Confidential: For Review Only - bmj.com · Mean weight loss in the TRD groups was -10.7 kg, and...

Confidential: For Review Only

Doctor Referral of Overweight People to Low Energy total

diet replacement Treatment (DROPLET): a pragmatic randomised controlled trial

Journal: BMJ

Manuscript ID BMJ.2018.044188

Article Type: Research

BMJ Journal: BMJ

Date Submitted by the Author: 19-Mar-2018

Complete List of Authors: ASTBURY, Nerys; University of Oxford, Nuffield Department of Primary

Care Health Sciences

Aveyard, Paul; University of Oxford, Nuffield Department of Primary Care

Health Sciences

Nickless, Alecia; University of Oxford, Nuffield Department of Primary Care

Health Sciences

Hood, Kathryn; University of Oxford, Nuffield Department of Primary Care

Corfield, Kate; University of Oxford, Nuffield Department of Primary Care

Health Sciences

Lowe, Rebecca; University of Oxford, Nuffield Department of Primary Care

Health Sciences

Jebb, Susan; Univerof Oxford, Primary Care Health Sciences

Keywords: Obesity, Primary Care

https://mc.manuscriptcentral.com/bmj

BMJ

Confidential: For Review OnlyDoctor Referral of Overweight People to Low Energy total diet replacement Treatment (DROPLET):

a pragmatic randomised controlled trial

Nerys M Astbury PhD 1, Prof Paul Aveyard PhD

1, Alecia Nickless MSc

1, Kathryn Hood

1, Kate

Corfield1, Rebecca Lowe

1, Prof Susan A Jebb PhD

1

1 Nuffield Department of Primary Care Health Sciences, University of Oxford, UK

Word Count : 3859

Running title: Total diet replacement programme for weight loss

Keywords: Obesity, total diet replacement programme, randomised controlled trial

Page 1 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


2

Abstract

Objective

The aim of this study was to test the effectiveness and safety of a total diet replacement programme

(TDR) for routine treatment of obesity in a primary care setting.

Design

Pragmatic, two-arm, parallel-group, open-label, individually randomised controlled trial

Setting

Ten primary care practices in Oxfordshire, UK

Participants

278 adult participants who were obese and seeking support to lose weight were recruited from

primary care registers and 73% were re-measured at 12 months. We excluded people already

following weight loss programmes or after bariatric surgery or who had contraindications to the

meal replacements. Participants were allocated to the total diet replacement programme (TDR) or

usual care (UC) groups using an independently generated randomisation sequence using randomly

permuted block sizes, stratified by general practice and BMI (more or less than 35kg/m2) with

concealed allocation.

Interventions

The TDR programme comprised weekly behavioural support for 12 weeks and monthly support for

three months with meal replacement products providing 810kcal/day as the sole food during the

first eight weeks followed by food re-introduction. Usual care comprised behavioural support for

weight loss from a practice nurse and a diet programme with modest energy restriction.

Main outcome measures

The primary outcome was weight change at 12 months analysed as intention-to-treat with mixed

effects models. Secondary outcomes included biomarkers of cardiovascular risk. We recorded any

adverse events.

Results

Participants in the TDR group lost more weight (-10.7kg) than usual care (-3.1kg); adjusted mean

difference of -7.2 kg (95% CI: -9.4, -4.9). Weight losses ≥10% occurred in 45% and 15% of participants

in TDR and UC groups respectively. There were greater improvements in biomarkers of

cardiovascular and metabolic risk in the TDR programme than in usual care. Adverse events classed

as moderate or greater occurred in 11% and 12% of participants in TDR and UC respectively.

Conclusions

Compared with regular weight loss support from a practice nurse, a programme of weekly

behavioural support and total food replacement providing 810kcal per day appears safe and

Page 2 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


3

tolerable, and leads to substantially greater weight loss and greater improvements in the risk of

cardiometabolic disease. It is suitable for routine use for the treatment of obesity in generalist

healthcare settings.

Trial registration: ISRCTN75092026

Page 3 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


4

What we already know on this topic

A systematic review of trials recommending energy intakes of 800kcal/d or less showed weight loss

at 1 year was -10.3 kg, a difference of −3.9 kg (95% CI −6.7 to −1.1) greater than the comparator

behavioural support programmes, but all were conducted in specialist obesity clinics or research

centres and none were conducted in routine primary care.

Recent results from the DiRECT trial among people with Type 2 diabetes showed that a total diet

replacement programme provided by healthcare professionals led to similar weight losses and 45%

of patients were in remission after 1 year1.

What this study adds

This study shows that similar weight losses can be achieved in a generalist primary care setting for a

people who are obese, through referral to a commercial provider offering a total diet replacement

programme.

Mean weight loss in the TRD groups was -10.7 kg, and -7.2 kg (95% CI: -9.4, -4.9) greater than usual

care. 45% of people achieved a weight loss of >10% at 1 year.

This model of primary care referral offers the potential for rapid roll out at scale across the NHS to

prevent and treat obesity-related disease.

Page 4 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


5

Introduction

Excess adiposity is one of the main risk factors for preventable morbidity and mortality.2 Weight loss

ameliorates the risk, improves functioning and wellbeing, 2 with growing evidence that benefits can

persist even if weight is regained.3 4

Primary care physicians have the opportunity to offer

treatments for obesity at the scale required to have a discernible effect on the prevalence of obesity

and related diseases5 and are encouraged to screen patients and offer support to lose weight.

6

Despite this, physicians rarely offer support to lose weight.7

There is good evidence that weight loss programmes provided in community groups by commercial

providers are more effective than comparator interventions delivered by primary care clinicians. 8-10

Moreover, such programmes are cost-effective and, over the long-term, cost-saving.9

Notwithstanding the population benefits, individuals who are referred lose only an additional 2kg

compared with self-help interventions and would benefit from interventions that lead to greater

weight loss.11

One option is a total diet replacement (TDR) programme, combining a low energy diet

with behavioural support. In a systematic review of trials we found that very low energy diets (VLED)

providing <800kcal/day led to significantly greater weight loss than behavioural weight management

programmes based on usual foods.12

Today, TDR programmes offer up to 1200 kcal/day, but

similarly use meal replacement products as the sole source of nutrition alongside a behavioural

support programme. Most trials of these programmes have been small scale, conducted in research

settings or specialist obesity clinics. There is a common perception that such programmes are

unacceptable to most people, possibly unsafe and lead to rapid weight regain, and guidelines do not

recommend their use for general treatment of obesity.13 14

The aim of this randomised controlled trial was to investigate the effectiveness and safety of primary

care physician referral to a commercially-provided low-energy total diet replacement programme

compared with usual care. Consistent with the pragmatic nature of the trial we did not attempt to

match treatment intensity in the two groups, rather we compared the Cambridge Weight Plan

programme, comprising specially formulated products and behavioural support, with the usual type

of weight management programmes offered by primary care staff, based on dietary advice and

behavioural support.

Methods

This trial was a pragmatic, individually randomised, two arm, open-label, parallel design allocating

participants to a TDR programme or routine support from a practice nurse.15

Ethical approval was

obtained from South Central Oxford B Research Ethics Committee (Ref: 15/SC/0337) and all

Page 5 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


6

participants gave written informed consent. The protocol was amended after registration to reduce

the number of secondary outcomes on the advice of the independent trial steering committee.

Participants and settings

Participants were recruited from primary care practices in Oxfordshire, UK who were willing and able

to offer a weight management programme within the practice. GPs searched their electronic health

records for adult patients with a BMI of at least 30kg/m2 and whose health would benefit with

weight loss and invited them by letter to participate. We excluded people who had received or were

scheduled for bariatric surgery, people participating in a weight management programme, or with

contraindications to the TDR according to the protocol.15

(Full inclusion and exclusion criteria are

presented in Supplementary Appendix) Following telephone screening, eligible participants

scheduled an appointment with a nurse at their local practice.

Randomisation and masking

An independent statistician generated a random computer-generated randomisation list with 1:1

allocation using stratified block randomisation with randomly permuted block sizes of 2, 4 and 6,

stratified by general practice and BMI (more or less than 35kg/m2). After participants’ eligibility was

confirmed by the nurse, they were enrolled in the study and the allocation was revealed using an

online randomisation programme, which ensures full allocation concealment. Due to the nature of

the intervention it was not possible to blind participants, clinicians or some of the researchers

working in the field to treatment allocation.

Interventions

The TDR programme was provided by Cambridge Weight Plan UK Ltd who manage a network of

counsellors providing behavioural support and food products.15

Participants were asked to contact a

local counsellor who was aware of the research study and the protocol for the provision of meal

replacement products, but who had not received any additional training to deliver the behavioural

support programme. For the first 12 weeks participants met with the counsellor weekly for support

which comprised goal setting and feedback, encouragement, reassurance and problem-solving.

Participants replaced all food with four meal replacement products daily (soups, shakes and bars),

750ml of skimmed milk, 2.25l of water or other very low/no energy drinks and a fibre supplement,

comprising 810kcal/day (3389kJ/d)(Supplementary Table S1). After eight weeks, there was a four-

week stepwise reduction in the use of the meal replacement products and re-introduction of

conventional food-based meals. During the weight maintenance phase from week 13 to 24

counsellors encouraged participants to attend monthly appointments and to consume one meal

replacement a day, with the remainder of the diet provided by food. If weight was regained,

Page 6 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


7

participants were recommended to return to the total diet replacement stage for periods of up to

four weeks. This programme was free of charge to week 24. Participants could choose to pay for

continued support thereafter.

Clinicians adjusted medication for hypertension and diabetes at the start of the programme at a

scheduled assessment at one month and as needed thereafter. We supplied guidelines for this to

clinicians (Supplementary Appendix).

For the comparator, participants followed each practice’s usual weight management protocol.15

Participants also received a copy of the 47-page booklet “ So you want to lose weight … for good”16

which includes information on goal setting, monitoring and feedback and advice about food types,

portion control and physical activity. They were not prohibited from undertaking any other weight

management intervention but none were offered by their healthcare provider.

Procedures

We measured height at baseline only, blood samples at baseline and 12 months and all other

measurements at baseline, three, six, and 12 months. Weight and body fat were measured on a

digital scale (TANITA SC-240). We measured waist circumference at the top of the iliac crest and

seated blood pressure in triplicate using an automated blood pressure monitor with the mean of the

last two measurements recorded. We recorded quality of life using the EQ5D and OWL-QOL.17 18

Fasting blood samples were collected to measure blood glucose, insulin, triglycerides, and

cholesterol fractions. Measurements were made by practice personnel at baseline and by the

research team at 3, 6 and 12 months.

Outcomes

The primary outcome was change in body weight from baseline to 12 months. The secondary

outcomes were: change from baseline in body weight at 3 and 6 months, the proportion of

participants achieving 5% and 10% weight loss at 12 months and change in fat mass, LDL cholesterol,

HbA1c, systolic and diastolic blood pressure at 12 months.

Pre-specified exploratory outcomes were change in fat mass and blood pressure at 3 and 6 months

and in waist circumference at 3, 6 and 12 months. The change from baseline to 12 months in fasting

triglycerides, HDL cholesterol, glucose, insulin was measured. We derived HOMA-IR, HOMA-%β and

HOMA-%S and calculated the change in 10-year cardiovascular risk using QRISK2.19 20

We measured

the change in self-reported quality of life using the EQ-5D18

and OWL-QOL17

between baseline and

12 months.

Page 7 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


8

We recorded all adverse events during the first three months of the programme and at six months

for gallstone-related events only, to allow for diagnostic delay. These were coded using MedDRA

version 18.1 and presented at the system organ class and preferred term level. We present events

that occurred in at least 2% of the participants.

Statistical Analysis

We determined that a difference between groups of 4kg was clinically relevant. Using data from

published studies on standard deviation and assuming 90% power with two-sided significance of 5%,

and 20% loss to follow-up, we needed a sample of 270 people. Accounting for multiple testing of

secondary outcomes, this gave 90% power to detect a standardized difference of 0.56 with 5%

significance for the secondary outcomes.

We followed a statistical analysis plan approved by the independent trial steering committee prior to

database lock. The primary, secondary and exploratory outcomes were analysed using an intention-

to-treat analysis using PROC MIXED in SAS Version 9.4 by an independent trial statistician. We used

linear mixed effects models with unstructured correlation matrix for repeated measures and

adjusted for baseline stratification variables, with practice set as a random effect. For binary

outcomes, we used analogous logistic models. Prior to analysis of outcomes, we assessed

association between baseline variables and loss to follow-up at 12 months. Age and gender were

associated with loss to follow-up and so were included as covariates as planned.

We assessed the sensitivity of the results to missing data using different imputation methods;

baseline and last observation carried forward, completers only, multiple imputation, and a pattern

mixture model assuming different degrees of missing not at random. We performed pre-specified

exploratory subgroup analyses to assess whether treatment effects differed by age, gender, BMI,

socio-economic status (based on the postcode of the participant) and practice. Following our

statistical plan, we did not compare quality of life using inferential statistics.

Patient involvement

Our extensive public involvement activities have shown that a large proportion of people are

interested to know whether weight-loss interventions are effective and welcome this kind of

research. In a previous trial involving opportunistic offers of support for weight loss patients

overwhelmingly reported that this was appropriate and helpful.6 Members of the public who have

expressed an interest in our research were invited to comment on the design of the study and the

patient-facing materials prior to ethical submission. Two lay people were members of the trial

Page 8 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


9

steering committee. Participants in the trial were offered the opportunity to hear the results of the

study upon completion and a lay summary and infographic have been provided.

Role of the funding source

This was an investigator-initiated study. The funders of the study had no role in the design of the

protocol, data collection, analysis or interpretation of the results. The study was conducted in

collaboration with an NIHR accredited Clinical Trials Unit. All authors had access to all data after the

statistical analysis was complete. The funders had no access to the data. SAJ had final responsibility

for submission for publication. The investigators have no personal financial relationships with the

funder.

Results

Participants were recruited between 12th

January 2016 and 28th

July 2016. We screened 283

participants from 10 practices and 278 were eligible and randomly allocated to either a total diet

replacement programme or usual care. Follow-up was completed on 4th

August 2017.

Most participants were middle aged, 61% were women, and 88% were white British. The average

BMI was 37.2 kg/m2. On enrolment 23% had a diagnosis of hypertension and 15% had diabetes

(Table 1).

We randomised 138 participants to the TDR group and 140 to UC. Four and two participants

withdrew consent for their data to be used after randomization and we followed-up 104 (78%) and

95 (69%) participants in the TDR and UC group respectively at 12 months (Figure 1).

Primary outcome

Weight loss was greater in the TDR group at all time points. At three months, the difference was -

9.6kg (-11.0; -8.2), p<0.0001 and at six months it was -9.6kg (-11.6; -7.7), p<0.0001. The difference

between groups reduced thereafter, with a difference in mean weight loss at 12 months, the

primary outcome, of -7.2kg, (-9.4; -4.9), p<0.0001. At this point, the TDR group had lost 10.7kg

(SD=9.6) and the control group 3.1kg (7.0) (Figure 2). This conclusion was unchanged by sensitivity

analyses for loss to follow-up (Table S2). In the pattern mixture modelling, even assuming extreme

bias in loss to follow-up in either the TDR or UC group, there was a treatment difference of more

than 5kg in favour of TDR. There was no evidence that the intervention effect on weight differed by

gender (p=0.22), age group (p=0.85), socioeconomic status (p=0.65), BMI (p=0.094), diabetes status

(p= 0.42), or practice (p=0.22) (Figure S2).

Page 9 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


10

Secondary outcomes

Secondary outcomes included change in fat mass (incidentally measured at the same time as body

weight) which showed similar, albeit slightly smaller differences in favour of the TDR group. The

exploratory outcome, change in waist circumference, showed a similar pattern (Table 2). At 12

months 73% and 32% of participants in the TDR and UC group respectively had lost ≥5% of their

baseline weight, an odds ratio of 6.5 (3.4; 12.2), p<0.0001. Forty-five percent and 15% participants

lost ≥10% baseline weight in the TDR and UC groups respectively; odds ratio 4.9 (2.4; 9.9), p<0.0001

(Figure S2). The number needed to treat to observe these benefits was 2.4 (1.8; 3.5) and 3.3 (2.5;

3.5) for ≥5% and ≥10% weight loss respectively.

At 12 months, participants in the TDR group had greater reductions in HbA1c (-2.2 mmol/mol [-4.4;

0.0]) and diastolic blood pressure (-3.1 mmHg [-5.5; -0.7]), but there was no significant difference in

the reduction in systolic blood pressure (-2.9 mmHg [-6.4; 0.6]) or LDL-cholesterol (0.0 mmol/l [-0.2;

0.2]). The exploratory outcomes assessed changes in glucose regulation, blood pressure and other

lipid fractions. These reflected the secondary outcomes, showing marked improvements in the TDR

group in glucose regulation, modestly greater reductions in blood pressure, a slight improvement in

triglycerides, but no difference in cholesterol fractions (Table 2). Thus, the TDR reduced overall

cardiovascular risk to a greater extent than UC (Table 2).

Summary statistics are presented for exploratory analyses of the effect of treatment on quality of

life. The EQ-5D-5L index, VAS scores and OWLQOL score all showed greater improvements from

baseline at 6 months and 12 months in the TDR group than in the UC group (Table 2).

Adverse events

Overall, adverse events were frequent and mild in both groups. Sixty-nine (51.5%) participants in

the TDR group experienced at least one AE and 41 (29.7%) did so in the UC group (Fisher’s Exact

Test: p = 0.0003) (Table S3), representing an additional one in five AEs in the TDR group. The most

frequent AEs where there was a greater incidence in TDR than UC were constipation, fatigue,

headache, and dizziness affecting approximately one in 7, one in 12 ,one in 17 and one in 22

respectively (Table 4). Most of these AEs were mild, with only 15 (11.2%) in the TDR and 17 (12.3%)

in the UC groups classed as moderate or severe, meaning they interfered with normal functioning

(Fisher’s Exact Test: p=0.82) (Table S2). There was one serious adverse event which occurred after

randomization but before the participant initiated the TDR programme.

Page 10 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


11

Discussion

Main findings

Primary care referrals of people who are obese to treatment within a total diet replacement

programme in the community reduced weight by 7.2 kg (-9.4; -4.9) more than usual care at one year,

with significantly greater improvements in glucose control, diastolic blood pressure and triglycerides

but not other lipid fractions. Among patients randomised to receive the total diet replacement

programme 73% and 45% lost ≥5 or ≥10% of their baseline weight respectively, compared to 32%

and 15% who received usual care.

Strengths and limitations

The strengths of this trial include recruiting patients who are typical of those seen in primary care

and who were seeking support for weight loss. There was no special training for clinicians to deliver

the programme other than written guidance since the intervention was delivered by a commercial

provider and weight loss did not vary by age, gender or SES. Together these factors mean this

programme could be readily implemented and realize benefits across the population. Loss to follow-

up was slightly lower than most weight loss trials9 and the observed difference in weight between

treatment groups was greater than 5 kg in all sensitivity analyses. It is however limited by the

relatively short duration of follow-up and the absence of direct evidence on the incidence of weight-

related disease or the cost-effectiveness of the intervention. The meal replacement products are

designed to be nutritionally complete, but we do not have data on nutrient intake during the

programme or following food reintroduction, nor did we measure physical activity. Although the

population we sampled was heterogeneous with respect to socioeconomic status they were not as

deprived as the UK population as a whole. The proportion of participants from non-white ethnic

groups, whilst representative of the population, was too small to allow any meaningful subgroup

analyses. Although the subgroup analysis showed no evidence that more deprived people received

less benefit, there is only weak evidence to suggest this effect applies to this group.

Comparisons with other studies

The TDR programme comprised 12 weekly support sessions followed by a further three monthly

sessions together with the use of food replacement products and it is likely that both the support

and food replacement were important. In the meta-analysis of VLED programmes, a very low energy

(<800kcal/day) diet proved less effective than behavioural support that aimed for modest energy

restriction, whereas programmes incorporating behavioural support alongside a very low energy diet

were more effective than support programmes aiming at modest energy restriction. 12

Currently

behavioural support for weight loss in primary care is constrained by limited interest in the topic, the

Page 11 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


12

lack of a defined programme, and competing demands and few patients receive support for weight

loss.7 In this pragmatic trial the nurse programme was planned to comprise about four sessions of

support over 12 weeks. This is likely to be more than the ‘usual care’ received by most people who

are obese in primary care, however it provided considerably less input than the TDR programme.

However, the level of behavioural support in the TDR group is comparable to that provided in

intensive lifestyle intervention studies such as the US Diabetes Prevention Program which offered 16

sessions in the first 24 weeks. The 10.7 kg weight loss seen with TDR is greater than the 7 kg weight

loss observed in the US Diabetes Prevention Program. Thus it appears that the weight loss observed

in the TDR programme reflects both the TDR diet component and the support provided, and it is

likely that either alone would be less effective than the package together.

Participants in this trial had a BMI of at least 30kg/m2 and many were at risk of weight-related

morbidity, though at baseline only 15% had diabetes and 23% hypertension. There was evidence

that the intervention enhanced blood glucose control across a number of measures and a

significantly greater improvement in triglycerides though not in cholesterol fractions. There were

greater improvements in systolic and diastolic blood pressure, though the former were not

significant beyond three months. However, we assessed blood pressure on one occasion only and

therefore this was subject to measurement error. Thus regression to the mean is likely to have

underestimated the benefit to mean blood pressure. Overall the changes we observed were

consistent with the magnitude of weight loss achieved 21

and comparable to those observed in the

DiRECT trial.1 Data from other weight loss studies would suggest that this is likely to reduce the

incidence of diabetes and cardiovascular disease and prevent premature mortality, commensurate

with the greater reduction in 10-year cardiovascular risk seen in the TDR group.4 22

Changes in quality

of life suggested greater improvements in the TDR group than usual care, consistent with evidence

from a systematic review that greater weight loss produces improvements in quality of life.23

In the

longer term there may be additional QALY gains as a result of a lower incidence of disease.

Anecdotal reports suggest that many people fear TDR programmes with severe energy restriction

will be unpopular and intolerable, but our results suggest otherwise. We used a similar recruitment

method to other trials and achieved a very similar take-up,9 suggesting that people did not

discriminate against this type of programme. Recruitment rates in this general sample of people who

were obese were 17%, which is similar to the DiRECT trial, which specifically recruited people with

type 2 diabetes.1 Adverse events were more common in the TDR group but almost all mild and there

was no significant difference in events of moderate or greater severity. Clinicians who were initially

unfamiliar with these programmes were able to adjust medication appropriately. Together these

data suggest that TDR programmes would not only be effective, but acceptable and well-tolerated if

Page 12 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


13

offered broadly. The system of referring patients to a commercial provider used in this study

achieved similar weight losses as seen in the DiRECT trial 1 where the behavioural support

component of the programme was provided by a healthcare professional. Referral to a commercial

provider, is already used for other types of weight loss support, and offers the potential to reduce,

rather than increase the workload for healthcare professionals.

Implications of this research

Current clinical guidelines recommend that this type of diet is reserved for patients in whom the

priority is short-term weight loss, for example, prior to bariatric or knee replacement surgery and

they are not recommended as routine weight loss interventions.13 14

This presumably reflects

concerns that weight loss is short-lived. This trial shows this treatment leads to greater weight loss

at one year than interventions based on usual food, nine months after the total diet replacement

phase of treatment. Although, on average, some weight was regained after the programme end at

24 weeks until the final follow up, this also occurred in the usual care group, and is common to all

weight-loss programmes.24

A previous experimental study in which patients were supported to

achieve 15% weight loss using either a very low energy diet or more moderate energy restriction

over a longer duration, observed similar rates of weight regain in the two groups after programme

end.25

Since the adverse health consequences of obesity relate both to the duration and magnitude

of excess weight this implies that the greater initial weight loss achieved with total diet replacement

programmes will be associated with greater improvements in long-term health outcomes.

The NHS does not routinely offer this type of programme and many primary care physicians are wary

about supporting patients who choose to use a total diet replacement programme because they are

unfamiliar with this approach or because of concerns about the safety of such interventions. This

trial should provide considerable reassurance. We included detailed elicitation of adverse events and

there were no unexpected and related adverse events during the 12 weeks of total diet replacement

and no cases of cholecystitis during an extended reporting period to 24 weeks. Moreover, there was

no excess of adverse events that interfered with participants’ ability to live their lives as normal,

showing that this approach was well-tolerated. GPs were provided with guidance to reduce, or

cease, medications for patients taking oral hypoglycaemic agents or treatments for hypertension at

the start of the diet and to monitor at 4 weeks. Weight loss at 4 weeks is a strong predictor of long

term success and provides an opportunity to adjust medication based on early weight change and

this approach appears to have been managed without giving rise to adverse events.26

In conclusion, a total diet replacement programme combining nutritionally complete meal

replacement products with behavioural support appears to be acceptable, well-tolerated and leads

Page 13 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


14

to greater weight loss with larger improvements in cardiovascular risk than currently available

weight loss programmes offered in primary care.

Page 14 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


15

Figure 1: Consort flowchart

286 Participants screened for eligibility

4 excluded:

4 BMI < 30kg/m2

4 declined to participate

278 were randomised

138 assigned to TDR programme 140 assigned to Usual Care

125 had weight measured at 1 month 114 had weight measured at 1 month

114 had weight measured at 3 months 97 had weight measured at 3 months

108 had weight measured at 6 months 94 had weight measured at 6 months

104 had weight measured at 12 months

17 Withdrawals

17 Loss to follow-up

4 Excluded from analysis

(Withdrew consent for use of data)

134 Were included in primary outcome

analysis

95 had weight measured at 12 months

25 Withdrawals

20 Loss to follow-up

2 Excluded from analysis

(Withdrew consent for use of data)

138 Were included in primary outcome

analysis

Page 15 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


16

Figure 2: Estimated weight change over 12 months in the intention to treat population

Page 16 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


17

Table 1: Baseline characteristics *

N Usual Care

Total Diet

Replacement

Age (yrs) 272 47.4 ± 12.8 48.2 ±11.5

Gender n (%)

Female 165 84 (60.9) 81 (60.5)

Male 107 54 (39.1) 53 (39.6)

Ethnicity

White British 240 119 (86.2) 121 (90.3)

Not White British 32 19 (13.8) 13 (9.7)

IMD decile § 272 7.3 ± 2.0 7.6 ± 2.0

Weight (kg) 272 105.2 ± 20 107.9 ± 18.9

Height (cm) 272 168.7 ± 9.7 169.2 ± 9.5

BMI (kg/m2) 272 36.8 ± 5.1 37.6 ± 5.7

Waist circumference (cm) 270 115.0 ± 12.5 116.4 ±13.5

Body fat (%) 268 42.1 ± 7.7 43.0 ± 7.8

Blood pressure (mmHg)

Systolic 269 130.1 ± 15.8 130.6 ± 16.4

Diastolic 269 81.3 ± 9.9 83.1 ± 9.7

Medical conditions n (%)

Diabetes 272 20 (14.5) 21 (15.7)

Hypertension 272 30 (21.7) 33 (24.6)

HbA1c (mmol/mol) 260 38.6 ± 10.9 39.6 ± 12.4

Fasting blood glucose (mmol/L) 261 5.6 ± 1.9 5.9 ± 2.4

Fasting insulin (pmol/L) 256 105.3 ± 85.7 96.4 ± 48.1

Page 17 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


18

Cholesterol (mmol/L)

Total 264 5.1 ± 1.1 5.1 ± 1.1

High-density lipoprotein 264 1.2 ± 0.4 1.2 ± 0.3

Low-density lipoprotein 256 3.2 ± 0.9 3.2 ± 0.9

Triglycerides (mmol/L) 264 1.6 ± 0.9 1.6 ± 0.8

*values represent means ± SD

§ IMD decile is an indicator of deprivation, with decile 1 being most deprived, and decile 10 the least

deprived.

Page 18 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


19

Table 2: Primary, secondary and exploratory outcomes by group

Change from baseline (mean ± SD) Treatment difference

Usual Care n

Total Diet

Replacement

n Adjusted difference (95% CI) p value

3 month

Weight (kg)2 -3.3 ± 4.2 97 -13.3 ± 6.3 114 -9.6 (-11.0, -8.2) <0.0001

Waist circumference (cm)3 -4.6 ± 4.9 94 -13.1 ± 7.7 111 -8.1 (-9.9, -6.4) <0.0001

Fat mass (kg)3 -3.0 ± 4.3 95 -10.7 ± 6.2 109 -7.1 (-8.6; -5.6) <0.0001

Systolic Blood Pressure (mmHg)3 3.5 ± 15.2 96 -2.6 ± 15.8 113 -5.8 (-9.1; -2.4) 0.0008

Diastolic Blood Pressure (mmHg)3 0.5 ± 8.9 96 -4.4 ± 9.3 113 -3.9 (-5.9; -1.8) 0.0002

6 months

Weight (kg)2 -4.5 ± 6.2 94 -15.1 ± 8.7 108 -9.6 (-11.6, 7.7) <0.0001

Waist circumference (cm)3 -7.0 ± 7.2 89 -15.4 ± 9.7 102 -8.3 (-10.6; -6.0) <0.0001

Fat mass (kg)3 -4.8 ± 5.6 86 -12.8 ± 9.6 94 -7.9 (-9.9; -5.9) <0.0001

Systolic Blood Pressure (mmHg)3 4.0 ± 14.0 92 0.3 ± 16.7 105 -3.3 (-6.9; 0.3) 0.0741


Page 19 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


20

Quality of life:

EQ-5D (Index)3 0.03 ± 0.15 73 0.07 ± 0.20 92

EQ-5D (VAS)3 7.0 ± 17.5 74 15.5 ± 18.2 93

OWL-QOL3 10.6 ± 14.8 74 17.4 ± 20.5 92

12 months

Weight (kg)1 -3.1 ± 7.0 95 -10.7 ± 9.6 104 -7.2 (-9.4,-4.9) <0.0001

Waist circumference (cm)3 -5.5 ± 7.3 91 -10.5 ± 9.1 99 -6.0 (-8.2; -3.7) <0.0001

Fat mass (kg)2 -4.1 ± 6.5 93 -10.4 ± 8.5 100 -5.8 (-7.9; -3.7) <0.0001

Systolic Blood Pressure (mmHg)2 2.9 ± 15.2 93 -1.6 ± 16.4 100 -2.9 (-6.4; 0.6) 0.1072


HbA1c (mmol/mol)2 -1.0 ± 7.7 75 -3.2 ± 8.8 91 -2.2 (-4.4; 0.0) 0.0511

Fasting glucose (mmol/L)3 0.1 ± 1.3 75 -0.5 ± 1.8 89 -0.4 (-0.8; -0.1) 0.0201

Fasting insulin (pmol/L)3 -10.4 ± 91.6 72 -21.8 ± 41.8 87 -18.0 (-32.0; -4.0) 0.0119

HOMA- IR3 -0.1 ± 1.5 70 -0.5 ± 1.2 86 -0.4 (-0.7;-0.2) 0.0026

HOMA β (%)3 -15.0 ± 83.8 70 -12.5 ± 39.7 86 -9.8 (-22.9; 3.4) 0.1452

HOMA S (%)3 -4.6 ± 70.4 70 28.8 ± 47.5 86 30.9 (16.4; 45.5) <0.0001

Page 20 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


21

Total Cholesterol (mmol/L) 0.0 ± 0.9 78 -0.2 ± 0.9 91 -0.2 (-0.5, 0.04) 0.1050

HDL cholesterol (mmol/L)3 0.1 ± 0.3 78 0.2 ± 0.3 91 0.1 (0.0; 0.2) 0.0911

LDL Cholesterol (mmol/L)2 -0.1 ± 0.7 73 -0.1 ± 0.6 87 0.0 (-0.2; 0.2) 0.8184

Triglycerides (mmol/L)3 0.1 ± 0.6 76 -0.3 ± 1.0 89 -0.4 (-0.6; -0.1) 0.0022

QRISK2 (%)3 0.0 ± 2.1 88 -0.9 ± 2.6 100 -1.0 (-1.7; -0.3) 0.0061

Quality of Life:

EQ-5D (Index)3 0.07 ± .014 93 0.09 ± 0.17 100

EQ-5D (VAS)3 9.2 ± 17.0 96 13.0 ± 18.7 101

OWL-QOL3 14.0 ± 16.7 94 17.0 ± 20.9 99

1 Primary outcome

2Secondary outcome

3Exploratory outcome

Page 21 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


22

Table 3: Frequency of all adverse events*

Participants reporting an event n (%)

Usual Care

Total Diet

Replacement Total

Gastrointestinal disorders:

Abdominal discomfort 3 (2.2) 3 (2.2) 6 (2.2)

Abdominal pain upper 0 (0) 3 (2.2) 3 (1.1)

Breath odour 0 (0) 3 (2.2) 3 (1.1)

Constipation 0 (0) 20 (14.9) 20 (7.4)

Dry mouth 0 (0) 4 (3.6) 4 (1.5)

Nausea 0 (0) 3 (2.2) 3 (1.1)

Painful defaecation 0 (0) 4 (3.0) 4 (1.5)

General disorders and administration site conditions:

Asthenia 0 (0) 3 (2.2) 3 (1.1)

Fatigue 1 (0.7) 12 (9.0) 13 (4.8)

Influenza like illness 4 (2.9) 3 (2.2) 7 (2.6)

Thirst 0 (0) 3 (2.2) 3 (1.1)

Infections and infestations:

Lower respiratory tract infection 0 (0) 3 (2.2) 3 (1.1)

Nasopharyngitis 7 (5.1) 4 (2.9) 11 (4.0)

Investigations:

Scan 3 (2.2) 0 (0) 3 (1.1)

Nervous system disorders:

Dizziness 2 (1.4) 6 (4.5) 8 (2.9)

Headache 3 (2.2) 11 (8.2) 14 (5.1)

Page 22 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


23

Psychiatric disorders:

Irritability 0 (0) 3 (2.2) 3 (1.1)

Respiratory, thoracic and mediastinal disorders:

Oropharyngeal pain 2 (1.4) 2 (1.5) 4 (1.5)

*Events of any severity that occurred in >2% of the sample.

Page 23 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


24

References

1. Lean ME, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2

diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet 2017 doi: 10.1016/s0140-

6736(17)33102-1.

2. World Health Organization. Global Health Risks: mortality and burden of disease attributable to

selected major risks., 2009.

3. Zomer E, Gurusamy K, Leach R, et al. Interventions that cause weight loss and the impact on

cardiovascular risk factors: a systematic review and meta-analysis. Obes Rev

2016;17(10):1001-11 doi: 10.1111/obr.12433.

4. Diabetes Prevention Program Research G. Long-term effects of lifestyle intervention or metformin

on diabetes development and microvascular complications over 15-year follow-up: the

Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinology

2015;3(11):866-75 doi: 10.1016/S2213-8587(15)00291-0.

5. Tsai AG, Wadden TA. Treatment of obesity in primary care practice in the United States: a

systematic review. J Gen Intern Med 2009;24(9):1073-9 doi: 10.1007/s11606-009-1042-5.

6. Aveyard P, Lewis A, Tearne S, et al. Screening and brief intervention for obesity in primary care: a

parallel, two-arm, randomised trial. The Lancet 2016;388(10059):2492-500 doi:

10.1016/S0140-6736(16)31893-1.

7. Booth HP, Prevost AT, Gulliford MC. Access to weight reduction interventions for overweight and

obese patients in UK primary care: population-based cohort study. BMJ Open

2015;5(1):e006642 doi: 10.1136/bmjopen-2014-006642.

8. Jebb SA, Ahern AL, Olson AD, et al. Primary care referral to a commercial provider for weight loss

treatment versus standard care: a randomised controlled trial. Lancet 2011;378(9801):1485-

92 doi: 10.1016/S0140-6736(11)61344-5.

9. Ahern AL, Wheeler GM, Aveyard P, et al. Extended and standard duration weight loss referrals for

adults in primary care (WRAP): a pragmatic randomised controlled trial. The Lancet

2017;389(10085):2214-25.

10. Jolly K, Daley A, Adab P, et al. A randomised controlled trial to compare a range of commercial or

primary care led weight reduction programmes with a minimal intervention control for

weight loss in obesity: the Lighten Up trial. BMC Public Health 2010;10:439 doi:

10.1186/1471-2458-10-439.

11. Hartmann-Boyce J, Johns DJ, Jebb SA, et al. Behavioural weight management programmes for

adults assessed by trials conducted in everyday contexts: systematic review and meta-

analysis. Obesity Reviews 2014;15(11):920-32 doi: 10.1111/obr.12220.

12. Parretti HM, Jebb SA, Johns DJ, et al. Clinical effectiveness of very-low-energy diets in the

management of weight loss: a systematic review and meta-analysis of randomized

controlled trials. Obesity Reviews 2016;17(3):225-34 doi: 10.1111/obr.12366.

13. National Institute for Health and Clinical Excellence (NICE). Obesity: identification, assesment and

management London, 2014.

14. Jensen MD, Ryan DH, Apovian CM, et al. 2013 AHA/ACC/TOS Guideline for the Management of

Overweight and Obesity in Adults A Report of the American College of Cardiology/American

Heart Association Task Force on Practice Guidelines and The Obesity Society. Journal of the

American College of Cardiology 2014;63(25):2985-3025 doi: 10.1016/j.jacc.2013.11.004.

15. Jebb SA, Astbury NM, Tearne S, et al. Doctor Referral of Overweight People to a Low-Energy

Treatment (DROPLET) in primary care using total diet replacement products: a protocol for a

randomised controlled trial. BMJ Open 2017;7(8):e016709 doi: 10.1136/bmjopen-2017-

016709.

16. British Heart Foundation. So you want to lose weight... for good. London: BHF Publications, 2005.

17. Niero M, Martin M, Finger T, et al. A new approach to multicultural item generation in the

development of two obesity-specific measures: the Obesity and Weight Loss Quality of Life

Page 24 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


25

(OWLQOL) questionnaire and the Weight-Related Symptom Measure (WRSM). Clin Ther

2002;24(4):690-700.

18. EuroQol G. EuroQol--a new facility for the measurement of health-related quality of life. Health

Policy 1990;16(3):199-208.

19. Hippisley-Cox J, Coupland C, Vinogradova Y, et al. Predicting cardiovascular risk in England and

Wales: prospective derivation and validation of QRISK2. BMJ 2008;336(7659):1475-82 doi:

10.1136/bmj.39609.449676.25.

20. Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation

uses the computer program. Diabetes Care 1998;21(12):2191-2.

21. Zomer E, Gurusamy K, Leach R, et al. Interventions that cause weight loss and the impact on

cardiovascular risk factors: a systematic review and meta-analysis. Obesity Reviews

2016;17(10):1001-11 doi: 10.1111/obr.12433.

22. Ma C, Avenell A, Bolland M, et al. Effects of weight loss interventions for adults who are obese

on mortality, cardiovascular disease, and cancer: systematic review and meta-analysis. BMJ

2017;359:j4849 doi: 10.1136/bmj.j4849.

23. Warkentin LM, Das D, Majumdar SR, et al. The effect of weight loss on health-related quality of

life: systematic review and meta-analysis of randomized trials. Obesity Reviews

2014;15(3):169-82 doi: 10.1111/obr.12113.

24. Dombrowski SU, Knittle K, Avenell A, et al. Long term maintenance of weight loss with non-

surgical interventions in obese adults: systematic review and meta-analyses of randomised

controlled trials. BMJ 2014;348:g2646 doi: 10.1136/bmj.g2646.

25. Purcell K, Sumithran P, Prendergast LA, et al. The effect of rate of weight loss on long-term

weight management: a randomised controlled trial. Lancet Diabetes Endocrinol

2014;2(12):954-62 doi: 10.1016/S2213-8587(14)70200-1.

26. Unick JL, Hogan PE, Neiberg RH, et al. Evaluation of early weight loss thresholds for identifying

nonresponders to an intensive lifestyle intervention. Obesity (Silver Spring) 2014;22(7):1608-

16 doi: 10.1002/oby.20777.

Page 25 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


26

Funding

Funding for this study was provided through a research grant from Cambridge Weight Plan UK Ltd to

the University of Oxford and also supported by the National Institute for Health Research (NIHR)

Collaboration for Leadership in Applied Health Research and Care Oxford at Oxford Health NHS

Foundation Trust. SAJ, PA and NA are supported by the Oxford NIHR Biomedical Research Centre

and National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health

Research and Care Oxford at Oxford Health NHS Foundation Trust.

The funders had no role in study design, data collection, data analysis, data interpretation, or writing

of the report. The views are those expressed by the author(s) and not necessarily those of the NHS,

the NIHR or the Department of Health. The University of Oxford holds the relevant clinical trials

insurance policy for this study and acted as the study sponsor.

Acknowledgements

We thank participants and the staff at the primary care practices who contributed to this study. We

also thank staff in the Nuffield Department of Primary Care NIHR Clinical Trials Unit for support in

the running of this trial, in particular Dr Sadie Kelly for acting as data manager and Mrs Sarah Tearne

who assisted in setting up the study.

Contributors

SAJ and PA designed the study, NA was the trial manager and AN was the trial statistician. NA, KH,

KC and RL were responsible for data collection. SAJ and NA drafted the paper and all authors

contributed to the interpretation of the data and critical review of the paper.

Potential Conflicts

All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf

and declare: The trial was in part funded by the Cambridge Weight Plan but the authors receive no

personal financial benefit. The study was an investigator-initiated protocol and the funder has no

access to the data. PA and SAJ have previously conducted studies in which weight-loss interventions

were provided to the NHS by WeightWatchers, Slimming World or Rosemary Conley.

Ethical Approval

The trial was approved by South Central Oxford B NHS Research Ethics Service Committee, (NHS REC

No 15/SC/0337 )

Data sharing

Requests for access to data from the DROPLET study should be addressed to the principal

investigator at [email protected]. All the individual participant data collected, after de-

identification (including the data dictionary) will be available on request following publication. All

proposals requesting data access will need to complete a data request form with details of the

Page 26 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


27

research question and analysis plan. All proposals will require the approval of the investigator team

before any data are released.

Transparency

The guarantor (SAJ) affirms that the manuscript is an honest, accurate, and transparent account of

the study bring reported; that no important aspects of the study have been omitted; and that any

discrepancies from the study as planned (and, if relevant, registered) have been explained.

Page 27 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Supplementary Appendix

Table of Contents:

Full Inclusion and Exclusion criteria 1

Medication adjustment guidelines 2

Figure S1: Proportion of participants losing ≥5% or ≥10% baseline weight 3

Table S1: Typical Nutritional Composition of meal replacements used in TDR 6

Table S2: Adjusted treatment effects under different missing data approaches 7

Table S3: Adverse events analysis 8

Page 28 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


2

Full list of Inclusion and Exclusion criteria

Inclusion:

Participants were included in the study if the met the following criteria:

• Participant is willing and able to give informed consent for participation in the study.

• Aged 18 years or above.

• Body Mass Index ≥30 kg/m2.

• Likely to benefit from weight loss in the Primary Care physician’s opinion.

Exclusion:

The following criteria were used to exclude individuals for whom weight loss might not be safe, those

who may have difficulty adhering to TDR intervention, or those with medical conditions that were a

contraindication to the TDR programme.

• Currently or recently (within 3 months of study entry) attended a weight management

programme or currently participating in another weight loss study.

• Had bariatric surgery, or scheduled bariatric surgery.

• Pregnant, breastfeeding, or planning to become pregnant during the course of the study.

• Receiving insulin therapy

• Heart attack or stroke within the last 3 months

• Heart failure of grade II New York Heart Association and more severe

• Angina, arrhythmia, including atrial fibrillation or prolonged QT syndrome

• Taking MAOI medication

• Taking anticoagulant medication (e.g. warfarin)

• Taking varenicline (smoking cessation medication)

• Chronic renal failure of stage 4 or 5

• Active liver disease (except NAFLD) a past history of hepatoma or within 6 months of onset of

acute hepatitis.

• People having active treatment for cancer other than skin cancer treated with curative intent by local

treatment only or people taking hormonal or other long-term secondary prevention treatment after

initial cancer treatment.

• Active treatment or investigation for possible or confirmed gastric or duodenal ulcer.

Maintenance treatment with acid-suppression is not a contra-indication.

• Porphyria

• Scheduled for surgery within 12 months

• A member of household is already enrolled in the study

• Unwilling to provide blood samples

• Patients that the Primary Care physician judges not able to meet the demands of either

treatment programme or measurement schedule. This may include severe medical problems

not listed above or severe psychiatric problems including substance misuse that make following

the treatment programme or adhering to the protocol unlikely.

Page 29 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


3

Medication adjustment guidelines

Page 30 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


4

Figure S1: Proportion of participants losing ≥5% or ≥10% baseline weight*

*Proportion of participants in each group in the intention to treat population who had lost ≥5% and

≥10% baseline weight at 12 months. (Participants with complete data at baseline and follow-up

only N=199). For ≥5% weight loss the OR (95% CI) was 6.5 (3.4, 12.2) and for ≥10% weight loss was

4.9 (2.4, 9.9) favouring the TDR group.

0

10

20

30

40

50

60

70

80

≥ 5% ≥ 10%

Pro

po

rtio

n o

f p

art

icia

pn

ts (

%)

Weight change from baseline at 12 months

Usual Care

Total Diet Replacement

P<0.0001

P<0.0001

Page 31 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


5

Figure S2: Treatment effect by sub-groups

*IMD decile is an indicator of deprivation, with decile 1 being most deprived, and decile 10 the least deprived. IMD groups were compared using

median split. IMD was not a pre-specified sub-group analysis, and was added after the statistical analysis plan was written, but before the

primary analysis was conducted.

Primary analysis

Practice 10

Practice 9

Practice 8

Practice 7

Practice 6

Practice 5

Practice 4

Practice 3

Practice 2

Practice 1

Practice

IMD decile >8

IMD decile ≤ 8

IMD decile*

BMI >35

BMI ≤ 35

BMI

55-78

44-54

19-43

Age group

Female

Male

Sex

Type 2

None

Diabetes status

Subgroup

-7.20 [ -9.40, -4.90]

-9.10 [ -24.70, 6.40]

-3.80 [ -19.10, 11.40]

4.90 [ -13.90, 23.70]

-10.60 [ -19.00, -2.30]

-5.80 [ -12.80, 1.10]

-2.10 [ -8.70, 4.50]

-2.70 [ -9.70, 4.40]

-11.80 [ -17.10, -6.50]

-14.00 [ -20.90, -7.00]

-6.30 [ -10.70, -1.80]

-7.70 [ -10.50, -4.90]

-5.90 [ -9.60, -2.20]

-8.70 [ -11.80, -5.70]

-5.00 [ -8.30, -1.70]

-8.80 [ -12.00, -4.50]

-6.90 [ -10.60, -3.20]

-7.10 [ -9.40, -4.80]

-5.90 [ -8.80, -2.90]

-9.20 [ -12.60, -5.80]

-10.20 [ -15.60, -4.80]

-6.60 [ -9.00, -4.10]

Treatment effect [95% CI]

199

4

4

3

15

21

22

21

36

21

52

73

123

105

94

78

67

54

113

86

35

164

N

-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30

Difference in weight change from baseline (kg) at 12 months (95% CI)

Favours intervention Favours control

Page 32 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


6

Table S1: Typical Nutritional Composition of meal replacement products used in the total diet

replacement programme

Per 100g Per serving

Energy

kJ 1565 845

kcal 370 200

Fat (g) 4.8 2.6

of which saturates (g) 0.9 0.5

Carbohydrate (g) 50.1 27.0

Of which sugars (g) 35.1 18.9

Fibre 5.2 2.8

Protein 29.2 15.8

Salt 1.2 0.6

Page 33 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


7

Table S2: Adjusted treatment effects under different missing data approaches

BOCF LOCF Multiple imputation Completers only

Usual Care TDR Usual care TDR Usual Care TDR Usual Care TDR

(N = 138) (N=134) (N= 138) (N=134) (N=138) (N= 134) (N=95) (N=104)

Unadjusted weight change from baseline* -2.1 ± 6.0 -8.3 ± 9.6 -2.7 ± 6.3 -10.2 ± 9.2 -3.5 ± 8.2 -10.2 ± 9.7 -3.1 ± 7.0 -10.7 ± 9.6

Difference between groups † -6.1 (-8.0, -4.3) -7.5 (-9.4, -5.6) -6.4 (-8.5, -4.4) -7.5 (-9.8, -5.1)

p-value <0.0001 <0.0001 <0.0001 <0.0001

*Mean weight (kg) ± SD

† Adjusted mean difference (kg) (95% CI) using linear mixed effects model with fixed effects for randomisation group, baseline weight, visit and

randomised group x visit interaction. Random effects accounting for practice and participant and within subject variance covariance matrix

specified as unstructured. Age and sex were included as covariates as baseline values were predictive of missingness.

Page 34 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


8

Table S3: Adverse Events analysis*

Usual Care Total Diet Replacement p value

Participants reporting at least one AE n (%) 41 (29.7) 69 (51.5) 0.0003

Participants reporting a moderate or severe AE n (%) 17 (12.3) 15 (11.2) 0.85

*A logistic model would not converge, therefore a Fisher’s exact test was used to test the associations between groups.

Page 35 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


1Jebb SA, et al. BMJ Open 2017;7:e016709. doi:10.1136/bmjopen-2017-016709

Open Access

AbstrActIntroduction The global prevalence of obesity has risen significantly in recent decades. There is a pressing need to identify effective interventions to treat established obesity that can be delivered at scale. The aim of the Doctor Referral of Overweight People to a Low-Energy Treatment (DROPLET) study is to determine the clinical effectiveness, feasibility and acceptability of referral to a low-energy total diet replacement programme compared with usual weight management interventions in primary care.Methods and analysis The DROPLET trial is a randomised controlled trial comparing a low-energy total diet replacement programme with usual weight management interventions delivered in primary care. Eligible patients will be recruited through primary care registers and randomised to receive a behavioural support programme delivered by their practice nurse or a referral to a commercial provider offering an initial 810 kcal/d low-energy total diet replacement programme for 8 weeks, followed by gradual food reintroduction, along with weekly behavioural support for 24 weeks. The primary outcome is weight change at 12 months. The secondary outcomes are weight change at 3 and 6 months, the proportion of participants achieving 5% and 10% weight loss at 12 months, and change in fat mass, haemoglobin A1c, low-density lipoprotein cholesterol and systolic and diastolic blood pressure at 12 months. Data will be analysed on the basis of intention to treat. Qualitative interviews on a subsample of patients and healthcare providers will assess their experiences of the weight loss programmes and identify factors affecting acceptability and adherence.Ethics and dissemination This study has been reviewed and approved by the National Health ServiceHealth Research Authority (HRA)Research Ethics Committee (Ref: SC/15/0337). The trial findings will be disseminated to academic and health professionals through presentations at meetings and peer-reviewed journals and to the public through the media. If the intervention is effective, the results will be communicated to policymakers and commissioners of weight management services.trial registration number ISRCTN75092026.

IntroductIonThe prevalence of obesity worldwide has more than doubled since 1980.1 According to the latest estimates from the WHO, more than 1.9 billion adults were overweight, of whom 600 million were obese, representing 39% and 13% of the world’s adult popula-tion, respectively.2 Obesity is associated with premature mortality,3 but also substantial morbidity, including significantly increased risks of diabetes, cardiovascular disease and most non-smoking-related cancers, as well as physical impairments linked to excess weight such as breathlessness, joint problems and back pain.4 Collectively this creates a burden of ill-health and reduced quality of life for individuals, additional treatment costs to

Doctor Referral of Overweight People to a Low-Energy Treatment (DROPLET) in primary care using total diet replacement products: a protocol for a randomised controlled trial

Susan A Jebb, Nerys M Astbury, Sarah Tearne, Alecia Nickless, Paul Aveyard

To cite: Jebb SA, Astbury NM, Tearne S, et al. Doctor Referral of Overweight People to a Low-Energy Treatment (DROPLET) in primary care using total diet replacement products: a protocol for a randomised controlled trial. BMJ Open 2017;7:e016709. doi:10.1136/bmjopen-2017-016709

► Prepublication history and additional material are available. To view these files please visit the journal online (http:// dx. doi. org/ 10. 1136/ bmjopen- 2017- 016709).

Received 3 March 2017Revised 13 June 2017Accepted 21 June 2017

Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford UK

correspondence toDr Nerys M Astbury; nerys. astbury@ phc. ox. ac. uk

Protocol

strengths and limitations of this study

► This study is the largest randomised controlled trial to date of a low-energy total diet replacement programme for weight management in routine primary care.

► This intervention is based on a model of care where general practitioners refer patients to a programme delivered in the community by a commercial provider using non-National Health Service staff, which, if successful, could be readily adopted into practice without the need for specialist training for the primary care workforce.

► The primary outcome is weight at 1 year. Although this is 9 months after the low-energy total diet replacement, epidemiological evidence suggests that any weight lost will continue to be regained beyond 1 year.

► The intention of obesity treatment programmes is to improve long-term health, but this study does not include morbidity or mortality outcomes.

► Longer term follow-up data would be helpful to better estimate the longer health impact and cost-effectiveness of the intervention.

group.bmj.com on March 19, 2018 - Published by http://bmjopen.bmj.com/Downloaded from Page 36 of 46


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

http://bmjopen.bmj.com/

http://dx.doi.org/10.1136/bmjopen-2017-016709



http://crossmark.crossref.org


http://group.bmj.com


2 Jebb SA, et al. BMJ Open 2017;7:e016709. doi:10.1136/bmjopen-2017-016709

Open Access

the National Health Service (NHS) and reductions in economic productivity.5 While high priority must be given to prevent future cases of obesity, in the short term, there is a pressing need to identify effective interven-tions to treat established obesity. Research has shown that even modest reductions in weight can bring significantly reduced risks of disease. For example, in the US Diabetes Prevention Program (DPP), individuals randomised to an intensive lifestyle intervention lost 7 kg by the end of the first year. Although some of this weight was regained, the intensive lifestyle group remained 4 kg lighter than the usual care group at 4 years, and this reduced the inci-dence of diabetes by 58% relative to usual care,6 with benefits persisting to at least 15-year follow-up despite weight regain.7

Primary care is an important setting for weight manage-ment interventions to reduce multimorbidity. However, although a number of interventions have been shown to be effective in intensive research studies, this success has not always been replicated in routine settings. For example, there was no significant reduction in weight when a weight loss programme adapted from the DPP was delivered by primary care teams.8 Our recent review of interventions suitable for use in routine care9 and a second review, using slightly different inclusion criteria, of interventions specifically delivered in primary care10 both concluded that behavioural weight management interventions led by primary care practitioners were ineffective. This may relate in part to the complexity of advice needed for successful dietary change and the need for frequent contact to provide support, which exceeds the capacity of routine primary care systems. However, although a number of interventions have been shown to be effective in intensive research studies, this success has not always been replicated in routine settings. General practitioner (GP) referral to a commercial provider offering group-based support is an effective option for weight management in primary care, and our meta-anal-ysis showed a mean reduction in weight of 2.3 kg over no intervention at 1 year.9 However, greater weight losses would be expected to bring greater health gains.

Very low-energy diets (VLEDs) have been used for weight loss over many years in specialist settings. A VLED is defined as a diet providing ≤800 kcal a day, based on the use of specially formulated products designed as the sole source of nutrition during periods of total diet replacement. When used as directed, these formula products meet 100% of the dietary reference values for vitamins, minerals and trace elements for healthy, weight-stable people and are enriched with high biolog-ical-value protein. Although most contain some dietary fibre, a fibre supplement may also be recommended. A recent systematic review and meta-analysis of the available randomised controlled trials showed that behavioural weight management interventions incorporating a VLED led to 3.9 kg greater weight loss at 1 year compared with intensive specialist-delivered behavioural programmes.11 However, most of the trials included in this review were

small, typically including only 50–100 participants who were treated by obesity specialists, and many trials had methodological limitations.

UK guidance from the National Institute for Health and Care Excellence (NICE) recommends that VLEDs may only be used for a maximum of 12 weeks in people who have a clinical need to lose weight rapidly, such as prior to a knee replacement surgery or those seeking fertility services, but recommends against their routine use to manage obesity.12 Clinical guidance in the USA does not recommend the routine use of VLEDs, but rather suggests that their use ‘may be reasonable in limited circum-stances, but only when provided by trained practitioners in a medical care setting where medical monitoring and high intensity lifestyle intervention can be provided’.13

Nevertheless, there has been growing interest in the potential for routine use of weight loss programmes similar to traditional VLEDs, in so far as they incorpo-rate a period of total diet replacement using specially formulated products as the sole source of nutrition, but where the energy content is more than 800 kcal/day but less than 1200 kcal/day. The NICE guidelines suggest that this type of low-energy diet could be consid-ered for weight management, providing care is taken to ensure they are nutritionally complete.12 There is one observational report (n=91) on the use of these low-en-ergy total diet replacement programmes in primary care which found that 64% of participants completed the 810 kcal/day dietary programme, defined as either 12 weeks or reaching 20 kg weight loss, with a mean weight loss of 16.9 kg (SD=6.0 kg). One-third of partici-pants starting the programme maintained a weight loss of ≥15 kg at 12 months.14 A large randomised controlled trial, the DiRECT (Diabetes Remission Clinical Trial) study, is currently underway to investigate whether this type of low-energy total diet replacement programme can be used to treat type 2 diabetes among people who are also overweight.15 It will compare the health effects of the current best-available type 2 diabetes care with those achieved through weight management based on a low-energy total diet replacement programme. While this will provide important mechanistic evidence on the links between weight loss and diabetes risk, it will be deliv-ered by NHS staff, whereas the present study will test the effectiveness of referral outside the NHS to a commercial provider.

To fill this evidence gap, we will conduct a randomised controlled trial to specifically test the effectiveness of a GP referral to a community-based low-energy total diet replacement programme for patients who are obese and likely to benefit from weight loss. It will assess the clinical effectiveness of a weight loss intervention by measuring weight loss and the change in biomarkers of cardiovascular risk at 12 months relative to weight loss advice provided by practice nurses. This comparator is intended to represent ‘usual care’, although in practice most patients who are obese are not offered support to lose weight.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

The context for this trial follows the established model for GP referral to community group-based weight loss programmes.16 This uses the generic authority and cred-ibility of health professionals to motivate patients to consider weight management and the specialist knowl-edge of the commercial provider to guide the intervention and offer frequent contact and behavioural support to the patient. If successful, it will provide another option for weight management that can be offered to patients in primary care, and GPs will be able to guide patients towards the treatment that best fits their circumstances and preferences. This trial will specifically test whether a partnership between GPs and providers will allow for the safe provision of low-energy total diet replacement programmes even for patients with multimorbidity who may gain the greatest benefits from such interventions but who may also need clinical oversight and adjust-ments to some of their medications as they lose weight. It will provide the opportunity for qualitative research to investigate the perspectives of patients and healthcare practitioners on this type of treatment.

objectiveThe aim of the Doctor Referral of Overweight People to a Low-Energy Treatment (DROPLET) trial is to determine the clinical effectiveness, feasibility and acceptability of referral to a low-energy total diet replacement programme compared with usual weight management interventions in primary care.

MEthodsDesign and settingThe study will take place in general practices in England. The study is designed as an individually randomised, two-arm and parallel group superiority trial with the primary endpoint as objectively measured changes in body weight from baseline to 12 months. Due to the nature of the intervention, it will not be possible to blind participants, clinicians or some of the study team to the treatment allocation after randomisation.

recruitmentAround 10 general practices will be identified to take part through the clinical research networks. Recruited prac-tices will be asked to conduct a search of their electronic health records in order to identify suitable patients for the DROPLET study. As a result of this search, eligible patients will be sent an invitation letter from their GP as part of a staggered mailout. Patients will be encouraged to call the research team if they are interested in taking part.

GPs may also identify eligible patients during routine consultations. The GP will provide the patient with an invitation letter and suggest that the patient ring the study team. The study team will provide the potential partici-pants with information on what taking part in the study will entail, and an initial assessment of suitability to take part. Those who make contact and self-report meeting

the eligibility criteria will be scheduled for a baseline/enrolment appointment.

Inclusion criteria ► participant is willing and able to give informed con-

sent for participation in the study ► aged 18 years or above ► body mass index (BMI) ≥30 kg/m2

► likely to benefit from weight loss in the GP’s opinion.

Exclusion criteria ► unable to understand English ► currently or recently (within 3 months of study entry)

attended a weight management programme or cur-rently participating in another weight loss study

► had bariatric surgery or scheduled bariatric surgery ► pregnant, breast feeding or planning to become preg-

nant during the course of the study ► receiving insulin therapy ► heart attack or stroke within the last 3 months ► heart failure of grade II New York Heart Association

and more severe ► angina, arrhythmia, including atrial fibrillation or

prolonged QT syndrome ► taking monoamine- oxidase inhibitor (MAOI) medi-

cation ► taking anticoagulant medication (eg, warfarin) ► taking varenicline (smoking cessation medication) ► chronic renal failure of stage 4 or 5 ► active liver disease (except non-alcoholic fatty liver

disease (NAFLD), a history of hepatoma or within 6 months of onset of acute hepatitis

► people having active treatment for cancer other than skin cancer treated with curative intent by local treat-ment only, or people taking hormonal or other long-term secondary prevention treatment after initial can-cer treatment

► active treatment or investigation for possible or con-firmed gastric or duodenal ulcer; maintenance treat-ment with acid suppression is not a contraindication

► porphyria ► scheduled for surgery within 12 months ► a member of household is already enrolled in the

study ► unwilling to provide blood samples ► patients that the GP judges not able to meet the de-

mands of either treatment programme or measure-ment schedule; this may include severe medical prob-lems not listed above or severe psychiatric problems including substance misuse that make following the treatment programme or adhering to the protocol unlikely.

Participant flowThe baseline/eligibility assessment will be scheduled with a practice nurse or healthcare assistant at their own GP practice, where informed consent for participation in the study will be obtained before eligibility will be formally



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

assessed. After demographic information and all base-line measurements have been collected, the participant will be randomised to the allocated treatment group using the online randomisation system. The patients’ own GP will be notified by letter of the enrolment and randomisation of their patient, so that it may be docu-mented on their medical record. Participants allocated to the low-energy total diet replacement programme and taking medications for type 2 diabetes, hypertension or high cholesterol will have their medications reviewed by a prescribing member of the clinical care team, usually the GP or trained nurse prescriber. During this medication review, the clinician will decide what changes to medica-tions are required at the time the participant commences the low-energy total diet replacement programme, with guidance provided by the study team (see online supple-mentary figure 1). In addition, participants randomised to the low-energy total diet replacement group and who take antihypertensive medications will be provided with a home blood pressure monitor and asked to record blood pressure once daily during the weight loss phase (weeks 1–12). These readings can be used to guide clinicians with any further changes in hypertension medications.

All participants will be invited to attend a 4-week follow-up appointment with the practice nurse. The main purpose of the visit is a clinical review of medica-tion, including any adjustments required. Any changes in medication will be recorded on the concomitant medi-cation log. Participants will be invited to attend further follow-up visits with a member of the trial team at the GP practice at 12 weeks, 6 months and 12 month following randomisation. Participant flow through the study is outlined in figure 1.

sample sizeThe total number of participants to be recruited for this study is 270. This is based on a sample size calculation for the primary outcome using equal variance indepen-dent samples t-test assuming a difference between groups at 12 months of 4 kg with an SD in both groups of 9 kg, obtained from a meta-analysis of published studies.11 The sample size has been inflated by 20% to account for attri-tion, and assumes 90% power and two-sided alpha of 5%.

randomisationAll eligible, consenting participants will be randomised with an allocation ratio of 1:1 to low-energy total diet replacement or usual care programmes using an online programme, which reveals group allocation as per a computer-generated randomisation list. The rando-misation criteria will be validated by an independent statistician. Allocation will be stratified by GP practice and baseline BMI (≤35 kg/m2 or >35 kg/m2) using stratified block randomisation with randomly varying block sizes of 2, 4 and 6. The randomisation software ensures full allocation concealment, with the allocation group only revealed to the person performing the randomisation once a study identifier and required stratification details have been entered.

InterventionsLow-energy total diet replacementThe programme offered to participants randomised to the active intervention will be provided by Cambridge Weight Plan, Northants, UK.

Following randomisation participants allocated to this group will be referred to a local Cambridge Weight Plan counsellor who will invite the participant to attend regular appointments for 24 weeks. These appointments consist of motivational support, encouragement, reassurance and problem-solving. All counsellors attend a 1-day in-person training course covering screening for suitability, nutri-tion, behavioural approaches and medical monitoring. They must pass an accreditation examination before they are allowed to deliver the programme in the community. Thereafter, they have a yearly training updates, a nomi-nated sponsor (experienced counsellor) and access to an online chat forum for sharing queries. Cambridge Weight Plan has a healthcare professional available for the counsellors to consult for advice on specific medical and nutritional queries. Counsellors delivering the inter-vention for the purposes of this trial received short trial specific training before being allocated study participants.

During the first 12 weeks the participant will meet with their counsellor weekly. Patients will be asked to follow a programme based on using formula meal replacement products (soups, shakes and bars) and milk comprising 810 kcal/day (3389 kJ/day). For the first 8 weeks, patients will be advised to replace all their usual foods and drinks with four of the formula products daily: 750 mL of skimmed milk, 2.25 L of water or other non-calorific drinks and a fibre supplement (total diet replacement stage). During the first 2 weeks, the formula products will be limited to liquid products (soups and shakes), but from week 3 onwards participants will have the option to include meal replacement bars as part of the formula product allowance. After 8 weeks there will be a 4-week stepwise reduction in the use of formula meal replacement products and a gradual reintroduction of food-based meals. The weight maintenance phase from week 12 to 24 participants attend monthly appointments at 16, 20 and 24 weeks, during this phase participants are advised to consume only one formula product a day, with the remainder of the diet to consist of self-selected food. This weight maintenance phase will include a recom-mendation to return to the total diet replacement stage for periods of up to 4 weeks if participants regain 1 kg or more than their weight measured at 12 weeks.

All consultations with counsellors and formula prod-ucts will be provided to participants by their nominated counsellor and will be free of charge for the first 24 weeks, after which the intervention will end. Participants in both groups will be free to choose whether or not to continue with the programme, but at their own cost.

ComparatorThe comparator intervention will consist of the usual weight management programme provided by a member



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

https://dx.doi.org/10.1136/bmjopen-2017-016709

https://dx.doi.org/10.1136/bmjopen-2017-016709





Open Access

Figure 1 Participant flow through the study. BMI, body mass index; GP, general practitioner.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

Figure 2 Schedule of measurements.18

of the practice nurse team who has been trained to offer a weight loss programme. The trial will take place only in practices where this is routine care. Participants

allocated to the usual care group will not be prevented from attending other weight management groups if they choose to do so, but no NHS referrals to these schemes



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

will be offered during the trial. The practice nurse will give participants a copy of the booklet ‘So you want to lose weight … for good’.17 This 47-page booklet provides advice akin to a behavioural weight management programme. The aim is to produce a weight loss goal of 0.5–1 kg/week. It includes goal setting, advice on portion control and physical activity, other behavioural strate-gies, and monitoring and feedback on progress. Nurses will be asked to offer a programme for 12 weeks, at a frequency that is usually used in the practice (eg, weekly or biweekly).

Physical activityWe recognise the importance of the role of aerobic and resistance exercise in facilitating weight loss and maintaining lean body mass to facilitate weight loss main-tenance.

Participants randomised to the low-energy total diet replacement arm are given appropriate advice based on their previous exercise history, current ability and what is appropriate for their stage weight loss programme. Clin-ical guidelines in the UK emphasise the importance of advice to increase physical activity, and we would expect this to be incorporated into the control ‘usual care’ inter-vention.

outcomesPrimary outcome

► change in body weight from baseline to 12 months.

Secondary outcomes ► change in body weight from baseline at 3 and 6

months ► proportion of participants achieving 5% and

10% weight loss at 12 months ► change in fat mass between baseline and 12 months ► change in low-density lipoprotein (LDL) cholesterol

concentrations between baseline and 12 months ► change in haemoglobin A1c (HbA1c) between base-

line and 12 months ► change in systolic and diastolic blood pressure be-

tween baseline and 12 months.

Exploratory outcomes ► change in fat mass from baseline to 12 weeks and from

baseline to 6 months ► change in waist circumference from baseline to 3, 6

and 12 months ► change in triglyceride and high-density lipopro-

tein (HDL) cholesterol concentrations between base-line and 12 months

► change in fasting glucose and insulin concentra-tions and change in Homeostatic Model Assesment (HOMA)of insulin resistance (HOMA- IR), insulin sensitivity (HOMA-%S) and beta cell function (HO-MA-%B) between baseline and 12 months

► change in systolic and diastolic blood pressure be-tween baseline and 3 months and between baseline and 6 months

► change in QRISK between baseline and 12 months ► change in the quality of Life measured using the EQ-

5D scale between baseline and 12 months ► change in obesity-related quality of life measured with

the Obesity-Specific Quality of Life (OWLQOL) be-tween baseline, 3, 6 and 12 months

► proportion of people continuing their weight loss at-tempt and following the prescribed programme at 4, 8 and 12 weeks

► the number of weight control behaviours that partic-ipants are using assessed using the Oxford Food and Activity Behaviours (OxFAB) questionnaire18 at 3 and 6 months

► qualitative interviews with a subsample of participants at 6 and 12 months

► adverse event (AE) reports up to 12 weeks, the end of the weight loss intervention or 6 months for AEs known or presumed to be related to gallstones.

MeasurementsFigure 2 provides a summary of the measurements collected.

Sociodemographic characteristicsParticipants will be asked to self-report age, sex and ethnicity.

Medical historyRelevant medical history and all concomitant medication will be recorded and checked against the participants’ medical record. Participants will also be asked to self-re-port items required to determine cardiovascular risk score using QRISK2.19

Physical measurementsHeight will be measured to the nearest 1 cm using stadi-ometers available in the practice. Weight will be recorded to the nearest 0.1 kg using an electronic scale (SC-240 MA, Tanita Japan), which will also record the proportion of body fat using bioelectrical impedance. Waist circum-ference will be measured in the horizontal plane at the upper border of the iliac crest at the end of expiration20 using a fibreglass non-stretch tape measure fitted with a tensioning device (Gulick II Tape Measure, Fitness Mart USA). Seated blood pressure will be measured in triplicate with 1 min between each measure. All physical measures are performed by assessors trained according to the study manual of procedures.

Fasting blood sampleA fasting venous blood sample will be collected (to be analysed for glucose, insulin, HbA1c, HDL and LDL cholesterol, triglycerides). When baseline/enrolment appointments are scheduled at times when it may be inappropriate to fast, participants will be asked to arrange for a fasting blood sample to be collected at an alterna-tive appointment within 7 days of the enrolment visit and before the participant commences the allocated weight loss programme.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

QuestionnairesParticipants will be provided with a questionnaire booklet which they will be asked to complete and return to the trial team in a postage paid envelope provided. The ques-tionnaire booklet contains the following measures:

► OWLQOL: a weight-specific instrument intended to be used to assess obesity-specific symptoms and quality of life, general functional status and well-being, and per-son-specific preference measurement.21

► Quality of life: EQ-5D will be used as a standardised val-idated instrument used for measuring general health status.22

► Programme adherence: self-reported adherence to the allocated programme and methods participants are using to attempt to lose weight will be recorded by questionnaire.

► Programme feedback: will be assessed using several 5-point Likert scales, including whether there is an aim to continue with the programme.

► OxFAB: a questionnaire to assess personal strategies used by individuals for the purposes of weight loss.18

Retention and withdrawalWe will seek to follow up all participants except those who expressly withdraw from the study. Participants who decide to withdraw from or discontinue the intervention allocated as part of the study will be asked to return for follow-up visits to collect outcome measures. To promote participant retention and complete follow-up, partici-pants will be offered a £10 gift card for attending each of the 6-month and 12-month follow-up visits.

Adverse eventsAEs are of relevance in this trial because many prac-titioners feel these programmes are poorly tolerated and unsuitable for routine use in primary care. We will record AEs following Good Clinical Practice. All serious and non-serious AEs that occur during the first 12 weeks of the study or until the termination of the weight loss programme will be recorded in participants who initiate one of the weight loss interventions. We will also record all AEs that are presumed to be or known to be related to gallstones up to 6 months.

data managementData will be recorded in a web-based data capture system (OpenClinica), which is hosted by the Primary Care Clin-ical Trials Unit of the University of Oxford. This system is customised and has an audit trail facility. Ranges and programmed validation checks are implemented in the system in order to aid reliable data entry.

statistical analysisThe primary and secondary outcomes will be assessed using an intention-to-treat analysis by an independent statistician. Each continuous outcome will be assumed to follow the normal distribution and be analysed by means of a linear mixed-effects model, adjusted for outcome at baseline. The model will include fixed-effects terms for

randomised group, visit, interaction between randomised group and visit, and baseline BMI (for non-weight outcomes only), and random effects to account for repeated measures on the same participant at 3, 6 and 12 months. No adjustment will be made for baseline BMI in the analysis of the weight outcomes due to its strong collinearity with baseline weight. A random effect will also be included for individual practice. An unstructured variance–covariance matrix will be specified between repeated measurements on the same individual, and the random effects for patient and practice will be assumed to be independent. The adjusted treatment effect together with the 95% CI and p value will be reported. The analysis will be performed using PROC MIXED in SAS Version 9.4. The proportion of participants who lose 5% and 10% of their initial weight at 12 months, respectively, will be presented, and the adjusted difference between the two arms and 95% CI will be reported. The binary outcome will be analysed by means of a logistic mixed-ef-fects model, adjusting for baseline BMI (fixed effect) and practice (random effect). The number needed to treat to achieve 5% or 10% weight loss, defined as the inverse of the absolute difference in proportions, will be reported if the differences between the treatment and control groups are statistically significant. A full statistical analysis plan will be prepared prior to any data analysis.

Qualitative substudyThe purpose of this study is to examine participants’ views of the programmes. In particular, we aim to examine the features that helped or hindered adherence to the programme and participants’ views of the behavioural support provided in the respective programmes. We will therefore purposively sample participants based on their responses to the satisfaction questionnaire, reflecting posi-tive, neutral and negative evaluations. Where possible, we will select participants to reflect both genders, socioeco-nomic status and ethnic group differences. We anticipate interviewing around 20 participants in the intervention group and 10 in the control group, but sampling will continue until saturation is reached, evidenced by no new themes occurring.

We will develop a semistructured topic guide for the interviews. The interviewer will encourage respondents to discuss their perceptions and experiences freely and in depth. The interview will set the context by asking about previous experience of weight management. Thereafter, we will ask for participants’ views on which component parts of their treatment they felt were effective and which they felt were not effective; thoughts about ability to continue to manage their weight when treatment has ended; and their views on medication adjustments where these occurred. The acceptability of the weight manage-ment treatment programmes and any preference they initially had for the total diet replacement programme or the usual care programme will be explored.

Data from participants will be collected in a confiden-tial, telephone interview, which will be audio-recorded.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960





Open Access

All interviews will be transcribed. To examine saturation, analysis will proceed concurrently with interviewing.

trial steering committeeAn independent Trial Steering Committee (TSC) will provide oversight of all matters relating to participant safety and data quality and value to the public. Due to the low risk nature of the DROPLET trial and that it is an open-label trial, the TSC also has the role of the Data Monitoring Committee, in addition to its role as the TSC. However, there are no early stopping rules, and all AEs are evaluated unblinded to allocation by the trial manage-ment group as well as the TSC.

The TSC includes an independent clinician, dietitian, statistician and two patient representatives. The TSC has reviewed the trial protocol, statistical analysis plan and the suitability of the proposed safety data to be collected. No interim analysis is planned for this trial due to the short recruitment period and low risk nature of the two dietary approaches.11 The trial may be subject to inspection and audit by the University of Oxford, under their remit as sponsor, the trial coordinating centre as the Sponsor’s delegate and other regulatory bodies.

Ethics and disseminationThe study protocol (V.4.0; 5 October 2016) was reviewed and approved by the South Central Oxford B REC Committee (Ref: 157/SC/0337). Any protocol modi-fications will be sent for review by the research ethics committee and will be amended at the trial registry.

It is planned that results will be disseminated to academic and health professional audiences via presen-tations at conferences and publication in peer-reviewed journals. Participants will be sent a summary of the trial findings at the time when the main article is published. If the trial shows this intervention is effective, the results will be communicated to policymakers and commissioners of weight management services through briefing papers summarising the main findings. We will also provide the results to all participants coincident with publication and disseminate the results to the public through a press release, regardless of what the results show.

Acknowledgements The low-energy total diet replacement programme including the formula meal replacement products will be provided by Cambridge Weight Plan, Northants, UK.

contributors SAJ and PA designed the study and secured the funding. NMA and ST helped to develop the protocol. NMA is the trial manger and AN is the trial statistician.

Funding This research is funded by research grants from Cambridge Weight Plan Ltdand NIHR Collaboration forLeadership in Applied Health Research and Care (CLAHRC) Oxford at Oxford HealthNHS Foundation to the University of Oxford. The sponsor of the trial isthe University of Oxford. The protocol was initiated and designed by the investigators who have nopersonal financial relationships with the Cambridge Weight Plan Ltd. Although Cambridge Weight Plan were consultedand commented on the protocol, the final decisions lay with the investigators.There are no restrictions on publication of results arising from this study andthe contract between the funder and the University ensures that the fundingbody will have no input into the decisions regarding publication.

competing interests SAJ and PA have led publicly funded trials in which the weight management intervention was provided free of charge by other commercial companies. They receive no personal financial benefits from these trials. NMA,

ST and AN have no competing interests. Cambridge Weight Plan, as the funder of this trial, is also the manufacturer of the nutritional products used in the trial and provided the products used in the trial free of charge to the participants.

Ethics approval NHS Research Ethics Committee (South Central Oxford B Committee).

Provenance and peer review Not commissioned; externally peer reviewed.

data sharing statement For access to the data set, a formal request should be sent to theDROPLET study group. The request will only be considered when the principalresults of the study have been published.

open Access This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http:// creativecommons. org/ licenses/ by/ 4. 0/

© Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

rEFErEncEs 1. Kelly T, Yang W, Chen CS, et al. Global burden of obesity in 2005 and

projections to 2030. Int J Obes 2008;32:1431–7. 2. World Health Organization. Obesity and overweight: fact sheet

Geneva: world Health Organization. 2016. http://www. who. int/ mediacentre/ factsheets/ fs311/ en/.

3. Whitlock G, Lewington S, Sherliker P, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet 2009;373:1083–96.

4. Guh DP, Zhang W, Bansback N, et al. The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 2009;9:88.

5. Butland B, Jebb SA, McPherson K, et al. Tackling obesities: foresight report: government Office for Science. 2007.

6. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403.

7. Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications over 15-year follow-up: the Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinol 2015;3:866–75.

8. Vermunt PW, Milder IE, Wielaard F, et al. Lifestyle counseling for type 2 diabetes risk reduction in Dutch primary care: results of the APHRODITE study after 0.5 and 1.5 years. Diabetes Care 2011;34:1919–25.

9. Hartmann-Boyce J, Johns DJ, Jebb SA, et al. Behavioural weight management programmes for adults assessed by trials conducted in everyday contexts: systematic review and meta-analysis. Obes Rev 2014;15:920–32.

10. Booth HP, Prevost TA, Wright AJ, et al. Effectiveness of behavioural weight loss interventions delivered in a primary care setting: a systematic review and meta-analysis. Fam Pract 2014;31:643–53.

11. Parretti HM, Jebb SA, Johns DJ, et al. Clinical effectiveness of very-low-energy diets in the management of weight loss: a systematic review and meta-analysis of randomized controlled trials. Obes Rev 2016;17:225–34.

12. (NICE) National Institute for Health and clinical excellence obesity: identification, assesment and management (NICE clinical guideline). London: NICE, 2006.

13. Jensen MD, Ryan DH, Apovian CM, et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice guidelines and the Obesity Society. Circulation 2014;129(25 Suppl 2):S102–38.

14. Lean M, Brosnahan N, McLoone P, et al. Feasibility and indicative results from a 12-month low-energy liquid diet treatment and maintenance programme for severe obesity. Br J Gen Pract 2013;63:115–24.

15. Leslie WS, Ford I, Sattar N, et al. The Diabetes Remission Clinical Trial (DiRECT): protocol for a cluster randomised trial. BMC Fam Pract 2016;17:20.

16. Ahern AL, Olson AD, Aston LM, et al. Weight Watchers on prescription: an observational study of weight change among adults referred to Weight Watchers by the NHS. BMC Public Health 2011;11:434.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

http://creativecommons.org/licenses/by/4.0/


http://dx.doi.org/10.1038/ijo.2008.102

http://www.who.int/mediacentre/factsheets/fs311/en/

http://www.who.int/mediacentre/factsheets/fs311/en/

http://dx.doi.org/10.1016/S0140-6736(09)60318-4

http://dx.doi.org/10.1186/1471-2458-9-88

http://dx.doi.org/10.1056/NEJMoa012512

http://dx.doi.org/10.1016/S2213-8587(15)00291-0

http://dx.doi.org/10.2337/dc10-2293

http://dx.doi.org/10.1111/obr.12220

http://dx.doi.org/10.1093/fampra/cmu064

http://dx.doi.org/10.1111/obr.12366

http://dx.doi.org/10.1161/01.cir.0000437739.71477.ee

http://dx.doi.org/10.3399/bjgp13X663073

http://dx.doi.org/10.1186/s12875-016-0406-2

http://dx.doi.org/10.1186/s12875-016-0406-2

http://dx.doi.org/10.1186/1471-2458-11-434





Open Access

17. British Heart Foundation. So you want to lose weight. for good. London: BHF Publications, 2005.

18. Hartmann-Boyce J, Aveyard P, Koshiaris C, et al. Development of tools to study personal weight control strategies: OxFAB taxonomy. Obesity 2016;24:314–20.

19. Hippisley-Cox J, Coupland C, Vinogradova Y, et al. Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2. BMJ 2008;336:1475–82.

20. NHLBI Obesity Education Initiative. The practical guide: identification, evaluation and treatment of overweight and obesity in adults: national Institutes of Health (NIH Publication Number 00-4084. 2000.

21. Patrick DL, Bushnell DM, Rothman M. Performance of two self-report measures for evaluating obesity and weight loss. Obes Res 2004;12:48–57.

22. EuroQol G. EuroQol--a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199–208.



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

http://dx.doi.org/10.1002/oby.21341

http://dx.doi.org/10.1136/bmj.39609.449676.25

http://dx.doi.org/10.1038/oby.2004.8



Confidential: For Review Onlyprotocol for a randomised controlled trialcare using total diet replacement products: a

primaryLow-Energy Treatment (DROPLET) in Doctor Referral of Overweight People to a

AveyardSusan A Jebb, Nerys M Astbury, Sarah Tearne, Alecia Nickless and Paul

doi: 10.1136/bmjopen-2017-0167092017 7: BMJ Open

http://bmjopen.bmj.com/content/7/8/e016709Updated information and services can be found at:

These include:

References http://bmjopen.bmj.com/content/7/8/e016709#ref-list-1

This article cites 17 articles, 4 of which you can access for free at:

Open Access

http://creativecommons.org/licenses/by/4.0/use, provided the original work is properly cited. See: others to distribute, remix, adapt and build upon this work, for commercialthe Creative Commons Attribution (CC BY 4.0) license, which permits This is an Open Access article distributed in accordance with the terms of

serviceEmail alerting

box at the top right corner of the online article. Receive free email alerts when new articles cite this article. Sign up in the

CollectionsTopic Articles on similar topics can be found in the following collections

(346)Nutrition and metabolism (724)General practice / Family practice

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:



BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

http://bmjopen.bmj.com/content/7/8/e016709

http://bmjopen.bmj.com/content/7/8/e016709#ref-list-1


http://bmjopen.bmj.com//cgi/collection/bmj_open_general_practice_family_practice

http://bmjopen.bmj.com//cgi/collection/bmj_open_nutrition_and_metabolism

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/



Confidential: For Review Only - bmj.com · Mean weight loss in the TRD groups was -10.7 kg, and...

Documents

Transcript of Confidential: For Review Only - bmj.com · Mean weight loss in the TRD groups was -10.7 kg, and...