Alastair Canaway’s journal round-up for 28th November 2016

Every Monday our authors provide a round-up of some of the most recently published peer reviewed articles from the field. We don’t cover everything, or even what’s most important – just a few papers that have interested the author. 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.

The cost-effectiveness of antibiotic prophylaxis for patients at risk of infective endocarditis. Circulation [PubMed] Published 13th November 2016

Did NICE get it wrong? In 2008 NICE recommended stopping antibiotic prophylaxis (AP) for those at risk of infective endocarditis (IE). For those unfamiliar with this research area, AP refers to the use of antibiotics or similar to prevent infection complications. IE is an infection of the endocardial surface of the heart which can have severe, and potentially fatal consequences. NICE stopped the recommendation of AP for those at risk of IE whilst undergoing dental procedures citing lack of evidence of efficacy and cost-effectiveness. This paper sought to fill the void in evidence and conduct an economic evaluation of AP using the latest estimates of efficacy and resource use. The paper constructed a decision analytic model to estimate costs and benefits. Both resource use and adverse event rates were sourced through Hospital Episode Statistics. The results were pretty conclusive: AP was less costly and more effective (than no AP) for all patients at risk of IE. Scenario analyses suggested that AP would have to be substantially less effective than estimated for it to fail on grounds of cost-effectiveness. The paper estimated that the annual savings of reintroducing AP in England would be between £5.5m and £8.2m with a health gain of over 2600 QALYs. Given the low costs of AP, the consequent cost saving and health improvements, perhaps NICE will be persuaded to reconsider their decision.

Maximizing health or sufficient capability in economic evaluation? A methodological experiment of treatment for drug addiction. Medical Decision Making [PubMed] Published 17th November 2016

The standard normative framework for economic evaluation within the UK is extra-welfarism, specifically, using health as the maximand (typically measured using QALYs). Thus, the evaluative space is health, with maximisation as the decision rule. Arguments have been made that health maximisation is not always the most appropriate framework. It has been suggested that the evaluative space be broadened to include capability wellbeing (based on the work of Sen), whilst a minimum threshold approach has been touted as an alternative approach to decision making. Such an approach is egalitarian and aims to ensure all members of society achieve a ‘sufficient’ level of capability wellbeing. This paper reports a pilot trial for the treatment of drug addiction to explore how i) changing the evaluative space to that of capability wellbeing, and ii) switching the decision-making principle to sufficient capability, impacts upon the decisions made. The drug addiction context is particularly pertinent due to non-health spill over impacts to the patient and others. The intervention considers three treatments: treatment as usual (TAU), TAU with social behaviour and network therapy (SBNT) and TAU with goal setting (GS). The two measures of interest within this study are the EQ-5D-5L and the ICECAP-A (capability measure for adults), QALYs and years of full capability (YFC) were calculated. Additionally, years of sufficient capability (YSC) were also calculated, sufficient capability was determined by a score of 33333: ‘a lot’ on each dimension of the ICECAP-A instrument. The study examined four situations: i) broadening the costing perspective from NHS/PSS to government, ii) broadening the evaluative space from QALYs to YFC, iii) broadening both costing perspective and evaluative space, and iv) changing the decision making rule to years of sufficient capability (YSC). The study found that changing from health maximisation to capability maximisation changed the treatment decision, as did changing the perspective: treatment recommendation is sensitive to choice of evaluative space and perspective. In the YSC analysis, the decision remained the same as the YFC analysis. The authors note a number of limitations with their study. The biggest for me, was the sample size of 83 – unsurprising given this was a pilot trial. As a result of the small numbers in each arm (30, 27, and 26) there is a surfeit of uncertainty, and just a handful of extreme cases in any one arm has the potential to change the results, and so it is difficult to draw any firm conclusions from this study. This paper however does provide a good starting point for the novel YFC approach, I’d be very interested in seeing this operationalised in a larger trial.

Does the EQ-5D capture the effects of physical and mental health status on life satisfaction among older people? A path analysis approach. Quality of Life Research [PubMed] Published 19th November 2016

This study sought to identify whether the EQ-5D captures impacts of mental and physical health on life satisfaction (LS) of older adults. This involved a retrospective cohort of 884 patients in Ireland. Path analysis was used to evaluate the direct and indirect effects. The EQ-5D-3L was used to measure health-related quality of life, whilst life satisfaction was measured with the life satisfaction index (LSI). Various specific measures of health status were also measured, e.g. co-morbidity level, activity limitation, and anxiety and depression. Within the analysis a number of assumptions were required, specifically around causation. The overall findings suggest that the EQ-5D-3L sufficiently captures the impact of physical health on life satisfaction, but not mental health. The author’s reflect that this may be due to a fundamental incommensurability of the general public’s preferences (who value the health states for the EQ-5D) and those who experience these health states. The authors conclude that the EQ-5D-3L should be used with caution within economic evaluations, and the use of the EQ-5D will underestimate benefits of treatment to mental health. The authors suggest alternative measures: HUI-3, AQoL and the ICECAP, and advocate their use alongside the EQ-5D within economic evaluation to better capture mental health impacts. A lot of this boils down to existing issues of debate: who should do the valuing (patient vs society), what are we trying to maximise (health vs well-being, or minimum threshold) and are existing measures doing the job they are supposed to be doing (is the EQ-5D fit for purpose). All these are interesting areas and it’s nice to see these issues being pushed to the fore once more.



Meeting round-up: Priorities 2016

This was my first experience of the biennial conference organised by the International Society on Priorities in Health Care. The society was founded in 1996 at the University of Birmingham in the UK and returned to its spiritual home 20 years on. As well as bringing bioethicists, philosophers, economists, health care practitioners and patient advocates together, the conference also saw the combined wits of the Health Service Management Centre (HSMC) and the Health Economics Unit (HEU), co-chaired by Iestyn Williams and Joanna Coast (now at Bristol), who organised a very insightful programme that stimulated plenty of debate between attendants.

After a recent bad experience of plenary talks, Priorities 2016 managed to return my faith in the power of good plenary sessions. The opening session of the conference by Angela Coulter, Rachel Baker and Sally Brearley, focusing on the application and practicalities of incorporating patient views into healthcare decision-making, set the tone for high quality presentations over the three days. Although impossible to summarise all the relevant contributions made with simultaneous sessions throughout, I will focus on my highlights.

Multi-criteria decision analysis (MCDA) is something that has been gaining a lot of attention in health economics, so I jumped at the chance to learn more from some of the key names involved in its use and development. I was slightly surprised then to hear Rob Baltussen make a convincing argument why going beyond the quantification of more than two criteria is likely to muddle more than help decision-making. Instead, he made an argument for a deliberate form of MCDA when presenting decision makers with more than two criteria, sounding somewhat similar to a cost consequence analysis in health economics. This deliberative form of MCDA was also argued to align more closely with Norman Daniels’s accountability for reasonableness priority setting ethical framework.

There were numerous relevant health economics talks of interest. In terms of commissioning health services in England, there was an organised session led by Hugh McLeod on a new project starting in Gloucestershire Clinical Commissioning Group (CCG), who are planning to trial the use of the ICECAP capability measure to aid their decision-making. At the same time of this talk, there was also a Health Foundation sponsored session on how to set priorities across the NHS, with speakers including Cam Donaldson and Muir Gray. By many accounts, it was the highlight of the conference for those who attended.

Other notable health economics sessions looked at how benefits are measured, with Lidia Engel presenting twice from her PhD research, including a helpful conceptual map of the multiple options available when considering how going beyond the quality adjusted life year (QALY) could be operationalised in practice. Yvonne Michel looked at issues in asking patients with spinal cord injury about their mobility in terms of walking, a common feature in health measures used in the generation of QALYs. My talk on how capabilities could be an appropriate evaluative space in renal care also took place in this lively session.

Another session with an economic evaluation focus included a talk by Hareth Al-Janabi on his research looking at incorporating health spillover effects on the family from children with health conditions, with his example drawing from data on children with meningitis in the UK. Lars Schwettmann discussed inconsistencies in the willingness to pay for QALYs in the German sample of the EuroVaQ project. Joseph Millum discussed his attempts to place different values on the disutility of death at different ages of childhood, prompting the largest proportion of hands raised by those in attendance following a presentation that I have seen. There was also a talk from three US-based researchers who presented a systematic review for looking at how social justice could be incorporated into an economic evaluation. This session was chaired by Stirling Bryan, who had previously discussed his recently published paper in Medical Decision Making with Graham Scotland at the conference, on the search for efficiency in current health care provision versus the current focus of the majority of most health economic analysis on new interventions.

It was also a good conference in terms of getting international perspectives on how health economics is used to aid priority setting in different countries. Key debates included the use of health/QALY maximisation alone, versus how it is combined with equity concerns around absolute shortfall as implemented in Norway, presented by Trygve Ottersen and proportional shortfall as implemented in the Netherlands, presented by Werner Brouwer. Another interesting development is the use of income as an equity consideration to be incorporated alongside the health outcome in economic analysis, with Ole Norheim and Richard Cookson working on this new area of research.

The above is only a microcosm of the Priorities 2016 conference through the perspective of one attendant. I would highly recommend keeping your eyes peeled for when this conference comes around again in 2018. It may not have had health economics in the title, but I would highly recommend health economists to attend and share their experience with others in related areas of research and practice at this very worthwhile meeting.

Chris Sampson’s journal round-up for 11th July 2016

Every Monday our authors provide a round-up of some of the most recently published peer reviewed articles from the field. We don’t cover everything, or even what’s most important – just a few papers that have interested the author. 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.

GPs’ implicit prioritization through clinical choices – evidence from three national health services. Journal of Health Economics [RePEcPublished 7th July 2016

Through economic evaluation we inform high-level prioritisation decisions about (for example) which drugs should and should not be available. Meanwhile, GPs are able to prioritise at the individual level through their prescribing behaviour. But do they prioritise? And in what ways? This study reports on a discrete choice experiment carried out with 907 GPs in England, Scotland and Norway to try and elicit prescription behaviour in different decision making contexts. A key aspect that the study considers is the presence of a double agency problem, whereby the GP advocates both maximum patient benefit (‘patient agency’) and cost containment for society (‘social agency’). GPs were asked a generic question about prescribing either ‘Medicine A’ or ‘Medicine B’ and the DCE included 5 attributes: total costs, effect, patient costs, patient preference and physician’s experience. All else equal, GPs in all countries preferred lower total costs. There was variation both within and between countries in the extent to which GPs were willing to accept high societal costs for greater patient benefit. GPs in England seem to exhibit stronger social agency in that they were less willing to accept high costs than GPs in both Norway and Scotland. However, in regard to patient costs and patient preferences, UK GPs were willing to accept greater societal costs. The authors discuss a variety of possible reasons for these findings but suggest that strong governance reinforces social agency, while cultural aspects moderate the effect.

Unrelated future costs and unrelated future benefits: reflections on NICE Guide to the Methods of Technology Appraisal. Health Economics [PubMed] Published 3rd July 2016

NICE would prefer that we disregard UFC. That’s unrelated future costs (not Ultimate Fighting Championship) – for example, the costs of dementia care having prevented death from a heart attack. But the availability of these unrelated treatments will likely confer benefit that is not excluded from the analysis. So it’s easy to see how we could end up with suboptimal allocations of resources. In this editorial, the authors consider the arguments against the inclusion of unrelated future costs, which can be broadly considered as relating to ‘principles’, ‘practicalities’ and ‘implications’. The authors argue that current approaches in principle are no more acceptable than the inclusion of costs and exclusion of benefits, as both are inconsistent in their handling of future payoffs. Practically, the authors argue that it is simpler to incorporate projected future costs than to tease out future benefits. Some have argued that the implications are limited, but the authors highlight that issues such as the level of comorbidity could have a major impact. There’s a lot of research still to be done in this area, but for now we should at least strive for consistency in our handling of future costs and benefits.

The capability approach: a critical review of its application in health economics. Value in Health Published 29th June 2016

Friends know I’ve been guilty of a bit of ICECAP-bashing in the past. Though I like the capability approach in principle, I am not a fan of how it has been applied in health economics. Naturally, I was drawn to this “critical” review. In fact, it was published as a working paper just before I finished writing my chapter for Jeff’s book but I didn’t have time to read it let alone incorporate its findings. So here we are with the real (published) deal. The primary purpose of the review is to evaluate the extent to which current questionnaires (e.g. ICECAP) can actually capture capabilities. The article does an excellent job of concisely identifying fundamental problems in the use of current measures. One issue is that the use of terms like “able to” does not allow for trade-offs between domains. A person may have maximum capability in all domains, but not be able to achieve maximal functionings in all of them simultaneously. As such, unachievable capability sets could be defined. Another problem is that these measures do not capture all of the possible combinations of functionings, only the dominant one. Therefore, these measures fail to capture the key basis for the capability approach – the value in choice. We haven’t yet figured out how to properly value a set, rather than a single combination. The authors suggest a way forward, based on the estimation of ‘approximate capability’. This could identify dominant functionings and the degree of choice. A key benefit of this approach would be the conceptual clarity for which it allows. As I have argued, I think this is the main failure of the application of the capability approach (and indeed health state valuation more broadly) in health economics.

Clinical guidelines: a NICE way to introduce cost-effectiveness considerations? Value in Health Published 28th June 2016

Most UK health economists will be familiar with NICE clinical guidelines. They outline what should (and should not) be taking place as part of care pathways in the NHS. The production of guidelines isn’t (usually) triggered by any new intervention but rather they are designed to improve current standard of care. The recommendations take into account economic considerations. This article outlines some of the advantages of the NICE guidelines programme and describes the role of health economists. One advantage of the guideline development process is that it is a joint enterprise between NICE and the various royal colleges of medicine. But there is also tension in this relationship from an economics perspective as optimal individual patient care may be at odds with broader societal objectives (if you’ve skipped ahead, see the first article summarised above). This article identifies a key advantage of NICE guidelines as being able to make recommendations on disinvestment. A key potential that I see, which isn’t discussed here, is for whole disease modelling studies to be routinely funded as part of the guideline development process.

Photo credit: Antony Theobald (CC BY-NC-ND 2.0)