86290SPRYT - Solving the patient no-show problemClosedCollaborative R&DInnovate UKEvery year 10% of all healthcare appointments in the UK are scheduled and missed without cancellation (25% in the US), at a cost of £1B. The no-show rate for routine immunisation and screening appointments is a particular problem with Do Not Attend (DNA) rates as high 25%-30%. No-show appointments negatively impact both patient and care teams. Missed screenings from no-shows may result in delayed disease detection, potentially putting patients in danger. "Missed hospital appointments waste patient and staff time, prevent sick people from being seen at the earliest opportunity and cost our amazing NHS an unjustifiable amount of money. Artificial intelligence has enormous potential to revolutionise healthcare and this is exactly the type of innovation our NHS needs to embrace to ensure every penny goes further as part of the Long-Term Plan" (The health secretary, Matt Hancock, The Guardian, Apr 2019). COVD-19 has greatly exacerbated this problem by driving up hospital waiting lists as routine procedures and checks were postponed and increasing the demand for digital health solutions to replace face-to-face appointments. Surprisingly, the no-show rate for virtual appointments is higher than in-person (\>30%). SPRYT is addressing this problem with AI by enabling patients to book or re-schedule medical appointments via conversational message or voice. **No new app to download**. Patients' use their preferred communication channel (WhatsApp/Alexa). SPRYT also predicts no-shows, maximizing the time available for patient care and increasing the efficiency of clinics and hospitals. Our solution combines 1) Conversational AI: free-flow conversational agent rather than a pre-scripted, decision-tree chatbot. If the patient can only answer yes or no to a question, the system's learning is very limited. 2) Reinforcement learning: for both predicting patient behaviour and to take action to minimise missed appointments. Knowing that a patient is likely to miss an appointment will not, by itself, make them attend. Therefore, even the most accurate model will only be truly valuable if its predictions allow us to take action and **change the outcome**. Changing the outcome of a no-show is our goal. Solving the no-show problem is a crucial step in removing a major obstacle in cost effective health care delivery and patient safety in a post COVID-19 world.