Is AI the answer to preventing the collapse of public healthcare?

As the ageing population expands, along with a plethora of associated health conditions, the strain on public healthcare systems continues to grow significantly. A recent study from the Institute for Public Policy Research​ revealed that the backlog of patients awaiting treatment through the NHS in the UK would take 10 years to clear, whilst current strikes by junior doctors are likely to compound the issue.

Meanwhile, understanding of the importance of personalisation for health interventions is on the rise, with nutritional needs varying vastly due to different combinations of biological, lifestyle and environmental factors. Research shows​ that tailored advice in response to such factors translates to effective and sustainable health improvements as compared to generalised population-based recommendations.

Consequently, the use of AI to deliver personalised nutritional and health advice has been a growing phenomenon, with brands increasingly implementing such services through the collation of individual data from wearable technology.

AI-powered personalised nutrition service Prevess, for example, has been doing just this.

“By providing tailored health advice, PrevessAI effectively prevents and manages health issues," Timo Spring, CEO and co-founder, told NutraIngredients.  "Additionally, AI enhances the human aspect of healthcare, freeing medical professionals from time-consuming tasks like documentation, research and data interpretation, allowing more time for patient care."

He added that AI enhances the healthcare system by automating routine tasks, rapidly analysing large datasets and offering personalized health recommendations to reduce the burden on healthcare professionals, making healthcare delivery more efficient and effective.

Larger healthcare organisations have implemented personalised AI, including international healthcare company BUPA which launched its AI-assisted mental health service 'Take Care Of Your Mind' through its 'Blua' app in Spain. Using data collated through wearables and daily questionnaires collating data information around sleep, exercise, and lifestyle, the platform provides advice with the aim to prevent the worsening of symptoms and detect early signs of mental health conditions.

Key challenges​

Despite its potential, Spring drew attention to the primary challenges in the implementation of AI for health advice.

“Number one, ensuring accuracy and reliability of AI-generated advice requires integration of diverse data sources while maintaining patient data privacy and security," he explained. "The effectiveness of AI advice hinges on high-quality data, sophisticated algorithms and continual learning from new data."

“In addition, many companies attempt to implement AI without considering the expertise of healthcare professionals.

"PrevessAI believes in developing medical protocols collaboratively with medical professionals, enhancing efficiency through AI. An analogy is ChatGPT's use in marketing, where AI generates multiple effective headlines quickly.

"In healthcare, however, even slight deviations in AI recommendations could lead to significant consequences, emphasizing the need for expert oversight in AI applications.”

Practical guidance

Meanwhile, the team behind personalised nutrition data insights firm Qina is "very concerned" about the ethics of nutritional AI systems and is about to publish a framework for developing ethical and trustworthy AI solutions as a practical guide for companies.

Mariette Abrahams, CEO and founder of the hub, agrees that although AI may be a useful tool to reduce the current strain on healthcare systems, there are many factors that influence adoption and acceptance of novel technologies.

“Such factors include digital literacy, access to technologies and affordability," she said. "Recent studies have clearly demonstrated that digital health integration is not evenly distributed, especially in the older generation. Therefore, I think that AI shows potential in helping with self-monitoring and tracking, remote healthcare, 24/7 support, but only if there is human oversight and if everyone benefits.”

She added that key challenges include data privacy and security issues, as well as transparency of data sources used to generate recommendations. She said that reliable AI systems will require training on quality data, with human oversight to enable the consistent addition of the latest scientific findings and individual data.

“However, there are many questions and concerns in terms of what we are recommending to whom, when and why, when we have limited representation in training datasets, or insight into how AI systems generate recommendations," she noted.

She pointed out the importance of ensuring that AI systems do not lead to unintentional behaviours and consumption patterns, adding that "we need to be considering the long-term social impact of AI systems".

“I believe there is a huge opportunity to ensure that the solutions that are being developed are ethical, accurate and reliable in order to build trust. To achieve this, we need to have the right strategies in place and build solutions with the right nutrition and behavioural approaches from the start."