We have seen increased use of artificial intelligence (AI) in healthcare during the pandemic to support NHS services to continue delivering care. With many trusts lacking the digital infrastructure needed to streamline even basic administrative tasks, however, it is vital that this is the starting point for widespread AI implementation. Improving efficiency and workflow will demonstrate the value of AI, building confidence in new solutions and laying the foundation for bigger changes.
One of the few positives we can take from the pandemic has undoubtedly been the innovation in healthcare during 2020. With the pressure on, the use of various technologies was adopted or accelerated to strengthen the NHS response to Covid-19 and keep services running.
The optimisation of healthcare that AI and machine learning offers has been particularly crucial in bolstering resilience. In July, for example, NHSX rolled out predictive technology that has allowed local teams to forecast Covid-19 hospitalisations up to three weeks ahead. Similar technology has been applied to predict hospital supplies of oxygen in order to mitigate shortfalls.
These developments are a significant step on the NHS journey towards greater digitisation and demonstrates a readiness to embrace new solutions. Beyond the pandemic, however, we need to ensure that we get the basics right before implementing the widespread, everyday use of advanced technology.
New solutions can sometimes make things harder for clinicians where the digital infrastructure is not in place to support their use, which has often been the case within the NHS. This has created a scepticism amongst some healthcare professionals about the potential benefits of new technology that cannot be ignored. So we must examine the long-term effects of solutions introduced during the pandemic and their impact on the staff experience to determine where technology is truly adding value.
Nuance’s whitepaper on administration and burnout during the pandemic, published last year, made clear that many of the IT systems currently in place do not promote efficiency around basic administrative tasks, so it is crucial to address these needs first. The target for electronic health records to be rolled out across all trusts, for example, must be met before the NHS can think about moving beyond that.
Additionally, there is still a perception that AI will reduce human interaction, when the reality is that automation of certain functions will release time for more of it.
Gaining trust by demonstrating the value of solutions first-hand will be vital to getting the right people on board. Leveraging AI to streamline basic functions is an important initial step and will reinforce the idea of AI as an enabler with healthcare professionals still in control. Using it to improve the efficiency and efficacy of care delivery, and to aid workflow, will increase confidence in new solutions in the future.
AI-powered speech recognition technology is an example of a good starting point. By consistently and accurately capturing patient interactions and reducing the time clinicians spend documenting information, speech-to-text solutions provide tangible value and could foster the digital infrastructure needed for bigger changes. These types of investments will build trust and lay a sustainable foundation for AI that can deliver real improvements in healthcare.