Chris Sampson’s journal round-up for 7th May 2018

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.

Building an international health economics teaching network. Health Economics [PubMedPublished 2nd May 2018

The teaching on my health economics MSc (at Sheffield) was very effective. Experts from our subdiscipline equipped me with the skills that I went on to use on a daily basis in my first job, and to this day. But not everyone gets the same opportunity. And there were only 8 people on my course. Part of the background to the new movement described in this editorial is the observation that demand for health economists outstrips supply. Great for us jobbing health economists, but suboptimal for society. The shortfall has given rise to people teaching health economics (or rather, economic evaluation methods) without any real training in economics. The main purpose of this editorial is to call on health economists (that’s me and you) to pull our weight and contribute to a collective effort to share, improve, and ultimately deliver high-quality teaching resources. The Health Economics education website, which is now being adopted by iHEA, should be the starting point. And there’s now a Teaching Health Economics Special Interest Group. So chip in! This paper got me thinking about how the blog could play its part in contributing to the infrastructure of health economics teaching, so expect to see some developments on that front.

Including future consumption and production in economic evaluation of interventions that save life-years: commentary. PharmacoEconomics – Open [PubMed] Published 30th April 2018

When people live longer, they spend their extra life-years consuming and producing. How much consuming and producing they do affects social welfare. The authors of this commentary are very clear about the point they wish to make, so I’ll just quote them: “All else equal, a given number of quality-adjusted life-years (QALYs) from life prolongation will normally be more costly from a societal perspective than the same number of QALYs from programmes that improve quality of life”. This is because (in high-income countries) most people whose life can be extended are elderly, so they’re not very productive. They’re likely to create a net cost for society (given how we measure value). Asserting that the cost is ‘worth it’ at any level, or simply ignoring the matter, isn’t really good enough because providing life extension will be at the expense of some life-improving treatments which may – were these costs taken into account – improve social welfare. The authors’ estimates suggest that the societal cost of life-extension is far greater than current methods admit. Consumption costs and production gains should be estimated and should be given some weight in decision-making. The question is not whether we should measure consumption costs and production gains – clearly, we should. The question is what weight they ought to be given in decision-making.

Methods for the economic evaluation of changes to the organisation and delivery of health services: principal challenges and recommendations. Health Economics, Policy and Law [PubMedPublished 20th April 2018

The late, great, Alan Maynard liked to speak about redisorganisations in the NHS: large-scale changes to the way services are organised and delivered, usually without a supporting evidence base. This problem extends to smaller-scale service delivery interventions. There’s no requirement for policy-makers to demonstrate that changes will be cost-effective. This paper explains why applying methods of health technology assessment to service interventions can be tricky. The causal chain of effects may be less clear when interventions are applied at the organisational level rather than individual level, and the results will be heavily dependent on the present context. The author outlines five challenges in conducting economic evaluations for service interventions: i) conducting ex-ante evaluations, ii) evaluating impact in terms of QALYs, iii) assessing costs and opportunity costs, iv) accounting for spillover effects, and v) generalisability. Those identified as most limiting right now are the challenges associated with estimating costs and QALYs. Cost data aren’t likely to be readily available at the individual level and may not be easily identifiable and divisible. So top-down programme-level costs may be all we have to work with, and they may lack precision. QALYs may be ‘attached’ to service interventions by applying a tariff to individual patients or by supplementing the analysis with simulation modelling. But more methodological development is still needed. And until we figure it out, health spending is likely to suffer from allocative inefficiencies.

Vog: using volcanic eruptions to estimate the health costs of particulates. The Economic Journal [RePEc] Published 12th April 2018

As sources of random shocks to a system go, a volcanic eruption is pretty good. A major policy concern around the world – particularly in big cities – is the impact of pollution. But the short-term impact of particulate pollution is difficult to identify because there is high correlation amongst pollutants. In this study, the authors use the eruption activity of Kīlauea on the island of Hawaiʻi as a source of variation in particulate pollution. Vog – volcanic smog – includes sulphur dioxide and is similar to particulate pollution in cities, but the fact that Hawaiʻi does not have the same levels of industrial pollutants means that the authors can more cleanly identify the impact on health outcomes. In 2008 there was a big increase in Kīlauea’s emissions when a new vent opened, and the level of emissions fluctuates daily, so there’s plenty of variation to play with. The authors have two main sources of data: emergency admissions (and their associated charges) and air quality data. A parsimonious OLS model is used to estimate the impact of air quality on the total number of admissions for a given day in a given region, with fixed effects for region and date. An instrumental variable approach is also used, which looks at air quality on a neighbouring island and uses wind direction to specify the instrumental variable. The authors find that pulmonary-related emergency admissions increased with pollution levels. Looking at the instrumental variable analysis, a one standard deviation increase in particulate pollution results in 23-36% more pulmonary-related emergency visits (depending on which measure of particulate pollution is being used). Importantly, there’s no impact on fractures, which we wouldn’t expect to be influenced by the particulate pollution. The impact is greatest for babies and young children. And it’s worth bearing in mind that avoidance behaviours – e.g. people staying indoors on ‘voggy’ days – are likely to reduce the impact of the pollution. Despite the apparent lack of similarity between Hawaiʻi and – for example – London, this study provides strong evidence that policy-makers should consider the potential savings to the health service when tackling particulate pollution.

Credits

Chris Sampson’s journal round-up for 6th June 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.

Development of a measure (ICECAP-Close Person Measure) through qualitative methods to capture the benefits of end-of-life care to those close to the dying for use in economic evaluation. Palliative Medicine [PubMedPublished 3rd June 2016

I’ve written somewhat critically about the use of ICECAP measures in economic evaluation. One thing that they do well is the development of the measures themselves. Here we have the latest: the ICECAP-CPM. End of life care is one area in which evaluations should take into account benefits (or burdens) experienced by family members or close persons, and government publications support this view. This paper reports on the development of a new measure to capture such spillover effects in terms of capabilities. People (n=27) who were either recently bereaved or had loved ones or relatives currently receiving end of life care were interviewed, and qualitative analysis was used to determine which attributes should be included in the measure. In the end, 6 attributes were selected: i) communication with those providing care services, ii) practical support, iii) privacy and space, iv) emotional support, v) preparing and coping, and vi) emotional distress. The measure allows for 5 levels within each of these domains. Notwithstanding my concerns about the use of ICECAP measures, the ICECAP-CPM represents an innovative and potentially useful way of capturing some of the wider benefits of end of life care. I expect it will start to be included in evaluative studies, though it may yet prove more useful as a routinely collected close-person-reported outcome measure. But the ICECAP-CPM also creates new questions. Who counts as a close person? Do we treat close person capabilities as additive? Do we really want to trade benefits to close persons against benefits to patients? This will keep researchers busy and means that the ICECAP-CPM (like all new outcome measures) should for now be used with caution.

QALY gain and health care resource impacts of air pollution control: A Markov modelling approach. Environmental Science & Policy Published 25th May 2016

Here’s something you don’t see every day: an evaluation of the health impacts and costs of a policy that falls outside of the remit of the Department of Health. The study reports on a Markov model based on 3 diseases: chronic obstructive pulmonary disease, coronary heart disease and lung cancer. The model is used to estimate the impact of changes in air quality for 40-90 year olds, and the main novelty of the study is the use of QALYs in this context. A 9% reduction in small particulate matter concentrations in England and Wales – in line with current targets – is evaluated. Data were taken from a variety of national sources to incorporate the impact of air quality on the risk of disease onset and death, and disease-related health service use. Health-related quality of life estimates were based on published EQ-5D index scores. The main finding is that the improvement in air quality would generate 540,000 QALYs. Due to improved longevity, additional health care costs would amount to around £263 million. Results are also presented by age and sex, though I can’t see why this would be important. The QALY benefit is on average greater for men than women and the savings from reduced morbidity are (of course) greater for younger people. On balance, the model probably produces underestimates as it does not include all possible health impacts of air pollution.

The costs of inequality: whole-population modelling study of lifetime inpatient hospital costs in the English National Health Service by level of neighbourhood deprivation. Journal of Epidemiology & Community Health [PubMedPublished 17th May 2016

People in more deprived areas have worse health outcomes and do not make equivalent use of health services. Some services are more heavily used in more deprived areas, while others are less likely to be used. The net impact on costs is therefore not clear, so this study looks at lifetime inpatient hospital costs for the English population. Hospital Episode Statistics for 2011/12 are used and analysed based on the 32,482 lower-layer super output areas (LSOA). Deprivation is measured using the index of multiple deprivation (IMD) and LSOAs are grouped by IMD quintile, age and sex. Average costs are estimated and then compared with the least deprived quintile. Mortality rates and survival curves are estimated. For 0-60 year olds, a greater number of episodes were observed with greater deprivation. For people over 75, the trend reversed; this seems to be caused by fewer people in deprived areas surviving into old age. Much of the inequality was driven by differences in emergency admissions, with a 71% higher rate for the most deprived compared with the least deprived quintile. The pattern for costs is very similar, so average cumulative lifetime costs were greater for people in the more deprived areas. It’s a bit of a leap to assert that the difference in costs is because of deprivation (and could be remedied by removing the inequality), but if we make that leap the total ‘cost’ of inequality in these terms was £4.8 billion in 2011/12.