Sam Watson’s journal round-up for 3rd June 2019

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.

Limits to human life span through extreme value theory. Journal of the American Statistical Association [RePEc] Published 2nd April 2019

The oldest verified person to have ever lived was Jeanne Calment who died in 1997 at the superlative age of 122. No-one else has ever been recorded as living longer than 120, but there have been perhaps a few hundred supercentarians over 110. Whenever someone reaches such a stupendous age, some budding reporter will ask them what the secret was. They will reply that they have stuck to a regimen of three boiled eggs and a glass of scotch every day for 80 years. And this information is of course completely meaningless due to survivorship bias. But as public health and health care improves and with it life expectancy, there remains the question of whether people will ever exceed these extreme ages or whether there is actually a limit to human longevity.

Some studies have attempted to address the question of maximum human longevity by looking at how key biological systems, like getting oxygen to the muscles or vasculature, degrade. They suggest that there would be an upper limit as key systems of the body just cannot last, which is not to say medicine might not find a way to fix or replace them in the future. Another way of addressing this question is to take a purely statistical approach and look at the distribution of the ages of the oldest people alive and try to make inferences about its upper limit. Such an analysis relies on extreme value theory.

There are two types of extreme value data. The first type consists of just the series of maximum values from the distribution. The Fisher-Tippett-Gnedenko theorem shows that these maxima can only be distributed according to one of three distributions. The second type of data are all of the most extreme observations above a certain threshold, and wonderfully there is another triple-barrelled theorem that shows that these data are distributed as a generalised Pareto distribution – the Pickand-Balkema-de Haan theorem. This article makes use of this latter type of data and theorem to estimate: (i) is there an upper limit to the distribution of human life spans? (ii) What is it, if so? And (iii) does it change over time?

The authors use a dataset of the ages of death in days of all Dutch residents who died over the age of 92 between 1986 and 2015. Using these data to estimate the parameters of the generalised Pareto distribution, they find strong evidence to suggest that, statistically at least, it has an upper limit and that this limit is probably around 117-124. Over the years of the study there did not appear to be any change in this limit. This is not to say that it couldn’t change in the future if some new miraculous treatment appeared, but for now, we humans must put up with a short and finite existence.

Infant health care and long-term outcomes. Review of Economics and Statistics [RePEc] Published 13th May 2019

I haven’t covered an article on infant health and economic conditions and longer term outcomes for a while. It used to be that there would be one in every round-up I wrote. I could barely keep up with the literature, which I tried to summarise in a different blog post. Given that it has been a while, I thought I would include a new one. This time we are looking at the effect of mother and child health centres in Norway in the 1930s on the outcomes of adults later in the 20th Century.

Fortunately the health centres were built in different municipalities at different times. The authors note that the “key identifying assumption” is that they were not built at a time related to the health of infants in those areas (well, this and that the model is linear and additive, time trends are linear, etc. etc. something that economists often forget). They don’t go into too much detail on this, but it seems plausible. Another gripe of mine with most empirical economic papers, and indeed in medical and public health fields, is that plotting the data is a secondary concern or doesn’t happen at all. It should be the most important thing. Indeed, in this article much of the discussion can be captured by the figure buried two thirds through. The figure shows that the centres likely led to a big reduction in diarrhoeal disease, probably due to increased rates of breast feeding, but on other outcomes effects are more ambiguous and probably quite small if they exist. Some evidence is provided to suggest that these differences were associated with very modest increases in educational attainment and adult wages. However, a cost-benefit calculation suggests that on the basis of these wage increases the intervention had a annualised rate of return of about 5%.

I should say that this study is well-conducted and fairly solid so any gripes with it are fairly minor. It certainly fits neatly into the wide literature on the topic, and I don’t think anyone would doubt that investing in childhood interventions is likely to have a number of short and long term benefits.

Relationship between poor olfaction and mortality among community-dwelling older adults: a cohort study. Annals of Internal Medicine [PubMed] Published 21st May 2019

I included this last study, not because of any ground-breaking economics or statistics, but because it is interesting. This is one of a number of studies to have looked at the relationship between smell ability and risk of death. These studies have generally found a strong direct relationship between poor olfaction and risk of death in the following years (summarised briefly in this editorial). This study examines a cohort of a couple of thousand older people whose smell was rigourously tested at baseline, among other things. If they died then their death was categorised by a medical examiner into one of four categories: dementia or Parkinson disease, cardiovascular disease, cancer, and respiratory illness.

There was a very strong relationship between poor ability to smell and all-cause death. They found that cumulative risk for death was 46% and 30% higher in persons with a loss of smelling ability at 10 and 13 years respectively. Delving into death by cause, they found that this relationship was most important among those who died of dementia or Parkinson disease, which makes sense as smell is one of the oldest limbic structures and linked to many parts of the brain. Some relationship was seen with cardiovascular disease but not cancer or respiratory illness. They then use a ‘mediation analysis’, i.e. conditioning on post-treatment variables to ‘block’ causal pathways, to identify how much variation is explained and conclude that dementia, Parkinson disease, and weight loss account for about 30% of the observed relationship. However, I am usually suspicious of mediation analyses, and standard arguments would suggest that model parameters would be biased.

Interestingly, olfaction is not normally used as a diagnostic test among the elderly despite sense of smell being one of the strongest predictors of mortality. People do not generally notice their sense of smell waning as it is gradual, so would not likely remark on it to a doctor. Perhaps it is time to start testing it routinely?


Paul Mitchell’s journal round-up for 6th 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.

A longitudinal study to assess the frequency and cost of antivascular endothelial therapy, and inequalities in access, in England between 2005 and 2015. BMJ Open [PubMed] Published 22nd October 2017

I am breaking one of my unwritten rules in a journal paper round-up by talking about colleagues’ work, but I feel it is too important not to provide a summary for a number of reasons. The study highlights the problems faced by regional healthcare purchasers in England when implementing national guideline recommendations on the cost-effectiveness of new treatments. The paper focuses on anti-vascular endothelial growth factor (anti-VEGF) medicines in particular, with two drugs, ranibizumab and aflibercept, offered to patients with a range of eye conditions, costing £550-800 per injection. Another drug, bevacizumab, that is closely related to ranibizumab and performs similarly in trials, could be provided at a fraction of the cost (£50-100 per injection), but it is currently unlicensed for eye conditions in the UK. This study investigates how the regional areas in England have coped with trying to provide the recommended drugs using administrative data from Hospital Episode Statistics in England between 2005-2015 by tracking their use since they have been recommended for a number of different eye conditions over the past decade. In 2014/15 the cost of these two new drugs for treating eye conditions alone was estimated at £447 million nationally. The distribution of where these drugs are provided is not equal, varying widely across regions after controlling for socio-demographics, suggesting an inequality of access associated with the introduction of these high-cost drugs over the past decade at a time of relatively low growth in national health spending. Although there are limitations associated with using data not intended for research purposes, the study shows how the most can be made from data routinely collected for non-research purposes. On a public policy level, it raises questions over the provision of such high-cost drugs, for which the authors state the NHS are currently paying more for than US insurers. Although it is important to be careful when comparing to unlicensed drugs, the authors point to clear evidence in the paper as to why their comparison is a reasonable one in this scenario, with a large opportunity cost associated with not including this option in national guidelines. If national recommendations continue to insist that such drugs be provided, clearer guidance is also required on how to disinvest from existing services at a regional level to reduce further examples of inequality in access in the future.

In search of a common currency: a comparison of seven EQ-5D-5L value sets. Health Economics [PubMed] Published 24th October 2017

For those of us out there who like a good valuation study, you will need to set yourself aside a good piece of time to work your way through this one. The new EQ-5D-5L measure of health status, with a primary purpose of generating quality-adjusted life years (QALYs) for economic evaluations, is now starting to have valuation studies emerging from different countries, whereby the relative importance of each of the measure dimensions and levels are quantified based on general population preferences. This study offers the first comparison of value sets across seven countries: 3 Western European (England, Netherlands, Spain), 1 North American (Canada), 1 South American (Uruguay), and two East Asian (Japan and South Korea). The authors in this paper aim to describe methodological differences between the seven value sets, compare the relative importance of dimensions, level decrements and scale length (i.e. quality/quantity trade-offs for QALYs), as well as developing a common (Western) currency across four of the value sets. In brief summary, there does appear to be similar trends across the three Western European countries: level decrements from levels 3 to 4 have the largest value, followed by levels 1 to 2. There is also a pattern in these three countries’ dimensions, whereby the two “symptom” dimensions (i.e. pain/discomfort, anxiety/depression) have equal importance to the other three “functioning” dimensions (i.e. mobility, self-care and usual activities). There are also clear differences with the other four value sets. Canada, although it also has the highest level decrements between levels 3 and 4 (49%), unusually has equal decrements for the remainder (17% x 3). For the other three countries, greater weight is attached to the three functioning dimensions relative to the two symptom dimensions. Although South Korea also has the greatest level decrements between level 3 and 4, it was greatest between level 4 and level 5 in Uruguay and levels 1 and 2 in Japan. Although the authors give a number of plausible reasons as to why these differences may occur, less justification is given in the choice of the four value sets they offer as a common currency, beyond the need to have a value set for countries that do not have one already. The most in-common value sets were the three Western European countries, so a Western European value set may have been more appropriate if the criterion was to have comparable values across countries. If the aim was really for a more international common currency, there are issues with the exclusion of non-Western countries’ value sets from their common currency version. Surely differences across cultures should be reflected in a common currency if they are apparent in different cultures and settings. A common currency should also have a better spread of regions geographically, with no country from Africa, the Middle East, Central and South Asia represented in this study, as well as no lower- and middle-income countries. Though this final criticism is out of the control of the authors based on current data availability.

Quantifying the relationship between capability and health in older people: can’t map, won’t map. Medical Decision Making [PubMed] Published 23rd October 2017

The EQ-5D is one of many ways quality of life can be measured within economic evaluations. A more recent way based on Amartya Sen’s capability approach has attempted to develop outcome measures that move beyond health-related aspects of quality of life captured by EQ-5D and similar measures used in the generation of QALYs. This study examines the relationship between the EQ-5D and the ICECAP-O capability measure in three different patient populations included in the Medical Crises in Older People programme in England. The authors propose a reasonable hypothesis that health could be considered a conversion factor for a person’s broader capability set, and so it is plausible to test how well the EQ-5D-3L dimension values and overall score can map onto the ICECAP-O overall score. Through numerous regressions performed, the strongest relationship between the two measures in this sample was an R-squared of 0.35. Interestingly, the dimensions on the EQ-5D that had a significant relationship with the ICECAP-O score were a mix of dimensions with a focus on functioning (i.e. self-care, usual activities) and symptoms (anxiety/depression), so overall capability on ICECAP-O appears to be related, at least to a small degree, to both health components of EQ-5D discussed in this round-up’s previous paper. The authors suggest it provides further evidence of the complementary data provided by EQ-5D and ICECAP-O, but the causal relationship, as the authors suggest, between both measures remains under-researched. Longitudinal data analysis would provide a more definitive answer to the question of how much interaction there is between these two measures and their dimensions as health and capability changes over time in response to different treatments and care provision.



Chris Sampson’s journal round-up for 5th September 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.

The effect of complementary private health insurance on the use of health care services. International Journal of Health Economics and Management Published 31st August 2016

Moral hazard is one of the key ideas taught to fledgling health economists, but having taken flight you don’t hear all that much about it. That’s because most of us live in Europe, enjoying our universal publicly funded health care systems. But I quite like papers that remind me that moral hazard is still a going concern and that my MSc was relevant. This paper looks at the impact of complementary private health insurance – that is, alongside a national health service. There aren’t so many studies of moral hazard in this setting. Private health insurance (let’s call it PHI) might decrease use of public health care (let’s call it NHS), but it might also increase pressure on the NHS by creating additional demand. For example, people might need a referral from an NHS GP in order to qualify for PHI coverage. This study uses cross-sectional questionnaire data from Denmark, collected from 5447 individuals for the purpose of this study. The questionnaire collected all sorts of data relating to health care use and sociodemographics. People who gave ‘Don’t know’ or ‘Other’ responses were dropped, meaning that only 4362 were analysed. 49% of the sample had PHI – the ‘treatment’ of interest. The authors use a bivariate probit model with propensity score matching to predict health care use. Furthermore, an instrumental variable is used to improve identification. Having PHI seems to increase use of services, with strong effects for prescription medicine, dentist visits and chiropractors. This suggests that PHI coverage may contribute to increasing national health care costs. There are some major limitations to this study, which the authors acknowledge. The response rate was 41%, and the sample wasn’t particularly representative. The one thing I can’t get my head around is the authors’ identification strategy. The instrumental variable chosen was whether or not an individual wears glasses, as in this case PHI is particularly favourable. Even controlling for the covariates used in this analysis, I cannot see (no pun intended) how this could be unrelated to health care use.

The value of disease prevention vs treatment. Journal of Health Economics Published 29th August 2016

The public’s view of pharma just keeps getting worse“, apparently. One probably-entirely-made-up-but-sort-of-reasonable-sounding thing I’ve heard Joe Public say in the past is that Pharma would like us all to remain sickly cash cows. New treatments = milk. Prevention is just… soya. That analogy made no sense, but there are also more reasoned arguments that we spend too much on treatment and too little on prevention. There are also numerous studies characterising people’s preferences regarding prevention and treatment under different conditions. This study builds on this background by developing a utility model of disease valuation in order to derive willingness-to-pay values for reductions in incidence (prevention), mortality (treatment) or deterioration in quality of life (palliative care). The basis for the model is 3 possible states – healthy, ill and dead – through which people can progress in only one direction (i.e. there is no cure). The ‘ill’ state relates to a specific disease and has a value somewhere between 1 (healthy) and 0 (dead). The authors use the model to determine – for example – how willingness to pay for improvement in the ‘ill’ state might be affected by the mortality rate. Two key implications of the model are that i) when the risk of dying from a disease is greater than the incidence rate, prevention is more valuable than treatment and ii) when the incidence rate is greater than the decline in quality of life, prevention is more valuable than palliative care. The model is also used to incorporate probability weighting to give a more realistic characterisation of people’s risk preferences. In most cases, the two previous findings will hold. An interesting finding of this part of the analysis is that it seems to partly explain people’s disproportionately strong preferences for treating more severe diseases. The model suggests that prevention is more valuable than treatment for most real-world situations, and so we’ve probably got the balance all wrong.

Does one size fit all? Assessing the preferences of older and younger people for attributes of quality of life. Quality of Life Research [PubMed] Published 23rd August 2016

There’s plenty of talk nowadays about the idea that QALYs don’t reflect the most important objects of value for particular groups of people, especially older people. Non-health improvements in quality of life might be more important. Whether we’re using EQ-5D, SF-6D, HUI3 or your personally preferred multi-attribute utility measure, the idea is that they’re measuring the same thing. But they’re not. They consistently give systematically different results. This study sought to find out if older people value quality of life attributes used in these measures differently to younger people. The authors elicit preferences for different domains using a web-based survey of two groups of 500 people: over 65s and 18-64 year olds. Individuals were presented with 12 descriptors from the EQ-5D, AQoL and ASCOT and asked to complete both a ranking and a best worst exercise. Socioeconomic data were also collected. The two cohorts ranked the domains differently, but perhaps not as differently as we might expect. ‘Independence’ was important to both groups, with 36% of over 65s and 20% of 18-64 year olds ranking it first. Physical mobility, mental health and pain also ranked highly for both groups. Older people ranked control, self-care and vision more highly than younger people, who in turn ranked safety, social relationships, dignity, sleep and hearing more highly. The results from the ranking exercise and the best worst exercise were similar. So, non-health attributes matter to everyone and older people’s preferences differ to younger people’s. But so what? We could probably find differences between a sample of men and a sample of women, or between an urban and a rural population. The question is: which differences matter? Studies like this are useful, but they can’t tell us how we ought to handle heterogeneous preferences.

From representing views to representativeness of views: Illustrating a new (Q2S) approach in the context of health care priority setting in nine European countries. Social Science & Medicine [PubMedPublished 22nd August 2016

Asking the public what they think; it’s a dangerous game (nb Brexit, Boaty McBoatface, Mrs Brown’s Boys). But there are good grounds for doing so when it comes to health care resource allocation. This paper comes from an ongoing research project that I’ve written about on a couple of occasions. A previous paper used Q methodology and identified 5 viewpoints regarding the fundamental basis for the allocation of resources in health care, titled: 1) ‘egalitarianism, entitlement and equality of access’, 2) ‘severity and the magnitude of health gains’, 3) ‘fair innings, young people and maximising health benefits’, 4) ‘the intrinsic value of life and healthy living’ and 5) ‘quality life is more important than simply staying alive’. This study developed a new methodology called Q2S, designed to extract features from the viewpoints elicited through the original Q study and create a survey to find out how these different viewpoints are represented in society. Data were collected from 39,560 respondents from 9 European countries. Participants were presented with a series of descriptions with which to identify agreement on a 7-point Likert scale from “very unlike my point of view” to “very much like my point of view”. 41% of respondents gave their highest score to a single viewpoint, while the rest tied across two or more viewpoints and were subsequently asked to identify which one would best reflect their view. 43% of respondents were allocated to Viewpoint 1. This viewpoint asserts that health care is a basic right, that treatment effectiveness is essentially irrelevant because all life has the same value, and that scarcity is not a concern. It was predominant in all 9 countries. Gulp! Next up with 17% was Viewpoint 2, which is a bit closer to health maximisation but with a preference for allocation to life-saving treatment and more severe health states. Viewpoint 3 was not popular, with only 4% of people identifying it as most like their point of view. The authors identify various associations between sociodemographic variables and likelihood of particular viewpoints. There’s a lot of food for thought in this paper. Where do you sit? My position changes depending on how revolutionary I’m feeling.

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