5 minute read
The value of patient feedback
Dr Tom Palser, Consultant Surgeon and Clinical Lead at Methods Analytics and Simon Swift, Managing Director at Methods Analytics examine the need for utilising patient feedback
Complaints are never something that we really want to hear - but they are vital to ensuring the level of care and continued development within health organisations. Without patient feedback, even organisations at the forefront of innovation may fall short of realising what their patients actually need. It is imperative that complaints are not only acknowledged and acted upon, but also that they are interacted with in the best manner possible - and AI may play a pivotal role in this.
For reference, complaints to the UK NHS by patients or families are incredibly common, with the NHS receiving more than 175,000 complaints about its services every year - this translates into more than 3,300 per week. Our partner NHS trust (University Hospitals of Leicester NHS Trust) alone receives 4,500 complaints or concerns per year. Allied to this, the cost of litigation is steadily increasing - £2bn to the wider NHS and £22m to UHL alone, annually.
Looking at the issue elsewhere in the world, in 2010, a report by the Harvard Business School placed the cost of medical malpractice in the USA at $55.6bn, increasing the cost of care by an estimated 5 per cent. The vast majority of this, though not entirely, originates with a patient complaint.
Yet despite this wealth of information which is presented to us, these complaints are not always used to understand systemic or specific problems in care organisations, disregarding the rich information that they contain. While some complaints may provide very little information to learn from, many provide a very comprehensive analysis of our care systems from the most important perspective - the patient.
The sheer volume of complaints causes significant problems in the way
that complaints are managed and, most importantly, learnt from by hospitals and care facilities. Having to manually analyse and process these complaints is incredibly labour-intensive and takes considerable time. As complaints can be very long and emotive, analysing them, responding to them and learning from them objectively, can be di icult. Human beings are not good at synthesising information from large numbers of narrative sources as primacy and latency e ects overwhelm us. Human beings, even with the best intentions, also get bored.
As several high-profile reports to the British Government have found, this means:
• Already distressed patients and families frequently su er delays • Many complaint responses do not address all the key issues or questions
that people have raised, causing further distress and cost to hospitals and sta time • Perhaps most worryingly, hospitals cannot maximise the learning from past experiences • In turn, this means mistakes are repeated and patient care does not improve.
The solution – how can AI help?
To tackle this problem, we’re using a combination of automation and a branch of AI known as Natural Language Processing (NLP) to focus on this problem and enable healthcare providers to derive real value from complaints, both immediately for the complainer but more broadly for the organisation and future patients.
NLP is a branch of artificial intelligence that allows computers to interpret human language by “reading” and then analysing large blocks of unstructured text. Put simply, the system processes the data into a format in which it can be understood, then classifies it, and identifies the underlying issues in the complaint - this is known as topic modelling.
Although this system is still in development, it is based on previous work which we have successfully deployed for organisations such as the UK Health Regulator, the Care Quality Commission and the UK Ministry of Defence.
Our aim is by using the AI system, we can address and enable complainants to be better served, and for organisations to use the rich patient feedback to mitigate future risk, targeting these key areas:
1) Reassuring patients and relatives
a. A faster, more complete, and more accurate response to the issues raised in their complaints. Automation equals speed and appropriate workflows in this instance, so the complaint gets to the right person to respond to the complainant quickly. There is a lot of evidence that speed is a major factor in resolving complaints simply, as people feel listened to. b. Reassurance that all the issues raised are being identified and patterns are being viewed by senior leaders. Not always, of course, but o en when you speak to complainants the term used is ‘we want to make sure it doesn’t happen to someone else’, so having a clear pathway and a fast process gives the signal you are listening and acting.
2) Helping hospitals
a. Accurately identifying key problem areas in near real-time so mitigation can be put in place. This is about risk: clinical risk and executive risk. Executives who do not have strong process in place to manage complaints have an unmanaged risk. That is not good governance. b. Improving the e iciency of complaint management resulting in: i. Improved experience for patients and relatives, reducing the onward e ects of the complaint in sta time to manage the issue and potential later litigation ii. Reduced burden on sta of responding to the risk now and through the process iii. Reduced time clinical sta spend dealing with complaints, thereby allowing them to focus on clinical care. c. Focusing quality improvement e orts where they are needed most, thereby improving patient outcomes. d. Potentially reducing both the number of complaints going forwards, as risks are better understood and mitigations can be put in place and therefore litigation costs, by improving learning from complaints and by improving patient satisfaction with the complaints process.
Conclusion
Key to note is that although it makes the human’s task much faster, easier and more objective, it does not remove the human oversight of what is obviously a sensitive area.
“AI in healthcare” usually conjures up images of robots on the wards or using computers to read CT scans. However, healthcare is so much more complex than this, and many of the ways in which AI will improve care and improve e iciency are in “back-room” tasks. Although less “glamorous”, they are no less important for both the patient and the hospital.