The app is back - yes really!

Remember the NHS Covid-19 contact tracing app, the one that caused the ‘pingdemic’ back in the summer? You know, the one everybody stopped using, then deleted because it kept sending them home - it’s dead as a doornail, right?

Wrong. In fact, it’s just broken the “pings” record it set during that pingdemic. Being a. bit of a nerd about the app I keep an eye each week on the statistics published every Thursday by the NHS.

Today’s numbers are for the week ending December 22nd, so aren’t affected by the Christmas holidays, and they show that across England and Wales 698,646 alerts were sent out to people who had been in contact with someone who later tested positive. The previous record back in July was 690,711.

When it came to the positive test results entered in the app and then triggering those contact alerts, the latest number at 243,406 far outstripped July’s 148,277. That suggests two things to me - first, the obvious one that this is the biggest wave of infection we’ve seen and the record will be broken again next week.

But secondly the lower ratio of positive tests to alerts triggered does suggest fewer people are using the app so each case’s phone will have come into contact with fewer other phones with bluetooth contact tracing active.

Sources on the app team concede that use has declined since July but say that has been overstated. Indeed, the latest figures show there were almost 200,000 new downloads of the app over the week. Of course the big difference between now and July is that the app, just like the wider contact tracing programme, no longer instructs people in its alerts to go into isolation if they are fully vaccinated.

(For a while the app was out of step with manual tracing in not ordering contacts of Omicron cases to isolate - which I wrote about here - but then the Test and Trace system dropped the isolation policy for Omicron too.)

So what are we to conclude? It seems to me that this app, with its troubled history and its poor PR, is actually once again doing exactly what it was designed to do. It is giving people information they would not otherwise have - that the woman they sat next to on the bus or the man next to them in the checkout queue later tested positive - and then offering them advice as to how they should respond.

Note that I say advice not orders - unlike the manual contact tracing system there is no legal force behind the anonymous app, although a decision to enter a test result into it indicates that you have already identified yourself into the system.

Certainly, the researchers who designed the NHS Covid-19 app believe it has had an impact. Back in the Spring they argued in a paper in Nature that it had cut the number of cases and thereby reduced deaths. Even though Omicron appears less deadly, I am sure that they would say that by nudging people to get tested (if they can find a lateral flow test!) the app is helping reduce the spread of infections and thereby lessening the variant’s impact.

The curious thing is how quiet the government has gone about the app. When the idea was unveiled in Spring 2020, ministers told us it would be a key weapon against the virus

. When, after a troubled birth, it was finally launched in the autumn of last year, there was a massive publicity campaign and the message was that it was our patriotic duty to install it.

But for the last few months - crickets. App, you say? What app? Perhaps ministers are mindful of the uproar from businesses in the summer when pings from the app were blamed for sending thousands of staff home. But this is one “non-pharmaceutical intervention” which does not involve compulsion and is now incredibly cheap to run - I am told just a couple of dozen people are on the app maintenance team. For some reason, however, a government which was never much good at explaining how the app worked now seems to want us to forget all about it..

Note - plans to turn this Substack into a newsletter about tech and health are proceeding, albeit quite slowly. Get in touch if you have ideas about stories, companies or people it should cover.