Professor Susan Hopkins, Chief Executive Officer (CEO) of the UK Health Security Agency (UKHSA)‍

In 2022, an infection prevention director in an English acute trust opens another revision to testing guidance. Two floors down, a bed manager is hunting for a single room for a patient who needs isolation. On the wards, staff are still swabbing, and fewer of them are uploading their negative results to the national portal. The effort is real and constant. Whether it is changing outcomes is a question nobody in the building can answer with confidence. A government-commissioned evaluation has now examined 122 trusts, and it suggests that the uncertainty was justified.

The headline finding invites misreading. Researchers found no observable link between greater test kit allocation and lower inpatient mortality or fewer delayed care home discharges. They did not find that testing failed. Reported results made up at most 0.3 per cent of tests ordered, and 89 of 119 trusts logged no Covid-related delayed discharge at all, which leaves analysts unable to separate a true zero from an unrecorded one. A null result built on records never designed to answer the question tells us about the records.

For NHS leaders the more useful reading is diagnostic. The study judged routine hospital data fit for real-time bed management, which is how trusts used it during the crisis. Evaluation is where it fell short, and the shortfalls are specific. Absence records did not separate Covid from other illness, and kit deliveries were a weak stand-in for kit use. Community infection levels drove care home transfer delays, which means hospital data alone could never have explained them. Each finding translates into a data requirement that a design team could write down tomorrow.

There is also credit in how this evidence arrived. A system that commissions, completes and publishes an unflattering evaluation of its own pandemic policy is doing something public bodies find hard. Trust in institutions seldom grows from announcements of success. A willingness to examine uncomfortable questions in the open gives accountability something real to work with.

The qualitative findings carry operational weight. Infection control directors described a shortage of single rooms and occupancy so high that deep cleaning was often impossible, with poor ventilation a further constraint. Staff grew sceptical about repeatedly swabbing healthy colleagues, and many stopped uploading negative results. The study cannot say whether that scepticism was justified, because the outcome data cannot settle it. It does show that compliance depends on staff seeing a purpose, and that the burden fell on infection prevention teams already stretched by constant changes to guidance. The estate findings give a concrete case for capital spending on isolation capacity and ventilation, though that case sits awkwardly in a settlement where maintenance backlogs already compete for every pound.

The timing is useful. Current health policy points towards a single patient record, a move from analogue to digital working, and a federated data platform in trusts. These are the instruments the study says were missing in 2022. The practical test is whether the next national intervention can be evaluated while it is being delivered. That means results captured automatically at the point of testing and hospital activity linked to community prevalence from the outset. The technology is not exotic. Delivery depends on procurement discipline and interoperability standards, both within the system's control, although the history of NHS IT programmes counsels caution about assuming they will be applied.

The implications for each audience are direct. Trust executives should treat data capture as part of intervention design, since a policy that cannot be measured cannot be defended or refined. Policymakers should fund evaluation at launch, because retrospective analysis of operational data has now been shown to have limits. Diagnostics and health-tech suppliers have an opening in passive, low-burden capture that spares staff another portal. Patients stand to gain most, because future decisions about screening in hospitals can rest on evidence.

The NHS cannot go back and learn whether mass testing worked. It can make sure the next question is answerable. This study, uncomfortable as its headline is, supplies the specification.

‍