Empowering local councils could save Test and Trace

The Excel testing fiasco this week has dealt a further blow to the government’s already battered credibility. Nearly 16,000 positive Covid cases were “lost” in Public Health England’s IT system, resulting in long delays in tracing the contacts of those infected.

PHE was running an outdated version of Excel, it seems. Some £12 billion has been budgeted for the Test and Trace system. A £100 upgrade would have prevented the problem.

It has now emerged that in the scramble to reach the 40,000 contacts on Monday, Ring Central, the call system used by NHS Test and Trace, was overwhelmed by the volume of calls and repeatedly crashed. Contact tracers struggled to work through the backlog and were locked out for 30 minutes at a time.

In the Commons, Matt Hancock was careful to blame PHE for the cock-up, deflecting attention from the already beleaguered Test and Trace system, which has struggled to cope with increased demand for tests in recent weeks.

But the latest howler raises a more fundamental question of design: Is a highly centralised system really the best way to run NHS Test and Trace?

A curriculum staple of management science courses is the trade-off that exists between the reliability and efficiency of centralisation and the responsiveness and flexibility of decentralisation. As Carl Propper and Nelson Phillips argue in a joint paper for Imperial College Business School, in the case of NHS PPE procurement, the problem was “a lack of the centralisation of purchasing when efficiency and reliability was the most important goal.”

By contrast, “the real problem is not a lack of resources or expertise in track and trace, but an overly centralised system when local responsiveness and flexibility are required.”

In the past, contact tracing for infectious diseases such as TB or polio has been a local affair, not just because outbreaks have tended to be small, but because speedily tracing contacts is essential in controlling outbreaks. The sooner an infected patient’s contacts can be tracked down, the less likely it will be that they go on to infect others. In this respect, localised systems have a big advantage.

“We have an irrational approach to testing at the moment,” Professor Allyson Pollock, a member of the Independent SAGE committee, told Reaction. “We’ve already got a very fragmented health service in England. [The Government] has compounded that fragmentation by creating a privatised, centralised system for tracing and testing – alongside NHS labs doing testing in hospitals – instead of using the existing structures and systems, as has been done in Germany.”

“Serco, Deloitte, Amazon – none of them have any experience of contact tracing. Regional public health systems do not start from scratch, they have a lot of experience and expertise. Testing needs to be embedded locally in primary care, but it’s been carved out.”

Germany is often touted as a testing success story. Its highly localised approach has been credited with helping to keep the country’s Covid death rate relative to population size one of the lowest in Europe. More than 400 local community public health teams were given the authority and resources to manage testing and tracing in their locality from the beginning of the pandemic. Regional health authorities were reinforced with redeployed civil servants and extra staff and supported by state institutions.

There are models for effective localised testing systems closer to home. In response to testing failures, councils have started to launch their own testing and tracing operations, independent of central government. Officials in Sandwell, in the West Midlands, have said that Test and Trace only reaches 60% of contacts in the area.

Lisa McNally, the director of public health in Sandwell, has said that “as soon as a new case comes in now, we’re not waiting for [PHE’s] test and trace to fail to reach them, we’re phoning the same day.” In Bradford, currently under a local lockdown, the city council has said it would like to copy Sandwell but lacks the funding to do so.

The council in the rural Welsh region of Ceredigion set up its own in-house contact tracing system in March and has been able to rapidly trace those potentially infected with council staff and volunteers going door-to-door when a contact cannot be reached by phone. Wales as a whole has kept its tracing system in the public sector and managed to trace 90% of listed contacts compared to around 70% in England. The vast majority have been contacted within 24 hours.

Other decentralised approaches appear to have worked. The University of East Anglia has used its own labs – with a capacity of 1000 tests a week – to test 1500 of its students since the start of term. Results are returned within 24 hours and draconian restrictions on the majority of its 17,000 students have been avoided. Most are back in the classroom.

Is there a balance to be struck? Even if NHS Test and Trace becomes more devolved, some level of national oversight and coordination would be required. And whichever approach is adopted, success will still depend on the capabilities of those implementing it.

It might not be a case of “the more devolved the better”. But there will certainly be a sweet spot and valuable lessons to learn from those places where a decentralised approach has been tried. Ultimately, there’s no substitute for competence.

What’s needed, of course, is both – a testing system optimised to the nature of the logistical challenge coupled with high-calibre leadership. The string of testing fiascos, of which the Excel glitch is simply the latest instalment, shows that right now we are lacking both.