clinical obesity 10.1111/cob.

12135
doi: 10.1111/cob.12135

Review of 3-year outcomes of a very-low-energy

diet-based outpatient obesity treatment programme

P. Sumithran1,2, L. A. Prendergast3, C. J. Haywood1,2, C. A. Houlihan1 and J. Proietto1,2

What is already known about this subject What this study adds
• A multidisciplinary approach is recommended for the treatment of • This review assessed long-term (3-year) outcomes of a dedicated
obesity. outpatient obesity treatment clinic.
• Weight loss achieved using lifestyle methods alone is usually • Clinically beneficial weight loss can be achieved using a medically
regained over time. supervised very-low-energy diet, continued follow-up and
• Attrition from long-term weight management programmes is high. consideration of pharmacotherapy to prevent weight regain.
• Several baseline characteristics were identified as predictors of
attrition.

1
Austin Health Weight Control Clinic, Summary
Heidelberg Repatriation Hospital, Heidelberg, Obesity is a complex disorder that requires a multidisciplinary treatment
Vic., Australia; 2Department of Medicine approach. This review evaluated 3-year outcomes of a very-low-energy diet
(Austin Health), University of Melbourne, (VLED)-based programme at a tertiary hospital multidisciplinary weight manage-
Heidelberg, Vic., Australia; 3Department of ment clinic. Medical records of all patients who agreed to undertake the VLED
Mathematics and Statistics, La Trobe programme and who did not undergo bariatric surgery during the 3-year
University, Bundoora, Vic., Australia follow-up period were examined. Baseline data collection included demographic
and anthropometric characteristics, childhood onset of obesity and co-existing
Received 11 November 2015; revised 9 medical conditions. Weight was modelled using a linear mixed effects analysis.
December 2015; accepted 11 December Logistic regression analyses were used to model the probability of continuing to
2015 attend the clinic and to identify pre-treatment factors associated with longer
duration of attendance. Data from 1109 patients were included. A total of 231
Address for correspondence: Professor J patients (19.2%) were still attending the clinic 3 years after their initial appoint-
Proietto, Austin Health Weight Control Clinic, ment. Mean weight loss among patients who attended the clinic for 3 years was
Heidelberg Repatriation Hospital, 300 6.4 kg (3.5%, 95% confidence interval [CI] 2.8, 4.2%). People who were pre-
Waterdale Rd., Heidelberg, Vic. 3081, scribed pharmacotherapy maintained greater weight loss at 3 years (7.7% vs.
Australia. E-mail: j.proietto@unimelb.edu.au 2.3% without pharmacotherapy, 95% CI for difference 3.9, 7.0%). People who
had an onset of obesity in childhood, who had co-existing hypertension or
coronary artery disease, and who did not currently smoke were more likely to
continue to attend the clinic for up to 3 years. In summary, in an outpatient weight
management clinic, patients who undertook a VLED-based programme and con-
tinued in follow-up achieved a clinically significant weight loss at 3 years, par-
ticularly if pharmacotherapy was used for weight loss maintenance.

Keywords: Outpatient, very-low-energy diet, VLED, weight management clinic.

factorial disorder, a multidisciplinary treatment approach is

Introduction
recommended, including dietary and exercise components,
In Australia, almost two-thirds (63%) of adults are over- behaviour modification, and consideration of pharmaco-
weight or obese (1). Since obesity is a complex and multi- therapy and bariatric surgery in selected patients (2).

clinical obesity 6, 101–107

© 2016 World Obesity. clinical 1
101
2102Review of of
Review weight clinic
weight 3-year
clinic outcomes
3-year outcomes P.P.Sumithran
Sumithranetetal.al. clinical obesity
obesity

Weight loss achieved using lifestyle methods alone is 15–40 mg daily, exenatide 5–10 μg bd in people with type
usually regained over time (2,3); therefore, it is imperative 2 diabetes, and following the withdrawal of sibutramine
that evidence-based weight-loss programmes including from the Australian market in 2010, the combination of
strategies to prevent weight regain are developed and phentermine 15 mg and topiramate 12.5–50 mg daily.
assessed. Patients who regain weight despite lifestyle modification
Our hospital, a tertiary referral centre in metropolitan and pharmacotherapy, or in whom pharmacotherapy is
Melbourne, established a Weight Control Clinic in Febru- contraindicated, are offered a referral for bariatric surgery.
ary 2004. The clinic offers an appointment to all adult There are no fees for clinic appointments. Patients purchase
patients referred with a body mass index (BMI) above the VLED products at their own expense, and the cost of
35 kg m−2, or >30 kg m−2 with a weight-related pharmacotherapy is subsidized (cost to the patient $12–
comorbidity, who have made a previous unsuccessful $50 [AUD] per month depending on medication type, dose
attempt to lose weight or maintain weight loss. The treat- and patient’s eligibility for concession). The aim of this
ment programme is based on the following premises: the report is to evaluate the 3-year outcomes of this treatment
predisposition to obesity has a genetic and/or epigenetic programme. The Austin Health Human Research Ethics
basis (4,5); body weight is physiologically defended (6–8); Committee approved this review.
a very-low-energy diet (VLED) is more effective than a
hypocaloric balanced diet for initial weight loss (3,9); over
Methods
the long-term, weight regain is likely (2,3) therefore long-
term follow-up is required, pharmacotherapy to suppress Medical records of all patients who attended the clinic
appetite should be offered to people who are unable to between February 2004 and December 2012 were exam-
maintain weight loss and bariatric surgery should be avail- ined. Patients whose first appointment was between Febru-
able if medical therapy is unsuccessful. ary 2004 and December 2009, who attended the clinic at
least once, and agreed to undertake the VLED programme
were included in this review of 3 year outcomes. All
Treatment programme
patients referred to the clinic were made an appointment,
The programme comprises an initial weight-loss phase and if an appointment was missed, another was booked, of
using a VLED and an indefinite long-term weight-loss which patients were notified by letter, on at least two occa-
maintenance phase. A medical evaluation of patients, sions, before patients were considered lost to follow-up.
including a comprehensive medical history and physical
examination, is undertaken before initiating the VLED.
Data extraction
During the period reviewed, patients were recommended
either a full VLED regimen (3 meal replacement products Information extracted from the record of the initial visit
plus 2 cups of non-starch vegetables daily; approximately included age, gender, height, weight, whether the onset of
2300 kJ [550 kcal] per day) or a partial regimen (2 meal obesity occurred in childhood, triggers for weight gain,
replacement products, plus one meal consisting of lean medical conditions, current medications and management
protein and 2 cups of non-starch vegetables; approximately plan (to identify eligibility for this review). Follow-up data
3350 kJ [800 kcal] per day), depending on clinical factors included body weight up to 3 years after the initial visit,
and patient preference. During the VLED phase, appoint- date of last attendance at the clinic and whether patients
ments are scheduled every 2–4 weeks for monitoring of underwent bariatric surgery. Whether or not pharmaco-
weight, blood pressure and medical conditions (e.g. adjust- therapy was prescribed was extracted for patients who
ment of medications for diabetes if required). attended clinic for at least 3 years. Information regarding
Before entering the long-term weight maintenance phase, changes in medical conditions and medications was not
dietary and lifestyle modifications are discussed. Patients systematically collected.
then make a transition from VLED to regular foods over
several weeks, with introduction of a relatively high-
Statistical analysis
protein reduced-carbohydrate diet. During the transition
and weight maintenance phases, patients are reviewed Descriptive data are expressed as means ± standard devia-
every 1–3 months. tions. Baseline characteristics were compared between men
For patients unable to maintain weight loss despite life- and women, and between completers (people who attended
style modification, pharmacotherapy may be used for appe- for at least 3 years) and dropouts using independent sample
tite reduction. During the period under review, the t-tests. Analyses of binary outcomes were carried out using
medication most often prescribed was sibutramine logistic regression with a logit link function. Profile likeli-
10–15 mg daily. Other pharmacotherapy prescribed hood 95% confidence intervals (CIs) were computed for
included topiramate 25–100 mg bd, phentermine the odds ratios. Due to the large number of covariates,

© 2016
© 2016 World World
Obesity. Obesity.
clinical clinical
obesity obesity
6, 101–107
clinical obesity Review
Review ofof weight
weight clinic
clinic 3-year
3-year outcomesP. P.
outcomes Sumithran
Sumithran et al.103
et al. 3

stepwise model selection based on the Akaike information Table 1 Baseline characteristics
criterion (AIC) was used. Prediction rates were calculated
Descriptives [means ± SD]
using 10-fold cross-validation. To account for the irregu-
larity of time points at which patients attended the clinic, Men/Women (n, %) 350 (31.6%)/759 (68.4%)
and the within-subject correlation of weight over time, Age (years) 48.3 ± 13.1
weight was modelled using a linear mixed effects (LME) Weight (kg) 125.0 ± 29.8
analysis with time treated as a continuous covariate, which BMI (kg m−2) 45.3 ± 9.3
SBP (mmHg) 134.1 ± 18.4
included polynomial terms up to and including order four
DBP (mmHg) 80.5 ± 13.6
to allow for non-linear change. Additionally, an intention- Waist circumference (cm) 127.8 ± 17.9
to-treat (ITT) analysis was undertaken, in which patients Hip circumference (cm) 137.6 ± 18.6
who stopped attending the clinic before 3 years’ follow-up
were assumed to have returned to their baseline weight. Comorbidities [n (%)]
Maximum percentage weight loss recorded during the three
Smoker 191 (18.5%)
years was analysed using multiple linear regression to
Hypertension 585 (56.7%)
adjust for covariates. Stepwise model selection based on the Dyslipidaemia 360 (35.0%)
AIC was used to select factors contributing most to the OSA 482 (46.8%)
model. T2DM 347 (33.7%)
Coronary artery disease 96 (9.3%)
Obese in childhood 374 (36.3%)
Mental illness* 336 (32.6%)
Results
For descriptives, n = 1109. For comorbidities, n = 1029–1032. *Includes
Clinic population depression, anxiety, bipolar disorder and schizophrenia.
BMI, body mass index; DBP, diastolic blood pressure; OSA, obstructive
Of the 2074 patients who attended their first appointment sleep apnoea; SBP, systolic blood pressure; T2DM, type 2 diabetes
between February 2004 and December 2009, 1287 (62%) mellitus.
agreed to commence the VLED programme. Of them, 178
underwent bariatric surgery during the follow-up period
and were excluded from analyses. Baseline characteristics continued attendance at 1 and 3 years with 76 and 83%
of the remaining 1109 patients are given in Table 1. accuracy, respectively (Table 2).
Approximately one-third of patients were male. Men were
heavier than women (137.9 ± 33.4 kg vs. 119.0 ± 26.0 kg;
P < 0.001), but BMI was not different between genders Weight loss
(44.9 ± 9.9 kg m‒2 vs. 45.4 ± 9.0 kg m‒2; P = 0.39). More
In plot A of Fig. 2, weight changes recorded at clinic
than half of patients had hypertension, nearly half had
appointments for up to 3 years for each of the 1109
obstructive sleep apnoea and one-third had type 2 diabetes.
patients who agreed to start the VLED programme are
depicted. For clarity, mean weight changes estimated using
the LME analysis at specified time points are provided in
Attrition
plot B of Fig. 2. Table 3 shows estimated means and asso-
Attrition from the clinic over time is depicted in Fig. 1. The ciated 95% CIs at 1, 2 and 3 years for all continuing
median duration of attendance was 248 days. Nine patients, and an ITT analysis assuming return to baseline
hundred and twenty-two patients (83.1%) returned for at weights for patients who stopped attending. The mean
least one follow-up visit, and 466 (42.0%) and 213 weight loss among patients continuing to attend the clinic
(19.2%) were still attending 1 and 3 years after their initial was 9.7 kg (6.2%) at 1 year and 6.4 kg (3.5%) at 3 years.
appointment. A logistic regression analysis of pre- The mean maximum weight loss percentage and contribut-
treatment factors associated with longer duration of ing factors identified in the analysis are given in Table 4 for
attendance, repeated at one and three years to model the the 1109 patients who agreed to commence the VLED
probability of patients continuing to attend the clinic at (including people who did not attend any further appoint-
these times, is given in Table 2. Several baseline variables ments and were assumed to have remained at their initial
were consistent predictors of attendance: people who were weight), and the 922 people who attended at least one
obese as children, who had hypertension or coronary artery follow-up visit. People who returned for at least one
disease, and who did not smoke were more likely to con- follow-up appointment achieved a mean maximum weight
tinue to attend for up to 3 years. Gender, age, blood pres- loss of 9.2% [95% CI 8.6, 9.7%]. Factors associated with
sure at the initial appointment and mental illness did not greater maximum weight loss were higher baseline BMI,
affect attrition. The models estimated the probability of systolic blood pressure and age, but effect sizes were small

© 2016 World Obesity. clinical

clinical obesity 6, 101–107
4104Review of of
Review weight clinic
weight 3-year
clinic outcomes
3-year outcomes P.P.Sumithran
Sumithranetetal.al. clinical obesity
obesity

100
80
Still in treatment (%)

60

248
da ys

50%
40
20
0

1 2 3
Figure 1 Percentage of patients still in
Years treatment over time.

Table 2 Estimated odds ratios [OR (95% CI)] for factors contributing to Of the 213 patients who attended the clinic for at least 3
probability of still attending the clinic at 1 and 3 years after the initial years, more than half (53%) had a final recorded weight at
appointment
least 5% below their initial weight, and 33% were main-
1 year 3 years taining a weight loss of 10% or more. Fifty (23%) of them
had been prescribed pharmacotherapy for appetite reduc-
Still attending (n; %) 466; 42.0% 213; 19.2% tion during the follow-up period. At their closest appoint-
Prediction rate (%) 75.9 83.7 ment date to 3 years, mean weight loss was significantly
Obese child 1.38 (1.03, 1.85)* 1.42 (1.00, 2.01)*
greater among people who were prescribed pharmaco-
Type 2 diabetes 1.42 (0.99, 2.05)
Body mass index 1.02 (1.00, 1.03)* therapy than those who were not (mean weight loss 7.7%
Diastolic blood pressure 1.01 (1.00, 1.02) vs. 2.3%, 95% CI for difference −3.9, −7.0%).
Asthma 1.45 (1.02, 2.07)*
Smoker 0.72 (0.49, 1.03) 0.53 (0.31, 0.87)*
English as first language 0.53 (0.28, 0.99)* Discussion
Hypertension 1.69 (1.26, 2.26)‡ 1.66 (1.13, 2.45)†
This retrospective review assessed 3-year outcomes after
Coronary artery disease 1.82 (1.13, 2.98)* 1.67 (1.00, 2.77)*
establishment of an obesity treatment clinic in a large
Prediction rate (%) refers to the strength of the model with respect to
metropolitan hospital using an evidence-based treatment
predicting whether or not a patient is still attending at the nominated programme comprising an initial VLED, structured rein-
time point. troduction of regular foods and continued follow-up with
*P ≤ 0.05. consideration of pharmacotherapy to prevent weight
†P ≤ 0.01.
regain. Patients who attended at least one follow-up
‡P < 0.001.
appointment achieved a mean maximum weight loss of
(e.g. for every 1 kg m−2 increment in initial BMI, maximum 9.2%, and those who continued to attend for 3 years
weight loss was estimated to increase by 0.09%). Con- maintained a mean loss of 6.4 kg (3.5%).
versely, smoking and a history of mental illness were asso- It is difficult to directly compare the weight outcomes
ciated with lower maximum weight loss. Gender, having from our clinic with others due to the heterogeneity of the
English as a first language, childhood onset of obesity and patient populations and treatment methods among reports.
a history of hypertension, coronary disease, asthma and Furthermore, there are few reports of outcomes beyond 1
diabetes were not associated with maximum weight loss year. In general, our patients have more severe obesity and
percentage. a greater prevalence of co-existing medical conditions, and

© 2016
© 2016 World World
Obesity. Obesity.
clinical clinical
obesity obesity
6, 101–107
clinical obesity Review
Review ofof weight
weight clinic
clinic 3-year
3-year outcomesP. P.
outcomes Sumithran
Sumithran et al.105
et al. 5

A B

0
50 All
At least 1 year
At least 3 years

−2
ITT

−4
Weight change (kg)

Weight change (kg)

0

−6
−8
−50

−10
−12
−100

1 2 3 1 2 3
Years Years

Figure 2 (A) Individual weight change (kg) over time (n = 1109). (B) Weight change modelled over time for all patients who started the
very-low-energy diet (solid line), those who attended at least once 1 (dashed line) and 3 years (dotted line) after their initial visit, and
intention-to-treat (ITT) analysis (dotted-dashed line) assuming return to baseline weight in people no longer attending the clinic.

Table 3 Estimated weight changes [mean

Still attending Intention-to-treat
(95% CI)] based on repeated measures
modelling of the raw data and
Weight change (kg) Weight change (%) Weight change (kg) Weight change (%)
intention-to-treat, assuming return to baseline
weight in people no longer attending
1 year −9.7 (−10.3, −9.2) −6.2 (−6.9, −5.7) −9.1 (−9.6, −8.6) −6.1 (−6.7, −5.5)
2 years −8.0 (−8.6, −7.3) −4.6 (−5.3, −3.9) −7.8 (−8.5, −7.2) −4.9 (−5.6, −4.2)
3 years −6.4 (−7.3, −5.6) −3.5 (−4.2, −2.8) −1.5 (−2.0, −1.0) −0.3 (−0.9, 0.2)

Table 4 Mean maximum % weight loss recorded during the 3 years for BMI 48.0 kg m−2) achieved a weight loss of 10.8% at 12
all 1109 patients, and for those (n = 922) who attended at least one months in the 57% of patients who completed the pro-
appointment after their initial visit [mean (95% CI)]
gramme (10). Rolland and colleagues reported outcomes of
All At least 2 visits another UK community-based programme (11), in which
5965 participants (mean BMI 36.3 kg m−2) undertook a
Max. % weight loss 7.6 (7.1, 8.1)‡ 9.2 (8.6, 9.7)‡ VLED for an average of 20 weeks, followed by a weight
Factors contributing to maximum % weight loss maintenance phase including weekly transactional cogni-
Age 0.07 (0.02, 0.12)† 0.06 (0.01, 0.11)*
tive behavioural therapy. In their analysis of completers,
Body mass index 0.08 (0.01, 0.14)* 0.09 (0.02, 0.16)*
Systolic blood pressure 0.03 (0, 0.07)* 0.03 (−0.01, 0.06) weight loss was 17.6% at 1 year and 12.9% at 3 years. The
Smoker −1.47 (−2.94, 0.01) −1.48 (−3.14, 0.17) weight losses achieved in these programmes are superior to
Mental illness −1.05 (−2.23, 0.13) −1.12 (−2.42, 0.18) our clinic results, which may partly be explained by the
exclusion from the community programmes of people with
Also included are the factors identified in the logistic regression model comorbidities which may confer greater difficulty in losing
as contributing to maximum % weight loss.
weight (such as psychiatric illnesses, and use of insulin or
*P ≤ 0.05.
†P ≤ 0.01.
sulfonylureas for diabetes), and the more intensive schedule
‡P < 0.001. of follow-up visits compared to our clinic.
Among outpatient-based reports, a retrospective analysis
the frequency of appointments at our clinic is less compared of a programme in 1887 patients (mean BMI 38.2 kg m−2)
with many others reported. A programme using a liquid using 12 weeks of liquid meal replacements (520 or
low-energy diet for 12 weeks (or 20 kg weight loss) fol- 850 kcal d−1) without pharmacotherapy, but with an inten-
lowed by structured re-introduction of food and optional sive follow-up period of weekly clinic visits, reported a
pharmacotherapy (orlistat) for 12 months in a primary care mean initial weight loss of approximately 15% at 12 weeks
setting in 91 people with severe obesity (mean age 46 years, and 10% at 48 weeks in patients who continued to attend

© 2016 World Obesity. clinical

clinical obesity 6, 101–107
6106Review of of
Review weight clinic
weight 3-year
clinic outcomes
3-year outcomes P.P.Sumithran
Sumithranetetal.al. clinical obesity
obesity

the clinic (12). A prospective multicentre German study in predictors of attrition (16). A retrospective review of pre-
8296 participants (mean BMI 40.8 kg m−2) which com- dictors of attendance at an outpatient obesity clinic in an
prised a 12-week meal replacement diet (800 kcal d−1) fol- Australian tertiary hospital with similar patient character-
lowed by structured re-feeding, weight stabilization and istics to ours found that almost half of patients referred did
weight maintenance phases, and also included weekly clinic not complete the questionnaire required to book their first
visits, reported an impressive 1-year weight loss of 17.9% appointment, and a further 17% did not attend their first
(13). However, in the subgroup of 301 participants with appointment (18). Fifty-five per cent of patients attending
longer-term follow-up, mean weight loss at 3 years was the clinic lost weight. Patients with diabetes were more
5.9 kg (4.2%). In a retrospective review of 51 older likely to return their questionnaire, and those with hyper-
outpatients with a high prevalence of obesity-related cholesterolaemia were more likely to lose weight (18). In a
comorbidities (mean age 65.2 years, BMI 38.5 kg m−2) pre- meta-analysis of predictors of attrition from 24
scribed a hypocaloric diet and a variety of pharmaco- randomized controlled trials of weight-loss medications,
therapy who continued follow-up for at least 6 months Fabricatore and colleagues (19) found that attrition at 1
(mean duration 39 months), initial weight loss was 5.7 kg year was greater in participants who received placebo com-
(approximately 6%) at 6 months, and at their last recorded pared with those treated with either orlistat or sibutramine
evaluation, mean weight loss was 6.65 kg, 67% of patients (34.9% vs. 28.6% and 28.3%, respectively), and in groups
were maintaining a loss of 5% and 33% a loss of 10% of which did not undergo a pre-randomization lead-in period
their initial weight (14). A similar result was reported in a than in those that did (39.9% vs. 29.1%, P < 0.01). In
long-term telephone follow-up of a 26-week programme univariable analyses, the presence of a weight-related
comprising an initial 12-week VLED, structured re-feeding comorbidity was related to lower total and non-adverse
and maintenance and weekly clinic appointments, in which event-related attrition, but this did not remain a significant
loss of ≥5% weight was maintained by 53% and ≥10% of factor in the final multivariable model. In keeping with our
weight by 35% of participants (15). These long-term findings, age was not related, and there was limited evi-
results are very similar to the weight loss achieved by dence for BMI as a predictor of attrition (19). Inelmen and
patients still attending our clinic at 3 years and indicate that colleagues reported that patients who completed a
although some weight regain is common after 6–12 12-month lifestyle intervention had a greater number of
months, clinically relevant weight loss in the longer term is obesity-related diseases than those who dropped out (17).
possible with a medical treatment programme if strategies Therefore, we and others (16–19) have found that the
are used to assist with maintenance of weight loss. presence of various physical conditions related to excess
Attrition from long-term obesity management pro- weight is predictive of continued attendance. Possible
grammes is notoriously high and is expected to be higher explanations for this are a greater perceived urgency for
from real-world outpatient clinics compared with research weight loss, stronger support from family members and
studies, in which methods to reduce drop-out are often healthcare providers, and greater concern about the health
employed, such as run-in periods in which non-adherent implications of obesity, which may increase tolerance of
patients are screened out, and offering patients financial treatment-related adverse effects and inconveniences (19).
and other incentives to encourage retention. Our rate of Retrospective chart reviews have inherent limitations.
attrition at 12 months is in keeping with that reported in Patients were considered to have co-existing medical con-
other non-surgical weight loss clinics (16,17). Gill and col- ditions (hypertension, dyslipidaemia, type 2 diabetes,
leagues retrospectively reviewed attrition from a multidis- mental illness) if they were taking medications for them or
ciplinary clinic in Canada, including 887 participants in a if the condition was noted in the medical history. Informa-
medical weight loss programme and 318 surgically treated tion regarding current symptoms, changes in medications
patients (16). Comparable to ours, this was a publicly or biochemical parameters was not systematically col-
funded urban clinic, with a predominance of female lected. Although we found that mental illness was not
patients, mean age of 43 years, initial BMI 50.6 kg m−2 and associated with attrition, we did not have information to
a considerable proportion of patients (46%) with a mental distinguish between different types of illness, or the pres-
illness. However, in contrast to our clinic, a rigorous lead-in ence of active symptoms. Others have reported mixed find-
period must be completed in order to receive an appoint- ings regarding the association between symptoms of
ment, and if an appointment is missed, an intensive effort depression and attrition (17,20). It was also not possible to
to contact the patient is made over 12 months before they determine the reasons for drop-out from the clinic. The
are considered lost to follow-up (16). Attrition at 12 fixed timing of the clinic (on a weekday morning) is likely
months was 53.9% in the medical clinic and 11.9% from to be a barrier for people who have conflicting commit-
the surgical clinic. Participation in the medical (vs. surgical) ments which are not flexible (17). Furthermore, the high
clinic, lower BMI, lack of co-existing musculoskeletal rate of refusal to commence the VLED programme (38%)
disease and presence of a mental illness were identified as and early discontinuation (17% who agreed to it but failed

© 2016
© 2016 World World
Obesity. Obesity.
clinical clinical
obesity obesity
6, 101–107
clinical obesity Review
Review ofof weight
weight clinic
clinic 3-year
3-year outcomesP. P.
outcomes Sumithran
Sumithran et al.107
et al. 7

to return after their initial appointment) suggests a discrep- 3. Anderson JW, Konz EC, Frederich RC, Wood CL. Long-term
ancy between patients’ expectations and the programme weight-loss maintenance: a meta-analysis of US studies. Am J Clin
Nutr 2001; 74: 579–584.
offered. In order to address this, newly referred patients are
4. Stunkard AJ, Foch TT, Hrubec Z. A twin study of human
now being sent information about the programme and obesity. JAMA 1986; 256: 51–54.
invited to make an appointment if they wish to proceed. 5. Gluckman PD, Hanson MA. Developmental and epigenetic
Among our patients, mean weight loss at 3 years was pathways to obesity: an evolutionary-developmental perspective.
significantly greater in those who had been prescribed phar- Int J Obes (Lond) 2008; 32(Suppl. 7): S62–S71.
6. Woods SC, D’Alessio DA. Central control of body weight and
macotherapy. Weight loss achieved without bariatric
appetite. J Clin Endocrinol Metab 2008; 93(Suppl. 1): S37–S50.
surgery is accompanied by adaptive changes in energy 7. Sumithran P, Prendergast LA, Delbridge E et al. Long-term
expenditure and several hormones involved in appetite persistence of hormonal adaptations to weight loss. N Engl J Med
regulation, which collectively facilitate weight regain (21). 2011; 365: 1597–1604.
During the period under review, most patients were pre- 8. Rosenbaum M, Hirsch J, Gallagher DA, Leibel RL. Long-term
persistence of adaptive thermogenesis in subjects who have main-
scribed either sibutramine, or combination phentermine
tained a reduced body weight. Am J Clin Nutr 2008; 88: 906–912.
and topiramate for appetite reduction. In Australia, the 9. Purcell K, Sumithran P, Prendergast LA et al. The effect of rate
only medication currently approved for long-term obesity of weight loss on long-term weight management: a randomised
treatment is orlistat, which is superior to placebo in reduc- controlled trial. Lancet Diabetes Endocrinol 2014; 2: 954–962.
ing weight regain in clinical studies, but which does not 10. Lean ME, Brosnahan N, McLoone P et al. Feasibility and
indicative results from a 12-month low-energy liquid diet treat-
attenuate the increased appetite which follows diet-induced
ment and maintenance programme for severe obesity. Br J Gen
weight loss (22). Since 2012, four new or combination Pract 2013; 63: e115–e124.
medications that reduce appetite have been approved by 11. Rolland C, Johnston KL, Lula S, Macdonald I, Broom J.
the U.S. Food and Drug Administration for the long-term Long-term weight loss maintenance and management following a
management of obesity – controlled-release phentermine- VLCD: a 3-year outcome. Int J Clin Pract 2014; 68: 379–387.
12. Bailey BW, Jacobsen DJ, Donnelly JE. Weight loss and main-
topiramate, lorcaserin, naltrexone-bupropion and
tenance outcomes using moderate and severe caloric restriction in
liraglutide (23). When available, pharmacotherapy with an outpatient setting. Dis Manag 2008; 11: 176–180.
proven long-term safety and efficacy would be a useful 13. Bischoff SC, Damms-Machado A, Betz C et al. Multicenter
adjunct to lifestyle modification for people who are other- evaluation of an interdisciplinary 52-week weight loss program for
wise unable to maintain weight loss. obesity with regard to body weight, comorbidities and quality of
life – a prospective study. Int J Obes (Lond) 2012; 36: 614–624.
14. Horie NC, Cercato C, Mancini MC, Halpern A. Long-term
Conflicts of interest statement pharmacotherapy for obesity in elderly patients: a retrospective
evaluation of medical records from a specialized obesity outpatient
JP is chairman of the medical advisory board for liraglutide
clinic. Drugs Aging 2010; 27: 497–506.
3 mg and has received payment for consultancy from Astra 15. Wadden TA, Frey DL. A multicenter evaluation of a propri-
Zeneca and Eli Lilly, and lectures from iNova pharmaceu- etary weight loss program for the treatment of marked obesity: a
ticals unrelated to this paper. PS has received payment for five-year follow-up. Int J Eat Disord 1997; 22: 203–212.
lectures from Nestlé and Novo Nordisk unrelated to this 16. Gill RS, Karmali S, Hadi G et al. Predictors of attrition in a
multidisciplinary adult weight management clinic. Can J Surg
paper. The other authors have no conflicts of interest rel-
2012; 55: 239–243.
evant to this paper. 17. Inelmen EM, Toffanello ED, Enzi G et al. Predictors of drop-
out in overweight and obese outpatients. Int J Obes (Lond) 2005;
Acknowledgements 29: 122–128.
18. Brook E, Cohen L, Hakendorf P, Wittert G, Thompson C.
We thank Abang M.N.A. Samatan for his work on an early Predictors of attendance at an obesity clinic and subsequent weight
version of this review, and Rodney Camba and Nick change. BMC Health Serv Res 2014; 14: 78.
Jandric for IT assistance. Author contributions are as 19. Fabricatore AN, Wadden TA, Moore RH et al. Predictors of
attrition and weight loss success: results from a randomized con-
follows: PS, CJH, CAH and JP collected the data; LAP and trolled trial. Behav Res Ther 2009; 47: 685–691.
PS analysed the data; PS, LAP and JP wrote the manuscript. 20. Mazzeschi C, Pazzagli C, Buratta L et al. Mutual interactions
between depression/quality of life and adherence to a multidisci-
References plinary lifestyle intervention in obesity. J Clin Endocrinol Metab
2012; 97: E2261–E2265.
1. Australian Bureau of Statistics, 2012. Australian health 21. Sumithran P, Proietto J. The defence of body weight: a physio-
survey: first results, 2011–12, cat. no. 4364.0.55.001. URL logical basis for weight regain after weight loss. Clin Sci 2013; 124:
http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/4364.0.55.001 231–241.
Chapter1002011-12 (accessed September 2015). 22. Sumithran P, Proietto J. Benefit-risk assessment of orlistat in
2. National Health and Medical Research Council. Clinical Prac- the treatment of obesity. Drug Saf 2014; 37: 597–608.
tice Guidelines for the Management of Overweight and Obesity in 23. Apovian CM, Aronne LJ, Bessesen DH et al. Pharmacological
Adults, Adolescents and Children in Australia. National Health management of obesity: an Endocrine Society clinical practice
and Medical Research Council: Melbourne, 2013. guideline. J Clin Endocrinol Metab 2015; 100: 342–362.

© 2016 World Obesity. clinical

clinical obesity 6, 101–107