Thesis Thursday: Frank Sandmann

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

Title
The true cost of epidemic and outbreak diseases in hospitals
Supervisors
Mark Jit, Sarah Deeny, Julie Robotham, John Edmunds
Repository link
http://researchonline.lshtm.ac.uk/4648208/

Do you refer to the ‘true’ cost because some costs are hidden in this context?

That’s a good observation. Economists use the term “true cost” as a synonym for “opportunity cost”, which can be defined as the net value of the forgone second-best use of a resource. The true value of a hospital bed is therefore determined by its second-best use, which may indeed be less easily observed and less obvious, or somewhat hidden.

In the context of infectious disease outbreaks in hospital, the most visible costs are the direct expenditures on treatments of infected cases and any measures of containment. However, they do not capture the full extent of the “alternative” costs and therefore cannot equal opportunity costs. Slightly less visible are the potential knock-on effects for visitors to the hospital who, unbeknown to them, may get infected and contribute to sustained transmission in the community. Least seen are the externalities borne by patients who have not been admitted so far but who are awaiting admission, and for whom there is no space in hospital yet due to the ongoing outbreak.

In my thesis, I provided a general overview of the historical development of the concept of opportunity costs of resources before I looked in detail at bed-days and the application for hospitals.

How should the opportunity cost of hospital stays be determined?

That depends on for whom you want to determine these costs.

For individual patients, it depends on the very subjective decision of how else they would spend their time instead, and how urgent it is to receive hospital care.

From the perspective of hospital administrators, it is straightforward to calculate the opportunity costs based on the revenues and expenditures of the inpatients, their length of stays, and the existing demand of care from the community. This is quite important because whether there are opportunity costs from forgone admissions will depend on whether there are other patients actually waiting to be admitted, which is somewhat reflected in occupancy rates and of course waiting lists.

Any other decision maker who is acting as an agent on behalf of a collective group or the public should look into the forgone health impact of patients who cannot be admitted when the beds are unavailable to them. In my thesis, I proposed a method for quantifying the opportunity costs of bed-days with the net benefit of the second-best patients forgone, which I illustrated with the example of norovirus-associated gastroenteritis.

How important are differences in methods for costing in the context of gastroenteritis and norovirus?

The results can differ quite substantially when using different costing methods. Norovirus is an ideal illness to illustrate this issue given that otherwise healthy people with gastrointestinal symptoms and no further comorbidities or complications shouldn’t be admitted to hospital in order to minimise the risk of an outbreak. Patients with norovirus are therefore often not the patient group that is benefitting the most from a hospital stay.

In one of the studies of my PhD, I was able to show that the annual burden of norovirus in public hospitals in England amounts to a mean £110 million using conventional costing methods, while the opportunity costs were two-to-three times higher of up to £300 million.

This means that there is the potential for a situation where an intervention is disadvantaged when using conventional methods for costing and ignoring the opportunity costs. When evaluating such an intervention against established decision rules of cost-effectiveness, this may lead to an incorrect decision.

What were some of the key challenges that you encountered in estimating the cost of norovirus to hospitals, and how did you overcome them?

There were at least four key challenges:

First was the number of admissions. Many inpatients with norovirus won’t get recorded as such if they haven’t been laboratory-confirmed. That is why I regressed national inpatient episodes of gastroenteritis against laboratory surveillance reports for ten different gastrointestinal pathogens to estimate the norovirus-attributable proportion.

Second was the number of bed-days used by inpatients that were infected with norovirus during their hospital stay. Using their total length of stay, or some form of propensity matching, suffers from time-dependent biases and overestimates the number of bed-days. Instead, I used a multi-state model and patient-level data from a local hospital.

Third was the bed-days that were left unoccupied for infection control. One of the datasets tracked them mandatorily for acute hospitals during winters, while another surveillance system was voluntary, but recorded outbreaks throughout the year. For a more accurate estimate, I compared both datasets with each other to explore their potential overlap.

Fourth was the forgone health of alternative admissions who had otherwise occupied the beds. I had to make assumptions about the disease progression with and without hospital treatment, for which I used health-state utilities that accounted for age, sex, and the primary medical condition.

If you could have wished for one additional set of data that wasn’t available, what would it have been?

I have been very fortunate to work with a number of colleagues at Public Health England and University College London who provided me with much of the epidemiological data that I needed. My research could have benefitted though from a dataset that tracked the time of infection for a larger patient population and for longer observation periods, and a dataset that included more robust estimates for the health gain from hospital care.

If I could make a wish about the existing datasets on norovirus that I have used, I would wish for a higher rate of reporting given that it became clear from our comparison of datasets that there is a highly-correlated trend, but the number of outbreaks reported and the details of reporting leave room for improvement. Another wish of mine for daily reporting of bed-days during winter became reality only recently; during my PhD, I had to impute missing values that were non-randomly missing at weekends and over the Christmas period. This was changed in winter 2016, and I have recently shown that the mean of our lowest-to-highest imputation scenarios is surprisingly close to the daily number of bed-days recorded since then.

Parts of your thesis are made up of journal articles that you published before submission. Was this always your intention and how did you find the experience?

I always wanted to publish parts of my thesis in separate journal articles as I believe this to be a great chance to reach different audiences. That is because my theoretical research on opportunity costs may be of broader interest than just to those who work on norovirus or bed-days given that my findings are generalisable to other diseases as well as other resources. At the same time, others may be more interested in my results for norovirus, and still others in my application of the various statistical, economic, and mathematical modelling techniques.

After all, I honestly suspect that some people may place a higher value on their next-best alternative use of time than reading my thesis from cover to cover.

Writing up my thoughts early on also helped me refine them, and the peer-review process was a great opportunity to get some additional feedback. It did require good time management skills though to keep coming back to previous studies to address the peer-reviewers’ comments while I was already busy working on the next studies.

All in all, I can recommend others to consider it and, looking back, I’d do it again this way.

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.

Sam Watson’s journal round-up for 21st November 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.

Model-based geostatistics for prevalence mapping in low-resource settings. Journal of the American Statistical Association Published 18th October 2016

Geostatistics refers to the set of statistical methods used to analyse discrete spatial data to unobserved underlying causal processes. The methods were first conceived in the 1950s for the South African mining industry to be able to make inferences about gold deposits from small sets of samples. Nowadays, these methods have wide applications, including in epidemiology as they are shown here. Peter Diggle and Emanuele Giorgi build on earlier work to give a broad overview of geostatistics for disease prevalence mapping and discuss specific applications in low resources settings where issues such as sparse data or the need to rely on different data sources of different quality arise. While this may not seem to be in the realm of health economics, the analysis of spatial data is often used for modelling things like hospital access or as a step in burden of disease calculations. In high income country settings, such as the United Kingdom, we often have access to high resolution data across all areas, but this could easily be combined with other data sources linked to specific locations in a way that this paper and others discuss. In a previous post, we discussed modelling and analysis of slums, methods such as these will no doubt prove invaluable as more data becomes available in the future. Better data along with valid and robust statistical methods can only improve policy analysis and policy design. This paper provides a good insight into this area of statistics for any statistically-minded health economist.

Assessment of economic vulnerability to infectious disease crises. The Lancet [PubMed] Published 12th November 2016

The risk register of the UK Cabinet Office provides a summary of the likelihood and potential impact of a wide range of possible civil emergencies. Pandemic influenza is rated as the highest risk both in terms of likelihood and potential impact. Indeed, an influenza pandemic of the scale of the 1918-9 Spanish Influenza would cause economic losses of around $3 trillion. Similarly, the ebola pandemic that struck a number of West African nations was estimated to have caused a cumulative loss worth about 10% of GDP for the affected nations. And yet infectious disease pandemics rarely feature in economic risk assessments. It may be due to the methodological difficulty of doing so, or poor judgement about the impact of low-probability, high-impact events. In any case, this article proposes a framework for assessing the potential consequences of a pandemic. The steps it outlines are pretty much as you would expect: identify pandemic risk, identify health system capacity, estimate economic vulnerability on a sectoral basis, and make an overall assessment. This may seem a fairly routine task when compared to the assessment of other economic risks, but as the authors show, it is often not done. They show that terrorism is widely considered as a major economic risk, and mentions of terrorism in reports change little from before to after a terrorist incidence. But, pandemics are often not discussed until after one has occurred. The Lancet can sometimes go awry when it ventures into economics. However, it can sometimes publish thoughtful and interesting arguments that economists should consider. Infectious disease crises are a neglected dimension of global security.

Distinguishing hypothetical willingness from behavioral intentions to initiate HIV pre-exposure prophylaxis (PrEP): Findings from a large cohort of gay and bisexual men in the U.S. Social Science & Medicine Published 18th November 2016

A short while ago we posted a piece about pre-exposure prophylaxis (PrEP). The media had kicked up a fuss when a court ruled that the provision of PrEP, which can significantly reduce the risk of contracting HIV, fell within the remit of the NHS. We argued that this was unwarranted since PrEP would likely go through the same scrutiny as any other potential treatment, and its cost-effectiveness would need to be established. Economic evaluations of interventions for infectious diseases can be tricky since the analyst should take into account transmission dynamics, which can amplify intervention effects in a population. A further concern, and one which this paper considers, is whether the patients targeted for the intervention would actually be willing to take it. A sub-sample of 880 PrEP-naïve men, from a longitudinal study of gay and bisexual men in the United States, were surveyed on their attitudes to PrEP. The men were asked about their perceptions of the efficacy of PrEP (about 90% reduction in risk) and whether they were unwilling, willing, or intending to take it. The results are broken down into various groups, including for those who had engaged in condomless sex with a partner who was HIV positive or of unknown status – a potentially high risk group to whom PrEP may be offered. Of this higher risk group, 27% responded that they were unwilling to take it, which may seem surprising given the high efficacy and low side effect profile. One explanation could be the lack of knowledge about PrEP: only one third of men in the sample knew its efficacy. Results like these are crucial to appropriately design and evaluate interventions in this area and it is suggestive of the need for greater provision of information to higher risk individuals.

Credits