Introduction
Health initiatives including the Improving Access to Psychological Therapies (IAPT) programme have aimed to identify new and effective approaches to the delivery of services to meet current demands and preferences. Dependent on presentation, a ‘relatively brief low-intensity intervention’ (Department of Health, 2008, p. 22) is recommended in the first instance, including computerized cognitive behavioural therapy (CCBT).
The shift within mental health services to promoting patient choice, adopting a holistic approach and the value of lived experiences is evident in guidance and policy. In acknowledging this, growing attention is being given to the fundamental role that peer workers can play (e.g. Mahlke et al., Reference Mahlke, Krämer, Becker and Bock2014). Peer support services are thought to support a unique relationship providing the opportunity for people with similar experiences, on a reciprocal basis, to offer ‘more authentic empathy’ (Bailie and Tickle, Reference Bailie and Tickle2015, p. 48) in using their own experiences as a tool in the journey to recovery.
The incorporation of peer support worldwide is expanding and the emerging evidence base is promising (e.g. Faulkner and Basset, Reference Faulkner and Basset2012; Fuhr et al., Reference Fuhr, Salisbury, De Silva, Atif, van Ginneken, Rahman and Patel2014) but additional research on effectiveness is required.
This paper presents an evaluation of the implementation of an innovative home-based CCBT service provided by Self Help Manchester, a service-user led, third sector organization based in Greater Manchester, commissioned by Manchester Clinical Commissioning Groups (CCGs). Outcomes from the service will be reported and compared with IAPT national service data.
Method
The @Home eTherapy service provided by Self Help provides telephone-supported CCBT to clients. Assessment and treatment appointments are offered over the telephone by an eTherapy coordinator, who has lived experience of mental health problems. eTherapy coordinators do not require any formal qualifications but take part in a two-week induction supported by monthly group supervision. Several specialist online computerized programmes are available, each based on CBT principles. These include: Living Life to the Full Interactive, SilverCloud Health, Beating the Blues, Breaking Free and Sleepio. An appropriate CCBT package is identified for those deemed suitable. Clients are offered one assessment appointment and between six and twelve 20-minute support calls, dependent on the programme.
Measures
Self-report measures used as part of the IAPT minimum data set (Department of Health, 2008) were used at each session: Patient Health Questionnaire (PHQ-9), Generalized Anxiety Disorder Scale (GAD-7) and the Work and Social Adjustment Scale (WSAS).
Individuals with scores above the clinical cut-off on the PHQ-9 (>9) and/or GAD-7 (>7) at assessment were identified as meeting ‘caseness’ for depression and/or anxiety. First and last scores were used to determine rates of recovery and extent of improvement, regardless of caseness threshold (reliable improvement). Reliable improvement is identified if there is a decrease in one or both outcome measures that surpasses the measurement error for that measure (PHQ-9 ≤ 6; GAD-7 ≤ 4) and no increase in the other beyond the error of measurement. On the other hand, if there is an increase in one or both scores that is more than the measurement error, clients are considered to have reliably deteriorated. Rates of reliable recovery, where a client is considered to have both recovered and reliably improved, was also calculated.
Data were managed and analysed using SPSS (v22). Descriptive statistics are presented. Inferential statistics are used to assess statistical and clinical significance of change on key measures. Data from the last attended appointment were used.
Clients entering the service consented for their anonymized data to be used in routine evaluations; therefore approval from a Research Ethics Committee was not required.
Results
2054 clients were referred to the service over 30 months, of whom 1355 (66.0%) attended an initial assessment and 724 (53.4%) attended at least two treatment appointments and were considered to have ‘finished a course of treatment’ [Health and Social Care Information Centre (HSCIC), 2016]. Most completing treatment were female (447, 61.7%), aged 45 years or under (579, 79.9%), with more than half (413, 57.0%) under 36 years. Almost all clients reported as being from a ‘white’ ethnic group (660, 91.2%).
Statistically significant improvements were found for treatment completers on all outcome measures. Large pre- to post-effect sizes were identified for changes in level of severity of depression (PHQ-9; Cohen's d = 0.98), and anxiety (GAD-7; d =1.07), whilst a medium effect size was identified for functioning (WSAS; d = 0.53).
On intake, 91.4% met caseness for anxiety, depression or both. Of those meeting caseness at initial presentation, 408 (61.6%) were considered ‘recovered’ at their final appointment. 410 (56.7%) of treated clients met the criteria for reliable recovery. 479 (66.2%) of treated clients met the criteria for reliable improvement from the start to end of treatment, and 32 (4.4%) were identified as having reliably deteriorated.
Not all programmes were available across all sites during the data collection period. Analysis therefore takes into account all programmes rather than individual programme outcomes. 60.6% accessed Living Life to the Full, 19.1% accessed Silvercloud Health, 4.4% accessed Beating the Blues, 3.6% accessed Sleepio and 0.1% accessed Breaking Free.
An independent samples t-test was conducted to compare appointment attendance with moving from caseness to recovery at the last attended session. There was a significant difference between the number of sessions attended for those moving to recovery (mean = 6.04, SD = 1.8) and those still meeting caseness (mean = 4.62, SD = 2.0); t (361) = 8.91, p < .01 (two-tailed). The magnitude of the differences in the means (mean difference =1.42, 95% CI: 1.11 to1.73) was large (d = 0.937).
Benchmarking outcome data
@Home eTherapy service data was compared with IAPT data. A direct comparison with IAPT guided CCBT data is provided where available (HSCIC, 2016). Clients were generally comparable in terms of demographics to those completing IAPT services overall (CCBT specific data is not available), with most being from a ‘white’ ethnic group (91.2% and 80.72%), just under two-thirds being female (61.7% and 64.9%) and similar distribution of referrals over the different age categories (HSCIC, 2016).
Table 1 benchmarks the eTherapy @Home service against current IAPT data (HSCIC, 2016). 96.3% of those finishing a course of treatment waited less than 6 weeks to enter the @Home eTherapy treatment, compared with 91.4% waiting less than 6 weeks within CCBT IAPT services nationally (HSCIC, 2016). Similar numbers completing CCBT treatment met caseness criteria at therapy intake (91.4% vs 88.6%).
Comparison of eTherapy @Home service data with IAPT service data
1 Data obtained from HSCIC (2016); data presented for guided self-help (computer) unless stated. 2Data relate to all IAPT referrals, irrespective of treatment as direct comparison with guided self-help (computer) data is not available.
An equivalent number of referrals completed an @Home treatment (53.4%) in comparison with national IAPT service data for individuals referred to guided CCBT (52.0%). For people accessing CCBT services, @Home service clients attended, on average, almost twice the number of CCBT appointments (5.6) than reported for guided CCBT treatment in national IAPT services (2.9). Recovery and reliable recovery rates exceeded national IAPT service averages and reliable improvement rates were found to be comparable to those identified in IAPT services; however, specific guided CCBT data is not available from NHS digital datasets. In line with recent work exploring the evaluation of evidence-based intervention measurement in routine practice (Delgadillo et al., Reference Delgadillo, McMillan, Lucock, Leach, Ali and Gilbody2014) the effect sizes identified for PHQ-9 and GAD-7 data are equivalent to ‘high performance benchmarks’ for national services.
Discussion
The service was associated with significant changes to outcomes, and treatment effect sizes were large. Those who completed treatment met IAPT key performance recovery target (61.6%, 56.7% reliable recovery) which exceeds national and local IAPT statistics for CCBT services. Recovery rates were also higher than previously reported for face-to-face peer-supported CCBT services (Cavanagh et al., Reference Cavanagh, Seccombe and Lidbetter2011).
Findings suggested that those who attended more treatment appointments were more likely to move from caseness to recovery, providing rationale to investigate ways to increase engagement or referral strategies for ensuring the needs of individuals who may have more complex presentations and find it difficult to engage. The flexibility of the service, which provides telephone support up to 8 p.m. on weekdays, and allows access to the CCBT programme 24/7, may have contributed to engagement and outcome.
Limitations of the study include the lack of opportunity to incorporate controlled conditions. Improvement data must therefore be interpreted with caution, as there is the possibility that clients may have fared as well with no treatment. Additionally, as follow-up data were not analysed, it is unclear if the outcomes were sustained following the last attended appointment and the reasons for discontinuing treatment. Qualitative interviews with clients, particularly those who did not engage at all or dropped out would be valuable, particularly to understand more about the ‘active ingredients’ of the service.
Directly comparing a CCBT peer supported intervention to the same intervention supported by psychological wellbeing practitioners may provide insight into the impact that such a workforce may have. The way in which the service is implemented and integrated into existing mental health services appears important, and has implications for sustainability. Literature has highlighted the importance of offering guided self-help, rather than pure self-help (i.e. Gellatly et al., Reference Gellatly, Bower, Hennessy, Richards, Gilbody and Lovell2007). Understanding more about the support clients are provided with may be important, exploring any differences and similarities to the support provided in IAPT services.
Acknowledgements
Thanks to service staff, volunteers and service users who allowed us to collect data for use in anonymized format for publication.
Financial support: This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of interest: Kate Cavanagh has been a consultant to Ultrasis plc, who market Beating the Blues.
Ethical considerations: The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, and its most recent revision.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1352465817000297
Comments
No Comments have been published for this article.