An AI (artificial intelligence) conversational agent that can aid with patient aftercare following cataract surgery is being trialled in NHS hospitals.
Researchers from Newcastle University, Oxford University Hospitals (OUH) NHS Foundation Trust and Imperial College Healthcare NHS Trust have been studying the AI, named Dora, and its effects on patients. Read on to find out how it works, and how people have reacted to virtual aftercare.
Contents:
- What is Dora?
- Trialling AI aftercare
- Why cataract surgery?
- Reactions from patients
- The benefits of AI in patient care
What is Dora?
Dora is an AI-powered automated voice system designed to talk to patients and assess their needs based on the clinical significance of five symptoms post-surgery: redness, pain, vision issues, new floaters, and flashing lights.
It is hoped that Dora will help identify patients who need additional help from a human doctor, cutting down on time-consuming tasks and freeing up NHS resources for those who need them the most.
Trialling AI aftercare
Published in The Lancet’s open-access journal eClinicalMedicine, the study used feedback and data from over 200 patients recovering from cataract surgery, which is the most common surgery performed by the NHS. After three weeks, the patients received a follow-up call from Dora, monitored by an ophthalmologist for the purpose of the study.
The acceptability evaluation of the research was led by Professor Edward Meinert, the study’s co-Principal Investigator and Professor of Digital Health and Clinical AI at Newcastle University.
“Patient response to the AI system indicates its potential suitability for wider adaptation in other clinical pathways that could free up NHS time and resources.” Professor Edward Meinert.
According to Dr Kanmin Xue, Honorary Consultant Ophthalmologist and co-Principal Investigator at OUH, “the Dora system is able to call patients autonomously by telephone and conduct a conversation that gathers the key information needed to determine whether a patient is recovering well or requires further medical assessment.”
After the participants were contacted, researchers compared the decisions of the AI to those of the supervising clinician and evaluated them for accuracy. Feedback about their experience using Dora was also gathered from the patients to analyse efficiency and address any concerns.
Why cataract surgery?
Cataract surgery is one of the most commonly performed surgical procedures in the UK, with a minimal complication rate of 1-2%. According to the Royal College of Ophthalmologists, almost half a million cataract operations take place every year. This routine procedure treats blurred vision and involves removing the eye’s natural lens and replacing it with an artificial one.
With growing demands on the NHS and constraints in terms of resources, digital technologies have the potential to help address workforce constraints.
Routine ‘high volume, low complexity’ clinical pathways account for many waiting list backlogs and offer an excellent opportunity for improved efficiency.
AI poses a potential solution in freeing clinician time and prioritising any patients presenting with complications.
Reactions from patients
Health Innovation Oxford and Thames Valley carried out a feasibility study, assessed feedback from clinicians, and evaluated the cost benefits of incorporating Dora into the post-operative patient pathway.
They found that Dora had a moderate-to-strong level of agreement with the supervising clinician, and an overall accuracy of 89%.
Most of the post-call reactions were neutral to positive, ranging from patients finding the aftercare “fine” to saying they “enjoyed talking to Dora”. There was a favourable response to Dora’s efficiency and consistency of tone, with some noting that an AI wouldn’t have “off-days” and that they “gave [doctors] time to take a breath”.
Many participants felt that Dora could benefit patients and clinicians by increasing convenience, saving time and costs, and providing reassurance by ensuring that all patients receive follow-up.
Others said they felt “a bit remote and lonely talking to a machine”, disliked the “mechanical voice” of Dora, and felt they would not be comfortable using AI if they experienced complications post-surgery. Some voiced their concern over a lack of empathy, reassurance, and understanding of nuance from the AI, and preferred the availability of a clinician as a safety net.
Despite any concerns, Dora was considered a generally acceptable, safe, and potentially cost-beneficial alternative to clinician-delivered cataract surgery follow-up assessments.
“[Dora] was successful in identifying and prioritising those individuals who needed additional clinical input.” Kanmin Xue
The benefits of AI in patient care
When used effectively and with high accuracy, there are clear advantages to using AI to ease pressure on doctors and patients alike. It is thought that the AI follow-up assessments in the study saved approximately £35 per patient compared to standard care. It also has the potential to reduce hospitals’ carbon footprint and increase the accessibility of healthcare for all patients.
The researchers acknowledged the importance of not dehumanising patient care by replacing patient-clinician interaction with automation, especially in sensitive cases or those with complications. They also flagged accessibility concerns to be tackled in a future trial, such as the adaptability of this technology for neurodiverse patients, those with hearing difficulties, and non-English speakers.
“Since the trial, an updated version of Dora has been deployed at OUH as the default system for post-cataract surgery assessment for thousands of patients,” says Dr Xue. “We have now also expanded its use to pre-operative assessments, where Dora runs through a checklist of screening questions, including whether or not the patient wishes to have surgery for their cataract. The information is logged on our electronic medical records, thus speeding up the face-to-face nurse-led pre-assessment.”
Further research is needed to monitor Dora’s safety and effectiveness while unsupervised. Yet the success of the study opens up the potential for future trials to investigate additional factors, such as time, cost, barriers and facilitators to implementation.
Dora emphasises the importance of implementing such automation to ensure that clinical resources can be provided in a timely manner to patients with the greatest need, rather than replacing patient-clinician interaction points.
If successful, it is hoped that there will be potential applications of similar technologies to other clinical pathways within and beyond ophthalmology.
You might also like:
- read the study: Edward Meinert, Madison Milne-Ives, Ernest Lim, Aisling Higham, Selina Boege, Nick de Pennington, Mamta Bajre, Guy Mole, Eduardo Normando, Kanmin Xue, Accuracy and safety of an autonomous artificial intelligence clinical assistant conducting telemedicine follow-up assessment for cataract surgery, eClinicalMedicine, Volume 73, 2024, 102692, ISSN 2589-5370, https://doi.org/10.1016/j.eclinm.2024.102692
- learn more about the researchers involved in this study:
- Professor Edward Meinert, Professor of Digital Health and Clinical Artificial Intelligence at Newcastle University
- Dr Kanmin Xue, Honorary Consultant Ophthalmologist and co-Principal Investigator at Oxford University Hospitals (OUH)
