There are many definitions of Population Health Management (PHM) today, but from a practitioner’s point of view, we adopt the following definition:
A set of clinical, non-clinical and business processes, capabilities and technology to engage providers and patients towards creating a strong partnership to reduce overall cost of healthcare and improve health outcomes for a defined population. For example, a diabetes center may employ a set of medications (clinical), lifestyle treatments (non-clinical), and value based bundled payments (business process) via a care management software (technology) to engage their diabetes doctors and patients. These all help to create intervention programs that reduce cost of said bundles and improve average HBA1C (health outcomes) for all diabetics in northern Illinois (defined population).
PHM has become a buzzword in the industry and appears important. But how does one apply PHM to help their specific problem at hand? Before transforming towards PHM, it is critical to know why you are doing so from a business and organizational point of view. Several business reasons exist:
Desire to take on risk whether it be global or partial capitation. Many health systems, physician groups, and plans are moving towards partial or global capitation. To be sustainable, it becomes necessary to employ PHM to reduce cost while not sacrificing quality and gravitating towards the 90’s HMO mode.
Ability to truly affect value based care beyond the “accidental” wins. The Centers for Medicare & Medicaid Services (CMS) has mandated a shift towards value based payments and the industry is definitely heading that way. But we all know that regression to the mean is a common phenomenon in PHM. Once you reduce the noise of randomness, how do you identify the truly risky patients and treat them in a way that reduces their cost trajectory? A sophisticated PHM solution is needed to do so.
Simply, to survive. Overall, the shift to value based payments is well on its way and fee for service business will soon become obsolete. However, today we are deeply involved in a FFS To remain relevant, providers must be able to adopt value based payments and have capabilities to survive in this new model.
There are several approaches that Providers take to implement a PHM solution:
Integrate PHM as part of the electronic medical record (EMR) solution. We’re all familiar with how challenging several EMR systems can be. Still, it’s easier to work with a system you know even if it isn’t as efficient as possible. Many providers have opted to either use the existing modules of their EMR or customize their EMR to meet the needs of their PHM solution. This often requires not only high capital expenditure investments but also sustained operating expense investments to keep the system up and running. You essentially become an extension of the software firm that provides your EMR and need to maintain it.
Use a parallel system and solution. There are a plethora of PHM solutions out in the market from analytics to care management, reporting solutions, etc. Many providers opt to integrate the data from their EMR solutions into such an off the shelf PHM solution. This often does not grant you all the features you would hope for (or design yourself) but does place the burden of software maintenance on the vendor.
Build it yourself. Some providers prefer to build their own homegrown system of analytics, care management, and overall PHM. Though this trend is becoming less common, there are merits to this approach as you will have full control over the iterative R&D process as you tailor your solution to your population.
Create a hybrid. This is probably the most common approach where a combination of homegrown and vendor led solutions are integrated.
Most providers tend to follow some variation of the approaches #1 and #2 mentioned above. Once an approach is determined, the following critical steps must be taken to realize your transformation:
Start with a clear vision of what you need. Not all aspects of PHM need be done by you. As a provider, your primary objective is to engage your patients in a meaningful partnership, not necessarily crunching numbers and doing statistics. Decide what you must do and what can be outsourced effectively. The vendor landscape is constantly changing, so consistently re-evaluate your approach and dynamically shift if and when appropriate.
Define what success looks like. It’s easy to get lost in a dozen goals. Create a few ultimate key performance metrics that define if you were successful or not. Cost of care per episode, per bundle, per patient, etc. can be good financial metrics. Create a clinical dashboard targeted towards your specialty that is relevant to your providers and to the patients. Providers don’t care about regulatory metrics (unless they tie to dollars) but primarily care about what they are doing day to day and how that affects patient outcomes (ex. how is the HBA1C level for my diabetes patients?).
Prioritize the features you will release in a roll out process. Each feature must be thought through and a business and clinical value must be placed on it to create a relative weighting of each feature set. This will help prioritize the solution so that value can be captured as the solution is rolled out.
Design the solution. It is imperative to understand what you are trying to change and why. PHM has become a buzzword and often theoretical in nature. Journey maps, user stories, whatever your preference in artifacts, this is a critical step that most organizations ignore and end up with less than effective solutions for a hefty cost.
Implement the solution. Whether you use waterfall or agile development processes, release your solution in meaningful phases. Adoption and change management here is key. You may have the best solutions scientifically, but if your user community is not engaged and not ready to adopt, it will not be effective. Again, the success measures should be inclusive of all things that must generate the ultimate results that you are after.
Use the success metrics in between releases to challenge your product. After each release, use the success metrics to challenge why your solution is not working or how it may not sustain the same results. Unlike some traditional implementation processes, this is a step that will require heavy business and clinical analysis and challenge towards the end of a release and not only in the beginning phases.
Rinse and repeat. Just like agile software development (a process via which one iteratively creates software and users get to experience the product throughout the development process), PHM is also an agile process by nature. Learn your population, not only their clinical needs but also their social, emotional and logistical needs so you can tailor your solutions to fit the needs of your customer community.