Journal round-up: The European Journal of Health Economics 24(2)

from The Academic Health Economists’ Blo… at https://bit.ly/3JOD96H on March 20, 2023 at 06:05AM

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.


I’ll be honest; I wasn’t super excited about any of the papers in this issue. But you never know what you might discover, so I thought I’d dig in. This issue’s opening editorial is about GPs and GP training in four European countries. It has little to do with health economics. The authors call for radical and wholesale reform of general practice in Europe along the lines of horizontal integration of care services within community facilities. I don’t really get it. Oh well. Moving on!

Having only recently returned from a EuroQol meeting, I’m still in the mood for research on the EQ-5D and friends, and there are a few papers in this issue that fit the bill. I especially enjoy reading about people’s preferences refusing to behave in the way researchers want or expect, so I was pleased to see an article with ‘lexicographic preferences’ in its title. The article reports on an exploratory study using time trade-off (TTO) experiments to value adult and child health states using the EQ-5D-Y. The authors are seeking an explanation for the tendency for health states to be given higher values for children than adults. The authors’ key suspect is time preferences, and the lexicographic preferences in question arise from the tendency for people to prefer anything to the death of a child. Unfortunately for the authors, they don’t find any satisfactory explanations. Nevertheless, there’s lots of interesting stuff to unpick here.

Sticking with the EQ-5D, we have an article comparing the EQ-5D-3L and 5L across four health conditions in China, as if you needed any more convincing about the 5L. You can probably guess that the researchers found reduced ceiling effects for the 5L and better discriminatory power.

Over now to the EQ-5D’s younger and less popular sibling, the SF-6D, with a mapping study. The article maps the Obesity Problem Scale to the SF-6D using a big dataset (n>36,000) from the Scandinavian Obesity Surgery Registry. As usual in mapping studies, the authors tried and tested a range of options for their algorithm, landing on good old OLS. The mapping model predicts SF-6D index scores. But the inputs to their mapping are not just the summary score from the Obesity Problem Scale; they also include age, sex, BMI, and comorbidities. And we’re offered several models for both pre- and post-surgery. At what point does a mapping algorithm become a predictive model for health state values (‘utilities’) that just happens to include another quality of life instrument? I find it hard to see how this could be widely useful. While we’re on the topic of obesity, this issue also includes a TTO study looking at preferences for different frequencies of GLP-1 receptor agonist injections.

I appreciate a conceptually ambitious systematic review, and this issue includes a good one on the relationship between unemployment and health care use. The authors included 28 studies in their review and plugged 13 of those into a meta-analysis. The meta-analysis suggests that unemployed people are 30% more likely to use services, driven primarily by mental health care. For primary care and hospital care, there was a lot of variation in the findings, with studies showing effects in both directions (and many with no effect at all). Despite the more evident impact on mental health care, it’s important to note that this was based on only five studies. This is a challenging effect to estimate in a meta-analysis, and the authors point to several mechanisms that might be explored with further research.

Another review in this issue assesses whether adopting a societal perspective changes the conclusions of economic evaluations in the context of multiple sclerosis. The short answer is yes, but you probably could have guessed that (why else would the authors bother).

The rest of the papers in this issue related to bed occupancy and hospital quality, costs associated with diabetic foot disease in Italy, cancer care in Portugal, an economic evaluation of ultra-hypo-fractionated radiotherapy for prostate cancer, and a simulation study exploring bias in probabilistic cost-effectiveness analysis.

Alright, point proven; don’t judge an issue by its ToC. I’ve learnt a few things from cracking into this one.


Credits