admitted adolescents required follow up by child and adolescent mental health services after discharge. Conclusion : We identifi ed several reform possibilities, including up-skilling emergency and adult mental health staff in child and adolescent mental health, exploration of alternatives to admissions and specialist service coverage. Key words: admission , adolescent mental health , adult psychiatric unit. doi: 10.3109/10398562.2011.601311 © 2011 The Royal Australian and New Zealand College of Psychiatrists Chohye Park Child and Adolescent Psychiatrist, Hamilton, New Zealand . Brett McDermott Associate Professor, Kids in Mind Research: The Mater Center for Service Research in Mental Health, and University of Queensland, Brisbane, Queensland, Australia . Jik Loy Child and Adolescent Psychiatrist, Waikato District Health Board, Hamilton, New Zealand . Peter Dean Forensic Psychiatrist, Regional Forensice Service, Waikato District Health Board, Hamilton, New Zealand. Correspondence: Dr Chohye Park, Child and Adolescent Psy- chiatry, PO Box 4139, Hamilton, New Zealand. Email:chpark@ xtra.co.nz There is widespread recognition that the care of adolescents presenting with acute and severe mental health problems poses a challenge for service provision. This is within the context of a recent Australasian survey that found 25% of people aged 16 – 24 years suffered a mental disorder in the previous 12 months compared to 20% for the total population. 1 Of participants with a disorder, only about one third sought professional help. Issues with acute adolescent presentations include lack of beds, geographical location delaying prompt admission to child and adolescent inpatient units, and by necessity use of acute adult inpatient psychiatric units or paediatric wards. 2 Where there are limited resources for adolescents who present in crisis, an acute adult psychiatric unit is often the only available admission option. Signifi cant issues arise from the admission of adolescents to adult units. As a generalization, adult services lack understanding of the developmental needs of adolescents, lack the skills to deal with age-specifi c behaviour and psychopathology, and there are safety issues in the collocation with severely disturbed adult psychiatric patients. 3 – 5 It is also recognized by child and adolescent psychiatrists that appropriate family work and a therapeutic milieu, including education support, cannot be provided in adult acute wards to the adolescent inpatient population. 6 Teens aged 16 – 17 years who have dropped out of education are one example of a vulnerable group at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/ 346 A us tr al as ia n Ps yc hi at ry • Vo l 1 9, N o 4 • A ug us t 2 01 1 whose mental health needs would be unlikely to be comprehensively assessed in general adult psychiatry. The advantage of child and adolescent units has been emphasized in the literature, with outcome studies indicating therapeutic goals being achieved even with short term admissions. 7,8 There are several studies indi- cating that child and adolescent groups with severe psychopathology and family dysfunction were better provided with treatment in inpatient or day program settings. 9,10 Therefore, it is argued that there is a need for age-appropriate inpatient psychiatric units to pro- vide evidence-based, effective services to the adolescent population. Another contemporary service provision issue is the appropriateness of some admissions for adolescents. It has also been argued that increasing numbers of ado- lescent admissions to inpatient units, especially in the US in the past, was related to social policy and the placement of troubled adolescents from one social insti- tution to another. It raised questions over the suitability of use of inpatient beds for adolescents in the US. 11 Behaviourally disturbed adolescents with social prob- lems are admitted to adolescent inpatient units, pre- dominantly with a diagnosis of conduct/disruptive behavioural disorder. 12 While appropriate admission of adolescents to a psychiatric inpatient facility may pre- vent deterioration of functioning and mental wellbe- ing, inappropriate admission would cause restriction of liberty, stigmatization, medicalization of social prob- lems, and future poor compliance. It is therefore neces- sary to investigate diagnoses and the relationship with involvement of other social agencies leading to adoles- cent inpatient unit admission. 13 This study investigates child and adolescent psychiatric patients under 18 years who were admitted to a regional acute adult psychiatric unit, approximately 130 km from a specialized child and adolescent inpatient unit. Patients who required a longer stay or more intensive treatment are generally transferred to this specialized unit within a few days of the acute adult admission. We report on the demographics and diagnostic profi le of admitted adolescents as well as aspects of the care path- way including referral, police involvement, use of the Mental Health Act and post-discharge follow-up. METHOD Participants Consecutive admissions to the general psychiatric inpa- tient ward in Hamilton from January 2002 to December 2007 were audited. Admissions during this period totalled 344 and we were able to audit 332 (96%) of these fi les. On review of computer records, the missing fi les (12) appeared predominantly to be a group who had overnight inpatient admissions with no further fol- low up arranged. The general population of the area is 374 000, of which 10% are under 18 [2001 census]. The acute general psychiatric inpatient unit has 54 beds. However, there is no child and adolescent inpatient unit; Hamilton and its rural area has 1.2 beds allocated in the specialized child and adolescent inpatient unit in Auckland. Procedure Information was obtained retrospectively from the clin- ical fi les held by the service. The fi rst and third authors applied a data collection template to capture in a system- atic process the relevant characteristics of patients admit- ted to the adult general psychiatric unit. Data collected included demographics, precipitating problems, referral source to psychiatric unit, past psychiatric history, MHA usage, family composition, family psychiatric history, geographic area of origin, learning disorder/intellectual disability, substance abuse history, history of sexual/ physical abuse, medication usage, seclusion history, length of stay, follow-up arrangement, discharge place and fi nal diagnosis. Health district business rules ensures all diagnoses were based on DSM-IV criteria. The fi rst author audited all fi les and the third author selected one out of every ten fi les audited in order to validate the fi ndings. The fi ndings of both authors were found to be consistent. As this study met the criteria for an audit, it was processed as an observational study for expe- dited review by the Northern Regional New Zealand Ethics Committee. Statistical analysis To facilitate analysis, data was often truncated into sum- mary categories. For example 16 “ causes ” of admissions were summarized into causes relating to: self-harm/ suicidality, mental state deterioration and aggression. Nominal and ordinal variables were analysed with χ 2 or logistic regression analysis; the latter included age and gender as typical covariates. Continuous data was analysed with t-tests. No variable had more than 10% missing data. The p level was set at 0.05. RESULTS Sample characteristics Of the 332 admissions audited, there was a small gender difference in admission numbers (169 girls and 163 boys). Two ethnic groups were strongly repre- sented: Caucasian 57.5% (191/332) and Maori 39.4% (131/332). Very small numbers of Pacifi c Island (6/332), Asian (3/332) and African (1/332) patients were admit- ted. The mean patient age was 16.5 years (SD 1.1). There were 88 patients who were under 16, the youngest of whom was 12 years old. The participant group had experienced numerous previous psychosocial stressors. Sixty percent of clients had a past history of mental illness (205/332) and over half of the clients had a family history of mental illness (177/332). Other family issues included that approximately two thirds of the patients had experienced parent separation or divorce at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/ 347 A ustralasian Psychiatry • Vol 19, N o 4 • A ugust 2011 (219/332). More than one third of the clients had a history of sexual and/or physical abuse (126/332). Pre- vious police and/or justice system involvement was common (116/332), more so in boys (69.1% of those involved) and more than one third of the patients had a history of learning disorder with or without intellec- tual disability (113/332). Referrals came from a variety of services, mostly from a family member (31%), police (28.9%) or via the hos- pital emergency department (16.3%) (Table 1). Two thirds of admissions occurred out of working hours fol- lowing psychiatrist/registrar assessment. More than half of the admissions (53%) were from outside the city boundaries. The principal DSM-IV Axis I diagnoses are presented in Table 2; 11.4% of clients (38/332) had a comorbid Axis I diagnosis. The most frequently diag- nosed conditions were mood (38.2%) and psychotic (25.7%) disorders followed by a diverse range of condi- tions of roughly similar frequency: anxiety, adjustment, substance and disruptive behaviour disorders. Only three individuals had a diagnosis of anorexia nervosa; 15.6% of patients (52/332) had no Axis I diagnosis, but rather were classifi ed as experiencing a personality dis- order (16/332), situational problem (12/332) or had no diagnosis (9/332) (Table 2). Mental Health Act: relationship to patient characteristics and cause of admission The majority of admissions to the general ward involved the Mental Health Act (61.4%, 204/332). There was a small but signifi cant over-representation of Maori under the Mental Health Act (68.2% of admissions of Maori versus 57.1% of Caucasians ( χ 2 � 4.04, p � 0.04)). Sig- nifi cantly more men were admitted using the Mental Health Act (121/163 versus 83/167 women, χ 2 � 21.03, p � 0.000). There was no signifi cant difference in the age of patients under the Mental Health Act or not under the Act. Fewer patients from Hamilton (85/154, 55.2% versus 119/176, 67.6%) compared to an outlying region were under the Mental Health Act ( χ 2 � 5.37, p � 0.02). The causes of admission were summarized into three categories: self harm or suicidal behaviour (184/332, 55.4%), mental health deterioration (84/332, 25.3%) and aggression/violence (60/332, 18.1%). There were signifi cant differences in the use of the Act and the cause of admission. The Act was signifi cantly more likely to be used if the cause of admission was a dete- rioration in mental health status (67/84, 79.7%) or aggression (45/60, 75.0%), but not self harm/suicidality (91/182, 50.0%, χ 2 � 26.74, p � 0.000)). Service provision, length of stay, medication, seclusion and follow up The durations of admission of two patients were con- sidered statistical outliers (lengths of stay 157 and 247 days); in both cases, admission duration was due to diffi culty fi nding post-hospital accommodation). When excluded from the analysis, the average length of stay in the unit was 7.18 days, (SD 12.6). However, over half of admitted patients (186/332) were discharged within three days. The longer stays were associated with use of the Mental Health Act (11.23 days versus 3.75 days) and the cause of admission being aggression or mental status deterioration rather than self harm/ suicidality (13.77 versus 4.59 days). There was a gender difference in length of stay with boys staying twice as long as girls (11.71 days versus 5.05 days ) , Seclusion occurred in 12.3% (41/332) of patients, most frequently those patients with a diagnosis of psychosis (17/41). Antipsy- chotic medication was the main medication (37/41) taken by a patient who experienced a period of seclusion. The most frequently used medication was a mixed reg- imen, most usually an antipsychotic plus antidepres- sant. Both a mixed regimen (57/86 versus 29/86) and antipsychotic alone (60/72 versus 12/72) were more often used in patients under the Mental Health Act. Antidepressants were used less often in patients under the Mental Health Act (27/67 versus 35/62; overall χ 2 4 � 26.54, p � 0.000). Over half of the inpatients were discharged to CAMHS services (188/332, 56.7%), followed by general practitio- ners (49/332, 14.8%), adult mental health services (43/332, 13%) and to the regional child and adolescent inpatient unit (40/332, 12.0%). There were 12 patients who were not referred for follow up. Compared to Euro- peans, twice as many Maori were transferred to the regional child and adolescent inpatient hospital (23/36, 63.4% versus 13/36, 36.1%; χ 2 � 9.12, p � 0.00). The most common diagnosis at transfer was psychosis (19/40). DISCUSSION Over the 6 year period, a consistent number of adoles- cents, 50 – 60 per year, were admitted to the acute general Table 1: Patient referral details and Mental Health Act usage Referral source Out of working hours admissions Mental Health Act usage N % N % N % Family 103 (31.0) 74 (30.1) 59 (28.9) Police assisted 96 (28.9) 84 (41.2) 68 (33.3) Emergency department 54 (16.3) 38 (15.4) 25 (12.2) CAHMS 21 (6.3) 4 (1.6) 18 (8.8) GP 10 (3.0) 6 (2.4) 6 (2.9) Other agencies 30 (9.0) 23 (9.3) 22 (10.7) Self-referral 18 (5.4) 17 (6.9) 6 (2.9) Total 332 (100) 246 (100) 204 (100) at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/ 348 A us tr al as ia n Ps yc hi at ry • Vo l 1 9, N o 4 • A ug us t 2 01 1 Table 2: DSM IV Axis I Diagnoses, Mental Health Act and Length of stay Diagnosis Total Use of Mental Health Act Length of stay Mean days (SD) N % N % Days SD Any mood disorder 107 (38.2) 71 (39.4) 8.7 (25.1) Major depressive disorder 43 28 3.0 (3.30) Depressive disorder NOS 37 20 4.83 (7.90) Dysthymic disorder 4 2 3.75 (3.00) Bipolar disorder 23 21 16.5 (22.0) Any anxiety disorder 27 (9.6) 11 (6.1) 5.6 (9.2) Obsessive-compulsive disorder 2 1 1 Posttraumatic stress disorder 18 8 7.9 (7.0) Acute stress disorder 7 2 1 Any psychotic disorder 72 (25.7) 50 (27.8) 14.4 (22.0) Psychotic disorder NOS 58 38 13.9 (23.9) Schizophrenia 8 8 20.4 (11.6) Schizophreniform disorder 1 0 15 Brief psychotic episode 5 4 9.4 (6.8) Any disruptive behaviour disorder 19 (6.8) 13 (7.2) 4.6 (11.1) Conduct disorder 18 12 4.8 (11.4) ADHD 1 1 1 Other diagnosis 55 Adjustment disorder 25 (8.9) 18 (10.0) 7.52 (22.1) Substance abuse 20 (7.1) 9 (5.0) 3.35 (2.6) Autistic spectrum disorder 3 2 17 (23.3) Asperger ’ s syndrome 4 3 17.5 (25.8) Anorexia Nervosa 3 3 13 (0.9) Total 280 (100) 180 (100) inpatient unit. The commonest pathway to inpatient admission was an out of working hours presentation by concerned family members, with or without police involvement, for a patient who resided out of the city boundaries. Clearly there is a burden both in terms of yearly bed-days by this group, as well the need for child and adolescent expertise in out of hours settings. Most CAHMS arranged admissions occurred during working hours. Specialist child and adolescent outpatient ser- vices, inclusive of early treatment opportunities for vul- nerable adolescents, may prevent out of hours crises and the need for families to seek urgent assistance. As the average length of stay was relatively short, it appears that most of the admissions provided crisis intervention and assessment. The shortest stay group were those who were referred on to the regional child and adolescent inpatient hospital, mainly diagnosed with psychosis, with high usage of the Mental Health Act, and use of antipsychotic medication. When ado- lescents under the age of 18 were admitted to the acute general inpatient unit, the child and adolescent service became involved during working hours. They were then triaged and referred to a more appropriate service. As the most severely unwell group were transferred to a specialized inpatient child and adolescent unit after a few days stay in the adult unit, there appeared to be a role for the acute adult unit to provide a containing environment for adolescents while awaiting transfer. In a recent Australian child and adolescent inpatient unit study, all admissions had an Axis I diagnosis. 14 In our audit, 52 patients did not have an Axis I diagnosis which may indicate the inappropriateness of some of these admissions. The signifi cant rates (28.9%) of police involvement in child and adolescent admissions require further investigation including whether there has been appropriate use of the Mental Health Act. Police involve- ment may impact on self-esteem and be a potentially traumatizing experience for this vulnerable group. It highlights the need for a resource to manage mentally unwell adolescents in crisis. In this context, there was no available child and adolescent specialist service in the admission process and there was no alternative means to contain these troubled adolescents. Managing abnormal behaviour and abnormal state of mind by using the Mental Health Act may have unfavourable consequences for future compliance and engagement with mental health services. A developmental task at this age is to gain a sense of self-control and self-deter- mination. Adolescents may be better served by allowing them to take a more active role in the treatment plan- ning of their illness, potentially improving engagement in their future management. at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/ 349 A ustralasian Psychiatry • Vol 19, N o 4 • A ugust 2011 Along with high use of the Mental Health Act, there were some adolescents who had been secluded in the acute adult inpatient psychiatric unit with severe men- tal illness. This practice may be useful to protect the adolescents from exposure to severely disturbed adult psychiatric patients, but it raises concern that opportu- nities for age-specifi c alternative methods to manage acutely disturbed adolescents are missed in a general adult unit. Seclusion in an adult unit may be the fi rst experience of psychiatric treatment for such patients, reducing opportunities for long-term engagement and therapeutic alliances. The Maori population (18% in the 2001 census in the region) is over-represented in admissions to the acute adult psychiatric unit. There is a signifi cantly greater risk of being given a diagnosis of psychosis compared to non-Maori. The Maori population has a small but statistically signifi cant over-representation of Mental Health Act use. An aspiration is to reduce this over- representation. Before this can be achieved, a more detailed understanding of psychosis in Maori adoles- cents is needed. This may lead to an argument for a specifi c early intervention strategy with this group. Other social factors such as family composition, family psychiatric history and family substance abuse appeared to be highly associated with admission, which was con- sistent with other research. 14,15 Working with limited resources and including local child and adolescent services, the 24-hour availability of specialist child and adolescent service trained staff for adolescents emergency mental health presentations may help to reduce admissions to the adult unit and facilitate the speedy transfer to the regional child and adolescent inpatient unit. Other possible alternatives to adult unit admissions include supported respite care beds, intensive community treatment or out of hours direct admissions to the regional child and adolescent unit. Training and increased support for the general inpatient unit staff and crisis team to improve their skills in dealing with vulnerable adolescents may improve the service provided to this group. Study limitations This was a retrospective clinical fi le review and thus relies on information recorded in clinical records. The study only described those who were hospitalized in an adult inpatient unit, and therefore young people who were admitted directly to the regional inpatient child and adolescent inpatient unit were not included in this study. Further categorization of the patient group, for example of mood and psychotic disorders, would require use of a structured diagnostic interview that was suitable for research purposes. DISCLOSURE The authors report no confl icts of interest. The authors alone are responsible for the content and writing of the paper. REFERENCES Australian Bureau of Statistics, 2008a. 1. National Survey of Mental Health and Wellbeing: summary of results, Australia, 2007 . ABS Cat No.4326.0. Canberra: ABS. O ’ Herlihy A, Worrall A, Banerjee S 2. et al . National in-patient child and adolescent psychiatry study(NICAPS) . Report submitted to the Department of Health, 2001 www. rcpsych.ac.uk/PDF/NICAPS. Scott D, McGilloway S, Donnelly M. Young people ’ s experience of adult inpatient 3. psychiatric care: a Northern Ireland case study. Mental Health and Learning Disabili- ties Care 2001; 4 : 305 – 308. Barker S. 4. Environmentally unfriendly: patients ’ view of conditions on psychiatric wards . London: MIND, 2000. 11 Million. 5. Pushed into the shadows: young people ’ s experience of adult mental health facilities 2007 . London: 11 Million (Offi ce of the Children ’ s Commissioner for England). Worrall A, O ’ Herlihy A. Psychiatrists ’ views of in-patient child and adolescent mental 6. health services: a survey of members of the child and adolescent faculty of the Royal College of psychiatrists. Psychiatric Bulletin of the Royal College of Psychiatrists 2001: 25 : 219 – 222. Mayers SD, Calhoun SL, Krecko VF, Vesell HP, Hu J. Outcome following child psychiatric 7. hospitalization. Journal of Behavioural Health Service and Research 2001; 28 : 96 – 103. Gavidia-Payne S, Littlefi eld L, Hallgren M, Jenkins P, Coventry N. Outcome evaluation of 8. a statewide child inpatient mental health unit. Australian and New Zealand Journal of Psychiatry 2003; 37 : 204 – 211. McDermott BM, McKelvey R, Roberts L, Davies L. Severity of children ’ s psychopathol-9. ogy and impairment and its relationship to treatment setting. Psychiatry Services 2002; 53 : 57 – 62. McShane G, Mihalich M, Walter G, Rey J. Outcome of patients with unipolar, bipolar 10. and psychotic disorders admitted to a specialist child and adolescent mental health service. Australasian Psychiatry 2006; 14 : 198 – 201. Warren T, Gutteridge R. Adolescent psychiatric hospitalization and social control 11. . In: Tesplin L, ed. Mental health and criminal justice . Beverly Hills, CA: Sage, 1984: 119–133. Weithorn O. Mental hospitalization of troublesome adolescents: an analysis of skyrock-12. eting admission rates. Stanford Law Review 1988; 40 : 773 – 838. Kaltiala-Heino R, Frojd S. Severe mental disorder as a basic commitment criterion for 13. minors. International Journal of Law and Psychiatry 2007; 30: 81 – 94. Tonge B, Huges C, Pullen J, Beaufoy J, Gold S. Comprehensive description of adoles-14. cents admitted to a public psychiatric inpatient unit and their families. Australian and New Zealand Journal of Psychiatry 2008; 42 : 627 – 635. Gutterman E, Markowitz J, LoConte J, Beier J. Determinants for hospitalization from 15. an emergency mental health service Journal of The American Academy of Child and Adolescent Psychiatry 1993; 32 : 114 – 122. at Bobst Library, New York University on October 8, 2014apy.sagepub.comDownloaded from http://apy.sagepub.com/