Coronavirus death distributions – has anything changed?

Today I was wondering whether much had changed since I had a look at some of the initial coronavirus death distributions back in, perhaps, April.

When looking at the age groups of people who “who have died in hospitals in England and either tested positive for COVID-19 [or where] COVID-19 was mentioned on their death certificate” (emphasis added) until mid August 2020 the total numbers for young people are still very low.1

Age group (years) n
Total 29,452
0 – 19 20
20 – 39 212
40 – 59 2,288
60 – 79 11,197
80 15,735

However there was quite a perceptible spike for the excess death rate of the group of 15-44 years across 22 European countries.2

Most of these probably had a pre-existing condition, as this NHS data3 running up to mid July shows:

Age group Yes (cond.) No
Total 27,767 1,379
0 – 19 yrs 16 4
20 – 39 177 33
40 – 59 1,993 265
60 – 79 10,499 569
80 15,082 508

Of these conditions, diabetes is by far the most common, associated with 26% of deaths.4 This rightly raises interest in the lifestyle and environmental factors contributing to diabetes.

  1. ↩︎
  2. Source: ↩︎
  3. ↩︎
  4. ↩︎

Is the widespread use of infant formula partly responsible for obesity rates?

Is the widespread use of infant formula partly responsible for obesity rates? The following passage suggests that there might be a link:

“Overfeeding during the postnatal period influences the development of the hypothalamus (such as neuronal connectivity) (Plagemann, 2006). Leptin, a hormone, is a produced by adipocytes and nutrition-induced changes in this hormone during development may result in abnormal hypothalamic development and function (Bouret, 2013; Plagemann, 2006). Animals exposed to postnatal overfeedings display leptin resistance that occurs before the animals become obese suggesting that this hormonal resistance may initiate the development and maintenance of obesity (Glavas et al., 2010).”1

  1. Donna L Mendrick, Anna Mae Diehl, Lisa S Topor, Rodney R Dietert, Yvonne Will, Michele A La Merrill, Sebastien Bouret, Vijayalaskshmi Varma, Kenneth L Hastings, Thaddeus T Schug, Susan G Emeigh Hart, Florence G Burleson, Metabolic Syndrome and Associated Diseases: From the Bench to the Clinic, Toxicological Sciences, Volume 162, Issue 1, March 2018, Pages 36–42, ↩︎

Mechanical errors

In response to my feedback on their essay, today a student asked me what ‘mechanical errors’ were. While it’s actually not that straight-forward to find a good definition online, I found a nice table that presents different categories of writing errors with their sub-categories. It is quite a useful schema to refer to when giving feedback (only skim-read the remainder of the article, so no endorsement — I’m in particular wondering what the authors mean by “Intelligible errors”, of which they seem to have found none ..).

Table taken from Wu, H.-p. and Garza, E. V. (2014). Types and attributes of english writing errors in the EFL context—a study of error analysis. Journal of Language Teaching & Research, 5(6).

Livin la vida loco2

Want to escape the Google/Apple duopoly in your mobile phone usage and still go soft on the climate in a convenient way? While lots of transport apps, for example the popular Citymapper, complain when you use an Android version (e.g. CrDroid) without the Google framework installed, train ride booking company Loco2‘s app runs well without Google.

Citymapper: there’s work to do!
Loco2: well done!

How to name that *@#! paper?!

Ever wondered how to give your research paper that precise / witty / punchy title? Read no further and leave the confines of this blog to indulge in Giorgos Kallis’ Getting the Title Right (you know how to spot a hyperlink when you see it, no?).  Enjoy!

Is mobile phone and wi-fi radiation something to worry about?

Is mobile phone and wi-fi radiation something to worry about? Here are really interesting comments by scientists on what they want the world to know about the importance of the International electromagnetic field (EMF) Scientist Appeal that calls upon the United Nations and its sub-organizations, the WHO and UNEP, and all U.N. Member States, for greater health protection on EMF exposure.

For people who are not biologists or medical experts it is certainly difficult to assess the risks of mobile phone and wi-fi radiation. That’s why we have experts. Plenty of them seem to agree that more safeguards are in order.

In the European Union “the precautionary principle may be invoked when a phenomenon, product or process may have a dangerous effect, identified by a scientific and objective evaluation, if this evaluation does not allow the risk to be determined with sufficient certainty.” What would need to happen for the European Commission to invoke the precautionary principle when it comes to, by-now ubiquitous, electromagnetic fields? How much happens currently the way it does as a result of lobbying and nobody wanting to spoil the party? Are we even able to seriously imagine a reversal of the rollout of wireless technologies?



Mark Hertsgaard and Mark Dowie (2018): The inconvenient truth about cancer and mobile phones, The Observer,

Roxanne Nelson (2014): Children Face Higher Health Risk From Cell Phones, WebMD,

UPDATE: Joel M. Moskowitz (director of the Center for Family and Community Health in the School of Public Health at the University of California, Berkeley) (2019): We Have No Reason to Believe 5G Is Safe, Scientific American Blog,

Cyberduck for working remotely on high-performance computing clusters

Are you working remotely with R on a  high-performance computing cluster? Does it not allow you to directly edit your files with R Studio by using R Studio Server?

I have spent endless hours tiresomely navigating my files with Midnight Commander, editing them with Nano from the shell and transferring them by hand with SCP.

Now I have discovered Cyberduck and my life is much easier. I simply edit files via Cyberduck with R Studio or an editor on my desktop, and transfer simply by drag and drop.

Syria and the taboo on chemical weapons

The US, in cooperation the the UK and France, have just launched military strikes on Syrian regime facilities, with the declared intention to punish the use of chemical weapons.

One of the main worries about these strikes is that they are targeted at a regime that is backed up by the presence of Russian troops who fight alongside the regime. Observers are justly worried that, if anything goes wrong, such moves may bring us closer to nuclear war.

In the run-up to the strikes, Friday’s Economist newspaper (14th-20th April, p. 11) argued, under the title of The duty to deter, that

Punishing the use of chemical weapons will not end the suffering in Syria, or unseat Mr Assad. But if the taboo on chemical weapons is allowed to fade away, other despots will be tempted to use them, too.

Let us look at this line of argument, taking into account the danger of nuclear escalation: if nuclear powers showed restraint in punishing the use of chemical weapons in a situation where the alleged perpetrator is fighting, on their own territory, alongside another nuclear superpower, would it follow that other despots will be tempted to use chemical weapons, too? I think we would need to give a qualified answer to that question, which goes beyond a simple yes or no. Yes, other despots working alongside a nuclear superpower may feel emboldened to use chemical weapons. No, it would be unwise for those despots, who are not fighting alongside the forces of a nuclear superpower, to assume that such action may not turn them into targets for retaliatory strikes.

Only last year, the Economist freely acknowledged that its support for the 2003 Iraq invasion was wrong. While it is understandable that the press seeks to make a neat and succinct case for or against something, the Economist should now be wary of any ill-founded sabre-rattling, and weigh each case for military action with due consideration for context.





How to properly sync Scrivener with the cloud

Many blogs/websites advocate using Scrivener together with Dropbox or other cloud services.

However, when you sync Scrivener with the cloud (such as Dropbox or OX Drive) you risk loosing texts. Scrivener saves the text snippets as separate files and when you have a big project, like a dissertation, you run the danger of not noticing that some of these texts were not correctly synced and you may end up noticing only much later that you are stuck with old versions. Good luck finding the desired versions of the text snippets then!

To minimise this risk, implement the following set-up:

  •  Store your Scrivener document file in a non-syncing folder on your hard disk
  • Automatically save backups to a syncing folder

Solve problems with Bibdesk / Bibtex and Ox Drive


Do you use the wonderful open source reference management solution Bibdesk / Bibtex and the synchronisation software Ox Drive, which is, for example, used by the privacy-aware mailbox provider Have you ever run into trouble with your .bib (bibliography files) getting messed up or do you want to avoid them getting messed up?

Make sure you always have sufficient free back-up space on your Ox Drive! Otherwise the .bib file gets duplicated, with one version upon another piling up and, on the way, references getting lost ..

However, even then data may get lost, for whatever reason. To avoid this, save your .bib file not in a synced folder but on an ordinary folder on your hard disk. Then set up an automator work flow to make daily backups from the non-synced folder to the synced folder.