Population health management (PHM) is a data-driven approach that ACOs use to improve outcomes and lower total cost of care for a defined population by identifying risk before it becomes utilization, coordinating care across settings, and addressing the clinical and social factors that drive spending.
For accountable care organizations (ACOs), PHM is not a philosophy. It is the operating model that determines whether you earn shared savings or owe losses.
The evidence is now unambiguous. In Performance Year 2024, Medicare Shared Savings Program (MSSP) ACOs saved Medicare $2.5 billion relative to benchmarks and earned $4.1 billion in performance payments — the highest results since the program began in 2012. But the aggregate number hides the more useful finding. ACOs composed predominantly of primary care clinicians generated $403 in net per capita savings versus $224 for ACOs with fewer primary care clinicians. Nearly double, from the same program, under the same rules.
That gap is not about size, brand, or capital. It is about how close an organization sits to the patients it is accountable for, and how well it converts data about those patients into action. That is population health management. This guide covers what it requires, what the current evidence proves, and why the timeline to build it just got shorter.
Executive Summary
- PHM is the operational foundation for ACO performance under shared savings, quality accountability, and downside risk.
- PY2024 MSSP results show primary-care-dense and physician-led ACOs substantially outperforming their peers, indicating that PHM capability — not organizational scale — predicts financial results.
- Participation has expanded sharply: 511 ACOs in PY2026, up from 476, with 82.8% now in a two-sided risk track.
- Chronic disease remains the core economic problem PHM exists to solve: 90% of the nation’s $5.3 trillion in annual health expenditures goes to people with chronic and mental health conditions.
- The runway is compressing. CMS has finalized changes intended to move more ACOs into two-sided risk sooner.
What Is Population Health Management?
Population health management is the practice of improving health outcomes across a defined group of people — an ACO’s attributed beneficiaries, a health system’s service area, a payer’s membership — rather than responding to patients one encounter at a time.
The distinction matters operationally. Traditional care is reactive and episodic: a patient presents, a clinician treats, the encounter closes. PHM is prospective and continuous. It asks which patients in the population are likely to generate avoidable utilization in the next six to twelve months, and what can be done now to change that trajectory.
That shift requires seeing the whole person. A patient’s health is shaped by far more than what happens in an exam room. The U.S. Department of Health and Human Services estimates that social determinants of health (SDOH) drive 30–55% of health outcomes — housing stability, food access, transportation, social support, economic security. An ACO that can see clinical risk but not social risk is working with half the picture.
Why PHM Matters More in 2026
Three forces have converged, and all of them shorten the time available to build the capability.
Accountable care is now the default, not the pilot
As of January 2026, 14.3 million Medicare beneficiaries receive care coordinated by an ACO, up from 13.7 million in 2025. The Shared Savings Program alone grew to 511 ACOs serving 12.6 million people with Traditional Medicare — a 12.3% increase and the largest population the program has ever covered.
More consequentially, 82.8% of Shared Savings Program ACOs now participate in Level E of the BASIC track or the ENHANCED track, both of which qualify as Advanced Alternative Payment Models. That is the highest share since the program’s inception. The majority of the market is carrying real downside exposure.
The on-ramp to risk is getting shorter
CMS has finalized changes in the CY 2026 Physician Fee Schedule Final Rule intended to increase the number of ACOs in two-sided risk — arrangements where organizations are accountable for losses as well as savings. In the proposed rule, CMS sought to reduce the maximum time an ACO can remain in a one-sided BASIC model from seven performance years to five.
The implication for planning is direct. Whatever grace period an ACO assumed it had to develop analytics, care management, and physician engagement, it should assume less. And the consequences are real: in PY2024, 16 ACOs owed shared losses totaling $20.0 million.
Chronic disease is the cost structure
Chronic conditions are not one category of spending among many. They are the spending.
- Chronic and mental health conditions account for 90% of $5.3 trillion in annual U.S. health expenditures.
- Chronic kidney disease affects more than 37 million adults, and more than 1 in 3 Medicare dollars — $141.1 billion — is spent caring for patients who have it. Nearly 9 in 10 people with CKD do not know they have it.
- Diabetes affects 40.1 million Americans, with another 115.2 million adults living with prediabetes. Total estimated cost in 2022: $640 billion in medical costs and lost productivity.
- Heart disease and stroke cost the health care system $223.2 billion per year, with an additional $191.5 billion in indirect costs.
- Alzheimer’s and other dementias cost an estimated $360 billion in 2024, projected to approach $1 trillion by 2050.
These conditions do not resolve. They are managed — well or poorly — over years. That is precisely the work PHM organizes.
What the PY2024 Results Actually Prove
Rather than a hypothetical case study, look at what actually happened across 476 ACOs in the most recently reconciled performance year. The CMS PY2024 financial and quality results are the most instructive dataset available on what PHM capability is worth.
Financial results
Two things stand out. First, performance improved year over year — $245 in net per capita savings in PY2024 against $207 in PY2023, with gross per capita savings rising from $515 to $651. Second, and more useful strategically: the segments that outperform are the ones structurally closest to the patient. Physician-led organizations beat hospital-led ones by 77%. Primary-care-dense organizations beat primary-care-light ones by 80%.
CMS also found that ACOs achieving shared savings had lower utilization than their benchmark across hospital discharges, emergency department (ED) visits, and skilled nursing facility (SNF) stays. That is the mechanism. Savings are not an accounting outcome; they are the downstream result of avoided utilization, which is the downstream result of PHM working.
Quality results
The clinical improvements are equally concrete. Among ACOs reporting all-payer digital quality measures in PY2024 versus PY2023:
- Patients with poor hemoglobin A1c control fell from 35.18% to 23.52%
- Patients with adequately controlled high blood pressure rose from 69.63% to 73.65%
- Patients screened for depression with a documented follow-up plan rose from 43.70% to 55.36%
ACOs also outperformed comparable physician groups: 53.53% versus 44.42% on depression screening and follow-up, and 71.21% versus 67.82% on controlling high blood pressure.
Those are population-level movements in exactly the measures that PHM programs target. They did not happen by accident.
The Five Capabilities That Make PHM Work
PHM is not a single tool. It is five capabilities that have to function together — a weakness in any one degrades the rest.
1. Data integration
You cannot manage what you cannot see, and most ACOs cannot see the whole picture. Clinical data lives in the electronic health record (EHR) — often several EHRs. Claims live with the payer. Pharmacy fills, lab results, and social risk screening live somewhere else again.
Fragmentation is not merely inconvenient. It produces incomplete risk scores, missed care gaps, and interventions aimed at the wrong patients. Healthcare data aggregation is the precondition for everything downstream.
What integrated PHM data includes:
EHR clinical data · medical and pharmacy claims · laboratory and diagnostic results · health risk assessments · SDOH screening · patient-reported outcomes and remote monitoring data
2. Risk stratification
Once the data is unified, the question becomes: who needs attention, and how much?
Risk stratification segments the population so that intervention intensity matches actual risk. Done well, it prevents both failure modes — under-serving patients heading toward a crisis, and over-serving stable patients who do not need intensive management.
Effective stratification is multi-dimensional, layering:
- Clinical complexity — chronic condition burden, recent admissions, medication adherence
- Utilization patterns — ED visits, specialist use, readmission risk
- Social risk — SDOH screening results, transportation and food access barriers
- Predictive signal — models that forecast future high-cost utilization rather than describing past spend
The critical distinction is prospective versus retrospective. A report telling you who was expensive last year is history. A model telling you who is about to become expensive is a work list.
3. Proactive intervention
Stratification without intervention is just an expensive report.
This is where PHM converts insight into contact: care manager outreach to rising-risk patients, chronic disease self-management programs, medication therapy management, transitional care after discharge, preventive screening campaigns, and remote monitoring for patients with unstable conditions.
The design principle is timing. Intervening before an acute event costs a phone call. Intervening after one costs an admission.
4. Care coordination
For a patient with heart failure and diabetes seeing a cardiologist, an endocrinologist, and a primary care physician, fragmented care is not just frustrating — it is expensive. Duplicated tests, conflicting medication regimens, and unmanaged transitions are among the most reliable sources of avoidable cost in an ACO.
Care coordination closes those gaps through designated coordinators for high-risk patients, shared care plans visible across settings, structured discharge protocols, medication reconciliation at every transition, and bi-directional communication between providers.
Post-acute care deserves particular attention. As we’ve covered in ACO cost drivers, variation in SNF length of stay and site-of-care decisions is one of the largest controllable cost levers most ACOs have.
5. Patient engagement
Every preceding capability depends on the patient’s participation. An identified, stratified, outreach-targeted patient who does not engage produces no savings.
Engagement works when it is specific: education matched to health literacy, medication reminders, health coaching, accessible patient portals, shared decision-making tools, and removal of practical barriers — transportation, cost, scheduling — that turn intent into non-adherence.
Where PHM Programs Break Down
Most failures are not strategic. Organizations understand what PHM requires. They struggle to operationalize it.
Data arrives too late to act on. By the time claims-based reporting surfaces a trend, the behavior driving it happened months ago and the performance year is largely locked. Timeliness is not a nice-to-have; it determines whether insight is actionable at all.
Analytics never reach the workflow. A dashboard that a care manager has to remember to open is not a program. PHM works when risk signals land inside the work people already do.
Physician engagement is treated as communication rather than data. Clinicians respond to credible, specific, provider-level information about their own panels. Generalized appeals to value-based care do not change referral patterns. Quality performance improves when physicians can see their own numbers against their peers.
Measurement sprawls. Organizations track everything and act on nothing. The measures that matter are the ones tied to a decision someone can make this quarter.
What to Measure
A disciplined PHM scorecard covers four domains and stays small enough that leadership actually reviews it:
- Clinical quality — HbA1c control, blood pressure control, preventive screening completion, depression screening with follow-up
- Utilization — inpatient discharges per 1,000, ED visits per 1,000, SNF days, 30-day readmission rate
- Financial — total cost of care, net and gross per capita savings against benchmark, financial impact against shared savings targets
- Equity and access — performance gaps across demographic and social risk subgroups
The Bottom Line
Population health management is the difference between managing a population and reporting on one.
The PY2024 results make the case better than any argument could. The ACOs earning the most are not the largest — they are the ones structurally and operationally closest to their patients, with the data infrastructure to know who needs what, and when. Primary-care-dense ACOs earned $403 per capita against $224 for their peers. That gap is a capability gap.
With 82.8% of Shared Savings Program ACOs now carrying two-sided risk and CMS working to shorten the path to it further, the organizations that build this capability now will be the ones still performing in three years.
Is Your PHM Capability Ready for Downside Risk?
- Can you identify rising-risk patients before they generate avoidable utilization? If your risk stratification is retrospective, you are managing history.
- Do your analytics reach the people who make care decisions? Insight that stops at a dashboard does not change outcomes.
- Can you see cost and quality variation at the provider level? Without it, physician engagement is a conversation instead of a plan.
If any of those answers is “no” or “not sure,” let’s talk. Koan Health delivers a population health analytics platform built for ACOs managing to a benchmark — unifying fragmented data, stratifying risk prospectively, and connecting insight to the care management workflows where performance is actually won. Our clients have achieved more than $800 million in Medicare shared savings.
Book a demo to see Datalyst™.
Sources
- Centers for Medicare & Medicaid Services, Medicare Shared Savings Program ACOs: Updated Performance Year 2024 Financial and Quality Results — September 29, 2025
- Centers for Medicare & Medicaid Services, 2026 Medicare Accountable Care Organization Initiatives Participation Highlights — February 4, 2026
- Centers for Disease Control and Prevention, Fast Facts: Health and Economic Costs of Chronic Conditions — May 26, 2026
- Centers for Medicare & Medicaid Services, National Health Expenditure Fact Sheet — 2024 data
- U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Social Determinants of Health Evidence Review
.svg)