What we do

As part of a multi-disciplinary team, we realise the value of curated and linked healthcare data to develop, apply and evaluate both digital and non-digital interventions to improve the effectiveness, safety, and patient experiences of care. Through data fusion we combine relevant information of different varieties (structured and unstructured) and multiple sources to provide a more accurate description of the interactions patients have with the healthcare system. Our Health Services Research applies big data analytics principles and techniques to better understand quality in healthcare using a mixed methods approach focusing on the intersection of healthcare policy and practice. We aim to apply robust evaluation to interventions and changes that enable the translation of an evidence-base to the real-world healthcare setting.

Why it is important

Healthcare providers collect a wealth of data concerning the delivery of patient care in multiple Electronic Patient Records systems, covering the full patient pathway, and other systems that capture safety and experience insights and complaints from patients and carers (Consumer informatics). In the increasingly digitally enabled and digitally mature NHS, the problem is no longer a lack of data, but making sense of the huge volumes of information to allow both healthcare professionals to access the right data, at the right time, in the correct context; and for organisations to be able to take these data from silos and use them in supporting clinical services and understanding healthcare utilisation.

Our work uses translational data analytics for organisational knowledge, supporting clinical services and direct patient care.

How it can benefit patients

The appropriate application of data analytic techniques (Natural Language Processing/Machine Learning/Artificial Intelligence) against curated and linked datasets provides new knowledge to improve the quality, safety and patient experiences of care and care delivery. Real-time data-driven insights, in addition, provides opportunities to feed this information back into clinical systems for direct patient benefit.

Summary of current research

• Scale, spread and embed. Using natural language processing of free-text patient experience
  
• Complaints: Patient Complaints for Organisational Learning
• RECAP: Developing and validating a risk score to predict hospital admission for patients with acute COVD-19 in Primary Care. Joint project with the University of Oxford. Funders UKRI and Imperial Jameel Community Fund. 
• inSIGHT: Impact of COVID-19 on the adoption of digital-first technologies in primary care.
• Using electronic health records to develop and validate a machine learning tool to predict type 2 diabetes outcomes.
• Listening to the public’s voice: impact of digital-first technologies on quality, safety and equity of primary care during the COVID-19 pandemic. 
• Deriving novel consumer insights to better understand patient needs and preferences for engagement, including information on discrepancies in patient needs and service provision