People with severe mental illness (SMI) often have their lives shortened significantly by physical conditions such as diabetes and cardiovascular diseases, Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly1 but less attention has been paid to the stark inequalities in oral health that they experience. According to a recent systematic review with meta-analysis, people with SMI were four times more likely to have periodontal (gum) disease compared with the general population. Reference Cai, Ng, Zhao, Siskind and Kisely2 This excessive burden of periodontal disease borne by people with SMI is partially caused by the shared social determinants of health. Reference Joury, Kisely, Watt, Ahmed, Morris and Fortune3 For example, social factors such as low socioeconomic position affect behavioural factors like tobacco and alcohol use, which are common risk factors for both periodontal disease and SMI. Additionally, there are complex bidirectional interactions among periodontal disease and SMI, and chronic inflammatory physical conditions such as type 2 diabetes, that tend to cluster in people with SMI. For example, periodontitis, especially in its advanced stages, can be a risk and prognostic factor for SMI and type 2 diabetes. It can induce or contribute to chronic systemic inflammation, a risk factor for both SMI and type 2 diabetes. Reference Hajishengallis4 This means that maintaining good periodontal health is important for both the mental and physical health of people with SMI. Although oral hygiene self-care is key for the prevention and management of periodontal disease, a systematic review showed that only 39% of individuals with SMI cleaned their teeth twice a day compared with 75% reported in the general population. Reference Turner, Berry, Aggarwal, Quinlivan, Villanueva and Palmier-Claus5 Flossing and mouthwash use were also lower in this group of people. SMI is associated with negative and positive symptoms, as well as reduced motivation for self-care behaviours including oral hygiene. Furthermore, the side-effects of the first generation of antipsychotics include tremulous movements, which can impede oral hygiene. Evidence syntheses have highlighted the absence of effective interventions to improve oral hygiene among people with SMI. Reference Khokhar, Khokhar, Clifton and Tosh6,Reference Macnamara, Mishu, Faisal, Islam and Peckham7 This is due to the lack of consideration of co-creation and theory in the development, implementation and evaluation of such interventions. Reference Joury, Kisely, Watt, Ahmed, Morris and Fortune3 While different theories exist to guide the development of behaviour change interventions, use of the Behaviour Change Wheel (BCW) seems to be on the rise. The BCW offers a systematic approach that draws on stakeholders’ input (through co-creation), as well as on using ‘integrative’ behaviour change theoretical framework and process to develop ‘multi-level’ complex behaviour change interventions. Reference Michie, van Stralen and West8
Aim
To address the above evidence gap, we conducted the current study aimed at co-creating a theory-driven oral hygiene intervention for people with SMI using the BCW approach.
Method
Reporting is in line with the Consolidated Criteria for Reporting Qualitative Research (Appendix 1). Reference Tong, Sainsbury and Craig9
Study design and theoretical framework
We followed a process split into two stages over eight steps, drawing on the framework of Hawkins et al (2017) for the co-production and prototyping of public health interventions and the above-mentioned BCW theoretical framework (Fig. 1). Reference Michie, van Stralen and West8,Reference Hawkins J.Madden, Fletcher, Midgley, Grant and Cox10 Stage 1 involved BCW steps 1–3 to define the problem in behavioural terms, and to select and specify the target behaviour. We completed this stage through conducting a critical review and stakeholder one-on-one and focus group engagements with people with SMI, carers, primary care, mental health and dental professionals, clinical and public leads, policymakers and commissioners. Reference Joury, Kisely, Watt, Ahmed, Morris and Fortune3 Stage 2 involved BCW steps 4–8 to design and refine the intervention. In step 4, we identified what needs to change to bring about change in the target behaviour (identifying barriers and facilitators). We adopted a phenomenological approach to explore the subjective experience of toothbrushing, or supporting toothbrushing, in people with SMI and identify barriers and facilitators to improve and sustain this behaviour. We mapped identified barriers and facilitators to the Capability, Opportunity, Motivation–Behaviour (COM-B) model and the Theoretical Domain Framework (TDF). In steps 5 and 6, we selected relevant intervention functions and policy categories that can facilitate behaviour change. In step 7, we identified and mapped behaviour change techniques (BCTs) accordingly using BCT Taxonomy v1. Reference Michie, Richardson, Johnston, Abraham, Francis and Hardeman11 This helped identify potential strategies that could be used to facilitate change in the target behaviour. In step 8, we specified the intervention mode of delivery.
Fig. 1 Stages and steps involved in the co-creation of the Restart Smiling toothbrushing intervention for people with severe mental illness (SMI).
Ethical standards
We 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 as revised in 2013. All procedures involving human participants were approved by the NHS Research Ethics Committee (no. 21/WA/0058). Written informed consent was obtained from all participants in the qualitative study.
Participants and recruitment
The lead author (E.J.) recruited participants through sharing of flyers with relevant national and regional experts-by-experience groups, clinical networks and communities of practice and snowballing. The study was also promoted at the Right to Smile event and ‘Mind Your Smile’ national live X Chat (#MindYourSmile). Participants were recruited purposively. Eligible people with SMI were those aged 18–65 years and diagnosed with either bipolar disorder, schizophrenia/schizoaffective disorder or other psychotic disorders. Other participants were carers, primary care, mental health and dental professionals and clinical leads who provide care for people with SMI. All participants were required to speak English and not to be participants in other studies. Interested participants completed an online eligibility and sociodemographic questionnaire. Eligible participants were contacted by the lead author to schedule the semi-structured interview. All participants were given £20 for their participation in the present study.
Data collection
To complete steps 4–8, the lead author conducted 20 one-on-one semi-structured interviews using Microsoft Teams (Microsoft Office, version 16.63 for Mac iOS) between July 2022 and June 2023. Two topic guides (one for people with SMI and the other for carers, professionals and clinical leads) were informed by stage 1 and structured following the COM-B and TDF domains (Appendices 2 and 3). Four experts (a person with SMI, a carer and general practitioner, a psychologist and mental health nurse, and a dentist) were invited to comment via email on the questions, the structure of the interview guides, missing discussion topics and any other comments. Interviews lasted between 39 and 99 min and were audio-recorded and transcribed verbatim by a trusted transcribing service. Field notes were taken by the lead author to help contextualise and interpret findings. Data collection stopped when saturation was reached. To refine the content of the intervention and mode of delivery, E.J. collected feedback from stakeholders. This included patient and public involvement activities and two stakeholder engagement workshops. Identified toothbrushing barriers and facilitators, intervention functions, policy categories and BCTs were presented, and feedback was obtained using the Acceptability, Practicability, Effectiveness, Affordability, Side-effects and Equity (APEASE) criteria. 12
Research team
Our multidisciplinary team have expertise across public health, psychology, special care dentistry, general practice, psychiatry and SMI carer experience. E.J. is a female clinical academic and specialist in dental public health with a PhD and an interest in co-creating complex interventions to tackle mental, physical and oral multimorbidity in people with SMI. E.S. is a female psychologist with a PhD and an interest in the development and evaluation of complex interventions for chronic illness. E.H. is a female specialist in special care dentistry with a PhD and an interest in facilitating access to dental care in people with mental ill health. F.F. is a female consultant in dental public health with an interest in tackling oral health inequalities experienced by vulnerable groups. M.B.A.A. is a male early career researcher with an interest in developing public health interventions. C.A.C.-G. is a female general practitioner with a PhD and an interest in the primary care management of people with mental, physical and oral multimorbidity. S.B. is a female carer for a person living with SMI. K.B. is a male psychiatrist with a PhD and an interest in mental, physical and oral multimorbidity in people with SMI.
Data analyses
Sociodemographic data were summarised using descriptive statistics. We analysed interview data using framework analysis as described by Ritchie et al. Reference Ritchie, Lewis, McNaughton-Nicholls and Ormston13 We thereafter coded data using the constructs of the COM-B model and TDF, and refined the content of the intervention and mode of delivery based on stakeholders’ feedback. First, an inductive and iterative thematic approach was used to understand participants’ responses and perspectives. Steps started with data familiarisation and development of an initial coding framework (E.J.), which was discussed and refined by the wider research team. The framework was then applied to the whole data-set. E.J. undertook complete coding of transcripts using Microsoft Excel spreadsheet (Microsoft Office, version 16.89 for Mac iOS), and developed and triangulated themes across different participant groups. To ensure reliability, transcriptions were independently coded by E.S., E.H., F.F. and M.B.A.A. Consensus on themes and interpretation was reached through wider team discussions. Two participants provided input on the themes and interpretation of findings for additional validation. Second, E.J. coded data using the constructs of the COM-B model and TDF. Codes were reviewed by E.S., and any discrepancies in coding were discussed by E.J. and E.S. until consensus was reached. Third, by meeting as a co-production team we refined the content of the intervention, including training for professionals and content received by people with SMI based on stakeholder feedback. We used the APEASE criteria to guide decisions about the intervention content and mode of delivery. 12 Throughout data analyses, we kept an audit trail and engaged in ongoing reflexivity.
Results
Twenty participants took part in the semi-structured interviews. Their sociodemographic characteristics are summarised in Table 1.
Table 1 Participants’ sociodemographic characteristics
SMI, severe mental illness.
a. Two participants also had major depressive disorder.
b. Professional background included occupational therapist, clinical psychologist, general practitioner, peer support worker, dentist, specialist in special care dentistry, special care dental nurse and oral health educator.
c. Clinical leadership background included clinical lead for adult mental health in an integrated care board, and head and assistant head of dental services.
d. Two participants (one with SMI and a carer) were peer support workers and provided their perspectives from both roles.
Target behaviour
In stage 1 (steps 1–3), the oral hygiene behaviour to be changed was identified and specified jointly by stakeholders. Oral hygiene practices include toothbrushing and the use of other aids for cleaning teeth, such as dental floss and mouthwash. 14 The target behaviour that was selected and specified by stakeholders was brushing the teeth and gum line at least twice daily (last thing at night or before bedtime and on at least one other occasion) for 2 min with a manual or electric toothbrush and appropriate fluoride toothpaste.
What needs to change to bring about change in toothbrushing
In stage 2 (step 4), the following 6 themes were identified for what needs to change to improve and sustain brushing among people with SMI. The links between these themes and COM-B and TDF theoretical constructs are mapped in Table 2.
Table 2 Matrix of the links between what needs to happen for brushing to occur, COM-B and TDF domains, interventions functions, policy categories and examples of identified behaviour change techniques (BCTs)
SMI, severe mental illness; COM-B, Capability,Opportunity, Motivation–Behaviour model; TDF, Theoretical Domain Framework.
Exploring the health, social and emotional consequences of improving oral health and brushing
Participants stressed the importance of understanding and discussing the health, social and emotional consequences of having good oral health and maintaining good oral hygiene (Table 2). This would motivate people with SMI to prioritise their oral health and improve their brushing regimen. They highlighted that information and discussion should focus on the importance of oral health to the physical and mental health of people with SMI, and the strong links between their mouth, mind and body.
Valued self-identity
In addition to simply understanding the benefits of good oral health, several participants explained that the condition of their oral health represented a deeper self-identity (Table 2). It was suggested that discussion could explore the link between good oral hygiene, healthy teeth and mouth, fresh breath and social interaction, emotions and self-identity.
Forming a brushing habit
Participants emphasised the need to support people with SMI to form a brushing habit that is incorporated into their daily morning and evening routines (Table 2). People with SMI stated that they often find the evening brushing routine more challenging, when they usually feel tired and might not remember to brush their teeth. This is partly related to the side-effects of their psychotropic medications. Participants suggested problem-solving with the person with SMI and their prescriber – for example, the possibility of amending the evening medication time to 30–60 min before bedtime. This would help people with SMI carry out their evening self-care routine (including brushing) and avoid evening/night binge eating, which is a major side-effect of psychotropic medications:
‘If it [medication] doesn’t have to be taken at eight o’clock and you have a little bit of food or you have a sandwich or something, and you take your tablet at 10 … brush your teeth … then go to bed yeah. That could be an idea’. Female participant with SMI (P315)
Participants also mentioned the use of a reminder and/or associating the new brushing habit with an existing habit to help with remembering (Table 2). They noted that different people prefer different types of reminders (e.g. visual reminders or setting phone alarms).
Receiving instructions and demonstration on brushing and considering cognitive capacity
Participants identified the need to receive evidence-based instructions and demonstration on brushing frequency, duration, technique and coverage (Table 2). Instructions should be simple and concise and supported by accessible resources, particularly demonstration videos:
‘She doesn’t really read a lot, so maybe a video format would be the best.’ Female carer (P309)
The above should be supported with a discussion to aid the selection of the type of toothbrush (manual or electric) considering the level of cognitive concentration required, manual dexterity and sensory issues, personal preference, reach (size of head), pressure sensitiveness and other characteristics and practicalities (such as the need to remember to recharge the electric toothbrush) (Table 2). Most participants with SMI preferred the electric toothbrush because of the low level of cognitive concentration required, and considered it a ‘trusted companion’ with which to clean their teeth.
Participants noted the prescribing of high-concentration fluoride toothpaste for people with SMI, which requires system-level liaison with prescribers. They also mentioned the need to explore the use of flavourless and/or foamless fluoride toothpaste should the person with SMI have sensory issues:
‘I know they recommend like higher fluoride toothpaste, don’t they? For people on prescription, but I’ve only been offered that once … if you do have severe mental illness, you’re more likely to be neglecting your teeth, so that seems to meet … the criteria that you’re at risk’. Female participant with SMI (P282)
Exploring the need for financial and practical social support
Participants mentioned that financial and social support, particularly practical social support, might be needed (Table 2). The latter might include reminding people with SMI to brush their teeth, buying them brushing equipment, changing their manual toothbrush or the head of their electric toothbrush. Such social support could be requested from a family member, carer and/or other people they identify from their social network.
Further strategies for ‘bad days’
Participants also identified the following strategies to help with sustaining brushing on ‘bad days’ (when their mood is very low and they lack energy or motivation):
-
(a) Identify reasons for maintaining brushing during bad days, such as knowing how quickly oral health can deteriorate if they don’t brush.
-
(b) Include brushing in the maintenance plan they create for bad days, which includes the simple tasks they could do for themselves on a bad day to help them feel better.
-
(c) Ask a family member or carer to include brushing in their support on bad days.
‘If you could get an understanding of how quickly your oral health can decline if you don’t brush them … maybe for me that might help that extra tiny bit of motivation of me thinking this is a really bad day, but these are my teeth’. Female participant with SMI (P308)
‘I mean for sure it helps when, for example, my kid is here and I see that it’s important for him to keep hygiene.’ Female participant with SMI (P314)
Intervention functions, policy categories and BCTs
In steps 5–7, 6 out of 9 intervention functions and 4 out of 7 policy categories, described in the BCW guide, were identified as relevant based on the findings of the semi-structured interviews, and are mapped in Table 2. With respect to BCTs, 36 out of 93 BCTs in BCT Taxonomy v1 were identified from interview data (Table 3). Ten BCTs were selected based on stakeholders’ feedback as key behavioural strategies to improve and sustain brushing among people with SMI (Table 3).
Table 3 Identified behaviour change techniques (BCTs) according to BCT Taxonomy v1 a
a. BCTs shaded were selected based on stakeholders’ feedback as key behavioural strategies to improve and sustain brushing among people with severe mental illness.
Intervention mode of delivery
In step 8, the following four key recommendations on intervention mode of delivery were specified.
Integrating the brushing intervention into SMI physical health checks
Participants suggested that a brief oral hygiene assessment and brushing support are best provided as part of SMI annual physical health checks and follow-up care:
‘I think it would sit really well as part of the annual physical health check if the nurse did talk to you about it … went over brushing technique, and those sorts of things’. Male participant with SMI (P304)
‘We want people to smoke less, we want people to be healthy, and the mouth is part of health, so why not assess their oral hygiene at that physical health check? … I mean if they’re already providing these other interventions there’s no reason why oral hygiene can’t be included there as well’. Female dental professional (P331)
Training and remunerating primary care and mental health professionals
Most participants indicated that physical health-checking professionals in primary care and mental health services should be trained and remunerated to provide a brief brushing assessment and support for people with SMI. Participants also highlighted the advantages of using these professionals over dental professionals to deliver brushing support. This is because the former have regular contacts and established rapport with people with SMI and are best placed to support brushing as part of an integrated care approach to rebuild broader lifestyle routines, after addressing pressing mental and physical health needs:
‘You’re seeing them regularly, so they might help you write a plan … and then they check in on you a week later, or whatever, so you’ve got that kind of contact, you don’t have like contact with dentists … it’s for those people that see you regularly to have those conversations’. Female participant with SMI (P282)
Using a personalised and trauma-informed approach
Participants explained that brushing support should be person-centred and trauma-informed. This starts with asking people with SMI whether they would like to improve their brushing. Next, it points out what support is available for improving their brushing, then it focuses on finding what works for them to improve their brushing and supports decision-making in a non-judgemental approach, emphasising that they are not alone:
‘Every patient is their own, we can’t treat that patient group all the same way, because it doesn’t work the same way for everyone else. So we have to be respectful of their thoughts and needs as well.’ Female dental professional (P333)
‘I think when these conversations do happen it has to be like in a very sensitive way with like a cultural understanding.’ Female carer (P309)
The role of peer support workers was highlighted in relation to delivering such trauma-informed brushing support:
‘You don’t need a dentist to say those things to you … Sometimes it’s a small little bit of information … [from] peer support or expert by experience … then everything feels less judgemental. It needs to come from something more collaborative’. Female participant with SMI (P314)
Maintaining engagement
Maintaining engagement through offering follow-up support appointments (in person or via the telephone) was considered highly important by participants:
‘They’ll have a series of appointments with the nurse to work on oral hygiene. And, they’ll build up rapport with the patient, show them how to brush their teeth, give them tips and hints as how to do it better, and that’s over a series of appointments.’ Female dental professional (P331)
Further recommendations
Some participants outlined that brushing support should be culturally and linguistically tailored to cater for the needs of people with SMI from diverse ethnic backgrounds, and for those whose English is not their first language. Furthermore, there should be special considerations when supporting brushing in in-patient settings, such as providing supervised brushing. The intervention could be delivered through up to four support sessions face-to-face (for the initial session) and/or remote (for the follow-ups):
‘I think the points with my mum that would be the most helpful is probably having other Bengali women holding that discussion. So, she’ll be more inclined to take it in and also there wouldn’t be that language barrier.’ Female carer (P309)
‘For the ones that are in-patients, hopefully they are being supervised with the toothbrushing, so they’re not going to do random things with it.’ Female dental professional (P318)
The components of the brushing intervention (named Restart Smiling), including materials and training resources, are summarised in Fig. 2 and the logic model is presented in Supplementary Table 1.
Fig. 2 The Restart Smiling toothbrushing intervention components. SMI, severe mental illness.
Discussion
This is the first study to co-create an oral hygiene intervention for people with SMI using the BCW process. Our study identified new behaviour change strategies that could address the limitations of previous studies in terms of forming and sustaining a new toothbrushing habit. Reference Adams, Wells, Clifton, Jones, Simpson and Tosh18–Reference Yoshii, Kitamura, Akazawa and Saito32 For example, our findings highlight the unique challenges of integrating brushing into the evening routine. The study also identified a facilitator to rebuild the evening routine in a way that avoids or minimises the side-effects of psychotropic medications on brushing and on weight management at night. Furthermore, our findings identified strategies that could be used to support sustaining brushing on bad days in the longer term. Our study also highlights that brushing support should ideally be provided as part of integrated physical care by non-dental professionals who, compared with their dental counterparts, have regular contacts and established rapport with people with SMI. They also have better capability to support the formation of a toothbrushing habit as part of their assistance in building back the broader daily lifestyle routines. Such support may include social prescribing, which involves referral to non-medical services, to help people with SMI identify their social needs (such as housing and employment), apply for support (including financial support that could help with the cost of brushing equipment) and develop ‘well-being’ action plans to promote, establish or re-establish integration and support in their communities. Reference Carnes, Sohanpal, Frostick, Hull, Mathur and Netuveli33
The use of BCW in the development of oral health interventions is quite new. Two recent studies aimed to develop oral hygiene interventions for people with learning disabilities and for Punjabi immigrant adults. Reference Kaur, Kandelman and Potvin15,Reference Mac Giolla Phadraig, Asimakopoulou, Faulks, Van Harten, Ledger and Spencer16 However, in the study by Mac Giolla Phadraig et al, the BCW process was neither fully utilised nor well reported. Reference Mac Giolla Phadraig, Asimakopoulou, Faulks, Van Harten, Ledger and Spencer16 The study of Kaur et al applied the BCW process retrospectively on qualitative data collected from five Punjabi immigrant participants, indicating major limitations with methodology rigour and participatory principles of the BCW process. Reference Kaur, Kandelman and Potvin15 Furthermore, the study of Mishu et al identified barriers and facilitators to oral hygiene in people with SMI and mapped them against the COM-B model and TDF domains. Reference Mishu, Faisal, Macnamara, Sabbah, Peckham and Newbronner17 Nonetheless, the authors did not link these to relevant intervention functions or policy categories, nor did they identify appropriate BCTs or design the oral hygiene intervention.
Strengths and limitations
One of the main strengths of the present study is the use of the BCW, which is highly participatory and heavily grounded in behaviour change theory. Compared with the above-mentioned studies that aimed to develop oral hygiene interventions for people with SMI, the BCW approach allowed us to co-create systematically a brushing intervention with multi-level components targeting personal capabilities and motivational aspects, as well as wider healthcare system and socioeconomic determinants. The transferability of the current findings is supported by the thick description of methods, participants’ characteristics and contexts, as well as by maintaining an audit trail and a reflexive journal and undertaking purposive sampling. Participants’ experiences covered a wide range of settings, including primary care, in- and out-patient secondary mental health services, social care and the voluntary, community and social enterprise sector. They had diverse oral hygiene self-care practices.
Despite the above strengths, our study is not without limitations. One of the limitations of using the BCW is that it is resource-intensive due to involving a wide range of stakeholders and intervention complexity. Another limitation is related to the need to tailor the present toothbrushing intervention to people with SMI in secure psychiatric settings. A further potential limitation for transferability is that the findings may be less applicable to people with SMI and healthcare systems outside of the UK, where routine physical health checks and follow-up care are not commonly offered.
Implications and next steps
In terms of policy and practice implications, our study informed NHS England national guidance (2024) on improving the physical health for people living with SMI, which included for the first time oral health within the comprehensive SMI physical health checks and cited Restart Smiling within training resources. 34 SMI physical health checks are provided in primary care and secondary mental health services in England. 34 We summarised, in Fig. 2 and Supplementary Table 1, the inputs needed to embed toothbrushing support in the SMI physical health checks practice, namely, (a) adding questions to assess brushing into the electronic SMI physical health checks template and (b) training and remunerating primary care and mental health professionals who deliver SMI physical health checks for supporting brushing. These inputs are in line with the Right to Smile consensus statement, and with recommendations for better physical health checks and integrated care for people with SMI. Reference Joury, Beveridge, Littlejohns, Burns, Copsey and Philips35–Reference Mishu, Aggarwal, Shiers, Peckham, Johnston and Joury37 The next steps include prototyping and evaluating the Restart Smiling intervention.
In conclusion, we have co-created a complex theory-driven oral hygiene intervention for people with SMI integrated into physical health checks underpinned by systems thinking and using the BCW in real-world settings.
Supplementary material
The supplementary material is available online at https://doi.org/10.1192/bjo.2025.10905
Data availability
The anonymised data that support the findings of this study are available from the corresponding author, E.J., upon reasonable request.
Acknowledgements
The authors thank all participants and stakeholders who took part and supported Restart Smiling development in stages 1 and 2. Special thanks to Professor Elizabeth Barley for her support on the interview topic guides. Some of the symbols used in the manuscript’s figures were downloaded from www.flaticon.com.
Author contributions
Conceptualisation of the study, study design and data acquisition: E.J.; data analysis and interpretation: E.J., E.S., E.H., F.F., M.B.A.A., S.B. and K.B.; writing of the manuscript: E.J.; critical revision of the manuscript: E.J., E.S., E.H., F.F., M.B.A.A., S.B., C.A.C.-G. and K.B. All authors gave their final approval and agree to be accountable for all aspects of the work.
Funding
This study was partly funded by the QMUL Clinical Lecturer Small Grant Scheme and National Institute for Health and Care Research (NIHR) Academic Clinical Lecturer Award. C.A.C.-G. is part-funded by NIHR Applied Research Collaboration West Midlands (no. NIHR200165). The views expressed are those of the authors and not necessarily those of the funders.
Declaration of interest
The authors declare no conflicts of interest with respect to the research, authorship and/or publication of this article. The views expressed are those of the author(s) and not necessarily those of NIHR or the Department of Health and Social Care. K.B. is a member of the British Journal of Psychiatry Open editorial board and did not take part in the review or decision-making process of this paper.
Transparency declaration
The authors confirm that this manuscript is an honest, accurate and transparent account of the study reported, that no important aspects of the study have been omitted and that any discrepancies from the study protocol have been explained.
Open access
eLetters
No eLetters have been published for this article.