Targeting acute kidney injury (AKI) to improve patient safety: Patient and carer inputs to design a learning healthcare system

Talk Code: 
Jung Yin Tsang
Benjamin Brown, Stephen Campbell, Niels Peek, Thomas Blakeman
Author institutions: 
University of Manchester


Acute kidney injury (AKI) is common, harmful and costly, making it a global priority for patient safety. Every year it is attributed to over 100,000 deaths and costs the NHS over £1 billion. By targeting AKI as a syndrome, it moves away from tackling individual diseases, allowing wider patient safety issues to be highlighted, such as polypharmacy and safer transitions of care, particularly for vulnerable patient groups. Much AKI research concentrates on hospital acquired AKI, yet nearly two-thirds occur in primary care, where earlier detection and intervention is more appropriate. Our research seeks to fill this gap by exploring the potential for using a computerised ‘learning healthcare system’ to improve the safety of patients with AKI in the community. A preliminary step to achieve this involves seeking views from patients and the public, aiming to elucidate any gaps in communication surrounding the care of patients with AKI and exploring how a computerised system could support improvements.


We consulted two established patient and public involvement (PPI) groups ‘H@PPI’ (Health E-Research Centre group) and ‘PRIMER’, (Primary Care Research in Manchester Engagement Resource) to outline the design and co-facilitation of a workshop. The co-design workshop was subsequently attended by ten members, including seven patients and carers, two patient facilitators and a PPI research officer. Following a brief presentation to introduce the project, participants were split into two smaller groups for personalised discussion. We also utilised the Ketso® kit to stimulate deeper reflection and discussion using colour coding and systematic representation.


The workshop helped identify potential merits and limitations of current practice not previously considered by our team. Gaps in communication included the isolation of primary and secondary information – especially at discharge and the delay in patients receiving the results of blood tests and AKI diagnoses. Potential solutions involved using a computerised system to join up certain primary and secondary health records, audit functions to pinpoint the most vulnerable patients and providing patient access to blood tests to remove the need for an extra communication step. Feedback was positive for the capability of a true ‘learning healthcare system’, where outputs are automatically fed back into the system to generate continuous improvements. There were also suggestions that were outside our project scope, but carries prospects for future research, such as using artificial intelligence to predict those at higher risk of AKI much earlier.


Patients and the public are integral to improving safety and outcomes, frequently highlighting gaps in basic care. Findings have informed the design of our system and suggested potential for future research. We aim to ensure ongoing engagement with patients and carers throughout the design of our learning healthcare system.

Submitted by: 
Jung Yin Tsang
Funding acknowledgement: 
This project is funded by the NIHR School for Primary Care Research