December 12, 2024

expert reaction to study looking at population health displacement versus benefits of new drugs recommended by NICE / the NHS in England during 2000–20

A study published in the Lancet looks at the heath impact of new drugs recommended by the National Institute for Health and Care Excellence (NICE) in England between 2000-2020. 

Prof Amitava Banerjee, Professor of Clinical Data Science and Honorary Consultant Cardiologist, Institute of Health Informatics, UCL, said:

“The authors are to be congratulated on a rigorous, comprehensive analysis of this very complicated issue of how we fund drug innovation in the NHS and other health systems, and whether we are getting it right, as measured by population health benefit.

“First, there has been an emphasis nationally on evaluating and funding new cancer (45% of the 339 new drug appraisals) and immunology (21%) treatments quickly.  This research shows that in these areas, in particular, there is a substantial opportunity cost to the NHS.  It is clear that more thought and research needs to systematically look at the gap between surrogate outcomes such as changes in tumour size on imaging versus long-term impact on reducing mortality and on improving quality of life.

“Second, drug innovation and the evaluation of new drugs is not reflecting the population disease burden.  For example, of the 339 new drug appraisals, only 8% were on vascular disease, whereas ischaemic heart disease, stroke and other cardiovascular diseases were some of the highest-ranking causes of mortality and morbidity during the study period (https://assets.publishing.service.gov.uk/media/5e1735f2ed915d3b0b00c7cc/GBD_NHS_England_report.pdf).  We need to consider at system-level, how we encourage drug innovation in the areas where potential and likely population benefit reflects the population health need.

“Third, successive governments have prioritised drug discovery, development and trials as part of their life sciences strategy.  On the basis of these analyses, there is a need to better align the incentivisation and the funding of drug innovation with population health needs and to better maximise the net health system benefit.

“Finally, this analysis relates to drug innovation only.  The incremental population health benefit of other health technologies such as new pacemakers, artificial hips or other medical devices may or may not be even more debatable.  More data and research are required in this area.”

Dr Daniel Howdon, Health Economist, Academic Unit of Health Economics, University of Leeds, said:

“The best-available estimates of the cost-effectiveness of relevant NHS activity in England have, for nearly a decade, implied that the acceptance of new health technologies at prices implied by the NICE cost-effectiveness threshold is highly likely to substantially reduce population health.  Despite comment articles, sometimes authored with explicit acknowledgement of pharmaceutical industry funding, seeking to call these estimates into question, no better estimates have to my knowledge been produced.

“This excellent and well-founded paper uses these well-established estimates to quantify the net health loss that may have arisen from NICE decision-making over a recent 20-year time period.  While it may seem counterintuitive that the provision of a new clinically effective treatment may damage population health, the health gain arising from such a new clinically effective treatment must be seen set against the health loss arising from the denied alternative use of the funds used to provide it – alternative uses that may have provided health in a more cost-effective way.  Where the health loss arising from not providing these alternative services exceeds the health gain from the allocation of funds to a new treatment, this causes a net loss to population health.

“This paper quantifies that the estimated net health loss in England from such decision-making from 2000 to 2020 has been to harm population health by 1,250,000 years of good quality life (estimated in quality-adjusted years, QALYs).  Indeed, for these headline figures, the authors use a relatively conservative estimate of the health-opportunity cost arising in the NHS in England; alternative such estimates would imply that this harm of 1,250,000 QALYs is an underestimate.

“Those who benefit from a newly-provided expensive treatment are identifiable – and are understandably given high prominence in reporting on these decisions.  Those whose health does not benefit from alternative services – services that are not provided due to the use of funds to provide this new treatment – are however no less real for their lack of identifiability, and it would seem ethically problematic to consider their health to be less important for this lack of identifiability.  Unfortunately, however, their relative lack of identifiability can make their suffering less salient and less visible in popular discourse.

“While strictly maximising population health might legitimately not be seen as the only goal of healthcare expenditure, the publication of this paper provides further solid grounding for a serious debate, between honest participants, about whether it might be appropriate to consider lowering the threshold or thresholds used for such decisions in England, and therefore to reduce the maximum reimbursement paid to the pharmaceutical industry.  As the paper alludes to, however, the most common political and lobbying pressures on the threshold are near-universally to push this in the opposite direction – that of, in all probability, harming population health even further.

“Just as medical professionals are mandated to do no harm to the patient in the room in front of them, we might see one valid ethical and practical mandate on policy- and decision-makers of this type to do no harm to population health.  Unfortunately, this paper suggests that this harm over a 20 year period in England has been substantial.”

‘Population-health impact of new drugs recommended by the National Institute for Health and Care Excellence in England during 2000–20: a retrospective analysis’ by Huseyin Naci et al. was published in the Lancet at 23:30 UK time on Thursday 12 December 2024.

Declared interests

Prof Amitava Banerjee: “Amitava Banerjee is Professor of Clinical Data Science at University College London and Honorary Consultant Cardiologist at UCL Hospitals and Barts Health NHS Trusts. There are no relevant conflicts of interest.”

Dr Daniel Howdon: “I am a previous co-author with one of the authors of this piece.  I have no committee memberships to declare.  I have never taken any payment or benefits in kind from any private consultancy or pharmaceutical company.”