Isca Evidence

Background

People living with overweight or obesity are at an increased risk of developing cardiovascular disease, type 2 diabetes, atherosclerosis and a number of life-limiting or life-threatening conditions. Approximately one in four adults in the UK is obese, with 67% of men and 60% of women in England living with overweight or obesity. Hospital admissions related to obesity continue to place significant demands on the NHS.

Recent evidence has emerged about the effectiveness of a number of ‘anti-obesity’ drugs, particularly Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) such as semaglutide (aka Ozempic, Wegovy), liraglutide (Saxenda) and tirzepatide (Mounjaro). These drugs increase hormones called 'incretins', which help the body make more insulin, reduce the amount of sugar the liver produces and slow digestion speed. They also reduce appetite. Initially these drugs emerged as treatments for type-2 diabetes mellitus, however evidence for weight loss is also beginning to be well-established.

We were tasked with reviewing evidence about the effectiveness of GLP-1RAs for weight loss, understanding how we could support people to use them in the community, and what people’s experiences are of using or prescribing the drugs. We developed three pieces of work to explore this evidence:

• Work package 1 (this review): A scoping review of network meta-analyses on GLP-1 RAs for weight loss in people living with obesity
• Work package 2: A systematic review of digital components of interventions to support weight loss
• Work package 3: A scoping review of the experience of people using GLP-1 RAs, or those prescribing or supporting people using GLP-1 RAs

Equality, Diversity and Inclusion (EDI)

To incorporate a diverse range of experience and views into this work, the research team drew upon the knowledge and expertise of the PERSPEX team throughout the conduct of this work.

The inclusion criteria for this review reflect that the definition of obesity differs amongst certain minority ethnic groups. Where possible, this was considered alongside reported health inequalities data within the synthesis of our findings. We used the PROGRESS PLUS framework, a collection of EDI characteristics, to extract data on participant characteristics and consider the relevance of specific findings to different population groups where applicable. These characteristics may impact health outcomes and include indicators such as age, ethnicity/race, socioeconomic status and gender/sex.

The quantity, quality and findings of network meta-analyses evaluating the effectiveness of GLP-1 RAs for weight loss: a scoping review

Status

Complete

What do we want to know?

GLP-1 RAs are a class of drug originally developed to treat type 2 diabetes but are now increasingly used for weight loss, especially in people living with obesity. Network meta-analyses (NMAs) allow us to look at all the evidence and compare the effectiveness of all the GLP-1RAs, even when they haven’t been tested in the same trials. Despite an abundance of evidence about the effectiveness and safety of GLP-1 RAs for weight loss, NMAs are inconsistent in their quality and scope, and this is a fast-moving field.

We sought to: identify the most recent NMAs evaluating the effectiveness of GLP-1 RAs for weight loss; provide an overview of the quality and findings of this evidence, identify any gaps in the evidence to date; and consider the value of updating the most recent, comprehensive and high-quality NMAs.

Research questions

1. What is the quantity, quality and scope of recent NMAs evaluating the effectiveness of GLP-1 RAs for weight loss in overweight and obese adults?
2. What is the effectiveness of GLP-1 RAs for weight loss in overweight and obese patients, according to recent, high-quality NMAs?
3. What adverse events are associated with GLP-1 RAs in overweight and obese patients, according to recent, high-quality NMAs?

What we did

We conducted a scoping review of systematic reviews with NMAs. We searched literature databases to identify relevant published systematic reviews.

We included systematic reviews of NMAs which studied adults aged 18 or above with a BMI of 25 or more (23 or more for Asian, Middle Eastern, Black African and Afro-Caribbean populations). The lower threshold is used because these populations experience health risks associated with overweight and obesity at a lower BMI, and aligns with national guidelines. The intervention must have included delivery of at least one of 6 GLP-1 RAs: semaglutide, liraglutide, tirzepatide, exenatide, dulaglutide or lixisenatide. Any dose and any mode of delivery (for example, tablet or injection) were included. Interventions could be pharmacological only or multimodal, for example drug treatment alongside lifestyle modifications. The comparator was another GLP-1 RA, placebo or usual care. The primary outcome was a measure of weight loss such as change in weight or BMI from baseline.

We described the key characteristics and quality of the evidence, the effectiveness of GLP-1 RAs for weight loss, alongside any safety outcomes reported. The drugs already or soon to be licensed for weight loss in the UK (in November 2024, when we completed the review) were prioritised in the synthesis (these are semaglutide, liraglutide and tirzepatide).

Public & Patient Involvement & Engagement (PPIE)

PERSPEX contributed to the protocol by reviewing a Plain English version. Discussion with the group highlighted safety and maintenance of weight loss as key areas of interest to patients and carers and contextualised the use of weight loss drugs within a broader societal context, with concerns about industry sponsorship. These discussions also fed into the interpretation and writing of the report.

What did we find?

Evidence

359 references were screened with 22 meeting the criteria for inclusion.

12 reviews investigated weight loss as a primary outcome, five focussed on HbA1C and three looked at safety outcomes. 12 reviews specifically investigated participants with Type 2 diabetes (T2DM) and seven specifically investigated participants with overweight or obesity.

Review characteristics varied greatly including the number of trials included and sample sizes. Nine reviews included semaglutide, 11 included liraglutide, two included tirzepatide and eight included exenatide. Two reviews each included dulaglutide and lixisenatide. Nine reviews reported safety outcomes.

Eight papers conducted analyses based on PROGRESS-PLUS criteria, a framework of EDI criteria which identify characteristics that influence health opportunities and outcomes. Most of these looked at ‘other personal characteristics’, such as other health conditions.

We used established tools to critically appraise the quality of the evidence found, which provides an indication of how much confidence we can place in the findings of the reviews. These tools showed moderate confidence in most of the reviews we included. Most of the NMAs were missing data on some of the drugs we were interested in and some included poor quality studies.

Results

GLP-1 RAs vs placebo/usual care

Compared with placebo or usual care, all GLP-1 RAs were associated with statistically significant increased weight loss at a minimum of one time point.

When combining data from multiple time points, the largest effects were seen for subcutaneous semaglutide 2.4mg with 11.5-12.5kg greater weight loss than placebo; tirzepatide combined doses was also effective, with over 8.5kg more weight loss than placebo.

Comparative effects of different GLP-1 RAs

Tirzepatide and semaglutide were the most effective drugs for weight loss, however there were no NMAs comparing tirzepatide and semaglutide 2.4 mg directly so it is not clear which of these may be better than the other. At six months, tirzepatide 15mg was more effective than semaglutide 1.0 mg by up to 4.4 kg. At combined time points (analysing the data collected at several points in time all together e.g. at 3, 6 and 12 months after starting medication), semaglutide 2.4 mg was associated with 5.8 and 8.7 kg greater weight loss than semaglutide 1.0 (data from two NMAs), suggesting that semaglutide 2.4 mg may be more effective for weight loss than tirzepatide.

Safety

The drugs associated with the greatest weight loss, tirzepatide and semaglutide 2.4 mg, were generally associated with an increased risk of safety issues (AEs, adverse events, and SAEs, serious adverse events) compared to placebo. Despite these concerns, data came from a single trial, with wide confidence intervals for all outcomes indicating uncertainty about the findings.

Other doses of semaglutide were associated with some increased risk of discontinuation, but not of SAEs.

Limitations to evidence

Systematic reviews with NMAs are considered amongst the highest levels of evidence, but scrutiny of the NMAs included in this scoping review reveals several limitations. It was often unclear which trials were included in each NMA, or how many informed direct comparisons between drugs or drugs and placebo/usual care. These alongside other limitations, such as the tendency to combine both multiple doses of drugs and outcomes at different time points, and frequent inclusion of studies at high risk of bias, restricts our understanding and confidence in the evidence.

The majority of trials of GLP-1 RAs (and two of the prioritised reviews) were funded by the pharmaceutical companies which make the drugs, and there may be conflicts of interest associated with this. It was fervently noted by PERSPEX, our PPIE group, that the role of industry sponsorship should be considered, particularly with respect to the possible diversion of funds and focus from societal/public health interventions to drug-based interventions.

Future research

More evidence is needed comparing semaglutide 2.4 mg with tirzepatide, and to explore longer-term safety and effectiveness. Despite an abundance of recent NMAs, findings are inconsistent, particularly for safety outcomes, and the methodological rigour of future NMAs could be improved.

Sharing our findings

• The protocol for this work is available here.
• The report has been accepted for publication by the NIHR Journals Library and this page will be updated with the details of publication when available.
• Please note a living NMA, a process of regularly updating the steps of an NMA to include the most current evidence, is ongoing on this topic.   

Background

People living with overweight or obesity are at an increased risk of developing cardiovascular disease, type 2 diabetes, atherosclerosis and a number of life-limiting or life-threatening conditions. Approximately one in four adults in the UK is obese, with 67% of men and 60% of women in England living with overweight or obesity. Hospital admissions related to obesity continue to place significant demands on the NHS.

Recent evidence has emerged about the effectiveness of a number of ‘anti-obesity’ drugs, particularly Glucagon -like peptide-1 receptor agonists (GLP-1 RAs) such as semaglutide (aka Ozempic, Wegovy), liraglutide (Saxenda) and tirzepatide (Mounjaro). These drugs increase hormones called 'incretins', which help the body make more insulin, reduce the amount of sugar the liver produces and slow digestion speed. They also reduce appetite. Initially these drugs emerged as treatments for type-2 diabetes mellitus, however evidence for weight loss is also beginning to be well-established.

We were tasked with reviewing evidence about the effectiveness of GLP-1RAs for weight loss, understanding how we could support people to use them in the community, and what people’s experiences are of using or prescribing the drugs. We developed three pieces of work to explore this evidence:

• Work package 1: A scoping review of network meta-analyses on GLP-1 RAs for weight loss in people living with obesity
• Work package 2 (this review): A systematic review of digital components of interventions to support weight loss
• Work package 3: A scoping review of the experience of people using GLP-1 RAs, or those prescribing or supporting people using GLP-1 RAs

Equality, Diversity and Inclusion (EDI)

To incorporate a diverse range of experience and views into this work, the research team drew upon the knowledge and expertise of the PERSPEX team throughout the conduct of this work.

The inclusion criteria for this review reflect that the definition of obesity differs amongst certain minority ethnic groups. Where possible, this was considered alongside reported health inequalities data within the synthesis of our findings. We used the PROGRESS PLUS framework, a collection of EDI characteristics, to extract data on participant characteristics and consider the relevance of specific findings to different population groups where applicable. These characteristics may impact health outcomes and include indicators such as age, ethnicity/race, socioeconomic status and gender/sex.

Which aspects of digital interventions to support weight loss programmes are associated with success? Systematic review with component network-meta-analysis

Status

Complete

What do we want to know?

While the effectiveness of digital interventions to support the delivery of GLP-1 RAs in the community has not been established, in large part due to the recency of these drugs as a treatment option, there is an abundance of literature on the use of digital technology to support weight loss interventions in general. For example, several systematic reviews have evaluated the effectiveness of interventions using mobile phones (or ‘mHealth’) and/or other electronic (or ‘eHealth’) and digital technologies to lose weight.

We sought to identify components of digital support for weight loss interventions that are most likely to be effective in supporting patients to achieve weight loss goals.

Research questions

1. What is the breadth and scope of evidence for the effectiveness of weight loss interventions which include digital components?
2. How do we categorise the nature and content of the digital components of included interventions?
3. What is the relative effectiveness of different digital intervention components for digitally led weight loss interventions?

What we did

We conducted a systematic review searching bibliographic databases and trial registries, with forward and backward citation searching of prioritised reports. Abstract screening of 1268 records lead to 99 trials meeting our inclusion criteria (all randomised controlled trials). Any type of weight loss intervention was eligible, so long as there was a digital component associated with its delivery, and it was not delivered as part of secondary or tertiary care weight loss management. Only RCTs and cluster RCTs were included.

From the 99 included trials we prioritised the analysis of 68 trials that we identified as being predominantly digital to improve the clarity of the findings. We performed an intervention component analysis (ICA) on 151 trial arms from the 68 trials with data from over 14,500 participants, to identify the key digital components of interventions. We then conducted a component network meta-analysis (NMA) to ascertain which of the components were associated with successful outcomes.

Public & Patient Involvement & Engagement (PPIE)

PERSPEX contributed substantially to this review throughout the process. Members shared their knowledge of digital support and highlighted digital exclusion and health inequalities as areas of concern. Feedback was then sought on the protocol; this led to changes to increase clarity and improve readability. PERSPEX reviewed the search strategy by reading an initial version and making suggestions for additional search terms. PERSPEX had further input on the early synthesis by reviewing the initial categories identified as part of the Intervention Component Analysis, to describe the different ways that digital technologies were used within the included studies.

What did we find?

Evidence

The majority of studies (52) evaluated weight loss as an outcome. With regards to EDI characteristics: all 68 prioritised studies reported age and gender/sex of participants; 38 reported race/ethnicity; 49 reported educational attainment; 19 reported employment and four reported occupation status; 18 reported income; five reported a deprivation/poverty score; two reported socioeconomic status; and one reported rural residency.

A roughly equal number were categorised as at low risk of bias, having some concern over risk of bias, or at high risk of bias.

Of 151 trial arms coded, 90 arms were digital only, 28 were digitally led, 12 were treatment as usual plus additional printed information but no counselling (TAU+), 10 were treatment as usual (TAU) and 11 were waitlist.

The ICA identified nine components facilitated by digital technology: goal setting, self-monitoring, peer support, reminders, feedback, specialist contact, information/education, competition/challenge, and incentives/rewards (see Table 1 below). These components were delivered or facilitated by a range of digital technologies, with websites and web-based platforms, applications (i.e. for smartphones), short messaging services (SMS) and emails being the most common.

Categories describing and grouping the components were discussed with PERSPEX, to check sense and to identify categories that might be missing, or which could be collapsed. These discussions led to the addition of the ‘incentives and rewards’ category, and consideration of whether the intervention was tailored or customised in any way for participants. We also coded whether the intervention more broadly targeted diet, exercise or both; and whether any face-to-face components were delivered individually or in groups.

Table 1. Descriptions of digital components of interventions

Results

We looked at absolute weight loss, absolute change in BMI, weight loss as a percentage of baseline weight and responders (the proportion of people achieving more than 5% weight loss from their baseline). The component NMA showed that no single digital component stood out as being singularly linked to effectiveness at 6-month or 12-month time points. There were some statistically significant findings for specialist contact, information and feedback at 6 months, but these were not consistent across time points and outcomes. However, there were three ‘best bet’ components which were consistently associated with favourable, if not always statistically significant, effect estimates across outcomes. These were: provision of information and education; contact with a specialist; and provision of incentives or rewards. An exploratory analysis modelling the potential effectiveness of an intervention comprising these three components together suggested significant 6-month impacts, and indicative 12-month impacts on absolute weight loss, with significant reduction in BMI at 12 months.

Limitations to evidence

As anticipated, we did not identify any relevant trials using digital approaches to support patients using specific weight loss drugs. As such, the validity of our findings in this context is restricted. However, our findings were not characterised by concerning levels of statistical heterogeneity despite the wide variation in intervention designs, which suggests that different studies found similar results for the same outcomes. We can be reassured that consistencies in patient experiences and behaviours across weight loss interventions should mean our findings can translate to patients using weight loss drugs, especially as they should be prescribed alongside changes to diet and exercise.

The component NMA used coding that captured the presence of a component, not its nature (i.e. frequency, intensity etc). The examples we highlight for each ‘best bet’ component demonstrate variation in their delivery, and it can be expected that different modes of delivery will yield variations in effectiveness. As such, further investigation is required to determine the optimal ingredients of future interventions, which may vary between individuals.

Many of the included trials imposed an upper limit of 39.9 kg/m2 for BMI for inclusion into the study (the upper range for patients living with obesity class 2). Therefore, the needs of patients living with obesity class 3 are not well represented.

Future research

Further evidence is needed to develop simplified digital interventions to support people with weight loss. These should provide high quality digital information, facilitate contact with a specialist e.g. a dietitian, and may include incentives and rewards for progress and intervention compliance. Trials with long-term follow up, and trials designed for patients specifically receiving weight-loss drugs are required.

Sharing our findings

• The protocol for this work is hosted on PROSPERO with the ID CRD42023493254, available here
• Our paper has been accepted for publication in the Journal of Medical Internet Research, and this page will be updated when it is published.

Background

People living with overweight or obesity are at an increased risk of developing cardiovascular disease, type 2 diabetes, atherosclerosis and a number of life-limiting or life-threatening conditions. Approximately one in four adults in the UK is obese, with 67% of men and 60% of women in England living with overweight or obesity. Hospital admissions related to obesity continue to place significant demands on the NHS.

Recent evidence has emerged about the effectiveness of a number of ‘anti-obesity’ drugs, particularly Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) such as semaglutide (aka Ozempic, Wegovy), liraglutide (Saxenda) and tirzepatide (Mounjaro). These drugs increase hormones called 'incretins', which help the body make more insulin, reduce the amount of sugar the liver produces and slow digestion speed. They also reduce appetite. Initially these drugs emerged as treatments for type-2 diabetes mellitus, however evidence for weight loss is also beginning to be well-established.

We were tasked with reviewing evidence about the effectiveness of GLP-1RAs for weight  loss, understanding how we could support people to use them in the community, and what people’s experiences are of using or prescribing the drugs. We developed three pieces of work to explore this evidence:

• Work package 1: A scoping review of network meta-analyses on GLP-1 RAs for weight loss in people living with obesity
• Work package 2: A systematic review of digital components of interventions to support weight loss
• Work package 3 (this review): A scoping review of the experience of people using GLP-1 RAs, or those prescribing or supporting people using GLP-1 RAs

Equality, Diversity and Inclusion (EDI)

To incorporate a diverse range of experience and views into this work, the research team drew upon the knowledge and expertise of the PERSPEX team throughout the conduct of this work.

The inclusion criteria for this review reflect that the definition of obesity differs amongst certain minority ethnic groups. Where possible, this was considered alongside reported health inequalities data within the synthesis of our findings. We used the PROGRESS PLUS framework, a collection of EDI characteristics, to extract data on participant characteristics and consider the relevance of specific findings to different population groups where applicable. These characteristics may impact health outcomes and include indicators such as age, ethnicity/race, socioeconomic status and gender/sex.

What are the experiences, views and perceptions of patients, carers and clinicians of glucagon-like peptide-1 receptor agonists (GLP-1 RAs)? A scoping review

Status

Completed

What do we want to know?

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), such as semaglutide and liraglutide, and the dual gastric inhibitory polypeptide (GIP) and GLP-1 RA, tirzepatide, are drugs authorised in the UK by the National Institute for Health and Care Excellence (NICE) for the management of obesity and/or type 2 diabetes mellitus (T2DM). Qualitative research provides the opportunity to gain insight into the perceptions of long-term efficacy, safety and acceptability of GLP-1 RAs, as well as the benefits and challenges associated with introducing this type of intervention for patients, carers and health service providers.

Initial scoping searches indicated the presence of a small body of primary research which focuses on the views and/or experiences of patients or clinicians on the use of GLP-1 RAs to help control diabetes and/or promote weight loss. However, given the rapidly developing research landscape pertaining to this topic and changes in how these types of medications are prescribed and delivered, it is not certain the extent to which this existing evidence-base is useful to inform the current or future configuration and delivery of services.

We aimed to conduct a scoping review to answer the research question:

What is the quantity, nature and key characteristics of primary evidence which explores the experiences, views and perceptions of patients, carers, and clinicians regarding the use of GLP-1 RAs for any indication?

What we did

We searched bibliographic databases alongside sources of grey literature. References from potentially relevant systematic reviews were also checked.

We sought the views of patients, carers or clinicians of any age, with experience of GLP-1 RAs, whether this was being prescribed GLP-1 RAs, supporting patients who have been prescribed them, prescribing them to patients, or supporting patient access to, or services delivering them. The drugs could be prescribed for any indication, in any health care or community setting. Only qualitative or mixed-methods primary studies were eligible.

Public & Patient Involvement & Engagement (PPIE)

In this scoping review, PERPSEX checked the initial search strategy and made suggestions for additional search terms. A plain English version of the draft protocol was sent to members to read prior to the June 2024 online PERSPEX meeting. This document informed discussions about the review, particularly the included population, and whether this should include only those prescribed GLP-1 RAs or those taking them through other means. This directly informed development of the review protocol. The preliminary findings of the review were discussed at the September 2024 online meeting.

Over the course of 12 months PERSPEX members and researchers discussed GLP-1 RAs at 6 online meetings and through various written and video material communications. These discussions highlighted safety as a key area of interest to patients and carers and contextualised the use of these drugs within a broader societal background, surfacing concerns about industry sponsorship. These discussions foregrounded patient, and carer matters for the review team and informed preliminary scoping work to define the research question for this review.

What did we find?

Evidence

Of 1545 records screened at title and abstract stage, 25 eligible studies meeting our inclusion criteria were found to be reported in 26 papers.

Of these studies: 11 spoke directly with patients/public; four with healthcare professionals (HCPs) or clinicians; three spoke to both patients/public and healthcare workers; and seven were social media based.

Most studies collected data through interviews with some additionally using focus groups, a survey or ethnographic observation of GP consultations. Ten studies undertook thematic analysis, four conducted content analysis and two used grounded theory.

With regards to the phenomenon of interest, of seven studies interviewing HCPs, six discussed prescribing or deprescribing. Of 14 studies speaking to patients/public, eight discussed taking GLP-1 RAs as a prescription, three discussed purchasing or choosing GLP-1 RA, two investigated experiences of taking the drug within a clinical trial, and one was specific to taking semaglutide as part of a specialist paid weight management intervention.

Results

The studies discussing prescribing and deprescribing of GLP-1 RAs by healthcare professionals highlighted some of the complexities, practicalities and costs involved in prescribing the medications and the considerations for adverse events. Two studies reported some uncertainties around the use of newer drugs including GLP-1 RAs, while another reported a general willingness and trust in the medication and familiarity with its previous use as a diabetic medication.

Studies investigating the purchasing or choosing of GLP-1 RAs by patients or the public discussed factors, thoughts and preferences in identifying reasons for choosing GLP-1 RAs medication.

Studies concerning patients or the public taking GLP-1 RAs were carried out in healthcare contexts, although the services that patients were seen within were generally unclear. Several of the overall findings did report some positive changes and satisfaction with GLP-1 RAs as a treatment option when compared to other medications. Common themes were identified around treatment practicalities, the want for an easy simple regime, routes of administration, treatment adherence, side effects and social and emotional factors. The impact on weight whilst taking these drugs for diabetic management was a key concept identified in six studies. Two studies highlighted follow-up and ongoing support and explanations with healthcare providers about clinically relevant information that would aid decision making, as potentially important factors for increased motivation and continuation of treatment.

Studies focussing on social media highlighted strong public interest in GLP-1 RAs and raised discussion points around public perceptions or experiences of use, however, the origin of social media posts and therefore the context of drug use was unclear.

There was no evidence about the views, experiences or perceptions of carers regarding the use of GLP-1 RAs.

Limitations to evidence

This review was limited by the quantity and quality of primary qualitative research exploring patient, clinician and especially carer experiences of using of GLP-1 RAs for any indication and especially for weight loss. There were limitations with data collection and data analysis techniques in the included studies, with often a limited focus on GLP-1-RAs, reducing the value of in-depth interpretation of findings. The lack of transparency around the potential impact of conflicts-of-interest on the findings is a further limitation with this body of evidence to date.

Future research

There is an urgent need for high-quality qualitative evidence to explore the experiences of patients, carers and clinicians with respect to the use of GLP-1 RAs, particularly those indicated for weight loss. This type of qualitative evidence is crucial to facilitate decision making, understand patients’, carers’ and clinicians’ support needs, and inform service development. Qualitative evaluations could be conducted alongside new randomised controlled trials, or as stand-alone studies seeking to recruit a diverse range of participants. Qualitative methods that elicit rich experience data would be particularly valuable.

Once a body of primary evidence is established, qualitative evidence synthesis is required to inform commissioning/clinical practice/government guidelines. Until a more complete understanding of patient, carer and clinician experiences’ is obtained, some caution should be applied to the commissioning of services supporting the delivery of GLP-1 RAs.

Sharing our findings

• The protocol for this work is registered on Zenodo.
• This report has been published by Health Expectations and is available here.