Vaccine delay: is Australia accidentally doing the right thing?

The federal government has overpromised and underdelivered on the COVID-19 vaccine. It deserves to be criticised for that.

But delaying immunisation means that Australia may — albeit inadvertently — be doing the right thing.

Vaccine nationalism was always to be expected – and it’s a shame that the rich world has  snapped up most of the current supply. The political economy of COVID-19 is merely following a long-standing trend of colonial extraction.

But calls to do “whatever it takes” to get ahead of others in acquiring and distributing the vaccines are unprincipled and short-sighted.

Medical care is a highly scarce resource. In most cases (especially in emergencies) it is allocated based on need. The person suffering a heart attack will be treated ahead of someone with a broken toe — even if the latter arrived first and complains loudly.

COVID-19 is a global emergency. Producing vaccines takes time and supply will be dwarfed by need in the short-run.

It’s a simple zero-sum game: every administered dose means someone misses out elsewhere. Immunising Australians – remember that we have 39 active cases and five deaths in the past three months – would unquestionably mean avoidable deaths in places where the virus is killing thousands each day.

So much for being a good global citizen.

Also, immunisation is not the silver bullet people seem to think it is. Scientists are increasingly convinced that COVID-19 will become endemic. This means that public health measures will be a feature of daily life for some time.

For those who remain unconvinced, waiting our turn is also in Australia’s self-interest in two ways.

First, as we’ve already witnessed with this one, viruses mutate. Some mutations can result in strains that are more contagious, more lethal or even one more resistant to vaccines. Uncontrolled spread anywhere in the world can yield variants that pose a threat everywhere.

This is why COVID-19 is estimated to cost developed nations around US$119 billion per year, while unequal allocation of vaccines is forecast to slow global economic output by US$1.2 trillion a year.

Second, the speed with which the vaccines are developed and approved for use is unprecedented and certainly not normal. Some questions that would be normally be investigated by developers and regulators before rollout remain unanswered. Is a vaccine more effective in some people than others? Is there an optimal interval between doses?

Data to help answer these questions are being generated in real time. Every administered dose increases the knowledge at our disposal — an enormous advantage for countries that have the luxury of being able to delay their rollout by a month or two.

Vaccinating against COVID-19 is not an Olympic sport. We’re dealing with a deadly pandemic that needs to be stopped as soon as possible. This means allocating vaccines based on need and Australia’s need is thankfully low.

Waiting our turn is in our best interest — morally, medically, and economically.

This is an edited version of a piece published in Crikey on 4 February 2021.

About Luke Slawomirski

Health economist, policy wonk. @LukeSlawomirski on Twitter
This entry was posted in Uncategorized. Bookmark the permalink.

36 Responses to Vaccine delay: is Australia accidentally doing the right thing?

  1. conrad says:

    It’s obvious why individual countries want vaccines before others — It’s basically a prisoners dilemma situation for many, and I don’t think politically too many countries would have people happy to sacrifice their chance for others, even if one might have quite reasonable arguments.

    Also, these arguments are always phrased in terms of rich vs. non-rich nations, as it has been here. But in terms of death rates, rich countries have some of the highest (and some moderate income countries too, like Mexico). Basically, if you have a fat, unhealthy, and demographically older population, like the UK, you will get high death rates. If you have a healthier population, you’ll get less. So if you just wanted to prevent deaths from covid, then a rich vs. poor split is not so meaningful. Similarly, the GDP of many very poor countries is almost irrelevant, so if all we cared about economically was an overall “it’ll be better for us in the long run” effect, we could eliminate very poor countries from the pool of those we need to worry about.

  2. paul frijters says:

    “Australia’s need is thankfully low”
    By what metric? What is the ‘running cost’ of Australia remaining closed to the world and having occasional brutal lockdowns that involve disrupting children, regular health care, social life, inward migration, and economic life? How much borrowing do you think happens each week to mitigate the economic disruption of being closed and of a highly uncertain situation?

    On side effects one of the key unknown statistics is just what percentage of serious side-effects is actually picked up by the monitoring systems that need to be alerted by the victims themselves or their families (so a case of ‘if no-one commplains, we didnt have a problem’). I have seen estimates of 1% to 10% registration rates. Care to give an estimate yourself?

    • Christopher Hood says:

      You, in estimating the harm of the virus itself, continue to disregard medium and long term health effects short of death.
      Treating side-effects of vaccines seriously requires treating impacts of COVID-19 short of death seriously too. And it requires revising upward the ‘0.2%’ death rate estimate on which your low version of COVID-19 impact depends towards the current statistical evidence: between 1 and 2%, with a ‘very likely’ range currently between 1.2 and 1.5%.

    • Conrad says:

      Paul, the side effects of vaccines are getting pretty well known — France has had a tracking system for the three different types they are using on nurses at the moment where all of them are tracked, so your question is answerable. The AstraZeneca one is giving the most side-effects (sickness and hypertension at about .75% each), followed by the Pfizer and then Moderna one, which has next to none. So nothing too serious. That is in a different order to the trial data, although I don’t suppose it matters much given the low prevelance. I’d personally have any of them.

      Chris the overall death rates are nothing like what you say unless you are looking at particular groups. The CDC was estimating about .5% ages ago (which is still very high for a virus but still seems reasonable), but they differ a lot in different countries, and the US is bound to be high due to the medical system and so many people being fat.

      • Christopher Hood says:

        Conrad, you are many months out of date. We already have some countries over .2% deaths as a proportion of the whole population, and several close to that level. We already have some major cities and administrative districts around .3% deaths as a proportion of the whole population.
        In none of those cases is the infection rate, including the undiagnosed, higher than 10% of the whole population.
        That means that a .5% COVID-19 death rate is simply too low to be possible.
        As known infection rates rise and estimated unknown rates fall, the death rates compared to known infections haven’t shifted down much. The current statistics don’t leave any room for the low measures, yet there’s been no recalculation accordingly raising the cost of COVID-19.
        And where are the estimates taking account of the substantial rate of long-term serious health impacts on those who aren’t killed? That’s absent too, though reanalysis is obviously needed.
        When you talk about tracking as revealing side effects, you are discussing systems in which there’s no attribution to the vaccine beyond an adverse event happening close(ish) to the administration of the vaccine. These are tracking systems that provide every possible case for examination, and historically most such cases prove to have nothing to do with the vaccine concerned. COVID-19 vaccines won’t be any different.

        • conrad says:

          I don’t think I’m out of date. There are populations with high death rates, but some of these presumably have large amounts of people that have had it and they may also be more vulnerable populations. Some areas in NY, for example, had around 43% of people that had had it in the first wave (as measured by antibodies), and they are a population likely to show a much higher death rate than others. You need to go from what you noted “major cities and administrative districts” to the population mean.

          Also, you don’t need argue with me about long-term side effects. I’ve pointed that out for ages, including stuff now generally ignored (e.g., long-long term effects like earlier dementia onset) which is now largely unknown and varies a lot across viruses that cross the BB barrier for no predictable reasons.

          • Conrad
            While it’s obviously too soon to tell re long term anything, I still haven’t seen any australian studies or data on it.
            By now we surely have had enough infections for it to be possible to get some kind of estimate of what percent after say three to six months are still crook.
            As for longer term things like Parkinson’s , dementia or cancer we would surely need a better idea about how many conditions of age are ,in at least part , down to a viral infection at some time in that person’s past, before we get too excited about one particular virus, no?

            • Conrad says:

              Things going on 3-6 months after acquiring covid are really quite a different story to longer term things (i.e., 2o years down the track). I haven’t seen anything definitive on it, but I may have missed it (have been rather busy at work). There are studies from groups that had gone to hospital, which shows it isn’t very pleasant which is unsurprising, but not so much from other sources. There is odd stuff going on as well in some people, but who knows the real rate.

              You should be excited about one particular virus — things HIV and HSE cause dementia in susceptible people by themselves. Prion diseases can obviously be especially nasty, but only tiny numbers of humans are susceptible to them. Be glad we found each other tasty in the long distant past and clearly evolved quite some immunity to it :).

        • Conrad says:

          Here’s a good example for you. Let’s take everyone’s favourite country, Sweden.

          They have had about about 12.5 thousands deaths for about 10 million people. So about a .125% death rate. If we wanted to get to your figures, we would need to times that by 10.

          You can work backwards here. If they still need 10 times as many deaths to get to your figures, it must be the case that only 10% of them have had covid, despite walking around like sheep to the slaughter.

          • Christopher Hood says:

            And the current seriological and other estimate for Sweden has an (unconfirmed by individual diagnosis) prevalence estimate of a bit over 7% of the population (admittedly, as at Nov 2020). And the careful and detailed seriological work for Spain just before the current wave showed under 5% of the whole population as an (unconfirmed by individual diagnosis) prevalence estimate.
            Greater London currently has an overall estimate of (unconfirmed by individual diagnosis) prevalence at over 5% and under 10%; it’s the highest prevalence region of the British Isles.
            The ‘43%’ prevalence for NY is, to put it kindly, an outlying result based on unusual methodology. Compare the careful work to estimate health worker risk in NY State, compared to the whole State population benchmark: as at the end of Oct 2020 the whole population benchmark was less than 5% and is unlikely even with the second wave to have reached 10%.

            • conrad says:

              Has it occurred to you that the virus has been going more than 1 year now and the seriological tests won’t pick up cases after the IgG and IgM antibodies have worn off?

              For example: https://www.nature.com/articles/s41591-020-0965-6

              If you follow asymptomatic cases, some 40% won’t get picked up after a mere 8 weeks and 12.9% of symptomatic ones won’t either. I imagine the rate of those that got covid a year ago that get picked up will be even less.

              So there are lots of people that had covid, didn’t have any symptoms they cared about and are running around giving negative tests.

      • paul frijters says:

        nurses are relatively young and healthy, so not much at risk to begin with. Its the vulnerable elderly that are the group to watch and they have to report side-effects themselves, which leads to difficult data because they often have various illnesses.
        Some scholars are trying to deduce the side-effects in this vulnerable group from age-specific excess death data following differences in vaccine roll-outs. Not ideal but probably the best one can do.

        • Tony says:

          In Australia, the average age of nurses was ~44 years old circa 2017. The average age in Australia is 37 IIRC. So, relatively older, I speculate this may be down to fewer career choices for women 40-50 years ago, pushing the average age higher.

          FWIW the nurses I have talked to have varying opinions about covid. Some don’t think its a big deal, others take it very seriously, others threatened to quit when the prospect of going into hard hit nursing homes was raised.

  3. Nicholas Gruen says:

    Thanks Luke,

    I agree on the moral point – though I presume our not hogging vaccines means they go to other developed countries who are snapping up supply. Still, they have more people at immediate risk than we do. So our waiting saves lives in the short term.

    Medically I’m suspicious. I’d like to know more than I do, but I would be very surprised if the set-up we have at present isn’t monumentally inefficient. It’s a game of cat and mouse between for-profit researchers (who mostly do commodity research on molecules on which the heavy lifting has been done in government and philanthropically funded research) and regulators whose incentives are hugely risk-averse.

    They should avoid more thalidomides but only in the context of cost-beneficial choices based on the costs of another thalidomide and the benefits of earlier access to drugs. Of course, just rushing out a drug (or vaccine) against the same regulatory regime can’t be expected to be a huge improvement – a lot of thought should go into improving incentives generally, but rushing this one out seems to be saving the world vast amounts. And the circumstances have finally forced regulators to take urgent account of the benefits as well as the potential costs of drugs. Too bad if you’ve got cystic fibrosis or some other life threatening drug and the regulator applies the ‘front page of the paper’ test to their decision as to whether or not you should be able to access it.

    I’m not sure I quite understand the idea that waiting is in Australia’s economic interest.

    • KT2 says:

      2 famous economists.

      8 links in article.

      And PF comes back with:

      ” By what metric? ” Paul, how many metrics are you able to list to grease the wheels of communication? Oh. You just want a talking point with zero basis.

      “What is the ‘running cost’” Again Paul – you are eminently able to provide an ‘opinion’ plus referenced ‘fact’. But no.

      ” brutal lockdowns ” by what metric are lockdowns brutal. A metric please Paul, not a rhetoric metrick. And a totally hypocritcal phrase consiidering your rhetoric of panic and terrorism. But hey, we’ve come to expect that of you Paul

      “How much borrowing do you think happens each week to mitigate the economic disruption of being closed and of a highly uncertain situation? ” no need of comment – see above.
      And a final flourish, from a professor, who has quoted, referenced, cited, alluded to, and we get
      “I have seen”… the dark. Keep digging all, maybe one day we will get to the bottom or pop out the other side.

    • Tony says:

      I would dispute they are ‘rushed’.

      Keep in mind the vast majority of stuff in the vaccines have been around for years, its just been tweaked for covid. For example, the adenovirus in Oxford/AZ and the nanoparticles in Pfizer/Moderna are just the delivery boys. Oxford had been making a MERS vaccine using the same adenovirus for several years. Apparently, Moderna makes personalized anti-cancer vaccines using the same technology. Pfizer gets their nanoparticles from the same vendor.

      They could get through phase 3 trials so quickly because so many people were getting infected, so rapidly.

      For the thalidomide comparison, different ballparks. Vaccine technology is super safe relative to other medications. Even the worst vaccines have caused awful issues In like 1 in 16,000 people, but in reality bad stuff happens in 1 person per several million doses. Certainly, in the past, some vaccines have skewed inflammatory responses in children – but this has not really been a problem for decades (and is a non-issue in 1st world countries with good healthcare systems).

  4. “Some questions that would be normally be investigated by developers and regulators before rollout remain unanswered.”
    “ Every administered dose increases the knowledge at our disposal — an enormous advantage for countries that have the luxury of being able to delay their rollout by a month or two.”
    So morally we should let the poor go into the water first ?

    BTW could be wrong but I think I heard that Australia is contributing significant amounts to funds for the pacific islands and because we have more vaccines on order than we need also looking at donations of vaccines as well.

    • Conrad says:

      I imagine we don’t need to worry about the poor going into the water first — we can be safe in assuming that the problem rate is so low that the benefit far outweighs the real risk and rich countries will use it immediately, as they are (and reporting the risks for us here in Aus — as France is doing now).

      It’s good that Aus bought vaccines for our poorer neighbours.

  5. Chris Lloyd says:

    What an incredibly annoying blog. Crikey long ago become the left wing organ of the Guardian.

    “The political economy of COVID-19 is merely following a long-standing trend of colonial extraction.” What have we extracted from the non-existent colonies?

    “It’s a simple zero-sum game.” Not so. The total (also known as a sum) is increasing as we sit in lockdown. CSL is manufacturing doses of AZ as fast as it can. Is it moral to use doses you have manufactured yourself?

    “So much for being a good global citizen.” Most folks are prepared to be ethical up to the time where it costs them their lives. So, it is with nations as well.

    Pity the author could not just say that we are in a good position to hang back and not commit ouourselves to the first vaccine, with out bringing in the BLM inspired framing. I guess I am thankful that we were not subjected to claims of white supremacy or privilege.

  6. Christopher Hood says:

    You, in estimating the harm of the virus itself, continue to disregard medium and long term health effects short of death.
    Treating side-effects of vaccines seriously requires treating impacts of COVID-19 short of death seriously too. And it requires revising upward the ‘0.2%’ death rate estimate on which your low version of COVID-19 impact depends towards the current statistical evidence: between 1 and 2%, with a ‘very likely’ range currently between 1.2 and 1.5%.

    • paul frijters says:

      I presume you mean me with “You” and not the author of the post?

      If you take the effort to look at the various calculations I have made on the subject (posted on clubtroppo many times) you will see I have addressed the issue of long-covid many times, putting its magnitude into context. I have also several times done future scanning as to what the situation would look like if covid is going to remain endemic and recurrent in the human population, as is now increasingly likely. I had an interesting exchange with Conrad many months ago on the topic that I encourage you to find. The bottom line is that if covid is the new flu and hence that there is no long-run escape then that weakens the case for restrictions even further (less benefit, more costs). The strategy of closing oneself off from the world at huge costs then makes even less sense, a discussion that is now inevitable in Australia.

      On the IFR of 0.2%, that remains my estimate for the world as a whole, but for certain there are variations in regions. What I have in mind with the IFR though is roughly the definition that Ioanidis and the general public have in mind, namely as the % of the whole population that would die prematurely if exposed to the virus. So it differs from the casuality rate of those that show up positive in tests but pertains to all exposed (a far higher number of people). The 1% IFR estimates belong to much more restrictive groups (such as particular groups of people sick enough to go to hospitals) and allows the readers to make false deductions by not telling them you mean something very different, which is the basic scare-tactic used at the start of the pandemic (when, incidentally, the CDC also put out a figure of 0.2-0.3% IFR). I encourage you to be honest about your definitions and make an effort to talk about the same numbers the population has in mind when being told how dangerous a disease is.

  7. Chris Lloyd says:

    interested in the thoughts of the Tropposphere.

    Is there evidence that the 5-day lockdown has worked? Cheer leaders will say that there are no cases outside the track and trace net. But the original reason for the lockdown was that it was suspected that the net had already been broken! It has not. Thus it seems that the hyper-infectious strain that travels faster than light was not so hyper-infectious as feared.

    It seems to me that if they HAD detected cases outside the net then this, and only this, would have shown that the lockdown was necessary. And if there were no more cases outside the net over the next couple of weeks this, and only this, would show that the lockdown has worked. None of this is to say that the decision was flawed. Decision making under uncertainty is tough.

  8. Nicholas Gruen says:

    Thanks Chris,

    Before I read your comment my view was that even if the decision wasn’t vindicated in hindsight it was worth it in foresight precisely because of the additionally infectious strain. To paraphrase Oscar, 5 days in lockdown is unfortunate, but without it, 5 fortnights in lockdown would have looked like carelessness.

    The additional point you make about trying to judge if the test and trace net had been breached is a good one but still doesn’t change my mind (though an initial three-day lockdown sd have been sufficient to get more intel on that and then make another decision.)

    • Chris Lloyd says:

      Interested in your, and others, reaction to Leigh Sales grilling the premier: https://www.youtube.com/watch?v=DAo3IQeWVnw
      I don’t think anything here contradicts your comment or mine. But it brings the issues into sharper relief.

      • Conrad says:

        I’m with Nick on that. A few days of caution is better than the alternative (assuming they really have trouble tracing it). It might also show they don’t trust their tracing methods, especially given someone serving food at the airport was infected, which seems a whole lot worse than more or less any other place.

        To me the problem is with the CHO and their yes-men. This is the guy that apparently said on the Q&A program (before this event) that airborne transmission was not proved and was unlikely, and clearly didn’t implement strategies to curtail to.

        To me this shows blinding ignorance and something that could have been found with a 5 second google search (surely when you get paid hundreds of thousands of dollars a year you can do this) — the two biggest super-spreader events from SARs-I in HK were both airborne events. One was with a nebuliser (!!) and the other due to air conditioning and wet conditions caused by plumbing. Since it was a coronavirus too, it isn’t too much of a stretch of the imagination think similar effects might be found with this one.

        The other annoying thing, which anyone can learn from Yes-minister, is the inability for them to take any responsibility. At present they get someone to say ‘it wasn’t in our audit’, which is probably true. This is probably true but is Yes-Minister speak for either the audit didn’t have that due to ignorance or the proper audit wasn’t done. Even if the person with the nebuliser was outright lying (which seems rather unlikely), this is still their fault — it’s not like luggage doesn’t go through screening before you get out of an airport. So this shows their system is exactly what you don’t want for safety critical things like this — people more interested in passing the buck and playing political game of thrones with each other than taking responsibility.

      • Nicholas Gruen says:

        Yes it does. And her bull-at-a-gate style is oblivious to risk. Even if things worked out better for NSW than for Victoria, that may have been luck. Victoria’s decision may still have been a better one at the time – though with hindsight it may be revealed to have worked out worse. Such is the nature of risk.

        • Chris Lloyd says:

          I think we all basically agree. It would be nice if Dan could say:

          I followed the advice of the health experts who thought that it has already escaped the net. It turns out that it hadn’t and the lockdown was not necessary but I still think it was a reasonable decision at the time. Perhaps we could have stopped the lockdown after a couple of days when there were no rogue cases. I think we can also keep in mind that the health experts get things wrong and are sometimes overly cautious. But I will continue to follow their advice.

        • Chris Lloyd says:

          I think her main point was that Dan promised us a good isolation system so that lockdowns would not be necessary but by his actions we now know he will lockdown if we have a handful of potential (not even confirmed) free cases.

          I think Dan had a good rebuttal, namely that the new strain is more infectious and the system needs to be rejigged for that. And the plan now seems to be to build quarantine in remote areas. I predict they will be full of refugees in a few years. ;(

          • The experts say that the problems were not due to a new strain, rather it was system failures yet again.
            It’s not chance that NSW has consistently outperformed Vic rather it’s because Victoria has , higher standards.

            • Chris Lloyd says:

              That makes little sense as an explanation. We had outbreaks last month which were controlled with contact tracing. We were promised that contact tracing would mean no more lockdowns required. The stated reason from the Premier for this lockdown was that it was unclear that contract tracing could contain the UK variant.

    • Chris Lloyd says:

      “…still doesn’t change my mind.” I do not think I was trying to change anybody’s mind. If you read my comment carefully, I am not making any hindsight assisted policy prescription. I specifically said: “None of this is to say that the decision was flawed. Decision making under uncertainty is tough.”

  9. Chris Lloyd says:

    Thanks for the Oscar angle! Very funny.

  10. john Walker says:

    Chris the primary reason for the latest lockdown was because they stuffed up quarantine.

  11. Tony says:

    There is some moral hazard here though. Places like Australia have virtually eliminated the virus and arguably curtailed some freedoms of its residents. At the same time, life has been very difficult for overseas for many overseas citizens (perhaps 40,000 at any one time) who wish to return. In some cases some families have probably been condemned to homelessness in foreign countries following loss of job and cancelled repatriation flights.

    I would also add new variants of SARS2 may perhaps have longer incubation times, potentially able to avoid hotel quarantine. At some point, hotel quarantine stays of 3-4 weeks become increasingly untenable. In the meantime, the Australian population is ‘dry-tinder’ for any disease outbreaks.

    Lastly, I would add, that Australian society has come together to fight covid via lockdowns, with a promise that vaccines are ‘just around the corner’. Vaccine delay literally destroys social capital.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.