Ebene News – GB – COVID-19: How Oxford University’s algorithm decides who should protect

New tool works by using details of a person’s medical records to assess their likelihood of catching COVID-19 and dying

Technology correspondent

Around 17 million more people will be added to the protection list in England, after being found to be potentially at serious risk of COVID-19

They have been identified by a new algorithm that examines several factors and will receive a letter from the NHS in the coming days

Some will have already been offered a vaccine against the coronavirus, but the 800,000 or so who have not will be pushed back to the priority list for vaccination

Very simply, this algorithm works using details from your medical records to gauge your likelihood of catching COVID-19 and dying

He sifts through this information, then gives you a score out of 100: a high score means you’re more likely to catch COVID-19 and die, a low score means you’re less likely

People with the highest scores are added to the protection list and prioritized for vaccination

Whether you get one of the highest scores depends on your personal characteristics such as your age, gender or ethnicity, as well as your weight relative to your height

Older people known to be more vulnerable to COVID-19 Men are more vulnerable so they will score higher The same goes for ethnicity

Other factors that will increase your score include whether you smoke, be homeless, or most importantly, take certain types of medication.

Some cancer patients, for example, were not on the first protective list This algorithm can search the NHS database, select these people, and then flag them for vaccination

Poverty is known to be linked to more severe COVID-19 outcomes and this algorithm includes a measure of deprivation, based on your zip code

It takes a whole host of measures, including unemployment, level of car and housing ownership, and level of household overcrowding, and then uses them to calculate poverty in your area.

This gives you a score for it, which feeds into your final score when calculated at the end

You might be wondering exactly how these scores are calculated This is a much bigger question: suffice to say that it involves some extremely difficult calculations

Fortunately, all the details of the model will be posted so that people who understand math can walk through it.If this is not you, you can get a feel for it yourself by using this online calculator that uses the ‘algorithm to estimate your score based on the information you give it

There are of course concerns, but before we get into them it is worth stopping to think about the bewildering complexity of the situation

The algorithm doesn’t just try to predict who will die from COVID-19 if they catch it, which we unfortunately know too much by now He’s trying to predict who will catch COVID-19 in the first place

Unless the Oxford University academics who developed it invented time travel at the same time, the results will inevitably be imprecise, especially because they do not include some factors that clearly put you at a greater risk of catching COVID-19, such as Occupation

As a result, it is highly likely that some low risk people will end up skipping the queue

After the chaos caused by attempting to score last year’s A levels with an algorithm, the idea of ​​an imprecise algorithm selecting who to vaccinate might send your shivers down your spine

But if we focus only on its downsides, we risk missing out on its pros: its ability to sift through more data than a human ever could.

Algorithms are not inherently bad These are just tools – and like all tools they can be misused or used well This one could help the NHS select a vulnerable person from the population, but it is as good as the data included in it

Perhaps the biggest challenge will not be performing it, but explaining it to patients who receive (or don’t receive) an invitation accordingly

© 2021 Sky United Kingdom


Ebene News – GB – COVID-19: How the University of Oxford algorithm decides who should protect

Source: https://news.sky.com/story/covid-19-how-the-oxford-university-algorithm-decides-who-should-be-shielding-12220116