How the U.K. managed to kill an estimated 1,500 people using COVID-19 and an Excel spreadsheet

0
10


The paper in question actually appeared on the MedRxiv preprint server just before Christmas (which is my excuse for missing it until now), and carries the title: “Does Contact Tracing Work? Quasi-Experimental Evidence from an Excel Error in England.”

So many times, research papers are so focused on delivering a complex story of results from an experiment that the years of work, and decades of previous research, barely shine through in the results. But this particular paper tells a very special story—one of a nation forced to wing it in the face of a massive threat, due to poor planning on the part of public officials, the sudden understanding that even a half-hearted effort could save thousands of lives, and how a choice that seemed absolutely meaningless caused more deaths than some of the worst battles of World War II.

The story goes like this: As COVID-19 was spilling out across the planet, the U.K., like the U.S., was doing essentially nothing. On top of this, decades of alternating dominance by neo-liberals in both major parties had starved the National Health Service of beds, doctors, and supplies, much less anything so frivolous as planning for future emergencies. Then, as the pandemic was officially becoming a pandemic, British officials seriously evaluated whether letting millions of citizens die might not be the most cost-effective response.

As The Atlantic explained a year ago, South Korea had already demonstrated the effectiveness of instituting a nationwide system of testing for COVID-19 and using dedicated case tracing to isolate those exposed to the disease. What had seemed to be an unstoppable disaster in that nation had been brought to a near halt through the simple means of taking down names and addresses, and telling a relatively small number of people to stay home while the rest of the nation continued on about its job. Still, Johnson was looking at a different proposal.

Instead of trying to stop the virus, there would be no national testing organization, and no hard social distancing rules. Instead the government would give “soft advice” to those over 70, telling them to stay home, while younger people were allowed to become infected.

Since the virus causes milder illness in younger age groups, most would recover and subsequently be immune to the virus. This “herd immunity” would reduce transmission in the event of a winter resurgence. On Sky News, Vallance said that “probably about 60 percent” of people would need to be infected to achieve herd immunity.

“Most” would recover. Though … 60% of the UK population being infected would have caused just a tad over 800,000 deaths if the nation matched the case fatality rate for those under 65 at the time.

This was not just the beginning of a disaster, but the major reason that the U.K. managed such a high death toll. Like Trump, Johnson failed to implement the programs that proved effective elsewhere in a timely manner, and refused to tighten down at a point where it would have done the most good. And like Trump, he defended his inaction until it was far too late to prevent the U.K. from seeing one of the worst outcomes on the planet.

It was only later, after Johnson himself had come near death, and the U.K. had outraced both France and Spain in the effort to murder its own citizens, that a serious attempt was made to implement a national system of case tracing. Because there had been no real preparation for that effort, the case tracing was carried out by saving the names and information of exposed individuals … in an Excel spreadsheet. 

Anyone who has ever worked in an office has a pretty good idea of what that means. They also likely have an idea of just how many ways things might go wrong. And the people keeping that list found one of those ways.

On October 3, 2020, UK authorities announced that due to a “technical error,” 15,841 COVID-19 cases that should have been reported between September 25 and October 1 have not entered the official case statistics and were not referred to the central contact tracing system.

The technical error in this case was that the Excel sheet was being saved in the older .xls format instead of the newer .xlsx format. For your family budget, this would not make a difference. However, the .xls format is limited to 65,536 rows. So, every day from September 25, 2020 to October 2, 2020, the people managing this sheet added thousands of names. And every day those names were simply lost, as the sheet was at the limit.

When this was realized on October 2, the format was changed, but looking back on the 15,841 contacts that had not been made provided one opportunity—it allowed researchers to see the results of outcomes from this group that had not been contacted, compared to the people both before and after who were contacted. As a result, they were able to estimate that the week-long failure of case tracing cost about 1,500 lives.

Bumbling incompetence by senior officials that leads a nation down a path to destruction. A belated effort to address the problem that sees plucky healthcare workers forced into using the tools at hand. An honest mistake made in an office that results in over a thousand deaths in homes across the nation. It’s a wonder the authors of this research paper aren’t fielding competing offers from the BBC and Netflix.

It’s almost hard to notice that, in the end, what this paper shows is that case tracing works. The researchers were able to effectively repudiate official statements from the U.K. government and their supposed team of expert advisors who insisted that case tracing would have “little impact” by showing that in just this case, the number of deaths connected to these missed calls was significantly higher than in the cases of those contacts.

Which is a very good reason to have teams ready to carry out this vital function throughout the remainder of the COVID-19 pandemic and to keep those teams in place in preparation for another such event, even if it never comes. Also … give these people an actual database.





Source link

LEAVE A REPLY

Please enter your comment!
Please enter your name here