Data in healthcare
Data is playing a bigger role than ever in healthcare. The amount of data available has exploded and the cost of storing and processing it has collapsed. Analytical techniques and artificial intelligence (AI) can reveal patterns and generate insights that were previously undetectable or unusable.
This data revolution creates the possibility of diagnosing and treating disease much more accurately and cheaply. Today, our understanding of the biology of disease is relatively limited and research evolves slowly. However, by bringing together information about lifestyle, pathology, genetics, and interventions, we will dramatically accelerate the cycle of discovery and application. Even with current treatments, using data to detect, understand and act on variations in care or to detect illness earlier could prevent over 100,000 deaths per year in England.
Healthcare costs are rising, and in many countries they exceed 10 per cent of GDP. To become more cost-effective, healthcare systems need to establish what works and what doesn’t, and encourage the right lifestyles, right care, and right provider.
Data can help make this possible:
• Sixteen hospitals combined data from staff rotas, theatre systems, and patient records to better understand operating theatre utilization. Analysis of the data enabled the hospitals to implement scheduling changes that increased the accuracy of operating times by 13 per cent and improved utilization by 12 per cent.
• A&E patients with sepsis have a 5 per cent mortality risk. Machine learning can be used to accurately identify patients with sepsis in A&E using data from Electronic Health Records. This approach has greater predictive performance than other clinical decision rules-based models .
• Predictive risk algorithms can be used by clinical research organisations to select sites most likely to recruit eligible participants and meet trial milestones on time. This has led to 15 per cent faster enrolment, 10 per cent lower costs for patient visits, and 40 per cent better targeting. This is expected to reduce the cost of getting a new medicine from phase 1 to launch by around 30 per cent, with better patient safety and trial quality.
Health systems everywhere will benefit from having more data as records are digitised and from diagnostic tests and wearables. The UK, though, is particularly well positioned. UK Biobank, cancer registries, and Genomics England are examples of globally-leading initiatives which position the UK as a hub for precision medicine, particularly in rare diseases and cancer. Our single health system is largely digitised, and NHS Digital is in the process of integrating patient records.
Capturing these benefits requires thoughtful leadership. Leaders must manage privacy concerns, keep data secure, and handle potential liabilities around the results of analysis. They must also overcome the barriers to making full use of data (in how information is structured, stored and understood) so that we can truly gain insights and act on them. If we do so, the benefits will be immense.
• A&E patients with sepsis have a 5 per cent mortality risk. Machine learning can be used to accurately identify patients with sepsis in A&E using data from Electronic Health Records. This approach has greater predictive performance than other clinical decision rules-based models .
• Predictive risk algorithms can be used by clinical research organisations to select sites most likely to recruit eligible participants and meet trial milestones on time. This has led to 15 per cent faster enrolment, 10 per cent lower costs for patient visits, and 40 per cent better targeting. This is expected to reduce the cost of getting a new medicine from phase 1 to launch by around 30 per cent, with better patient safety and trial quality.
Health systems everywhere will benefit from having more data as records are digitised and from diagnostic tests and wearables. The UK, though, is particularly well positioned. UK Biobank, cancer registries, and Genomics England are examples of globally-leading initiatives which position the UK as a hub for precision medicine, particularly in rare diseases and cancer. Our single health system is largely digitised, and NHS Digital is in the process of integrating patient records.
Capturing these benefits requires thoughtful leadership. Leaders must manage privacy concerns, keep data secure, and handle potential liabilities around the results of analysis. They must also overcome the barriers to making full use of data (in how information is structured, stored and understood) so that we can truly gain insights and act on them. If we do so, the benefits will be immense.
Leaders must manage privacy concerns, keep data secure, and handle potential liabilities around the results of analysis
Leaders must manage privacy concerns, keep data secure, and handle potential liabilities around the results of analysis