This week we had one of those workshops where everything’s going fine until it turns out different people are working with different assumptions.
Other parts of NHS England (NHSE) and Department for Health and Social Care (DHSC) have come to similar conclusions to us about how users will access prevention services in the future. We want to align our work and theirs and start working together to help more people get support to reduce their health risks.
The delivery team (and I) think we can help people improve their health by:
- asking people to tell us about themselves (a health check)
- working out their risks based on what they tell us
- connecting them to local services based on their risk level and what they want
Policy from NHSE and DHSC has a different model. They want to get to the same outcome by:
- using data we already have to find people who are at risk – there’s already a system that finds people at risk of heart disease
- connecting them to an online service – online services are cheaper than face‑to‑face ones (although actually only cheaper if they are effective)
I assumed the simplest way to help understand someone’s risk was to ask them about their health. We thought using data already in the system would come later.
We have some evidence that people prefer face‑to‑face support when offered a choice. Outcomes are better too. We also know that connecting people to local services will be our biggest challenge. Lots of people have tried and failed to get accurate, up‑to‑date information on what services are available. The team has done good work on how we can start to make this work.
If we follow the policy approach, it’s a big change for our work. We go from something that feels like a normal online service to something much lighter. It’s more like a marketing campaign. It’s an interesting and maybe challenging idea – someone getting a message saying the NHS thinks they’re at risk and should take action.
The targeted approach could work well. It cuts out the need for users to give us information before we can help them. But I’m also worried about what we lose. The biggest loss is not asking people what they want to focus on and not connecting them to in‑person support. Without those we risk not taking advantage of the services that have the best outcomes.
Digital behaviour change support has benefits and some advantages over in‑person. Digital services fit more easily around people’s lives. There’s less of a commitment necessary to get started, small steps are good. But maybe also lower engagement.
I sit between the team and the policy people. There’s something that happens with each layer you move away from users. The messy reality of someone’s experience looks simpler and more solvable the further you get from it.
Our focus for this phase is to get things live and start learning as quickly as possible. My view is that starting with policy’s model will speed‑up learning if we commit to building towards face‑to‑face support. We will need to be pragmatic about who this will and won’t work for. But building a proven pipeline of users that we can offer local services to, will give us a better proposition when we engage with local service providers.
We need to understand a lot more:
- how well does the digital support work for our intended users?
- how well can we target people using the data we have (is it accurate, can we differentiate levels of risk)?
- how soon can a model be built and tested?