# Sam Watson’s journal round-up for 9th July 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.

Evaluating the 2014 sugar-sweetened beverage tax in Chile: an observational study in urban areas. PLoS Medicine [PubMedPublished 3rd July 2018

Sugar taxes are one of the public health policy options currently in vogue. Countries including Mexico, the UK, South Africa, and Sri Lanka all have sugar taxes. The aim of such levies is to reduce demand for the most sugary drinks, or if the tax is absorbed on the supply side, which is rare, to encourage producers to reduce the sugar content of their drinks. One may also view it as a form of Pigouvian taxation to internalise the public health costs associated with obesity. Chile has long had an ad valorem tax on soft drinks fixed at 13%, but in 2014 decided to pursue a sugar tax approach. Drinks with more than 6.25g/100ml saw their tax rate rise to 18% and the tax on those below this threshold dropped to 10%. To understand what effect this change had, we would want to know three key things along the causal pathway from tax policy to sugar consumption: did people know about the tax change, did prices change, and did consumption behaviour change. On this latter point, we can consider both the overall volume of soft drinks and whether people substituted low sugar for high sugar beverages. Using the Kantar Worldpanel, a household panel survey of purchasing behaviour, this paper examines these questions.

Everyone in Chile was affected by the tax so there is no control group. We must rely on time series variation to identify the effect of the tax. Sometimes, looking at plots of the data reveals a clear step-change when an intervention is introduced (e.g. the plot in this post), not so in this paper. We therefore rely heavily on the results of the model for our inferences, and I have a couple of small gripes with it. First, the model captures household fixed effects, but no consideration is given to dynamic effects. Some households may be more or less likely to buy drinks, but their decisions are also likely to be affected by how much they’ve recently bought. Similarly, the errors may be correlated over time. Ignoring dynamic effects can lead to large biases. Second, the authors choose among different functional form specifications of time using Akaike Information Criterion (AIC). While AIC and the Bayesian Information Criterion (BIC) are often thought to be interchangeable, they are not; AIC estimates predictive performance on future data, while BIC estimates goodness of fit to the data. Thus, I would think BIC would be more appropriate. Additional results show the estimates are very sensitive to the choice of functional form by an order of magnitude and in sign. The authors estimate a fairly substantial decrease of around 22% in the volume of high sugar drinks purchased, but find evidence that the price paid changed very little (~1.5%) and there was little change in other drinks. While the analysis is generally careful and well thought out, I am not wholly convinced by the authors’ conclusions that “Our main estimates suggest a significant, sizeable reduction in the volume of high-tax soft drinks purchased.”

A Bayesian framework for health economic evaluation in studies with missing data. Health Economics [PubMedPublished 3rd July 2018

Missing data is a ubiquitous problem. I’ve never used a data set where no observations were missing and I doubt I’m alone. Despite its pervasiveness, it’s often only afforded an acknowledgement in the discussion or perhaps, in more complete analyses, something like multiple imputation will be used. Indeed, the majority of trials in the top medical journals don’t handle it correctly, if at all. The majority of the methods used for missing data in practice assume the data are ‘missing at random’ (MAR). One interpretation is that this means that, conditional on the observable variables, the probability of data being missing is independent of unobserved factors influencing the outcome. Another interpretation is that the distribution of the potentially missing data does not depend on whether they are actually missing. This interpretation comes from factorising the joint distribution of the outcome $Y$ and an indicator of whether the datum is observed $R$, along with some covariates $X$, into a conditional and marginal model: $f(Y,R|X) = f(Y|R,X)f(R|X)$, a so-called pattern mixture model. This contrasts with the ‘selection model’ approach: $f(Y,R|X) = f(R|Y,X)f(Y|X)$.

This paper considers a Bayesian approach using the pattern mixture model for missing data for health economic evaluation. Specifically, the authors specify a multivariate normal model for the data with an additional term in the mean if it is missing, i.e. the model of $f(Y|R,X)$. A model is not specified for $f(R|X)$. If it were then you would typically allow for correlation between the errors in this model and the main outcomes model. But, one could view the additional term in the outcomes model as some function of the error from the observation model somewhat akin to a control function. Instead, this article uses expert elicitation methods to generate a prior distribution for the unobserved terms in the outcomes model. While this is certainly a legitimate way forward in my eyes, I do wonder how specification of a full observation model would affect the results. The approach of this article is useful and they show that it works, and I don’t want to detract from that but, given the lack of literature on missing data in this area, I am curious to compare approaches including selection models. You could even add shared parameter models as an alternative, all of which are feasible. Perhaps an idea for a follow-up study. As a final point, the models run in WinBUGS, but regular readers will know I think Stan is the future for estimating Bayesian models, especially in light of the problems with MCMC we’ve discussed previously. So equivalent Stan code would have been a bonus.

This is an economics blog. But focusing solely on economics papers in these round-ups would mean missing out on some papers from related fields that may provide insight into our own work. Thus I present to you a politics and sociology paper. It is not my field and I can’t give a reliable appraisal of the methods, but the results are of interest. In the global fight against non-communicable diseases, there is a range of policy tools available to governments, including the sugar tax of the paper at the top. The WHO recommends a large number. However, there is ongoing debate about whether trade rules and agreements are used to undermine this public health legislation. One agreement, the Technical Barriers to Trade (TBT) Agreement that World Trade Organization (WTO) members all sign, states that members may not impose ‘unnecessary trade costs’ or barriers to trade, especially if the intended aim of the measure can be achieved without doing so. For example, Philip Morris cited a bilateral trade agreement when it sued the Australian government for introducing plain packaging claiming it violated the terms of trade. Philip Morris eventually lost but not after substantial costs were incurred. In another example, the Thai government were deterred from introducing a traffic light warning system for food after threats of a trade dispute from the US, which cited WTO rules. However, there was no clear evidence on the extent to which trade disputes have undermined public health measures.

This article presents results from a new database of all TBT WTO challenges. Between 1995 and 2016, 93 challenges were raised concerning food, beverage, and tobacco products, the number per year growing over time. The most frequent challenges were over labelling products and then restricted ingredients. The paper presents four case studies, including Indonesia delaying food labelling of fat, sugar, and salt after challenge by several members including the EU, and many members including the EU again and the US objecting to the size and colour of a red STOP sign that Chile wanted to put on products containing high sugar, fat, and salt.

We have previously discussed the politics and political economy around public health policy relating to e-cigarettes, among other things. Understanding the political economy of public health and phenomena like government failure can be as important as understanding markets and market failure in designing effective interventions.

Credits

# Chris Sampson’s journal round-up for 1st August 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.

Individualised and personalised QALYs in exceptional treatment decisions. Journal of Medical Ethics [PubMedPublished 22nd July 2016

I’ve written previously about the notion of individualised cost-effectiveness analysis – or iCEA. With the rise of personalised medicine it will become an increasingly important idea. But it’s one that needs more consideration and research. So I was very pleased to see this essay in JME. The starting point for the author’s argument is that – in some cases – people will be denied treatment that would be cost-effective for them, because it has not been judged to be cost-effective for the target population on average. The author’s focus is upon people at the extremes of the distribution in terms of treatment effectiveness or costs: exceptional cases. There are two features to the argument. First, cost-effectiveness should be individualised in the sense that we should be providing treatment according to the costs and effects for that individual. Second, QALYs should be ‘personalised’ in the sense that individual’s own (health) preferences should be used to determine whether or not treatment is cost-effective. The author argues that ‘individual funding requests’ (where patients apply for eligibility for treatment that is not normally approved) represent an ideal context in which to use individualised and personalised QALYs. Unfortunately there are a lot of problems with the arguments presented in this essay, both in terms of their formulation and their practical implications. Some of the ideas are a bit dangerous. That there is no discussion of uncertainty or expectations is telling. If I can find the time I’ll write a full response to the journal. Nevertheless, it’s good to see discussion around this issue.

The value of medicines: a crucial but vague concept. PharmacoEconomics [PubMed] Published 21st July 2016

That we can’t define value is perhaps why the practice of value-based pricing has floundered in the UK. Yes, there’s cost-per-QALY, but none of us really think that’s the end of the value story. This article reports on a systematic review to try and identify how value has been defined in a number of European countries. Apparently none of the identified articles in the published literature included an explicit definition of value. This may not come as a surprise – value is in the eye of the beholder, and analysts defer to decision makers. Some vague definitions were found in the grey literature. The paper highlights a number of studies that demonstrate the ways in which different stakeholders might define value. In the countries that consider costs in reimbursement decisions, QALYs were (unsurprisingly) the most common way of measuring “the value of healthcare products”. But the authors note that most also take into account wider societal benefits and broader aspects of value. The review also identifies safety as being important. The authors seem to long for a universal definition of value, but acknowledge that it cannot be a fixed target. Value is heavily dependent on the context of a decision, so it makes sense to me that there should be inconsistencies. We just need to make sure we know what these inconsistencies are, and that we feel they are just.

The value of mortality risk reductions. Pure altruism – a confounder? Journal of Health Economics Published 19th July 2016

Only the most belligerent of old-school economists would argue that all human choices can be accounted for in purely selfish terms. There’s been much economic research into altruistic preferences. Pure altruism is the idea that people might be concerned with the general welfare of others, rather than just specific factors. In the context of tax-funded initiatives it can be either positive or negative, as people could either be willing to pay more for benefits to other people or less due to a reluctance to enforce higher costs (say nothing of sadism). This study reports on a discrete choice experiment regarding mortality reductions through traffic safety. Pure altruism is tested by the randomised inclusion of a statement about the amount paid by other people. An additional question about what the individual thinks the average citizen would choose is used to identify the importance of pure altruism (if it exists). The findings are both heartening and disappointing. People are considerate of other people’s preferences, but unfortunately they think that other people don’t value mortality reductions as highly as them. Therefore, individuals reduce their own willingness to pay, resulting in negative altruism. Furthermore, the analysis suggests that this is due to (negative) pure altruism because the stated values increase when the notion of coercive taxation is removed.

Realism and resources: towards more explanatory economic evaluation. Evaluation Published July 2016

This paper was doing the rounds on Twitter, having piqued people’s interest with an apparently alternative approach to economic evaluation. Realist evaluation – we are told – is expressed primarily as a means of answering the question ‘what works for whom, under what circumstances and why?’ Economic evaluation, on the other hand, might be characterised as ‘does this work for these people under these circumstances?’ We’re not really bothered why. Realist evaluation is concerned with the theory underlying the effectiveness of an intervention – it is seen as necessary to identify the cause of the benefit. This paper argues for more use of realist evaluation approaches in economic evaluation, providing an overview of the two approaches. The authors present an example of shared care and review literature relating to cost-effectiveness-specific ‘programme theories’: the mechanisms affecting resource use. The findings are vague and inconclusive, and for me this is a problem – I’m not sure what we’ve learned. I am somewhat on the fence. I agree with the people who think we need more data to help us identify causality and support theories. I agree with the people who say we need to better recognise context and complexity. But alternative approaches to economic evaluation like PBMA could handle this better without any express use of ‘realist evaluation’. And I agree that we could learn a lot from more qualitative analysis. I agree with most of what this article’s authors’ say. But I still don’t see how realist evaluation helps us get there any more than us just doing economic evaluation better. If understanding the causal pathways is relevant to decision-making (i.e., understanding it could change decisions in certain contexts) then we ought to be considering it in economic evaluation. If it isn’t then why would we bother? This article demonstrates that it is possible to carry out realist evaluation to support cost-effectiveness analysis, but it isn’t clear why we should. But then, that might just be because I don’t understand realist evaluation.

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

# A taxonomy of behavioural interventions

claimtoken-539efc0fe87b4
Back in March I made a note to myself to write a paper – or, more likely, a blog post – presenting a taxonomy of behavioural interventions. I had gotten tired of everything being called a ‘nudge’ and with debates about whether nudges are ethical. I even bought a copy of Nudge so that I could use it to populate the taxonomy with examples.

Thankfully, someone else was already working on this and has beaten me to it – producing almost exactly what I had in mind. Mira Fischer from the University of Cologne and Sebastian Lotz from Stanford have written a working paper titled ‘Is soft paternalism ethically legitimate? – the relevance of psychological processes for the assessment of nudge-based policies‘. They differentiate between 4 types of behavioural intervention – or ‘nudge’ – and discuss the ethical implications associated with each by considering the psychological processes at play. It’s far better than any blog post I could have written, and I recommend reading it.

Fischer and Lotz’s taxonomy

Consider a utility-maximising individual with 2 choices (A or B), each with 2 possible outcomes (1 and 2), such that the utility associated with choice A would be $U_{A} = \pi_{A1}(u_{A1M}+u_{A1N})+\pi_{A2}(u_{A2M}+u_{A2N})$, where ‘π’ is the probability and ‘u’ the utility of the outcome and the ‘M’ and ‘N’ refers to monetary and non-monetary utility. Based on this, the authors then discuss the ways in which various types of nudge might influence the individual’s choice. The table below is not from the paper and is my interpretation of the taxonomy.

 Type Name Point of influence Means of impact on expected utility of choice Examples 1 ‘discomfort nudge’ choice evaluation non-monetary utility default settings on electronic devices; communication of social norms 2 ‘probability nudge’ choice evaluation subjective probability of realisation informational campaigns 3 ‘indifference nudge’ preference formation monetary or non-monetary utility positioning of healthy/unhealthy products 4 ‘automatism nudges’ ? ? changes in road markings

Is the taxonomy complete and well-defined?

In my opinion, it is not.

I do not believe that Type 4 nudges exist in the way described. The authors use the example of changing road markings to make drivers think they are travelling faster than they actually are and thus reduce their speed. It seems clear to me that this is an example of Type 2; the driver has been made to believe that the probability of them crashing at their current speed is greater than they would otherwise have believed. The idea that there is an ethical difference between nudges to our ‘automatic’ behaviour and nudges to our considered behaviour – given that so much of our behaviour is automatic – I believe is unfounded.

When I was considering writing my own taxonomy of behavioural interventions, I was approaching it from a decision analysis perspective. Simply imagining the structure of an individual’s decision process and considering the different points at which an individual could be influenced. Based on the Thaler/Sunstein definition, a nudge can affect any part of a person’s decision process.

Based on this I believe there are 3 points of influence: i) before an individual’s preferences are defined ii) after the definition of preferences but before the observation of the choice set and iii) once the choice set has been recognised. Once preferences are defined and the choice set has been recognised there are 2 means of influencing choice; utility or probability.

As such, I think the taxonomy should look like this:

 Type Point of influence Means of impact on expected utility of choice Examples A preference formation values/priorities education; positioning of food B choice set observation choice set expansion/compression positioning of food; introduction of cycle lanes C choice evaluation subjective probability informational campaigns D choice evaluation utility defaults; communication of social norms

As the authors outline in their paper, particular nudges will cross type boundaries. I have included the ‘positioning of food’ nudge under 2 types to highlight this. If positioning causes an individual to choose a healthy item – where they otherwise would have chosen a less healthy one – this could either be because they saw the healthy item first or because they simply didn’t see and fully consider the unhealthy option. In the former case Type A is at work, while in the latter case Type B is at work. I believe that educational interventions could fall into any of the above types because they can improve an individual’s ability to satisfy their own preferences. Type D could, of course, include a tax or a subsidy.

Furthermore, the ethical implications may be different depending on whether the impact on types B, C or D is positive or negative, and also whether the impact on utility is monetary or non-monetary, which would increase the total number of types to 9.

I don’t know whether I, the authors or both of us are right, but there’s one thing we can agree on. One nudge isn’t necessarily as ethical as the next, so we need better ways of defining behavioural interventions.