Tier 3 specialist weight management service and pre‐bariatric multicomponent weight management programmes for adults with obesity living in the UK: A systematic review

Summary Background NHS England has recommended a multidisciplinary weight management services (MWMS—Tier 3 services) for patients requiring specialized management of obesity, including bariatric surgery, but clinical and measurable health‐related outcomes from these services remains fragmented. We therefore undertook a systematic review to explore the evidence base of effect on body weight loss and comorbidities outcomes of Tier 3 or UK pre‐bariatric MWMPs. Methods AMED, CINAHL, EMBASE, HMIC, MEDLINE, PsycINFO, PubMed, HDAS search and Google Scholar were searched from January 2000 to September 2017 in a free‐text fashion and crossed‐references of included studies to identify potential illegibility. Inclusion criteria were as follows: (a) published Tier 3 original study abstracts/articles; (b) intervention studies with before and after data; (c) studies that included any sort of MWMPs conducted on British residents with obesity; and (d) studies included T2DM measurements in a MWMPs. Results In total, 19 studies met the inclusion criteria. The total number of participants analysed was N = 11,735. Baseline accumulative average BMI was calculated at 42.54 kg/m2, weight 117.88 kg and waist circumference 126.9 cm. And at 6 months, 40.73 kg/m2, 112.17 kg and 120.3 cm, respectively. Secondary outcome variables were as improved with reduction in HbA1c, fasting blood sugars, insulin usage and blood pressure. Physical activity increased at 3 months then declined after 6 months with no significant changes in cholesterol levels. Conclusion Tier 3 and MWMPs have a short to mid‐ranged positive effect on obese patients (BMI ≥30 kg/m2) living in the UK regarding accumulated reduction in weight, glycaemic control, blood pressure and with subtle improvements in physical activity.


| BACKG ROUND
Morbid obesity is an increasing lifelong chronic condition that no country has yet succeeded to tackle. 1 In England, the prevalence of obesity is among the highest in Europe. 2 Two-thirds of adults are overweight and one in four are obese (Body Mass Index (BMI) of >30 kg/m 2 ). 3,4 McKinsey Global Institute reported that, second to smoking, obesity has the largest impact on the public health budget with an estimated annual cost to the United Kingdom ' s (UK) National Health Service (NHS) of £44.7b. 5 The importance of a range of obesity prevention initiatives comes from the increasing number of health complications and their related high cost. High Blood Pressure (BP), type 2 diabetes mellitus (T2DM), heart attacks, strokes, cancers and other health issues, for instance, are evidently associated to the conditions of being overweight or obese. 4 Even though bariatric surgical intervention is a proven effective approach for treating chronic obesity, access and eligibility for bariatric surgery remains low. 6 The reasons for this are multifactorial, but may include a lack of developed infrastructure for medical assessment and services, unclear referral procedures, as well as uncertainties regarding costs and long-term outcomes. 7 In England, the rate of bariatric surgical operations dropped by 31% between 2011-2012 and 2014-2015 (from 8794 to 6032 operations, respectively). 4 It is much worse in Scotland and Wales, and there is no NHS bariatric surgery performed in Northern Ireland. 8 Provision of bariatric intervention by NHS is, therefore, less than 1% of the national need. 8 In the UK, obesity is managed through a 4-levels tiered pathway.
Tier 1 and 2 are focused on universally environmental and population-wide prevention services. 4,9 Following this, individuals with more complex obesity and/or medical needs are considered for Tier 3 Multidisciplinary Weight Management Service (MWMS), 10 which may lead to a Tier 4 service for consideration of bariatric surgery. 4,11 Tier 3 MWMS consists of a (bariatric) physician, a dietitian, a specialist nurse and a clinical psychologist with access to physical therapy. 4 All adults identified with a BMI of ≥40 kg/m 2 , or ≥35 kg/m 2 with comorbidities are eligible for bariatric surgery following assessment and input from Tier 3 services. Tier 3, in this context, could also apply to a "Weight Assessment and Management Clinic" provided by primary or secondary care. 4 Within a Tier 3 service, strategies are implemented to make critical changes about eating and physical activity habits to improve health and identify risk factors so that the planned intervention addresses and improves all elements comprehensively. 4 Screening for hormonal or genetic causes of excessive weight as well as all related comorbidities and disabilities are conducted by the bariatric physician and each individual should have their own tailored lifestyle and healthful eating advice provided by a specialist dietitian. In addition, patients are screened for signs of psychiatric comorbidities due to the well-recognized link between obesity with many psychological disorders such as anxiety, depression, self-harm and suicidal behaviours, eating disorders (such as binge eating and bulimia nervosa), borderline personality disorders, alcohol and substance misuse, childhood adversity, among others. Patients with proven effort, an adequate timeframe prescribed by the multidisciplinary team, and with right weight criteria and medically optimized for surgery, will then be advised to progress towards the Tier 4 bariatric surgical intervention. 4,12 Although our understanding of the benefits of a Tier 3 service is growing-based on our appraisal of current literature, 11,4,13 current evidence remains fragmented and needs to be synthesized to produce a more comprehensive picture which will help to translate to a safe and cost-effective approach to the management of morbid obesity in the UK. We, therefore plan to explore the evidence base of effect magnitude on body weight loss in addition to other healthrelated outcomes of severely obese adults undergoing a Tier 3 or pre-bariatric Multicomponent Weight Management Programmes (MWMPs) in the UK. We include obese adults in the UK with a BMI ≥30 kg/m 2 who have been enrolled in a Tier 3 service or in any form of MWMP for losing weight. PubMed. An extended search was conducted using Google Scholar after reviewing additional studies that were included by Brown et al (2017) systematic review. 12 Terms used were related to "obesity" and "overweight" in conjunction with geographical restrictions to the UK (eg, England, Wales, Scotland, North Ireland). Terms related to MWMS, Specialist Weight Management (SWM) and Tier 3 (eg, weight management services, weight reduction programmes, weight management interventions, multidisciplinary weight loss initiatives and multicomponent weight loss schemes) were utilized on the titles and abstracts search. In addition, we screened reference sections of all included studies to identify potential illegible articles that meet the inclusion criteria of this review. See Figure 1 flow chart.

| Study selection
In this review, we use a similar pragmatic selection approach to Brown et al (2017). 12 Tier 3 studies for adults (18 years and over with no upper age limit) with a mean baseline BMI of ≥40 or ≥35 kg/m 2 with a comorbidity or ≥30 kg/m 2 with T2DM are included. In addition, all UK multicomponent pre-bariatric weight loss interventions that were planned and delivered for obese adults with BMI ≥30 kg/m 2 published since January 2000 until September 2017 were screened for potential inclusion. Inclusion criteria follow: (a) published Tier 3 original study abstracts and articles; (b) intervention studies with before and after data; (c) studies including any sort of MWMP planned for morbidly obese British residents; and (d) studies that included T2DM measurements in a MWMP for overweight adults. We excluded studies on children or adolescents and all studies conducted within non-British weight reduction intervention programmes. The decision to include or exclude studies was initially made based on the article title, then abstract and finally review the full-text article.

| Data extraction
We evaluated each of the included studies and extracted four data aspects: (a) descriptive to study design and intervention (Table S1 ); (b) sample size and demographic characteristics (Table S2 ); (c) assessed measurements (Table S3 ); and (d) health outcome records at baseline followed by points of time intervals (Tables S4-S9 ). For each segment, authors (year and country where intervention was delivered) are indicated.
In the descriptive of study design and intervention, we included the following: sitting, study design, aim, type of intervention, a brief description of intervention, inclusion and exclusion criteria, duration and lost-to-follow-up or drop-out data rate. In the demographics: We were not able to extract food intake records because of heterogeneity of stratification methods used by a number of studies in addition to concerns of related recall bias. We support Brown et al's (2017) decision regarding the difficulty in producing a meta-analysis in reviewing Tier 3 and all MWMPs due to heterogeneity. 12 The increased rate of patient drop-out and apparent risk of bias are also preventive factors to a meta-analysis. Thus, narrative synthesis is carried out.

| Risk of bias
All studies showed high risk in selection, performance, detection and attrition bias. This is because all included studies, except for the only RCT, 16 were designed as evaluation (before and after), retrospective analysis or uncontrolled prospective investigation. The risk of publication or reporting bias was low to unclear for all studies which may add to the overall reliability ( Figure 2 ). Attrition bias was evaluated high in consequence of the increased pattern of patients' drop-out rate; which was not fully investigated or discussed. P < 0.01). 16 In total, eight studies (42%) reported a change in BMI and/or weight at three months from their baseline, and the majority reported statistically significance weight reduction with an accumulative average of 114.48 kg. 16,[20][21][22]26,[29][30][31] Six studies (31%)

| Participants' characteristics
reported a percentage of participants who lost 5% or more of their initial weight (calculated mean: 22.95% of participants). 21,22,24,27,30,31 Jennings et al (2014) was the only study to report a 10% or more weight reduction rate among participants (3.6%). 21 Details on rates are summarized in Table 3 .

| Secondary outcome variables
The included studies reported secondary outcome variables in a heterogeneity that made tracking a set of health outcome variables problematic. Eight studies (42%) reported secondary health outcome variables at baseline: waist circumference, glycaemic control, lipids, BP and physical activity. 16,17,20,21,26,28,30,32 Details on baseline results are in Table 2 .  Table 2 ). 26 At six months, three studies (16%) reported further significant reduction in waist circumference with an average of 6.6 cm ( P < 0.001). 21,28,30 The waist circumference averaged at 120.3 cm.
T A B L E 3 Calculated average rates of participants who have lost weight covered and reported by the included studies (%) Drop-out rate increased to an average of 74.13% at two years point; ranging from 62.0% to 80.5%, as reported by 4 studies. 21,29,31,32 12 We have only excluded two studies from their selection, as one was of non-British origin and the other was comparing groups in post-bariatric. 13

| LIMITATI ONS
Studies published on Tier 3 and UK MWMPs are limited in number.
Yet, most if not all of included studies are of high risk of bias in terms of allocation sequence, allocation concealment, blinding, incomplete outcome data. The only RCT reviewed has shown a modest change in weight compared to all included studies. 16 The high rate of dropouts was present in most if not all included studies with inadequate reasoning. The majority have excluded non-completers' data from their final analysis.

| CONCLUSION
The reviewed evidence for the Tier 3 service and MWMPs suggests a short-to mid-ranged positive effect on British patients with obesity (BMI ≥30 kg/m 2 ) regarding accumulated reduction in weight, glycaemic control, BP and subtle improvement in physical activity.
The high drop-out rate might have contributed to limiting longer terms' progress in all positive results, especially those related to physical activity. More randomized trial investigations and drop-out explorations would improve overall reliability. Tier 3 service and MWMPs can assist obese adults living in the UK to lose weight and may improve their overall health status.

E TH I C S S TATE M E NT
Since this is a systematic review, ethical request is not applicable.

C O N F L I C T O F I N T E R E S T
No conflict of interest is declared for all authors.

AUTH O R S CO NTR I B UTI O N
Alkharaiji, M undertook data (study) collection, analysis and wrote the first draft of the manuscript. Anyanwagu, U acted as an independent second reviewer for study selection, supported analysis and supported the final draft of the manuscript. Donnelly, R. provided crucial academic input on the content of the manuscript and interpretation of data analysis. Idris, I. conceived the study, provided academic supervision, supported data analysis and interpretation, and wrote the final draft of the manuscript.

DATA ACC E S S I B I LIT Y
Supporting data are provided as supplementary information.