Alphabetical order mismatch and 52 of 78 neighborhoods had wrongly merged data. I spend a lot of time teaching advanced inference methods, but boring research data management remains the most essential skill. And that includes auditing for merge mistakes.

Drug-resistant tuberculosis is much more expensive to treat

Today's data insight is by Saloni.

You can find all of our Data Insights on their dedicated feed: ourworldindata.org/data-insights

Exciting news: I will be teaching a 4-day summer school on open science this August!
The course is for ECRs in the social & behavioural sciences who want to get up to speed with open & reproducible practices and/or learn how to handle specific problems.

utrechtsummerschool.nl/courses/data-s…

New Nature Medicine
The longest randomized trial assessment—4 years— of semaglutide (Wegovy) in participants without diabetes
nature.com/articles/s4159…
Sustained weight loss, no new side effects noted

New article by me:

The rise in reported maternal mortality rates in the US is largely due to a change in measurement.
ourworldindata.org/rise-us-matern…

The change was adopted by different states at different times, resulting in what appeared to be a gradual rise in maternal mortality.

In the UK you can be arrested for noticing a prosecutor's case is shoddy.

Be My Eyes with GPT-4o

US maternal mortality rates seem to have risen severely since the early 2000s.

But rather than an actual rise in mortality, analysis by the US National Center of Health Statistics shows that this was due to a change in measurement.

Saloni explains:
ourworldindata.org/rise-us-matern…

I am honored and excited to be taking on the role of President of CGD! Looking forward to working with all the brilliant researchers there.

This is why I enjoyed the book 'Jane Austen: Game Theorist' and frequently recommend it
amazon.co.uk/Jane-Austen-Th…

This is an amazing result. If the whole country had adopted the new measurement approach at once, the issue would be obvious from the sudden leap up, but because it was done state-by-state, nationally it looks like there's a gradual trend upwards. Should be a big story.

Great post, and this phenomenon is not uncommon in health data. In fact it’s thought that some of the observed rise in self-harm among young people may be an artefact of similar changes to recording guidelines.

Very important, complex article, re: maternal mortality. These procedural changes in coding: better aid identification of events, create larger short term error/variance, complicates country comparables. Going forward, hopefully we have more consistent data to see true rate, etc.

Great article. Beyond the headline, note that maternal mortality (and morbidity) is believed to be underestimated in most settings, and surveillance in low and middle income countries is very poor.