Why The WHO Shouldn't Grant Diet Drug Request To Be Added To Essential Medicine List - Part 3
This is the Weight and Healthcare newsletter! If you like what you are reading, please consider subscribing and/or sharing!
In part 1 we talked about a request that has been submitted for the World Health Organization (WHO) to add diet drugs (specifically GLP1 agonists like Novo Nordisk’s Saxenda and Wegovy) to their list of “essential medicines.” We discussed who was making this request and the justification that they were using. In part 2 we took a deeper dive into the research that they used to try to support this request, and in this final installment, we will look at the research around efficacy, harm, and cost-effectiveness.
First I’ll offer a summary for each issue and then I’ll give a breakdowns of the research that they cite. Just a quick reminder that this request is asking the World Health Organization (WHO) to add these drugs to their list of “essential medications” globally.
Before we get into the sections, I want to mention two overarching issues that are found throughout the entirety of this request and the studies that are used to support it.
First, in general, a belief has been fomented (predominantly by those in the weight loss industry) that being higher-weight is so terrible then it’s worth “throwing anything at the problem.” This leads to acceptance of poor, short-term, and/or incomplete data as “good enough” to foist recommendations onto higher-weight people, which means that part of weight stigma in healthcare is that higher-weight people are afforded less right to ethical, evidence-based medicine than thinner people.
Second, is clinging to correlation (without any mechanism of causation) when it comes to weight, health, and health outcomes, including the abject failure to consider confounding variables. So throughout these studies “being higher-weight is associated with [health issue(s)]” stated uncritically in support of weight loss interventions. There is an utter failure to explore the idea that the reason for the outcome differences is not weight itself but, instead, exposure to weight stigma, weight cycling (which these medications actually perpetuate by their own admission,) and healthcare inequalities.
Issues with research supporting effectiveness, harms, and benefits
Study Duration:
This is the main issue. While there was one study that went up to 106 weeks, the vast majority of the studies are between 14 and 56 weeks. We know that these drugs can have significant, even life-threatening side effects (earning them the FDA’s strongest warning.) 14-56 weeks is not not nearly enough time to capture the danger of long-term effects, or to capture long-term trends around weight loss/weight regain.
Study Population
Many of the studies included have small samples. Many have study populations are overwhelmingly white, which is a huge issue when making a global recommendations.
Small effect and overlap
Many of the studies show only a bit of weight loss (often 15lbs or less) and often there was overlap in weight lost between the treatment group and the placebo group. Even using the “ob*sity” construct that this request is based on, for many people, this amount of weight loss wouldn’t even change their “class” of “ob*sity.”
Failure to capture adverse events
Much of the research they use to support their claims of safety didn’t actually capture individual adverse events or serious adverse events. Often they only captured subjects who reported leaving treatment due to side effects.
Issues with research supporting cost effectiveness
The cost-effectiveness analyses they cite are based on Quality Adjusted Life Years (QALYs). This is a measurement of the effectiveness of a medical intervention to lengthen and/or improve patients’ lives.
The calculation for this is [Years of Life * Utility Value = #QALY]
So if a treatment gives someone 3 extra years of life with a Health-Related Quality of Life (HRQL) score of 0.7, then the treatment is said to generate 2.1 [3 x 0.7] QALYs.
This is a complicated and problematic concept that deserves its own post sometime in the future, but looking just at this request I think it’s important to note that they are working on two main unproven assumptions:
1. That being higher weight causes lower health-related quality of life and/or shorter life span (rather than any lower HRQL being related to experiences that higher-weight people have including weight stigma, weight cycling, healthcare inequalities et al.)
2. That this treatment induces weight loss and/or health benefits that increase the life span and/or health-related quality of life of those who take it.
I don’t believe either of these assumptions are proven by the material cited in the request to the WHO. Specifically, it’s very possible that it’s not living in a higher-weight body, but rather the experiences that higher-weight people are more likely to have (weight stigma, weight cycling, healthcare inequalities) that impact their HRQL.
Further, the short-term efficacy data available (and Novo Nordisk’s own admission about high rates of regain) fall far short of proving any assumptions about these drugs ability to actually improve or extend life. Further, the failure of the literature to adequately capture negative side effects of the drugs, both short and long-term, means that this calculation cannot be properly made.
Incremental Cost-Effectiveness Ratio (ICER)
ICER is how QALYs are turned into a monetary value. It is calculated by dividing the difference in total costs by the difference in the chosen measure of health outcome or effect.
[(Cost of intervention A -Cost of Intervention B) / (Effectiveness of Intervention A – Effectiveness of Intervention B)]
The result is a ratio of extra cost per extra unit of health effect of a more vs less expensive treatment which can then be measured in QALYs.
Again, this is worthy of its own post because there are all kinds of ethical issues around things like how we value life, how we define “healthy” and the ethics of determining whether or not prolonging someone’s life is “cost effective.” I’m not going to do a deep dive into that today, but I do want to note that it is a serious issue in these kinds of calculations.
In this specific case, even if one was to get past the ethical issues, an accurate calculation is impossible to make on both of the measures of the equation.
Cost of these drugs varies wildly between countries and sometimes within countries because, for example, Novo Nordisk is a for-profit corporation whose goal is to create as much profit as possible. Per the WHO request letter, the monthly cost of liraglutide is $126 in Norway and $709 in the US. Semaglutide is $95 per 30 days in Turkey, but $804 per 30 days in US.
When it comes to effectiveness of the treatment, again, there is virtually no long-term data. We do know that in Novo Nordisk’s own studies, weight is regained rapidly and cardiometabolic benefits are lost when the drugs are discontinued and even when people stay on the drugs, weight loss levels off after about a year, at 68 weeks weight cycling begins, and at 104 weeks (when follow-up ended) weight was trending up. It’s possible that these drugs are utterly ineffective over the long-term and/or that the prevalence of long-term side effects renders any treatment effects moot. We simply do not know.
I do not think that this is a remotely appropriate basis from which to request that these drugs be declared globally essential by the WHO.
Here are the citation breakdowns. These are not deep dives since there are enough issues with the research on a simple surface analysis.
Breakdowns of evidence of comparative effectiveness
Effects of liraglutide in the treatment of ob*sity: a randomised, double-blind, placebo-controlled study, Astrup et al.)
This is a 20-week study funded by Novo Nordisk. It included 564 people on various doses of liraglutide and a placebo group who didn’t get the drug and a group on orlistat. There were no more than 90-98 people in each group.
The study explains “Participants on liraglutide lost significantly more weight than did those on placebo” by which they meant that those on the highest dose of liraglutide lose about 9.7lbs more than those on the placebo over the 20 weeks.
III LEAD studies
These are four studies that look at liraglutide in combination with other drugs for the treatment of Type 2 Diabetes that also included some information on weight changes. One was 52 weeks, the others were 26, the maximum amount of weight lost was only about 5lbs.
The first [Liraglutide, a once-daily human GLP-1 analogue, added to a sulphonylurea over 26 weeks produces greater improvements in glycaemic and weight control compared with adding rosiglitazone or placebo in subjects with Type 2 diabetes (LEAD-1 SU), Marre et al] was a study that looked at the efficacy of adding liraglutide or rosiglitazone 4 to glimiperide in subjects with Type 2 Diabetes to test effects on blood sugar and body size.
The study followed 1041 adults for 26 weeks. The study found that those on .6mg of liraglutide gained 0.7kg, those on 1.2mg gained 0.3kg, and those on 1.8mg of liraglutide lost 0.2kg, while those on placebo lost 0.1kg.
The second [Efficacy and safety comparison of liraglutide, glimepiride, and placebo, all in combination with metformin, in type 2 diabetes: the LEAD (liraglutide effect and action in diabetes)-2 study. Diabetes Care, 2009. 32(1): p. 84-90. Nauck, M., et al.,]
looked at the efficacy of adding liraglutide to metformin therapy for those with Type 2 Diabetes. They found that over the 26-week study those on liraglutide lost 1.8 ± 0.2, 2.6 ± 0.2, and 2.8 ± 0.2 kg for 0.6, 1.2, and 1.8 mg doses. Those on placebo lost 1.5 ± 0.3kg.
The third [Liraglutide versus glimepiride monotherapy for type 2 diabetes (LEAD-3 Mono): a randomised, 52-week, phase III, double-blind, parallel-treatment trial. Lancet, 2009. 373(9662): p. 473-81. Garber, A., et al.,]
This was a study of the comparative effectiveness of Liraglutide versus glimepiride for type 2 diabetes, with small weight loss as an ancillary finding. Those in the liraglutide group lost an average of 2kg.
The final study [Efficacy and safety of the human glucagon-like peptide-1 analog liraglutide in combination with metformin and thiazolidinedione in patients with type 2 diabetes (LEAD-4 Met+TZD), Zinman et al.] was a 26-week study with 533 total subjects. The goal was to study the efficacy of liraglutide when added to metformin and rosiglitazone for people with type 2 diabetes. They found that those on liraglutide lost between 0.7 and 2.3kg (1.5lbs to 5.1lbs) in 26 weeks.
Meta-Analyses and Systematic Review Findings
Efficacy of Liraglutide in Non-Diabetic Ob*se Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Barboza, J.J., et al.,
None of the included studies were more than 56 weeks and one was only 14 weeks. One had as many as 3731 subjects, but one had only 40. Some had body weight loss as a primary outcome, but some did not. Maximum doses ranged from 1.8 to 3.0mg. The mean body weight reduction was 3.35 kg (7.4lbs) but in one study there was no difference in weight loss. The maximum difference was 6.3kg (13.9lbs)
They also refer to Iqbal et al which we discussed in part 2.
Effects of glucagon-like peptide-1 receptor agonists on weight loss: systematic review and meta-analyses of randomised controlled trials. Vilsbøll, T., et al.
The included studies are between 20 and 53 weeks long, and include some of the studies they already cited individually above. Of the 25 included studies only 3 had “ob*sity” as the main inclusion criteria, the rest were Type 2 Diabetes.
The mean weight loss for those on the highest dose of the drug was between 0.2kg and 7.2kg. For those in the control group it was 2.9 kg, so there was actually overlap between the treatment and placebo groups.
Summary of evidence of safety and harms
They begin with the claim “The safety profile of GLP-1 receptor agonists is also well studied”
To support this they cite:
Efficacy and Safety of Liraglutide 3.0 mg in Patients with Overweight and Ob*se with or without Diabetes: A Systematic Review and Meta-Analysis, Konwar, M., et al.,
This included 14 total studies, many of which the authors of the WHO request had cited individually and were included in other systematic reviews and meta-analyses above. The smallest study included 19 people, the largest included 2,487. The total number of subjects was 4,142.
Their conclusion was “Liraglutide in 3.0 mg subcutaneous dose demonstrated significant weight reduction with a reasonable safety profile for patients with overweight or ob*sity regardless of diabetic status compared to placebo.”
Their methodology says that they omitted studies from analysis due to “short duration.” They included studies that had a minimum of 12 weeks and a maximum of 56 weeks of follow-up.
While they included 14 studies, only 11 of them actually included information about adverse events.
In terms of adverse effects (AEs,) they found that the pooled estimate of nine studies in nondiabetic patients and two studies in diabetic patients revealed a significant proportion of patients experiencing the adverse events in liraglutide 3.0 mg group when compared with placebo., and the pooled estimate of the eleven studies showed that liraglutide 3.0 mg had higher risk of AEs compared to placebo.
When it came to “serious adverse events” they found that there was a similar risk level between the drug and placebo groups, but remember that’s for only 12 to 56 weeks, and Novo Nordisk is recommending that people take these drugs for the rest of their lives. A few months to a little over a year is not enough time to capture long-term serious adverse events.
The efficacy and safety of liraglutide in the ob*se, non-diabetic individuals: a systematic review and meta-analysis. Zhang, P., et al.,
This included five RCTs (which were included in various of the above systemic reviews and meta-analyses) ranging in follow-up from 14 to 56 weeks.
The only adverse event information captured was the number of people who withdrew from treatment due to adverse events (which they found was similar between drug and placebo) and nausea (which was experienced more by people on the drug.)
So, in addition to being short in duration, this was far from a comprehensive list of side effects. They made no attempt to capture serious adverse side effects and their short-term nature would have made this difficult anyway.
Association of Pharmacological Treatments for Ob*sity With Weight Loss and Adverse Events: A Systematic Review and Meta-analysis. Khera, R., et al.
This looked at weight loss and adverse events with a number of different weight loss drugs. Interestingly liraglutide did not show the highest amount of weight loss but was associated with the highest odds of adverse event–related treatment discontinuation. It should also be noted that high drop-out rates of 30-45% plagued all of the trials which the study authors admit means that “studies were considered to be at high risk of bias.“
Given that those who drafted the WHO request are asking that these drugs be considered essential globally, it is disappointing that they included this study and didn’t bother to mention this issue in their written request.
This included 28 RCTs (most of which were included in other citations above) and only 3 that included liraglutide. They didn’t capture individual adverse events, but only “Discontinuation of Therapy Due to Adverse Events.” They only evaluated a year of data so, again, while it is likely that these studies would have captured common adverse events had they bothered to try, there isn’t long enough follow-up to have any information about serious (possibly life-threatening) long-term adverse events.
Association of Glucagon-like Peptide 1 Analogs and Agonists Administered for Ob*sity with Weight Loss and Adverse Events: A Systematic Review and Network Meta-analysis. Vosoughi, K., et al.,
This study included 64 RCTs with durations from 12 to 160 weeks, with a median of 26 weeks. As is common in these studies, the majority of the sample (74.9%) was white.
Like those above, they only looked at treatment discontinuation from adverse events, they did not capture specific adverse events (common or serious.) Of the seven GLP-1 drugs they tested, liraglutide was tied with taspoglutide for the highest discontinuation of treatment due to adverse events.
The study authors also note that “Risk of bias was high or unclear for random sequence generation (29.7%), allocation concealment (26.6%), and incomplete outcome data (26.6%).”
Breakdowns for Comparative Cost-effectiveness Studies
First, the WHO request authors themselves admit that when it comes to cost-effectiveness, “the analyses have generally been performed only for high-income countries.” This is significant since they are asking the WHO to consider these drugs essential for the entire world.
It’s also important to understand that none of the data looks at a comparison of cost effectiveness for weight-neutral health interventions to these drugs. Without that information there is no way to calculate actual “cost effectiveness” since it’s possible that weight-neutral health interventions would have greater benefits with less risk and dramatically lower cost.
NICE’s guidance: Liraglutide for managing overweight and ob*sity Technology appraisal guidance [TA664]Published: 09 December 2020.
Do recall that NICE is involved in the current scandal with Novo Nordisk for influence peddling.
These guidelines are created based on a submission of evidence by Novo Nordisk. The committee’s understanding of “clinical need” was based on the testimony of a single “patient expert” who “explained that living with ob*sity is challenging and restrictive. There is stigma associated with being ob*se.”
Once again we see a rush to blame body size for any “challenges” and “restrictions” of living in a higher-weight body, accompanied by the immediate decision that those bodies should be subjected to healthcare interventions that risk their lives and quality of life in order to be made (temporarily, by Novo and NICE’s own admission) thinner. There did not seem to be a patient expert to discuss the weight-neutral options.
It was not immediately apparent if the patient expert was provided/paid by Novo Nordisk, but they certainly forwarded their narrative that simply living in a higher-weight body is a disease requiring treatment.
It should be noted that while the trial Novo Nordisk submitted covered a wider range of people, they specifically submitted for this recommendation only the subgroup of that population who were diagnosed with “ob*sity,” pre-diabetes, and a “high risk of cardiovascular disease based on risk factors such as hypertension and dyslipidaemia.”
So, even if we accept this guidance as true, the WHO Essential Medicines request applies to a population much wider than this and so this fails to justify the cost-effectiveness for that population.
This guidance is also based on the costs associated with obtaining the drugs through a “specialist weight management service” since an agreement is in place for Novo Nordisk to give a discount to these services.
In calculating the ICER per QALY gained, the recommendations note that “Because of the uncertainties in the modelling assumptions, particularly what happens after stopping liraglutide and the calculation of long-term benefits, the committee agreed that an acceptable ICER would not be higher than £20,000 per QALY gained”
Again, this recommendation is based on a trial submitted by Novo Nordisk that included 3,721 people and lasted for three years, but only 800 met the criteria for this cost-effectiveness recommendation. The trial failed to show a significant reduction in cardiovascular events. Novo’s calculation of risk reduction was based on surrogate outcomes, which NICE points out “introduces uncertainty because causal inference requires direct evidence that liraglutide reduces cardiovascular events. This was not provided in the company submission because of lack of long-term evidence.”
The NICE committee admits “relying on surrogates is uncertain but accepted that surrogate outcomes were the only available evidence to estimate cardiovascular benefits.”
I just want to point out that another option would be to refuse to experiment on higher-weight people without appropriate evidence.
These cost-effectiveness calculations are based on someone using the drug for two years, with no actual data on reduction in cardiovascular events, and with the admitted assumption that “any weight loss returned to the base weight 3 years after treatment discontinuation.”
Said another way, this committee decided that it was cost effective to spend up to £20,000 per QALY for people to take a weight loss drug with significant side effects for two years, with no direct evidence of reduced cardiovascular events, and with the acknowledgment that people will be gaining all of their weight back when they stop taking it.
Those who wrote the request for WHO to consider these drugs “essential” chose to characterize this as “At the chosen threshold of £20,000 per quality-adjusted life year (QALY) gained, the report concluded that liraglutide is cost-effective for the management of ob*sity.”
I do not think that is an accurate characterization of the findings.
The request cites “A report by the Canadian Agency for Drugs and Technologies in Health (CADTH) found that compared to standard care, the ICER for liraglutide was $196,876 per QALY gained”
For the US, they cite a study that found that to achieve ICERs between $100,000 and $150,000 perQALY or evLY gained, the health-benefit price benchmark range for semaglutide was estimated as $7500 - $9800 per year, which would require a discount of 28-45% from the current US net price.
They also cite “Cost-effectiveness analysis of semaglutide 2.4 mg for the treatment of adult patients with overweight and ob*sity in the United States, Kim et al.
Let’s take a look at their conflict of interest disclosure (emphasis mine)
“Financial support for this research was provided by Novo Nordisk Inc. The study sponsor [that means Novo Nordisk] was involved in several aspects of the research, including the study design, the interpretation of data, the writing of the manuscript, and the decision to submit the manuscript for publication.
Dr Kim and Ms Ramasamy are employees of Novo Nordisk Inc. Ms Kumar and Dr Burudpakdee were employees of Novo Nordisk Inc at the time this study was conducted. Dr Sullivan received research support from Novo Nordisk Inc for this study. Drs Wang, Song, Wu, Ms Xie, and Ms Sun are employees of Analysis Group, Inc, who received consultancy fees from Novo Nordisk Inc in connection with this study.”
Given that, you probably won’t be shocked to learn that this concluded that Novo Nordisk’s drug, semaglutide, was cost-effective. The reason I bolded the text above is that this study is based on modeling – they are taking what is, by their own admission, a “new drug” and making predictions for 30 years. Everything was simulated based on trial data (you know, those trials that we’ve been discussing that often have horrendous methodology…) and “other relevant literature.” The construction of the modeling and the interpretation of the results was directed by the company who stands to benefit financially from the findings, and carried out by that company’s employees and consultants.
Also, and I’ll just quote again here since I don’t think I can improve on their text “Cost-effectiveness was examined with a willingness-to-pay (WTP) threshold of $150,000 per QALY gained” I do not think that this WTP is based on a global assessment.
In their (and by their I mean Novo Nordisk’s) modeling they find that semaglutide was estimated to improve QALYs by 0.138 to 0.925 and incur higher costs by $3,254 to $25,086 over the 30-year time horizon vs comparators.
And, again, this is without any kind of actual long-term data.
I think that the best way to characterize this information is “back of the envelope calculations” at best.
To sum up, I do not think that the research they cite comes anywhere close to proving that these drugs have levels of efficacy, safety, or cost-effectiveness that warrant their addition to the WHO list of essential medicines. I believe that if the WHO grants this request I think it will be an affront to medical science, it will cheapen the concept of “essential medicines,” and it will harm untold numbers of higher-weight people all over the world.
Did you find this post helpful? You can subscribe for free to get future posts delivered direct to your inbox, or choose a paid subscription to support the newsletter and get special benefits! Click the Subscribe button below for details:
Liked this piece? Share this piece:
More research and resources:
https://haeshealthsheets.com/resources/
*Note on language: I use “fat” as a neutral descriptor as used by the fat activist community, I use “ob*se” and “overw*ight” to acknowledge that these are terms that were created to medicalize and pathologize fat bodies, with roots in racism and specifically anti-Blackness. Please read Sabrina Strings’ Fearing the Black Body – the Racial Origins of Fat Phobia and Da’Shaun Harrison’s Belly of the Beast: The Politics of Anti-Fatness as Anti-Blackness for more on this.