Inclusive use of technology can improve patient outcomes and service quality, says Alex Kafetz, Director of International Projects at Beamtree
Healthcare organisations are overwhelmed by the amount of data they produce, and the number of market solutions offering to solve this problem. It’s easier if senior executives have an idea of the value of their data and how it can be used to improve their services and thus their patient outcomes.
In today’s digital world, it should be simple enough to research a surgeon or the department he works in. But in practice, it’s been difficult to quantify rates of success and failure without the numbers to back them up.
Similarly, how can unused data in hospital systems be repurposed to create exciting new areas that will engage clinicians to adopt it?
For Alex Kafetz, Beamtree’s Director of International Projects, it has to be done in such a way that doesn’t complicate further the lives of busy frontline professionals. He says it’s about making their lives easier, giving them more time to spend with patients to discuss treatment options, and preventing risk.
Beamtree is an Australian company that understands how to collect and integrate data, interpreting information and gathering insights. It focuses on allowing providers to reinvest their resources, so they can deliver better quality care, reducing avoidable deaths, readmissions rates and length of stay. In practice this means improving the accuracy and timeliness of data to enable tracking and enhance the quality of care, by providing clinical support for better decision-making, and delivering cost efficiency through activities such as price benchmarking.
“With respect to the big American EMR vendors, clinicians haven’t always been able to realise the full value of these systems,” Alex says. “However, they are a fantastic data repository, so much of our activity involves taking the data from the electronic medical record and turning it into meaningful applications, primarily for doctors. But we also spend a great deal of time with health information managers and clinical coders, and we spend a large amount of investment in data quality as well.”
Beamtree’s solutions include PICQ®, the industry standard for or assessing and reporting on the accuracy and specificity of clinical coding data. A quality assessment tool, it uses more than 900 indicators to assess and measure coding accuracy against the ICD-10 classification enabling benchmarking across hospitals, and provides insights to improve reported activity. There is also RippleDown Expert™, which automatically applies clinical expertise to generate patient-centric reports, replicating the expert’s unique decision-making process at scale. Developed by pathologists for pathologists, it improves the quality and speed of information delivered to clinicians. RippleDown Expert™ provides treatment recommendations, real-time alerts, and advanced reflex and appropriate testing recommendations. Added to the mix is RippleDown Auditor ™, which automates real time data reviews, ensuring accuracy in data entry and providing visibility. Beamtree’s clients operate across 25 countries and over 1,100 locations.
The importance of data quality
Beamtree is currently working with Saudi Arabia to improve the consistency of coding and data reporting. “In Saudi Arabia we audited a large number of hospitals on the ground actually during COVID restrictions, which was another challenge,” Alex says. “The aim was to help them with Vision 2030 by understanding the baseline for the quality of data in the Kingdom. Now we’re moving forward to help them consider how it can be improved, and the first step is top quality data.”
The complexity stems from local coding customs. The International Classification of Diseases (ICD10) system turns the patients journey into a set of codes which are used for planning and analysis, but the issue lies in the recording by humans – this is where the errors creep in. They can result in apparently poor performance by a provider, but on inspection there could be a simple reason. Alex tells the story of a UK hospital that had a high instance of foreign objects left in patients following surgery. On inspecting the data, it turned out to be synthetic lenses after cataract operations, but the coder had thought they were errors.
Other examples can lead to price benchmarking – the classic story is hip replacements where one code can make a procedure more expensive than another. “Worldwide, we don’t have enough clinical coders,” Alex says. “It’s a very specialised profession and people with that mindset and that kind of skills are being diverted into data science roles. If we can automate the easy end of the scale, which frees up the coders to code manually the more complex material, which is intellectually stimulating, we can protect that part of the workforce and also code quicker.”
Preventing patient deterioration
A fascinating part of Beamtree’s Ainsoff Deterioration Index product is around patient deterioration, tracking their progress and seeing which interventions have benefit. This is based largely on the work of Beamtree’s lead clinician, cardiothoracic surgeon Dr Levi Bassin who is based in Sydney. Having spent many years unable to perform scheduled operations on patients because they had deteriorated unexpectedly, he decided to develop algorithms that could determine the risk of deterioration.
The system alerts clinicians if there is a high risk of deterioration in seemingly healthy patients and enables clinical conversations to manage the scenario. “We have fewer false positives than other systems, but our sensitivity remains,” says Alex, referring to a published paper. “Clearly this improves mortality rates, reduces admissions to intensive care and impacts the bottom line.” In addition, clinicians are able to determine whether patients could be discharged on an expedited pathway.
As much as anything, the company’s aim is to enable doctors to be part of the digital health revolution, involving clinicians and getting their buy in for technology that will make their job easier and patient outcomes better.
“Everything we do is underpinned by this amazing platform called RippleDown, which was invented in Sydney University 20 years ago and has been mainly used in pathology,” he says. “Now we’re widening it out, such as the deterioration example. It’s a rule builder, but it’s really simple. A clinician can build rules, write it in plain English, and it understands and matches it. With other systems, the clinicians have to send the rule to be hard coded in. In addition, it allows infinite rows and will immediately test that rule against any other rule you’ve written previously, even if it were three years ago, and it can have an infinite number of rules.”
So clinicians like the system – and that’s always a good start!