Let’s say that there is a very severe disease—let’s call it horriblitis—with significant impacts on patient morbidity and mortality. The only available treatment for horriblitis is drug called BlackPill. BlackPill improves health outcomes by 1 QALY over best supportive care (BSC), but costs $500,000 over the patient’s lifetime. While not cost effective by traditional standards, payers felt that it as unethical to not give patients with horriblitis BlackPill because of the severity of the disease.
Now, a new drug called GreenPill has come to market. GreenPill is twice as effective as BlackPill, producing 2 QALYs vs. BSC, and only costs $1 more ($501,000 lifetime cost). Should payers cover GreenPill?
While most HTA bodies and payers would say ‘yes’, a paper by Walton et al. (2026) argues that the answer is ‘no’. Historically, HTA bodies have operated in an incremental world. Since standard of care is the BlackPill, paying $1 for an extra QALY is a great deal relative to the BlackPill. However, Walton makes the point that BlackPill is not cost effective vs. BSC. In fact, GreenPill is not cost-effective vs. BSC ($500,001 extra cost / 2 extra QALYs ≈$250,000/QALY). Walton argues that neither BlackPill or GreenPill should be reimbursed. Why? And is this practical?
The authors claim that incremental cost-effectiveness analysis rests on the premise that the standard of care (in this case BlackPill) is cost effective. This is not always the case. To address this issue, the Walton and co-authors propose 3 options:
- Reassess all treatments. The authors argue that in the UK NICE could repurpose its multiple technology appraisal (MTA) process to periodically re-evaluate all technologies within an indication in a comprehensive guideline and recommend reimbursement accordingly. While possible in theory, this would be far from practical in reality as many treatments are used without formal HTA evaluation.
- Reassess all treatments only when new treatments are introduced. This approach is much more feasible—fewer evaluations to be conducted—but also highly problematic. First, the evaluation of existing medical innovations would be done in an ad hoc manner (based on whether new treatment is introduced). Second, why would any pharmaceutical firm bring a new product to market with so much uncertainty?
- Adjust WTP thresholds when current standard of care is cost-ineffective. If the standard of care was not cost effective, the WTP threshold would be lowered. This is de facto what ICER does with its ‘shared savings’ approach. One problem is that diseases—like our made-up horriblitis—that have cost-ineffective treatments covered are often the most severe diseases. My own research (Shafrin et al. 2025) found that the shared savings approach would target “rare, severe, and pediatric diseases”…however these are exactly the types of diseases we want to create new medicines for!
For manufacturers of GreenPill, receiving $501,000 for their treatment may be a good return on investment. However, if GreenPill was compared to BSC, they would only get $300,000 (assuming a $150,000/QALY threshold). That is a 40% price cut. This likely would mean that pharmaceutical companies would be less likely to develop drugs for disease where the current standard of care is cost ineffective.
Implementing the regime proposed by Walton and co-authors would be highly problematic for society. If GreenPill were never developed, patients would be stuck with the less good BlackPill technology (1 QALY gain only) or be relegated to have BSC. Moreover, the analysis does not take into account dynamic pricing; it is likely that BlackPill (and eventually GreenPill) will become generic drugs and deliver low-cost health benefits for many years to come.
While my scenario is hypothetical, this may soon be coming to the UK. The 2025 NHS 10-Year Plan grants new NICE statutory powers to withdraw access to cost-ineffective therapies. It’s not clear for which types of therapies (and when) this would be applied.
I do agree that re-evaluating BlackPill after the initial evaluations is helpful when new information comes to light (e.g., the drug is more/less effective or more/less safe based on real world data). This new evidence could serve to increase or decrease the drug’s price.
If no new information is available, however, and BlackPill was determined to merit coverage–perhaps due to significant unmet need, diseases severity, etc.—it does not make sense to relitigate this. If BlackPill was determined to not merit reimbursement, then it is true that GreenPill would need to have a larger health benefit to be considered cost effective.
The key question HTA bodies aim to look at is: would reimbursing GreenPill make the world better? Because you can pay only $1 for 1 extra QALY, the answer is certainly yes. Asking whether patients with horribilitis deserve to receive any treatment (say with BlackPill) is also a valid question, but one that has previously been answers (during the BlackPill review—either positively or negatively). While HTA bodies performing MTA could reduce cost, it would put a significant impediment to innovation as it would lead to significant pricing uncertainty and would—de facto—result in HTA decisions being changed without new evidence.
In short, I view the Walton et al. approach as highly problematic. What do you think?