Rita Faria’s journal round-up for 20th January 2020

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.

Opportunity cost neglect in public policy. Journal of Economic Behavior & Organization Published 10th January 2020

Opportunity cost is a key concept in economics, and health economics is no exception. We all agree that policy-makers should consider the opportunity cost alongside the benefits of the various policy options. The question is… do they? This fascinating paper by Emil Persson and Gustav Tinghög suggests that they may not.

The paper reports two studies: one in the general population, and the other in a sample of experts on priority setting in health. In both studies, the participants were asked to choose between making a purchase or not, and were randomised to choices with and without a reminder about the opportunity cost. The reminder consisted of the “no” option having the comment “saving the money for other purchases“. There were choices about private consumption (e.g. buying a new mobile phone) and health care policy (e.g. funding a new cancer screening programme).

In the study in the general population, the participants were 6% less likely to invest in public policies if they were reminded of the opportunity cost. There was no effect in private consumption decisions. In the study with experts on health care priority setting, the participants were 10% less likely to invest in a health programme when reminded about opportunity costs, although the result was “marginally significant“. There was a numerical difference of -6% regarding private consumption, but non-significant. The authors concluded that both lay people and experts neglect opportunity cost in public policy but much less so in their own private consumption decisions.

It struck me that this effect is driven by quite a small difference between the scenarios – simply stating that choosing to reject the policy means that the money will be saved for future purchases. I wonder about how this information affects the decision. After all, the scenarios only quantify the costs of the policy, without information about the benefits or the opportunity cost. For example, the benefits of the cancer screening programme were that “cancer treatment will be more effective, lives will be saved and human suffering will be avoided” and the cost was 48 million SEK per year. Whether this policy is good or bad value for money all depends on how much suffering it avoids and how much would be avoided by investing the money in something else. It would be interesting to have coupled the survey with interviews to understand how the participants interpreted the information and their decision making process.

On a wider note, this paper agrees with health economists’ anecdotal experience that policy-makers find it hard to think about opportunity cost. This is not helped by settings where they hear about the experience of people who would benefit from a positive recommendation and from doctors who would like to have the new drug in their medical arsenal, but not much about the people who will bear the opportunity cost. The message is clear: we need to do better at communicating the opportunity cost of public policies!

Assessment of progression-free survival as a surrogate end point of overall survival in first-line treatment of ovarian cancer. JAMA Network Open [PubMed] Published 10th January 2020

A study about the relationship between progression-free survival and overall survival may seem an odd choice for a health economics journal round-up, but it is actually quite relevant. In cost-effectiveness analysis of new cancer drugs, the trial primary endpoint may be progression-free survival (PFS). Data on overall survival (OS) may be too immature to assess the treatment effect or for extrapolation to the longer term. To predict QALYs and lifetime costs with and without the new drug, the cost-effectiveness model may need to assume a surrogate relationship between PFS and OS. That is, that an effect on PFS is reflected, to some extent, in an effect on OS. The question is, how strong is that surrogate relationship? This study tries to answer this question in advanced ovarian cancer.

Xavier Paoletti and colleagues conducted a systematic review and meta-analysis using individual patient data from 11,029 people who took part in 17 RCTs of first-line therapy in advanced ovarian cancer. They assessed the surrogate relationship at the individual and at the trial-level. The individual-level surrogate relationship refers to the correlation between PFS and OS for the individual patient. As the authors note, this may only reflect that people who have longer life expectancy also take longer to progress. At the trial-level, they looked at the correlation between the hazard ratio (HR) on OS and the HR on PFS. This reflects how much of the effect on OS could be predicted by the effect on PFS. They used the surrogate criteria proposed by the Follicular Lymphoma Analysis of Surrogacy Hypothesis initiative. As this is outside my area of expertise, I won’t comment on the methodology.

One of their results is quite striking: in 16/17 RCTs, the experimental drug did not have HRs for PFS and OS statistically different from the control. This means that there have not been any new drugs with statistically significant benefits! In terms of the surrogate relationship, they found that there is an individual-level association – that is, people who take longer to progress also survive for longer. In contrast, they did not find a surrogate relationship between PFS and OS at the trial-level. Given that the HRs were centred around 1, the poor correlation may be partly due to the lack of variation in HRs rather than a poor surrogate relationship.

Now the challenge remains in cost-effectiveness modelling when OS is immature. Extrapolate OS with high uncertainty? Use a poor surrogate relationship with PFS? Or formal expert elicitation? Hopefully methodologists are looking into this! In the meantime, regulators may wish to think again about licensing drugs with evidence only on PFS.

After 20 years of using economic evaluation, should NICE be considered a methods innovator? PharmacoEconomics [PubMed] Published 13th January 2020

NICE is currently starting a review of the methods and process for health technology assessment. Mark Sculpher and Steve Palmer take this opportunity to reflect on how NICE’s methods have evolved over time and to propose areas ripe for an update.

It was very enjoyable to read about the history of the Methods Guide and how NICE has responded to its changing context, responsibilities, and new challenges. For example, the cost-effectiveness threshold of £20k-£30k/QALY was introduced by the 2004 Methods Guide. This threshold was reinforced by the 2019 Voluntary Scheme for Branded Medicines Pricing and Access. The funny thing is, although NICE is constrained to the £20k-£30k/QALY threshold, the Department of Health and Social Care routinely uses Claxton et al’s £13k/QALY benchmark.

Mark and Steve go through five key topics in health technology assessment to pick out the areas that should be considered for an update. The topics are: health measurement and valuation, broader benefits, perspective, modelling, and uncertainty.  For example, whether/how to consider caregiver burden, and benefits (and opportunity costs) on caregivers, guidance on model validation, and formal incorporation of value of information methods. These are all sorely needed and would definitely cement NICE’s position as the international standard-setter for health technology assessment.

Beyond NICE and the UK, I found that this paper provides a good overview on hot topics in cost-effectiveness for the next few years. Must read for cost-effectiveness analysts!


Thesis Thursday: Caroline Chuard

On the third Thursday of every month, we speak to a recent graduate about their thesis and their studies. This month’s guest is Dr Caroline Chuard who has a PhD from the University of Zurich. If you would like to suggest a candidate for an upcoming Thesis Thursday, get in touch.

Three essays on the health effects of family policies
Hannes Schwandt, Josef Zweimüller
Repository link

Is there a strong health economics evidence base on family policies?

The literature on parental leave and family health is relatively young. This literature emphasises that the returns depend on several key features. First, the timing of measurement matters. Therefore, the effects differ according to whether they are measured in the short- versus long-run. Second, the initial level of parental leave and the extent to which parental leave is increased are both key influencing factors. As such, an introduction is more beneficial than an increase at an already generous level of parental leave. Third, the results depend on the targeted group.

But keep in mind that the effects of family policies on health outcomes are just one part of a large literature that studies the effect on other outcomes such as maternal labour market outcomes, fertility, and child cognitive and non-cognitive development (e.g. Ruhm (2000), Lalive and Zweimüller (2009), Baker and Milligan (2008), Dustmann and Schönberg (2012), Lalive et al. (2014), Carneiro et al. (2015), Dahl et al. (2016), Danzer and Lavy (2018), Butikofer et al. (2018) and many more which have recently been reviewed by Olivetti and Petrongolo (2017) and Rossin-Slater (2018)).

What policy changes were you able to evaluate in your research?

I exploit two types of family policy changes in two countries. On the one hand, I use three changes in parental leave duration in Austria and, on the other hand, I use cantonal variation in family allowances across Switzerland.

More specifically, Austria increased parental leave by 1 year to 2 years in July 1990. This was partially reversed again in July 1996, by exclusively reserving 6 months to fathers so that maternal leave was essentially reduced to 1.5 years. Finally, in July 2000, there was another large extension in paid parental leave by 1 year to 2.5 years. Enforcement of all these changes was very strict, changing from one day to another depending on giving birth in June or July. This sharp discontinuity allows me to employ a regression discontinuity design.

In the case of Switzerland, I analyse the impact of birth allowances (so-called baby bonuses) on fertility, newborn health and birth scheduling. I exploit a unique quasi-experimental setting of Switzerland’s family allowances system. In this system, cantons are free to choose whether they want to implement birth allowances and how much they want to pay. During the last 50 years, 11 cantons have introduced a baby bonus, all increase the amount paid thereafter, and two cantons even abolished the baby bonus after all. This gives rise to a lot of cantonal variation. Thus, I use a difference-in-differences setting where I can analyse both the introduction and the intensity of the treatment.

What were the key strengths of the data sets that you used?

For all my studies I rely on administrative data. Thus, I can use the universe of observations delivered with high quality, as both Austria and Switzerland have very reliable administrative data.

In the Austrian case, I can even combine several different data sets. Namely, I use the Austrian Social Security Database (ASSD), which covers the complete working history of every worker in Austria. The ASSD covers every birth of employed mothers and their actual duration of parental leave. I can link the ASSD to the Austrian Birth Register (ABR) recording newborn health outcomes and additional individual-level characteristics of the mother. Finally, for a part of Austria, I additionally merge the data to health outcomes recorded in the health insurance data. This data set records every outpatient doctor visit, prescribed medication, and hospital stays including diagnosis code.

All of this, together, gives a huge variety of different variables on an individual basis allowing me to study a broad set of outcomes (such as health outcomes next to the directly targeted labour market outcomes). Furthermore, the detailed level of information allows me to study the impact of labour market behaviour on two margins—the extensive margin of mothers who choose to work or not and the intensive margin of how much mothers choose to work. The richness of the data also makes it possible to analyse heterogeneous effects across mothers and by work environment.

Did the policies achieve what they were designed to achieve?

This is a little hard to tell from looking at my results only. For example, in Austria the initial increase of parental leave duration by 1 year was introduced so that fathers could take up to 6 months of the full duration. This policy reform was a result of parliamentary procedural requests which wanted to introduce paternal leave. Due to the flat benefit structure almost no fathers were taking up parental leave, which essentially resulted in an increase of maternal leave from 1 to 2 years and, ultimately, led to the second policy change by exclusively reserving 6 months out of the total 2 years for fathers.

However, what I want to mention here, note that I explicitly evaluated side effects. All three chapters of my dissertation highlight the importance of studying alternative and indirect outcome measures in addition to the direct measures targeted by policymakers.

For example, in the Swiss study, we only find little fertility effects, the directly targeted outcome measure of birth allowances, but a sizable and significant reduction in the stillbirth rate as well as a positive impact on birth weight. A policymaker, who would now only study fertility, would argue that birth allowances were expensive to implement with little to no result, which, however, does not capture the full story.

Is there heterogeneity in how family policy reforms affect families?

The answer depends on the person affected and the studied outcome. For example, the Austrian parental leave duration reform affects maternal work behaviour during pregnancy regardless of the mother’s socioeconomic background and the industry. This change in prenatal maternal work status doesn’t affect newborn health at all.

However, when I study the same reforms with respect to maternal health, there is substantial heterogeneity. The initial increase in leave length is especially good for low-wage and unmarried mothers. Reducing leave duration harms mothers with unhealthy babies, proxied by a preterm birth or low birth weight baby. Substantially increasing leave duration is, though, especially bad for maternal health of those mothers who already suffered from mental diseases pre-birth. Also, for the paper on the Swiss baby bonus, we find a more beneficial impact in the decline of stillbirths for low socioeconomic status mothers.

Based on your research, how would you design parental leave policies?

With my research, I tried to give a more complete picture on the impact of family policies by taking into account health outcomes which have vastly been neglected so far. Nevertheless, for a policy recommendation it is crucial to take the findings from the previous literature into account.

Firstly, introducing parental leave has generally been shown to be very beneficial for the cognitive development of children (Carneiro et al., 2015). Secondly, these returns are, however, quickly declining (Butikofer et al., 2018). In combination with my findings of no impact of working during pregnancy on child health and a negative impact of too long parental leave policies for maternal health (Chuard, 2018), I would clearly put the focus on mandatory leave in the first months of a newborn’s life. While this might seem obvious for many European countries, this is still not the case in the US. And even Europe might face the risk on the other end of the parental leave duration scale. Many European countries tend to expand leave rather generously both pre- and post-natal, which seems from my research not necessary (always keep in mind, these policies are extremely expensive) and could potentially even be harmful in the long-run.

Agent relationships and information asymmetries in public health

The agent relationship and information asymmetry are two features of healthcare economics – but how do they apply to public health policy around processed foods?

Why is health different to other goods?

Arrow’s 1963 seminal paper helped lay the foundations for health economics as a discipline. The Nobel-winning economist talks about what makes healthcare different to other types of market goods. Two of the principal things are agent relationship – that a clinician often makes choices on behalf of a patient (Arrow calls them a “controlling agent”); and information asymmetry – that a clinician knows more than the patient (“informational inequality”). Whereas if someone is buying a new car, they make their own choices, and they might read up on the extensive information available so that they are reasonably knowledgeable about what to buy. These two factors have evolved and possibly diminished over time, especially among highly educated people in developed countries; people often have more choice over their treatment options, and some people have become ‘expert patients‘. Patients may no longer believe that the Götter in Weiß (Gods dressed in white) always know best.

Agent relationship and information asymmetry are features of healthcare economics but they also apply to public health economics. But where people accept clinicians as having more knowledge or acting as their agent, people don’t always accept advice on food from public health policy makers in the same way. People may think, “well I know how to buy a bottle of beer, or a can of coke, or a pizza”, and may not see any potential information asymmetry. Some of it might be ‘akrasia’ – they know that food is unhealthy, but they eat it anyway because it is delicious! However, few people may be aware that poor diet and obesity are the biggest risk factors for ill health and mortality in England.

People might ask “why should a nanny state agent make my food or drink decisions for me?” Of course, this is ignoring the fact that processed food companies might be making those decisions, and reinforcing them using huge marketing budgets. Consumers see government influences but they don’t always see the other information asymmetry and agent relationship; the latent power structures that drive their behaviours – from the food, drinks, alcohol industry, etc. Unsustainable food systems that promote obesity and poor health might be an example of market failure or a tragedy of the commons. The English food system has not moved on enough from post-world war 2 rationing, where food security was the major concern; it still has an objective to maximise calorie supply across the population, rather than maximise population health.

Some of the big UK misselling scandals like mortgage PPI are asymmetries. You could argue that processed foods (junk food high in salt, sugar and saturated fats) might be missold because producers try to misrepresent the true mix of ingredients – for example, many advertisements for processed foods try to misrepresent their products by showing lots of fresh fruit and vegetables. Even though processed foods might have ingredients listed, people have an information asymmetry (or at least, a deficit around information processing) around truly understanding the amount of hidden salt and sugars, because they may assume that the preparation process is similar to a familiar home cooked method. In the US there have been several lawsuits from consumers alleging that companies have misled them by promoting products as being wholesome and natural when they are in fact loaded with added sugars.

The agent relationship and information asymmetry as applied to food policy and health.

How acceptable are public health policies?

A 2012 UK poll carried out by YouGov, funded by the Adam Smith Institute (a right wing free market think tank), found that 22% of people in England thought that the government should tell people what to eat and drink, and 44% thought the government should not. Does this indicate a lack of respect for public health as a specialism? But telling people what to eat and drink is not the same as enacting structural policies to improve foods. Research has shown that interventions like reducing salt in processed foods in the UK or added sugar labelling in the US could be very cost effective. There has been some progress with US and UK programmes like the sugary drinks industry levy, which now has a good level of public support. But voluntary initiatives like the UK sugar reduction programme have been less effective, which may be because they are weakly enforced, and not ambitious enough.

A recent UK study used another YouGov survey to assess the public acceptability of behavioural ‘nudge’ interventions around tobacco, alcohol, and high-calorie snack foods. It compared four types of nudges: labelling (adding graphic warning labels to products); size (reducing pack size of snacks, serving size for alcohol, and number of cigarettes in packets for tobacco); tax (increasing the price to consumers); and availability (banning sales from corner shops). This study found that labelling was the most acceptable policy, then size, tax, and availability. It found that targeting tobacco use was more acceptable than targeting alcohol or food. Acceptability was lower in people who participated in the relevant behaviour regularly, i.e. smokers, heavy drinkers, frequent snackers.

What should public health experts do?

Perhaps public health experts need to do more to enhance their reputation with the public. But when they are competing with a partnership between right wing think tanks, the media and politicians, all funded by big food, tobacco and alcohol, it is difficult for public health experts to get their message out. Perhaps it falls to celebrities and TV chefs like Jamie Oliver and Hugh Fearnley-Whittingstall to push for healthy (and often more sustainable) food policy, or fiscal measures to internalise the externalities around unhealthy foods. The food industry falls back on saying that obesity is complex, exercise is important as well as diet, and more research is needed. They are right that obesity is complex, but there is enough evidence to act. There is good evidence for an ‘equity effectiveness hierarchy‘ where policy-level interventions are more effective at a population level, and more likely to reduce inequalities between rich and poor, than individual, agentic interventions. This means that individual education and promoting exercise may not be as effective as national policy interventions around food.

The answer to these issues may be in doing more to reduce information asymmetries by educating the public about what is in processed food, starting with schools. At the same time understanding that industries are not benevolent; they have an agent relationship in deciding what is in the foods that arrive at our tables, and the main objectives for their shareholders are that food is cheap, palatable, and with a long shelf life. Healthy comes lower on the list of priorities. Government action is needed to set standards for foods or make unhealthy foods more expensive and harder to buy on impulse, and restrict marketing, as previously done with other harmful commodities such as tobacco.

In conclusion, there are hidden agent relationships and information asymmetries around public health policies, for instance around healthy food and drinks. Public health can potentially learn from economic instruments that have been used in other industries to mitigate information asymmetries and agent relationships. If Government and the food industry had shared incentives to create a healthier population then good things might happen. I would be curious to know what others think about this!