Your country needs you. Or to be specific, it needs your phone.
The new NHS coronavirus tracing app is to be trialled from this week on the Isle of Wight and ministers are pushing it with all the fervent appeal to moral duty they can muster. Since it’s being sold as the nearest thing to a safe way out of this nightmare, unsurprisingly early surveys suggest most islanders are willing at least to try something that promises to alert them when they’ve been in contact with other app users who later show symptoms of Covid-19. Throughout this epidemic the British have proved more willing than anyone imagined to do whatever’s asked of them, to protect the NHS or their own loved ones. But can it really be that simple?
The good news is that new mathematical modelling from the London School of Hygiene & Tropical Medicine (LSHTM) suggests mass “test and trace” strategies – variations on the theme of identifying cases quickly, working out who they’ve been around and getting those people to self-isolate before they unwittingly spread the virus – could cut transmission rates by 50-65%, compared to just 2-30% for mass testing or self-isolation alone. In other words, it beats the pre-lockdown strategy of getting people to stay at home only once they start coughing or showing a temperature.
The bad news, however, is that the researchers also suggest tracing apps may not by themselves be enough to contain the virus. The paper hasn’t yet been peer reviewed so shouldn’t be treated as gospel but under its model, keeping the rate of transmission (or the R rate) anywhere near the magic number of one, even with a tracing app, might still mean people restricting themselves to fewer than 10 contacts a day. That’s not so much a return to life as we knew it, as one measly bus journey plus a couple of close encounters with joggers in the park; mass home working and social restrictions would probably still be necessary. And then there’s the single biggest hurdle to adopting the app for the long term, which isn’t the threat to privacy (valid as that may be) so much as old-fashioned human irritation and inconvenience.
The one thing likely to get people deleting the app is being told to stay home for what feels like no good reason. It’s no good if, say, an alert goes off every time you’re washing up in the kitchen and someone walks past on the street outside. So instead of blindly pinging alerts to any user coming within six feet of a potentially infected person, the NHS app will use clinically informed algorithms to work out which encounters are actually risky, teaching itself as it goes.
The bonus is that the trial could give doctors critical insights into how the virus is still spreading even under lockdown, or which patterns of behaviour are risky. The snag is that means collecting data centrally, which isn’t the way Apple and Google want the app to work on their devices. To cut a long technical story short, the fear is that the resulting compromise might be a battery hog, draining mobiles annoyingly fast.
The “test, track and trace” mantra also stops intriguingly short of what happens if people refuse to self-isolate on receiving an alert. Britain isn’t likely to copy China in forcibly quarantining people against their will, but we may discover that ultimately there’s no substitute for the old school, labour intensive method of manual tracing – or employing trained staff to identify, call and coax the potentially infected into doing the right thing. A real live nurse is a lot more persuasive than a text message, which may be one reason the LSHTM team concluded manual tracing could work better than apps at reducing the R rate.
Matt Hancock’s promise to recruit an army of 18,000 contact tracers by the middle of this month suggests the government may already be formulating a plan B, or at least envisaging manual tracing working alongside the app to bolster its impact. It’s not impossible that the app itself will eventually become the equivalent of the half-empty Nightingale hospitals; a technically impressive feat to pull off in the time available, which still ends up being overtaken by events.
There’s no point in being overly churlish about decisions made on a vertiginously steep learning curve. If the Nightingales hadn’t been built yet their beds had turned out to be needed, the outrage would rightly have been visible from space; if the Isle of Wight trial ends up exposing some gaping holes, then arguably that’s the point of trials, as opposed to springing a half-baked idea on the country without having first tested it. Tracing apps may still be a piece in a bigger jigsaw of multiple small interventions, which aren’t in themselves silver bullets but collectively make the difference.
But this isn’t a magic key unlocking the door to normal life, and the risk of overhyping it is that disillusionment may quickly set in. Even with a successful app, we won’t be out of jail free, so much as on probation; and like prisoners released on a tag, still living with the knowledge that liberty could at any moment be curtailed.
• Gaby Hinsliff is a Guardian columnist
