COVID and ICER: a look at the cost-effectiveness of 'flattening the curve'

You may also be wondering what exactly is happening around us with this Covid-19 pandemic. My generation has probably never experienced such turmoil and disruption. (I am forty-one.) We have few reliable points of reference. Leadership seems to have broken down in many corners of the world, and one cannot help noticing worrying signs of incompetence both at national levels and in international organizations. If a pandemic was only a question of time, as we are being reminded (in all likelihood correctly), then the global leadership failure is even more astonishing. Crisis management seems to be intertwined with petty (and less petty) politics—think of Taiwan and the WHO, or the scramble for face masks. Governments (populists, ‘progressives’ and the large middle field alike) seem to be panicking as are ordinary people. Voices which point at the erosion of rational policymaking and emphasize that it is hard to gauge the exact dimension and seriousness of a pandemic without data and especially testing, are outshouted by clickbait headlines across all channels of media (’the world will never be the same again’ and the like). Global thinking has become unidimensional, we see an escalation of commitment towards ‘flattening the curve’, and I have the impression that leaders try to maintain an air of credibility through communication exercises.

What can a health economist do to kill time in such a crisis? Calculate an ICER, for example ??. In this case, an ICER (incremental cost-effectiveness ratio) for ‘flattening the curve’ – quite a dramatic health care intervention. This is exactly what my long-time friend and co-founder of Ideas & Solutions, Márk Molnár did on the back of an envelope for Facebook-sharing, modelling Hungary, where he and I come from. Based a large set of best available assumptions (keep in mind: we know practically nothing reliable about Covid-19, and feel free to experiment with your own data), his calculations show that, with the lockdown of our economy and society,

a mid-sized, upper-middle income country with a health care system comparable to Hungary may be paying somewhere around 260K euro to provide one additional quality-adjusted life year to a person who would otherwise die from Covid-19.

This is not the usual threshold of 30-40-50K euro, but 5 or even 12 times more. In the case of Hungary, this is not 3x GDP per capita, but 22-23 times that.

What does this mean? Either governments are doing something totally not cost-effective (translate: we are doing something wrong, possibly because of governments being caught unprepared and panicking, scared of ’visual’ deaths for the lack of ventilators, and having no better idea than following the unidimensional approach of flattening the curve, without reliable data, mixing up incidence of disease and incidence of symptoms, not knowing how many people had got through Covid-19 before the world shut down.) Or they have revealed a much higher willingness-to-pay than ever before (translate: governments in most countries have never really been keen on saving human lives at 260K euro per year, now they seem eager to do so). In the latter case, it is fair to expect them to continue and accept ICERs well above, say, 200K euro for live-saving diagnostics, medicines, vaccines etc. in the future. On my end, to be intentionally populistic and provocative, I am ready to pay 260K per QALY (and to be locked down indefinitely) to save predominantly frail people with probably limited individual life expectancies (average Covid-population, certainly), but then I ask that my primary school friend (age 41, he, too) treated with lymphoma (a father of three and with an effectively treatable disease, just to be even more populistic) be given the same resources and chance to survive. I also ask that he not face any delays in treatment and follow-up imaging because all resourced are bogged down with Covid-19, and I also expect decision-makers to pay, again, 200K per quality-adjusted life year for diseases like diabetes, cardiovascular conditions and cancer which kill many more people each year, but certainly in a less dramatically visible way.

Márk’s figures are admittedly back-of-the-envelope calculations, there are a lot of assumptions and the methodologist’s discerning eye will certainly spot assumptions and formulae which it may not like. His results are very sensitive to the blow to the GDP (which we don’t know), the mortality of people requiring and receiving ventilation (which we are starting to see), the health status of Covid-19 patients at the onset of disease (which he assumes to be 0.5, but may be higher, who knows) and the average age of death from Covid-19. You may certainly arrive at significantly lower ICERs but I fiddled around a bit with sensitivity analyses and I find it quite hard to decrease the ICER below 130-150K euro while keeping a realistic set of assumptions. This is still way above the usual reference value.

But this is really NOT the point here. The point is, repeat, that we are either doing something wrong now, or we had been doing something wrong up until Covid-19 knocked on our doors.

While I tend to fear that ‘flattening the curve’, like any unidimensional thinking, has a high probability of being inefficient and/or dysfunctional, I can’t help thinking that striving for so much efficiency in health care may not have been the best policy idea of the past few decades. Neither would I rule out that, albeit way behind political leaders, health economists may also have some collective and indirect responsibility in where we are now. While we cannot re-write past policies I would like to think that we, health economists, will be more cognizant of our responsibility and possible impact, and will help usher decision-makers away from the ’efficiency-only’ logic in the period which will follow Covid-19.

One additional point—the calculations do not take into consideration indirect societal costs of ’flattening the curve’. People are freaking out, there are micro-scale (for the time being) hunger revolts in southern Italy, ordinary people are beaten up in Asia and Africa for not obeying rules likely impossible to obey, people are queuing up for ’essential’ guns in the US, and a probably not-only-public-health-surveillance system is being put into practice in Moscow. There are also broader considerations about legitimacy and governance. In times when rigorous and consistent guidance would be essential, we first hear that masks are not needed, then we hear that we should put on masks (or in some countries we are already obliged to wear them, just in case). In January and early February, public health bodies across Europe still assessed the risk of local transmission as very low – in retrospect, it is hard to understand what they were thinking and expecting. In January and early February, we were told that border closures and travel bans were unnecessary, ineffective, unfair and discriminatory, now we all are pretty much locked up, not within our country borders but often in our homes. ’If people want to travel they will travel anyway’ – was said, and the same media was nodding to such mind-blowingly nonsensical ’expert opinion’ which is now admonishing us to ’social distance’.

’Social distancing’ and ’flattening the curve’ – these two pretty words will be the lasting legacy of the spring of 2020, and they will keep reminding us of the unpreparedness of governments around the globe. As for Covid-19, we have likely manoeuvred ourselves into a one-way street, but if we have any economic resources left when this is over, it will be worth re-thinking how we use them in health care. And also whether it would make more sense in general to focus on a diagnosis first (for example, through testing), and then to design an intervention which protects the vulnerable and lets others go on with their lives (and keep running the economies on which health care systems are dependent).