CEOs and board members will be held increasingly responsible for preventable mortality and harm in hospitals. However, harness AI to help make sure your hospital is safe and then use it to prove that it is, advises Dr Mark Ratnarajah, managing director of C2-Ai
In August this year, the Prime Minister pledged £250m to boost artificial intelligence and genomic testing in the NHS citing, among other things, the benefits of harnessing NHS data. The long-term vision is to help the NHS become a world leader in utilising innovation and technology.
At the time, Health Secretary, Matt Hancock, commented: “We are on the cusp of a huge health tech revolution that could transform patient experience by making the NHS a truly-predictive, preventive and personalised health and care service.”
“We are on the cusp of a huge health tech revolution that could transform patient experience by making the NHS a truly-predictive, preventive and personalised health and care service.”
The importance of utilising accurate health data in relation to patient safety, isn’t a new phenomenon. The Keogh Review in 2013 identified areas of urgent action, as well as some common barriers to delivering high-quality care, one of which was the 'ability of hospital boards and leaders to use data to drive quality improvement'. The review recognised that this was made more difficult by how hard it is to access data held in different places and in different ways across hospitals systems, which is where AI comes into play.
The NHS Board Principles for Good Governance document stresses the need for 'evidence-based decision making' in support of a board’s 'statutory duty of quality'. Boards are required to endorse and sign off declarations of assurance to regulators in relation to quality. Making sure you are certain of the performance of your safety monitoring systems is vital and whether the right quality metrics are being measured effectively – particularly when thinking about future liability for inaction today.
Dr Mark Ratnarajah, Managing Director (UK), C2-Ai
Put simply, clinical Ai systems help healthcare institutions reduce avoidable harm and therefore mortality, as well as generating significant operating expenditure savings. Good-quality healthcare costs less to deliver (Institute for Health Improvement (IHI) 2003). Now, innovative technology allows healthcare organisations to set a benchmark and gold standard for improving quality, reducing harm and variation in care, as well as delivering much-needed cost efficiencies. AI can be used to highlight potentially-harmful issues that need resolving and then demonstrate, by continual monitoring, that a hospital is safe.
So, if you’re upgrading or implementing a new hospital reporting system, what key questions do you need to ask to ensure that the technology provides you with the data set you need to ensure your organisation is safe operationally?
Some reporting systems produce data and information 12 months in arrears, whereas others provide a current accurate baseline picture in real time or after a ‘30 day’ lag. If you are basing safety and cost efficiencies on this information, immediacy is key to minimising harms and mortality.
Assuming your system has identified an issue, how is this issue then coded and described? Most systems work from Disease Resource Groups (DRG) coding, but DRG codes do not reflect the reality of how clinical departments are organised. Moreover, many systems focus on the first 2-3 diagnostic codes for a patient, but this overlooks how patients will often be referred between a series of doctors before the correct diagnosis is reached. And finally, because these systems then aggregated the results into a statistical prediction, they cannot identify which patients may have been affected when generating an alert. So let’s say you have received an alert for mortality in ‘bronchitis’. Where do you look? Which specialty is this happening in? And which cases does it refer to? Your clinicians will naturally, and understandably, pose these questions and will be mistrustful of data that cannot answer them.
However, with the right technology it is possible to accurately identify and code issues in a way which identifies the root cause, and the cases that need to be reviewed. This will then enable clinical teams to drill down and resolve issues in a far-more-efficient, targeted and effective way.
The Institute for Healthcare Improvement states that a 'total systems approach' to patient safety must address harm as well as mortality. Identifying avoidable harm allows hospitals to save lives and costs in terms of re-admissions, admin, medications, bed blocking etc. So, make sure reporting also highlights avoidable harms as part of a comprehensive solution within truly evidence-based safety. Mortality in hospital is fundamental to providing safe care. But it is by focusing on the avoidable harm, which is far more common, that one can reduce avoidable mortality, and at the same time improve overall quality for the much larger group of patients who may survive, but still suffer unnecessary complications.
Improving patient outcomes is only possible if you measure performance accurately. The diagnostic history and risk of each individual patient should be taken into consideration to give a true and accurate reflection of both performance and outcomes.
The diagnostic history and risk of each individual patient should be taken into consideration to give a true and accurate reflection of both performance and outcomes
If you don’t adjust for risk, systems will, for example, deliver many false high complication rate alarms that don’t consider the high-risk nature of a set of patients. The end result is that a significant amount of clinical time is wasted investigating a problem that doesn’t exist, or simply reflects the fact that a group of patients were admitted with a set of complex problems.
When genuine risk-adjusted methodology is used an accurate picture of performance can be obtained, set against national and international benchmarks, and underlying issues can be resolved.
Most reporting systems look at a shortlist of harms that are easy to count, such as pressure ulcers and wound infections, but then produce prodigious amounts of data about them. Or they look at single issues like Acute Kidney Injury, but focus on identifying and treating the issue rather than preventing and pre-empting it. This barely scratches the surface of what is really going on. There are more than 150 traceable complications and harms. These are often interrelated in ways that point to the root cause of a problem that is much further upstream. And in some cases it is the lack of treatment rather than an active error that has led to the problem. It’s only possible to see the whole picture and find the solution if all of these parameters are captured and analysed.
However, it is also the case that the ultimate causes and solutions are simple to report and act upon. The best systems – especially those backed by AI to do the complex groundwork – will not drown you in reams of unassociated data points, but rather present the problem and the solution in a nutshell so that you can focus on making the improvements and reducing the risk of past mistakes.
Prevention is the best cure and identifying those patients at the greatest risk of harm, either before or at admission to hospital, will ensure that they are on the correct treatment pathway and optimise care from the outset to reduce care variation, anticipate and avoid harm from occurring in the first place, thereby reducing avoidable costs. Can your current system identify those patients at the greatest risk of harm from problems, such as acute kidney injury or hospital acquired chest infections, before they appear, rather than reporting on the harmful result once it has already occurred?
South Tees Hospitals NHS Foundation Trust and North East Academic Health and Science Network acknowledged that acute kidney injury (AKI) is a major problem among hospitalised patients.
As recommended nationally, South Tees Hospitals NHS Foundation Trust operated an AKI alerting system – flagging up potential AKI indicative test results to doctors. The trust recognised that it needed to embed this better into current practice and developed an ‘AKI aware’ culture to measure rates, with the overall aim to reduce results by 20% within 12 months.
The trust commissioned CRAB Clinical Informatics (C2-Ai) as they were able to baseline data in near real time, extracting patient episodes involving ICD-10 codes either as primary or secondary diagnoses. The system enabled the trust to track the incidence of AKI in surgical and medical patients. Whilst it confirmed that prevalence was in line with national averages, the trust was convinced it could deliver an improvement.
An AKI Awareness Programme was set up, including workshops, the introduction of guidelines, as well as appointing an AKI advanced nurse practitioner. The results have been impressive – whilst initially aiming for a 20% reduction over a year, within as little as five months, the data showed a 36% reduction.
While many trusts run awareness campaigns around AKI, the difference at South Tees was that it had embedded a multi-strand approach, with a change in culture. Long after the campaign had ended, AKI rates continued to fall. Furthermore, there was a ‘halo-effect’ with reduction in other avoidable harm events and improvement in care of the deteriorating patient, the sickest in hospital.
The trust’s patient-level cost team calculated that AKI was costing £1.65m a year at the 1.8% prevalence rate. The fall in incidence of AKI equated to 118 episodes a year and with the difference in costs of patient with and without AKI calculated at an average of £4,500 – total savings have been estimated at £500,000. This converts into a 6.8-fold return on investment. Reduced length of stay and avoidance of critical care were the key factor in these savings and the trust has since developed a business case to expand its existing service.