Population health management real time state-of-health analysis
To lower health costs, physician networks and medical homes must employ a closed loop population management program that focus on patient SOH stratification, chronic disease management, care coordination and incentive management. This approach will enable them to consistently reduce ER and inpatient admissions, which are the greatest expenditures in health care today.
Published on: Mar 4, 2016
Transcripts - Population health management real time state-of-health analysis
Page 1 www.PSCISolutions.com
Population Health Management:
Real-Time State-of-Health Analysis Solutions
Leveraging EMR Data to design and execute provider-driven care management programs.
The Affordable Care Act opens the door to a wealth of opportunity for hospitals and physician
groups. They are beginning to adapt to the new pay-for performance and bundled payment
systems, and develop population-based care management programs. While the goal of this piece
of legislation is to hold hospitals and physicians jointly responsible for quality and cost of care, the
new payment models span the entire care continuum including primary care physicians, specialists,
hospitals, and other medical professionals. The biggest winners will be the ones who can improve
quality of care while driving down costs. Those that focus first on preventive care for top chronic
illnesses will be the first to reach the finish line.
Innovative healthcare providers take the lead by
developing coordinated care systems that embody
the core principles of preventive care: Patient-
Centric Medical Homes (PCMHs) and Accountable
Care Organizations (ACOs). Physician networks
are adopting the PCMH model, relying on primary
care physicians (PCPs) and care coordinators as
the central hub for care, and looking to specialists
when necessary. Medical homes deliver
preventive care to the entire spectrum of patients,
from healthy to chronic, with the goal of avoiding
admissions to acute care facilities.
WHY CHRONIC CONDITIONS FIRST?
Chronic diseases account for the majority of acute
care costs (in-patient, out-patient, and ER).
Controlling acute care admissions for chronic disease is essential to control healthcare costs.
Therefore, the effort to minimize healthcare costs must begin with managing top chronic conditions.
According to the CDC, “Chronic diseases are the leading cause of death and disability in the US,”
and Healthcare Cost Monitor underscores this fact revealing that, “Seventy-six percent of Medicare
"Chronic disease and health promotion." Centers for Disease Control and Prevention. Center for Disease
Control, 2010. Web. 26 Feb 2012. http://www.cdc.gov/chronicdisease/overview/index.htm.
CDC on Chronic Diseases
7 out of 10 deaths among Americans each year
are from chronic diseases. Heart disease, cancer
and stroke account for more than 50% of all
deaths each year.
In 2005, 133 million Americans – almost 1 out of
every 2 adults – had at least one chronic illness.
Obesity has become a major health concern. 1 in
every 3 adults is obese and almost 1 in 5 youth
between the ages of 6 and 19 is obese (BMI ≥
95th percentile of the CDC growth chart).
Diabetes continues to be the leading cause of
kidney failure, non-traumatic lower-extremity
amputations, and blindness among adults, aged
Page 2 www.PSCISolutions.com
spending is on patients with five or more chronic diseases.”
The Agency of Healthcare Research
and Quality also emphasizes the high cost of chronic conditions.
Regardless of which payment model becomes
predominant (shared savings, bundled payment, or
ACO), in order to bend the health care cost curve,
health care providers must focus on preventive
care for chronic care patients. Provider
organizations will need an innovative approach to
redesign care processes, with a focus on keeping
chronic care patients healthy and out of ERs and
YESTERDAY | CLAIMS-BASED PREDICTIVE MODELS
For years, healthcare insurance companies (payers) have mined claims data for chronic patients
and have built predictive models to identify high-risk patients. Armed with historical reports, case
managers designed intervention programs that were meant to prevent complications among
chronic patients and reduce ER visits and hospitalizations.
While this approach has seen some success,
limitations far outweigh merits. Data used by
payers to flag high risk patients is historical claims
data — primarily costs, admissions, and diagnoses.
Because this view is retrospective and heavily
biased toward cost, patients with past high acute
care costs are flagged as “risky”, regardless of their
current state of health. Furthermore, regression
and time series risk models are typically updated
Most physicians are highly skeptical of claims-
based predictive models because they have no
clinical basis, and give no consideration to an
individual’s current state of health. Moreover, there
is a complete lack of causation, “Why is a patient
considered high-risk? What are the clinical reasons for the score? How do we lower the patient’s
risk score? How does the score measure the effectiveness of my care management program?”
Swartz, Kimberly. "Projected Cost of Chronic Diseases." Health Care Cost Monitor. Health Care Cost Monitor,
n.d. Web. 26 Feb 2012. http://healthcarecostmonitor.thehastingscenter.org/kimberlyswartz/projected-costs-of-
Stanton, M. W.. "The High Concentration of US HealthCare Expenditures." Agency for Healthcare Research and Quality.
AHRQ, 2006. Web. 26 Feb 2012. http://www.ahrq.gov/research/ria19/expendria.htm.
Health Care Cost Monitor on Chronic Disease
Seventy-six percent of Medicare spending is on
patients with five or more chronic diseases.
Currently 10% of health care dollars are spent on
overall direct costs related to diabetes, amounting
to $92 billion a year (1.5 times the amount spent
on stroke or heart disease). The Centers for
Disease Control and Prevention predicts that
spending on diabetes care will reach $192 billion
According to the Milken institute, overall cost of
heart disease is predicted to reach $186 billion in
AHRQ on Cost of Chronic Conditions
The 15 most expensive health conditions account
for 44 percent of total health care expenses.
Patients with multiple chronic conditions cost up to
seven times as much as patients with only one
Page 3 www.PSCISolutions.com
These models lack a correlation to clinical information. For example, a physician will acknowledge
a high risk score if there is evidence that the patient has a high BMI, and the HbA1C has been
consistently high over the past year and is trending higher. The score becomes even more credible
when there is evidence of ER admissions or acute care inpatient admissions.
Unfortunately, an individual’s current state of health has no bearing on his or her claims-based risk
score. Claims-based risk scores are created with regression analysis at a population level to
predict scores at the patient level. Individual scores are relative to the population, therefore could
change as the population changes, even with no change in the individual’s state of health.
Not only are today’s calculations unsuitable for determining a patient’s true risk, they provide no
insight on how an individual’s score improves or deteriorates after each clinical visit. Information
lags so far behind; physicians are given no insight to actively manage ongoing care. Claims-based
risk scores are not actionable – they provide no insight for care at the provider level.
Claims-based risk scores are also deficient because they do not adequately represent the
population. Reports provided by payers are used primarily by case managers, who in most cases
work for a payer. Physicians reject these reports as a basis for their own effectiveness in
managing patients, because they are only a subset of their total population. Furthermore, payer
reports are not typically useful for evidence-based care, to identify and implement clinical best
practices. Finally, they are inadequate for measuring physician performance to design incentive
In order to use payer reports across an entire population, a provider would first need to normalize
multiple payers’ risk scoring systems, then aggregate the information. Because each payer has a
unique methodology, there is little chance that the resulting information would be accurate or
meaningful for developing care management programs.
Considering today’s approach to developing care management programs and understanding
physician effectiveness, it’s important to remember that CMS does not provide patient risk scores.
The fact that Medicare patients account for the majority of chronic patients and populations, other
payers’ risk reports incorporate only a small fraction all chronic patients. Therefore, the impact of
using individual (or even combined) claims-based payer risk reports is minimal in any effort to bend
the overall patient population health care cost curve at the provider level.
FURTHER CONSIDERATIONS FOR A NEW APPROACH
Current thinking and efforts create a disproportionate focus on existing chronic patients. The
intervention approach is designed specifically for this group, while wellness programs reflect only
the hope that the healthy population will remain so. Because today’s healthy patients are largely
ignored, yet will become tomorrow’s chronic patients, this approach is deeply flawed. If providers
delay uncovering and examining causes until a chronic diagnosis emerges, there is no opportunity
to avoid a chronic scenario. A better approach is to monitor all patients, healthy and chronic, for
risk of hospitalizations. Unfortunately, current claims-based predictive risk models allow no room
for this approach.
Page 4 www.PSCISolutions.com
Claims-based risk models create a grave conflict for today’s physicians. To realize bonuses, they
must choose cost of care over effective care. To make matters worse, incentives do not reward
every health care professional that has an impact on patient health. Conversely, payers strive to
minimize bonuses to physicians and networks. Physicians perceive that payers have an “upper
hand” and can deny bonuses as models change, and assert that costs were higher than
“reasonable” against the statistical model. As a result, there is inherent conflict between physicians
Progressive medical groups do not use claims-based patient risk reports created by payers to
develop care management programs. And, until today, there has been a stark absence of credible
decision support hindering proactive care management. As a result, health professionals have not
had the ability to focus on population state of health as a means to reduce ER and hospital
VITAL PROGRESS | CLINICAL MODELS FOR POPULATION MANAGEMENT
Today, most large physician groups and medical homes already use at least a basic EMR system.
CMS predicts that by 2014, more than fifty percent of all eligible medical professionals in the U.S.
will use EMR. According to Frost & Sullivan, the ambulatory EMR market will explode to $3 billion
by 2013. This is a transformational shift, because for the first time in history, clinical information is
digitally available in real time, with reasonable availability of laboratory results and patient vital
In this age of EMR, the healthcare industry is proclaiming a new wave of decision support for
primary and acute care, leveraging data from EMR applications. The new generation of primary
care management solutions delivers real-time meaningful use clinical data from EMR records.
These systems use patient medical records to measure state of health, and evaluate the
effectiveness of care programs and evidence-based medicine. Real-time clinical data from EMR
records is also being used to create sustainable, repeatable programs to reduce the number of
high-risk patients, and design individualized care management programs. Using current clinical
data for analysis rather than historical claim data means that health care providers create programs
that are meaningful and effective for their specific population.
The new care management decision support systems use actual clinical data, and there is little or
no analysis or interpretation required by the physician. As a result, a care coordinator can take
ownership of care management, so that primary physicians can focus on delivering patient care. In
light of predictions for the short supply of doctors over the next few years, this is good news for
patients and providers alike.
Page 5 www.PSCISolutions.com
CLOSED-LOOP CARE MANAGEMENT PROGRAMS
Using real-time clinical data from EMR records,
health care providers now have the capacity to
design a closed-loop population care management
program (Figure 1). A well-designed program
delivers primary care to drive higher quality, reduce
costs, and deliver greater value in health care.
The very foundation of the well-designed program is
population state of health stratification, the ability to
categorize patients into high (red), moderate
(yellow), and low (green) risk groups by chronic condition.
Population stratification makes it possible to design customized programs for high-risk patients,
execute and monitor programs, and measure the performance of clinical teams for incentive
Population State of Health (SOH) Stratification
State of health stratification provides actionable and measurable information about actual health
status at the population and patient levels, with visibility of controllable and non-controllable factors.
An SOH model takes into consideration every patient’s age, gender, ethnicity, family history, all
clinical factors (like BMI, lipid panel, blood HM, PFTs) and co-morbidities, and delivers an accurate
SOH score for every encounter and for the entire population (score ranges 0 to 100). A low score
indicates excellent health, and as the number increases so does the likelihood of complication(s)
and hospitalization within 12 to 18 months. SOH is a “risk predictor”.
However, it is also an indicator of the quality of care delivered. If the score trends down, the quality
of care is good, because health is improving. In this sense, the trend of the SOH score is a
measure for quality of primary care.
While payers have their own calculation and definition for “risk”, the remainder of this article uses
the terms “Risk” and “State of Health” interchangeably.
Origins of State of Health (SOH) Models
Nationally accepted clinical models are the basis for state of health models. In some cases, when
the data did not contain all the parameters required to compute SOH scores, assumptions and
approximations were considered and validated with physicians to ensure the integrity of the
models. The SOH models were then validated against historical data.
Figure 1 - Closed-Loop Care Management Program
Page 6 www.PSCISolutions.com
SOH scores are calculated at the patient level and rolled up to a population level (Figure 2). In this
example, each row corresponds to a physician's patient population. It shows the patient count, the
number of office visits (encounter) and the average population SOH score for each chronic
disease. “Red” signifies “high risk” scores. Physicians and care coordinators use the easy-to-
digest visual information to focus on high risk populations, and drill down to individual patients to
understand factors that contribute to scores.
Figure 2 Population SOH (Risk) Stratification by Physician.
Focus on prevention and screening; monitor compliance for chronic conditions.
This approach allows health care providers to design meaningful preventive care programs for the
exact population, and create individualized programs for specific patients.
Chronic Disease Management
Patients who comply with prescribed care programs are typically more successful in managing
chronic conditions. This is where care coordinators play an important role. Leveraging state of
health scores, care coordinators pinpoint high risk patients by chronic condition, and best evidence
guidelines become the basis for customized care management programs. The care management
program is integrated with the care management execution system that includes patient
scheduling, outbound call centers, home visits, patient portals and emails. While the disease
management program identifies needs, the execution system promotes compliance with
treatments, medications, scheduling laboratory tests, attending educational counseling sessions,
and other prescribed activities.
Page 7 www.PSCISolutions.com
Monitoring gaps in care established by evidence-based care, patients’ SOH trends, and underlying
clinical drivers over time, care coordinators can identify patients that need their attention.
Physicians who improved the state of health for their population (i.e. lower the score) over a one to
three year period established and used better clinical protocols (i.e. best practice care management
programs). In one instance, one physician’s CHF population risk increased to 55%, while another’s
dropped to 5% (Figure 3). Analyzing SOH population trend by physician population, the team of
physicians identified the most effective clinical protocols for the patient population and standardized
around best evidence care. The physician team also used SOH scores as a measure of quality of
primary care, resource utilization, costs, and patient experience to establish best evidence care
protocols, to lower cost and improve patient experience. (Figure 4).
Figure 3 - Effectiveness of two physician CHF populations.
Use best practices within the risk group for evidence-based care coordination: medicines, treatment levels,
frequency of visits; by risk group.
Page 8 www.PSCISolutions.com
Figure 4 - Population Chart / Cost-Quality Grid, High-Medium-Low Risk
Population performance: Map patients on quality and total cost across the continuum-of-care
(ambulatory and acute). Identify optimal preventive care levels to minimize lifecycle cost over a time period
by chronic condition.
It is not enough to simply design and launch new programs. If financial incentives for health care
professionals are not aligned with performance, success may be temporary and hard to sustain.
Effective incentive programs distinctly drive higher quality and reduce costs for greater value in
Align team incentives with population quality and cost performance targets (physicians
and care coordinators)
Establish and share best evidence practices by chronic condition
Encourage teamwork to lower healthcare costs
Illustrate accurate physician and clinical coordinator population performance, and the
impact to incentives
Incentive programs reward care teams for reducing population risk scores, improving patient
satisfaction scores, and reducing overall patient costs. Continuum of care dashboards (ambulatory
and acute) are useful in designing incentive programs and illustrate risk-cost-quality details for each
patient (Figure 5).
Page 9 www.PSCISolutions.com
Figure 5 - Continuum of Care Analysis by Patient, Preventive Care Impact on Acute Care Costs
Monitor how much total inpatient and outpatient care (cost and quality) is being provided to the risk panel;
identify patient outliers.
Patient SOH scores can be rolled up to population averages. For example, one incentive program
dashboard maps physician/care coordinator teams on a cost-quality grid (Figure 6). In this case,
the quality metric captures population SOH, ACO quality measures, and patient satisfaction scores.
The intersection of the crosshairs is the target for quality and cost for the specific patient
population. Each bubble corresponds to a specific physician- care coordinator team, and the size
of the bubble illustrates the size of the population they manage. The distance of each bubble from
the crosshair indicates the positive or negative variance from the target and is proportional to each
team’s bonus or penalty.
Figure 6 – Physician value index used for incentive management for care teams.
Report shared savings by plan by physician on a periodic basis and show the impact of actions on their
Page 10 www.PSCISolutions.com
Results | Validating the SOH Model APPROACH
Using SOH models as a surrogate for primary care quality and the indicator of possible
hospitalization is a new concept and will become the contemporary paradigm for chronic disease
management. Therefore, it was important to understand how effectively the model and scores
could predict hospitalizations against historical patient population data. To validate the models,
researchers compared the new SOH model against that of a leading claims-based risk model (the
The insurance payer used claims data (patient age, gender, ethnicity, previous ER, IP admissions,
costs, diagnosis and other claims files data). They calculated a risk score, a number between 0
For the SOH model, researchers used real-time clinical data (patient age, gender, ethnicity, vital
signs, lab results and treatment medications). The SOH model did not include past ER or IP
admissions data. The SOH model established a risk score between 0 and 100 for diabetes (Table
Total diabetes patients
(type 1 and 2, complicated and uncomplicated)
Time period (2010) 1 year
IP admissions 53
ER visits 95
Table 1 - SOH Validation
Next, researchers calculated a SOH score for each patient using historical data over two years
(2008-2009), and stratified the population based on SOH scores. Researchers compared SOH
scores to actual IP admissions and ER visits.
Figure 7 shows total hospitalized patients as a ratio of the total diabetic patients for that SOH band.
For example, in the SOH band 50-60, 20% of all patients were hospitalized. As the score
increased, the ratio of patients within that band also increased. At very high scores, all patients
were hospitalized. Thus, Figure 7 validates the accuracy and predictive power of the SOH score.
Page 11 www.PSCISolutions.com
Figure 7- Ratio of Hospitalized Patients to Total Diabetic Patients
Next, researchers compared the SOH results to the payer’s claims-based actuarial model. Figure 8
shows the relationship between the payer risk scores and IP admissions. In the 250-500 risk band,
the ratio of admitted patients is higher than the SOH model. Since most patients are in this band,
the predictive power of the payer’s model at low risk scores is diminished. Similarly, at higher risk
scores, the predictive power of the payer’s model is only 50% whereas the researchers’ SOH
model is closer to 100% accurate.
Page 12 www.PSCISolutions.com
Figure 8 - Relationship between the payer risk scores and IP admissions.
Figure 9 shows a similar comparison, SOH bands and ER visits. This comparison and the uniform
curve further substantiates the SOH model as a valid and accurate predictor of ER admissions.
Similarly, Figure 10 shows the payer’s model for ER admissions, which is clearly weak at both low
and high risk bands.
Page 13 www.PSCISolutions.com
Page 14 www.PSCISolutions.com
EVERYONE WORKS SMARTER USING SOH MODELS
State of health models are highly accurate and predictive, and ideally suited for chronic care
population management by chronic condition. Using SOH scores, care coordinators can correctly
identify and focus on high risk patients with a great risk of hospitalization in the short term. Given
the rapid adoption of EMRs among primary care physicians and groups, the data required to build
SOH models is readily available now, and will continue to expand over the next two years.
Healthcare providers can enable continuous improvement using SOH models together with care
management programs. This approach has already been institutionalized in a number of leading
medical homes like Medical Clinic of North Texas (MCNT). Within these organizations, there are a
wide variety of individuals who actively use these models in their daily work, and can be described
Administrators & Management, to quantify the effectiveness of care management
programs, measure productivity, and monitor incentive programs.
Physicians, to define and/or leverage best practices in managing disease, in line with their
desire for evidence-based care. By analyzing SOH scores and understanding drivers,
they have more insight to deliver better care.
Care coordinators, who are primarily interested in identifying high risk patients, to
understand risk factors, develop individual care programs, and monitor patient
Medical Clinic of North Texas (MCNT), a Level 3 Recognition by the National Committee for Quality
Assurance (NCQA) Physician Practice Connections ® - Patient Centered Medical Home™ (PPC-
PCMH) has been a pioneer user of SOH based population management approach.
MCNT demonstrated a stellar FY 2010 performance with Total Medical Cost trend for their
managed population of 2.4% better than market, is a culmination of various quality of care drivers:
Potential avoidable ER visits decreased by 13.3%
OP Diagnostics trended only 1.9% vs. market trend of 9.7%
OP Surgical trended 5.6% vs. market trend of 15%
Utilization of CCD Specialists increased by 18.3% while drugs administered trended 10%
less than market
High tech scans/1000 decreased by 12%
Overall performance index improved in Facility Outpatient (-5%), Other Medical Services
(-6%) and Professional (-1%) relative to the market
An enviable performance considering the challenges the healthcare provider markets are facing
with influx of changes in the market.
Page 15 www.PSCISolutions.com
To lower health costs, physician networks and medical homes must employ a closed loop
population management program that focus on patient SOH stratification, chronic disease
management, care coordination and incentive management. This approach will enable them to
consistently reduce ER and inpatient admissions, which are the greatest expenditures in health
To become masters in their population management programs, they need decision support
systems such as population SOH (risk) stratification and predictive models. With the growth of
EMR systems, claims-based risk analysis will replaced by clinically-driven models that leverage up-
to-the-minute clinical data to accurately determine state of health scores.
SOH scores are more accurate in relation to actual patient risk, and have extremely strong
predictive power. Because they are based on actual clinical data rather than claims history, they
are widely embraced by the physician community. Physicians actively look to SOH models to
understand causes, predict outcomes, and focus on controllable factors to improve patient health.
Since EMR, laboratory and pharmacy data is now widely available, SOH models are easy to build
for most physician practices and medical homes.
Ultimately, today’s physician has real power to bend the healthcare cost curve by focusing on high-
risk chronic patients, designing appropriate care management programs, and helping to keep
patients out of hospitals.
PSCI serves healthcare providers with decision support to analyze PERFORMANCE, identify
improvement STRATEGIES, stimulate CHANGE, and monitor IMPROVEMENT across DRGs, with
process- and outcome-centric intelligence, predictability and transparency in cost-of-care and
quality-of-care. We place powerful analysis capabilities at the fingertips of every hospital executive,
at a fraction of the cost of those historically available only to a few large HMO giants. Hospital
executives gain insight within days, not months or years, to meet Triple Aim objectives (cost,
outcomes, patient satisfaction), meaningful use requirements, and financial goals. PSCI transforms
healthcare CIOs from "data and technology officers" to "intelligence and insight officers" and helps
them avoid the hassle and cost of delay associated with multi-year electronic health record and
ICD-10 projects. PSCIs closed-loop decision support solutions break down silos across hospital
systems to deliver actionable intelligence and what-if predictive modeling for improving overall
hospital financial performance. For more information, please call (469) 261-4840, or visit
© 2012 PSCI, Inc. All rights reserved.