Introduction
NHS Talking Therapies for Anxiety and Depression services [formerly known as Improving Access to Psychological Therapies (IAPT) services] have been providing cognitive behavioural therapy (CBT) for people with common mental health problems (such as anxiety disorders and depression) within the UK since 2008 (Clark, Reference Clark2011). Demand for services outstrips the funding provided (Dalal, Reference Dalal2019) and, alongside difficulties recruiting and retaining staff (Cartmell, Reference Cartmell2017; Kell and Baguley, 2019), many services experience waiting lists longer than national guidance recommends (BACP, 2019). Part of the solution to this has been computerised cognitive behavioural therapy (cCBT), which provides access to CBT treatments using an online platform (Andrews et al., Reference Andrews, Cuijpers, Craske, McEvoy and Titov2010). Research suggests that this method of delivering therapy is as beneficial as other low-intensity treatments (Arnberg et al., Reference Arnberg, Linton, Hultcrantz, Heintz and Jonsson2014; Gellatly et al., Reference Gellatly, Chisnall, Seccombe, Ragan, Lidbetter and Cavanagh2018), although there are some conflicting findings (Gilbody et al., Reference Gilbody, Littlewood, Hewitt, Brierley, Tharmanathan, Araya, Barkham, Bower, Cooper, Gask, Kessler, Lester, Lovell, Parry, Richards, Andersen, Brabyn, Knowles, Shepherd and White2015). Practically, cCBT offers the potential to increase the capacity of the practitioner to support more clients at any one time; thus, increasing access to low-intensity CBT (Andrews et al., Reference Andrews, Cuijpers, Craske, McEvoy and Titov2010), whilst also reducing the high levels of burn-out experienced by IAPT workers (Koomson et al., Reference Koomson, Haque, Sivananthan, Jayaraajan, Hussain and Ahmad2020; Scott, Reference Scott2018).
There are a range of NICE-approved cCBT programmes designed in line with the CBT model (Palacios and Richards, Reference Palacios and Richards2019). This study focused on one particular platform called SilverCloud which is popular within NHS Talking Therapies services, with 85 services across England offering this to their clients (National Institute for Health and Care Excellence, 2018). However, there is no reason to suspect that these findings are specific to this one package of cCBT or are not equally applicable across other similar packages.
The cCBT platform used in the study includes a range of CBT interventions and can be tailored to the client’s presenting problem, with a variety of base programmes for anxiety, depression, co-morbid anxiety and depression, stress and specific anxiety disorders. Each base programme can then have additional modules added by the supporter to give the client access to additional content if it is relevant (e.g. sleep, anger, bereavement). The content varies on the programme, but it generally includes a mixture of psychoeducation and guided exercises to help develop core CBT skills such as self-monitoring, awareness of thoughts and emotions, and ways to change cognitive content and processes (Eilert et al., Reference Eilert, Timulak, Duffy, Earley, Enrique, Kennedy, McCormack, Palacios, Wogan and Richards2022). Psychoeducation is presented in video and written formats, with interactive exercises that support the client to reflect on and apply the techniques to their own thoughts, feelings and behaviours. If a client is allocated to cCBT following an assessment they are given instant access to the online content, which they can complete at their own pace at a time that is convenient for them. Within the service in the study, they are always supported by a Psychological Wellbeing Practitioner (PWP). The client’s progress on the platform is visible to their allocated PWP, which allows the PWP to provide encouragement and additional relevant support. The client and PWP mainly communicate via asynchronous messaging via the programme, allowing the client to highlight any problems they are having with the content or techniques that they may not understand, for example. Clients usually complete the programme within 8 weeks but have access to the online content for a year following discharge to support with relapse prevention. Within the service in the study, cCBT was always only offered to patients identified as suitable for low-intensity CBT intervention and was purely offered as a PWP-supported option.
Unfortunately, despite the extensive evidence base for the efficacy and effectiveness of cCBT (Barak et al., Reference Barak, Hen, Boniel-Nissim and Shapira2008; Sharry et al., Reference Sharry, Davidson, McLoughlin and Doherty2013), uptake in the service had been extremely limited. A systematic review of cCBT in rural areas by Vallury et al. (Reference Vallury, Jones and Oosterbroek2015) reported evidence that this mode of treatment is more acceptable to patients in rural areas when compared with urban services. However, health care professionals’ views towards cCBT as a form of treatment for patients appears mixed (Simon et al., Reference Simon, Ploszajski, Lewis, Smallman, Roberts, Kitchiner, Brookes-Howell and Bisson2021). For example, a study by Waller and Gilbody (Reference Waller and Gilbody2009) found that staff views around using cCBT to treat patients presenting with mental health difficulties were largely negative. The researchers predicted that these views were likely to have had an adverse effect on cCBT being offered as a form of treatment to patients.
This service evaluation adds to the current literature exploring staff perspectives of online CBT interventions. This study focuses on factors that may influence a practitioner’s opinion of cCBT, specifically job role, experience of cCBT and training. This evaluation collected data before and during the COVID-19 pandemic and therefore the impact arising from changes to remote working during this time has been considered. For example, Wind et al. (Reference Wind, Rijkeboer, Andersson and Riper2020) hypothesised early in the COVID-19 pandemic that this unexpected event might act as a catalyst for change and cause staff to be increasingly accepting of cCBT as a helpful and accessible form of treatment.
The study was carried out in an NHS Talking Therapies service offering a range of talking therapies to people with depression and anxiety disorders in Northern England. Unfortunately, due to a mismatch between the NHS commissioning areas and the local council areas, it is impossible to provide detailed and precise data on the demographics of the area included. However, the available data suggests that the area is much more sparsely populated than the national average, being a mix of rural and urban areas with pockets of extreme deprivation within both. The area also has a much older age profile than the national average and lower levels of ethnic diversity. Public transport tends to be extremely limited, but there may also be issue around digital inclusion related to socioeconomic factors as well as limited to access to broadband in some areas.
Using the same methodology as Meisel et al. (Reference Meisel, Drury and Perera-Delcourt2018), this evaluation surveyed NHS Talking Therapies staff (those that offer assessments and the leadership team) on their perspectives of offering cCBT as a treatment option following an initial telephone appointment. This original service evaluation aimed to replicate this study within a largely rural NHS Talking Therapies service. However, 2 months after the original data collection, the COVID-19 pandemic occurred, and the service (like all others) was transformed with respect to service delivery methods, and this provided an opportunity to evaluate if this also changed staff beliefs in a commensurate manner. A further data collection time point was added in to see if further training impacted on staff beliefs as well. The data were therefore collected at three distinct time points. Due to the initial data collection being intended to replicate and extend the Meisel et al. (Reference Meisel, Drury and Perera-Delcourt2018) study, no formal hypotheses were established for time point 1 and the data collection at time point 2 was purely opportunistic. However, the training between time points 2 and 3 was intended to provide additional information for practitioners and help them feel more confident in understanding cCBT and allocating patients for this treatment.
The service in the study has consistently allocated fewer people than expected to cCBT and does not force patients to access cCBT to be able to access other options (nor is it offered unsupported or for people waiting for high-intensity CBT). Their aim has been to use cCBT as one treatment option that patients could choose (if desired) from a range of low-intensity treatment options. The initial aim of the study was to understand what staff beliefs might be acting as barriers to cCBT being offered or accepted and was expanded to include post-pandemic beliefs. If key beliefs could be identified that were not in line with the evidence, the plan was to find a way to provide targeted information that might help clinicians to appropriately offer cCBT, thus improving access for patients.
Method
A descriptive methodology was applied in this service evaluation study to explore staff views on providing cCBT as a treatment option to patients. Using a repeated measures design, data were collected at three distinct time points.
Time point 1 (collected pre-COVID pandemic in January/February 2020). This original data collection aimed to identify staff perspectives on cCBT including advantages, barriers/problems, utility and confidence.
Time point 2 (collected post- COVID pandemic in January/February 2022). This time point aimed to identify if there had been changes in staff beliefs following the large increase in remote working due to the COVID-19 pandemic.
Time point 3 (collected in September/October 2022). Data were collected at this time point after staff members attended a 2-hour training session focused on the particular cCBT package, to further investigate if this intervention had altered staff perceptions of advantages, disadvantages, etc.
Due to the changeability in staff turnover over time, we were not able to guarantee that the same staff members completed all three questionnaires. Informed consent to complete the questions was provided by each participant at the beginning of the questionnaire, prior to completing the remaining responses.
Questionnaire
The current study employed a questionnaire, which was developed and implemented by Meisel et al. (Reference Meisel, Drury and Perera-Delcourt2018) and administered to staff members in the service. The questionnaire was designed in line with literature that looked at the positives and negatives of cCBT (Donovan et al., Reference Donovan, Poole, Boyes, Redgate and March2015; Stallard et al., Reference Stallard, Richardson and Velleman2010) and from discussions that emerged in staff meetings (Meisel et al., Reference Meisel, Drury and Perera-Delcourt2018). For this service evaluation, additional questions were added at later time points to explore the impact of the COVID-19 pandemic, as well as incorporating the shift to offering remote therapy. A theme that arose from the previous literature was staff wanting more guidance and training in using and offering cCBT. Therefore, it was felt that a question around what staff would like training to look like was important to incorporate in the questionnaire. This would help staff tailor any training to the needs of the team and allow this to be explored in the third data collection time point.
The questionnaire collected information about the participants’ current position in the team, and the level of training that they had completed to deliver cCBT. The next section of the questionnaire focused on exploring staff views on the strengths and difficulties of cCBT as a treatment option, compared with face-to-face and remote therapy, and allowed them opportunity to comment further on their thoughts. Participants were also asked to consider what would assist someone to encourage cCBT as a treatment option to clients and what they would like to be included in any staff training.
Time point 1
For the first time point, responses were collected from 62 members of staff (81 were employed in clinical and/or leadershipFootnote 1 roles within the service at the time and provided with the opportunity to take part). Questionnaires were distributed during face-to-face team meetings, and time was provided for staff to complete them during the meeting. Those who were not able to attend the team meetings were contacted by email and offered the opportunity to complete the questionnaire online at a time convenient to them.
Time point 2
For the second time point, there were a total of 79 practitioners (employed in clinical and/or leadership roles) working within the service. A total of 54 participants completed the questionnaire. The questionnaire used was the same as time point 1, but with the addition of two questions: ‘Please write any comments on how you feel the COVID-19 pandemic has impacted on supporting clients through cCBT at [service name]?’ and ‘If you were to receive training on cCBT, what do you feel you would like it to involve to make you feel more confident in supporting cCBT?’. The questionnaire was administered via a Qualtrics link during the online team meetings in which staff were provided with an allotted time to complete the questionnaires. Alternatively, any staff that missed the meetings were emailed the questionnaire and invited to take part.
Time point 3
For the third timepoint, there were a total of 81 practitioners (employed in clinical and/or leadership roles) working within the service. A total of 32 participants completed the questionnaire. However, only 56% (n=18) of respondents had attended the training session delivered by the platform provider on how to use the cCBT program. Again, the questionnaire used was the same as at time point 2, but with the addition of an additional question asking participants whether they had attended and completed the recent cCBT training. The questionnaire was administered via a Qualtrics link during the online team meetings in which staff were provided with an allotted time to complete the questionnaires, with additional staff emailed the questionnaire and invited to take part.
Training
The training offered before time point 3 was developed and delivered by the cCBT platform provider and covered the challenges and barriers to using cCBT in an NHS Talking Therapies service, an overview of research evidence, testimonials from previous clients, a description of the role of the supporter, the benefits of cCBT for an NHS Talking Therapies service, and examples of how to talk to clients about cCBT. The training also included a demonstration of the cCBT programs and supporter portal, and an opportunity to ask questions.
Data analysis
Staff perspectives gathered using Likert scale responses to set statements were analysed as a percentage agreement for each statement. The data were analysed in this way for all three time points and are displayed in graphs in the Results section, with a description of how opinions may have changed over time.
The open questions from all respondents were analysed drawing on a content analysis approach (White and Marsh, Reference White and Marsh2006), with the aim of identifying patterns in the qualitative data. The author initially coded the data to identify themes in the content and capture staff views. Using criteria from a developed framework of trustworthiness (Guba and Lincoln, Reference Guba and Lincoln1989; Lincoln and Guba, Reference Lincoln and Guba1985) for qualitative research, the data for all three time points was initially coded and then re-coded later to further capture the themes. A peer debriefing process was engaged in as part of the analysis: a second researcher reviewed the coding of questions and themes to check on the meaning. Quotes are also provided within the Results section for each question to evidence the patterns and themes that were identified.
Results
Client characteristics
We asked staff at all three time points about how much they agreed with how appropriate cCBT would be for the following groups of people:
-
people with mild/moderate mental health difficulties;
-
people with severe mental health difficulties;
-
older adults, people with few years of formal education;
-
unemployed;
-
people whose first language is not English.
Figure 1 shows the percentage of participants who agreed that cCBT would be a suitable treatment option for each of these groups at each time point.
Figure 1. Client characteristics.
Seventy-two per cent of staff agreed at time point 1 that cCBT would be effective for treating mild/moderate mental health difficulties. This increased over time to 80% and 79% at time points 2 and 3, respectively. Similarly, the percentage of staff who disagreed that cCBT would be effective for severe mental health problems decreased across the time points. These findings suggest a more positive view of what cCBT can be effective for over time. cCBT is designed as a low-intensity intervention and the stepped care model (National Collaborating Centre for Mental Health, 2023) suggests that severe mental health issues should be treated at high intensity or secondary care. More staff indicated ‘neither disagree or agree’ at time points 2 and 3 for the cCBT being effective for severe mental health difficulties and so it is possible that having seen cCBT used more often during the pandemic, that they were more open-minded to how this could be used.
Across all three time points, suitability of cCBT for the client and their presenting problems was reported to impact whether it would be offered as a treatment option. Some staff provided examples of mental health issues that they thought would not benefit from a cCBT intervention. Thus, there was a pattern around case complexity influencing the appropriateness of cCBT. This supports the participants’ differences in agreement for whether cCBT would be effective for mild/moderate or severe mental health difficulties (Fig. 1), which indicate that cCBT is seen as more acceptable to staff for treating mild/moderate difficulties than severe difficulties.
We also provided a free text space, asking participants if there were any other client characteristics which would make them less likely to offer cCBT as a treatment option.
‘Too severe/chronic MH problems’ [time point 1]
‘If have social anxiety may perpetuate problem. Those lacking social contact perhaps?’ [time point 1]
‘I think if there are specific risk or safeguarding concerns to consider’ [time point 2]
‘Presenting problems which low intensity interventions are not recommended for i.e. PTSD’ [time point 2]
‘Complex presenting problem’ [time point 3]
‘Complex trauma’ [time point 3]
When asked about age, education, employment and language, staff opinions differed on whether they felt that the effectiveness of cCBT would be hindered (Fig. 1). When considering age, over 50% of staff disagreed that cCBT would not be suitable for older people at all time points. Although the percentage of staff who disagreed that cCBT would not be suitable for people with fewer years of formal education decreased between time points 2 (10%) and 3 (0%), no participants agreed with this statement at time point 3, with more participants feeling that they did not agree or disagree (52%). The percentage of staff who disagreed that cCBT would not be suitable for people who are unemployed increased over time points, with no staff agreeing with this statement at time points 2 and 3. This suggests some shift in opinions about who can benefit from cCBT, with practitioners finding cCBT a more acceptable option for people with fewer years of education or who are unemployed.
Staff agreement and disagreement about whether cCBT would not be suitable for people whose first language is not English varied more widely than for the other statements, with less disagreement over the time points. However, this statement does not clarify the level of English that a patient may have, even as their second language. Some staff may have considered clients who need interpreters when responding, whereas others may have been thinking of clients who speak and understand English very well as a secondary language.
When given the opportunity to describe additional client characteristics that might make them less likely to offer cCBT as a treatment option, staff reported that a lack of computer literacy and access to appropriate technology as a reason that would make staff less likely to offer cCBT, across all three timepoints.
‘Clients would have to have access to a computer and be computer literate’ [time point 1]
‘People with no access to necessary equipment or very limited IT skills’ [time point 2]
‘Clients not comfortable with modern technology’ [time point 2]
‘Lack of confidence in using computers/tablets’ [time point 3]
At all time points, patient choice was also raised, suggesting that a patient’s preference for individual face-to-face CBT would be an important factor for practitioners. However, it is not clear from the data if the choice stated by the client would only influence allocation to cCBT or if practitioners make judgements about what to offer clients based on their perception of what their preference will be.
‘Patient preference’ [time point 1]
‘Clients preferring to speak to therapist’ [time point 2]
‘People who prefer face to face’ [time point 3]
‘Client wish and preference for talking therapy’ [time point 3]
At time point 2 only, staff highlighted that a lack of opportunity for a therapeutic relationship with a clinician would be an important factor.
‘Clients preferring to speak to a therapist’ [time point 2]
‘Benefiting from conversation during treatment’ [time point 2]
Ten per cent or fewer of respondents at each time point thought that cCBT would be better than face-to-face CBT for depression, panic disorder, OCD and GAD. However, 45–89% of staff thought that cCBT would be as good as face-to-face CBT for these conditions. For all of the mental health conditions asked about, the percentage of staff that thought cCBT would be as beneficial as face-to-face CBT was higher at time point 3 (post-training) than at time points 1 and 2. The percentage of staff that thought cCBT would be worse than face-to-face CBT was also lower for all of the mental health conditions mentioned at time point 3 than at time points 1 and 2.
For stress, over 69% of staff felt that cCBT would be as good as face-to-face CBT at each time point. The percentage of staff who felt that cCBT would be worse than face-to-face CBT reduced at each time point. For social anxiety, however, over 60% of staff felt that cCBT would be worse than face-to-face CBT at each time point. However, this decreased at each time point. At time point 2, 12% of staff thought that cCBT would be better for social anxiety than face-to-face CBT, but this reduced to 4% at time point 3. The percentage of staff who thought that cCBT would be as good as face-to-face CBT remained similar at time points 1 and 2 but increased to 36% at time point 3.
At time points 2 and 3, participants were also asked their opinion on whether they thought that cCBT would be better than, the same as, or worse than remote (via video link) CBT with a therapist. Few staff felt that cCBT would be better than remote CBT, and this reduced at time point 3 to 0% for four out six of the mental health issues (Fig. 3). However, more than half of respondents at both time points felt that cCBT would be the same as remote CBT with a therapist for all the mental health issues, except for social anxiety. Social anxiety showed higher rates of staff believing that cCBT would be worse than remote CBT compared with the other mental health issues asked about. This was also the case when asked about face-to-face CBT (Fig. 2).
Figure 2. Comparison with face-to-face CBT.
Figure 3. Comparison with remote CBT.
Advantages
Participants were asked at all three time points how much they agreed with 12 suggested benefits of cCBT. Figures 4 and 5 show the percentage of participants at each time point that agreed with these statements. There was high agreement (<80%) with the following statements across all three time points: outcome measures can easily be embedded into cCBT; one advantage of cCBT is that it is available 24/7; cCBT can be easily used at home; cCBT can provide earlier access to evidence-based treatment for mental health difficulties. Agreement with these statements suggest that staff see the practical benefits of cCBT. Conversely, staff were mixed in their opinions for the statement about cCBT as a viable solution to a lack of qualified CBT therapists. Almost half (46–51%) of the participants disagreed with this statement at all three time points. However, a quarter of participants at time points 1 and 3, and a third of participants at time point 2, agreed that cCBT was a viable solution, with the rest of participants rating their opinion as neutral. However, cCBT is a low-intensity intervention and so it may be that staff who disagreed with this statement believed that cCBT was not a viable alternative to high-intensity CBT intervention. When comparing cCBT with face-to-face CBT (Fig. 2) and with remote CBT (Fig. 3), beliefs that cCBT would be worse than the alternative options were strongest for social anxiety and OCD, which are typically treated at step 3 of the IAPT model.
Figure 4. Advantages of cCBT (Part 1).
Figure 5. Advantages of cCBT (Part 2).
Agreement with the following statements increased across time points: face-to-face CBT can easily be adapted to be delivered online; I feel my work as a therapist is valued when offering clients cCBT; supporting clients using cCBT requires a high level of skill. Agreement was also higher at time point 3 for the following statements when compared with time points 1 and 2 (although agreement was slightly lower at time point 2 when compared with time point 1): cCBT makes it easy for clients to share information about their current difficulties; cCBT reduces the stigma of accessing mental health services.
Interestingly, opinions about if cCBT saves time for therapists were mixed at all three time points. At time points 1 and 2, 22% of staff disagreed with this statement. However, neutral and agreement responses differed between time points 1 and 2, with only 37% agreeing at time point 1 and 51% agreeing at time point 2. At time point 3, disagreement with this statement increased from the previous time points to 29%, and agreement reduced from time point 2 to 46%.
When asked if they agreed that cCBT is preferable to other forms of self-help, over 50% of staff said ‘neither agree nor disagree’ at all three time points. Agreement with this statement increased at time point 2 but reverted to the same level of agreement as time point 1 at time point 3. However, without further feedback from the participants, it is not possible to know what other forms of self-help (i.e. psychoeducation groups, bibliotherapy, low-intensity CBT sessions with a PWP) participants were comparing with when they shared their beliefs for this question.
Participants were then given the opportunity to identify any other advantages to using cCBT. Across all three time points, staff reported that an advantage of cCBT is the accessibility and convenience of it for clients; they have increased choice, and it is more accessible and flexible than CBT.
‘I suppose it allows for working at slower/faster pace – patient doing more some weeks and less others’ [time point 1]
‘Flexibility with time able to spend – can fit around other responsibilities’ [time point 2]
‘More accessible for those with young children who have no childcare options’ [time point 2]
‘Access patients in remote rural locations’ [time point 3]
At all three time points, a theme was identified around cCBT benefiting the service, with an emphasis at time point 1 that cCBT is advantageous to the therapists themselves.
‘Possibly keeps the therapist more emotionally removed from the client’s difficulties and so helps prevent therapist burn-out/fatigue’ [time point 1]
‘Can reduce burden on therapists’ [time point 2]
‘Less therapist time’ [time point 2]
‘Helps waiting lists’ [time point 3]
‘Can be used as a build up to face to face’ [time point 3]
Barriers
Participants were asked at all three time points how much they agreed with 13 suggested disadvantages of cCBT. Figures 6 and 7 show the percentage of participants at each time point that agreed with these statements. For nine of these, disagreement increased over the three time points. Additionally, two showed higher agreement at time point 3 compared with time point 1, although agreement with these statements increased at time point 2: most clients will not complete the assigned exercises using cCBT; most clients will drop out of cCBT.
Figure 6. Barriers to using cCBT (Part 1).
Figure 7. Barriers to using cCBT (Part 2).
Figure 8. Impact of the pandemic on practice.
When asked if they agreed with the statement that most clients will have difficulty understanding the principles of CBT using cCBT, agreement was at 5% or lower for all three time points, reducing to 0% agreement at time point 3. Agreement also decreased over time points when participants were asked if they agreed with the following statements: the lack of direct therapist contact will be perceived negatively by clients using cCBT; the lack of a therapeutic relationship will result in worse treatment outcomes for patients using cCBT. When responding to ‘It is difficult to pitch materials appropriately to clients’ understanding of CBT principles using cCBT’, agreement decreased over time to 4% at time point 3, although disagreement or neither agree nor disagree were equal at 48%. Agreement to ‘It is difficult to tailor homework materials appropriately to clients’ needs using cCBT’ also decreased over time to 4% at time point 3, with over half of the participants disagreeing (52%).
The responses when asked if the lack of direct therapist contact will be perceived negatively by clients using cCBT were mixed. Agreement with this statement increased from 31% at time point 1 to 41% at time point 2, but then decreased to 22% at time point 2. Disagreement, however, was at 22% for time points 1 and 2, but increased to 37% at time point 3.
Participants were also asked about any other barriers or problems they could foresee with using cCBT in the service. At time points 1 and 2, staff highlighted that barrier to using cCBT was a poor evidence base. This was not identified at time point 3 (post-training).
‘Realistically the evidence base for cCBT is poor’ [time point 1]
‘The evidence base for it is not actually very good’ [time point 2]
‘Realistically the evidence base for cCBT is poor’ [time point 2]
At time points 1 and 2, staff reported a barrier associated with the client’s feelings towards cCBT as an intervention and identified patterns around client preferences and expectations. This was not identified at time point 3.
‘Clients do like a face to face appointment when offered’ [time point 1]
‘I think there is a perception that this is 2nd best to direct therapy contact’ [time point 2]
At time point 3, staff identified a barrier around availability of resources, particularly access to a stable internet connection, lack of finances and electricity. This was not identified at time points 1 and 2.
‘A growing number of people are struggling with finances, and affording phones, internet or electricity, this is likely to worsen’ [time point 3]
‘Living in rural remote areas, access to stable internet connection can be an issue’ [time point 3]
At all three time points, staff reported that they were concerned that the use of cCBT would not have a therapeutic impact, with the opportunity to manage emotions being limited and some participants felt that there was a lack of a therapeutic relationship in cCBT.
‘Lack of relationship’ [time point 1]
‘Projection/transference’ [time point 1]
‘The absence relational dynamic would be my primary concern’ [time point 2]
‘From my experience there is lack of therapeutic relationship’ [time point 3]
Increasing confidence
Staff were asked for feedback on what might be helpful in supporting them to refer more clients to cCBT. They were provided with six suggestions (see Table 1) and asked to rank them in order of helpfulness (one being the most helpful). Responses differed across time points, although training on what cCBT involves was perceived to be one of the most helpful solutions at all three timepoints. At time point 1, staff ranked raising awareness of cCBT with GPs as the most helpful suggestion, but this was later ranked in position four at time points 2 and 3. Offering telephone support alongside cCBT to maintain rapport was ranked as one of the least helpful suggestions at time points 1 and 2; however, at time point 3, staff ranked this as one of the most helpful. Training on the evidence base of cCBT was ranked at position three at time points 1 and 2 but was ranked at position five at time point 3; whereas ensuring that supervisors are familiar with cCBT to provide effective supervision was ranked as one of the least helpful suggestions at all three time points.
Participants were also asked to offer additional suggestions that they felt would be helpful in supporting them to refer more clients to cCBT. Across all three time points, staff reported that receiving training on cCBT would be helpful when offering it as a treatment option, so they are more familiar and confident with using it. Staff felt that more training and guidance was required for them to feel more confident referring clients to cCBT. They suggested written instructions, training sessions, guidance on booking cCBT appointments, information about the evidence base and access to the platform for clinicians so that they can experience it themselves.
‘Have us all work through it ourselves. Given time to do this like SP/SR [self-practice/self-reflection] exercise. What we like and don’t like’ [time point 1]
‘Handout guide or reminder sheet of the process of pathway and what [cCBT package name] offers as a prompt at assessment’ [time point 1]
‘Access to specific training on using e-CBT to come from a more informed position’ [time point 2]
‘Provide training to become more familiar with what cCBT involves’ [time point 2]
‘All therapists should have a demo account so they know what [cCBT package name] looks like to the patient; trainee PWPs should get [cCBT package name] training’ [time point 3]
At time points 1 and 2, staff reported that feedback on cCBT within the local service would be helpful from clients that have used cCBT, as this would to support them to refer more clients to it. For example, information on recovery rates, engagement, and patient experience. This was not identified by staff at time point 3.
‘Seeing comments from patients who have been offered and accessed cCBT’ [time point 2]
‘Could regularly mention it in team meetings and to hear some positive feedback from clients who have participated in cCBT would be useful’ [time point 2]
‘Hearing feedback from people who have used it’ [time point 2]
At all three time points, staff felt that increased awareness of the cCBT platform was important. They felt that staff needed to be more aware of cCBT, but that referrers also needed more awareness of this being an option. They felt that referrers should be given more awareness of cCBT as a form of treatment.
‘To have more training on what the process involves so I am able to explain this more to patients’ [time point 1]
‘Raising awareness among other referrers such as health visitors and CMHART’ [time point 2]
‘Making the referral process as easy as possible’ [time point 3]
Impact of COVID-19 pandemic
At time point 3, staff were asked an additional question about whether their practice had changed following the pandemic (see Fig. 8). The majority of staff reported that the number of clients that they offered (50%) or allocated to cCBT (52%) following the pandemic had not changed. However, a third of staff reported that they had increased the number of clients they had offered (36%) and allocated (33%) to cCBT.
Please write any comments on how you feel the COVID-19 pandemic has impacted on supporting clients through cCBT at [service name]?
This question was only asked at time points 2 and 3 when there was largely agreement that the COVID-19 pandemic led to a general acceptability towards cCBT in terms of increasing options and alternative to option to face-to-face CBT, when this was not available.
‘Clients seem to see it as a more acceptable option’ [time point 2]
‘I think it has made remote working more normal and expected’ [time point 2]
‘Greater willingness for therapists and clients to undertake cCBT’ [time point 2]
‘It was a god send’ [time point 3]
‘Opened patients up to the prospect that remote cCBT is viable option’ [time point 3]
Future training
If you were to receive training on cCBT, what do you feel you would like it to involve to make you feel more confident in supporting cCBT?
This question was only asked at time points 2 and 3. At time point 2, staff felt that they would benefit from training being focused on raising their awareness on what the evidence base for cCBT is as a treatment option compared with other forms of treatment. Some staff at time point 3 did not feel further training was required.
‘Group session on how to use [specific cCBT package] and what it offers’ [time point 2]
‘No further training required’ [time point 3]
A consistent theme around wanting access to further information and support around cCBT was identified at time points 2 and 3. Specifically, time point 2 identified more knowledge of the evidence base and being given feedback and time point 3 highlighted the importance of supervision when supporting clients with cCBT.
‘A review of the evidence base’ [time point 2]
‘Evidence based/ data’ [time point 2]
‘Supervision support and how best to supervise’ [time point 3]
At time point 2, staff reported that having training on using the programme directly themselves, with the chance to try it out would be important. Staff at time point 3 in particular identified that they wanted practical experience of working through cCBT such as through role plays and examples.
‘Experiencing it first hand’ [time point 2]
‘Staff have access to programme to practice’ [time point 2]
‘Role play as a patient’ [time point 3]
‘Going through what each programme involves’ [time point 2]
Discussion
This service evaluation aimed to add to the current literature exploring staff perspectives of online CBT interventions. This evaluation collected data before and during the COVID-19 pandemic. Following the sudden and unexpected national quarantine, the service (like all others) was transformed with respect to service delivery methods, and therefore we were provided with a timely naturalistic opportunity to evaluate if this also changed staff beliefs. A further data collection time point was added to see if further training impacted on staff beliefs. It was thus an exploratory study to explore staff perspectives across three time points.
Comparing across time points, there appear to be some changes in staff attitudes towards cCBT. Specifically, there was increased agreement with time that cCBT was a suitable option for supporting those with mild to moderate mental health difficulties and that it can allow patients earlier access to evidence-based treatments. There was also a notable shift with some practitioners finding cCBT a more acceptable option for people with fewer years of education or who are unemployed.
Overall, despite these few exceptions there appeared to be limited change in staff perspectives towards cCBT across the three time points. Specifically, this evaluation showed that across all three time points, there was consistency in staff identifying that case complexity influences the appropriateness of cCBT, suggesting that staff believe that it is more effective for those with mild to moderate than severe difficulties. Furthermore, staff were consistent in their perspectives on the advantages of cCBT, particularly in terms of accessibility and convenience of it for clients; they have increased choice, and it is more accessible and flexible than CBT. However, there were some advantages of cCBT that were not agreed with, consistent across time points. For example, practitioners disagreed with the statement that cCBT saves therapists’ time. Research has shown that it benefits the therapist in terms of reducing time and burden on them, which supports previous research (Koomson et al., Reference Koomson, Haque, Sivananthan, Jayaraajan, Hussain and Ahmad2020; Scott, Reference Scott2018). A proportion of practitioners disagreed that cCBT was a viable solution to a lack of qualified CBT therapists. It would be interesting to explore and make sense of this through further research.
Consistently across time points staff reported a view that a lack of therapeutic relationship was a barrier to supporting clients with cCBT. This is in line with the staff beliefs reported in the original study by Meisel et al. (Reference Meisel, Drury and Perera-Delcourt2018), who reported that only 10% of their participants believed that there is strong evidence to support the use of cCBT. A paper by Kiluk et al. (Reference Kiluk, Serafini, Frankforter, Nich and Carroll2014) highlighted that there was limited evidence to indicate that computerised forms of therapy negatively impacted on the therapeutic alliance. Furthermore, research conducted by Aafjes-van Doorn et al. (Reference Aafjes-van Doorn, Békés and Prout2020) during the COVID-19 pandemic showed that the majority of practitioners felt online therapy sessions supported a good therapeutic alliance. This suggests that the staff beliefs reported within this paper were staff perceptions only and not consistent with the evidence base. This is also interesting given the positive feedback that has been received by patients accessing cCBT and extensive evidence base for the efficacy and effectiveness of cCBT (Barak et al., Reference Barak, Hen, Boniel-Nissim and Shapira2008; Sharry et al., Reference Sharry, Davidson, McLoughlin and Doherty2013). Misinterpretation of the evidence base is something that was also prevalent in the study of Meisel et al. (Reference Meisel, Drury and Perera-Delcourt2018). However, Aafjes-van Doorn et al. (Reference Aafjes-van Doorn, Békés and Prout2020) observed that therapists with more online therapy experience, who had more confidence in their own ability to deliver therapy online, and reported a good therapeutic alliance were reportedly more accepting of online therapy delivery. It is unclear how the specific demographics of the area covered by the service have influenced the participant beliefs. Whilst some of the service practitioners may choose to live in rural or isolated areas, the nature of their demographics and employment means that in general they are less likely to rely on a very limited public transport system and are less likely to experience digital exclusion or social deprivation. Despite the rurality of the area, there is not the high usage of digital solutions in general or within the mental health system that might be expected. Currently it is not clear what role digital exclusion plays in the low access to cCBT, but this should be considered alongside ways to overcome this (Greer et al., Reference Greer, Robotham, Simblett, Curtis, Griffiths and Wykes2019).
In terms of the impact of the COVID-19 pandemic on supporting clients through cCBT (this question could only be asked at time points 2 and 3), there was largely agreement that it led to a general acceptability towards cCBT in terms of increasing options and it being an alternative option to face-to-face CBT, when this was not available. However, findings also suggested that staff believed patient preference for face-to-face therapy would make them less likely to offer cCBT as a treatment option. Thus, findings appear mixed, and we cannot directly support the hypothesis of Wind et al. (Reference Wind, Rijkeboer, Andersson and Riper2020) that staff will appear more accepting of cCBT as a form of treatment with increasing levels of remote working.
One difference across time, namely something that was discussed at time point 3 only, was staff reporting that the availability of resources and electricity would be a barrier to cCBT. This is interesting and could be linked to the timing of time point 3 in which staff completed the survey as we were entering a heightened period of financial unrest in the United Kingdom with significant rising costs of living.
A barrier identified by staff, at time points 1 and 2, was that cCBT had a poor evidence base, thus suggesting it was not an effective form of treatment. This was not highlighted at time point 3, suggesting that perhaps the staff training contributed to staff learning more about the evidence base for cCBT. At time points 1 and 2, staff reported that feedback on cCBT within the service would be helpful from clients that have used cCBT, as this would support them to refer more clients to it, for example, information on recovery rates, engagement, and patient experience. This was not identified by staff at time point 3, which again might suggest that the staff training was effective in providing this feedback.
However, despite the prior staff training, at time points 2 and 3, staff identified that having more knowledge of cCBT, as well as direct practical experience, would make them feel more confident in supporting clients with cCBT; more training was identified as being needed. This is an interesting finding as staff are provided with a log in when they start in the service, which provides them with the opportunity to practise working through cCBT in their own time. Perhaps there is a need for training to incorporate dedicated time on working through modules. However, we have no evidence that we are aware of that having this dedicated time is effective and would impact staff perspectives on cCBT. Given the stuckness of some of the staff beliefs (which are not in line with the research evidence) perhaps more innovative training methods may be required. This is speculative, but perhaps some form of training exercises required to directly target the beliefs in a more direct way might be more effective. This could be working through a module and then engaging in a reflective task which directs the individual to hold their experience of the programme in mind alongside their competing beliefs. For example, someone could work through a single module and then be asked to directly consider what they have experienced and how this fits with their belief that only highly educated could benefit from the programme, or they could be asked what specifically they have noticed that supports or contracts their beliefs that only employed people could benefit from the programme. Clearly these types of strategies would need evaluating but making them more experiential, Socratic and targeted may potentially have a greater effect.
It is curious that the results of this service evaluation were so similar across three time points, despite the enforced increase in remote working with the COVID-19 pandemic; particularly in terms of how practitioners perceive what patients think about cCBT. It thus might be worth conducting a similar study with the public, patients waiting for therapy or previous patients to examine whether staff are accurately identifying barriers for patients or if they are creating ones with their own assumptions and narratives.
A limitation of this study was the differences in samples sizes across the time points. Specifically, time point 3 had a smaller sample size than time points 1 and 2, which could perhaps suggest caution in the comparisons made across the three time points. For time point 1, staff were asked to complete the questionnaire face-to-face in a team meeting whereas for the remaining time points, questionnaires were completed online. This could have impacted on uptake due to availability and time given to staff to complete the questionnaires. A second limitation is that the findings are based on staff within one service. Ideally data would have been collected across multiple services in order to provide stronger and more generalisable results. A final limitation is that the study focuses on one specific cCBT package, but there is no evidence that these beliefs would not be generalisable to other similar guided self-help packages. In conclusion, this service evaluation was a timely naturalistic way to examine staff beliefs about one specific online cCBT package over the course of time. Generally, staff perspectives were influenced by pre-existing beliefs about cCBT. There appeared to be some benefits of staff training, although staff still felt that more training was needed to help them feel more comfortable supporting clients through cCBT.
Table 1. Ranked suggestions for improving confidence
Key practice points
-
(1) Staff offering treatment options such as cCBT should be offered appropriate training in content and ideally given time to work through example modules in order to familiarise themselves with the content and be in a more informed position to offer such treatments.
-
(2) Having said this, the availability of relevant information may not always lead to accurate beliefs about treatment options. Staff attitudes towards treatment options are not always in line with the evidence base and may not always be amenable to change when new information is provided either by training or by experience (e.g. during the pandemic). This would suggest a need for practitioners to reflect on their own beliefs and explicit discussions with services around beliefs in order to ensure that patient have access to evidence-based interventions in a timely manner.
-
(3) Further research is required to understand how best to help clinicians to make treatment allocation decisions based on the evidence base rather than being unduly influenced by their own personal beliefs especially when relating to cCBT.
Data availability statement
Data are available upon request from the authors.
Acknowledgements
An important acknowledgement of thanks to Briony Guy for her time and support across the time span of this project. We would also like to thank Andrea Robinson and Laura Sirey. The study was completed in part fulfilment of the Doctorate in Clinical Psychology (Newcastle University).
Author Contributions
Kathryn Ragan: Conceptualization (equal), Data curation (equal), Formal analysis (equal), Investigation (equal), Methodology (equal), Project administration (equal), Resources (equal), Supervision (lead), Validation (equal), Visualization (equal), Writing - original draft (equal), Writing - review & editing (equal); Claire Kelly: Conceptualization (equal), Data curation (equal), Formal analysis (equal), Investigation (equal), Methodology (equal), Project administration (equal), Resources (equal), Validation (equal), Visualization (equal), Writing - original draft (equal), Writing - review & editing (equal); Richard Thwaites: Conceptualization (equal), Formal analysis (equal), Methodology (equal), Supervision (supporting), Validation (equal), Writing - original draft (supporting), Writing - review & editing (equal).
Financial support
This project received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
None of the three authors is employed by SilverCloud or have received any payment from them. No-one from SilverCloud was involved in the project other than to provide a 2-hour training session based on findings from time points 1 and 2. The submitted version of the article was not read or influenced by any SilverCloud employee. Richard Thwaites is the Editor of the Cognitive Behaviour Therapist. He was not involved in the review or editorial process for this paper, on which he is listed as an author. The other authors have no declarations.
Ethical standards
Data collection for each time point was separately registered with the Research and Development department of the NHS Mental Health Trust involved. As a service evaluation, further ethical approval was deemed to not be required. The researchers have all abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the BABCP and BPS.
Open access
Comments
No Comments have been published for this article.