Precision at Scale: Step 3 of the Population Health Sycle™ – Segment

May 11, 2026

This week, we move into the Segment step of the Population Health Sycle™. At this stage, you’ve gathered a wealth of data during the Survey step. Now, the challenge is to transform that raw information into a roadmap.

The ultimate goal of the Segment step is simple yet ambitious: achieving precision medicine at a population health scale.

The Data Foundation: Moving Beyond the Basics

To segment effectively, we must look past simple diagnosis codes. A strong segmentation model requires a multi-dimensional view of the individual. As we discussed in the "Survey" phase, your model should integrate:

• Clinical & Utilization Data: Chronic conditions and historical ER/hospital usage.

• Health Literacy: Does the individual understand their condition and how to navigate the system?

• Readiness to Change: Where do they sit on the spectrum of behavioral change?

• SDoH Data: Social determinants like housing stability, transportation, and food security.

By using these diverse data points, you move beyond the standard, basic "Low, Moderate, High" risk model and begin to identify the actual drivers behind the numbers.

Identifying the "Why": A Case for Targeted Intervention

In the Segmentation step, we can ask things like: Do we move frequent ER utilizers to a higher level than their clinical condition alone might justify?

In many cases, the answer is yes. We often see ER utilization as one of the most solvable challenges in population health if the "why" can be determined. It’s one of the areas to go after early by assigning them to the correct Segment.

Real World Example: In one of our programs, we identified an individual with over 180 ER visits in a single year. While their clinical needs were real, the driver was social—they were living on the street. It took a substantial effort to locate them (they were living under a bridge), but by segmenting them correctly, having an on-the-ground outreach program to find them,  and providing the right social and clinical support, we transitioned them to a physician clinic. The result was a dramatic improvement in their Quality of Life and a massive reduction in unnecessary costs.

The Power of Predictive Personas

While high-utilizers require immediate attention, your program must also not ignore the Rising Risk—those individuals most likely to be hospitalized in the next six months, or those who are most likely to see significant, preventable reductions in health. Segmenting them into a bucket that gets focused on in the Solve phase is critical.

Today’s AI modeling tools have an ever-increasing ability to identify patterns that traditional analysis might miss. By identifying these individuals with unique and impactful issues and creating "Personas" of like individuals based on this data, we can apply standard interventions earlier and with much higher specificity.

The Bridge to Engagement

Finally, segmentation is the prerequisite for our next step: Sell.

It is vital to understand an individual’s willingness to enroll in a program during the Segment phase. A high-risk individual who is likely to remain high-risk without significant intervention requires a specific strategy to get them enrolled and stabilized. While this requires more resources upfront, the outcomes—measured in clinical success, Quality of Life, and cost containment—are substantial.

Is your organization ready to move beyond "one-size-fits-all" healthcare? Effective segmentation is the difference between a generic outreach list and a high-impact clinical strategy. If you’re looking to refine your risk models, better identify your "rising risk" populations, or integrate SDoH data into an actionable framework, I’d love to help you apply the Population Health Sycle™ to your specific goals. Reach out today to start a conversation about how Accountable Health can drive better outcomes and smarter resource allocation together.

Next Week: We’ll discuss how to take these identified segments and move into the Sell phase to drive actual enrollment and engagement.

The content and writing of this post were created by Fred Goldstein, and the format and images were assisted by AI.