Four step app ‘endorsement’ process

Healthcare apps will go through a four-stage recommendation process as part of NHS England’s plans to create an endorsement model for these tools.

Speaking at the National Information Board meeting at the King’s Fund Digital Health and Care Congress, Diarmaid Crean, deputy director, digital at Public Health England, said the model was being developed to help the developers of good apps differentiate themselves from the “dross of apps that are of low quality.”

The first stage of the process will allow app developers to go through their own assessment and make sure the app is secure and works for the user.

The next stage will involve what Crean describes as a “crowd-sourcing panel”, which will see a community of experts from around the world look an app and validate the developer’s assessment.

Once an app passes this panel, it is then down to developers to work with academia to provide evidence that an app has efficacy and is able to deliver value.

The final stage is reserved for apps that Public Health England, NHS England and other partners feel can be massively beneficial to the healthcare system and will be subject to further independent evaluation.

Crean added that apps will be available as they go through the process, and that the endorsement could apply at any stage.

Further details on the process will be announced in September with the first ‘endorsed’ apps expected to be available by the end of the year.

The model is being developed by one of several NIB workstreams, which are working on ways to deliver goals to improve the use of digital in the NHS as outlined in the ‘Personalised Health and Care 2020’ framework.

Crean said NHS England and its partners need to provide guidance in the area as the current health application marketing is “not working”, referencing a study that showed that the cost and rating of smoking cessation apps had no real bearing on their use for people trying to give up smoking.

“The market isn’t giving citizens access to the apps they desperately need – the quality ones – nor is there any wisdom of the crowd.”

He added: “The endorsement model needs to give these good app producers a unique selling point so they can be found by citizens and patients. It must also help [developers] support revenue so they can survive.”

Questions still remain on how best to demonstrate the efficacy of apps in a healthcare system used to relying lengthy randomised clinical trials that don’t suit the fast-paced nature of mobile technology, as pointed out in a panel session earlier in the day at the King’s Fund Congress.

Phil O’Connell fellow of Staffordshire University's health faculty and creator of the NHS telehealth solution Simple Telehealth and Florence made a point that not everything a clinician does is based on a randomised clinical trial.

“Any clinician that tells you everything has an randomised clinical trial. Ask them to show you a paper that shows you if you don’t drink water it’s terminal.

“Some things need randomised clinical trials, where there is a change in clinical practice. What we are doing it’s not about that. It’s about changing the service. It’s about changing the way we do something not actually what we are doing.”

Session chair Charles Lowe, managing director of the Digital Health and Care Alliance, suggested that a form of randomised clinical trial was already in widespread use in a technology setting in the form of AB testing, which is used by online firms such as Amazon to test multiple versions of a tool or design in order to see what gets the best result.

“There is a way RCT can be done very quickly. One of DHACAs’s submissions to the NIB made the point that someone needs to sit down and redesign the way these things are done.

“The system can be designed in such a way that apps are evaluated in a way that is far, far faster than they are at present.”