Every Monday our authors provide a round-up of the latest peer-reviewed journal publications. We cover all issues of major health economics journals as well as some other notable releases. Visit our Resources page for links to more journals or follow the HealthEconBot. If you’d like to write one of our weekly journal round-ups, get in touch.
Volume 23, Issue 6
The latest issue of Value in Health is bookended with two new definitions. Leading the way is a report from the ISPOR Patient-Centred Special Interest Group. The group set out to define patient engagement in research by reviewing existing definitions. They settle on a definition that incorporates the nature of patient involvement and the value of patients’ input in research. The issue closes with the announcement of a ‘new’ definition of health technology assessment. Somebody should update Wikipedia.
As a pedant, I’m a fan of definitions. But the studies from this issue that interested me most were two on outcome measurement.
I’m currently working on the development of an EQ-5D bolt-on for cognition, so I was inevitably intrigued by a study assessing the acceptability of the EQ-5D(-3L) in people with dementia. The authors compared the EQ-5D with the condition-specific QoL-AD – for which the first (I think) preference-based value set was recently published – using data from a German observational study with 560 patient-carer dyads. Perhaps surprisingly, the EQ-5D performed quite well, correlating with the QoL-AD and differentiating between known groups. But the real challenge in this context is responsiveness to change, which this study doesn’t address. The authors hurtfully refer to the need to study “bold-ons” in this context.
Also on outcome measurement is an article that studies five different preference-based quality of life instruments for use with informal carers: the Carer Experience Scale, CarerQoL, ASCOT-Carer, EQ-5D-5L, and ICECAP-A. The findings are a mixed bag. Responsiveness to change (in hours of care) was pretty poor across the pack. But construct validity was pretty good for the carer-specific measures. In general, the authors find favourable results for the ICECAP-A, which performed as well as the carer-specific measures and better than the EQ-5D-5L. However, I can’t help but be a little suspicious of the pro-ICECAP studies that arise from the centre that developed the ICECAP in the first place.
With more than a dozen studies in this issue on a diverse range of topics, I’ll just focus on a selection of other studies that caught my attention.
I was drawn to a couple of articles in this issue on the subject of evidence-based policy. One study looked at the impact of NIHR-funded trials using information from citations and Altmetric. It’s a good news story, with plenty of evidence that this research is being used to inform both future research studies and policy-making. On a related topic, another study finds that HTA agencies don’t necessarily insist upon evidence from peer-reviewed sources. Depending on who you are, that could be good news or bad. But the study highlights the need for some consistency and transparency in the selection of evidence for decision-making.
If decision modelling is your thing, there are a few useful studies in this issue. One study claims novelty in the use of Markov models that are based on the incidence of hospitalisation. I’m not sure that is novel, but the application to an evaluation using hospital episode statistics is certainly interesting and the authors provide some valuable guidance on the use of such an approach. A discrete event simulation from Thailand also uses real-world data to populate a decision model. The authors evaluate different approaches to treating Alzheimer’s disease and tote their model as an example of the high-quality cost-effectiveness analyses that could be conducted in low- and middle-income countries. And for all of you VoI fans, there’s a study report from the Collaborative Network for Value of Information, who are trying to make your VoI analyses less taxing.
Volume 20, Issue 3
This issue includes the usual mix of opinion and applied analyses that you get in the Expert Reviews. Leading the way is a perspective on ‘value-added medicines’ – a term I still don’t fully understand – and a literature review of trials looking at cardiovascular outcomes from glucose-lowering therapies. Of particular interest to me in this issue are two studies in the context of mental health.
A new standard of care was recently developed in the Netherlands for the prevention and treatment of psychosis. Alongside this, a decision model was developed that goes by the name of PsyMod. This issue includes an article outlining the development of PsyMod, which involved input from a range of stakeholders. It’s a relatively simple model, though some validation is attempted. The authors advertise the model as being freely available to researchers “upon request”. What a shame it isn’t open source. Another study looks at the treatment of schizophrenia in China, comparing two widely used antipsychotics. Based on a Markov model, the study finds that the drug amisulpride is very likely to be cost-effective.
The issue also includes other modelling studies in the contexts of hepatitis C, multiple sclerosis, and venous thromboembolism. One article that stands out as being something a bit different is a study of drug stock management for pharmacies. Based on a study conducted in Indonesia, the authors find that different strategies are optimal depending on whether the drugs in an inventory are stagnant or in shortage.