Putting People Ahead of the Maths: Addressing the Limitations of Today’s Approach to HTA

Contributed by


Mike Glover, Vice President Global Head of Access and Pricing Strategy Oncology at GSK outlines that while new mathematical approaches to health technology assessment (HTA) have been hugely beneficial, a broader approach that includes the insights of individual people is necessary to truly drive forward better health outcomes.


The words “people” and “maths” have been chosen deliberately.

In a cross industry/payor meeting that was held recently, we were discussing the changing world of Multiple Myeloma treatment. One of the delegates believed the goal of the meeting should be to agree on the “most cost effective sequence of treatments for MM”. They explained that “we should not be drawn into discussions over various treatments as all pharma companies will want theirs to be used first.” This is an interesting start to a joint Provider/pharma meeting, but unfortunately it is not unusual.

Let’s focus on two words: people and maths. In our Pharma world there is a lot of real tension between these two domains. There is no doubt, though, that over the past three decades the disciplines that encompass the study of population health and health economics – such as epidemiology, biostatistics, medicine, and economics – have come together to advance the methods that underpin what today we call Health Technology Assessment, or HTA for short.


Foremost within the field of HTA is the metric that we are all familiar with called the Quality Adjusted Life Year, or QALY, which is uniformly equated with healthcare decision making. This decision making is often through the application of a population-level mathematical threshold or, in lay terms, an “upper limit”, at which society has deemed that technologies have not demonstrated enough value to justify making them available to the population.

Yet in making this very statement we easily moved from people to “population”, which embraces a mathematical/computational element such as an ICER threshold. Basic economics tells us that an economic analysis should inform the decision, not drive the decision. Over the course of this discussion we will look at six research and evidence-based observations with a purpose. We need to move beyond the constraints of the population-level QALY when we consider the value of novel health technologies.

However, before reviewing each of the six points, let’s address that this is not purely another anti-HTA, anti-ICER-threshold discussion. In fact, some of the concepts and positions we discuss are widely held. Recently, the International Society of Pharmacoeconomics & Outcomes Research (ISPOR) published a member survey. Remember that ISPOR represents the very discipline within pharma, academia, and government that develops the methods, trains the researcher, and sets the standards for HTA.

The ISPOR member research reveals that over three-quarters of respondents believe that the cost effectiveness/cost utility modeling approach has limitations. Those limitations center on a lack of representation of the perspectives of other stakeholders. Central to these perspectives are the people we are seeking to help, i.e. the patient.

In fact, 94% of respondents noted that Cost Effectiveness/Cost Utility analyses fail to fully account for the patient perspective. In other words, the perspectives of individual people are being excluded.

Encouragingly, the limited perspective that is inherent in the HTA-equals-a-QALY approach has not been lost on the ISPOR leadership. In a much-debated paper in November 2018, an ISPOR Task Force discussed what it called “Elements of Value”. This is now affectionately called the “Value Flower”.


The Value Flower identifies 12 elements of value: (in the “clock” face below)

  • The QALY and net costs = foundational elements (petals 12 o’clock and 1)
  • Productivity and adherence = widely studied elements of value (the red petals at 2 and 3 o’clock)
  • And 8 more novel elements of value that are less frequently studied (the orange petals from 4-to-11 o’clock).

Made with Visme Infographic Maker


In summary, there is a widespread recognition that when we assess the value of a health technology, we ought to take a broader and more holistic view.


Societal value is dynamic, changing over time subsequent to evolving provider, patient and product characteristics

A study done in 2012 by Lu et al makes the point that at the time of launch the inherent “value” of a drug is often at its lowest. In fact, it could be argued that the worst time to assess the value of a drug is at launch.

Let’s think about this:

  • The drug is new and the only doctors who are familiar with how to use it are clinical trialists.
  • It takes time for doctors to figure out how best to use a new drug, what type of patient it is best for, when to prescribe it, when not to use it, etc…
  • The drug only has its first indication. The data is constantly evolving, often uncovering further valued benefits that were considered in its initial valuation. It may be months or years until further indications are approved.
  • The cost is – at least in the majority of markets ex-US – at its highest on the day of launch.
  • The drop-in price that accompanies loss-of-patent-protection is several years away.

What Lu et al concluded is that value is ever changing and thus the cost effectiveness should be considered dynamic. Yet, the HTA process focuses on a once-and-done, shortly-post-approval assessment. Even further, subsequent payor-initiated assessments look to deliver greater downward pressure on costs. This is sometimes driven by conditional approvals at launch, and it’s sometimes driven by newer entrants into the arena.

Two recent examples of drugs that may have been undervalued at launch are the SGLT-2 class (in diabetes) and the PCSK9 class (in hyperlipidemia),both of which have been subject to extensive HTA and public scrutiny on their value. Yet, both of these drugs have demonstrated meaningful long-term health outcomes (in heart failure, renal disease and CV events respectively) several years after launch. These are perfect cases to argue for a dynamic value.

Pharma companies in partnership with payor communities should evolve and explore dynamic pricing methodologies to ensure that the full value of a medicine is captured where possible. Recently, there has been an increasing receptiveness to novel outcome based thinking in payor relationships, partly driven by the more expensive medicines in the cell and gene therapy arena. Perhaps a more dynamic approach to valuing medicines could partner this thinking and not be another way of dressing up discount under the banner of outcomes based.


Patient preferences can be misrepresented by HTA focusing on an average treatment response

In what many view as a landmark paper on the topic of novel approaches to value, Lakdawalla et al (2012) described “The Value of Hope”.

In simple terms, think of a case where a patient is faced with a cancer diagnosis and two treatments to choose from. Treatment A guarantees an average response of 24 months of survival. Treatment B guarantees 10 months response. However, there’s a 20% chance of being a good responder, which can then lead to up to 54 months of survival. Almost three quarters (71%) of patients preferred to gamble and pick Treatment B.

What does this tell us?

  • People are able to put treatment choices regarding their life and mortality in context.
  • Individual people have different views and make different trade-offs.
  • The hope of being a good responder is clearly of value.

The concept of the value of hope is not captured in the population-level QALY. In fact, another tension exists, namely that providers, payers, and regulators tend to focus on the mean and median response while patients focus on the “what if”. Another element of value that one can identify is what is called “Option Value”. In simple terms, that part of the function of today’s drug is to keep you alive long enough or well enough to receive tomorrow’s new advance.

Two examples illustrate “Option Value” nicely:

  • Taxol kept women with breast cancer alive to receive Herceptin, in effect adding 0.4 years of life for those surviving to receive Herceptin.
  • The first generation TKI (in this case imatinib (Gleevec)) kept some people with Chronic Myeloid Leukemia (CML) alive long enough to receive the second generation TKI.
  • When one incorporates the value of the 2nd Gen TKI into the value of the original 1st Gen TKI the survival benefit raises by almost 10%

However, the conventional QALY does not embrace such option value.


Societal benefits from innovation over-estimate the value retained by the innovator

There is a widespread belief that all the value from a health technology advance, such as a new drug, flows to the pharma company. This is not the case.

David Grabowski and colleagues applied methods originally developed by Gary Becker, the Nobel Laureate in Economics. They demonstrated that over a 20 year period, the first class of blockbuster statins had delivered to US society a staggering $1.2 trillion dollars of value, helping middle-age adults improve their lipid profile and in-turn have fewer CV events. Of this $1.2T, the manufacturers kept 24%. This is not an inconsequential amount, but it’s also not the major proportion of the total value generated.

This benefit is not restricted just to the lipid example. In the case of HAART treatments for people with HIV/AIDS, only 5% of the value was retained by the pharma innovator. 95% of the value, $1.4 trillion, was retained by society.

In a cancer example, and returning to the TKI class for CML, society retained 90% of the value of the 1st Line TKI class and 90% of the value of the 2nd Line TKI class. In short, society appropriates the vast, vast majority of the value of novel pharmaceuticals.

Again, this element of societal value is not embraced within the conventional cost-per-QALY HTA approach.


Payors focus on short term expenses rather than longer term costs

Payers have incentives to focus on short-term costs and offsets rather than long-term outcomes because of patient churn. Mechanical budget limits are more likely to be exceeded by cures or other therapies that involve high short-term cost but larger long-term benefits. In 2015, Van Nuys K. et al. looked at the Hepatitis C arena where clearly Hep C treatments returned substantial health gains. Payers are reluctant to cover novel therapies beyond the sickest individuals, but short-term savings mean higher cumulative costs after just 13 years.

Health policy should seek to align payers with long-term benefits, not focus even more on short-term cost. Namely that a short-term view supports affordability but does not allow society to ever get ahead of the clinical, humanistic and economic burden of the disease


Patients describe a willingness to pay for coverage of innovative specialty medicines

In 2012, John Romley and colleagues published research that demonstrates that healthy people are willing to pay an additional up-front premium, in this case of $13 or about 10 Euro per month, to have full access to novel pharmaceuticals. Which means that, when one extrapolates this to the likely demand for and use of such new drugs, people are willing to pay more than 2x what the coverage costs.

That is, people are willing to pay extra for generous drug coverage of new treatments.


Finally, providers and patients embrace innovation particularly for severe, life-threatening diseases.

Previously we looked at an example where healthy people are willing to pay additional premiums now for drug coverage that includes access to novel treatments.

More broadly, there is a wealth of evidence that society embraces innovation. Note these quotes from major players in cancer care, ASCO and NCCN, in the USA.


“ … age of rapid innovation in cancer care and introduction of novel immunotherapies, NCCN and ASCO recognize the crucial need to support physicians’ decision-making through publication of timely, evidence-based clinical guidelines,” said Robert W. Carlson, MD, Chief Executive Officer, NCCN. “We anticipate that, through our common mission, the collaboration between NCCN and ASCO indeed will improve the lives of patients with cancer.”


Clearly, these stakeholders recognize and appreciate the value provided by pharma in making new treatment options available to patients and providers. The market more broadly embraces this innovation. Take the uptake and rapid expansion of the immunotherapy agents, the growth of these agents has continued and the immunotherapy class of cancer drugs emerging as a major advance in cancer care.

Whether this success in the market represents pent up demand for a novel therapy or frustration at the lack of effectiveness of older treatments is a subject for a different discussion. However, from the viewpoint of society, the need to embrace and support biopharmaceutical innovation, and the need to make such innovation accessible to people with cancer, the data speaks for itself.


In summary:

The technology appraisal movement has advanced the standards and methods to assess the value of innovative treatments and the QALY deservedly has a place within the HTA approach.

However, there are limitations in letting the maths drive the decision. The QALY has significant limitations and does not embrace elements of value that at a simple level make common sense. Other broader methods could capture a much wider view of the value of medicines and living with disease that individual people can express. These broader assessment methods, along with others, clearly exist and warrant a much wider discussion. We must ensure that evolution in disease management is accompanied by evolution in drug appraisal.



Although I am an Employee of GSK the views represented here are independent and I am not representing GSK in this piece.



Dynamic Cost-Effectiveness of Oncology Drugs (Am J Manag Care. 2012;18:S249-S256). https://www.ajmc.com/journals/supplement/2012/a386_12nov_oncology/a386_12nov_onclogy_lu_s249

The Large Social Value Resulting from Use of Statins Warrants Steps to Improve Adherence and Broaden Treatment  https://pubmed.ncbi.nlm.nih.gov/23048109/

The Option Value of Innovative Treatments for Non-Small Cell Lung Cancer and Renal Cell Carcinoma https://pubmed.ncbi.nlm.nih.gov/29087638/

Darius N. Lakdawalla, John A. Romley, Yuri Sanchez, J. Ross Maclean, John R. Penrod and Tomas Philipson
How Cancer Patients Value Hope And The Implications For Cost-Effectiveness : Assessments Of High-Cost Cancer Therapies Health Affairs, 31, no.4 (2012):676-682

https://www.nccn.org/about/news/newsinfo.aspx?NewsID=846, article dated 2/17/17,

https://www.nccn.org/about/news/newsinfo.aspx?NewsID=846, article dated 2/17/17,

Romley JA et al. Health Affairs 2012; 31(4):683-690

Hodi et al, NEJM, 2010; and Lakdawalla et al, Health Affairs, 2012.

Sanchez Y, et al. Am J Manag Care. 2012.

Philipson T and Jena AB. Who Benefits from New Medical Technologies? Estimates of Consumer and Producer Surpluses for HIV/AIDS Drugs. Forum for Health Economics and Policy. 2006;9(2).

Yin W, Penrod JR, Maclean JR, Lakdawalla DN, Philipson T. Value of survival gains in chronic myeloid leukemia. AJMC 2012;18:S257-S64.

Lu Y et al, AJMC 2012; 18: s249-256

Hartung et al. “The cost of MS drugs in the US and the pharmaceutical industry” Neurology May 2015

Moreno et al. Costs and spillover effects of private insurers’ coverage of hepatitis C treatment. AJMC, 2016

Van Nuys K. et al. “Broad Hepatitis C Treatment Scenarios Return Substantial Health Gains, But Capacity Is A Concern.” Health Affairs 2015



Add Your Comment

You must be logged in to post a comment.

Related Content

Latest Report