This post makes two points:-

  1. The current approach to the pandemic in the UK (and in most other countries) involves costs that are disproportionate to any benefits from deaths averted.  It fails the UK Government’s own criteria for cost effective health interventions by a wide margin. The extreme measures being taken have been introduced without any of the evidence that has been required before medical interventions can be used.
  2. Meanwhile, HMG has banned the use of what may be an effective treatment that is low cost and low risk, using drugs that have long been in use. The reason given is that we are awaiting the results of clinical trials – meanwhile people may be dying needlessly when the risk of approving the treatment now for general use seems to be minimal.

Costs of Social Distancing Measures versus Benefits

It is unclear what level of damage we are inflicting on our economy, but a conservative estimate would be that it probably exceeds 10% of our GDP- the Government has already committed more than that amount in grants and loan Guarantees, and there are additionally long term economic costs of lost businesses and lost growth.

We don’t actually know how many deaths might be averted as a result, because even countries like Germany have only tested a small share of the population, and hence the infection rate is unknown and so is the overall fatality rate. The death rate of confirmed cases in Germany is about 1.3%, the death rate of all those infected including those without symptoms who have not been tested is presumably much lower. If we assume in the UK that, left unchecked, the epidemic might infect 80% of the population with a mortality rate of 1%, then we would have about 480,000 excess deaths. Allowing for the deaths that will occur even with the measures, and for some excess deaths that will occur in a ‘do nothing’ scenario due to the capacity of the NHS being exceeded, we might guess that, at most, a social distancing strategy might reduce COVID 19 deaths by up to 500,000. The great majority of the deaths that are averted would be patients who are elderly and/or with other significant underlying health problems. Moreover, as yet we do not know how to exit from the virus without a resurgence. The reduction in mortality may just turn out to be a deferment that enables the NHS to flatten the peak in deaths, but does not prevent a second peak.

How much is it worth spending to achieve this reduction in mortality? Most people are uncomfortable with the idea of putting a price on a death averted, but it is essential to do so if we are to decide how big the health budget should be and how it should be prioritised.

The National Institute for Clinical Excellence (NICE) uses ‘Quality Adjusted Life Years’ (QALYs) when assessing the cost effectiveness of treatment. This is a measure of how many extra years of life a patient might expect to have with a health intervention, compared with what would have happened without the intervention. It adjusts for the quality of life, recognising that an extra year spent in pain or with serious disability is worth less to the patient than a year in good health.

The figure used by NICE for evaluating whether a treatment is cost effective or not is that it should cost not more than £20,000 per QALY saved, though it will consider the case for treatments costing up to £30,000 taking into account other factors. These figures are from NICE QUALY guidelines that were in use in 2017. I am not aware if they have been updated, but the numbers are unlikely to have changed significantly.(Source: Joy Ogden, QALYs and their role in NICE decision-making process, prescriber.co.uk, April 2017).

If we make what now seems a relatively conservative estimate that the social distancing measures will cost us 10% of national GDP, then the total cost would be £252 billion pounds. At the maximum acceptable cost per QALY saved of £30,000, then simple division will tell us that these measures would need to save 8.4 million disability adjusted life years if the cost per QALY saved is to be kept within the NICE ceiling. This means that, if a total of 500,000 people do not die of COVID 19, they would have to enjoy on average another 16 QALYs in good health for this level of costs to be worth incurring. It seems far more likely that an average patient who does not die from COVID 19   will enjoy only a relatively short extension to their life expectancy, certainly a lot less than 16 years. Moreover, any benefit to those who don’t die from COVID 19 needs to be offset by those who die from other causes related to increased poverty and hardship caused by the economic shutdown.

To summarise, even taking a high estimate of the deaths that might be prevented or delayed, and a low estimate of the economic damage we are incurring, the costs of the extreme measures being taken are not justifiable in terms of the Government’s own test of cost-effectiveness.

Treatment with Hydroxychloroquine

I am not medically qualified, but I am struck that public health scientists have been happy to encourage ruinously costly measures that will do severe and permanent damage to our economy, way of life, and health based on no clear evidence of their effectiveness, whereas the approach to medical intervention has been extremely cautious.

There is some evidence that hydroxychloroquine given early enough can prevent the progression of the illness to the point where it causes severe breathing difficulties. It is a cheap drug, has been in use for decades to treat malaria and a range of other conditions, and has few and manageable side effects when used for a short period of time. The evidence so far is limited to promising results in the laboratory, and two small clinical studies that have been criticised. However, given the lack of alternative treatments, it is not clear to me that anything will be lost by making it available. Malaysia has been using it from the start; Italy has started to use it in Rome and the wider region of Lazio, and is already reporting some promising results. At the ICU in Marseilles, it has been reported that over 1500 patients have been treated, with only one death, a staggeringly low fatality rate. (Sources, for a variety of materials on this see COVEXIT- Towards Hydroxychloroquine and other treatments, public group on Facebook.)  

The UK Government has prohibited health professionals from using hydroxychloroquine to treat the disease except in clinical trials (gov.uk/government/news/chloroquine-and-hydroxychloroquine-not-licensed-for-coronavirus-covid-19-treatment, published 25 March 2020,accessedn30 March). In normal times, that would be a reasonable response, but we are in an emergency situation. When the epidemic is killing patients at an accelerating rate, hospitals are being overwhelmed, and we have no alternative treatment to offer that can prevent the progression of the disease to the second stage where breathing is impeded, it seems reasonable to try a treatment that is cheap, has few significant side effects, and has some promising if incomplete evidence of being effective.

There may be some good reasons for caution that I don’t understand, but it does seem that different levels of proof are being required for drug treatments compared to drastic economic measures.. We should obviously be undertaking urgent clinical trials, but in the current emergency it is not evident to me why we should not meanwhile authorise the use of the drug, as an increasing number of health administrations have already begun to do.  The worst that will happen from using it now on a large scale is that it might not work, or might not work as well as we hope, but there seems little reason to believe that it will make matters worse.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s