Paul and Brett's Alpha

In the most simplistic and reductive terms, the investment strategy pursued by the Trust has three elements. The first is to understand the evolution of the healthcare delivery paradigm; how it is changing and what the future delivery of products, technologies and services might look like from a medium- to long-term perspective.  

The second element is to take a view on which of these changes seems the most attractive from an investment point of view, considering the addressable market for future products, technologies or services and also the defensibility of those future revenue streams (given our intention to hold investments for three to seven years, all other factors being equal).  

The third element is company specific. Where there may be more than one way to invest in a particular theme, we need to decide which option represents the best strategy to benefit from the changes that we foresee. Sometimes, we may conclude that there is no attractive way to play a particular theme, in which case we move on to something else.  

In many ways, the most intellectually interesting part of the continuum described previously is the first stage; speaking with key opinion leaders across the sphere of payment, regulation and care as to how things could or should improve in the future.  

There is always a degree of uncertainty in these kinds of prognostications. As we have noted many times (most recently in regard to the current enthusiasm for all things related to Artificial Intelligence and Machine Learning), the “future” often arrives in a different form and in a different way than first imagined, at which point, you need to pivot and think about alternative outcomes to those previously envisaged.  

To our minds at least, this flexibility of thought goes to the heart of the job that we are paid to do. It is part of the reason that we put the effort into these discursive factsheets – we want you to gain insight into both what we are thinking about and how we are thinking about it. 

A good illustration of this latter point would be the Trust’s historical holding in Teladoc. We first began to buy the shares in February 2018, when they traded around $38. We rapidly scaled this to be the second-largest position in the portfolio, arguing that telemedicine (or, more specifically, ‘electronic triage’ and ‘case management’, Teladoc’s two principle services at the time), was the most compelling opportunity to cut costs in primary care that we had yet come across.  

Recall that one in four primary care physician appointments are concluded to have been medically unnecessary after they have taken place, which is a bit of a problem. Tele-visits cost between 50% and 75% less than face-to-face appointments and are generally much shorter, which increases physician productivity (employed doctors can be viewed as a fixed cost).  

All of this went well enough until early 2020, when excitement over virtual appointment capabilities grew exponentially due to the restrictions imposed during the COVID-19 pandemic. The necessity to continue to see patients but minimise physical contact greatly accelerated the adoption of tele-medicine services across the globe, but it began to play out in a slightly different manner than we expected.  

Simply put, every large provider and various new entrant software companies began to build their own services. There was even a company that went public that year, Amwell, whose traditional business was building ‘white label’ services of this type for healthcare companies. We began to worry that Teladoc’s opportunity to gain market share and establish itself as the ‘go-to’ provider would be eroded because the pandemic accelerated market conversion to a speed that literally no-one could keep up with. All of a sudden, commoditisation looked to be inevitable in a short period of time. We began to scale back our holdings (having made a very nice IRR).  

The situation changed again in August 2020, when Teladoc announced that it was spending $18.5bn buying another digital health company called Livongo. We never liked this business, so we sold our holding down to zero (the share price at this stage was >$200). We likened the deal to putting ice cubes in champagne: you don’t add anything and you dilute what was good in the first place! Clearly, one would not do such a deal unless one worried about the core business prospects.  

The rest is history. The shares went sideways for a bit, briefly exploded to a high of $294 in February 2021 and then began an almost unbroken decline to this day. Over the course of 2022, Teladoc booked impairment charges on its Livongo acquisition totalling $13.4bn. At the end of this process, the shares stood at $26. We had to wait a little longer for the denouement, but this came in February 2024 when the CEO reset expectations for FY24, guided for low-to-mid single digit revenue growth over the coming three years and announced a cost cutting programme. On the results call, the CEO said: 

“It's important to remember that most U.S. healthcare consumers have access to virtual urgent care today, so it's largely a replacement market at this point. We've consistently taken share in this market, and we expect to continue to do so, but it's fairly well penetrated” 

In other words, it’s over. You have become a utility. It is no longer about growth, it’s about margins now. The shares are currently trading at $15, but this still feels like a generous forward multiple for what is a low revenue growth business. The company expects profits will continue to rise faster than revenues over the coming few years, but that margin improvement will struggle too once the low hanging fruit has been picked from the cost base.  

Talking of fruit, the conference call would have been a fruitful session for buzzword bingo, AI of course got a mention, but everyone is doing that, so whilst it will drive cost out and prices down; it will probably not boost margin (a win for the consumer, not the Teladoc shareholder).  

Perhaps the company can buy out some of its struggling peers to further consolidate the sector (think mobile telecoms as an analogy). Maybe it could start with Amwell. Those shares stand at $83c today, down 98% from the early 2021 high. Unlike Teladoc, it is still nowhere near profitable and may well run out of cash before it gets to breakeven. Doximity is a slightly different model for online primary care and has fared much better. It is profitable, still growing double digits and has no debt. The shares have nonetheless halved since 2022, but the multiple still looks dizzying to us. 

Our decision to sell out of Teladoc was much discussed with investors at the time. It was, not unreasonably, viewed as a ‘big call’, given that it was previously our second biggest holding and telemedicine was something that we had been very vocal about at the time. Moreover, it was a very successful investment for the Trust and was also not obviously “going wrong” at the time of sale. Finally, many analysts loved the Livongo deal, and of course the shares continued to rise after our exit, making us appear to have got it “wrong”, at least for a time.  

We recall comparable discussions around our exit of Illumina for very similar reasons in 2021 (it bought back the GRAIL stake, which set off Livongo-like alarm bells for us). In what should be a surprise to literally no-one, the core business is not doing well and group revenues are still forecast to be lower in 2024 than they were in 2021. The shares have ‘only’ lost 74% of their value since, but there is a valuable residual annuity within Illumina that will be hard for peers to compete away (this is less clear for Teladoc, in our view).  

The same types of discussion do not seem to be far from BBH investor’s minds today, albeit in a mirror-like form. The oft-asked question of the moment is not “why did you sell out of X” but rather “why don’t you own Novo Nordisk/Eli Lilly (or both)”. We are well aware that GLP-1 drug plays account for >12% of the portfolio at both our closest Trust peers and Lilly is the largest position in each case. These companies have been, and continue to be the largest contributors to the wider healthcare index’s performance. With this being the case, our decision not to own these shares has unarguably hurt us on a relative performance basis.  

It did not impact our relative and absolute performance for calendar 2023, when we still outperformed the benchmark (and our peers), but it hurt us a lot in the middle of that year and is hurting again in 2024, hence the discussion comes back to the fore.  

Eli Lilly and Novo Nordisk jointly accounted for just over 11% of the MSCI World Healthcare Total Return Index at the turn of the year and their share prices are up 34% and 24% respectively so far this year. With the index having risen 7.5% year-to-date in dollar terms, one can attribute more than 40% of the year-to-date index performance to these two names. Surely people ask, “You must regret your positioning?”.  

With investing, you are wrong until you are right and you are right until you are wrong. The funny thing of course about being right by selling out of something, or not owning it in the first place, is that everyone forgets what you did (or rather, did not do). History seldom recalls the avoided risk.  

When one is investing, all you can do is make the best decisions that you can with the information that you have at the time. We are not being inflexible on the obesity market and we have spent a lot of time researching it. What is it that we do and do not like about the Novo/Lilly obesity investment case? We will lay out our view again, in the hope of putting this issue, and our views on it to bed once and for all.  

Let us first acknowledge the theoretical addressable market (“TAM”) for effective weight loss therapies (note the intentional use of the word ‘theoretical’ and the phrase ‘effective weight loss therapies’ rather than GLP-1 drugs): it is not unfair or unreasonable to say that the majority of people in western markets are fat. Too fat.  

According to the UK government, nearly a quarter of 11 year olds in the UK are obese. That is really scary, because it sets you on a path toward Type 2 diabetes and fatty liver disease. Weight has always been much easier to gain than to lose (cf. previous factsheets on obesity drugs).  

Turning to the adults: 26% of UK adults are obese (BMI 30+) and a further 21% are overweight (BMI 25-30). We are barely in the same league as our US cousins though. There, 42% of adults are obese and 9% (or almost a quarter of the obese) are severely obese (BMI 40+; we used to call this morbidly obese, but like most ‘bad’ things it has unhelpfully been given a less scary name, to the obvious benefit of no one in particular). According the US CDC, 38% of American adults have pre-diabetes and 80% of them have no idea about this diagnosis. We had to double check that factoid. “Timebomb” springs to mind. 

Serious as it is, diabetes is not the only consideration. Around 220 human diseases have been linked to obesity (i.e. being obese makes them much worse) and obesity is arguably the causative agent in a small subset of serious ones, especially cardiovascular/cerebrovascular disease and cancer.  

Putting some numbers to the obesity crisis, there are some 50m potential patients in Europe and a further 70m in the US. That is a huge potential marketplace. We will probably approach 10m patients on GLP-1 therapy by the end of 2024. One can easily see how a simplistic analysis could forecast enormous category sales for safe and effective weight loss drugs and this seems to be where the market’s mindset is.  

We actually heard one CEO in this space claim that 150m Americans could be on anti-obesity therapy by the end of the decade. On the other hand, it might be sobering to consider that statins for high cholesterol, which are generic and thus cost very little, and have very robust evidence around the benefits of their usage, are still only being used by around 36m Americans today.  

The size of this theoretical market from a patient identity/prevalence standpoint is not something that is worth debating, it is objectively very large. A good proportion of these patients (the most obese and those with the highest overall levels of risk for obesity-linked diseases) will not struggle to gain reimbursement for treatment, at least for a time.  

There is also a very robust argument for bringing GLP-1 forward in the treatment continuum for Type 2 diabetes as an early intervention, due to the cardiovascular benefits of the therapies. As good as all this may sound for the two companies in question, there are some sensitivities to consider.  

Firstly, the cost/benefit of these therapies becomes much less positive as you move into lower risk, less obese patients. Who will pay for their (currently very expensive) therapy? Secondly, around a third of patients have tolerability issues; these are much higher in the real world than they are in clinical trials. Thirdly, there is the question of optimal therapy duration.  

Some commentators (i.e. the pharma industry) would argue that therapy needs to be life-long. As Prince sagaciously noted, ‘forever is a mighty long time’. What does GLP-1 maintenance therapy look like for a non-diabetic patient? Payors will only fund this if the overall benefits are positive. The challenge here is confounded by the knowledge that weight rebound on cessation of therapy (any therapy or diet, to be fair) is very significant.  

According to a recent report from the benefits consultancy Milliman, adherence to GLP-1 therapy for obesity beyond one year is <32%. How much of this is due to human nature, how much due to tolerability/side effects and how much is due to coverage restrictions/cost issues is unclear but the two-year health outcomes for the 70%-odd who come off therapy are probably not going to be hugely positive versus baseline and the benefit to the payor is going to be negative if the weight is regained despite spending all that money on the drugs.  

Human nature being what it is, we all want a quick fix; the easy option. However, gaining significant excess weight is not the work of a moment. It is estimated that you need to consume at least 7,000 excess calories to gain 1kg of additional body fat. The more active you are, the higher this number goes. It is thus illogical to imagine that there is ever going to be a quick fix to reverse this process, not to mention the issues around fat metabolism discussed in the September 2023 Musings.  

For facilities operators, the road back to normal post-pandemic was complicated by severe labour shortages. Like airlines, hospitals are one of those businesses where you simply cannot run short-staffed due to safety risks, so capacity must be curtailed to match staffing levels. Many people left the healthcare industry out of exhaustion and frustration in the aftermath of the pandemic. Some left because they could earn more money elsewhere.  

In a perverse circularity, the providers fall back on contract labour to bridge critical shortages and these same agencies were offering premium rates to those same staff leaving direct employment. This caused significant short-term margin pressure. Over time, this has ameliorated since procedural reimbursement rates reflect labour force cost inflation. The acceptance of higher wages has also allowed operators to hire new employees directly.  

Although higher wages represent a like-for-like cost increase, it is a cost saving versus using contract labour and the combination of a falling total wage bill and rising procedure rates has allowed for robust margin expansion during 2023, driven by a combination of lower unit wage cost and higher capacity utilisation, owing to a reduced capacity impact from labour shortages (although it is important to point out that labour shortages are still widely cited as a constraint for facilities operators).  

We do not expect this margin expansion to continue at pace, but operators seem confident they can maintain current margins while investing in additional capacity.  

All major operators cited reduced usage of contract labour as a significant profit driver in 2022 vs 2023. Does this increased capacity utilisation count as “COVID catch-up/backlog effects”? One could see how it might be considered part of this and thus why J&J might have taken a slightly different view on what drove the market in 2023 versus some of its peers.  

Another factor to consider is what the industry refers to as ‘patient activation’. You may well have a non-urgent medical issue that is in some way life-limiting, and it may well be that your insurance coverage would allow you to seek authorisation for an elective procedure to address this. Sometimes though, for whatever reason, patients do not follow up with doctors despite having a diagnosis. 

Medical device companies within our portfolio, especially in the cardiology, sleep apnoea and incontinence categories have been investing heavily in patient outreach and support to drive this patient funnel toward treatment, and this is increasing the market penetration of some product categories. It is not so much that any of this is new, but rather expanded use of social media channels and smart phones is making it easier to reach these people in a targeted fashion and drive them toward physicians who undertake that company’s intervention. 

Helping people change the habits that led to excess weight gain is critical to ensuring their weight loss becomes durable and they do not end up back where they started, or even in a worse position. This is not something that GLP-1 drugs do alone; ask any endocrinologist. That is why people tend to fair better on expensive programmes like WeightWatchers that provide support rather than just trying to do it all on their own. Primary care physicians (i.e. “GPs”) do not provide this type of support (assuming you can even get to see one). Perhaps what we need are other drugs to help people keep the weight off at a much lower cost.  

These points having been made, GLP-1 nonetheless represents the first broadly safe (if not very well tolerated) option for material weight loss. Assuming you can manage to tolerate the therapy (and pay for it), the majority of people could expect to lose a teens percentage of body weight in a year. This level of pharmacological efficacy has never really been possible before and so many of the questions posed previously have not really been debated by/with primary care doctors, as they have never come up before!  

The next question then is whether the GLP-1-based approach can be improved upon. When one begins to consider these questions, the conclusion that comes to our minds is a very simple one – GLP-1 monotherapy is not the answer, nor is the use of a combination that enables faster weight loss than monotherapy alone. Perhaps this is why Novo and Lilly both have multiple additional compounds in development for the treatment of obesity. It is very easy to conclude that there is considerable room for improvement here. 

If we ignore the self-pay aesthetics crowd in New York and Beverly Hills for a moment, and focus on the genuinely obese, the goal of therapy is not to look better, but to be healthier. Obesity is a health risk, after all. Sadly, measuring long-term health outcomes in clinical trials ceases to be practicable due to size and expense considerations. We will need to look toward patient registries. What do those longer-term outcomes look like? We do not yet know, but we can say for sure that this is not just about cardio-metabolic parameters.  

The core of this problem comes down to how GLP-1 drugs work. In the simplest sense, they cause food aversion. You want to eat less, so you do eat less. The body is in constant turnover and, without a ready supply of macronutrients, it will begin to waste away. What this means in practical terms is that, whilst you will preferentially lose fat mass, you will also lose lean mass (muscle, bone, organs).  

It has long been known that the ideal ratio of fat mass to lean mass loss when shedding body weight is at least 75% to 25%. A subgroup analysis from Novo’s STEP-1 study for Wegovy suggested a ratio of 10.4 to 6.9, or 60% to 40%.Whilst you can make the claim that body composition is improving when taking the drug (because you are losing more fat than lean mass), it is far from idealised.  

Moreover, the more weight you lose, the more the ratio will pivot away from fat mass toward lean mass (as there is less fat to lose). The same turns out to be true with the speed of weight loss; the faster you lose weight, the greater the risk that you are losing lean mass alongside fat mass. Starvation is a highly effective and rapid weight loss technique, but there is a reason why it is not one advocated by medical professionals.  

This problem is confounded by the fact that age ruins body composition as it is. Post menopausal women in particular can struggle to maintain healthy bone and muscle mass without regular exercise. The problem is less pronounced in middle-aged men, but we catch up in our later years.  

Although obesity is increasingly an issue for the young, it has been most commonly diagnosed in the middle aged and GLP-1 monotherapy may thus not be the best route to successful weight loss in some patient groups. To be clear, there is no robust evidence at this stage that long-term GLP-1 usage conveys any serious issues, but there again we have not got very much long-term data on its use in non-diabetic subjects.  

What we do know is that GLP-1 is but one of many hormones involved in the control of appetite and thus one of many potential druggable targets for pharmaceutical interventions.  

Our view is that the long-term outlook for this market will be one where the focus shifts to weight maintenance, body composition and the cost effectiveness of therapies and you need only look at some of the smaller biotechnology companies developing new drugs to see this shift being underway (i.e. disclosure of lean mass to fat mass ratios and ease of volume production in headline data releases).  

This is also an area where non-peptide (i.e. small molecule) drugs could have a huge advantage, since they are much easier to make at scale; both Novo and Lilly are struggling to meet demand for their products and investing huge amounts into capex to increase volumes.  

Combination approaches probably allow for gentler impacts, resulting in less nausea and thus better tolerability. We feel strongly that the goal of next generation therapies should not be to lose more weight in the first 12 months, but to be able to keep weight off with good preservation of lean tissue.  

And this is where the field opens up. There are many companies working on such products and we cannot see why, if they are successful in proving the concept, other ‘big pharma’ players will not want to get in on the action via M&A or in-licensing. For this reason, we see a multi-player, fragmented market. And this is the reason why we do not own Novo Nordisk and Eli Lilly at their current valuations.  

Both companies will continue to build an ‘evidence moat’ around their drugs, linking their use to improved outcomes in the commonest of those 220 obesity-linked diseases but let us not forget that it is the loss of the weight, not the mechanism of the weight loss per se that is driving the results here.  

The positive outcomes are almost a foregone conclusion to our minds. Let us also not forget that short-term symptom improvement is not the same as long-term outcomes, which of course rely on the weight staying off and that, in and of itself, will rely on maintenance therapy which at this stage is an assumption not a reality outside of clinical trials.  

Do we have some skin in the obesity game, that is to say – do we have exposure to potential second/third generation incretin obesity products? The answer is yes. We never said we didn’t think obesity was a real market, all we have said is that we question the attribution of sales and market share to a duopoly of established players and questioned whether or not GLP-1 therapy as a mechanism was the “solution” to the obesity puzzle in a demonstrably obesogenic environment.  

Some of you may also wonder if we have any exposure to the NASH/MASH fatty liver disease market. The answer to this is no. If obesity therapies are effective, then NASH/MASH risk will be greatly reduced, leaving only the F3/F4 patients with established fibrosis as a stand-alone market.  

Sadly, we have yet to see any compelling efficacy in this group, but we have looked at several projects across many different companies. One day, someone will unlock the biology of fibrosis (in lungs as well as livers), but so far the reversal of fibrotic deposition remains elusive to our minds. 

Sometime, in the next few years, we will find out the answer to the question of how these two markets (obesity and NASH/MASH treatment) will unfold. The approach we have taken is the same one that has served us well in the past. We will not invest if we cannot make the numbers work. We will also happily change tack if the situation develops differently to our current expectations and we are quite happy to explain, in detail via these factsheets, why we hold the opinions that we do.  

We always appreciate the opportunity to interact with our investors directly and you can submit questions regarding the Trust at any time via:  

shareholder_questions@bellevuehealthcaretrust.com  

As ever, we will endeavour to respond in a timely fashion and we thank you for your continued support during these volatile months.  

Paul Major and Brett Darke