We developed Supported Discharge service, an intensive community treatment team that provides an alternative to inpatient care. The teams were set up in 2011, and have been constantly evaluated. The SDS intervention was delivered by two SDS teams, one based in London and one in a rural area in Kent. Each SDS team included one consultant child and adolescent psychiatrist, one administrator, two-four whole time equivalents of CAMHS practitioners with nursing background and two-four whole time equivalents of clinical support workers.
What We Did
We developed Supported Discharge service, an intensive community treatment team that provides an alternative to inpatient care. The teams were set up in 2011, and have been constantly evaluated. The SDS intervention was delivered by two SDS teams, one based in London and one in a rural area in Kent. Each SDS team included one consultant child and adolescent psychiatrist, one administrator, two-four whole time equivalents of CAMHS practitioners with nursing background and two-four whole time equivalents of clinical support workers. The nature of the work included intensive case management, home treatment, day care in the hospital setting or any combination of the three according to need. The intensity of work provided was flexible, up to a maximum of five weekly contacts. Staff tasks included assisting young people with creating and carrying out customized care plans, psychiatric care, psychological interventions, school reintegration, optimising physical health care and social support.
The duration of treatment varied according to individual need, but with the aim of completing treatment within approximately 12 weeks followed by a referral to usual community mental health service if needed and using the Care Programme Approach as required. The SDS teams operated 8:00 to 20:00 with out-of-hours 24/7 cover available if required. The following elements of Assertive Community Treatment were used: small caseloads (four-five families per whole time equivalent), team approach, practising team leader, weekly team meetings and daily informal meeting, continuity of staffing, full responsibility for treatment services, responsibility for hospital discharge planning, no drop-out policy, assertive engagement mechanisms and work with informal support systems. The teams worked closely with in-patient services. Staff from SDS and inpatient services worked across both services to minimize the potential bias associated with staff enthusiasm in new services. SDS teams aimed to establish contact with each young person within the first 72 hours following admission. As soon as the young person’s clinical profile was consistent with intensive community treatment the young person and their family were offered supported discharge in consultation with in-patient professionals and SDS staff and the relevant community services.
Wider Active Support
Key partners included young people and their families, local authorities, emergency services, inpatient services, local commissioners, paediatric services and oversees partners
Young people have been included in all stages of the service development and service evaluation. Key features:
1. Young people helped to organize a conference on alternatives to inpatient care and offered a workshop on engagement https://www.evensi.uk/home-or-hospital-alternatives-to-camhs-inpatient-care/195463702
2. Young people helped to present our service in Germany https://www.ptk-bayern.de/ptk/web.nsf/gfx/F758E77E6BA6469AC125808E0036B03B/$file/DGKJP_Kongress_2017_Flyer.pdf
3. Young people have helped to develop the original proposal on the service development and evaluation. 20 young people (all in-patients) were involved in an extensive consultation on the priority of researching alternatives to in-patient care and the way these services should be delivered. In addition these young people indicated whether or not they would personally like to receive an alternative to inpatient
care. In addition 32 young people have been interviewed as part of a qualitative study comparing usual in-patient care with a community alternative. The key elements of feedback incorporated into this proposal included the need for out-of-hours support, flexible visiting hours, ability to support young people with school re-integration and providing timely access to national and specialist services. Patients and their families were consulted regarding the design of the evaluation study. As a result two changes were made (1) a qualitative component was developed to allow a wider evaluation of the trial conditions (2) the information sheets and one of the questionnaires (the Self Harm Questionnaire) were modified so that their language was more accessible for young people.
Looking Back/Challenges Faced
The biggest challenge was keeping all the relevant stakeholders engaged and happy during the setting up period. Better communication would have been useful. However, as the teams started delivering results, a period of stability and good collaboration ensued.
We recruit more and more professionals to join our team and have never been short of applicants.
Evaluation (Peer or Academic)
We undertook several evaluations (please also see below for specific outcomes) In a pilot study, the first 20 patients referred to SDS, largely those with delayed discharges were evaluated. Service use and clinical outcomes were collected and predictors of total contact time were investigated. Results: Significant improvements had been made with SDS. Higher baseline Strengths and Difficulties Questionnaire (SDQ) conduct scores were associated with less contact time. SDS seemed like a feasible model of care.
In a qualitative study we examined the young people’s experience of SDS and inpatient services, on a social and emotional level. Twenty young people, (10 SDS and 10 TAU) participated in a semi-structured visual interview study to examine their experiences of admission, ward-life and treatment. A thematic decomposition analysis was conducted on the data and specific themes relevant to satisfaction and engagement with inpatient services was examined in-depth. These include a) Behavioural surveillance as care surrogate and b) Managing the delicate emotional ecology of the ward: openness, triggering, sterility and relational engagements.
Finally, we explored some of the implications of these inpatient experiences for supported discharge services. Supported discharge aims to provide an alternative, more rapid pathway to community services for young people who are engaged with inpatient treatment, and this approach appeared to be very much in keeping with what the young people in this study considered to be important. Though psychiatric wards are broadly effective at producing desired clinical outcomes, inconsistencies in how emotional expressions and peer relationships are tolerated present challenges that generate additional, unnecessary distress for young inpatients.
We undertook a Randomised Controlled Trial comparing SDS and usual inpatient care. 287 patients were referred for inpatient admission during the study recruitment period. 123 patients were eligible for the study. 15 (12%) refused to participate. 108 patients were randomly assigned to a treatment group. 23 patients (21%) were from the Southwark site, 18 (17%) from the Lambeth site, 42 (39%) from the Croydon site, 16 (15%) from the Lewisham site and nine (8%) from the Kent site. For both groups, 82 patients (77%) were assessed at six-months follow-up, although hospital use data were available for 100% of patients.
Two participants, one in each treatment group, were withdrawn from the study. One patient, in the SDS arm, withdrew their consent and another, in the TAU arm, was withdrawn as he had no adequate provision of community clinical care and had to be looked after by the SDS team. The final sample comprised 106 patients.
Data were available for all 106 patients included in the final sample, 53 in each treatment group. Both treatment groups were similar for age at six-month follow-up, (mean years (SD): 16·34 (1.70) for the TAU group and 16·23 (1·54) for the SDS group. Both treatment groups had similar proportions of male and female patients: 20 (38%) and 33 (62%) respectively for the TAU group and 17(32%) and 36(68%) respectively for the SDS group. Both treatment groups had similar proportions of patients from a White British ethnicity versus those from other ethnicity groups: 24 (45%) and 29 (55%) respectively for the TAU group and 28 (55%) and 25 (45%) respectively for the SDS group. There were no significant differences in social class distribution between the two treatment groups except for the proportion of the patients from social class one (1/53 and 6/53) and four (9/53 and 5/53) in TAU and SDS respectively.
Main outcome measures
In unadjusted per-protocol analysis, there was a significant difference in overall hospital use among patients randomised to SDS (mean 47·25 days) versus TAU (mean 84·32). The ratio of the geometric mean total of inpatient hospital days between the TAU treatment group and the SDS treatment group was 1·67 (95% CI: 1·02 to 2·81), t (101) = 2·08, p=·04 as per the primary hypothesis. In adjusted analyses, treatment difference on the log scale was non-significant (-·05, 95%CI: -1·02 to ·01, p=0·057).
At 6-months follow-up, CGAS data were available for 102 patients in total, 50 in the TAU group (94%) and 52 in the SDS group (98%). Mean CGAS scores at six-month follow-up were 6% lower in the TAU group (59 ·7, SD 17·8) compared to the SDS group (63·2, SD 16·67). ANCOVA using bootstrapping and controlling for CGAS scores?? revealed a treatment difference of 4·88 (95% CI: -1·27 to 11·02, effect size 0·15) which was not significant.
At 6-months follow-up, SDQ data were available for 89 patients, 41 in the TAU group (77%) and 48 in the SDS group (91%). The mean SDQ scores at six-month follow-up were 16·17 (SD = 7·3) for the TAU group and mean = 17·64 (SD=7·07) for the SDS group. ANCOVA using bootstrapping, revealed a treatment difference of – ·26, 95%CI: -2·55 to 2 ·12, effect size <0·001. There was not enough evidence to suggest a difference in clinical symptoms (SDQ) at 6-month follow-up between the TAU and the SDS treatments, controlling for SDQ scores at baseline.
The proportion of patients who reported multiple (five or more) episodes of self-harm at six-month follow-up was 16/38 (42%) in the TAU group and 11/45 (24%) in the SDS group. Binomial logistic regression revealed that adolescents randomised to SDS were significantly less likely to report multiple episodes of self-harm, OR = 0·18, 95%CI: ·05 to ·64, p = 0·008. The odds of patients in the SDS group having multiple self-harm episodes was 82% lower than the odds of patients in the TAU group.
Employment, education and training
There was a significant difference in the proportion of adolescents who reintegrated to community schools between the SDS and TAU groups (0·81 SDS vs 0·51 TAU) Binomial logistic regression revealed that adolescents randomised to SDS were significantly more likely to attend a community school at 6-months follow-up OR = 4·14, 95%CI: 1·73 to 9·92, p = 0·001. There was a significant difference between the total number of days spent not in employment, education or training favouring the SDS group (SDS Median=49, TAU Median= 95·5, U= 665·00, p< ·004).
There was no evidence of differential effects on any of the above variables in adolescents with psychosis, with low global functioning or adolescents from minority ethnic groups.
Full economic data were available for x SDS (x%) and y TAU (x%). The cost of SDS day patient services were significantly higher in the SDS arm by around £24,000 (95% CI £18,321 to £29,783, p<0·001). Health and social care costs excluding SDS day patient services were significantly lower in the SDS arm by around £29,000 (95% CI -£53,647 to -£4,396, p 0·021). Combining all costs, the SDS group remained cheaper but differences between the groups were no longer significant with an adjusted mean difference of -£3,675 (95% CI -£27,559 to £20,209, p0·772). EQ-5D based QALYs and CGAS scores were similar between the groups at all time points and there were no statistically significant differences between the groups on either outcome at six-months follow-up. Results based on imputation for missing data were similar with no changes in terms of statistical significance in costs or economic outcomes
The cost-effectiveness acceptability curve based on QALYs suggests that the SDS has around a 60% probability of being cost-effective compared to TAU at the NICE preferred willingness-to-pay of £20,000 to 30,000 per QALY. The cost-effectiveness acceptability curve based on the CGAS suggests that SDS has a probability of at least 58% of being cost-effective compared to TAU, irrespective of a willingness-to-pay. Cost-effectiveness acceptability curves for both QALYs and the CGAS with missing data imputed, suggest that the probability of SDS being cost-effective compared to TAU is 50% or greater, irrespective of willingness to pay
The service has been evaluated in Germany. Admission rate to child and adolescent mental health inpatient units in Germany is high (54 467 admissions in 2013), resources for providing necessary beds are scarce. Alternative pathways to care are needed. Objective of this study was to assess the cost-effectiveness of inpatient treatment versus Hot-BITs-treatment (Hometreatment brings inpatient-treatment outside), a new supported discharge service offering an early discharge followed by 12 weeks of intensive support.
Methods: Of 164 consecutively recruited children and adolescents, living within families and being in need of inpatient mental health care, 100 patients consented to participate and were randomised via a computer-list into intervention (n = 54) and control groups (n = 46). Follow-up data were available for 76 patients. Primary outcome was cost-effectiveness. Effectiveness was gathered by therapist-ratings on the Children’s Global Assessment Scale (CGAS) at baseline (T1), treatment completion (T2) and an 8- month-follow-up (T3). Cost of service use (health care costs and non–health care costs) was calculated on an intention-to-treat basis at T2 and T3.
Results: Significant treatment effects were observed for both groups between T1/T2 and T1/T3 (P < 0.001). The Hot-BITs treatment, however, was associated with significantly lower costs at T2 (difference: 6900.47s, P = 0.013) and T3 (difference: 8584.10s, P = 0.007). Bootstrap cost effectiveness ratio indicated that Hot-BITs was less costly and tended to be more effective at T2 and T3. Conclusions: Hot-BITs may be a feasible cost-effective alternative to long inpatient stays in child and adolescent psychiatry.
Is there any other information you would like to add?
The service is now widely available in the UK and has received federal funding in Germany https://www.gkv-spitzenverband.de/krankenversicherung/krankenhaeuser/psychiatrie/stationsaequiv_psych_behandlung/st_aequ_beh.jsp