Stuart R. Gallant, MD, PhD
This post is the first in a two-part series on the value medications. “The Value of Medications” is a bit of a misnomer. Medications don’t have one value (“The Value”). Medications have different values, depending on how they are evaluated. These values can be depicted as spheres, as in the figure below. Sometimes the spheres overlap (signifying agreement on value), but sometimes they diverge (signifying conflicting valuations):
At the bottom of the figure is the sphere of scientific and medical knowledge. Science is an empirical endeavor (i.e., based on evidence), but the evidence is gathered on groups of patients. “On average, lisinopril decreases blood pressure by X amount once the dose is adjusted to the patient.” Of course, some patients will respond more strongly than others. Also, scientific studies are by their nature abbreviated, in timeframe and in the number of effects that they look at. As a result, scientific knowledge is limited, and in some cases incorrect. Many first-year medical students have the experience of one of their lecturers intoning, “Half of medical knowledge is wrong, we just don’t know which half.” Medicine tries to formulate a one single best treatment that works for each given medical condition—unfortunately that one correct answer may not work for every patient.
The sphere of healthcare economics, shown on the left of the figure, takes scientific and medical knowledge and runs with it. The most commonly cited healthcare economic tool is the Quality-Adjusted Life Year which is discussed later in this series. The advantage offered by QALY is that it reduces the medical knowledge about a particular treatment regime down to one number that can be assessed and compared—that is also its chief disadvantage—QALY analysis assumes that the knowledge of a particular disease or condition can be reduced down to one number.
Finally on the right side of the figure, there is the sphere of patient-centered care. Patient-centered medicine begins with the idea that each patient is unique and focuses on providing options and choices to the patient which are consistent with the patient’s needs and beliefs, and providing those options in a caring and compassionate way. Implicitly, patient-centered care assumes that medical problems have more than one solution and attempts to provide the patient a degree of control of their medical care.
As seen from the figure, each of these ways of valuing medicine has some areas in which it overlaps with the other ways. However, it is also possible that for some patients these ways of looking at value will not overlap—they will in fact conflict. It is far beyond the scope of these two posts to attempt to fully delineate these systems of value, their relationships to each other, and the tradeoffs that these relationships may entail. What I would like to do instead is shine a light on a few areas that pharmaceutical scientists need to be aware of as they engage in the drug development process. In the first post, the recent history of healthcare spending and cost-effectiveness assessment is discussed, with a particular focus on the limitations of a solely healthcare-economic viewpoint. In the second post, a recent example of cost-effectiveness analysis is considered and trends in cost effectiveness data are examined. Also, some thoughts are offered regarding the implications of this type of analysis in drug development.
Trends in Healthcare and Healthcare Spending
If you imagine what it was like to be a physician in the early 1960s, doctors had very few tools at their command, compared to physicians today. For example, many patients (especially older people) did not have insurance, so often even simple conditions such as hernia could not be corrected. At the same time, physicians had few medications to treat common chronic conditions, such as hypertension, diabetes, and hypercholesterolemia. For example, beta blockers and calcium channel blockers only became available to treat hypertension toward the end of the 1960s.
But starting in the mid-1960s, the combination of a booming economy for working-age Americans and government insurance for elderly Americans in the form of Medicare greatly expanded the reach and power of the American medical system. This expansion began in an environment of relatively few FDA approved medicines, but as time went by the drug portfolio of American physicians increased. Two plots allow us to see this process of growth in American healthcare as it stretched out from the 1960s to the 2010s [1]:
Healthcare spending as a percent of GDP rose steadily from the 1960’s through the early 2000’s before plateauing at just below 18% in the 2010s.
Prescription drug spending as a percent of healthcare spending actually dropped through the 1960s and 1970s. This was due to the limited number of options for drug treatment for common conditions (such as hypertension, diabetes, and hypercholesterolemia)—doctors could evaluate patients, but physicians had few pharmaceuticals to prescribe to their patients. However, with the advent of medications like oral treatments for diabetes, recombinant insulin, and statins, doctors had many more options to improve the quality and length of the lives of their patients. Through the 1980s and 1990s, the share of healthcare spending on prescription medications increased, before plateauing at a little less than 10% around 2005.
Measuring Healthcare Cost-Effectiveness
As the amount the United States spends on healthcare has risen (in real dollars), attention has focused on how to control (if not reduce) the size of America’s healthcare investment. Warren Buffett famously and waggishly described healthcare spending as “the tapeworm of American economic competitiveness” [2]. In January 2018, Buffett’s Berkshire Hathaway got together with Amazon and JP Morgan Chase to start Haven Healthcare (havenhealthcare.care) with the stated goal of providing low-cost, high-quality healthcare for the families of their employees. Over the following three years, Haven attempted to develop a workable business model. Periodic press releases appeared on the Haven website announcing progress, but by the end of 2020 momentum had been lost as evidenced by the revolving door of appointment and departure in the Haven C-suite. Very few patients were ever served through the Haven system, and in January 2021, Haven announced that it would be ending its work as a healthcare insurer/provider.
This writer doesn’t have any visibility into the death spiral of Haven, but there are a couple of lessons that should be obvious, even from the street-level view. First, there are a lot of very smart and well-resourced folks who are interested in reducing healthcare costs. Second, it’s a hard nut to crack. My personal opinion is that one significant issue is that medicine isn’t one business—it’s really many businesses existing together in a close geographic space.
One component of the healthcare system is of course pharmaceutical development, manufacturing, and supply. To better understand what it means for a medication to be cost-effective, healthcare economists working in the 1970s developed the concept of the Quality-Adjusted Life Year (QALY) [3]. The two central concepts of the QALY are that: 1) healthy life is worth more to humans than life burdened with illness and 2) a longer life is a preferred outcome, compared to a shortened life. (Those may seem like obvious statements, but there is a degree of controversy about the QALY which will be discussed below.) With those two assumptions in place, it is possible to turn the cost-effectiveness of a medication into a numerical calculation. In the schematic below [4], a theoretical medical treatment is assessed. Without the treatment, the individual (or group of individuals) decline in health over a period of years. With treatment, the decline is slower. The benefit provided by the medication can be quantified by integrating the brown area of the graph (area B).
Calculation of the QALY benefit of a treatment is the first step in assessing the cost-effectiveness of a treatment, the second step is to calculate the ratio of cost to the benefit (dollars per QALY gained). QALYs for diverse conditions can be tabulated and compared. Cut offs for treatments which are too expensive can be imposed, and costs can be controlled—or at least that’s how some healthcare administrators may attempt to use this type of analysis.
Here are a few thoughts on the benefits and the risks QALY analysis:
- QALY analysis can be one element of the discussion of a drug’s price: For novel medications pricing involves a discussion between the drug manufacturer and payors and formulary committees. The starting point of this discussion for manufacturers is the cost of drug development (factoring in failed drug development efforts). (See PharmaTopoTM post on “Cost of Pre-Clinical Development.”) But, these discussions also involve the benefit to the patient which is quantified using QALY analysis. This type of analysis will be discussed further later in this series.
- There are several significant weaknesses of QALY:
- QALY analysis doesn’t consider how medical progress occurs. Consider the case of HIV medications. The first medications on the market had limited efficacy (resistant strains quickly arose before the advent of multidrug treatment) and unpleasant side effects. But over time, more was learned about HIV, and medications improved. The first medications were a bridge to later medications, and therapy improved over time. Another way to say the same thing is that QALY systematically undervalues medications which exist in areas of medicine in which substantial progress is made because those medications exist as bridges to better medications—and this bridging is not included in the QALY analysis.
- The knowledge generated by clinical trials is limited in outcome measures (both what is measured and how long it is measured). For example, in the treatment of muscular dystrophy, a small clinical efficacy of a drug can set the course of the entire disease on a different path. A patient who might die early in life without the treatment may be able to attend college with the treatment [5]. QALY may not be able to quantify these kinds of differences over years because clinical studies are confined to shorter periods of time.
- Patients may value particular benefits differently. For instance, a small difference in clinical outcome (for example in the ability to control and predict the behavior of one’s stool for a cystic fibrosis patient) may provide great relief to the patient—it might make them feel more comfortable in traveling away from home, for instance. It could be hard to quantify that benefit in a QALY, but it might be important to the patient.
- Patients with disability are valued less than patients without disability—this aspect is particularly controversial. If a patient with a disability that reduces their health score from 1.0 to 0.7 is treated in a way that extends their life 5 years, the resulting QALY is scored at 70% compared to someone who started with a health score of 1.0—so the fully abled patient would have a QALY benefit of 5, but the disabled patient would have a QALY benefit of 3.5. This will make medical treatments for patients with disability appear less cost-effective, perhaps preventing some treatments from being developed or marketed in countries where QALY analysis is commonly used.
- The effect of the treatment on the patient’s family or caregivers is ignored in QALY analysis. For example, a treatment could reduce the number of dressing changes that a patient requires per day, potentially allowing a parent who is primary caregiver more time for work or for self-care, but those effects are not included.
In summary, QALY analysis is probably strongest when it is used in a very narrow context—comparing two very similar circumstances that only differ in one or a few small ways. Yet, that isn’t how QALY analysis (and similar measures such as DALY analysis and HYE analysis) is being used in many circumstances [6, 7], and some patients and care providers find that frustrating. That frustration became embodied in the Affordable Care Act which specifically prohibits use of QALY or similar measures to determine thresholds of coverage.
[1] Data was curated from data on these webpages and others:
www.statista.com/statistics/184968/us-health-expenditure-as-percent-of-gdp-since-1960/
www.statista.com/statistics/184914/prescription-drug-expenditures-in-the-us-since-1960/
[2] www.nytimes.com/2017/05/08/business/dealbook/09dealbook-sorkin-warren-buffett.html
[3] Zeckhauser, R. and Shepard, D. “Where Now for Saving Lives?,” Law and Contemporary Problems, Vol. 40, No. 4, Valuing Lives (Autumn, 1976), pp. 5-45.
[4] Sanofi. “Sanofi Short Guide: The QALY Explained,” www.sanofi.co.uk/-/media/Project/One-Sanofi-Web/Websites/Europe/Sanofi-UK/Home/our-responsibility/Working-in-partnership/patient-groups/The-QALY-Explained.pdf
[5] Butcher, L. “Is the Medication You’re Taking Worth Its Price?,” www.salon.com/2020/02/01/is-the-medication-youre-taking-worth-its-price_partner/
[6] Tan-Torres Edejer, T. et al. WHO Guide to Cost Effective Analysis, World Health Organization, Geneva 2003.
[7] Edlin, R. et al. “Cost-effectiveness analysis and efficient use of the pharmaceutical budget: the key role of clinical pharmacologists,” Br J Clin Pharmacol, 70:3, 350–355.
Disclaimer: PharmaTopoTM provides commentary on topics related to drugs. The content on this website does not constitute technical, medical, legal, or financial advice. Consult an appropriately skilled professional, such as an engineer, doctor, lawyer, or investment counselor, prior to undertaking any action related to the topics discussed on PharmaTopo.com.