Workforce management and rising care costs are just two of the pressing issues healthcare needs to address today. Dr Chris Bates, Director of Research and Analytics at UK-based clinical software company TPP, said health systems will need to adopt big data approaches to target models. new patient-centered care and technology to address such issues.
In a TPP-sponsored presentation at HIMSS22 APAC, Dr. Bates shared some of the latest uses of machine learning and data analytics in driving better care delivery and outcomes during the pandemic. .
Use big data
Clinically, TPP can use millions of patient EMRs to develop algorithms that can predict whether a patient has an ovarian cancer tumor, and another that provides a consistent view. of patients with the highest risk of developing cancer.
Meanwhile, on the operational side, TPP has helped to optimize the work of nurses in rural communities by developing ML algorithms. This technology can increase nurse-patient contact time by 40% and improve both patient and nurse satisfaction.
During the pandemic, the TPP was tasked by the UK government with working on an analytical platform for COVID-19 research. It powered the OpenSAFELY open-source platform for six weeks, which was then used by epidemiologists and public health analysts to run analysis on “tens of millions of records,” including care data. primary and secondary, immunization records and laboratory data. Bates said researchers can do this safely at home because the platform is built with tight layers of security.
“What we’ve built there is a safe, reliable, research environment that works at scale and huge data. It gives you all the power you need without compromising your health. privacy balance,” he said.
In developing the algorithms, Dr. Bates emphasized the importance of closely involving clinicians in the process.
“We’ve been working very closely with the doctors – and I think that’s important. We’re not trying to fight the doctors. We’re working with them to try to better reflect clinical practice. ready and use all those items of data to try and reflect on how real doctors work using multiple sources of information to make that decision,” he said.
Furthermore, while patients are willing to let the health system use their data to improve healthcare, they also want their data to be handled “particularly responsibly,” said Dr. Bates. emphasize. “We need anonymity and data aggregation. We need levels of security in place.”
Novel data source
From all of that work, TPP can identify new data sources that have the potential to “impact the future of analytics.” This is textual information, citizen data, and genomic information.
According to Dr Bates, the same AI algorithms they use to detect ovarian cancer also apply to helping governments respond to the pandemic. For example, textual symptom data is run through these algorithms to detect exactly where outbreaks are occurring. AI is also used on text-based data to detect emerging COVID-19 symptom patterns.
Measure data capabilities
Reflecting on the future of analytics, Dr Bates urged the audience to “continue to build” capabilities in analytics and machine learning. They should also apply metrics to measure their improvement in getting analytics to both clinicians and patients. “SIR [Adoption Model for Analytics Maturity] I think here is absolutely amazing and takes organizations on a wonderful journey from fundamental to descriptive and predictive analytics,” he suggests.
“In case [analytics] If you don’t empower the workforce if they don’t provide better, safer care, then all this work is completely useless,” he warned.
