One of the major healthcare benefits being created by mobile technology and associated digitalisation developments is enabling individuals to monitor and manage their health status.
And by doing so on an ongoing basis, they will also be able to assess their own health risks and manage themselves accordingly.
“If asked if personalised healthcare is becoming a reality, I would have to say ‘yes’,” Mediswitch Data Scientist, Paul Saunders, told delegates to the Africa Health Management Conference in Midrand last week: “For one, it can predict a chronic disease.”
Introducing the personal nature of what he referred to as the cognitive healthcare concept, Saunders presented a life cycle example with its inherent pressures and health impact along the way (simplified here): young healthy sportsman eventually marries, has children, has to earn to feed and educate family, pressure builds in 40s and 50s, suffers a heart attack and requires a bypass operation.
Had this individual had the benefit of a personal healthcare solution, i.e. being fully aware of his health status at all times via, for example, wearable monitoring devices and a mobile account of his health history, his story, Saunders said, would have been quite different.
A personalised healthcare solution, therefore, would have to make use of a patient’s medical history, lab results, genetic markers, wearable and other IoT devices: “And it should be mobile-enabled empowering the patient to take action.”
It’s advantages, he noted, had already been shown in a study of patients with diabetes with the development and utilisation of a predictive algorithm. Saunders described his organisation’s modus operandi as follows:
· Worked with one medical scheme
· One year’s claim info containing clinical information and lab results
· 19402 diabetes patients
· Submitted 138482 claims
· Amounting to R110 546 418, 09 (67% being hospital submissions)
“Based on this data,” Saunders explained, “we built an artificial intelligence algorithm that was 77% accurate and are now working towards adding a time dimension to our predictions and increasing the volume and variety of data used.”