Lee 

et al. BMC Psychiatry (2023) 23:130 BMC Psychiatry

https://doi.org/10.1186/s12888-023-04517-1

RESEARCH Open Access

Development and initial evaluation

of a behavioural intervention to support
weight management for people with serious
mental illness: an uncontrolled feasibility
and acceptability study
Charlotte Lee1,2*   , Felicity  Waite3,4   , Carmen Piernas1    and Paul Aveyard1,2    

Abstract 
Background  The rates of obesity and associated health problems are higher in people with serious mental illness
(SMI), such as schizophrenia and bipolar disorder, than the general population. A primary care referral to a behavioural
weight management programme can be an effective intervention, but people with SMI have reported barriers to
engaging with them and bespoke options are rarely provided in routine practice. It is possible that adjunct support
addressing these specific barriers could help. Here we report the development, feasibility and acceptability of an
intervention to improve uptake and engagement with a mainstream weight management programme for people
with SMI.
Methods  We worked with people with a lived-experience of SMI and used the person-based approach to develop

stream weight management programme (WW®) to be attended once a week, in-person or online, for 12-weeks. The
the ‘Weight cHange for people with sErious mEntal iLlness’ (WHEEL) intervention. It comprised a referral to a main-

adjunct support comprised a one-off, online consultation called Meet Your Mentor and weekly, telephone or email
Mentor Check Ins for 12-weeks. We assessed the feasibility of WHEEL through the number of programme and adjunct
support sessions that the participants attended. We analysed the acceptability of WHEEL using a thematic analysis of
qualitative interviews conducted at baseline and at 12-week follow-up. Our exploratory outcome of clinical effective-
ness was self-reported weight at baseline and at end-of-programme.
Results  Twenty participants were assessed for eligibility and 17 enrolled. All 17 participants attended Meet Your Men-
tor and one was lost to follow-up (94% retention). Nine out of 16 attended ≥50% of the weekly programme sessions
and 12/16 attended ≥50% of the weekly check-ins. Participants reported in the interviews that the adjunct support
helped to establish and maintain a therapeutic alliance. While some participants valued the in-person sessions, others
reported that they preferred the online sessions because it removed a fear of social situations, which was a barrier for
some participants. The mean change in self-reported weight was − 4·1 kg (SD: 3·2) at 12-weeks.

*Correspondence:
Charlotte Lee
[email protected]
Full list of author information is available at the end of the article

© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Lee et al. BMC Psychiatry (2023) 23:130 Page 2 of 13

Conclusions  A mainstream weight management programme augmented with brief and targeted education and
low-intensity check-ins generated sufficient engagement and acceptability to warrant a future trial.
Keywords  Mental illness, Schizophrenia, Bipolar, Weight, Intervention, Programme

Background comparable to clinical agoraphobia are observed in peo-

Serious mental illness (SMI) refers to chronic illnesses ple with SMI [20, 21]. This anxiety is likely to arise from
like schizophrenia spectrum and bipolar disorder that are and exacerbate low confidence, with high rates of social
characterised by hearing, seeing and believing things not isolation and unhelpful eating behaviours that follow
based on reality. People with SMI have an increased risk [22]. Interventions that target these barriers could there-
for poor physical health [1–3], with a 1·8 to three times fore support people with SMI to benefit from current
higher risk of obesity (body mass index [BMI] > 30 kg/m2) options for improved health.
than people without a diagnosis [4]. This contributes to A bespoke programme is one option to addressing
a 3·8 higher risk for mortality from avoidable cardiovas- these barriers for people with SMI. There is evidence that
cular disease (CVD) before the age of 50 than the general they can be modestly effective, with a meta-analysis of
population [5]. Meta-analyses of randomised controlled 41 RCTs across 4267 participants reporting 2·2 kg more
trials (RCTs) have shown antipsychotics, which are the weight loss than no or minimal intervention in follow-
mainstay treatment for most SMIs, cause weight gain; ups ranging from 8 to 52 weeks [23]. However, bespoke
especially second-generation drugs like quetiapine, olan- programmes are rarely part of routine care for people
zapine and clozapine [6, 7]. Other meta-analyses have with SMI as  a cost-effectiveness analysis has  indicated
shown people with SMI to have higher dietary energy these intensive programmes are not viable [24]. An alter-
intake and lower levels of physical activity than the gen- native – arguably more sustainable – approach is to sup-
eral population – behavioural patterns linked with obe- port people with SMI to engage with and benefit from
sity and CVD [8, 9]. mainstream programmes.
Behavioural weight management programmes that In a systematic review of 34 RCTs testing 36 behav-
support people to reduce their energy intake and increase ioural weight management programmes in people with
their physical activity can be an effective intervention for SMI, we found those that offered regular contact (e.g.,
weekly telephone calls), tools to support enactment
to the 12-week behavioural programme WW® (formally
obesity. In one RCT of 1267 participants, those assigned
(e.g., handouts) and tailored materials (e.g., shorter or
WeightWatchers®) lost 4·8 kg at one-year follow-up [10] repeated sessions) promoted engagement and were asso-
and 2 kg at five-year follow-up [11]. Furthermore, evi- ciated with clinical effectiveness [17]. Is it possible that
dence supports that these programmes are cost-effective providing these characteristics through adjunct sup-
in the short and long-term [10]. Accordingly, guidelines port, alongside mainstream weight management, might
in the United Kingdom (UK) recommend a primary care enhance engagement with these programmes for people
referral for any patient with obesity to this type of main- with SMI. However, the feasibility and acceptability of
stream programme, which are widely and freely available this approach is unknown. Here we report the develop-
– at the point of prescription – in some countries’ health ment and initial evaluation of the ‘Weight cHange for
systems such as England [12]. people with sErious mEntal iLlness’ (WHEEL) inter-
Yet, even when referred, people with SMI may not vention, which aimed to offer people with SMI adjunct
engage with mainstream weight management pro- support to access and engage with a mainstream weight
grammes [13]. One trial of 1384 participants with SMI management programme.
found the proportion of uptake was fewer than 2% com-
pared with >50% in 12 studies included in a review of 30
studies in the general population [14, 15]. Meta-anal-
Methods
Intervention design
yses of both survey data and qualitative accounts have
We worked with people with a lived-experience of SMI
reported physical (e.g., low energy), psychological (e.g.,
and used the person-based approach (PBA) to develop
low confidence) and socio-ecological (e.g., lack of social
WHEEL. This helped us to create an intervention rel-
support) barriers that can preclude access for people with
evant for people with SMI and thereby improve the likeli-
SMI [16, 17]. While people from the general population
hood of it being effective [25]. The PBA has four stages:
may also experience these barriers [18, 19], people with
(1) plan, (2) design, (3) evaluate, and (4) implement and
SMI have reported additional issues that can prevent
this paper covers the first three (see Fig. 1). Our patient
engagement. High levels of co-morbid anxious avoidance
and public involvement (PPI) and stages 1 and 2 (August
 Lee et al. BMC Psychiatry (2023) 23:130 Page 3 of 13

Fig. 1  The three stages to develop and evaluate WHEEL

2020 to June 2021) are described in outline here and in concentration difficulties. The first part covered a gen-
full elsewhere [17]. Here we focus on stage 3 (June 2021 eral discussion about their weight (e.g., the history of
to July 2022). We followed the Template for Interven- their weight gain, contributors to their weight gain and
tion Description and Replication (TiDieR) criteria [26] to attempts and challenges to change). The second part cov-
describe the intervention and the Consolidated Criteria ered their knowledge or experience of mainstream weight
for Reporting Qualitative Studies (COREQ) criteria [27] management programmes, with suggested improvements
to evaluate it (see Additional file 1: Supplementary mate- that we could enact in this study. All consultations were
rial 1 and 2). facilitated, audio-recorded and transcribed by the first
author. Overall, the contributors recognised the need to
Patient and public involvement (months 0–2) manage their weight and were positive about the oppor-
We elicited views on the priority and direction of this tunity for support. There was some trepidation about
work by consulting 12 people with a lived-experience of mainstream weight management and we were cognisant
SMI. In total, we conducted five telephone interviews of these concerns during our development. The contrib-
and two focus groups – one of four contributors, one utors worked with us throughout the next stages to co-
of three contributors. Each consultation lasted 2 hours design the intervention (e.g., by reviewing themes during
divided in two parts, with regular breaks to account for stage 1 and intervention handouts in stage 2).
Lee et al. BMC Psychiatry (2023) 23:130 Page 4 of 13

Stage 1: plan (synthesise qualitative data; months 0–13) be adjunctively provided alongside WW® (see Additional
We systematically reviewed 20 qualitative studies to file 1: Supplementary material 5).
identify the barriers to engaging with behavioural weight
management programmes and the programme charac- The prototype intervention
teristics that may help overcome them, as reported by
intervention. It comprised a referral to WW® for
In the third phase, we constructed the prototype WHEEL
people with SMI [17]. The nine barriers and correspond-
ing characteristics are summarised in Additional file  1: 12-weeks to be attended in-person. The adjunct support
Supplementary material 3. was a single, 1·5 hour, online, one-to-one consultation
called Meet Your Mentor. It aimed to address partici-
Stage 1: plan (synthesise quantitative data; months 0–13) pants’ specific concerns and prime them with education
We systematically reviewed RCTs of bespoke weight on the modifiable contributors to weight gain for people
management programmes for people with SMI to iden- with SMI. More specifically, Meet Your Mentor covered:
tify which characteristics were associated with weight
loss [17]. We used a crisp-set qualitative compara- 1. Your story and reasons for joining the intervention;
tive analysis (CsQCA) to establish causal relationships 2. The causes of weight gain in people with SMI – rec-
through systematic comparisons [28, 29]. Across 34 ognising the weight gain effects of antipsychotics
RCTs testing 36 programmes in 4305 people with SMI, while emphasising self-regulation;
those that offered interim booster support, supporting 3. The challenges to losing weight in people with SMI –
tools and tailored materials were more likely to result in addressing barriers like negative self-beliefs, reason-
greater weight loss than in the control group, compared ing bias and social avoidance;
with programmes that did not. The interventions resulted 4. Activities to promote personalisation and boost
in more weight loss (mean = − 4.37 to + 1 kg at 6 weeks to engagement.
18 months follow-up) compared with controls (− 1.64 to
+ 3.08 kg). The mentor was a knowledgeable facilitator (i.e., a
graduate-level psychologist; first author) who used a
Stage 2: design (construct a theoretical framework; months 17-page booklet, grounded in the principles of psychoe-
10–11) ducation and cognitive behavioural therapy for psychosis
We constructed a theoretical framework that was guided (CBTp), that was reviewed by our PPI contributors. We
by the National Institute for Health and Care Excellence offered regular breaks every 30 minutes. In addition, par-
(NICE) recommended processes [30]. At this stage, we ticipants were given supporting tools (i.e., their person-
drew upon the results from stage 1 [17], relevant lit- alised booklet), which they were encouraged to review in
erature [31] and included key stakeholder feedback (i.e., their own time.
from people with SMI and healthcare professionals). After Meet Your Mentor, the mentor offered light
touch practical support, activated the participants
Define the guiding principles 12-week membership for the mainstream programme
In the first phase we defined our guiding principles. This and found the date, time and location of their nearest in-
comprised the target behaviour(s); the hypothesised person session. We felt that it was important to set clear
mechanism of change; the target barrier(s) that we had expectations on their attendance at the first session of
identified from stage 1 and considered necessary to the mainstream programme to mitigate avoidance; how-
address; the proposed characteristic(s) that we thought ever, we permitted moderate flexibility if participants
may help overcome each barrier; and the function of that were unable to attend that week. Thereafter the men-
characteristic (see Additional file 1: Supplementary mate- tor provided 15-minute interim booster support over
rial 4). the telephone for 12-weeks to emphasise successes and
achievements. We scheduled these Mentor Check Ins at
Code the proposed characteristics a fixed day and time to increase the participants’ sense of
In the second phase we coded the proposed characteris- accountability and support.

stream programme WW®. It is a commercially-provided,

tics against ones that were already present in the main-
Stage 3: evaluate (initial feasibility and acceptability;
open-group that primary care clinicians can offer freely to months 12–24)
patients in England with sessions usually attended once a Study design

WW® did not include some of the proposed character-

week in-person for 12-weeks. This stage highlighted that We assessed the initial feasibility and acceptability of
WHEEL in a single-arm, uncontrolled intervention study.
istics. We therefore assessed which characteristics could The Medical Research Council (MRC) framework for
 Lee et al. BMC Psychiatry (2023) 23:130 Page 5 of 13

developing complex interventions recommends identi- people with SMI [6, 7]. None had extensive weight loss
fying key uncertainties to be answered during feasibility during or at end-of-programme. Recruitment continued
testing [32]. Based on the literature [16, 17], we identified until we obtained sufficient data to address the research
two key uncertainties to be answered in this study: aims, which in this case were the indicative measures
of feasibility and acceptability. Based on the sample size
1. Feasibility defined as engagement with the main- sometimes used in proof-of-concept and qualitative stud-
stream programme and adjunct support, which we ies, we estimated 12 participants could be sufficient to
assessed using descriptive statistics (number [n, %] of meet this aim [33, 34].
Meet Your Mentor, programme sessions, and Mentor
Check Ins attended);
2. Acceptability defined as positive responses for the Procedure
mainstream programme and adjunct support, which Once participants finished their Meet Your Mentor ses-
we assessed using two qualitative interviews (one at sion and agreed on their Mentor Check Ins dates, we sent
the baseline and one at 12-week follow-up). them a personalised WHEEL booklet and voucher to
access the mainstream programme. Attended their Meet
We also assessed the number (n, %) of participants Your Mentor and first programme sessions. This inter-
retained at 12-weeks and self-reported weight at baseline view aimed to understand the acceptability of Meet Your
and end-of-programme as exploratory outcomes. Mentor; specifically, how it addressed their concerns and
influenced their attendance at the first of 12 programme
sessions. We invited participants to the second interview
Study sampling and recruitment at 12-week follow-up. This interview aimed to understand
We advertised the study through PPI networks and used the acceptability of the Mentor Check Ins; specifically,
snowball sampling to recruit potentially eligible partici- how it supported their engagement with the remaining
pants. We also advertised it through the McPin Founda- 11 sessions (see Additional file 1: Supplementary material
tion – a charity that exists to put the lived experience of 6). The participants were remunerated £25 per hour per
people affected by mental health problems at the heart of interview as UK guidelines recommend [35]. In line with
research (www.​mcpin.​org). We provided interested par- the PBA, we analysed the qualitative interviews through-
ticipants with the information sheet prior to a telephone out programme delivery to modify WHEEL, which we
call to assess eligibility, obtain informed consent and col- recorded using the Table of Changes (TOC) method [25].
lect demographic information. We assumed that no further responses after an enacted
change was indicative of acceptance.
Eligibility criteria 
Data collection
Demographic data were self-reported (including age,
• Aged ≥18 years; diagnosis, height and weight) at enrolment over the tel-
• Given a primary diagnosis of SMI (e.g., schizophre- ephone. The mentor conducted all interviews over the tel-
nia, schizophreniform disorder, schizoaffective disor- ephone and she was trained in qualitative interviews. To
der, bipolar disorder, or depression with psychosis); mitigate respondent bias, the mentor prefaced questions,
• Wanting to lose weight; spotted inconsistencies to probe them and reassured par-
• Willing and able to join an in-person mainstream ticipants that there were no right or wrong responses [36].
programme and discuss their experience of it in an To mitigate interviewer bias, the mentor kept a reflexivity
audio-recorded interview; log to be cognisant of biases and debriefed with the study
• Willing and able to give informed consent for partici- team on a fortnightly basis. The study team developed
pation in the study. semi-structured topic guides for the two interviews (see
Additional file  1: Supplementary material 7). Questions
We assessed the suitability of participants without over- were initially open-ended, however, we modified our topic
weight and obesity willing to join the study on a case-by- guides to include closed questions, which helped to scaf-
case basis. They were admitted to the study if they were fold conversations and break down difficult concepts. The
taking medications known to cause weight gain and were mentor probed responses that she perceived as useful to
concerned to limit that trajectory. All were keen to pre- the study aim. All interviews were audio-recorded, inde-
vent excess weight gain and learn weight management pendently transcribed verbatim and checked by the men-
strategies, especially given the weight gain trajectories in tor for accuracy against the recording.
Lee et al. BMC Psychiatry (2023) 23:130 Page 6 of 13

Fig. 2  Descriptive data

Data analysis April 2022. Once we met the recruitment target (n = 12),
We used trial literature to categorise feasibility data as: we purposefully recruited only males to ensure participants
≤2 (low engagement), 3–5 (sporadic engagement), 6–10 across the gender spectrum were equally represented in the
(good engagement) and 10–12 (strong engagement) data. Twenty participants were assessed for eligibility and 17
[37, 38]. For acceptability data, we used an inductive- enrolled (i.e., three did not meet the inclusion criteria). Most
deductive descriptive thematic analysis to code, catego- participants were female (n = 13 [76%]), white (n = 8, [47%])
rise, identify and describe patterns in our transcriptions and living with schizophrenia spectrum disorder (n = 8
of the interview data [39]. This analysis followed four [47%]). The average age of participants was 48 years (range
steps. First, we deductively developed a priori codes 29–70). Fourteen (82%) had baseline overweight and obesity.
based on our PPI consultation. Second, we inductively
added codes from each transcript. These codes were spe-
cific to the research aim. Third, codes were organised Feasibility
into sub-categories and then categories to produce top- We followed-up 16 out of 17 participants at 12-weeks (i.e.,
level themes. Fourth, the study team discussed themes one lost without reason) and included them the analy-
to reach a consensus agreement. We acted on Miles and sis (95%, high retention). All 16 participants joined Meet
Huberman’s recommendations to test and confirm our Your Mentor (100%, strong engagement). For the pro-
findings [40]. This meant we scrutinised the mentor’s gramme sessions, five participants joined ≤2 (31%, low
reflexivity log during fortnightly discussions. We also engagement); two participants joined 3–5 (13%, sporadic
invited participants’ feedback on the summary report of engagement); seven participants join 6–10 (44%, good
findings to ensure our interpretations remained close to engagement); and two joined 10–12 (13%, strong engage-
their own accounts. Our ontological position was rela- ment). For the Mentor Check Ins, four participants joined
tivism and our epistemology was rooted in subjectivism. ≤2 (25%, low engagement); no participants attended 3–5
All data were managed in NVivo 1 software [41]. We pre- check-ins (0%, sporadic engagement); four participants
sent selected quotations under pseudonyms. We used a attend 6–10 (25%, good engagement); and 8 attended
Cochrane publication to categorise the retention rate as: 10–12 check ins (50%, strong engagement) (see Fig.  2).
≤49% (low), 50–79% (medium), and 80% + (high) [42]. The one participant lost without reason joined Meet Your
Mentor and neither attended any subsequent programme
sessions or responded to the Mentor Check Ins.
Results
Demographic characteristics
Acceptability
We recruited participants between 12th August 2021 and
We approached data saturation on some categories
31st January 2022 and completed data collection on 30th
after 13 first interviews although we interviewed all 16
 Lee et al. BMC Psychiatry (2023) 23:130 Page 7 of 13

participants retained at 12-weeks, which confirmed The information in the booklet was novel for some par-

initiating a weight loss activity (i.e., either starting WW®

that we did not need to construct new categories or ticipants. They reported that they felt optimistic about
themes. First interviews lasted on average 53 minutes
(range: 32–75) and second interviews lasted 45 minutes or changing their food choices).
I felt encouraged to start the Weight Watchers®, but I
(range: 9–95). We constructed seven themes centred on
WHEEL’s acceptability, which we present according to
didn’t feel pressured. (Fionee, second interview).
the programme components. These themes focussed on
how helpful or not the adjunct support was in accessing
I think what I took away from that as the most use-
and engaging with the programme.
ful thing was about making the decision, having the
power to make the decision at that particular time,
Meet your Mentor
when I’m about to eat something. I think that’s stuck
Theme 1: establishing a therapeutic alliance  This theme with me. (Denise, first interview).
considers the acceptability of the person delivering the Others valued the opportunity to share their own and
adjunct support. hear others’ experience of weight management in the
context of SMI.
Participants welcomed the opportunity to disclose their
concerns prior to starting the programme, with some I found it useful going through the booklet… not nec-
expressing that the mentor established a psychologically essarily things useful to know, although that craving
safe space. part was the biggest thing that stuck with me. But
going through the booklet was good and just feeling
that you know there’s obviously a lot of other people
I think, personally… the empathy, the kind of active
in the same situation and you’re not alone with it
listening [the mentor] had going on, I think that was
can really help. (Fionee, second interview).
really good. That was something I think most people
want… for me [it] was quite beneficial because it Two participants, Denise and Marcella, noted that
kind of made me think about weight a little bit more. the common challenges to losing weight in people with
I think that people with mental health problems, SMI (e.g., negative self-beliefs, reasoning bias and safety-
we have a, we have a way of kind of ignoring things, seeking behaviours) were irrelevant to them. They both
especially things that make stress bad or make us suggested tailoring the booklet to their specific concerns,
not feel great. (Abdel, first interview). which we enacted after Marcella (e.g., by asking partici-
pants which challenges were relevant to them before dis-
Participants reported that their weight gain was due to
cussing it further).
the side effects of antipsychotics, which left some of them
feeling out of control and demotivated. Hearing the men- Theme 3: a mentor that helped pinpoint specific barriers
tor reflect on their experiences in a clear and respectful to joining  This theme describes the outcome of Meet
manner validated some participants’ experiences and Your Mentor and if it supported participants to initiate
authenticated the credibility of the information that the programme.
followed. Most participants expressed concern to be in an unfa-
So hearing… there are… a lot of people that do have miliar place or attend a social interaction. For some, it
this problem from this particular type of medica- was because auditory-verbal hallucinations (e.g., critical
tion, was, was very reassuring I suppose because or threatening voices) felt frightening, which negatively
you know when it is being brushed off all the time by affected their desire to engage with other people. For oth-
psychiatrists you do start to think, ‘Well is it just me ers, it was fear or judgement or paranoid concerns about
then?’ (Fionee, second interview). the potential harm from others.
… [The voices say, they say] you can’t do anything
I’m on medication that literally forces me to eat… it’s right (Matthew, second interview).
either that or I end up in a psychiatric unit, like, I’m
sorry but I don’t have a choice (Abdel, first interview). It could be quite [er] hard just to introduce your-
self… like a [er] completely new… environment…
Theme 2: value of the booklet  This theme describes the because you can feel people are staring at you, or,
content of the booklet and if participants perceived it as or you could feel people talking about you… So that
useful or not in addressing their barriers to initiating the is quite, yeah that’s quite, can be quite scary. (Mar-
weight management programme. cella, first interview)
Lee et al. BMC Psychiatry (2023) 23:130 Page 8 of 13

A few participants spoke of how low self-esteem [Online] is probably [er] better because you could
seemed to exacerbate their fears of being rejected, nega- [er] switch the, [um] the video off if [er] if you had
tively judged or attacked by the group. Some participants [um] if you didn’t want to, to talk or to, to engage.
also expressed that it was difficult to wake up and join a And [er] that’s probably quite… useful for [er] for
session because of the antipsychotic induced fatigue. people with [er] with psychosis [er] because [er] the
online features that let you just [er] hide yourself.
As I have a mood disorder, low self-esteem is already
(Marcella first Interview)
an issue for me (as I presume for others in the group).
I have no intention of being weighed in front of The online modality enabled participants to still engage
others or (a)nother. I know I am overweight and it with the open-group. This was because it removed
causes me great distress and anxiety. I do not want fear of social situations, which was a barrier for some
to add to this. It is embarrassing to be weighed. A participants.
judgement is being made. (Jane, email correspond-
[The online session] was really good… because we
ence shared with permission)
talked about [um] I… just [typed] in the chat I said,
Access to a mentor with whom they had established a ‘Look I’m just listening really and learning, here
therapeutic rapport allowed participants to share these today, so I won’t say very much’. (Tansi, first inter-
concerns. This, in turn, opened opportunities for collabo- view)
rative reflection (i.e., using if-then statements, reframing
reasoning style). So [online is] great and the things that I liked about
[the coach] was that he spoke to everyone individu-
[Me and the mentor] talked in a very gentle way
ally [using the one-to-one chat function]. (Abdel,
about the problem that I was having and it offered a
first interview)
kind of solution… that was such a long time ago and
my, my life has changed a lot since then, that I don’t However, other participants expressed difficulties with
have those feelings anymore because it’s like [the using technology to access the online sessions. The frus-
mentor] addressed them in the first meeting that we tration and rumination that followed sometimes led to
had. (Alice, second interview) unhelpful coping strategies (i.e., over-eating and binges)
… for me it was the stress of trying to get on [the
Mainstream weight management programme online session]… when you’ve got a mental health
problem and you’re trying to get somewhere and you
Theme 4: acceptability of the in‑person vs online pro- can’t get in and then you’re trying and trying, it just
gramme  This theme centres around the acceptability of makes stress worse, which then exacerbates the con-
the mainstream programme, specifically the modality in dition, which meant that I ended up opening a box
which it is delivered. of Lindt chocolate and having most if it last night
after, after the session because I just didn’t feel… well
We initially invited participants to join the in-person it was really, you know, I was just so angry… I mes-
sessions. However, most participants expressed a strong saged a friend and I said “I’m I don’t think I should
desire to attend the online sessions, which we enacted do Weight Watchers at all… I’m going to give up.
in our TOC. That was because some felt able to manage (Abdel, first interview)
worries more easily from a distance (e.g., reducing eye
contact, staying mute and keeping the video off ). They
Theme 5: joining the programme amplified their sense
reported that these strategies reduced their concerns
of vulnerability  This theme focuses on the ongoing
around being in unfamiliar social settings, though it pre-
concerns reported by our participants that negatively
vented them from learning they could manage without
affected the acceptability of the mainstream programme.
such safety-seeking strategies.
I’ve just gone into a room full of people that I didn’t Whether it was the in-person or online programme
get and thank god I could keep myself mute and session, some participants shared an ongoing sense of
nobody could see me. Where, where do I look [if I feeling either unsafe or that they did not fit in the group,
went in-person] when I say something that [others] which they attributed to their diagnosis of SMI. They
just don’t agree with? You know, how do I manage described this as amplifying their sense of vulnerability,
my emotions? (Tansi, first interview) which meant they avoided attending the sessions.
 Lee et al. BMC Psychiatry (2023) 23:130 Page 9 of 13

I wish I could think of another word other than I back on the achievements of the past week and have
didn’t feel emotionally safe… because… I can’t afford someone to share that with. If I was left to my own
to be putting myself in  situations where I may feel devices, I think it would have been a very different
a little bit vulnerable. (Christie, first interview after experience. (Fionee, second interview)
they declined to join any more in-person sessions).
One participant shared concerns that the mainstream Suggested improvements
programme was not culturally appropriate to their needs,
particularly the foods discussed in the session. This Theme 7: define the nature of the mentorship  The par-
added to a sense of feeling pushed out from the group. ticipants offered recommendations for our future devel-
opment, which this theme summarises.
People like myself, the people who still eat their own
foods, who don’t really venture into westernised type
All participants expressed that they valued the adjunct
meals and foods… [the programme] doesn’t really
support, though it was notable that most also recom-
cater towards people like us and how do we make
mended a peer supporter with a lived-experience of SMI
it more culturally competent and able so that peo-
to address their ongoing feeling of isolation.
ple like myself who unwittingly have to gain weight
because of the medication can then hopefully be able It’s because a peer support worker can relate to you
to manage our weight. (Abdel, second interview) more easily and they’re cheaper ‘cos the NHS pay
them less. (Alice, second interview).
Mentor check in The content of the mentor check-ins needed refine-
ment. One participant wanted a more structured check-
Theme 6: maintaining an interaction with the men- in to account for their  occasional disorganised thoughts
tor  This theme circles back to theme 1 and discusses the and speech.
continuity of care that the Mentor Check Ins provided.
No [ the check-in was] really good when I
could [answer]. I just wasn’t clear on what the
Most participants described the check-ins as a chance to
check-in, I mean I know it was about how I…
talk with the mentor about their weight loss, which they
maybe a set of three questions [like] ‘how are
thought was lacking in the mainstream programme.
you today’, ‘how has your week been’, ‘anything
I did really enjoy that side of it… because even you’re looking forward to’ or ‘what was a great
though I did go to the workshops, you don’t always moment’. The reason I say that is because I’ve
get the opportunity to speak when you’re there, and got fast acting bipolar so the thoughts are
although you do have a check-in with the coach it’s, there, they just go really quickly through my
it’s still, it’s very different to having that conversation mind. (Tansi second interview).
with [the mentor] each week… it was just more per-
The frequency and scheduling of the Mentor Check-
sonal. (Fionee, second interview)
Ins were also noted. Some participants felt that regular,
This sense of protected time motivated participants to scheduled, weekly check-ins were helpful and others
continue going to some of the sessions. wanted more flexibility.
I think mental health plays a big role … ‘cos some- No I think once a week was good, and I enjoyed it
times your mental health can make something very being a phone call. I think it was about the right
small turn into something big task… but I I think for amount of time because it gave time to think to
me one of the motivators was the fact that we had change. (Fionee, second interview)
catch up calls… there was something about them
that made me think ‘I’ve got to see this through, I There’s nothing wrong with the mentor, it’s just I
can’t back out’. If I had joined on my own will and thought if I can’t do a Thursday and I’m free on a
without the involvement of [the mentor], maybe Wednesday that week, it means at least I’m not
after session one I probably would have given up, but missing a session. (Hassan, second interview)
[the mentor] helped me stay motivated. (Hassan,
second interview)
Weight loss outcome
I found them really good, really motivating to just, The mean self-reported weight change was − 4·1 kg (SD:
and encouraging, you know, to feel like I could look 3·2) at 12-weeks from enrolment.
Lee et al. BMC Psychiatry (2023) 23:130 Page 10 of 13

Intervention changes bias, we kept a reflexivity log to be cognisant of biases

All proposed changes, and those subsequently enacted, and debriefed with all other researchers on a fort-
were recorded using the TOC method (see Additional nightly basis. We also followed published recommen-
file  1: Supplementary material 8). The key changes were dations [40] (i.e., returning summary reports to the
as follows. Meet Your Mentor: (1) tailor the booklet to participants to ensure our interpretations were cred-
each participant’s concern and strengthen the perceived ible; triangulating evidence through comparisons with
relevance of the information provided; Meet Your Men- other published literature, reporting our work accord-
tor: (2) extend from 15 to 30-minute telephone calls to ing to published  guidelines), which strengthened the
allow for participant story telling; mainstream pro- rigour of our findings. Participants also self-reported
gramme: (3) send email reminders to avoid participants’ session attendance and weight change and one cannot
forgetting their sessions. rule out that social desirability also meant participants
over-estimated their results, especially given reports of
Discussion over-reporting in people with schizophrenia [43]. How-
Main findings ever, the reported weight loss resembles the results
This paper described how we combined person-, evi- seen across meta-analyses suggesting the intervention
dence- and theory-based approaches to develop, eval- likely led to weight loss [23].
uate and iterate a mainstream weight management The participants were ethnically diverse from across
programme tailored for people with SMI. In our develop- the UK and represent the intended people that could
ment, we highlight the number of stages, level of detail benefit from this support. Although most were female
and the importance of continued refinement at the initial this reflects the demographics of people from the gen-
evaluation stage. In our evaluation, we have shown that eral population that take up mainstream programmes
a mainstream weight management programme bolstered [10]. One cannot rule out that the intense contact, for
with adjunct support generated sufficient feasibility and the purpose of arranging the interviews and not the
acceptability to run as an RCT. intervention itself, offered inadvertent support that
the participant would not receive if it were offered as
intended; although this is likely to be negligible. In
Strengths and limitations
terms of the intervention itself, WHEEL is designed to
We have clearly documented how we developed our
be opportunistically offered and delivered in primary
intervention using established methods [32] and best
care, however, there is no telling whether people with
practice guidance [26, 27]. This included finding and
SMI would take up such a programme given the lack of
synthesising data using systematic reviews (e.g., of quali-
empirical evidence.
tative studies and RCTs) and our transparent decision-
making on the augmented characteristics. We improved
the prototype programme based on the participants’
direct experiences (e.g., from the qualitative interviews) Comparison with the literature
and researcher observations, rather than hypotheti- A previous trial testing the effectiveness of a main-
cal scenarios, which streamlined our development and stream weight management programme in 11 people
allowed us to address challenges as they arose. However, with schizophrenia and 11 matched controls reported a
we included a small sample size for a feasibility evalua- 3·3 kg weight loss in men only and a 50% follow-up rate
tion. We also made changes throughout delivery, which at 12-weeks [44]. In another trial, 1384 participants with
meant the interviews did not discuss a consistent inter- SMI were offered a choice between weight management
vention. This makes the number of participants discuss- interventions with fewer than 2% choosing the main-
ing the same version of the intervention even smaller and stream programme compared with 60% who chose the
this is likely to reduce reliability of the findings. same programme with extra support from a health coach
The mentor collected and analysed the data, which [14]. The data from these trials suggest that uptake and
included qualitative interviews on the value of the men- retention in mainstream programmes is likely to be low
torship. Since the mentor had built rapport with the in people with SMI, perhaps due to unaddressed barri-
participants, we may have we elicited responses that ers to engagement, which we sought to investigate in this
might not have been otherwise mentioned. However, study.
this dual role is likely to have biased the data. Social We found that low-intensity support (i.e., a one-off ini-
desirability in qualitative methods is intractable and we tial consultation followed by weekly telephone or email
prefaced questions, spotted inconsistencies to probe exchanges from a mentor) is associated with 4·2 kg weight
them and reassured participants that responses were loss and a 94% follow-up rate, which is higher than previ-
not wrong to mitigate it [36]. To mitigate interviewer ously reported in trials. These results are indicative and
 Lee et al. BMC Psychiatry (2023) 23:130 Page 11 of 13

warrant further investigation but align with the  results Supplementary Information
observed from other, more intensive, bespoke pro- The online version contains supplementary material available at https://​doi.​
grammes. One RCT reported 54% of the 414 recruited org/​10.​1186/​s12888-​023-​04517-1.
participants had ‘good’ engagement with a bespoke pro-
Additional file 1: Supplementary material 1. TiDieR checklist. Sup-
gramme and an 84% retention rate [24], compared with plementary material 2. COREQ checklist. Supplementary material 3.
our 56% and 94%, respectively; though a subsequent Summary of results from stage 1 and 2. Supplementary material 4. The
cost-effectiveness analysis suggested that this type of sup- guiding principles of WHEEL. Supplementary material 5. Proposed char-
acteristics coded against a mainstream weight management programme.
port is unlikely to be cost-saving for health economies to Supplementary 6. A flow diagram of WHEEL. Supplementary material
provide [24]. The augmented elements of the interven- 7. Interview schedule. Supplementary material 8. Table of changes
tion we provided are likely to be a feasible alternative of (TOC).
a potentially scalable augmented weight management
programme. Acknowledgements
We thank all PPI contributors and participants for their commitment to our
work. We thank the McPin Foundation for their support in this study. Our grati-
Implications for future research tude to the anonymous referees for their helpful review of our article.
The positive responses in our qualitative interviews Authors’ contributions
largely centred on the weekly mentorship. In our devel- CL, FW, CP, and PA conceived in the design of the study. CL co-ordinated the
opment, we grounded this mentorship in the principles study and undertook it. CL performed all analyses, wrote the paper, and had
primary responsibility for its final content. All authors contributed to data
of CBTp (e.g., strong therapeutic alliance and collabora- interpretation, and read, edited, and approved the final manuscript. CL is the
tion exploration), which gave participants the opportu- study guarantor.
nity to gather information that may lead them to update
Funding
or change their beliefs. However, the modality (i.e., tel- This research received no specific grant from any funding agency, commercial,
ephone or email), frequency (i.e., once a week or flex- or not-for-profit sectors. CL is funded by the Engineering and Physical Sciences
ible) and content (i.e., open or structured discussion) Research Council (EPSRC) and the National Institute for Health Research (NIHR)
Oxford Biomedical Research Centre (BRC). FW is funded by a Wellcome Trust
needs further investigation. Our participants explicitly Clinical Doctoral Fellowship. CP is funded through the Oxford and Thames Val-
raised the potential value of working with peer support- ley NIHR Applied Research Centre (ARC). PA is funded by the NIHR Oxford BRC,
ers, with some evidence that they can lead to improved the NIHR Oxford and Thames Valley ARC, and is a NIHR Senior Investigator.
health outcomes in people with SMI [45]. Peer support- Availability of data and materials
ers are a part of a new workforce in some countries’ The data generated or analysed during this study are available from the cor-
healthcare systems and they are likely to be more cost- responding author on reasonable request.
effective than  other healthcare professionals like psy-
chologists. The adjunct support in our study provided Declarations
by or in conjunction with peer supporters is an exciting Ethics approval and consent to participate
area for future research. The University of Oxford Medical Sciences Interdivisional Research Ethics Com-
mittee granted ethical approval for this work (R74763/RE006). All methods
were carried out in accordance with relevant guidelines and regulations. All
Conclusions participants provided informed consent to participate in the study.
We describe the development and initial evaluation of
Consent for publication
adjunct support to improve uptake and engagement with Not applicable.
a mainstream weight management programme for peo-
ple with SMI. The results of this study potentially indicate Competing interests
PA is an investigator on a trial of a weight loss intervention for treating non-
that low-intensity mentorship in a randomised controlled alcoholic steatohepatitis, where the intervention is donated by Oviva and
trial might be feasible and acceptable for people with Nestle to the NHS. The remaining authors do not have any conflicts of interest.
SMI. The views expressed in this paper are those of the author(s) and not neces-
sarily those of the funder(s). No funder(s) had a role in the study design, data
collection, analysis, or interpretation.
Abbreviations
Author details
CBTp Cognitive behavioural therapy for psychosis 1
 Nuffield Department of Primary Care Health Sciences, Radcliffe Observatory
COREQ Consolidated criteria for reporting qualitative research
Quarter, University of Oxford, Woodstock Road, Oxford OX2 6GG, UK. 2 Oxford
MRC Medical research council
Biomedical Research Centre, Oxford University Hospitals NHS Foundation
NICE National Institute for Health and Care Excellence
Trust, Oxford, Oxfordshire, UK. 3 Department of Psychiatry, University of Oxford,
PPI Patient and public involvement
Warneford Hospital, Warneford Lane, Headington, Oxford, Oxfordshire OX3
PBA Person-based approach
7JX, UK. 4 Oxford Health NHS Foundation Trust, Warneford Lane, Headington,
RCTs Randomised controlled trials
Oxford, Oxfordshire OX3 7JX, UK.
SMI Serious mental illness
TIDieR Template for intervention description and replication
Received: 1 July 2022 Accepted: 2 January 2023
WHEEL Weight cHange for people with sErious mEntal iLlness
Lee et al. BMC Psychiatry (2023) 23:130 Page 12 of 13

References 18. Lemstra M, Bird Y, Nwankwo C, Rogers M, Moraros J. Weight loss inter-
1. Correll CU, Druss BG, Lombardo I, et al. Findings of a U.S. national cardio- vention adherence and factors promoting adherence: a meta-analysis.
metabolic screening program among 10,084 psychiatric outpatients. Patient Prefer Adherence. 2016;10:1547–59. https://​doi.​org/​10.​2147/​PPA.​
Psychiatr Serv. 2010;61(9):892–8. https://​doi.​org/​10.​1176/​ps.​2010.​61.9.​ S1036​49.
892. 19. Leung AWY, Chan RSM, Sea MMM, Woo J. An overview of factors associ-
2. De Hert M, Correll CU, Bobes J, et al. Physical illness in patients with ated with adherence to lifestyle modification programs for weight Man-
severe mental disorders. I. Prevalence, impact of medications and dispari- agement in Adults. Int J Environ Res Public Health. 2017;14(8). https://​doi.​
ties in health care. World Psychiatry. 2011;10(1):52–77. https://​doi.​org/​10.​ org/​10.​3390/​IJERP​H1408​0922.
1002/j.​2051-​5545.​2011.​tb000​14.x. 20. Freeman D. Personal view persecutory delusions: a cognitive perspec-
3. Vancampfort D, Stubbs B, Mitchell AJ, et al. Risk of metabolic syndrome tive on understanding and treatment. Lancet Psychiatry. 2016;3:685–92.
and its components in people with schizophrenia and related psychotic https://​doi.​org/​10.​1016/​S2215-​0366(16)​00066-3.
disorders, bipolar disorder and major depressive disorder: a systematic 21. Rosebrock L, Lambe S, Mulhall S, et al. Understanding agoraphobic avoid-
review and meta-analysis. World Psychiatry. 2015;14(3):339–47. https://​ ance: the development of the Oxford cognitions and Defences question-
doi.​org/​10.​1002/​WPS.​20252. naire (O-CDQ). Behav Cogn Psychother. 2022;50(3):257–68. https://​doi.​
4. National Mental Health Intelligence Network. Severe mental illness (SMI) org/​10.​1017/​S1352​46582​20000​30.
and physical health inequalities: briefing - GOV.UK. Public Health England. 22. Wilton J. More than a Number.; 2020. https://​www.​centr​eform​ental​health.​
https://​www.​gov.​uk/​gover​nment/​publi​catio​ns/​severe-​mental-​illne​ org.​uk/​sites/​defau​lt/​files/​2020-​03/​Centr​eforMH_​HWBA_​Weigh​tMana​
ss-​smi-​physi​cal-​health-​inequ​aliti​es/​severe-​mental-​illne​ss-​and-​physi​cal-​ gement.​pdf. Accessed 30 Nov 2020.
health-​inequ​aliti​es-​brief​i ng#​fn:1. Published 2018. Accessed 14 Nov 2019. 23. Speyer H, Jakobsen AS, Westergaard C, et al. Lifestyle interventions for
5. Hayes JF, Marston L, Walters K, King MB, Osborn DPJ. Mortality gap for weight Management in People with serious mental illness: a systematic
people with bipolar disorder and schizophrenia: UK-based cohort study review with meta-analysis, trial sequential analysis, and meta-regression
2000–2014. Br J Psychiatry. 2017;211(3):175–81. https://​doi.​org/​10.​1192/​ analysis exploring the mediators and moderators of treatment effects.
BJP.​BP.​117.​202606. Psychother Psychosom. 2019;88(6):350–62. https://​doi.​org/​10.​1159/​
6. Correll CU, Detraux J, De Lepeleire J, De Hert M. Effects of antipsychotics, 00050​2293.
antidepressants and mood stabilizers on risk for physical diseases in peo- 24. Holt RIG, Gossage-Worrall R, Hind D, et al. Structured lifestyle education
ple with schizophrenia, depression and bipolar disorder. World Psychiatry. for people with schizophrenia, schizoaffective disorder and first-episode
2015;14(2):119–36. https://​doi.​org/​10.​1002/​WPS.​20204. psychosis (STEPWISE): randomised controlled trial. Br J Psychiatry.
7. Wu H, Siafis S, Hamza T, et al. Antipsychotic-induced weight gain: dose- 2019;214(2):63–73. https://​doi.​org/​10.​1192/​bjp.​2018.​167.
response meta-analysis of randomized controlled trials. Schizophr Bull. 25. Yardley L, Ainsworth B, Arden-Close E, Muller I. The person-based
2022;48(3):643–54. https://​doi.​org/​10.​1093/​SCHBUL/​SBAC0​01. approach to enhancing the acceptability and feasibility of interven-
8. Teasdale SB, Ward PB, Samaras K, et al. Dietary intake of people with tions. Pilot Feasibility Stud. 2015;1(1):37. https://​doi.​org/​10.​1186/​
severe mental illness: systematic review and meta-analysis. Br J Psychiatry. s40814-​015-​0033-z.
2019;214(5):251–9. https://​doi.​org/​10.​1192/​bjp.​2019.​20. 26. Hoffmann T, Glasziou P, Boutron I, et al. Better reporting of interventions:
9. Stubbs B, Firth J, Berry A, et al. How much physical activity do people template for intervention description and replication (TIDieR) checklist
with schizophrenia engage in? A systematic review, comparative meta- and guide. BMJ. 2014;348:g1687.
analysis and meta-regression. Schizophr Res. 2016;176(2–3):431–40. 27. Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative
https://​doi.​org/​10.​1016/J.​SCHRES.​2016.​05.​017. research (COREQ): a 32-item checklist for interviews and focus groups. Int
10. Ahern AL, Wheeler GM, Aveyard P, et al. Extended and standard duration J Qual Health Care. 2007;19(6):349–57. https://​doi.​org/​10.​1093/​intqhc/​
weight-loss programme referrals for adults in primary care (WRAP): a mzm042.
randomised controlled trial. Lancet. 2017;389(10085):2214–25. https://​ 28. Ragin C. What is Qualitative Comparative Analysis (QCA)? 1984. http://
doi.​org/​10.​1016/​S0140-​6736(17)​30647-5. eprints.ncrm.ac.uk/250/1/What_is_QCA.pdf. Accessed 28 Nov 2020.
11. Ahern AL, Breeze P, Fusco F, et al. Effectiveness and cost-effectiveness of 29. Thomas J, Brunton G. Using Qualitative Comparative Analysis (QCA) in
referral to a commercial open group behavioural weight management systematic reviews of complex interventions: a worked example. Syst Rev.
programme in adults with overweight and obesity: 5-year follow- 2014;3(1):1–4. https://​doi.​org/​10.​1186/​2046-​4053-3-​67.
up of the WRAP randomised controlled trial. Lancet Public Health. 30. Recommendations | Behaviour change: individual approaches | Guidance
2022;7(10):e866–75. https://​doi.​org/​10.​1016/​S2468-​2667(22)​00226-2/​ | NICE. National Institute for Health and Care Excellence. https://​www.​
ATTAC​HMENT/​65197​5C9-​3160-​4901-​B61C-​A464B​2D31F​B7/​MMC1.​PDF. nice.​org.​uk/​guida​nce/​ph49/​chapt​er/1-​Recom​menda​tions#​recom​menda​
12. The National Institute for Health and Care Excellence (NICE). Obesity tion-5-​plan-​behav​iour-​change-​inter​venti​ons-​and-​progr​ammes-​taking-​
prevention clinical guideline [CG43]. United Kingdom. https://​www.​nice.​ local-​needs-​into-​accou​nt. Published 2014. Accessed 17 May 2022.
org.​uk/​guida​nce/​CG43. Published 2015. 31. Carey ME, Barnett J, Doherty Y, et al. Reducing weight gain in people
13. Lawrence D, Kisely S. Inequalities in healthcare provision for people with with schizophrenia, schizoaffective disorder, and first episode psychosis:
severe mental illness. J Psychopharmacol. 2010;24(4 Suppl):61–8. https://​ describing the process of developing the STructured lifestyle education
doi.​org/​10.​1177/​13597​86810​382058. for people with SchizophrEnia (STEPWISE) intervention. Pilot Feasibility
14. Pratt SI, Brunette MF, Wolfe R, et al. Incentivizing healthy lifestyle behav- Stud. 2018;4(1). https://​doi.​org/​10.​1186/​s40814-​018-​0378-1.
iors to reduce cardiovascular risk in people with serious mental illness: 32. Skivington K, Matthews L, Simpson SA, et al. A new framework for devel-
an equipoise randomized controlled trial of the wellness incentives oping and evaluating complex interventions: update of Medical Research
program. Contemp Clin Trials. 2019;81:1–10. https://​doi.​org/​10.​1016/j.​cct.​ Council guidance. BMJ. 2021:374. https://​doi.​org/​10.​1136/​BMJ.​N2061.
2019.​04.​005. 33. Bucci S, Barrowclough C, Ainsworth J, et al. Actissist: proof-of-concept
15. Uptake and retention in group based weight management services trial of a theory-driven digital intervention for psychosis. Schizophr Bull.
- GOV.UK. https://​www.​gov.​uk/​gover​nment/​publi​catio​ns/​uptake-​and-​ 2018;44(5):1070–80. https://​doi.​org/​10.​1093/​SCHBUL/​SBY032.
reten​tion-​in-​group-​based-​weight-​manag​ement-​servi​ces. Accessed 3 Oct 34. Braun V, Clarke V. To saturate or not to saturate? Questioning data satura-
2022. tion as a useful concept for thematic analysis and sample-size rationales.
16. Firth J, Rosenbaum S, Stubbs B, Gorczynski P, Yung AR, Vancampfort D. 2019. https://​doi.​org/​10.​1080/​21596​76X.​2019.​17048​46.
Motivating factors and barriers towards exercise in severe mental illness: 35. INVOLVE. Briefing notes for researchers - public involvement in NHS,
a systematic review and meta-analysis; 2016. https://​doi.​org/​10.​1017/​ health and social care research. National Insitute for Health Research
S0033​29171​60017​32. (NIHR). https://​www.​nihr.​ac.​uk/​docum​ents/​brief​i ng-​notes-​for-​resea​rch-
17. Lee C, Piernas C, Stewart C, et al. Identifying effective characteristics of ers-​public-​invol​vement-​in-​nhs-​health-​and-​social. INVOLVE. https://​www.​
behavioral weight management interventions for people with serious nihr.​ac.​uk/​docum​ents/​brief​i ng-​notes-​for-​resea​rchers-​public-​invol​vement-​
mental illness: a systematic review with a qualitative comparative analy- in-​nhs-​health-​and-​social-​care-​resea​rch/​27371. Published 2012. Accessed
sis. Obes Rev. 2022;23(1):e13355. https://​doi.​org/​10.​1111/​OBR.​13355. 16 May 2022.
 Lee et al. BMC Psychiatry (2023) 23:130 Page 13 of 13

36. Bergen N, Labonté R. “Everything is perfect, and we have no problems”:

detecting and limiting social desirability bias in qualitative research.
Qual Health Res. 2020;30(5):783–92. https://​doi.​org/​10.​1177/​10497​32319​
889354.
37. Astbury NM, Tudor K, Aveyard P, Jebb SA. Heterogeneity in the uptake,
attendance, and outcomes in a clinical trial of a total diet replacement
weight loss programme. BMC Med. 2020;18(1):1–9. https://​doi.​org/​10.​
1186/​S12916-​020-​01547-4/​FIGUR​ES/3.
38. Piernas C, Maclean F, Aveyard P, et al. Greater attendance at a community
weight loss Programme over the first 12 weeks predicts weight loss at 2
years. Obes Facts. 2020;13(4):349–60. https://​doi.​org/​10.​1159/​00050​9131.
39. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis.
Qual Health Res. 2005;15(9):1277–88. https://​doi.​org/​10.​1177/​10497​
32305​276687.
40. Miles M, Huberman M. Qualitative data analysis: an expanded source-
book. 2nd ed. London: Sage Publications, Inc.; 1994.
41. QSR. NVivo qualitative data analysis software; 2021.
42. Gillies K, Kearney A, Keenan C, et al. Strategies to improve retention in
randomised trials. Cochrane Database Syst Rev. 2021;2021(3). https://​doi.​
org/​10.​1002/​14651​858.​MR000​032.​PUB3.
43. Firth J, Stubbs B, Vancampfort D, et al. The validity and value of self-
reported physical activity and Accelerometry in people with schizo-
phrenia: a population-scale study of the UK biobank. Schizophr Bull.
2018;44(6):1293–300. https://​doi.​org/​10.​1093/​SCHBUL/​SBX149.
44. Ball MP, Coons VB, Buchanan RW. A program for treating olanzapine-
related weight gain. Psychiatr Serv. 2001;52(7):967–9. https://​doi.​org/​10.​
1176/​appi.​ps.​52.7.​967.
45. Chinman M, George P, Dougherty RH, et al. Peer support services for indi-
viduals with serious mental illnesses: assessing the evidence. Psychiatr
Serv. 2014;65(4):429–41. https://​doi.​org/​10.​1176/​APPI.​PS.​20130​0244.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission

• thorough peer review by experienced researchers in your field
• rapid publication on acceptance
• support for research data, including large and complex data types
• gold Open Access which fosters wider collaboration and increased citations
• maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions