Sam Watson’s journal round-up for 11th December 2017

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.

Can incentives improve survey data quality in developing countries?: results from a field experiment in India. Journal of the Royal Statistical Society: Series A Published 17th November 2017

I must admit a keen interest in the topic of this paper. As part of a large project looking at the availability of health services in slums and informal settlements around the world, we are designing a household survey. Much like the Demographic and Health Surveys, which are perhaps the Gold standard of household surveys in low-income countries, interviewers will go door to door to sampled households to complete surveys. One of the issues with household surveys is that they take a long time, and so non-response can be an issue. A potential solution is to offer respondents incentives, cash or otherwise, either before the survey or conditionally on completing it. But any change in survey response as a result of an incentive might create suspicion around data quality. Work in high-income countries suggests incentives to participate have little or no effect on data quality. But there is little evidence about these effects in low-income countries. We might suspect the consequences of survey incentives to differ in poorer settings. For a start, many surveys are conducted on behalf of the government or an NGO, and respondents may misrepresent themselves if they believe further investment in their area might be forthcoming if they are sufficiently badly-off. There may also be larger differences between the interviewer and interviewee in terms of education or cultural background. And finally, incentives can affect the balance between a respondent’s so-called intrinsic and extrinsic motivations for doing something. This study presents the results of a randomised trial where the ‘treatment’ was a small conditional payment for completing a survey, and the ‘control’ was no incentive. In both arms, the response rate was very high (>96%), but it was higher in the treatment arm. More importantly, the authors compare responses to a broad range of socioeconomic and demographic questions between the study arms. Aside from the frequent criticism that statistical significance is interpreted here as the existence of a difference, there are some interesting results. The key observed difference is that in the incentive arm respondents reported having lower wealth consistently across a number of categories. This may result from any of the aforementioned effects of incentives, but may be evidence that incentives can affect data quality and should be used with caution.

Association of US state implementation of newborn screening policies for critical congenital heart disease with early infant cardiac deaths. JAMA [PubMedPublished 5th December 2017

Writing these journal round-ups obviously requires reading the papers that you choose. This can be quite an undertaking for papers published in economics journals, which are often very long, but they provide substantial detail allowing for a thorough appraisal. The opposite is true for articles in medical journals. They are pleasingly concise, but often at the expense of including detail or additional analyses. This paper falls into the latter camp. Using detailed panel data on infant deaths by cause by year and by state in the US, it estimates the effect of mandated screening policies for infant congenital heart defects on deaths from this condition. Given these data and more space, one might expect to see more flexible models than the differences in differences type analysis presented here, such as allowing for state-level correlated time trends. The results seem clear and robust – the policies were associated with a reduction in death from congenital heart conditions by around a third. Given this, one might ask: if it’s so effective, why weren’t doctors doing it anyway? Additional analyses reveal little to no association of the policies with death from other conditions, which may suggest that doctors didn’t have to reallocate their time from other beneficial functions. Perhaps then the screening bore other costs. In the discussion, the authors mention that a previous economic evaluation showed that universal screening was relatively costly (approximately $40,000 per life year saved), but that this may be an overestimate in light of these new results. Certainly then an updated economic evaluation is warranted. However, the models used in the paper may lead one to be cautious about causal interpretations and hence using the estimates in an evaluation. Given some more space the authors may have added additional analyses, but then I might not have read it…

Subsidies and structure: the lasting impact of the Hill-Burton program on the hospital industry. Review of Economics and Statistics [RePEcPublished 29th November 2017

As part of the Hospital Survey and Construction Act of 1946 in the United States, the Hill-Burton program was enacted. As a reaction to the perceived lack of health care services for workers during World War 2, the program provided subsidies of up to a third for building nonprofit and local hospitals. Poorer areas were prioritised. This article examines the consequences of this subsidy program on the structure of the hospital market and health care utilisation. The main result is that the program had the consequence of increasing hospital beds per capita and that this increase was lasting. More specific analyses are presented. Firstly, the increase in beds took a number of years to materialise and showed a dose-response; higher-funded counties had bigger increases. Secondly, the funding reduced private hospital bed capacity. The net effect on overall hospital beds was positive, so the program affected the composition of the hospital sector. Although this would be expected given that it substantially affected the relative costs of different types of hospital bed. And thirdly, hospital utilisation increased in line with the increases in capacity, indicating a previously unmet need for health care. Again, this was expected given the motivation for the program in the first place. It isn’t often that results turn out as neatly as this – the effects are exactly as one would expect and are large in magnitude. If only all research projects turned out this way.

Credits

Chris Sampson’s journal round-up for 4th December 2017

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.

Funding breakthrough therapies: a systematic review and recommendation. Health Policy Published 2nd December 2017

One of the (numerous) financial pressures on health care funders in the West is the introduction of innovative (and generally very expensive) new therapies. Some of these can be considered curative, which isn’t necessarily the best way for manufacturers to create a steady income. New funding arrangements have been proposed to facilitate patient access while maintaining financial sustainability. This article focuses on a specific group of innovative therapies known as ‘Advanced Therapy Medicinal Products’ (ATMPs), which includes gene therapies. The authors conducted a systematic review of papers proposing funding models and considered their appropriateness for ATMPs. There were 48 papers included in the review that proposed payment mechanisms for high-cost therapies. Three top-level groups were identified: i) financial agreements, ii) performance-based agreements, and iii) healthcoin (a tradable currency representing the value of outcomes). The different mechanisms are compared in terms of their feasibility, acceptability, burden, ‘financial attractiveness’ and their appeal to payers and manufacturers. Annuity payments are identified as relatively attractive compared to other options, but each mechanism is summarily shown to be imperfect in the ATMP context. So, instead, the authors propose an ATMP-specific fund. For UK readers, this will likely smell a bit too much like the disastrous Cancer Drugs Fund. It isn’t clear why such a programme would be superior to annuity payments or more inventive mechanisms, or even whether it would be theoretically sound. Thus, the proposal is not convincing.

Supply-side effects from public insurance expansions: evidence from physician labor markets. Health Economics [PubMed] Published 1st December 2017

Crazy though American health care may be, its inconsistency in coverage can make for good research fodder. The Child Health Insurance Program (CHIP) was set up in 1997 and then, when the initial money ran out 10 years later, the program was (eventually) expanded. In this study, the authors use the changes in CHIP to examine the impact of expanded public coverage on provider behaviour, namely; subspecialty training (which could become more attractive with a well-insured customer base), practice setting and prevailing wage offers. The data for the study relate to the physician labour market for New York state for 2002-2013, as collected in the Graduate Medical Education survey. A simple difference-in-differences analysis is conducted with reference to the 2009 CHIP expansion, controlling for physician demographics. Paediatricians are the treatment group and the control group is adult physician generalists (mostly internal medicine). 2009 seems to be associated with a step-change in the proportion of paediatricians choosing to subspecialise – an increased probability of about 8 percentage points. There is also an upward shift in the proportion of paediatricians entering private practice, with some (weak) evidence that there is an increased preference for rural areas. These changes don’t seem to be driven by relative wage increases, with no major change in trends. So it seems that the expanded coverage did have important supply-side effects. But the waters are muddy here. In particular, we have the Great Recession and Obamacare as possible alternative explanations. Though it’s difficult to come up with good reasons for why these might better explain the observed changes.

Reflections on the NICE decision to reject patient production losses. International Journal of Technology Assessment in Health Care [PubMedPublished 20th November 2017

When people conduct economic evaluations ‘from a societal perspective’, this often just means a health service perspective with productivity losses added. NICE explicitly exclude the inclusion of these production losses in health technology appraisals. This paper reviews the issues at play, focussing on the normative question of why they should (or should not) be included. Findings from a literature review are summarised with reference to the ethical, theoretical and policy questions. Unethical discrimination potentially occurs if people are denied health care on the basis of non-health-related characteristics, such as the ability to work. All else equal, should health care for men be prioritised over health care for women because men have higher wages? Are the unemployed less of a priority because they’re unemployed? The only basis on which to defend the efficiency of an approach that includes productivity losses seems to be a neoclassical welfarist one, which is hardly tenable in the context of health care. If we adopt the extra-welfarist understanding of opportunity cost as foregone health then there is really no place for production losses. The authors also argue that including production losses may be at odds with policy objectives, at least in the context of the NHS in the UK. Health systems based on privately-funded care or social insurance may have different priorities. The article concludes that taking account of production losses is at odds with the goal of health maximisation and therefore the purpose of the NHS in the UK. Personally, I think priority setting in health care should take a narrow health perspective. So I agree with the authors that production losses shouldn’t be included. I’m not sure this article will convince those who disagree, but it’s good to have a reference to vindicate NICE’s position.

Credits

Chris Sampson’s journal round-up for 20th November 2017

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.

Effects of health and social care spending constraints on mortality in England: a time trend analysis. BMJ Open [PubMed] Published 15th November 2017

I’d hazard a guess that I’m not the only one here who gets angry about the politics of austerity. Having seen this study’s title, it’s clear that the research could provide fuel for that anger. It doesn’t disappoint. Recent years have seen very low year-on-year increases in public expenditure on health in England. Even worse, between 2010 and 2014, public expenditure on social care actually fell in real terms. This is despite growing need for health and social care. In this study, the authors look at health and social care spending and try to estimate the impact that reduced expenditure has had on mortality in England. The analysis uses spending and mortality data from 2001 onwards and also incorporates mortality projections for 2015-2020. Time trend analyses are conducted using Poisson regression models. From 2001-2010, deaths decreased by 0.77% per year (on average). The mortality rate was falling. Now it seems to be increasing; from 2011-2014, the average number of deaths per year increased by 0.87%. This corresponds to 18,324 additional deaths in 2014, for example. But everybody dies. Extra deaths are really sooner deaths. So the question, really, is how much sooner? The authors look at potential years of life lost and find this figure to be 75,496 life-years greater than expected in 2014, given pre-2010 trends. This shouldn’t come as much of a surprise. Spending less generally achieves less. What makes this study really interesting is that it can tell us who is losing these potential years of life as a result of spending cuts. The authors find that it’s the over-60s. Care home deaths were the largest contributor to increased mortality. A £10 cut in social care spending per capita resulted in 5 additional care home deaths per 100,000 people. When the authors looked at deaths by local area, no association was found with the level of deprivation. If health and social care expenditure are combined in a single model, we see that it’s social care spending that is driving the number of excess deaths. The impact of health spending on hospital deaths was less robust. The number of nurses acted as a mediator for the relationship between spending and mortality. The authors estimate that current spending projections will result in 150,000 additional deaths compared with pre-2010 trends. There are plenty of limitations to this study. It’s pretty much impossible (though the authors do try) to separate the effects of austerity from the effect of a weak economy. Still, I’m satisfied with the conclusion that austerity kills older people (no jokes about turkeys and Christmas, please). For me, the findings also highlight the need for more research in the context of social care, and how we (as researchers) might effectively direct policy to prevent ‘excess’ deaths.

Should cost effectiveness analyses for NICE always consider future unrelated medical costs? BMJ [PubMed] Published 10th November 2017

The question of whether or not ‘unrelated’ future medical costs should be included in economic evaluation is becoming a hot topic. So much so that the BMJ has published this Head To Head, which introduces some of the arguments for and against. NICE currently recommends excluding unrelated future medical costs. An example given in this article is the case of the expected costs of dementia care having saved someone’s life by heart transplantation. The argument in favour of including unrelated costs is quite obvious – these costs can’t be ignored if we seek to maximise social welfare. Their inclusion is described as “not difficult” by the authors defending this move. By ignoring unrelated future costs (but accounting for the benefit of longer life), the relative cost-effectiveness of life-extending treatments, compared with life-improving treatments, is artificially inflated. The argument against including unrelated medical costs is presented as one of fairness. The author suggests that their inclusion could preclude access to health care for certain groups of people that are likely to have high needs in the future. So perhaps NICE should ignore unrelated medical costs in certain circumstances. I sympathise with this view, but I feel it is less a fairness issue and more a demonstration of the current limits of health-related quality of life measurement, which don’t reflect adaptation and coping. However, I tend to disagree with both of the arguments presented here. I really don’t think NICE should include or exclude unrelated future medical costs according to the context because that could create some very perverse incentives for certain stakeholders. But then, I do not agree that it is “not difficult” to include all unrelated future costs. ‘All’ is an important qualifier here because the capacity for analysts to pick and choose unrelated future costs creates the potential to pick and choose results. When it comes to unrelated future medical costs, NICE’s position needs to be all-or-nothing, and right now the ‘all’ bit is a high bar to clear. NICE should include unrelated future medical costs – it’s difficult to formulate a sound argument against that – but they should only do so once more groundwork has been done. In particular, we need to develop more valid methods for valuing quality of life against life-years in health technology assessment across different patient groups. And we need more reliable methods for estimating future medical costs in all settings.

Oncology modeling for fun and profit! Key steps for busy analysts in health technology assessment. PharmacoEconomics [PubMed] Published 6th November 2017

Quite a title(!). The subject of this essay is ‘partitioned survival modelling’. Honestly,  I never really knew what that was until I read this article. It seems the reason for my ignorance could be that I haven’t worked on the evaluation of cancer treatments, for which it’s a popular methodology. Apparently, a recent study found that almost 75% of NICE cancer drug appraisals were informed by this sort of analysis. Partitioned survival modelling is a simple means by which to extrapolate outcomes in a context where people can survive (or not) with or without progression. Often this can be on the basis of survival analyses and standard trial endpoints. This article seeks to provide some guidance on the development and use of partitioned survival models. Or, rather, it provides a toolkit for calling out those who might seek to use the method as a means of providing favourable results for a new therapy when data and analytical resources are lacking. The ‘key steps’ can be summarised as 1) avoiding/ignoring/misrepresenting current standards of economic evaluation, 2) using handpicked parametric approaches for extrapolation in order to maximise survival benefits, 3) creatively estimating relative treatment effects using indirect comparisons without adjustment, 4) make optimistic assumptions about post-progression outcomes, and 5) deny the possibility of any structural uncertainty. The authors illustrate just how much an analyst can influence the results of an evaluation (if they want to “keep ICERs in the sweet spot!”). Generally, these tactics move the model far from being representative of reality. However, the prevailing secrecy around most models means that it isn’t always easy to detect these shortcomings. Sometimes it is though, and the authors make explicit reference to technology appraisals that they suggest demonstrate these crimes. Brilliant!

Credits