Sam Watson’s journal round-up for 12th 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.

Estimating health opportunity costs in low-income and middle-income countries: a novel approach and evidence from cross-country data. BMJ Global Health. Published November 2017.

The relationship between health care expenditure and population health outcomes is a topic that comes up often on this blog. Understanding how population health changes in response to increases or decreases in the health system budget is a reasonable way to set a cost-effectiveness threshold. Purchasing things above this threshold will, on average, displace activity with greater benefits. But identifying this effect is hard. Commonly papers use some kind of instrumental variable method to try to get at the causal effect with aggregate, say country-level, data. These instruments, though, can be controversial. Years ago I tried to articulate why I thought using socio-economic variables as instruments was inappropriate. I also wrote a short paper a few years ago, which remains unpublished, that used international commodity price indexes as an instrument for health spending in Sub-Saharan Africa, where commodity exports are a big driver of national income. This was rejected from a journal because of the choice of instruments. Commodity prices may well influence other things in the country that can influence population health. And a similar critique could be made of this article here, which uses consumption:investment ratios and military expenditure in neighbouring countries as instruments for national health expenditure in low and middle income countries.

I remain unconvinced by these instruments. The paper doesn’t present validity checks on them, which is forgiveable given medical journal word limitations, but does mean it is hard to assess. In any case, consumption:investment ratios change in line with the general macroeconomy – in an economic downturn this should change (assuming savings = investment) as people switch from consumption to investment. There are a multitude of pathways through which this will affect health. Similarly, neighbouring military expenditure would act by displacing own-country health expenditure towards military expenditure. But for many regions of the world, there has been little conflict between neighbours in recent years. And at the very least there would be a lag on this effect. Indeed, in all the models of health expenditure and population health outcomes I’ve seen, barely a handful take into account dynamic effects.

Now, I don’t mean to let the perfect be the enemy of the good. I would never have suggested this paper should not be published as it is, at the very least, important for the discussion of health care expenditure and cost-effectiveness. But I don’t feel there is strong enough evidence to accept these as causal estimates. I would even be willing to go as far to say that any mechanism that affects health care expenditure is likely to affect population health by some other means, since health expenditure is typically decided in the context of the broader public sector budget. That’s without considering what happens with private expenditure on health.

Strategic Patient Discharge: The Case of Long-Term Care Hospitals. American Economic Review. [RePEcPublished November 2018.

An important contribution of health economics has been to undermine people’s trust that doctors act in their best interest. Perhaps that’s a little facetious, nevertheless there has been ample demonstration that health care providers will often act in their own self-interest. Often this is due to trying to maximise revenue by gaming reimbursement schemes, but also includes things like doctors acting differently near the end of their shift so they can go home on time. So when I describe a particular reimbursement scheme that Medicare in the US uses, I don’t think there’ll be any doubt about the results of this study of it.

In the US, long-term acute care hospitals (LTCHs) specialise in treating patients with chronic care needs who require extended inpatient stays. Medicare reimbursement typically works on a fixed rate for each of many diagnostic related groups (DRGs), but given the longer and more complex care needs in LTCHs, they get a higher tariff. To discourage admitting patients purely to get higher levels of reimbursement, the bulk of the payment only kicks in after a certain length of stay. Like I said – you can guess what happened.

This article shows 26% of patients are discharged in the three days after the length of stay threshold compared to just 7% in the three days prior. This pattern is most strongly observed in discharges to home, and is not present in patients who die. But this may still be just by chance that the threshold and these discharges coincide. Fortunately for the authors the thresholds differ between DRGs and even move around within a DRG over time in a way that appears unrelated to actual patient health. They therefore estimate a set of decision models for patient discharge to try to estimate the effect of different reimbursement policies.

Estimating misreporting in condom use and its determinants among sex workers: Evidence from the list randomisation method. Health Economics. Published November 2018.

Working on health and health care research, especially if you conduct surveys, means you often want to ask people about sensitive topics. These could include sex and sexuality, bodily function, mood, or other ailments. For example, I work a fair bit on sanitation, where frequently self-reported diarrhoea in under fives (reported by the mother that is) is the primary outcome. This could be poorly reported particularly if an intervention includes any kind of educational component that suggests it could be the mother’s fault for, say, not washing her hands, if the child gets diarrhoea. This article looks at condom use among female sex workers in Senegal, another potentially sensitive topic, since unprotected sex is seen as risky. To try and get at the true prevalence of condom use, the authors use a ‘list randomisation’ method. This randomises survey participants to two sets of questions: a set of non-sensitive statements, or the same set of statements with the sensitive question thrown in. All respondents have to do is report the number of the statements they agree with. This means it is generally not possible to distinguish the response to the sensitive question, but the difference in average number of statements reported between the two groups gives an unbiased estimator for the population proportion. Neat, huh? Ultimately the authors report an estimate of 80% of sex workers using condoms, which compares to the 97% who said they used a condom when asked directly.

 

Credits

Thesis Thursday: Matthew Quaife

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 Matthew Quaife who has a PhD from the London School of Hygiene and Tropical Medicine. If you would like to suggest a candidate for an upcoming Thesis Thursday, get in touch.

Title
Using stated preferences to estimate the impact and cost-effectiveness of new HIV prevention products in South Africa
Supervisors
Fern Terris-Prestholt, Peter Vickerman
Repository link
http://researchonline.lshtm.ac.uk/4646708

Stated preferences for what?

Our main study looked at preferences for new HIV prevention products in South Africa – estimating the uptake and cost-effectiveness of multi-purpose prevention products, which protect against HIV, pregnancy and STIs. You’ll notice that condoms do this, so why even bother? Condom use needs both partners to agree (for the duration of a given activity) and, whilst female partners tend to prefer condom-protected sex, there is lots of evidence that male partners – who also have greater bargaining power in many contexts – do not.

Oral pre-exposure prophylaxis (PrEP), microbicide gels, and vaginal rings are new products which prevent HIV infection. More importantly, they are female-initiated and can generally be used without a male partner’s knowledge. But trials and demonstration projects among women at high risk of HIV in sub-Saharan Africa have shown low levels of uptake and adherence. We used a DCE to inform the development of attractive and usable profiles for these products, and also estimate how much additional demand – and therefore protection – would be gained from adding contraceptive or STI-protective attributes.

We also elicited the stated preferences of female sex workers for client risk, condom use, and payments for sex. Sex workers can earn more for risky unprotected sex, and we used a repeated DCE to predict risk compensation (i.e. how much condom use would change) if they were to use HIV prevention products.

What did you find most influenced people’s preferences in your research?

Unsurprisingly for products, HIV protection was most important to people, followed by STI and then pregnancy protection. But digging below these averages with a latent class analysis, we found some interesting variation within female respondents: over a third were not concerned with HIV protection at all, instead strongly caring about pregnancy and STI protection. Worryingly, these were more likely to be respondents from high-incidence adolescent and sex worker groups. The remainder of the sample overwhelmingly chose based on HIV protection.

In the second sex worker DCE, we found that using a new HIV prevention product made condoms become less important and price more important. We predict that the price premium for unprotected sex would reduce by two thirds, and the amount of condomless sex would double. This is an interesting labour market/economic finding, but – if true – also has real public health implications. Since economic changes mean sex workers move from multi-purpose condoms to single-purpose products which need high levels of adherence, we thought this would be interesting to model.

How did you use information about people’s preferences to inform estimates of cost-effectiveness?

In two ways. First, we used simple uptake predictions from DCEs to parameterise an HIV transmission model, allowing for condom substitution uptake to vary by condom users and non-users (it was double in the latter). We were also able to model the potential uptake of multipurpose products which don’t exist yet – e.g. a pill protecting from HIV and pregnancy. We predict that this combination, in particular, would double uptake among high-risk young women.

Second, we predicted risk compensation among sex workers who chose new products instead of condoms. We were also able to calculate the price elasticity of supply of unprotected sex, which we built into a dynamic transmission model as a determinant of behaviour.

Can discrete choice experiments accurately predict the kinds of behaviours that you were looking at?

To be honest, when I started the PhD I was really sceptical – and I still am to an extent. But two things make me think DCEs can be useful in predicting behaviours.

First is the data. We published a meta-analysis of how well DCEs predict real-world health choices at an individual level. We only found six studies with individual-level data, but these showed DCEs predict with an 88% sensitivity but just a 34% specificity. If a DCE says you’ll do something, you more than likely will – which is important for modelling heterogeneity in uptake. We desperately need more studies following up DCE participants making real-world choices.

Second is the lack of alternative inputs. Where products are new and potential users are inexperienced, modellers pick an uptake number/range and hope for the best. Where we don’t know efficacy, we may assume that uptake and efficacy are linearly related – but they may not be (e.g. if proportionately more people use a 95% effective product than a 45% effective one). Instead, we might assume uptake and efficacy are independent, but that might sound even less realistic. I think that DCEs can tell us something about these behaviours that are useful for the parameters and structures of models, even if they are not perfect predictors.

Your tread the waters of infectious disease modelling in your research – was the incorporation of economic factors a challenge?

It was pretty tricky, though not as challenging as building the simple dynamic transmission model as a first exposure to R. In general, behaviours are pretty crudely modelled in transmission models, largely due to assumptions like random mixing and other population-level dynamics. We made a simple mechanistic model of sex work based on the supply elasticities estimated in the DCE, and ran a few scenarios, each time estimating the impact of prevention products. We simulated the price of unprotected sex falling and quantity rising as above, but also overlaid a few behavioural rules (e.g. Camerer’s constant income hypothesis) to simulate behavioural responses to a fall in overall income. Finally, we thought about competition between product users and non-users, and how much the latter may be affected by the market behaviours of the former. Look out for the paper at Bristol HESG!

How would you like to see research build on your work to improve HIV prevention?

I did a public engagement event last year based on one statistic: if you are a 16-year old girl living in Durban, you have an 80% lifetime risk of acquiring HIV. I find it unbelievable that, in 2018, when millions have been spent on HIV prevention and we have a range of interventions that can prevent HIV, incidence among some groups is still so dramatically and persistently high.

I think research has a really important role in understanding how people want to protect themselves from HIV, STIs, and pregnancy. In addition to highlighting the populations where interventions will be most cost-effective, we show that variation in preferences drives impact. I hope we can keep banging the drum to make attractive and effective options available to those at high risk.

Paul Mitchell’s journal round-up for 17th July 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.

What goes wrong with the allocation of domestic and international resources for HIV? Health Economics [PubMedPublished 7th July 2017

Investment in foreign aid is coming under considered scrutiny as a number of leading western economies re-evaluate their role in the world and their obligations to countries with developing economies. Therefore, it is important for those who believe in the benefits of such investments to show that they are being done efficiently. This paper looks at how funding for HIV is distributed both domestically and internationally across countries, using multivariate regression analysis with instruments to control for reverse causality between financing and HIV prevalence, and domestic and international financing. The author is also concerned about countries free riding on international aid and estimates how countries ought to be allocating national resources to HIV using quintile regression to estimate what countries have fiscal space for expanding their current spending domestically. The results of the study show that domestic expenditure relative to GDP per capita is almost unit elastic, whereas it is inelastic with regards to HIV prevalence. Government effectiveness (as defined by the World Bank indices) has a statistically significant effect on domestic expenditure, although it is nonlinear, with gains more likely when moving up from a lower level of government effectiveness. International expenditure is inversely related to GDP per capita and HIV prevalence, and positively with government effectiveness, albeit the regression models for international expenditure had poor explanatory power. Countries with higher GDP per capita tended to dedicate more money towards HIV, however, the author reckons there is $3bn of fiscal space in countries such as South Africa and Nigeria to contribute more to HIV, freeing up international aid for other countries such as Cameroon, Ghana, Thailand, Pakistan and Columbia. The author is concerned that countries with higher GDP should be able to allocate more to HIV, but feels there are improvements to be made in how international aid is distributed too. Although there is plenty of food for thought in this paper, I was left wondering how this analysis can help in aiding a better allocation of resources. The normative model of what funding for HIV ought to be is from the viewpoint that this is the sole objective of countries of allocating resources, which is clearly contestable (the author even casts doubt as to whether this is true for international funding of HIV). Perhaps the other demands faced by national governments (e.g. funding for other diseases, education etc.) can be better reflected in future research in this area.

Can pay-for-performance to primary care providers stimulate appropriate use of antibiotics? Health Economics [PubMed] [RePEcPublished 7th July 2017

Antibiotic resistance is one of the largest challenges facing global health this century. This study from Sweden looks to see whether pay for performance (P4P) can have a role in the prescription practices of GPs when it comes to treating children with respiratory tract infection. P4P was introduced on a staggered basis across a number of regions in Sweden to incentivise primary care to use narrow spectrum penicillin as a first line treatment, as it is said to have a smaller impact on resistance. Taking advantage of data from the Swedish Prescribed Drug Register between 2006-2013, the authors conducted a difference in difference regression analysis to show the effect P4P had on the share of the incentivised antibiotic. They find a positive main effect of P4P on drug prescribing of 1.1 percentage points, that is also statistically significant. Of interest, the P4P in Sweden under analysis here was not directly linked to salaries of GPs but the health care centre. Although there are a number of limitations with the study that the authors clearly highlight in the discussion, it is a good example of how to make the most of routinely available data. It also highlights that although the share of the less resistant antibiotic went up, the national picture of usage of antibiotics did not reduce in line with a national policy aimed at doing so during the same time period. Even though Sweden is reported to be one of the lower users of antibiotics in Europe, it highlights the need to carefully think through how targets are achieved and where incentives might help in some areas to meet such targets.

Econometric modelling of multiple self-reports of health states: the switch from EQ-5D-3L to EQ-5D-5L in evaluating drug therapies for rheumatoid arthritis. Journal of Health Economics Published 4th July 2017

The EQ-5D is the most frequently used health state descriptive system for the generation of utility values for quality-adjusted life years (QALYs) in economic evaluation. To improve sensitivity and reduce floor and ceiling effects, the EuroQol team developed a five level version (5L) compared to the previous three level (3L) version. This study adds to recent evidence in this area of the unforeseen consequences of making this change to the descriptive system and also the valuation system used for the 5L. Using data from the National Data Bank for Rheumatic Diseases, where both 3L and 5L versions were completed simultaneously alongside other clinical measures, the authors construct a mapping between both versions of EQ-5D, informed by the response levels and the valuation systems that have been developed in the UK for the measures. They also test their mapping estimates on a previous economic evaluation for rheumatoid arthritis treatments. The descriptive results show that although there is a high correlation between both versions, and the 5L version achieves its aim of greater sensitivity, there is a systematic difference in utility scores generated using both versions, with an average 87% of the score of the 3L recorded compared to the 5L. Not only are there differences highlighted between value sets for the 3L and 5L but also the responses to dimensions across measures, where the mobility and pain dimensions do not align as one would expect. The new mapping developed in this paper highlights some of the issues with previous mapping methods used in practice, including the assumption of independence of dimension levels from one another that was used while the new valuation for the 5L was being developed. Although the case study they use to demonstrate the effect of using the different approaches in practice did not result in a different cost-effectiveness result, the study does manage to highlight that the assumption of 3L and 5L versions being substitutes for one another, both in terms of descriptive systems and value sets, does not hold. Although the authors are keen to highlight the benefits of their new mapping that produces a smooth distribution from actual to predicted 5L, decision makers will need to be clear about what descriptive system they now want for the generation of QALYs, given the discrepancies between 3L and 5L versions of EQ-5D, so that consistent results are obtained from economic evaluations.

Credits