Our authors provide regular round-ups of the latest peer-reviewed journals. We cover all issues of major health economics journals as well as other notable releases. Visit our journal round-up log to see past editions organised by publication title. If you’d like to write a journal round-up, get in touch.
This issue of Value in Health kicks off with an editorial looking at the growth of patient-preference studies. Specifically, it looks back to the 2008 paper by Bridges et al., Things are looking up since we started listening to patients. The editorial highlights the further positive impact of patient-preference studies on drug development, regulatory decisions, and health technology assessment since this paper’s publication in 2008. This leads us nicely to the key highlight of this edition by the very same Bridges: new ISPOR guidance.
Being simple-minded, I love a bit of guidance. This new ISPOR guidance outlines good practices for enhancing the usefulness and impact of patient-preference studies in decision-making. It includes five key elements: context, purpose, population, method, and impact. The report provides guidelines for patient-preference researchers to collaborate with decision-makers, patients, and other stakeholders, ensuring studies are impactful. Key questions are presented to help stakeholders assess the ongoing impact of preference studies. If you’re planning a patient-preference study, this is the place to start!
For the modellers out there, you’ll be stoked to see quality-adjusted life expectancy (QALE) norms for the English population. This short article provides updated QALE population norms for England based on the EQ-5D-5L instrument, establishing the benchmark against which shortfalls can be assessed. This is in the context of NICE introducing severity-of-disease modifiers that assign greater value to QALY gains for patients with a greater absolute or relative expected shortfall in QALE under the current standard of care. Again, I expect this will get a lot of use by health economists in the near future.
What about real-world data (RWD)? More and more is being collected, but how useful is it from an HTA perspective? We typically rely on well-executed RCTs, which provide evidence of safety and efficacy. However, RCTs may have limited utility in certain instances due to narrow eligibility criteria and highly selected care settings. A paper in this issue by Crown et al. sought to evaluate whether observational analyses using RWD to emulate trials can produce effect estimates similar to those of the trials. The article describes the design of the project, summarises the approaches, and presents feasibility results for the emulations using new-user designs. The TLDR: RWD analyses can complement RCTs by providing insights into treatment effects in more diverse patient populations and clinical practice settings. While RCTs remain the preferred approach, understanding when reliable inferences can be drawn from observational data may prove useful in cases where trials are infeasible.
Elsewhere in the journal was the usual array of economic evaluation results, including various conditions I’ve never heard of and COVID-19 (I have heard of that one). The COVID study was interesting, looking at the distributional impacts of funding COVID treatment in the US. Unsurprisingly, it found that it led to improved overall health and reduced inequalities (given the disproportionate impact of COVID on vulnerable communities, i.e. those least likely to be able to afford treatment).
The methodology section brings us a paper asking whether intention to treat is still the gold standard, or whether HTA agencies should embrace a broader estimands framework. The answer is yes; ITT is the gold standard. It goes a bit deeper than that, giving circumstances where you may want to consider more than one maximand.
One for the child health economists now: a new generic preference-based measure. This one is aimed at 2-4 year-olds and is called the Health Utilities Preschool, or HuPS. Why the little ‘u’ and big ‘S’? This measure is developed from the HUI3. The essential advantage of the HuPS is that measurements are commensurate and continuous with those of HUI3 for group analyses. HuPS, in combination with HUI3, makes the HUI3-based scoring system applicable to the greatest range of ages among the leading child health utility measurement contenders. When I last looked, the HUI measures cost a significant amount to use, so that may put a dampener on its use in the academic environment. Conflict of interest: the CHU-9D has a special place in my heart, so my bias against HUI3 is strong.
In the mapping arena, there was a nice little win for good old regression against the ever ominous ‘machine learning’ using something called Gradient Boosted Tree Approaches – yep, me neither. The main point was that it was worse than regression at predicting health states. I do like to see negative results being published – nice one, Value in Health.
Well done if you’ve read this far; I have run out of steam. There were numerous other papers that I didn’t get around to looking at, so you can browse their titles here.