Make this article social: #NHITweek | #PopHealthIT | #Health2Con | #InvisiblePatient
I outlined the four main components of a good population health program last year, and these continue to hold true today:
- Identification and stratification of risk within a discrete population.
- Dissemination of information to physicians, care coordinators or others designated to contact patients and arrange follow up.
- Appropriate follow up to further understand the risks for individual patients, identify gaps in care and design a care plan to help the patient improve his/her health status.
- Ongoing care individualized to each patient’s need. That might be coaching, medication reminders, telehealth visits, remote monitoring or other strategies customized to each person’s condition and socio-economic environment.
But what’s still standing in the way of successful implementation and execution of the above components? One of the biggest barriers to effective population health improvement remains the friction in the flow of information between health plans, hospitals and health systems, and physicians. An effective program requires that all four of these components are not only in place, but working together seamlessly, and if there’s a break anywhere in the chain, you lose the opportunity to improve patients’ health.
It’s evident that we have not solved this friction challenge within the last year. However, do I think we can make progress by next year? I do.
Data plays a key role is addressing this problem. The truth is, if healthcare had a magic 8 ball, it would likely be data. One of the biggest advancements I’ve seen over the past 365 days is the emphasis on artificial intelligence (AI) and machine learning, and how we can use these developments to add meaning to our data. In fact, global market intelligence firm IDC predicts that 30 percent of providers will be running cognitive analytics against patient data to personalize treatments by as early as next year.
But beyond personalizing treatments, what if predictive analytics could help us identify the unknown risk within a population, allowing us to detect individuals who are unsuspectingly on the brink of catastrophic health problems?
I believe it can. One of the most profound uses of healthcare data is to define the “Invisible Patient.” These are the people who look healthy on paper, until they (seemingly) suddenly hit a tipping point – perhaps cardiac arrest – and then continue to suffer one adverse health ailment after the next.
These are the pockets of our population that are invisible to healthcare providers. They’re not the folks who already have a chronic disease diagnosis and are being actively monitored with intervention plans in place to mitigate future risk. Rather, they are the undiagnosed. However, with the right medical intervention at the earliest opportunity, their healthcare journey can take a drastically different more positive course.
But prior to intervention comes identification, and in order to identify these unknown individuals, healthcare providers need help sifting through the knowledge from millions of medical journals and patient studies, in combination with individual records, to identify the invisible patients within a population. Advancing technology, like AI and machine learning, is the help these providers need.
By using this data and technology, we’ll be able to not only give physicians the information they need to create intervention plans for their invisible patients, but we’ll also help payers and Accountable Care Organizations (ACOs) develop the best plan for their population’s needs and more cost effectively manage and support these populations.
It’s truly a matter of looking at the road ahead, rather than the retrospective of the rear-view mirror. This change in thinking will finally allow us to begin to deliver an effective population health management solution.
This week, I’ll be at Health 2.0 discussing the latest innovations, topics and trends in health technology. There’s no better place to be for National Health IT Week, as I can chat with others in the healthcare community who all share the same end goals – goals that can be realized by identifying and preventing risk before it manifests.
At the show, there will also be an #InvisiblePatient social media campaign – encouraging both show attendees and others following via social media – to show their support of the Invisible Patient by tagging a selfie or sharing a related story with the #InvisiblePatient hashtag on Twitter.
If you’ll be at the show, let me know (@drnic1) – I’d love to continue the conversation live.