Connecting lonely people at the push of a button

The Department of Digital, Culture Media and Sport (DCMS) funded a first-of-its-kind project in Liverpool to trial 5G in Health and Social Care services. The testbed area included Kensington and Fairfield, some of Liverpool’s most deprived areas, with some of the lowest levels of connectivity. If the project proved successful it could be transformative for the community it served. 

The Challenge

9 million people in the UK (about a fifth of our population) say they often or always feel lonely. Even more worryingly, a third of those people didn’t feel comfortable admitting they were lonely or would be willing to talk about it with someone. Loneliness can affect any age group, especially those with mental or physical disabilities. We wanted to tackle this problem head on. What if we could ease loneliness at the touch of a button? Push to Talk aims to do just that. 

Some of the key challenges included:

  • Connecting people in communities. The device only works if other people are on the network, connected with the people they’d like to talk to and can do it at a time that suits them.
  • Making the device discreet. People were already struggling to open up about their loneliness so the solution needed to be unobtrusive and welcoming to use.
  • Integrating the device into existing environments. We needed to make sure the device would work in a regular home and be easily understood by its user.
  • Be free and accessible. Due to the vulnerable nature of the user groups we wanted to make the device as readily available as we could. 

Services

  • Iterative design
  • Prototyping
  • Data analysis
  • Software development 

The Project

Working in partnership with an iconic group of businesses called The Consortium (including Sensor City, Liverpool University, Liverpool John Moores, The Royal and Broadgreen University Hospital, E-Health Cluster, Liverpool City Council, BluWireless and CGA Simulations) we were able to begin to deliver on the ambitious 5G project.

To make conversations possible on the network there were some key elements that needed to be in place:

  • Bespoke button design. We needed to create a device that was attractive, discreet and easy to use. The button uses a LoRaWAN PCB antenna and microcontroller to make it possible for a message to be sent via the gateway.
  • The Things Network. We chose to use the largest operator on the planet for IoT, The Things Network. The network enables the project to securely send messages across an extended area at an extremely low cost.
  • LoRaWAN Gateway.  Gateways are mounted on buildings and lampposts. The gateways use The Things Network as the LoRaWAN service provider
  • 5G network. The LoRaWAN gateway is mounted onto a lamppost with the 5G node and is connected via an onboard ethernet port, which provides the network route to the internet. 5G enables high bandwidth and stable connection in outdoor locations. Plus, there are no extra costs associated with digging or wiring the entire route.
  • API and Twilio server. The API server has the ability to connect the phone numbers of the service users wanting to talk. Twilio provides the call service (i.e. being able to start and stop conversations). The software used in Push to Talk is bespoke to us.
  • Systems dashboard. As well as the end users, we also had to consider the services administrator. They needed to be able to see how many buttons were online how many button presses there have been, how many calls, when those calls took place, how long each call was and to which button numbers.

Collaborating with key partners

Once we had the technology in place it was vital we reached the people who needed the service. To do this we partnered with several collaborators to make the project a reality. These partners included Local Solutions, The Irish Centre and Irish Community Care, The Hope Centre Tuebrook, Everton In the Community, Asda in the Community, Live Wire, LCVS, Prima Homes, Bradbury Fields and Breckside Park Care Home.

Getting people to participate 

As well as imagining, building and implementing Push to Talk Defproc has also played a crucial role in participation. We wanted to get a hands-on feel of how people might respond to the device, who would benefit from it most and how we could reach more people. To do this, we’ve taken many approaches to collecting user data –

  • In-person interviews. For months our team worked tirelessly to reach out to people in communities to gather their initial impressions, measure their desire for the device and implement improvements based on their feedback.
  • Attendance at relevant local events. To understand the people the device would be helping we attended many events such as coffee mornings and Carers Week. By meeting with people, seeing the challenges they face day-to-day and asking them their opinion, we were able to gain a deep understanding of our users.
  • Analysing data through the dashboard. As well as qualitative data we also gathered quantitative data. By analysing how people were using the device in the real world we were able to iterate the device and make it even more helpful to its intended user.

What the future holds

Push to Talk is a living and breathing project. We are currently developing what the best process will be to allow organisations easy access to Push to Talk, so that their service users can access it quickly and easily. As well as looking at new ways to make the device more accessible to different user types. For example, there have been suggestions to add a voice as well as lights when a voice is connecting. In this way, someone with a visual impairment will understand what is happening when a call is being connected. 

What people think about Push to Talk

“People who are caring and living on their own don’t always have someone to talk to. I love using it. I contact other carers and we have a laugh, it does everybody good”

– Mary, full-time carer 

2022 Update

Push to Talk has secured new funding and undergone a technical (and physical) transformation. Read all about it in our latest blog post.