Healthcare executives are putting money into AI and predictive analytics to help them move away from data saturation and toward insights at scale. However, each business encounters its own set of needs and problems along the way. Successful informatics leaders use a platform strategy, prioritise clinical expert collaborations, and break down organisational goals into manageable chunks.
No two digital transitions in healthcare are the same, and health informatics leaders are emphasising this point. For example, at the HIMSS 2022 conference, we heard from some healthcare executives who have teams of experts applying machine learning to 35 years of healthcare data to generate clinical and operational insights, as well as others who are still dealing with a fragmented IT system that requires them to spend the majority of their budget just keeping the lights on.
The benefits of machine-generated insights are evident, no matter where you are on your transformation journey. We looked at the difference between data and insights, and how AI and predictive analytics may help today’s healthcare businesses deliver insights at scale, from newfound operational efficiencies to more accurate clinical judgments and more successful treatment routes.
Many informatics leaders agree on the importance of translating data into insights. More than three quarters of informatics experts believe predictive analytics can have a positive impact on the cost of care (78 percent) and overall staff experience (76 percent), two critical components of the Quadruple Aim, according to our 2022 Future Health Index report, which surveyed nearly 3,000 healthcare leaders from 15 countries. Here’s a link to a new news centre story about informatics findings from the Future Health Index report.
So, while shifting from data to insights is a terrific approach to progress the Quadruple Aim, how can you get insights at scale, when insight production is built into the very fabric of your organization’s workflow?
The path begins by determining where your health system falls on the digital maturity curve, a set of global guidelines developed by HIMSS to help healthcare organisations benchmark and improve their digital transformation progress. In areas like continuity of care, analytics, diagnostic imaging, and electronic health records, there are seven maturity models to choose from (EHRs). Each model has eight stages, starting with minimal digital infrastructure and progressing to multi-vendor, networked, and dynamic capabilities.
Even healthcare professionals who are just starting out on this path may encounter hurdles, such as accessing, organising, and sharing data, as well as worries about data privacy.Overcoming these challenges to move smoothly from data to actionable insights at scale takes four steps:
1. Invest in interoperability – a fundamental (and very human) enabler
Ask five people to define interoperability and you’re likely to get five very different answers. So, let’s be clear: we will define interoperability here as the ability of software systems to exchange and make use of data. Though, the real power of interoperability is not in the technology itself. It’s in the human benefit it delivers by unlocking insights for patients, providers, and health systems.
The reality today is that healthcare data lives in a huge variety of locations, ranging from isolated server racks to cloud platforms. The siloed nature of data repositories is widely recognized, with almost two thirds (57%) of informatics leaders surveyed in our 2022 Future Health Index thinking that data silos hinder their ability to use data effectively.
To realize the potential of insights at scale, data – and the actionable insights it holds – needs to be available in formats that can be shared effortlessly, and above all securely, between points of care, whether within or between hospital systems, to the home, or even on the move. For example, your hospital department head could integrate operational data from one or more EHRs to generate analytics and real-time workflow intelligence that improve operations effectiveness. Leveraging open APIs and approved standards like IHE-HL7 can help facilitate data exchange across multiple sources across the continuum of care, so that healthcare providers can deliver the right care at the right time with minimal friction.
In other words, by breaking down siloed data and aggregating that data into actionable patient insights, doors to further innovation can be opened to achieve greater returns, such as improving clinical confidence and patient outcomes – other vital components of the Quadruple Aim.
With interoperability in place, the next step is to be able to deploy innovation at scale.
2. Power your system of action through the cloud
At the heart of every advanced insight strategy is a secure cloud-based platform that can serve as an agile toolkit to power your system of action by scaling machine learning quickly and easily throughout your organization and spurring further innovation. By 2024, research suggests that healthcare providers that have adopted a digital health platform will outpace competition and partners by 80% in the speed of digital transformation and new feature implementation.
Alongside secure, scalable storage, computing power, and AI toolkits, the cloud also enables dynamic new business models like software as a service. These subscription-based solutions for healthcare providers deliver more predictable, controllable payments and a different level of service upkeep. At Philips, we’ve taken a platform approach to informatics innovation to deliver data-driven actionable insights that advance precision care, support patient-centric, connected care and enable transitions of care. Our HealthSuite platform combines the core strengths of industry-leading cloud hosting and security, with deep clinical and operational knowledge. To date, more than 100 types of medical devices have been integrated into HealthSuite, with over 145 billion clinical images securely archived on the cloud platform.
3. Prioritize strategic partnerships grounded in clinical expertise
No care provider is an island. Organizations need to prioritize partnerships in order to successfully implement digital health technologies and climb the digital maturity curve. We know from the 2022 Future Health Index report that healthcare leaders are keen to partner with health technology companies – especially those who cover a broad range of areas including strategic vision, specialized healthcare management consulting services, guidance for data analysis, access to innovative technology, and flexible payment models.
They’re also keen to learn from each other. For example, late adopters of predictive analytics have expressed a desire to set up ‘mentorships’ with early adopters. Co-creating with an expert health technology company can enable data scientists, software developers, and clinicians to build a model that fits each provider’s unique needs and help ease the adoption challenges of deploying AI in practice.
4. Break your goals down into discrete, achievable, and measurable steps
For any leader starting out on the journey to insights at scale, the fourth crucial enabler is to begin with discrete, achievable goals in mind.
As we discussed in my previous blogpost, a good place to start is operational forecasting. For example, a Chief Nursing Officer (CNO) not only wants to know exactly what staffing levels are like today, but more importantly they want to know what staffing needs will be – say – tomorrow, or next weekend. This kind of care traffic control system can help a CNO predict potential patient care bottlenecks in a hospital, and anticipate patient flow capacity and match resources and staffing to patient care needs.
It’s vital to measure the impact of your insights too. Let’s take another example in imaging, where health systems in many countries face a shortage of radiology technicians. Vendor-agnostic imaging solutions can enable technicians in a central command center to seamlessly connect with technologists at scan locations across their organization and support them to ensure optimal patient imaging, reducing cost of recalls, and improving patient satisfaction. One such implementation in the US has enabled one health system to increase its imaging staff capacity and retain more than $350K of MRI procedure revenue that would have otherwise been lost to another provider or deferred, together with a saving in personnel travel cost of over $60K per year.