A week ago England started trialling a new app for digital contact tracing. I am a participant in the trial.
There is some progress from the last trial on the Isle of Wight that is praiseworthy. There is lots of documentation about what has been built. But unfortunately I can find very little information about how the trial is designed, how it is being evaluated, and how it is progressing.
This is an important gap.
While I am a strong supporter of responsibly and proportionately using technology to effectively tackle the pandemic there is little evidence that digital contact tracing is effective.
Governments need to be open-minded that the app might not be effective, and should be honest with the public about the progress and outcome of the trial. This is a necessary part of making any public service trustworthy.
If the government does not publish information about the trial we risk speculation and rumours filling the gap.
Increasing transparency and openness about the trial will help it progress and, if the trial shows that the app is effective, help with the subsequent rollout of any public service.
Government recommends that clinical trials are transparent and open
Part of the English trial is a clinical investigation.
This has been registered and approved by the Health Research Authority. The Health Research Authority recommends that research protocols are published in the interests of transparency. The app team have decided not to do this.
The HRA say that such transparency will both help with the research and reduce duplicated efforts – for example in other countries, and UK nations, carrying out similar initiatives. While some of this information is being shared privately, it could also be shared openly.
Other governments have not published information about their trials. This is a chance for the UK to lead by example.
As Chris Whitty, the UK’s Chief Medical Officer, said on the launch of the HRA’s new MakeItPublic campaign: ‘Transparency and openness is essential in making the most of the commitment of patients, service users and healthy volunteers who take part in research.”
Government recommends that public services are built openly
Meanwhile part of the trial is non-clinical and not covered by the HRA approvals process.
This part appears to be about user experience and features – for example the new app publishes an area risk score and allows people to check-in to venues using a QR code.
GDS, the Government Digital Service team that leads on design practices for public services, says people building services should make things open, it makes things better. This is not only about sharing code and final designs, it is about sharing why and how design decisions are made.
There is a wealth of prior art and experience in the public sector that the app team could usefully learn from. It is another chance for the UK to lead by example.
The last English digital contact tracing trial failed
This is the trial of a second English digital contact tracing app.
The first app crashed on the Isle of the Wight. Government did not publish results from that trial either, but two reasons were given for its failure.
Government Health Minister Lord Bethell told the House of Lords that the trial had taught “one important lesson: that people wanted to engage with human contact tracing first, and quite reasonably regarded the app as a supplementary and additional automated means of contact tracing”.
Later Matt Hancock, the UK Secretary of State for Health, confirmed rumours that there were technical problems with the app. It would not work on Apple iPhones unless it used a protocol designed by Apple and Google.
But, let’s be honest, many people suspected a third reason.
The first app was not sufficiently trusted because of the loud, public debate about it and the resulting lack of trust among part of the public. Why would the public trust something with unclear benefits and potentially large risks?
The overall test and trace programme was subsequently found to have broken the law, so perhaps the public has good judgement.
While we do not have a loud, public debate about the new trial yet there are many people both in and outside the UK who might have a reason to start such a debate. Meanwhile the public’s trust in this government’s use of technology will only have worsened with the recent exam algorithm debacle.
If the government does not publish information about the trial we risk speculation and rumours filling the gap. If the app turns out to be effective then this could damage the UK’s response to the pandemic.
A step towards trustworthiness?
England’s new digital contact tracing app trial needs to be more transparent and open. People should be able to see the design of the trial, how the app is being evaluated, and how the trial is progressing. If the trial shows that the app is effective, this will help with the subsequent rollout of any public service.
The trial could go further still. The evaluation could be independently performed. But it remains to be seen whether this government can take this initial step towards trustworthiness.