Content uploaded by Linda Solbrig
Author content
All content in this area was uploaded by Linda Solbrig on Jul 22, 2019
Content may be subject to copyright.
International Journal of Obesity
https://doi.org/10.1038/s41366-018-0122-1
ARTICLE
Clinical research
Functional imagery training versus motivational interviewing for
weight loss: a randomised controlled trial of brief individual
interventions for overweight and obesity
Linda Solbrig 1,2 ●Ben Whalley1●David J. Kavanagh3,4 ●Jon May1●Tracey Parkin 5●Ray Jones6●Jackie Andrade1
Received: 20 December 2017 / Revised: 18 March 2018 / Accepted: 10 April 2018
© Macmillan Publishers Limited, part of Springer Nature 2018
Abstract
Objective Functional Imagery Training (FIT) is a new brief motivational intervention based on the Elaborated Intrusion
theory of desire. FIT trains the habitual use of personalised, affective, goal-directed mental imagery to plan behaviours,
anticipate obstacles, and mentally try out solutions from previous successes. It is delivered in the client-centred style of
Motivational Interviewing (MI). We tested the impact of FIT on weight loss, compared with time- and contact-matched MI.
Design We recruited 141 adults with BMI (kg/m²) ≥25, via a community newspaper, to a single-centre randomised con-
trolled trial. Participants were allocated to one of two active interventions: FIT or MI. Primary data collection and analyses
were conducted by researchers blind to interventions. All participants received two sessions of their allocated intervention;
the first face-to-face (1 h), the second by phone (maximum 45 min). Booster calls of up to 15 min were provided every
2 weeks for 3 months, then once-monthly until 6 months. Maximum contact time was 4 h of individual consultation.
Participants were assessed at Baseline, at the end of the intervention phase (6 months), and again 12 months post-baseline.
Main outcome measures Weight (kg) and waist circumference (WC, cm) reductions at 6 and 12 months.
Results FIT participants (N=59) lost 4.11 kg and 7.02 cm of WC, compared to .74 kg and 2.72 cm in the MI group (N=
55) at 6 months (weight mean difference (WMD) =3.37 kg, p< .001, 95% CI [−5.2, −2.1], waist-circumference mean
difference (WCMD) =4.3 cm, p< .001, 95% CI [−6.3,−2.6]). Between-group differences were maintained and increased at
month 12: FIT participants lost 6.44 kg (W) and 9.1 cm (WC) compared to the MI who lost .67 kg and 2.46 cm (WMD =
5.77 kg, p< .001, 95% CI [−7.5, −4.4], WCMD =6.64 cm, p< .001, 95% CI [−7.5, −4.4]).
Conclusion FIT is a theoretically informed motivational intervention which offers substantial benefits for weight loss and
maintenance of weight reduction, compared with MI alone, despite including no lifestyle education or advice.
Introduction
Obesity is a leading cause of premature morbidity and
mortality worldwide [1,2], but relatively small changes in
weight (5% reduction) produce substantial health benefits
[3–5]. In individuals with overweight and obesity, a sus-
tained reduction of only 2–5 kg reduces cardiovascular risk
factors and can prevent progression to type 2 diabetes
mellitus [6–8]. Reductions in waist circumference bring
their own health benefits, as excess abdominal fat increases
diabetes risk fivefold in men with waist > 102 cm and
threefold in women with waist > 88 cm [9].
*Jackie Andrade
jackie.andrade@plymouth.ac.uk
1School of Psychology, Cognition Institute, University of
Plymouth, Plymouth, UK
2NIHR CLAHRC South-West Peninsula, Plymouth, UK
3Institute for Health & Biomedical Innovation, Centre for
Children’s Health Research, Brisbane, Australia
4School of Psychology & Counselling, Queensland University of
Technology, Brisbane, Australia
5School of Health Professions (Dietetics), University of Plymouth,
Plymouth, UK
6School of Nursing & Midwifery, University of Plymouth,
Plymouth, UK
Electronic supplementary material The online version of this article
(https://doi.org/10.1038/s41366-018-0122-1) contains supplementary
material, which is available to authorized users.
1234567890();,:
1234567890();,:
Most weight loss programs focus on lifestyle education
and peer support. These interventions produce some
reduction in weight [10,11], but reductions are often
reversed within 1–2 years: typically 40–50% of the original
loss [12,13]. Few studies document weight loss past the
end of treatment [14]. Maintenance of weight loss has been
reported primarily after high-intensity lifestyle interventions
that provided comprehensive weight loss counselling for up
to 8 years [15], or in trials that delivered at least
6–18 months of extended care, with therapist contact, after
the successful completion of a weight loss program [16].
Current weight loss programmes place little emphasis on
maintaining motivation [17], despite its central role in long-
term weight control [18–20]. Participants struggle to stay
motivated when trying to maintain weight loss [21–23].
Motivational interventions may therefore be as effective as
skills based and educational approaches. Motivational
Interviewing (MI) [24] is the best-established motivational
intervention, but trials that include MI to support weight
loss or physical activity typically achieve only modest
effects [15,25,26]. In consequence, MI’s clinical sig-
nificance is limited [25,27,28]. There is a need for more
effective motivational support.
Two features of MI may produce sub-optimal impact.
Participants are not taught how to apply MI strategies by
themselves, reducing its effectiveness in real-life decision
situations [29,30]. The latest cognitive models of desire
[29,31–33] and emotion [34–36] highlight the critical role of
episodic multi-sensory mental imagery in motivation [37–41],
but MI remains a heavily verbal style of counselling.
Functional Imagery Training (FIT) [42] is a motivational
intervention based on Elaborated Intrusion theory, which
recognises the importance of vivid mental imagery in
desires [32]. FIT uses imagery to increase desire and self-
efficacy for change. It is delivered in the empathic,
accepting and respectful spirit of MI [24], trusting partici-
pants to be experts on themselves, and assisting them to
identify their own goals and related behaviours rather than
trying to convince them to adopt a preset regimen. The FIT
protocol prescribes similar steps to MI [24]: elicitation of
the person’s incentives for change, exploring discrepancies
between core values and current behaviour, boosting self-
efficacy, and, when people are committed to change,
developing specific action plans for implementing this
commitment. At each step, FIT also invites participants to
develop personalised multisensory imagery to maximize the
veridicality and emotional impact of each aspect. In
essence, FIT is imagery-based MI, with a strong additional
focus on training self-motivation through the use of ima-
gery. When participants have developed plans for change,
they transition into training to become their own FIT
therapist, supporting their autonomy and ability to flexibly
respond to self-management challenges in the natural
environment using imagery. They are shown how to
develop a cognitive habit of practising emotive goal-related
imagery in response to cues from a routine behaviour, and
encouraged to generate this imagery whenever motivation
needs to be strengthened. Because imagery is more emotive
than verbal thought [34–36], this vivid, affectively charged
functional imagery sustains desire for change until each new
behaviour becomes habitual. Functional imagery also
interferes with cravings when temptations occur, by com-
peting for working memory with craving imagery [43].
Detailed episodic imagery that is firmly grounded in
experience allows participants to anticipate problematic
situations, and plan and rehearse effective responses to
them, increasing self-efficacy through ‘symbolic practice’
[44]. Initial small-scale tests of FIT have shown benefits for
reducing snacking [42].
In this study, we recruited members of the public with
overweight and obesity to compare the effects of MI and
FIT on weight loss over 6 months of low-intensity treat-
ment, and an additional 6 months of follow-up. We pre-
dicted that FIT would produce greater initial weight loss,
because imagery amplifies the effective components of MI,
and better maintenance at 12 months because FIT teaches
the cognitive skills individuals need to stay motivated.
Method
Ethics and trial registration
Study approval was granted by the Faculty of Health and
Human Sciences research ethics committee, University of
Plymouth, on 23 March 2015. The International Standard
Randomised Controlled Trials registration was http://www.
isrctn.com/ISRCTN17292316, 18 July 2016.
Participants and recruitment
We advertised once for potential participants in the Ply-
mouth Shopper, a free local community newspaper, reach-
ing around 64,000 homes. The advertisement sought adult
participants with a BMI ≥25 kg/m², to test motivational
interventions for losing weight and becoming more active.
Exclusion criteria were current pregnancy, diagnosed eating
disorder, and failure to complete baseline assessments.
Sample size: A recommended sample size of 191 was
estimated using G*Power 3.1.9.2, assuming power of .8 to
detect a small effect size of .2 for between-group difference.
Design and overview
The trial was a single-centre, two-armed, single-blind,
randomised controlled parallel design with matched
L. Solbrig et al.
therapist contact time, comparing FIT with MI (1:1 ratio of
participants). The trial had 6 months of active intervention,
followed by a 6-month follow-up with no therapist contact.
Analyses are reported by intention-to-treat.
Randomisation and masking
Participants completed demographic details and assess-
ments before randomisation. Participants were randomized
to MI or FIT by the lead researcher using https://www.ra
ndomizer.org/ (random pairs option). In the two post-
treatment assessment sessions, research assistants (RAs)
who were blind to the intervention group, collected and
recorded primary outcomes. Participants were informed that
new RAs would take their measurements, and were asked
not to give away whether they were in the FIT or MI group.
The lead researcher was present in the room and could
verify and record if un-blinding occurred. Analysis of pri-
mary outcomes and quality of life was performed by an
analyst who was blind to intervention and not otherwise
involved in the trial.
Assessment measures
Primary outcomes were body weight, BMI and waistline.
Weight (kg) was measured in street clothes, with shoes
removed, using Omron BF511 Family Body Composition
Monitor. Height was measured to the nearest centimetre,
allowing calculation of BMI (kg/m²). For waistline mea-
surement, participants removed coats and sweatshirts, but
no other clothing. Waistline was measured to the nearest
centimetre, at the height of the umbilicus, using a tailoring
tape measure.
A secondary outcome was the global quality of life,
measured using the 1-item Global Quality of Life Scale
(GQOL; 43). Participants in both groups were asked if they
would recommend the treatment they had been allocated to
a family member, friend or colleague.
We collected the data on participants’experiences and
process variables, including frequency of motivational
cognitions, self-efficacy for diet and physical activity, self-
reported diet and physical activity. These results will be
reported separately.
Interventions
Both interventions were delivered individually by the lead
author. Face-to face sessions were all conducted in the same
counselling room on the University of Plymouth campus.
Session 1 immediately followed the collection of baseline
assessments and randomisation, and lasted 1 h. Session 2
(~35 min) was delivered by telephone a week later. Parti-
cipants then received fortnightly 5–15 min-long ‘booster’
phone calls until 3 months, followed by monthly calls to
6 months post-baseline.
We developed scripts to guide delivery of MI and FIT
and ensure treatment fidelity and consistency (available
from the authors), but the order of segments was flexible
and guided by the participant’s needs and responses, in
keeping with the spirit of MI. Active listening (open
questions, affirmation, simple and complex reflections,
summarising) was used in both treatments. The initial ses-
sion of MI and FIT had the same general structure, incor-
porating a negotiated agenda, discussion of the treatment
allocation, assessment feedback, existing or potential goals,
incentives for adoption of those goals, and past successes
with weight loss efforts. The therapist checked degree of
goal commitment. Once participants were committed to
behaviour change, they developed a plan for action over the
following few days, including strategies to address potential
barriers to its implementation.
Session 2 reviewed and developed the themes from
Session 1, in the light of experiences since the initial ses-
sion. Booster calls provided opportunities to review pro-
gress, reaffirm successful aspects of performance and
incentives for behaviour changes, and set additional sub-
goals. Exclusively in FIT, all sessions and booster calls
included mental imagery exercises.
If participants had requested advice on diet or physical
activity they would have been referred to the UK’s National
Health Service’s (NHS) publicly available and accessible
NHS Choices website which provides general information
and advice on health, for example, staying fit, losing weight,
or beating stress (https://www.nhs.uk/pages/home.aspx).
We did not have to refer anyone to this option however.
Qualitative participant experience data will be published in
a separate paper, delivering insight on how participants felt
about not being provided with pre-set regimen, diets, or
information and advice.
FIT
After discussion of assessment feedback in Session 1, the
therapist explained the rationale for using imagery and gave
an experience of affectively-charged imagery. After discuss-
ing incentives for potential behaviour change, participants
imagined these outcomes occurring, as specific future events
that were created as vividly as possible. Similar images were
elicited about past successes and about detailed plans for the
coming days, including successful achievement of each step
and success in reaching their ultimate goal.
Participants nominated a routine behaviour that could
prompt their imagery practice. They carried out this beha-
viour in the counselling room, while imagining their action
plan and goal. They were also encouraged to practise imagery
before engaging in their chosen behaviour, and offered the
Functional imagery training versus motivational interviewing for weight loss: a randomised controlled. . .
simple ‘Goal in Mind’app (https://itunes.apple.com/au/app/
goal-in-mind/id1289557359?mt=8;https://play.google.com/
store/apps/details?id=com.goalinmind&hl=en) to download.
They could use the app to upload motivational photos, tick
off when they had remembered to do imagery practise (the
app did not provide reminders), input a goal they would like
to achieve and access a five minute audio that guided them
through imagining how they would work on their goal
today and how good it would feel to achieve it. The audio
was emailed as an MP3 file if participants did not want to
use the app, but wanted the audio to practise.
Session 2 reviewed progress, including participants’
efforts at practising imagery. Imagery was used to help
solve any problems with progress towards their goal and to
motivate new sub-goals.
Booster calls developed imagery about recent successes,
problem solutions, or new goals behaviours. If required,
additional imagery exercises included: ‘Cravings Buster’
(deliberate switching of attention from craving imagery to goal
imagery) and ‘Plateau’(reflecting on benefits experienced so
far and exploring additional ways of working on goals).
MI
The therapist did not explicitly evoke imagery, and avoided
language that was likely to trigger it. Some additional ques-
tions were added to the manual, to ensure that the MI sessions
had similar time and intensity as the FIT sessions. A few
examples are: “When you think about that list of things, how
does it make you feel?”,“Would you mind summarising the
things that are likely to get better if you change your beha-
viour?”and “Is there anyone who could help you follow it
through?”MI participants were offered a goal sheet with the
action plan they developed with the therapist in the first
session; FIT did not have this. They took the goal sheet home
and were encouraged to review their statements, goals and
strategies, especially when they felt they needed extra moti-
vation. Two examples of this additional exercise from the
script are: “Would you like to write that down so you have a
summary to take away?”and”If you need a bit of a boost to
your motivation over the next few days, you could try reading
that over to remind yourself about what you said.”The sec-
tions were as follows: What I am going to do…Why I want to
do it…How I’ll do it…I know I can do it because…. Parti-
cipants were encouraged to add to the sheet as their goals or
reasons for change evolved. During the booster calls, they
were asked if they had added any new goals or ticked off
achievements on their sheet.
Intervention fidelity
The therapist was trained in FIT by two of the creators of
FIT and undertook a 3-day MI course. She attended weekly
clinical supervision meetings with the senior author, to
review individual sessions. A random 20% sample of initial
FIT and MI sessions was rated on the Motivational Inter-
viewing Treatment Integrity (MITI) 3.1.1 [46] by an RA not
involved in the project. FIT sessions were rated on a 15-
item checklist based on the manual. Additionally, two RAs
listened to the session recordings independently and cate-
gorised them according to the intervention they thought the
participants had received.
Procedure
Participants gave informed consent a week before their
initial session. After completing demographic details and all
assessments, they were randomly allocated to MI or FIT.
The treatment sessions followed, as described above. After
the first booster call, all participants were asked to complete
reassessments of the expected process variables (results
reported separately). At the end of treatment (6 m), they
attended a 15-min post-treatment assessment session in the
counselling room. Quality of Life was assessed via emailed
questionnaire 1 week before this session. RAs blind to
intervention measured waist and weight, and participants
completed process measures online. They were told that the
therapist would be available if they were experiencing
distress (none took up this offer). Participants were
reminded that they were entering the unsupported main-
tenance period and that the therapist would be in touch
2–3 weeks before the final weigh-in, to arrange the
appointment. They received £15 for their time and travel.
At 12 months, participants returned for the final weigh-
in. This session did not include self-report instruments.
They received £5 for completing this assessment.
Data analysis
Weight and waist circumference were analysed separately.
To estimate differences between MI and FIT, outcome
measurements were regressed onto baseline score (kg or
cm), a time indicator (6/12 months), and a group indicator
(MI/FIT), using linear mixed-effects models [47]. These
models also included baseline BMI and its interaction with
time and group; baseline BMI was included because it
captures additional information about the severity of parti-
cipants’condition when entering the study. These models
are analogous to repeated measures ANCOVA, but allowed
us to make efficient use of all available data without
imputation of missing values. Of primary interest were the
between-groups contrasts for weight and waist cir-
cumference at 6 and 12 months. Tests of parameter values
and other contrasts are reported with Satterthwaite approx-
imation for degrees of freedom. In a secondary analysis,
GQOL scores at 6 months were regressed on baseline
L. Solbrig et al.
GQOL scores, baseline BMI, group, and the interaction of
baseline scores with group. Alternative model para-
meterisations, in which treatment effects were estimated as
a linear slope from baseline to 6 or 12 months, produced
equivalent inferences.
To support probability statements about the average
effect of FIT vs. MI, and likely prognoses of future parti-
cipants selecting FIT or MI, we re-ran our mixed models
using a Bayesian estimation procedure with pessimistic but
weakly informative priors [48]; full details are available in
our data supplement, but for regression coefficients these
priors were Gaussian, zero-centred, and with a scale
adjusted to 2.5 × SD(y) / SD (Gabry, J., & Goodrich, B.
(2016). rstanarm: Bayesian applied regression modeling via
stan. R package version,2[1].) Based on these models, we
provide summaries of the posterior density for the average
treatment effect, and for the predicted prognoses of new
individuals selecting FIT or MI. All models appeared to
converge satisfactorily based on visual inspection of
MCMC traces and parameter R-hat statistics [49]. All the
data and R code for the analyses presented here are avail-
able in an online supplement.
Economic costing
We used Public Health England’s weight management
economic assessment tool No. 2 [50], to estimate the
increase in quality-adjusted life-years (QALY) associated
with the additional weight lost by FIT participants at
12 months compared with MI.
Results
A total of 141 participants were recruited in the time
available (March -May 2016). One hundred and 21 were
randomised, 58 to MI and 63 to FIT. One hundred and 14
were included in the analysis of the 6-month follow-up, and
of those 112 completed both 6 and 12-month follow-ups
(Consort Diagram, Fig. 1; Table A in the supplementary
materials). No statistically significant differences were
found between groups at baseline (Table 1). Twenty-one out
of 58 FIT participants reported having used the app or audio
when asked at 6 months. Twenty-five out of 55 MI parti-
cipants had continued to use their goal-sheet past the first
MI session when asked at 6 months.
MI and FIT fidelity checks
MI skills were rated on the MITI’s[44] 5-point scale: 1:
Never, 2: Rarely, 3: Sometimes, 4: Often, 5: Always. For
MI, ratings for Evocation, Collaboration, Autonomy, Sup-
port, Direction, and Empathy ranged from 3.9 to 4.9
(median =4.5). For FIT, they ranged from 3.8–4.7 (median
Fig. 1 CONSORT Flow
Diagram
Functional imagery training versus motivational interviewing for weight loss: a randomised controlled. . .
=4.5). For FIT, 15 session elements were rated as 0 absent,
or 1 present. Totals ranged from 13–15 (median =15).
Independent raters correctly assigned 100% of audio
recordings to intervention.
To visualise changes in weight and waist, we plotted
unadjusted means and 95% confidence intervals for each
group. Figure 2indicates that participants treated with MI
experienced little to no reduction in weight or waist from
baseline to either 6 or 12-month follow-ups. Those treated
with FIT experienced large reductions in weight and waist
circumference. Relative to both the MI group and baseline,
participants treated with FIT continued to lose weight after
treatment ended.
Our primary statistical models estimated the differences
between-groups, conditional on baseline BMI, at month 6
and 12; results are presented in Table 2. We found sub-
stantial and statistically significant differences between the
MI and FIT groups at both follow-ups.
To make probability statements about the size of benefit
obtained by participants undergoing FIT, we re-estimated
our mixed models using a Bayesian procedure. Table 2
shows treatment effects, and associated 95% credible
intervals from these models.
For weight and waist circumference, there was over-
whelming evidence that FIT was beneficial. Similarly, for
GQOL [45] the difference between groups at month 6 was
statistically significant. In the FIT group, 58/59 partici-
pants would recommend the intervention to others; one
might recommend. In MI, 53/55 would recommend, 2
might.
Table 1 Baseline demographics,
split by intervention Group/intervention MI (range, median, mean) FIT (Range, Median, Mean)
N55 59
Gender (N) Female 40 Female 43
Male 15 Male 16
Age (N years) Range 19–70 Range 20–72
Median 43 Median 45
Mean 42 Mean 45
BMI (kg/m²) Range: 24.51–53.33 Range 25.98–47.97
Median 31.34 Median 31.85
Mean 32.54 Mean 33.21
Weight (kg) Range 131–59 Range 62.30–140.5
Median 87.40 Median 89.40
Mean 89.66 Mean 91.46
Waistline (cm) Range 80–148 Range 79–144
Median 105 Median 103
Mean 106.01 Mean 106.78
Employment status (N) Employed full-time 12 Employed full-time 20
Employed part-time 18 Employed part-time 14
Retired 5 Retired 6
In education 4 In education 7
Unemployed 3 Unemployed 5
Other 11 Other 4
Self-employed 1 Self-employed 2
Highest level of education (N) GCSE 15 GCSE 14
NVQ/Diploma 4 NVQ/Diploma 5
Trade 8 Trade 6
A or O-Levels 14 A or O-Levels 15
Access course 2 Access course 4
Foundation degree 2 Foundation degree 1
Degree Degree
Postgraduate 4 Postgraduate 3
Degree 1 Degree 4
No info given 4 No info given 5a
aThere were no statistically significant differences between groups at baseline
L. Solbrig et al.
To help clinicians and others evaluate the likely benefit
of FIT in clinical practice, we computed the posterior-
probability that the benefit of FIT for a new participant
would exceed a range of values between 0 and 15 kg lost,
and between 0–15 cm of waist reduction (Fig. 3).
When considering the risks or benefits of interventions,
clinicians, participants and researchers benefit from prob-
ability information presented as ‘natural frequencies’or in
‘pictographs’[51]. Consequently, we used the same model-
based simulations to calculate the range of likely prognoses
from the participants’perspective (Fig. 4). After treatment,
only 22% of MI participants were predicted to lose 5% or
more of their initial weight, compared with the NICE target
[52] that 30% do; after 12 months this figure was 23%. In
contrast, 54% of new FIT participants were predicted to lose
at least 5% of their initial weight after treatment, and 75%
are predicted to lose this much by 12-months. Stated dif-
ferently, nine of every 10 participants would have benefitted
more from FIT than from MI (probability FIT > MI for kg
lost at 12 m =0.94; for cm lost it is 0.85); for half of these
participants, the expected additional benefit is substantial
(>5 kg difference in projected outcomes, see data
supplement).
Economic assessment
We based our inputs to the PHE assessment tool [50] on the
conservative assumption that MI was equivalent to no-
treatment, and that FIT participants would begin to gain
weight immediately after the 12-month follow-up. We
modelled costs on the basis that 58 new patients would be
treated with FIT and that they would reduce their BMI by an
average of 2.148 kg/m2over 1 year (based on our primary
outcome models). We very conservatively assumed that
each hour of individual treatment would cost £250 to deli-
ver, and entered a worst-case per-participant cost for FIT at
£1000.
Even based on these highly conservative assumptions,
the PHE model suggests that FIT would be cost-effective
from the healthcare perspective within 3 years, judged by
NICE’s conventional willingness-to-pay threshold of
between £20,000 and £30,000 per QALY. Cost-per-QALY
after 3 years was £22,036, falling to £12,363 after five
years, and £7,229 after 10 years. Including costs of social
care and the prospect of increased employment, cost per
QALY was only £12,968 after 3 years, £3,739 after 5 years,
and was cost-saving from a 10-year perspective.
Discussion
In this first randomised controlled trial of FIT, we have
shown that two FIT interviews and nine brief booster phone
calls, amounting to under 4 h of therapist contact over
6 months, resulted in substantially greater and clinically
meaningful weight loss and waistline reductions at
6 months, compared with MI. Participants in the FIT arm,
but not MI, continued to lose weight and waist cir-
cumference in the unsupported 6-month maintenance phase.
Participants in both treatment groups reported improved
Fig. 2 Mean weight and waist
circumference with 95%
confidence interval, by group
Functional imagery training versus motivational interviewing for weight loss: a randomised controlled. . .
quality of life at 6 months, but FIT participants reported
greater improvements. Seventy-one percent of the FIT
group lost >5% of their initial weight, easily exceeding the
NICE weight management target that at least 30% of
service-users should do so [52]. MI did not meet the NICE
target, with only 23% losing 5%.
Importantly, we found that FIT was acceptable to parti-
cipants and that they would recommend FIT to a family
member, friend or colleague. Because the delivery of FIT
closely matches the protocol of an existing intervention
(MI), scaling delivery to larger numbers of participants
should be straightforward, with existing MI practitioners
requiring only minimal additional training.
To put these findings into a broader context, FIT per-
formed favourably compared to a longer, more intensive
intervention in a recent UK trial: Ahern and colleagues [53]
tested weight loss in participants with overweight and obesity
referred by GPs to Weight Watchers. Participants randomised
to the Weight Watchers programme for 12 months had lost an
average of 6.8 kg at 12 months, only 0.4 kg more than
participants in our RCT, who received <4 h of FIT spread
over 6 months. Participants on the standard 12-week Weight
Watchers programme lost less than FIT participants in this
study: only 4.8 kg on average [53].
The approach tested in this trial, of providing solely
motivational support, differs from the strategy recom-
mended by Public Health England [54,55] of combining
behaviour change techniques with lifestyle education and
advice. NICE recommends including MI and imagery in
behaviour change strategies. FIT combines both in a
coherent, structured intervention that trains users to become
their own therapist; the present results support this
approach. It remains to be determined if combining FIT
with diet and physical activity education would generate
superior outcomes.
Sustained reductions of around 5% of body weight can
effect significant health improvements, such as decreased
blood lipids, precursors of Type 2 diabetes and improved
blood pressure [3,4]. Weight loss of between 5 and 10% is
associated with significant improvements in cardiovascular
disease risk factors [5]. The reduction in waist circumference,
from an average of 106 cm to 97 cm by 12 months in FIT,
brings its own health risk reductions: excess abdominal fat
indicated by a high waist circumference in men (>102cm),
presents a five-fold increase in the risk of developing diabetes
[9]. For women, across a range of baseline BMIs between 25
and 50 kg/m², waist reduction of 5–10 cm is associated with a
reduction in cholesterol and systolic blood pressure [56].
Results from the Public Health England economic assessment
tool suggest that these health benefits would make delivery of
FIT cost effective, although detailed cost-effectiveness eva-
luation must form part of additional large-scale evaluations of
FIT.
The fact that FIT outperformed the best established
motivational intervention by such a margin is encouraging.
We demonstrated a mean weight difference of 2 kg for FIT,
compared with the active MI intervention group at
6 months. This benefit of FIT over MI was larger than the
Table 2 Between group contrasts (with Satterthwaite corrected degrees of freedom for Kg and Cm) and posterior mean differences (and 95%
credible intervals) for the effect of FIT vs. MI at month 6 and 12.
Outcome Follow-up FIT (mean) FIT (sd) MI (mean) MI (sd) dft p Treatment effect (MCMC) lower upper
Cm Baseline 106.07 13.75 105.50 12.51
Cm Month 6 99.05 12.61 102.78 13.37 205.4 −4.727 <.001 −4.444 −6.328 −2.560
Cm Month 12 96.97 12.59 103.04 12.45 206.1 −7.012 <.001 −6.697 −8.602 −4.801
Kg Baseline 90.48 15.90 89.13 14.76
Kg Month 6 86.37 15.07 88.39 15.72 161.5 −4.877 <.001 −3.670 −5.203 −2.139
Kg Month 12 84.04 15.96 88.46 15.34 163.4 −7.707 <.001 −5.929 −7.482 −4.418
GQOL Baseline 62.13 10.59 61.71 14.51
GQOL Month 6 75.81 11.66 72.53 10.42 109.0 2.107 .037 2.831 0.091 5.565
Cm waist circumference in cm, kg mean weight in kg, GQOL mean score of Global Quality of Life assessment, MCMC Markov Chain Monte
Carlo estimates from Bayesian model fits
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
02468101214
Benefit to new patient choosing FIT vs. MI.
Probability
Outcome
Cm
Kg
Fig. 3 Prognosis for new participants randomised to FIT vs. MI,
expressed as the probability the benefit will equal or exceed the value
on the x-axis
L. Solbrig et al.
benefit of MI over minimal control interventions in Arm-
strong et al’s. [24], meta-analysis of MI for weight loss:
mean difference =1.47 kg 95% CI =−2.05 to −0.88 at
around 6 months. However, we note thfat weight in the MI
group stabilised after six months, and successfully pre-
venting weight gain is an important health focus [57,58].
That FIT demonstrates such an improvement over an
existing treatment, and produces continued weight loss after
the end of the intervention period, highlights the benefitof
developing and adapting existing interventions based on
recent developments in cognitive science.
Limitations and future directions
We focused purely on motivation, providing no lifestyle
advice or information. We did not assess participants’
knowledge of nutrition or physical activity. First, this was in
keeping with the autonomous spirit of MI which FIT
retains. The ethos of both interventions is to motivate par-
ticipants to seek the information or support they need to
achieve their goals. Second, we focused on process out-
comes in this trial, rather than content outcomes. Assessing
previous knowledge at baseline could have led participants
to believe they should discuss their knowledge, or to expect
advice or prescribed diet and exercise plans. Although we
recruited from the general public, we acknowledge that we
may have attracted a well-informed sample (e.g., the kind of
person who would sign up for a study conducted at a uni-
versity). Perhaps this sample was more knowledgeable than
a random sample of people with overweight, but this cer-
tainly did not make them more successful; they still strug-
gled with overweight or obesity when they signed up for the
study. There is little evidence that knowledge alone moti-
vates behaviour. For example, a recent, large cross-sectional
study on self-management of glycemic control found no
correlation between knowledge about type 2 diabetes and
actual self-management behaviours [59]. Even with
healthcare professionals, education did not lead to beha-
viour change. McCluskey and Lovarini, (2005) tested an
educational intervention to improve evidence-based practise
amongst allied health professionals. Improved knowledge
was maintained at 8 month follow-up but behaviour change
was very limited, with nearly two thirds of health profes-
sionals still not reading any research literature [60]. It is
conceivable, however, that education is beneficial in
populations with very limited knowledge of nutrition and
exercise. Future studies should therefore assess whether FIT
works best as an alternative to established weight-loss
programmes based on lifestyle education and advice, or as
an adjunct to them.
As far as possible, we matched MI and FIT for inter-
vention intensity. There were the same number of sessions,
scheduled for the same duration, in both conditions.
Although the dialogue for FIT incorporated the same
essential components as for MI, we added some extra ele-
ments to the MI manual to add depth to the interview and
equate the time taken. In FIT, all participants were offered
the Goal in Mind app to guide their imagery practice. In MI,
all participants were given goal sheets to review at home. If
anything, MI participants had more time to talk to the
therapist because they did not do imagery exercises.
FIT achieved larger effect sizes than expected. Because
of practical constraints on recruitment, our sample was
smaller than that recommended by our power calculations to
detect modest to moderate effect sizes, therefore it is
MI FIT
Fig. 4 12 month prognosis for
100 new participants undergoing
MI or FIT
Functional imagery training versus motivational interviewing for weight loss: a randomised controlled. . .
plausible that our trial over-estimates the true effect of FIT,
i.e. a ‘Type M’(magnitude) Error [61]. Nonetheless, the
posterior probability for an effect of FIT <1 kg lost was
extremely low, and the evidence that FIT was preferable to
MI was substantial.
This was the first efficacy trial of FIT for weight loss, and
our results must be replicated in a larger multi-centre trial to
be confident that FIT can be delivered effectively at scale by
other therapists. However, we have achieved proof of
concept, and compared the intervention with a well-
established intervention for weight loss. Although the FIT
group maintained their weight loss at our 12- month follow-
up, further trials should follow participants for several years
to ensure that FIT does indeed help participants maintain
the weight loss achieved during treatment, and to properly
assess the health economic benefit of these reductions.
Conclusion
Less than 4 h of Functional Imagery Training, a novel
intervention that combines motivational interviewing with
mental imagery training, led to substantially greater weight
loss over 6 months than MI alone, despite the fact that
specific education on lifestyle and activity was absent from
this intervention. The benefits of FIT persisted to
12 months; participants continued to make substantial
reductions in weight even after therapy ended.
Disclaimer
The views expressed are those of the author(s) and not
necessarily those of the NHS, the NIHR or the Department
of Health.
Acknowledgements We would like to thank the research assistants
who have tirelessly supported this project: Despina Djama, Lloyd
Taylor, Kirsten Woodman, and Marina Khalil.
Funding This research was funded by the National Institute for Health
Research (NIHR) Collaboration for Leadership in Applied Health
Research and Care South West Peninsula.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
References
1. Eastwood P. Statistics on obesity, physical activity and diet:
lifestyles. Health Soc Care Inf Cent. 2013;2:15–35.
2. Health and Social Care Information Centre. Statistics on obesity,
physical activity and diet: England 2015. Heal Soc Care Inf Cent.
2015;103. http://www.hscic.gov.uk/catalogue/PUB16988/obes-
phys-acti-diet-eng-2015.pdf.
3. Jakicic JM, Clark K, Coleman E, Donnelly JE, Foreyt J, Melanson
E, et al. Appropriate strategies for intervention weight loss and
prevention of weight regain for adults. Med Sci Sports Exerc.
2001;33:2145–56.
4. Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW,
Smith BK. Appropriate physical activity intervention strategies for
weight loss and prevention of weight regain for adults. Med Sci
Sports Exerc. 2009;41:459–71.
5. Wing R, Lang W, Wadden T, Safford M, Knowler W, Bertoni A,
et al. Benefits of modest weight loss in improving cardiovascular
risk factors in overweight and obese individuals with type 2 dia-
betes. Diabetes Care [Internet]. 2011;34:1481–6.
6. Espeland M. Reduction in weight and cardiovascular disease risk
factors in individuals with type 2 diabetes: one-year results of the
look AHEAD Trial. Diabetes Care. 2007;30:1374–83.
7. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H,
Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by
changes in lifestyle among subjects with impaired glucose toler-
ance. N Engl J Med. 2001;344:1343–50. https://doi.org/10.1056/
NEJM200105033441801.
8. Williams G, Hamm MP, Shulhan J, Vandermeer B, Hartling L,
Avenell A, et al. Reduction in weight and cardiovascular disease
risk factors in individuals with type 2 diabetes: one-year results of
the look AHEAD trial received. Natl Inst Heal Clin Excell.
2012;8:1585–93. http://www.pubmedcentral.nih.gov/articlerender.
fcgi?artid=3996838&tool=pmcentrez&rendertype=abstract.
9. Gatineau M, Hancock C, Holman N, Outhwaite H, Oldridge L,
Christie A, Ells L. Adult obesity and type 2 diabetes. About Public
Health England. London: Public Health England; 2014. p. 1–39.
10. Hartmann-Boyce J, Johns DJ, Jebb SA, Aveyard P, Ogden J,
Onakpoya I, et al. Effect of behavioural techniques and delivery
mode on effectiveness of weight management: Systematic review,
meta-analysis and meta-regression. Obes Rev. 2014;15:598–609.
11. Truby H, Baic S, DeLooy A, Fox KR, Livingstone MBE, Logan
CM, et al. Randomised controlled trial of four commercial weight
loss programmes in the UK: initial findings from the BBC “diet
trials”. BMJ. 2006;332:1309–14. http://www.pubmedcentral.nih.
gov/articlerender.fcgi?artid=1473108&tool=pmcentrez&renderty
pe=abstract.
12. Wadden TA, Butryn ML, Byrne KJ, Efficacy of lifestyle mod-
ification for long-term weight control. Obes Res. 2004;12
(S12):151S–62. http://doi.wiley.com/10.1038/oby.2004.282.
13. Bessesen DH. Systematic review: an evaluation of major com-
mercial weight loss programs in the United States. Yearb Endo-
crinol. 2006;2006:153–7.
14. Kozica S, Lombard C, Teede H, Ilic D, Murphy K, Harrison C,
Initiating and continuing behaviour change within a weight gain
prevention trial a qualitative investigation. PLoS ONE.
2015;10:1–14. https://doi.org/10.1371/journal.pone.0119773.
15. Wadden TA. Eight-year weight losses with an intensive lifestyle
intervention: The look AHEAD study. Obesity. 2014;22:5–13.
16. Ross Middleton KM, Patidar SM, Perri MG. The impact of
extended care on the long-term maintenance of weight loss: A
systematic review and meta-analysis. Obes Rev. 2012;13:509–17.
17. West DS, Gorin AA, Subak LL, Foster G, Bragg C, Hecht J, et al.
A motivation-focused weight loss maintenance program is an
effective alternative to a skill-based approach. Int J Obes.
2011;35:259–69. http://www.scopus.com/inward/record.url?eid=
2-s2.0-79951683411&partnerID=tZOtx3y1.
18. Elfhag K, Rössner S, Rossner S, Wu T,Gao X, Chen M, et al.
Who succeeds in maintaining weight loss? A conceptual review of
factors associated with weight loss maintenance and weight
regain. Obes Rev. 2005;6:67–85.
L. Solbrig et al.
19. Silva MN, Markland D, Carraca EV, Vieira PN, Coutinho SR,
Minderico CS, et al. Exercise autonomous motivation predicts 3-
yr weight loss in women. Med Sci Sports Exerc. 2011;43:
728–37.
20. Teixeira PJ, Going SB, Sardinha LB, Lohman TG. A review of
psychological pre-treatment predictors of weight control. Obes
Rev. 2005;6:43–65.
21. Sabinsky MS, Toft U, Raben A, Holm L. Overweight men’s
motivations and perceived barriers towards weight loss. Eur J Clin
Nutr. 2007;61:526–31.
22. SharifiN, Mahdavi R, Ebrahimi-Mameghani M. Perceived bar-
riers to weight loss programs for overweight or obese women.
Heal Promot Perspect. 2013;3:11–22. http://www.pubmedcentral.
nih.gov/articlerender.fcgi?artid=3963684&tool=pmcentrez&
rendertype=abstract.
23. Solbrig L, Jones R, Kavanagh D, May J, Parkin T, Andrade J.
People trying to lose weight dislike calorie counting apps and
want motivational support to help them achieve their goals.
Internet Interv. 2017;7:23–31. https://doi.org/10.1016/j.invent.
2016.12.003.
24. Miller W, Rollnick S. Motivational interviewing: helping people
change. Guilford press; New York, 2012.
25. Armstrong MJ, Mottershead Ta, Ronksley PE, Sigal RJ, Campbell
TS, Hemmelgarn BR. Motivational interviewing to improve
weight loss in overweight and/or obese patients: A systematic
review and meta-analysis of randomized controlled trials. Obes
Rev. 2011;12:709–23.
26. Hardcastle SJ, Taylor AH, Bailey MP, Harley RA, Hagger MS.
Effectiveness of a motivational interviewing intervention on
weight loss, physical activity and cardiovascular disease risk
factors: a randomised controlled trial with a 12-month post-
intervention follow-up. Int J Behav Nutr Phys Act. 2013;10:40.
http://www.ncbi.nlm.nih.gov/pubmed/23537492.
27. Martins RK, McNeil DW, Review of motivational interviewing in
promoting health behaviors. Clin Psychol Rev. 2009;29:283–93.
https://doi.org/10.1016/j.cpr.2009.02.001.
28. VanWormer JJ, Boucher JL. Motivational interviewing and diet
modification: a review of the evidence. Diabetes Educ.
2004;30:404–19.
29. Kavanagh DJ, Andrade J, May J, Connor JP. Motivational inter-
ventions may have greater sustained impact if they trained
imagery-based self-management. Addiction. 2014;109:1062–3.
30. Olander EK, Fletcher H, Williams S, Atkinson L, Turner A,
French DP. What are the most effective techniques in changing
obese individuals’physical activity self-efficacy and behaviour: A
systematic review and meta-analysis. Int J Behav Nutr Phys Act.
2013;10:29. http://www.pubmedcentral.nih.gov/articlerender.
fcgi?artid=3639155&tool=pmcentrez&rendertype=abstract.
31. Andrade J, May J, Kavanagh D, Sensory imagery in craving: from
cognitive psychology to new treatments for addiction. J Exp
Psychopathol. 2012;3:127–45.
32. Kavanagh DJ, Andrade J, May J. Imaginary relish and exquisite
torture: the elaborated intrusion theory of desire. Psychol Rev.
2005;112:446–67.
33. Kavanagh DavidJ, Andrade Jackie, May Jon, Connor JP. Do brief
alcohol motivational interventions work like we think they do?
Addiction. 2014;109:1062–3.
34. Blackwell SE, Rius-Ottenheim N, Schulte-van Maaren YWM,
Carlier IVE, Middelkoop VD, Zitman FG, et al. Optimism and
mental imagery: A possible cognitive marker to promote well-
being?. Psychiatry Res. 2013;206:56–61. https://doi.org/10.1016/j.
psychres.2012.09.047.
35. Holmes E, Mathews A. Mental imagery and emotion: a special
relationship? Emotion. 2005;5:489–97.
36. Renner F, Ji JL, Pictet A, Holmes EA, Blackwell SE. Effects of
engaging in repeated mental imagery of future positive events on
behavioural activation in individuals with major depressive dis-
order. Cogn Ther Res Springe US. 2017;41:369–80.
37. Parham SC, Kavanagh DJ, Gericke CA, King N, May J, Andrade
J. Assessment of motivational cognitions in diabetes self-care: the
motivation thought frequency scales for glucose testing, physical
activity and healthy eating. Int J Behav Med. 2016;24:447–56.
https://doi.org/10.1007/s12529-016-9607-2
38. Knäuper B, Roseman M, Johnson PJ, Krantz LH. Using mental
imagery to enhance the effectiveness of implementation inten-
tions. Curr Psychol. 2009;28:181–6.
39. Knäuper B, McCollam A, Rosen-Brown A, Lacaille J, Kelso E,
Roseman M. Fruitful plans: adding targeted mental imagery to
implementation intentions increases fruit consumption. Psychol
Heal. 2011;26:601–17.
40. Michie S, Ashford S, Sniehotta F, Dombrowski S, Bishop A,
French D. A refined taxonomy of behaviour change techniques to
help people change their physical activity and healthy eating
behaviours: the CALO-RE taxonomy. Psychol Health.
2011;26:1479–98. http://discovery.ucl.ac.uk/1311528/.
41. Neill JO, Oluyomi T, Epstein LH, Eating Behaviors Episodic
future thinking reduces eating in a food court. Eat Behav.
2016;20:9–13. https://doi.org/10.1016/j.eatbeh.2015.10.002.
42. Andrade J, Khalil M, Dickson J, May J, Kavanagh DJ. Functional
imagery training to reduce snacking: testing a novel motivational
intervention based on Elaborated Intrusion theory. Appetite.
2016;100:256–62. http://linkinghub.elsevier.com/retrieve/pii/
S0195666316300502.
43. May J, Andrade J, Panabokke N, Kavanagh D, Visuospatial tasks
suppress craving for cigarettes. Behav Res Ther. 2010;48:476–85.
https://doi.org/10.1016/j.brat.2010.02.001.
44. Bandura A. Self-efficacy mechanism in human agency. Am
Psych. 1982;37:122–47.
45. Hyland ME, Sodergren SC. Development of a new type of global
quality of life scale, and comparison of performance and pre-
ference for 12 global scales. Qual Life Res. 1996;5:469–80.
http://www.jstor.org/stable/pdf/4034400.pdf.
46. Moyers TB, Martin T, Manuel JK, Miller WR, Ernst D. Revised
Global Scales: (MITI 3. 1. 1). 2010;1–29. https://casaa.unm.edu/
download/miti3_1.pdf
47. Bates D, Mächler M, Bolker BM, Walker SC, Fitting linear
mixed-effects models using lme4. J Stat Softw. 2015;67:1–48.
http://www.jstatsoft.org/index.php/jss/article/view/v067i01/v67i
01.pdf.
48. Gabry J, Goodrich B. rstanarm: Bayesian applied regression mod-
eling via stan [Computer software manual]. R package version. http://
cran.r-project.org/package=rstanarm. 2016. Accessed on 1 Septem-
ber 2017.
49. Gelman A, Carlin JB, Stern HS, Dunson DB, Vehtari A, Rubin
DD. Bayesian data analysis. CRC Press; Boca Raton, FL, USA,
2014.
50. Copley V. User guide: weight management economic assessment
tool Economic assessment of adult weight management inter-
ventions About Public Health England. London: Public Health
England; 2016.
51. Tait AR, Voepel-Lewis T, Zikmund-Fisher BJ, Fagerlin A. The
effect of format on parents’understanding of the risks and benefits
of clinical research: A comparison between text, tables, and gra-
phics. J Health Commun. 2010;15:487–501.
52. NICE. Obesity: identification, assessment and management. Natl
Inst Health Care Excell. 2014. https://www.nice.org.uk/guidance/
cg189/ifp/chapter/obesity-and-being-overweight.
53. Ahern AL,Wheeler GM,Aveyard P,Boyland EJ, Halford JCG,
Mander AP, et al. Extended and standard duration weight-loss
programme referrals for adults in primary care (WRAP): a ran-
domised controlled trial. Lancet. 2017;389:2214–25. https://doi.
org/10.1016/S0140-6736(17)30647-5.
Functional imagery training versus motivational interviewing for weight loss: a randomised controlled. . .
54. Thompson L, Aveyard P, Jebb S, Blackshaw J, Coulton V. Let’s
talk about weight: a step-by-step guide to brief interventions with
adults for health and care professionals. London: Public Health
England; 2017. https://assets.publishing.service.gov.uk/governm
ent/uploads/system/uploads/attachment_data/file/649095/child_w
eight_management_lets_talk_about_weight.pdf.
55. NICE. W eight management: lifestyle services for o ov verweight
or obese children and young people. Natl Inst Heal Clin Excell.
2017. https://www.nice.org.uk/guidance/ph53.
56. Han TS, Richmond P, Avenell A, Lean MEJ, Waist circumference
reduction and cardiovascular benefits during weight loss in
women. Int J Obes [Internet]. 1997;21:127–34. https://doi.org/10.
1038/sj.ijo.0800377.
57. Lloyd-Jones DM, Liu K, Colangelo LA, Yan LL, Klein L, Loria
CM, et al. Consistently stable or decreased body mass index in
young adulthood and longitudinal changes in metabolic syndrome
components: The coronary artery risk development in young
adults study. Circulation. 2007;115:1004–11.
58. Strong Ka, Parks SL, Anderson E, Winett R, Davy BM. NIH
public access. J Am Diet Assoc. 2008;108:1708–15.
59. Chen Q, Wang H, Wang Y, Wang Z, Zhao D. CAIY exploring
self-management on glycemic control using a modified
information–motivation–behavioral skills model in type 2 diabetes
mellitus patients in Shanghai, China: a cross-sectional study
IMB model of self-management on HbA. J Diabetes. 2018;86:
973–83.
60. McCluskey A, Lovarini M, Providing education on evidence-
based practice improved knowledge but did not change behaviour:
a before and after study. BMC Med Educ. 2005;5:40. https://doi.
org/10.1186/1472-6920-5-40.
61. Gelman A, Carlin J, Beyond power calculations. Perspect Psychol
Sci. 2014;9:641–51. https://doi.org/10.1177/1745691614551642.
L. Solbrig et al.
- A preview of this full-text is provided by Springer Nature.
- Learn more
Preview content only
Content available from International Journal of Obesity
This content is subject to copyright. Terms and conditions apply.
