Sam Watson’s journal round-up for 26th November 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.

Alcohol and self-control: a field experiment in India. American Economic Review Forthcoming

Addiction is complex. For many people it is characterised by a need or compulsion to take something, often to prevent withdrawal, often in conflict with a desire to not take it. This conflicts with Gary Becker’s much-maligned rational theory of addiction, which views the addiction as a choice to maximise utility in the long term. Under Becker’s model, one could use market-based mechanisms to end repeated, long-term drug or alcohol use. By making the cost of continuing to use higher then people would choose to stop. This has led to the development of interventions like conditional payment or cost mechanisms: a user would receive a payment on condition of sobriety. Previous studies, however, have found little evidence people would be willing to pay for such sobriety contracts. This article reports a randomised trial among rickshaw drivers in Chennai, India, a group of people with a high prevalence of high alcohol use and dependency. The three trial arms consisted of a control arm who received an unconditional daily payment, a treatment arm who received a small payment plus extra if they passed a breathalyser test, and a third arm who had the choice between either of the two payment mechanisms. Two findings are of much interest. First, the incentive payments significantly increased daytime sobriety, and second, over half the participants preferred the conditional sobriety payments over the unconditional payments when they were weakly dominated, and a third still preferred them even when the unconditional payments were higher than the maximum possible conditional payment. This conflicts with a market-based conception of addiction and its treatment. Indeed, the nature of addiction means it can override all intrinsic motivation to stop, or do anything else frankly. So it makes sense that individuals are willing to pay for extrinsic motivation, which in this case did make a difference.

Heterogeneity in long term health outcomes of migrants within Italy. Journal of Health Economics [PubMed] [RePEc] Published 2nd November 2018

We’ve discussed neighbourhood effects a number of times on this blog (here and here, for example). In the absence of a randomised allocation to different neighbourhoods or areas, it is very difficult to discern why people living there or who have moved there might be better or worse off than elsewhere. This article is another neighbourhood effects analysis, this time framed through the lens of immigration. It looks at those who migrated within Italy in the 1970s during a period of large northward population movements. The authors, in essence, identify the average health and mental health of people who moved to different regions conditional on duration spent in origin destinations and a range of other factors. The analysis is conceptually similar to that of two papers we discussed at length on internal migration in the US and labour market outcomes in that it accounts for the duration of ‘exposure’ to poorer areas and differences between destinations. In the case of the labour market outcomes papers, the analysis couldn’t really differentiate between a causal effect of a neighbourhood increasing human capital, differences in labour market conditions, and unobserved heterogeneity between migrating people and families. Now this article examining Italian migration looks at health outcomes, such as the SF-12, which limit the explanations since one cannot ‘earn’ more health by moving elsewhere. Nevertheless, the labour market can still impact upon health strongly.

The authors carefully discuss the difficulties in identifying causal effects here. A number of model extensions are also estimated to try to deal with some issues discussed. This includes a type of propensity score weighting approach, although I would emphasize that this categorically does not deal with issues of unobserved heterogeneity. A finite mixture model is also estimated. Generally a well-thought-through analysis. However, there is a reliance on statistical significance here. I know I do bang on about statistical significance a lot, but it is widely used inappropriately. A rule of thumb I’ve adopted for reviewing papers for journals is that if the conclusions would change if you changed the statistical significance threshold then there’s probably an issue. This article would fail that test. They use a threshold of p<0.10 which seems inappropriate for an analysis with a sample size in the tens of thousands and they build a concluding narrative around what is and isn’t statistically significant. This is not to detract from the analysis, merely its interpretation. In future, this could be helped by banning asterisks in tables, like the AER has done, or better yet developing submission guidelines around its use.

Credits

Sam Watson’s journal round-up for 23rd January 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.

Short-term and long-term effects of GDP on traffic deaths in 18 OECD countries, 1960–2011. Journal of Epidemiology and Community Health [PubMedPublished February 2017

Understanding the relationships between different aspects of the economy or society in the aggregate can reveal to us knowledge about the world. However, they are more complicated than analyses of individuals who either did or did not receive an intervention, as the objects of aggregate analyses don’t ‘exist’ per se but are rather descriptions of average behaviour of the system. To make sense of these analyses an understanding of the system is therefore required. On these grounds I am a little unsure of the results of this paper, which estimates the effect of GDP on road traffic fatalities in OECD countries over time. It is noted that previous studies have shown that in the short-run, road traffic deaths are procyclical, but in the long-run they have declined, likely as a result of improved road and car safety. Indeed, this is what they find with their data and models. But, what does this result mean in the long-run? Have they picked up anything more than a correlation with time? Time is not included in the otherwise carefully specified models, so is the conclusion to policy makers, ‘just keep doing what you’re doing, whatever that is…’? Models of aggregate phenomena can be among the most interesting, but also among the least convincing (my own included!). That being said, this is better than most.

Sources of geographic variation in health care: Evidence from patient migration. Quarterly Journal of Economics [RePEcPublished November 2016

There are large geographic differences in health care utilisation both between countries and within countries. In the US, for example, the average Medicare enrollee spent around $14,400 in 2010 in Miami, Florida compared with around $7,800 in Minneapolis, Minnesota, even after adjusting for demographic differences. However, higher health care spending is generally not associated with better health outcomes. There is therefore an incentive for policy makers to legislate to reduce this disparity, but what will be effective depends on the causes of the variation. On one side, doctors may be dispensing treatments differently; for example, we previously featured a paper looking at the variation in overuse of medical testing by doctors. On the other side, patients may be sicker or have differing preferences on the intensity of their treatment. To try and distinguish between these two possible sources of variation, this paper uses geographical migration to look at utilisation among people who move from one area to another. They find that (a very specific) 47% of the difference in use of health care is attributable to patient characteristics. However, I (as ever) remain skeptical: a previous post brought up the challenge of ‘transformative treatments’, which may apply here as this paper has to rely on the assumption that patient preferences remain the same when they move. If moving from one city to another changes your preferences over healthcare, then their identification strategy no longer works well.

Seeing beyond 2020: an economic evaluation of contemporary and emerging strategies for elimination of Trypanosoma brucei gambiense. Lancet Global Health Published November 2016

African sleeping sickness, or Human African trypanosomiasis, is targeted for eradication in the next decade. However, the strategy to do so has not been determined, nor whether any such strategy would be a cost-effective use of resources. This paper aims to model all of these different strategies to estimate incremental cost-effectiveness threshold (ICERs). Infectious disease presents an interesting challenge for health economic evaluation as the disease transmission dynamics need to be captured over time, which they achieve here with a ‘standard’ epidemiological model using ordinary differential equations. To reach elimination targets, an approach incorporating case detection, treatment, and vector control would be required, they find.

A conceptual introduction to Hamiltonian Monte Carlo. ArXiv Published 10th January 2017

It is certainly possible to drive a car without understanding how the engine works. But if we want to get more out of the car or modify its components then we will have to start learning some mechanics. The same is true of statistical software. We can knock out a simple logistic regression without ever really knowing the theory or what the computer is doing. But this ‘black box’ approach to statistics has clear problems. How do we know the numbers on the screen mean what we think they mean? What if it doesn’t work or if it is running slowly, how do we diagnose the problem? Programs for Bayesian inference can sometimes seem even more opaque than others: one might well ask what are those chains actually exploring, if it’s even the distribution of interest. Well, over the last few years a new piece of kit, Stan, has become a brilliant and popular tool for Bayesian inference. It achieves fast convergence with less autocorrelation between chains and so it achieves a high effective sample size for relatively few iterations. This is due to its implementation of Hamiltonian Monte Carlo. But it’s founded in the mathematics of differential geometry, which has restricted the understanding of how it works to a limited few. This paper provides an excellent account of Hamiltonian Monte Carlo, how it works, and when it fails, all replete with figures. While it’s not necessary to become a theoretical or computational statistician, it is important, I think, to have a grasp of what the engine is doing if we’re going to play around with it.

Credits

Paul Mitchell’s journal round-up for 26th December 2016

Every Monday (even if it’s Boxing Day here in the UK) 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.

Out-migration and attrition of physicians and dentists before and after EU accession (2003 and 2011): the case of Hungary. European Journal of Health Economics [PubMedPublished 2nd December 2016

Medical staff migration is an important cross-national policy issue given the international shortage of supply of doctors to meet healthcare demand. This study uses a large administrative survey collected in Hungary from 2004-2011 and focuses on the trends of medical doctors (GPs, specialists, dentists) since Hungary joined the EU in 2004 and the introduction of full freedom of movement between Hungary with Austria and Germany in 2011. The author conducted a time-to-event analysis with monthly collection of data on a person’s occupation used as a guide for outward-migration. A competing-risks model was used to also consider medical doctors exiting the profession, becoming inactive or dying. From the 18,266 medical doctors found in this sample over the nine year period, 12% migrated, 17% exited the profession and 14% became inactive. A five-fold increase in migration was seen when the restrictions on freedom of movement between Hungary and Austria/Germany were lifted, a worrying sign of brain drain from Hungary. For those who stayed but exited the profession, relative income is argued to have been a contributory factor, with incomes increasing by on average 40% in their new line of work (although this does not account for the “thank you money” received by doctors in Hungary for healthcare access). Generous maternity leave was argued to play a key role in absence from employment. A recognised limitation in this study is the inability to conduct robust analysis on the migration patterns of new medical graduates who are likely to be more prone to migration than their established colleagues (estimated to be 40% of all medical graduates in Hungary between 2007-2010 who migrated, before restrictions on freedom of movement between Austria and Germany were lifted). Nonetheless, the study still manages to shine a light on the external (competing against countries with larger economies) but also the internal (“attrition and feminisation of workforce”) challenges to national doctor staffing policy.

Does the proportion of pay linked to performance affect the job satisfaction of general practitioners? Social Science & Medicine [PubMedPublished 24th November 2016

The impact of pay for performance (P4P) on healthcare practice has been subject to much debate surrounding the pros and cons of incentives for medical staff to achieve specific goals. This study focuses on the impact that the introduction of the Quality and Outcomes Framework (QOF) for GPs in the UK in 2004 had on their subsequent job satisfaction. Job satisfaction for GPs is argued to be an important topic area due to it having an important role in retaining GPs and the quality of care they provide to their patients. Using linked data from the the GP Worklife Survey and the QOF, that rewards GPs performance based on clinical, organisation, additional services and patient experience indicators, across three time points (2004, 2005 and 2008), the authors model the relationship between P4P exposure (i.e. the proportion of income related to performance) and job satisfaction. Using a continuous difference-in-difference model with a random effects regression, the authors find that P4P exposure has no significant effect on job satisfaction after 1 and 4 years following the introduction of the QOF P4P system. The introduction of the QOF did lead to a large increase in GP life satisfaction; this is likely to be due to the large increase in average income for GPs following the introduction of QOF. The authors argue that their findings suggest GP job satisfaction is unlikely to be affected by changes in P4P exposure, so long as the final income the GP receives remains constant. Given the generous increases on GP final income from the initial QOF, it remains to be seen how generalisable these results would be to P4P systems that did not lead to such large increases in staff income.

Country-level cost-effectiveness thresholds: initial estimates and the need for further research. Value in Health [PubMed] Published 14th December 2016

National thresholds used to determine if a health intervention is cost-effective have been under scrutiny in the UK in recent years. It has been argued on the grounds of healthcare opportunity costs that the NICE £20,000-30,000 per QALY gained threshold is too high, with an estimate of £13,000 per QALY gain proposed instead. Until now, less attention has been paid to international cost-effectiveness thresholds recommended by the WHO, who have argued for a threshold between one and three times the GDP of a country. This study provides preliminary estimates of cost-effectiveness thresholds across a number of countries with varying levels of national income. Using estimates from the recent £13,000 per QALY gain threshold study in England, a ratio between the supply-side threshold with the consumption value of health was estimated and used as a basis to calculate other national thresholds. The authors use a range of income elasticity estimates for the value placed on a statistical life to take account of uncertainty around these values. The results suggest that even the lower end of the WHO recommended threshold range of 1x national GDP is likely to be an overestimate in most countries. It would appear something closer to 50% of GDP may be a better estimate, albeit with a great amount of uncertainty and variation between high and low income countries. The importance of these estimates according to the authors is that the application of the current WHO thresholds could lead to policies that reduce instead of increase population health. However, the threshold estimates from this study rely on a number of assumptions based on UK data that may not provide an accurate estimate when setting cost-effectiveness thresholds at an international level.

Credits