Harnessing analytics to improve care

COVID-19 has profoundly changed the way people think about their health data and this is setting the scene for wider adoption of precision medicine, or the customisation of treatments to an individual.

Last year, vast troves of previously untouchable data sources were corralled around the world to track and trace infection, from mobile phones to credit card data.

The pandemic also pushed consumers to be more aware of their health and health data, which is changing how they communicate with doctors, say Deloitte experts David Betts, Leslie Korenda and Shane Giuliani.

‘‘The COVID-19 pandemic has turned the healthcare system upside down and challenged consumers’ sense of wellbeing,’’ they wrote last August. ‘‘Many consumers show agency and engagement … [and] more consumers are using technology for health monitoring and are willing to share their data.’’

And even highly conservative national health regulators have been forced to accept the realities of the now-digital world, as entities such as the powerful US Food and Drug Administration embrace ‘‘real-world data’’, information collected outside a clinical trial structure.

The foundation for the change in mindset has been the ability to collect, store, process and analyse data, using machine learning algorithms and advanced computing resources, and taps into new ideas about preventive and precision medicine.

But this digital infrastructure didn’t emerge overnight.

The US pioneered health data interoperability standards in 2011 and from that grew a healthcare analytics industry that includes behemoths such as IBM, which provides a digital backbone for a range of health-related companies, Epic Systems which runs electronic health records, and telehealth giants Teladoc and Livongo which completed a $US18.5 billion merger last year.

With that infrastructure in place, organisations were able to act on the need for health data integration just as consumers went from suspicious to almost demanding their data be used, says Eric Chung, co-founder and CEO of predictive healthcare analytics start-up Prospection.

‘‘We’ve seen places like Israel and the UK use analytics to accelerate medical decision making and effectively roll out COVID-19 vaccines. So people are asking the question ‘if we can do it for COVID-19, then what about other diseases?’,’’ he says.

Prospection, a Sydney-based company that is rapidly expanding in Asia, specialises in big longitudinal data analytics for cancers, rare diseases and speciality diseases such as heart failure, psychosis, hepatitis and HIV, where there is an unmet need for better tailoring treatment to patients.

‘‘Our machine learning algorithms are primarily about following patients through time. The way that can help is it enables a doctor to easily draw on insights derived from millions of patients. Doctors don’t usually have the ability to look at the history of patients over, say, 10 years and see how it impacts them today, but an algorithm can,’’ Chung says.

That data is funnelled into four key areas for which Prospection is building products.

Patient insights is one key area, closely tied with the mass collection of real-world evidence. Via a product called PharmDash, it considers a person’s treatment ‘journey’ over time and can offer recommendations for care if the data suggests a particular course is not working.

PharmMap locates patients in need of better treatment options across a geographical area, thus identifying where governments and organisations need to allocate more resources to improve patient health and, ultimately, save money.

Prospection is using innovation to figure out how to get all this data into the hands of decision makers via clinical decision support software.

‘‘We have the technological innovation piece, which is the predictive algorithms to anticipate things like early diagnoses for rare diseases doctors may never have seen or better treatment options,’’ Chung says.

‘‘And we have partnerships, and finding the right ones that can show these insights to doctors at the right time by integrating our algorithms into their software.’’

Several partnerships are already in development in Australia, Japan, Korea, Britain and the United States, and Chung expects the software to be live by the end of 2021.

The company’s challenges in accessing data and delivering insights to medical decision makers give an indication of how far the industry has come, but how far it still has to go.

‘‘There’s been two challenges for us. We have very good algorithms and we have the power to deeply analyse volumes of information, but we need access to more data and ideally linked data that allows for a more complete view of how patients are treated,’’ he says.

‘‘The other challenge is how to get the evidence and insights into the hands of the decision maker and into software they already use. This is the next part of the puzzle and requires the right technology and partnerships.’’

Prospection has a strong presence in Australia and Asia, and is now looking at Britain and the US. The initial disease it has targeted for both countries is the rare condition of pulmonary artery hypertension, with plans to expand the range of diseases covered.