Launch of UK’s first condition-specific coronavirus chatbot

UK charity Versus Arthritis has developed the first AI chatbot for the 18 million people with arthritis and other MSK conditions affected by COVID-19. It’s the first coronavirus chatbot to be developed specifically for people with a long-term health condition in the UK.

COVA was rapidly developed by the in-house digital team with support from AI specialists Filament after experiencing an unprecedented demand on its helpline, website, online community and social media channels.

COVA is Versus Arthritis’ response to people needing information since the Government announced sweeping measures to stem the spread of coronavirus in mid-March.

The chatbot has been developed in just 14 days – showing how important it is for health charities to have an agile approach to digital service delivery. COVA incorporates health information developed specifically for people with arthritis and MSK conditions on the impact of COVID-19 with answers from Versus Arthritis as well as the British Society for Rheumatology, the NHS and GOV.UK.

COVA is available on the Versus Arthritis website but also can be placed on any website with a simple piece of code.

The virtual assistant is designed to guide users to the most relevant answer with buttons on four key areas: medical advice (e.g. medication), guidance on isolation, maintaining mental and physical health and managing work and finances. The charity is looking to expand COVA with a full text-chat interface in the coming weeks.

Launch of UK’s first condition-specific coronavirus chatbot

Crucially it enables those with arthritis and other MSK conditions which are among the 1.5 million having to shield themselves for the next 12 weeks or think they are at risk, to gain useful health information quickly and easily from a trusted source.

COVA utilises a coronavirus risk calculator developed by the British Society of Rheumatology to help people with arthritis and MSK conditions assess whether they are more vulnerable and should self-shield.

Versus Arthritis has a history of working with artificial intelligence, including its existing chatbot, AVA, the Arthritis Virtual Assistant, which has been providing information for people with arthritis and MSK conditions for over three years. COVA is being tutored by AVA’s experience to improve its interaction with users.

Speaking on the launch of COVA, Amanda Neylon, Director of Insight, Data and Technology said:

“Getting up-to-date, high-quality, trusted health information and services that are available 24/7 is important at the best of times, but particularly during our current crisis. COVA is a one-stop-shop where people with arthritis and MSK conditions can go with confidence to get the information they need to maintain their health and wellbeing.

“Our helpline and face to face services provide essential support for people every day. During this period of uncertainty, we need to increase our digital services – which include our online community, website health information, social media, and virtual assistant – to support the millions of people with arthritis and MSK conditions particularly those who are most at risk from the coronavirus.

“We’ve been pioneering the use of AI technologies in the charity sector for several years with our chatbot AVA. Our nimble and innovative response will have a big impact for people in a time of great change and uncertainty.”

Versus Arthritis is currently exploring partnerships with other health charities to see if they can add their relevant information to COVA, and so expand its capability.

Versus Arthritis teamed up with chatbot and AI specialists, on the development of COVA.

Pete Kemp, Filament’s EBM Lead said:

“It’s been fantastic to work with the Versus Arthritis team to build a trusted service supporting people with musculoskeletal conditions such as arthritis at this difficult time. The EBM platform empowers the Versus Arthritis team to improve their own services and health information continuously. The platform can be scaled to thousands of conversations and deployed to multiple chat and voice channels.”