Ever since the Dutch government uttered the words “herd immunity” after the UK gov framed it as meaning doing mostly nothing, internationally the Netherlands is mentioned in lists of countries that are supposedly taking a wait and see approach. That isn’t what’s happening of course. The country is in lock-down all but name, but as is the Dutch way it is implemented as an appeal for collaboration and solidarity (and people are indeed complying), and isn’t implemented heavy handed from the top down yet. It doesn’t mean that in the background there are no emergency decrees signed allowing the enforcement of every measure by police and even military police (it’s just that these are done on regional level, signed by the mayors in that region, so it isn’t on the radar of most people these things have been signed.) It also doesn’t mean no enforcement in practice.

One such example is Tomas Pueyo ‘Hammer and Dance’ piece. I’d have commented there, but it needed jumping to all kinds of hoops to be allowed to comment on Medium, and posting it here anyway is easier for future reference. He writes “Some countries, like France, Spain or Philippines, have since ordered heavy lockdowns. Others, like the US, UK, Switzerland or Netherlands, have dragged their feet, hesitantly venturing into social distancing measures.” “Governments around the world today, including some such as the US, the UK, Switzerland or Netherlands have so far chosen the mitigation path.

The French and Dutch measures are actually much on the same level, only with different cultural accents on the role of central authority, and from what I can see the same is true for Switzerland.

The Netherlands isn’t following a ‘mitigation strategy’ in the ‘weak tea’ meaning in the article as such, but doing much as Pueyo’s article suggests. Bring down the transmission rate, and then release and tighten the measures for the coming months to keep the transmission rate low.

Two quotes from the director of the infectious disease unit (RIVM) who is in charge from an interview (in Dutch).
Based on infection models we use, we estimate the effectiveness of interventions. If we add up the current interventions we conclude from those models that the transmission rate is reduced sufficiently. That is the most important goal.. (Dutch: “Op grond van de infectiemodellen die we hanteren, kunnen we het effect van de interventies inschatten. Als we de huidige interventies optellen, volgt uit die modellen dat de overdracht van het virus voldoende teruggaat om de verdere toename te doen afnemen. Dat is het belangrijkste doel,….”)

…That is why we must align everything, and next to interventions research their actual impacts. It could be we can reduce measures, and later need to introduce them again. It could also be we’ll introduce additional measures, it can go both ways. (Dutch: Daarom moeten we alles zorgvuldig afstemmen op elkaar en naast interventies ook onderzoek doen naar het effect daarvan. Het zou kunnen dat maatregelen versoepeld worden en later weer worden ingevoerd. Het zou ook kunnen dat er dan nog éxtra maatregelen volgen, het kan beide kanten op.”)

Actual behaviour of people will determine this mostly he adds. In short, it’s the ‘hammer and dance’ suggested in the linked article.

In conclusion, what you assume to see from further afield based on choice of words (‘herd immunity’ e.g.), headlines and your own cultural reference points, isn’t necessarily what’s happening on the ground. The same is true here inside the country, many people are making very different assumptions based on what they perceive the RIVM and government doing or not doing. Part of it is that the RIVM isn’t very pro-active in getting information and data out there to put it mildly. The interviews with its director in the past week have provided me with much more clues and understanding of what they are working on than their press communiques or their own website. I assume the director has better things to do than give interviews, so there’s room for much more pro-active disclosure and transparency, as it would free up his time at least. RIVM doesn’t have a knowledge, data or decisiveness issue, they’re clearly following every bit of available science, it has a communication and transparency issue. Key question for me is what we can do to assist RIVM in taking on their data publication and communication needs.

9 reactions on “It’s Easy to Conclude the Wrong Thing

  1. Agreed that the Netherlands are going for some sort of ‘gentle’ lockdown. To the point of rigorous testing of suspected cases and contact tracing I don’t think NL is collecting enough critical data the country needs to make informed decisions, similar to the US and the UK

  2. true, because limited tests avaliable for broad use. However NIVEL network used for general tests provide statistical measure of overall & geo spread. Those not reflected in published stats AFAIK but do take place. Means actively measure top of pyramid & establish sense of base too.

      • Sure. The NIVEL network ‘Zorgregistraties Eerstelijn‘ monitors for all kinds of things, like regular influenza, and self reported complaints by people to GPs. They added alertness and actual testing for Corona from the end of January (4 weeks before the first confirmed Dutch case). An announcement of that was posted on their site 7 Feb. RIVM mentions positive tests found through the NIVEL network in their daily Corona updates every now and then. The incidence level of all complaints are reported weekly by NIVEL on their site. General method description, and map of GPs involved is available. Their weekly bulletins with incidence and test numbers, graphs with multi-year comparison are published as PDFs

        • Thanks. I did a quick reading through the documents but didn’t see much information about their testing samples and how they are related to RIVM data, so it is difficult to put things in a context.

          A quick and dirty estimation based on the numbers of their last report (17% of the population with reported flu symptoms and 8% of flu symptoms tested for coronavirus) gives 1,36% of the population with coronavirus. Which would definitely lower than the actual, since it doesn’t take into account asymptomatic cases and some of those with the symptoms that do not get reported.

          And, an article citing the numbers of tests done a week before says that 57 samples were tested and only 2 gave coronavirus (https://www.nrc.nl/nieuws/2020/03/14/corona-stresstest-premier-slaagt-volk-zakt-a3993747). I do not expect that the numbers for this week are much bigger. If this is the data that the government is relying upon to make decisions next to testing reported by RIVM than I am even more worried than before.

          It was also amazing to see how the tone of the articles changes from “don’t panic, we know what we are doing” (above) to “nobody could see it coming” (https://www.nrc.nl/nieuws/2020/03/20/dreigingen-werden-hier-nooit-echt-nu-wel-a3994476). And yes, the author is very opinionated, but this is also how the tone of the messages from the government has changed in practically a week time. I just got an emergency alert via sms while a week ago they were still discussing if schools should be closed.

          So, I do not share your optimism about well-grounded decision-making in the Netherlands. I do not see the data and the methods used to interpret it to be so sure.

          • We don’t see more than a few glimpses of what is done or measured. Yet we all are inclined to think that what we do see is all there is. That is simply a false assumption, as it is in any situation. Yes, the data that is visible to us is very limited, yes the RIVM could be much better at transparency which is my key point. Yes you are worried, we all are. No, you and I do not know enough to second guess the RIVM because we don’t know what they do, and lack the expertise. Yes, they do tap into an enormous reservoir of knowledge, ánd they are crappy communicators (have been for a long time). The latter doesn’t mean they’re also crap at the former. Public communication also isn’t their assigned role. Yes they make choices on how to measure, but it’s not their role to proof the validity of that to me. Crappy transparency should be addressed, but it’s not a sign of something being wrong in and of itself. Yes, I trust the RIVM above my own ability to get an overview.

            Of course you are getting an NL alert now, because the sun is out, it’s weekend and people prove not to follow directions to stay home and keep distance, by flocking to the beach etc. Interventions are always a toolkit and their severity and strictness depends on actual response and behaviour. If people will not follow advice, we’ll have a more enforced lock-down somewhere this week. (The needed regulation is already in place, as it is the same regulation underneath the current measures, they’ll just implement the enforcement part of it more stringently, security region by security region. BOA’s and police first, then marechaussee) Which you could have concluded weeks ago. We’re not on a chaotic path, but on a spectrum of scenario’s created to deal with high uncertainty.

            I’m not optimistic, we’ll be in this situation for a year (RIVM says much the same thing by the way. ), and I am currently assuming lock-down in whatever form will continue well into May if not June (10-12 weeks total, length depending on how well people follow advice). It also isn’t a theoretic exercise to me. A friend lost a close relative, other peers are ill. On the economic side I need to ensure ten or so people stay employed. Meanwhile we’ll double cases every 4 days, slower if measures have an effect, faster if people don’t stick to them.

            Data is only useful if it can be reliably used for decisions. That shapes the design of data gathering especially when constraints w.r.t. capacity and resources are in place, as data is never intrinsically informative. So it is useless to gather data that won’t inform decisions (such as larger scale testing for its own sake, but not being able to test everything you encounter, meaning larger scale testing is just partly testing in a random fashion), that’s just window dressing because the crowd out here is hungry for more mental crutches. The decisions taken here do not differ much from those in other countries, so that tells you something.

            I’m not here to make anyone optimistic, because I’m not. I also don’t hear the government nor RIVM saying there’s room for optimism. I’m not here to convince you the people at RIVM, GGDs and hospitals are acting based on their knowledge which surpasses all of us not on the inside. I will not convince anyone to trust professionals in institutions if they’re not willing to be convinced, nor do I set out to convince anyone.

            Stay home, stay at distance from others. It’s what the data and the professionals already tell you. More data, while definitely welcome, will not tell you anything different, nor will it reduce your worry. Stay home, stay at distance from others, is also the foremost thing to act on within your individual power at this moment that is meaningful in dealing with this epidemic.

  3. Nuttig perspectief na de adrenaline rush na het lezen van Hammer and Dance. Heb zelf bij het KNMI gewerkt en net als bij alle andere expert/wetenschappelijke instellingen van de NL overheid zoals RIVM, PBL, etc wordt er werk van wereld klasse gedaan alleen weten doorgaans alleen insiders dat. Jammer genoeg is het niet meer vanzelfsprekend in de ‘post fact world’ dat er wordt vertrouwd op de uitspraken van experts. Meer transparantie zou het vertrouwen inderdaad ten goede komen, daar mag in geïnvesteerd worden. Dank voor uw post.

Comments are closed.