Skip to main content
Blog

COVID-19 testing: Tackling eligibility

Caroline Finucane, Lead Content Manager in the COVID-19 testing programme, explains her team’s challenges when ensuring only eligible users could order a free COVID-19 test.

At the start of the COVID-19 pandemic, free coronavirus tests were only for key workers so they could continue their critical work. But eligibility for tests would soon change as infection rates started to spike.

We were ready for the first surge, which came in October 2020. The government swiftly kickstarted the expansion of testing with rapid lateral flow tests. We wanted everyone to ‘Get a free NHS COVID-19 test’.

A child watches a woman puts the end of a rapid covid test swab into a small tube of liquid

But by 1 April 2022, free testing for the general public ended as part of the government’s Living with COVID plan. When this happened, my team faced the challenge of adapting the online ordering service to provide tests only to those who qualified, and to inform users early on in their journey if they were not eligible. Here, I outline the challenges we faced controlling access to tests, and I share our solutions.


Challenge 1: Simplifying a complex issue

The changing eligibility criteria kept us on our toes, but what really challenged us were these 2 things:

  • the rules were different based on where in the UK you lived
  • eligibility was self-determined, and our description of eligible groups could sometimes be misinterpreted; for example, volunteers who were not eligible may have described themselves as NHS workers

When people visited GOV.UK to request a rapid lateral flow test, they typically did not know the complex eligibility criteria and wanted to know straight away if they qualified. Yet we know from experience that most users will not read lots of information on the start page – they go straight for the green button.

It’s much easier for people to answer simple questions than try to understand complex eligibility information.

The GOV.UK solution is asking questions within the service to determine if the user is eligible. It's much easier for people to answer simple questions than try to understand complex eligibility information.  

But if a user invests time answering questions only to discover at the end they can’t have a test, it can cause frustration and anger.  

Therefore, to manage expectations, we used a ‘gate page’ early in the user journey. This is an alert that interrupts the user flow with important information - in our case, informing the user who is eligible for a test.

 

This type of alert grabs attention after the start page because it looks different and is unexpected. And at around 325 characters, it’s easy to read and understand.

Testing showed it worked: most people understood who was eligible before they clicked ‘continue’. Then the rules changed.


Challenge 2: Conveying that eligibility has changed

We found that people only read the interruption screen once. The next time they used the service, they’d skip past the updated screen without reading it; “I’ve read this before”.

So how, in future, would we convey that eligibility had just changed?

Changing the H1 heading to state that ‘The rules have changed’ did grab attention, although it would only work until the rules changed again.

And regardless of this, there would always be some ineligible people who, for whatever reason, would plough ahead.


Challenge 3: Telling people they are not eligible

The testing service was often used by vulnerable or ill people, and simply ending the journey on an exit page with the standard message ‘you are not eligible based on the answers you gave’ without any further information could cause anger or anxiety.

That’s why we had to consider the user beyond the end of their digital journey. We offered guidance on what they could do next if they were worried or had symptoms, providing them with links for further help

 

Challenge 4: Verifying people who are eligible

Another challenge we faced is that users who were determined to order a free test could backtrack and answer differently.

Our solution was to add more questions so that people had to verify the information they had given. So, for example, if someone answers that they need a test because they work for the NHS, they must specify their area of work and their job title.

There are understandable concerns about adding these extra questions to an already long journey. But for policy teams, it had the desired effect: the number of completed journeys, and traffic, dropped sharply after we deployed it. Tests were no longer needed for most people, because it was now time to live with coronavirus.


Conclusion

After many iterations, we had a system that was working well. Only a small number of ineligible people were qualifying for tests. We could find a better way to do this, but it would come at a price.

Validation technology, for example, is technically tricky to deliver. And asking busy, reluctant GPs to order tests on behalf of patients they deemed eligible wasn’t the answer.

Another argument for a self-determined eligibility system is that users are largely honest. Experience tells us that many people who end up end up in the wrong place in the journey misinterpret the question rather than knowingly lie.

We cannot soften the coronavirus testing policy, but we can make some pragmatic allowances.

Also, many ineligible people who get through have an understandable motive. Perhaps they had a health condition ‘not on the list’ and pretended it was, because they were worried and thought they’d been forgotten. Or they did not want to be unknowingly infected while visiting their elderly relative, so they pretended they were in the NHS cohort. 

We cannot soften the coronavirus testing policy, but we can make some pragmatic allowances because any system that controls eligibility will never be 100% accurate.

The system meant that eligible people could get free lateral flow tests without too much difficulty, while the number of ineligible people getting through was at an acceptably low level.

It was a balance, and we got the balance about right.



Related subjects

Caroline Finucane, who leads the content team for the COVID-19 testing service, shares some principles that will be useful for anyone working on a fast-paced, complex government project.

Author

Last edited: 5 October 2022 9:36 am