Health IT and Electronic Health Activate your FREE membership today |  Log-in

Community Blog

Apr 2 2014   11:18AM GMT

The challenges and promise of population health management



Posted by: adelvecchio
Accountable Care Organizations, ACOs, EHR, population health management

greg-chittim-mugGuest post by Greg Chittim, senior director, Arcadia Healthcare Solutions

It’s no secret that healthcare is changing. Traditional fee-for-service payment models are going away, and new capitated, quality-based payment models are here to stay. The challenge for providers is to position themselves for success under these new models. Now more than ever, it is critical for healthcare organizations to align their care delivery systems and payment models to support population health management to ensure they are delivering the highest value of proactive care to individual patients and patient populations.

The purpose of population health management is to create a mutually beneficial environment that incentivizes both plans and providers to manage the health of their patients by stratifying the population based on risk, engaging high-risk patients, and proactively improving the health of the overall population. Proactive management of high-risk patients has shown to drastically reduce costs by driving down the number of costly emergency department (ED) visits.

With this transition comes many challenges. New technologies, laws and regulations, lack of expertise, and the inherent difficulties of change management are causing all players supporting the transformation to be spread thin. To be successful in the healthcare industry of tomorrow, change is required today. More and more organizations are beginning to engage with accountable care organizations (ACOs) and other capitated quality-centric contracts, implement tools and technology that support population health, and break ground on initiatives that will better position them for success in the new era of healthcare. The industry is realizing that population health is not a fad — it is the future of healthcare.

At Arcadia Healthcare Solutions, we define population health management as a provider’s ability to answer the following three questions with certainty, and have the data to support those answers:

  1. Who are my patients?
  2. How sick are my patients?
  3. Am I effectively caring for my patients?

Below are steps organizations can take to address these questions.

Step 1: Provider-patient attribution

Maintaining a high degree of precision around defining a patient population has never been as important as it is now. Historically, patients scheduled visits, came in for their appointments, and follow-ups were conducted as necessary. With new payment and delivery models — such as ACOs and patient centered medical homes (PCMH) — provider reimbursements are increasingly dependent on the quality of care delivered and overall risk of their patient population. Providers are now forced to better manage the health of their patient population by proactively engaging their highest risk patients, while appropriately managing the health of the rest of their population. They must also support their better health management practices with high quality data from their EHR. As a result, the ability for health plans or other communities (such as health information exchanges) to successfully attribute patients to their primary care provider (PCP), and all parties having an answer to the question, “Who are my patients?” is absolutely critical.

In theory, this seems like a simple task; assign patients to a PCP, document it, and reconcile as needed — what’s the big deal? But of course, the devil is in the details. With the disconnected nature of healthcare and the numerous parties involved in the attribution process, aligning this data and creating a single source of truth is actually extremely challenging. We recently came across an example of this at an organization who thought they had a good handle on attribution — around 70-80% by their estimation. Upon initial analysis between their EHR and health plan membership files, we discovered that attribution rates were actually hovering around 13%, creating a barrier for future population health and quality improvement initiatives.

Attribution is where all data analytics, quality improvement, and PCMH transformation projects should begin. The most effective way to do this is by integrating clinical and demographic data from EHRs with claims and eligibility data. We do this by tapping into the back end of the EHR, pulling claims data, and loading all data into a centralized data warehouse. Data is then scrubbed and merged utilizing a master patient index, creating a single record for all patients. This rich view of a patient’s activity gives providers the level of detail captured in the EHR, as well as the breadth of data from claims. This allows them to see patient activity across the entire care continuum.

The initial attribution process involves reconciling conflicts between a provider’s perception of attribution (from EHR data) and a plan’s perception (from claims data). This inevitably requires some degree of manual intervention by both payers and providers, but only needs to happen once. A recurring, automated process — requiring far less manual intervention — is implemented after the initial attribution process is complete. We have seen this process improve attribution by upwards of 80% in as few as 90 days, and maintain sustainable attribution rates upwards of 90%.

Once an attribution process is in place, the value of any reporting, quality improvement, PCMH transformation, or overall population health initiative will increase dramatically. Providers will have transparency into their patient population, creating the foundation to begin tackling the challenges that will position them for success in the new era of healthcare.

Step 2: Measurement

Of the patient population you identified, how many of these patients are diabetic? Hypertensive? How many are at risk for becoming hypertensive? Are you certain that you can properly identify all of them?

Being able to answer these questions and having the data to back up your answers is crucial. Doing so is more difficult than you may think. First, it is important to be able to identify patients with chronic or other high-risk conditions so you can proactively manage those conditions. Second, new fee-for-value reimbursement models are directly dependent on the level of risk associated with your patient population. By not identifying high-risk patients, reimbursement dollars are left on the table. Finally, you must be able to prove the scope and beneficial impact of the interventions you deliver.

Traditionally, tools used to stratify the population and calculate risk have been primarily based on claims data. High-risk patients were identified based on the diagnosis codes for chronic or high-risk conditions. Outreach and engagement strategies leveraged this data to proactively manage these conditions. However, claims data does not always tell the whole story. What if the condition was not relevant to the visit or was not properly coded? What if the patient has not been in for a visit in that calendar year? What if the patient is at risk for becoming hypertensive but that risk is only captured in specific lab results or diagnostic values? These are all examples of cases where claims data may not tell the whole story. Each of these situations present missed opportunities to both improve quality of care and maximize risk reimbursement.

So, how do you fix this issue? How do you identify these patients, and how do you ensure that you are maximizing your risk reimbursement? One effective way to do this is by leveraging EHR data. Integrating claims with EHR data gives you a richer view of the patient population. Things EHR data can tell you that claims will often miss include the following things.

Vitals signs — Who are patients that have not yet been diagnosed, but are at high risk for becoming hypertensive?

Medical history — Who are patients that may not have had a visit in the last year, but are considered to be high-risk?

Transitions of care — Have any of your patients recently been to the ED?

These are only a few of the questions an integrated data set can answer, but the value of this information is sizable. This level of transparency allows providers to proactively manage their high-risk patients, as well as manage those on the brink of the ‘high-risk’ category. Internal studies conducted by Arcadia have revealed that millions of dollars in risk reimbursement opportunities have been missed due to the incomplete picture presented by claims data.

Although alternative strategies may exist that give you the transparency you need to measure the health of your population, an integrated, data-driven approach is crucial to doing this effectively at scale.

Step 3: Intervene and track

Are you doing what you need to do to effectively care for your patient population? You may be, but do you have the data to back it up? An integrated, high quality data set gives you the data required to support your answer to this question, and gives you the foundational tools needed to identify opportunities and drive improvement.

Once your patient population has been identified, the health and quality of care has been measured, and the population has been stratified based on risk, providers and care teams can begin driving more effective population health and quality improvement initiatives. Visibility into the specific needs of a patient population allows providers to begin providing preventative care to manage the health of their high-risk patients, and engage patients that are in danger of becoming high-risk. Combined, these tools allow providers to maximize the value of their ambulatory networks by aligning delivery models with fee-for-value payment models, and managing costs by providing proactive care, as opposed to costly ED visits or, reactive care.

Although tackling certain quality improvement issues in this model is obvious, others aren’t as apparent. For this purpose, we built a platform that delivers a library of best practice blueprints that can be deployed to drive quality improvement programs. This solution allows you to track the progress of that intervention, and can attribute the results to the intervention. A good change process is usually iterative, but this tool can take some of the iterative cycles and guesswork out of the process. Ultimately, it is designed to drive rapid, sustainable results from quality improvement initiatives.

As provider organizations begin to transform and adopt population health-based tools and methodologies, the improved level of visibility into the organization and patient population will fuel opportunities to drive sustainable change and improvement. With the shift from fee-for-service to fee-for-value payment models, the topic of population health is getting more and more attention. There is optimism about the effect that this shift will have on the industry, as our clients are already yielding positive outcomes in the early phases of the transformation.

Comment on this Post

Leave a comment:

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to: