A CAMHS Intensive Treatment Service: Clinical outcomes in the first year

Abstract

In 2005, the Scottish Executive recommended that young people with severe mental health difficulties should be managed in the community wherever possible. This study reports on the clinical outcomes associated with the development of a Child and Adolescent Mental Health (CAMH) Intensive Treatment Service (ITS) embedded within a Tier Four CAMHS structure. Following intervention, mean CGAS ratings significantly improved by 16 points (p<0.001) and HoNOSCA Clinical Scores significantly decreased by a mean of 6.94 points (p<0.001). This trend was replicated in self-report measures, where service users reported significant clinical improvements in mental health symptoms and indicators of quality of life. A CAMHS ITS, with close links to an adolescent inpatient unit, can provide a balanced care approach where young people with severe mental health difficulties can be treated in the community, where possible, without compromising on patient safety and quality of care.

Keywords

Outcome intensive outreach child adolescent mental health services

Introduction

In the UK, child and adolescent mental health services (CAMHS) are described according to a four-tier strategic framework, characterised by increasingly specialised provision (NHS Health Advisory Service, 1995). Tier Four CAMHS consists of specialised services for young people with severe and/or complex problems and refers to inpatient units (IPUs), day programmes and a range of intensive community services.

While there is strong evidence that adolescent IPUs are effective, (Blantz & Schmidt, 2000; Green et al., 2007; Pfeiffer & Strzelecki, 1990) there is a national shortage of age-appropriate inpatient beds (O’Herlihy, Worrall, & Banerjee, 2001) and an inequality in access (O’Herlihy et al., 2007). Also, in recent years, international, UK and Scottish policy (where health is a devolved power) has emphasised a shift in mental health provision towards preventative community-based care and away from acute hospital-centred activity (Scottish Executive, 2005a; World Health Organization, 2001). Within CAMHS, the Scottish Executive recommended that young people with severe mental health difficulties should be managed in the community wherever possible (Scottish Executive, 2005b). Service users have emphasised the need for more flexible, specialist services which are easier to access in an emergency (Lavis & Hewson, 2011). Furthermore, with increasing financial pressure within the NHS, the expense of inpatient care is being highlighted, with the proposal that intensive treatment services could be cost effective (National Mental Health Development Unit, 2010), although others stress that only some models have shown cost savings (Kurtz, 2009). In response, a number of models of intensive community treatment are being evaluated and implemented including multi-systemic therapy, day provision, case management, specialist outpatient services (including rapid outreach and crisis intervention), intensive home treatment, family preservation services, therapeutic foster care and short-term residential care programmes.

However, in contrast to adult mental health, there is comparatively little research into intensive community treatment for young people (Worrall-Davies & Kiernan, 2005). In a systematic review of the literature, Shepperd et al. (2007) highlighted problems associated with the current evidence base, including methodological issues, difficulties implementing randomised control trials with this population, and problems associated with extrapolating findings from a mainly North American evidence base, some of which only focus on specific problem areas, for example antisocial behaviours. Furthermore, a number of studies implied that intensive community treatment can provide a substitute or competitive model of care to adolescent IPUs. This is despite international and UK policy stressing the need for a balanced care model (Thornicroft & Tansella, 2004) where assertive outreach services and psychiatric IPUs are seen to be components in a spectrum of intensive mental health care for children and young people (Scottish Executive, 2004).

Taking into account the difficulties associated with implementing randomised control trials with this client group, Shepperd et al. (2007) recommended the need for comparative systems of audit of intensive treatment services to include baseline measurement at admission, demographic data and outcomes measured on standardised instruments. This paper will report on a CAMHS Intensive Treatment Service (ITS), which provides outreach to community settings, but is embedded with a Tier Four NHS CAMHS structure and maintains close links with an adolescent IPU to facilitate effective and efficient hospital stays when necessary, combined with a preventative, early intervention approach jointly with Tier Three CAMHS.

Lothian CAMHS Intensive Treatment Service

The CAMHS ITS discussed here is a multi-disciplinary service designed to meet the needs of young people with severe mental health difficulties, through intensive community treatment. The ITS is not a crisis service – this remains the remit of on-call psychiatry – but instead offers responsive planned care. It is designed to prevent hospital admissions where possible by offering community-based care packages, but also facilitates appropriate admissions to a 12-bed generic NHS adolescent IPU serving the South East of Scotland. Development of this CAMHS ITS and associated service redesign has been shown to significantly reduce median length of IPU stay and reduce admissions of young people to adult psychiatric wards by 65% (Duffy & Skeldon, 2012).

The ITS covers five Tier Three outpatient teams, three specialist teams, and supports existing Tier Four teams. It is designed to supplement existing services, rather than hold independent caseloads, producing fewer transitions for young people as they have contact with outpatient CAMHS workers and community resources throughout ITS intervention. Young people meet criteria if, as a result of severe mental health difficulties, the assessment or management of their risk or functioning requires appointments in excess of what Tier Three CAMHS is able to provide. Furthermore, all young people admitted to the IPU are allocated an ITS Community Psychiatric Nurse (CPN) at the point of admission to encourage transfer of therapeutic gains outside the unit and assist in integrating the young person back into their community.

The ITS covers a mainly urban area where 160,379 are estimated to be under 18 (National Records of Scotland, 2011). It is staffed by a 0.5 WTE Senior Charge Nurse, five CPNs, one Assistant Psychologist, 0.6 WTE Psychiatry, 0.5 WTE Occupational Therapy, 0.75 WTE Clinical Psychology, and dietetic and social work input. Overall case management and packages of care are determined by multi-disciplinary care planning meetings and reviewed every 4–6 weeks. Interventions are multi-disciplinary and eclectic, and include risk assessment; monitoring of physical and mental health; evidence-based individual, family and group therapy; medication; case management; and support in accessing external community agencies and education. Flexible packages of care aim to reflect ongoing need, and interventions take place within the young person’s home or community.

This CAMHS ITS was launched in April 2010 and began to consistently use a range of standardised outcome measures in October 2010 to inform clinical interventions and determine the effectiveness of the service. This study aims to report demographic information, baseline admission data and clinical outcomes.

Method

A combination of self-report and clinician-reported outcome measures were collected from 20 October 2010 until 31 December 2011. Measures were administered within 2 weeks of acceptance and discharge from the CAMHS ITS. Referral data for this time period were obtained via the Patient Information Management System.

Measures

The following measures were completed at entry and discharge from the service: the Children’s Global Assessment Scale (CGAS) (Shaffer et al., 1983); the Health of the Nation Outcome Scales for Children and Adolescents (Clinician version) (HoNOSCA) (Gowers et al., 1999); and the short World Health Organisation Quality of Life Instrument, (WHOQOL-BREF) (WHOQOL Group, 1998). Clinicians received appropriate training in HoNOSCA and CGAS administration and the same clinicians scored the young person at admission and discharge from the service. Furthermore, appropriate Beck Youth Inventories-Second Edition, (BYI-II) (Beck, Beck, Jolly, & Steer, 2005), were administered at admission and discharge according to service users’ presentations (minimum of one and maximum of three, measuring symptoms of depression, anxiety, anger, disruptive behaviour, and self-concept), and the Paddington Complexity Scale (PCS) (Yates, Garralda & Higginson, 1999), was completed at the point of admission.

Results

Participant characteristics and service use

Between 20 October 2010 and 31 December 2011, the CAMHS ITS accepted 113 referrals consisting of young people aged between 7 and 17 years (mean: 14.75, S.D.: 1.93) where 66% (n=75) were female. Service user deprivation indices (McLoone, 2004) were equivalent to the local population. Mood disorders and self-harming behaviour accounted for 41% of referrals (n=47), while eating disorders were the second highest presentation (27%, n=31). Median length of ITS intervention was 15 weeks (range: 1–48 weeks) and median number of appointments was 15 (range: 1–118). Of the 113 accepted referrals, 28 (24.8%) were admitted to the IPU (and therefore concurrently ITS) at first contact with Tier Four services, where the median length of admission was 17.5 days (range: 1–117) prior to discharge to the ITS. A further 13 young people (11.5%) were admitted to the IPU while being treated by the ITS after a median of 24 days (range: 1–90) of ITS intervention, and they remained in the IPU for a median of 39 days (range: 6–120).

Baseline admission data

Out of 113 accepted referrals, 95 (84%) completed pre-intervention measures. Uncompleted measures were a result of difficulties engaging with the service or service users being too unwell to complete them. Table 1 summarises the baseline admission data.

Table 1.

Baseline Admission Data for the CAMHS ITS.

Measure		n (% of admissions)	Mean (SD)
CGAS	Total Score	95 (84.1%)	43.37 (10.86)
HoNOSCA	Clinical Score	95 (84.1%)	22.94 (5.57)
	Behavioural Score (items 1-4)	95 (84.1%)	5.06 (2.82)
	Impairment Score (items 5-6)	95 (84.1%)	2.04 (1.76)
	Symptomatic Score (items7-9)	95 (84.1%)	5.73 (1.76)
	Social Score (items 10-13)	95 (84.1%)	9.12 (2.72)
WHOQOL -BREF	Psychological	70 (61.9%)	35.93 (22.39)
	Social	70 (61.9%)	51.83 (24.55)
	Physical	70 (61.9%)	54.23 (21.27)
	Environmental	70 (61.9%)	67.27 (17.58)
BYI-II*	Self Concept	65 (57.5%)*	34.69 (11.99)
	Anxiety	41 (36.8%)*	64.93 (15.05)
	Anger	40 (35.4%)*	60.10 (13.16)
	Depression	63 (55.7%)*	66.68 (16.34)
	Disruptive Behaviour	21 (18.6%)*	56.61 (12.39)
PCS	Total Score	95 (84.1%)	10.03 (1.96)

Up to three BYI-II were administered per young person.

Mean CGAS admission scores (43.37; S.D.: 10.86) indicate moderate impairment in functioning in most areas or severe impairment in one area (Shaffer et al., 1983). Of the 95 young people with a completed CGAS all received a score of 70 or below, indicating impairment in functioning, and the large majority (77%, 73 service users) scored lower than 50, indicating significant impairment. Of the 95 service users with a completed HONOSCA, mean HoNOSCA Clinical Scores at admission were 22.94 (S.D.: 5.57). The three highest-rated HoNOSCA items were emotional symptoms, where 77% of young people (n=73) were rated as severe (a score of 3 or above); poor school attendance (61%, 58 young people rated 3 or above); and problems with family life and relationships (46%, 44 young people). The PCS mean total score (10.03, S.D.: 1.96) indicates slightly more complex cases than found within a Tier 3 CAMHS (Total Score 9, Yates et al., 1999).

Mean transformed admission scores on the WHOQOL-BREF were below average, indicating difficulties across all domains. The mean score in the Psychological Domain (35.93; S.D.: 22.39) was over two standard deviations below average normed scores (Hawthorn, Herrman, & Murphy, 2006) and mean scores in the Social Relationship (51.83; S.D.: 24.55) and Physical Domains (54.23; S.D.: 21.27) were over one standard deviation below average. The mean Environment Domain score (67.27; S.D.: 17.58) was the closest to an average normed score of 75.1 (S.D.: 13.0).

A T-score of 40 or above on the BYI-II Self Concept is within the normal range (Beck et al, 2004). Mean T-score at presentation to the ITS was 34.69 (S.D.: 11.99), indicating self-concept within the much lower than average range. A T-score of 55 or below for all other BYI-II’s is within the average range. At admission, mean Anxiety (64.93; S.D.: 15.05), Depression (66.68; S.D.: 16.34) and Anger (60.10; S.D.: 13.16) T-scores were within the moderately elevated category, and the mean Disruptive Behaviour T-score (56.61; S.D.: 12.39) was within the average category.

Outcomes

In total, 71 young people were admitted and discharged from the ITS between October 2010 and December 2011 and therefore eligible for analysis of pre and post-intervention measures, equating to 63% of accepted referrals within this time frame. Outcome measures were unable to be collected for 18 of these young people at admission, and of the 53 remaining young people, clinician-rated outcomes were completed for 92% (49) of young people, but only 49% (26 young people) of self-report measures were completed as a large number were not returned following discharge. Median ITS admission for these young people was 17 weeks (range 2–48).

Statistical analysis of admission and discharge data are reported in Table 2 where, following a Shapiro–Wilk test, parametric Paired Sample t-tests or non parametric Wilcoxon Signed Ranks tests were used. Mean HoNOSCA Clinical Scores decreased by an average of 6.94 points following intervention (Z=-4.725, p<0.001) and 65% of young people (n=32) showed a clinically relevant change of four or more points (Sharma, Wilkinson, & Fear, 1999; Simpson, Cowie, Wilkinson, Lock, & Monteith, 2010). Mean CGAS scores increased by an average of 16.42 points, indicating significant improvements in service users functioning following intervention (Z=-5.521, p<0.001) and clinical improvements from ‘obvious’ to ‘some noticeable problems’ (Shaffer et al., 1983).

Table 2.

Outcome Data for the CAMHS ITS.

Measure		n	Admission Mean (SD)	Discharge Mean (SD)	Wilcoxon Z	Significance
CGAS	Total Score	49	43.44 (9.96)	59.86 (14.19)	−5.52	p<0.001
HoNOSCA	Clinical Score	49	22.94 (5.57)	16.0 (8.39)	−4.73	p<0.001
	Behavioural Score (items 1–4)	49	5.24 (2.83)	3.54 (2.57)	−4.12	p<0.001
	Impairment Score (items 5–6)	49	2.18 (1.65)	1.76 (1.52))	−1.75	n.s , p=0.79
	Symptomatic Score (items7–9)	49	5.82 (2.54)	4.10 (2.95)	−3.57	p<0.001
	Social Score (items 10–13)	49	9.40 (2.56)	6.45 (3.72)	−4.50	p <0.001
WHOQOL – BREF	Psychological	23	33.43 (22.22)	50.09 (27.92)	−2.12	p <0.05
	Social	23	52.13 (23.77)	58.04 (28.54)	−0.35	n.s , p=0.73
	Physical	23	53.43 (24.10)	65. 26 (24.71)	−2.73	p<0.05
	Environmental	23	70.0 (21.69)	77.87 (20.18)	−2.02	p<0.05
					Paired Sample t
BYI-II*	Self Concept	22*	30.57 (10.82)	36.32 (10.59)	−2.617	p<0.05
	Anxiety	10*	64.60 (13.71)	55.18 (18.63)		n.a.
	Anger	8*	63.13 (11.25)	58.5 (17.01)		n.a.
	Depression	22*	68.36 (16.32)	59.5 (15.66)	2.24	p<0.05
	Disruptive Behaviour	3*	64.00 (12.17)	60.33 (15.37)		n.a.

Up to three BYI-II were administered per young person.

Analysis of changes in the four domains of the WHOQOL-BREF indicated significant improvements within the Psychological (mean: 33.43 to 50.09), Physical (mean: 53.43 to 65.26) and Environment Domains (mean: 70.0 to 77.87) although, despite trends of improvement within the Social Relationship Domain (mean: 52.13 to 58.04), it did not meet significance. The mean Psychological Domain improved from two standard deviations to one standard deviation below average normed scores (Hawthorn et al., 2006) and the Social and Physical Domains went from over one standard deviation below average to within the average range. The Environment Domain was within the average range on admission therefore remained within the average clinical norm at discharge.

Administration of BYI-IIs was dependent upon clinical presentation, therefore restricted sample sizes means that statistical analysis was only appropriate for the Self Concept and Depression Inventories, which both showed significant improvements at the point of discharge (p<0.05). Although Self Concept ratings improved significantly, mean scores post intervention still remained with the ‘much lower than average’ clinical range. Clinically significant changes were noted in mean depression and anger ratings, which improved from moderately to mildly elevated, and mean anxiety scores which improved from moderately elevated to within an average range. Although the disruptive behaviour mean score improved, mean scores at discharge remained within the ‘mildly elevated’ range.

Factors associated with change in CGAS and HoNOSCA clinical scores

Gender differences were not associated with improvement in HoNOSCA (t=0.17, df=48, p=0.43) or CGAS scores (U (48)=-0.44, p=0.66) following intervention. Correlation between age and change in HoNOSCA narrowly missed significance (r=-0.06 N=49, p=0.06) but a significant association was found between age and change in CGAS (r=-0.32, N=49, p<0.05), indicating that younger children showed greater improvements. Length of intervention was not significantly correlated with changes in HoNOSCA (r=-0.10, N=49, p=0.49) or CGAS (r=0.22, N=49, p=0.13); however, number of ITS appointments offered was significantly correlated with change in both measures (HoNOSCA: r=0.35, N=49, p<0.05) (CGAS: r=-0.30, N=49, p<0.05), where more ITS appointments were associated with greater clinical and functional improvements. Significant correlations were found between admission CGAS scores and total change in CGAS (r=-0.41, N=49, p<0.05) and admission HoNOSCA scores and total change in HoNOSCA (r=3.35, N=49, p<0.05). Furthermore, a significant negative correlation was found between change in CGAS and HoNOSCA Scores (r=-0.75, N=49, p<0.001) indicating consistency across clinician measures, but no other significant correlations were found between change in HoNOSCA or CGAS and self-report measures.

Outcome of combined ITS and inpatient unit care

Some 28 young people were admitted to the IPU at the first point of contact with Tier Four and, due to the policy of allocating an ITS CPN at the point of admission to the IPU, they were also classified as being admitted to the ITS. Median length of IPU stay was 17.5 days (range: 1–117) for these young people, significantly lower than the national median of 82 days (Tulloch et al., 2008). After completion of a short-term assessment with ITS, four of these young people were discharged promptly to alternative CAMH services, therefore outcome measures were not completed, and five young people are still active with ITS, therefore not included in the following analysis. Mean CGAS ratings at IPU/ITS admission for the remaining 19 young people was 46.0 (S.D.:11.12) which, although higher than the mean ITS CGAS rating of young people who were never admitted to the IPU (44.28; S.D.: 10.95), is not significantly different (p=0.55). Combined IPU and ITS length of stay for these young people was a median of 17 weeks (119 days, range: 9–247 days) and CGAS ratings significantly improved to a mean of 57.11 (S.D.: 11.59) following this package of care (t (18)=-4.863 p<0.001); a mean clinical improvement of 11.4 points (S.D.: 9.8) (effect size 1.12).

Discussion

Demographic information infers that the CAMHS ITS studied here is targeting an equivalent profile to the national demographic of inpatient admissions (Tulloch et al., 2008), despite the majority of young people never being admitted to the IPU (63%). The predominant presenting difficulties of mood disorders with self-harming behaviour and eating disorders were therefore to be expected. Intervention for young people at risk of a psychotic disorder is provided within a separate Tier Four CAMH service, therefore relatively low numbers are accountable.

A mean CGAS score of 43 at the point of admission to the ITS highlights significant impairments in service users’ functioning, indicating severe problems in at least one area (Shaffer et al., 1983). When service users who were admitted to the IPU during ITS involvement were excluded, the mean CGAS rating of 44.28 is still within reported IPU admission ranges (Green et al., 2000; mean CGAS: 44.0). The mean HoNOSCA Clinical Score at admission (22.94) is comparable with that reported by another Scottish ITS (Simpson et al., 2010; mean: 19.12). Self-report measures indicate significant difficulties across range of clinical areas at the point of admission. Therefore, the ITS appears to be targeting a population with clinical and functional impairments in excess of that reported in a Tier 3 CAMHS setting (Garralda, Yates, & Higginson, 2000).

Service users’ functioning was found to significantly improve following intervention by an average of 16 CGAS points over a median 17-week admission (range 2–48). A lack of control group means that these improvements cannot wholly be attributed to the ITS, particularly as 37% of accepted referrals experienced an IPU admission at the start, or during the course, of ITS intervention. However, this balanced care approach has facilitated functional improvements for ITS service users which are greater than outpatient CAMHS treatment alone (Garralda et al., 2000), where a 7-point CGAS change over a 6-month treatment period was reported, and an average inpatient admission (Mean CGAS change: 12 points; Green et al., 2007). Indeed, while sample sizes are significantly smaller, the 19 young people who were admitted to the IPU at first contact with these CAMH Tier Four services experienced an equivalent level of clinical improvement on CGAS score (11.4 points), with a combined IPU/ITS length of stay of 17 weeks (of which IPU median admission: 18 days, range 1–109 days), as that reported by Green et al. (2007) over an average inpatient admission of 16.4 weeks.

Health gains were noted on HoNOSCA Clinical Scores, which significantly decreased following intervention by an average of 6.94 points. This is greater than HoNOSCA change found following community outpatient treatment, where a 3.61 point decrease was noted at 6 months (Garralda et al., 2000) and is roughly equivalent to change following an average inpatient admission (mean HoNOSCA Clinical Score change: 5.76; Green et al., 2001). These results cannot be attributed wholly to ITS intervention; however, a similar CAMHS ITS reported mean HoNOSCA Clinical Score changes of 10.95, from an average of 19.12 at admission to 8.17 at discharge (Simpson et al., 2010), although median length of intervention appeared to be longer at 20 weeks in comparison with the CAMHS ITS described here.

HoNOSCA and CGAS change was associated with initial case severity, a finding which has been replicated in other studies (Garralda et al., 2000; Simpson et al., 2010). Finally, although length of intervention was not significantly correlated with changes in HoNOSCA or CGAS, number of appointments offered was, where more ITS appointments was associated with greater clinical and functional improvements, highlighting the importance of intensive community treatment as a component of overall CAMHS care.

Study limitations

The nature of this clinical population, particularly the severity and nature of difficulties experienced, means that control groups are practically and ethically difficult, but the absence of such a group is a major limitation in this study. The significant clinical and functional improvements cannot wholly be attributed to intervention by the CAMHS ITS studied here; however, they do represent outcomes associated with an ITS which is embedded within a Tier Four CAMHS, implementing a balanced care model where outreach services and inpatient units are part of a spectrum of intensive mental health care (Scottish Executive, 2004). A further limitation is a potential positive bias, as ITS clinicians, instead of independent assessors, administered clinician-rated measures, particularly as clinician measures showed significantly greater improvements than self-report measures. However, this effect has also been observed within CAMHS inpatient settings (Gowers, Levine, Bailey-Rogers Shore, & Burhouse, 2002) in comparison with outpatient settings, where the authors highlighted the role of service users’ attitude to admission and discharge, insight and the use of objective intensive clinical observation as contributing factors, which may also be applicable to an ITS. Finally, although self-report measures were collected and showed significant improvements, completion rates were comparatively low and it may be that young people who did not return these measures represent service users who have not engaged with the service or who did not experience the same level of clinical change.

Summary and conclusions

This study highlights the potential of a balanced care approach where young people with severe mental health difficulties are managed in the community wherever possible (Scottish Executive, 2005b) while patient safety and quality of care is maintained, as emphasised by significant clinical and functional improvements on clinician-rated and self-report outcome measures.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author biographies

Fiona Duffy is a Clinical Psychologist in NHS Lothian CAMHS Intensive Treatment Service. Her research interests include the effectiveness of interventions and services for adolescents with severe and complex difficulties, including eating disorders, self harm and impulsive risk taking behaviours.

Joanne Skeldon is a Trainee Clinical Psychologist at the University of Glasgow. She previously worked as an Assistant Psychologist in NHS Lothian CAMHS Intensive Treatment Service.

References

Beck

J. S.

Beck

A. T.

Jolly

J. B.

Steer

R. A.

(2005). Beck Youth Inventories, Second Edition for Children and Adolescents (BYI-II). San Antonio, Texas: Harcourt Assessment, Inc.

Blantz

Schmidt

M. H.

(2000). Practitioner review: Preconditions and outcome of inpatient treatment in child and adolescent psychiatry. Journal of Child Psychology and Psychiatry, 41, 703–712

Duffy

Skeldon

(2012). The impact of the development of a CAMHS Intensive Treatment Service and service redesign on psychiatric admissions. Child and Adolescent Mental Health, (in press).

Garralda

M. E.

Yates

Higginson

(2000). Child and adolescent mental health service use HoNOSCA as an outcome measure. British Journal of Psychiatry, 177, 52–58.

Gowers

S. G.

Harrington

R. C.

Whitton

Lelliott

Beevor

Wing

Jezzard

(1999). Brief Scale for measuring the outcomes of emotional and behavioural disorders in children: Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA). British Journal of Psychiatry, 174, 413–416.

Gowers

S. G.

Levine

Bailey-Rogers

Shore

Burhouse

(2002). Use of a routine, self report outcome measure (HoNOSCA-SR) in two adolescent mental health services. British Journal of Psychiatry, 180, 266–269.

Green

Jacobs

Beecham

Dunn

Kroll

Tobias

Briskman

(2007). Inpatient treatment in child and adolescent psychiatry - a prospective study of health gain and costs. Journal of Child Psychology and Psychiatry, 48, 1259–1267.

Green

Kroll

Imrie

Frances

F. M.

Begum

Harrison

Anson

(2001). Health gain and outcome predictors during inpatient and related day treatment in child and adolescent psychiatry. American Academy of Child and Adolescent Psychiatry, 40, 325–332.

Hawthorn

Herrman

Murphy

(2006). Interpreting the WHOQOL-Bref: Preliminary Population Norms and Effect Sizes. Social Indicators Research, 77, 37–59.

10.

Kurtz

(2009). The Evidence Base to Guide Development of Tier 4 CAMHS. Department of Health. Retrieved from http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_10344.

Accessed 27/02/2012

11.

Lavis

Hewson

(2011). How many times do we have to tell you? A briefing from the National Advisory Council about what young people think about mental health and mental health services. National Advisory Council. Retrieved from http://nationaladvisorycouncilcmh.independent.gov.uk/downloads/How%20Many%20Times.pdf. Accessed 27/02/2012.

12.

McLoone

(2004). Carstairs scores for Scottish postcode sectors from the 2001 Census. MRC Social & Public Health Sciences Unit, University of Glasgow. Retrieved from http://www.sphsu.mrc.ac.uk/library/other%20reports/Carstairs_report.pdf. Accessed 27/02/2012.

13.

National Mental Health Development Unit. (2010). Factfile 3. The costs of mental ill health. Retrieved from http://www.nmhdu.org.uk/silo/files/nmhdu-factfile-3.pdf. Accessed 27/02/2012.

14.

National Records of Scotland. (2011). Retrieved from http://www.gro-scotland.gov.uk/index.html. Accessed 27/02/2012.

15.

NHS Health Advisory Service. (1995). Together we Stand: The commissioning, role and management of child and adolescent mental health services. London: HMSO.

16.

O’Herlihy

Lelliot

Bannister

Cotgrove

Farr

Tulloch

(2007). Increased inequity in provision of child and adolescent mental health inpatient services in England between 1999 and 2006. Psychiatric Bulletin, 31, 454–456.

17.

O’Herlihy

Worrall

Banerjee

(2001). National Inpatient Child and Adolescent Psychiatry Study (NICAPS). Final report to the Department of Health. London: Royal College of Psychiatrists Research Unit.

18.

Pfeiffer

Strzelecki

(1990). Inpatient psychiatric treatment of children and adolescents: A review of outcome studies. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 847–853.

19.

Scottish Executive. (2004). Psychiatric inpatient services for children and young people in Scotland: A way forward. Child Health Support Group: Inpatient Working Group Report.

20.

Scottish Executive (2005a). Delivering for Health. Retrieved from http://www.scotland.gov.uk/Resource/Doc/76169/0018996.pdf. Accessed 27/02/2012.

21.

Scottish Executive. (2005b). The mental health of children and young people: A framework for promotion, prevention and care. Retrieved from http://www.scotland.gov.uk/Resource/Doc/77843/0018686.pdf. Accessed 27/02/2012.

22.

Shaffer

Gould

M. S.

Brasic

Ambrosini

Fisher

Bird

H. R.

Aluwahlia

(1983). A children’s global assessment scale (CGAS). Archives of General Psychiatry, 40, 1228–1231.

23.

Sharma

V. K.

Wilkinson

Fear

(1999). Health of the Nation Outcome Scales: A case study in general psychiatry. British Journal of Psychiatry, 174, 395–398.

24.

Shepperd

Harris

Doll

Bull

Gowers

James

…Pollack

(2007). Systematic review and mapping study of alternatives to inpatient care for children and adolescents with complex mental health needs. Retrieved from http://www.sdo.nihr.ac.uk/files/project/141-final-report.pdf. Accessed 27/02/2012.

25.

Simpson

Cowie

Wilkinson

Lock

Montheith

(2010). The effectiveness of a community intensive therapy team on young people’s mental health outcomes. Child and Adolescent Mental Health, 15, 217–223.

26.

Thornicroft

Tansella

(2004). Components of a modern mental health service: A pragmatic balance of community and hospital care. Overview of systematic evidence. The British Journal of Psychiatry, 185, 283–290.

27.

Tulloch

Lelliott

Bannister

Andiappan

O’Herlihy

Beecham

Ayton

(2008). The Costs, Outcomes and Satisfaction for Inpatient Child and Adolescent Psychiatric Services (COSI-CAPS) study. Report for the National Co-ordinating Centre for NHS Service Delivery and Organisation R&D. Retrieved from http://www.rcpsych.ac.uk/pdf/COSI%20CAPS.pdf

28.

World Health Organization. (2001). The World Health Report. Mental health: New understanding, new hope. Retrieved from http://www.who.int/whr/2001/en/whr01_en.pdf. Accessed 27/02/2012.

29.

WHOQOL Group. (1998). Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychological Medicine, 28, 551–558.

30.

Worrall-Davies

Kiernan

(2005). Using a virtual team: An evaluation of the Bradford CAMHS intensive home treatment approach. Leeds, UK: Bradford Children and Young People’s Partnership. University of Leeds.

31.

Yates

Garralda

M.E.

Higginson

(1999). Paddington Complexity Scale and Health of the Nation Outcome Scales for Children and Adolescents. The British Journal of Psychiatry, 174, 413–416.