The health industry in the UK is under immense pressure, with staff shortages and funding restraints leading to unsustainable heavy workloads and low retention rates. It is clear that digital transformation is key to supporting healthcare professionals by relieving the administrative burden, reducing unpaid staff overtime and improving employee job satisfaction. The early August pledge of £250m for NHS artificial intelligence is one step towards realising this metamorphosis that has been detailed by Health and Social Care Secretary Matt Hancock’s Future of Healthcare document and the NHS Long Term Plan.
So what role does artificial intelligence and automation play in the health sector?
Artificial intelligence and automation already play a role in health. A highly publicised example of this is Ai-assisted robotic surgery to perform complex procedures. However, there are other examples that might fly under the radar, like reading patients vital signs and streamlining billing processes. The application of automation from an operational and administrative capacity is far-reaching, and new developments in technology are opening up further possibilities all the time. Some of the critical areas we at Alphalake can see automation benefiting in healthcare include:
In 2018, Theresa May announced an NHS rollout of Ai to cross-reference medical records, genetics and national data for early cancer diagnosis. Medical records, along with information about patients’ habits and genetics, will be cross-referenced with national data to spot those at an early stage of cancer. A Danish company has deployed Ai to analyse emergency services calls by listening to the background noises, tone of voice and words used in the call. The technology can correctly diagnose cardiac arrest in 93% of cases, compared to a human average of 73%. Additionally, automation will assist doctors to check large databases of medical information and verify diagnoses for less commonly found health issues. It will also collate health records across various departments and systems, so doctors have the patient’s medical history at their fingertips when they see a patient.
The unpredictable nature of emergency services has traditionally made it difficult to manage staff rosters effectively. Artificial intelligence can analyse the probable influx of patients based on data from a wide variety of sources, to predict the staffing requirements at different times of the year. Members of the public can also use apps or health department websites to check the wait times at emergency departments nearby. This technology should lead to a more even spread of patients at each hospital, a more manageable staff workload and a better patient experience.
From appointment scheduling to progressing scripts through the system, automation is set to revolutionise hospital administration. The health sector has long struggled to move past outdated business practices such as paper records. Automation promises to save much needed time dealing with administrative work, which can be channelled into improving patient experience.
Machine learning and Big Data
The introduction of machine learning allows for enormous quantities of data to be processed, filed, stored, expanded and accessed from around the world. As we move towards a global view of healthcare, the elimination of research silos will lead to a centralised database of healthcare information and clinical trial results. Machine learning also gives Ai an advantage when supporting human health workers, as each new software has a base point which includes all of the information and knowledge that has come before.
But there is a long road ahead
As health experts have been prompt to remind us, the NHS, and indeed many large multi-level organisations have a poor record with the integration of new technology. Careful planning and consultation are essential to successfully launch a well-planned ecosystem of infrastructure into a complex multi-layer organisation. Key considerations include:
Early involvement of healthcare professionals at all levels of the organisation
To view the implementation of automation as a high-level function, or the responsibility of the IT department is a grave error. As the impact of automation impacts on the workflows across all levels of organisations, it is essential that all levels are involved in the planning, strategising and discussion stages. This allows for early troubleshooting and robust analysis of where automation will add the most value. It also ensures that new systems can be integrated effectively with current hardware and software with minimum interruption to daily business.
Selection of processes suitable for automation
A recent McKinsey study found that there is a 36% automation potential within healthcare. The study calculated the technical feasibility, cost comparison of technology vs. labour, and the benefits such as reduced error rate and higher levels of output. It is essential to consider these points when determining the best areas of the business for automation. Generally speaking, manual, repetitive and rules-based tasks are the best place to start.
Strategies to address health inequalities
While we know that health inequalities exist around geographic locations, socio-economic groups and ethnicities, the introduction of digitally-enabled primary and outpatient care should make healthcare more accessible. As the sector transforms, consideration will be needed to avoid excluding those with low levels of digital literacy and ensure they have continued or improved access to care.
Ongoing Training and Support
As technology evolves and develops at a rapid rate, there will be an onus on organisations to look at ongoing training and up-skilling of their workforce, to keep pace.
As Health Secretary Matt Hancock summarised, we are “on the cusp of a huge health tech revolution that could transform patient experience by making the NHS a truly predictive, preventive and personalised health and care service”. The future of health is barrelling towards us, with enormous benefits for patient care.