Introduction
As the U.S. health care system increasingly prioritizes the value of
services provided to patients and establishes reimbursement models
that incentivize efficiency and quality, our physician training programs
must similarly adapt.
In primary care graduate medical education (GME), physician training
programs have recognized the importance of prioritizing new models
of primary care to ensure that the future primary care workforce
is adequate in supply, skilled at meeting the needs of patients and
populations, and ready to meet the demands of high-value care. These
skills are the “building blocks” of a high-functioning primary care
practice and are outlined in a 2016 AAMC report, High-Functioning
Primary Care Residency Clinics: Building Blocks for Providing Excellent
Care and Training.1
Population health management (PHM) is one building block of high-
functioning, high-value primary care but is inconsistently understood
and operationalized in practice and training settings. As we train the
next generation of physicians and as our delivery system shifts to focus
on managing populations and individuals, a framework for approaching
this work is necessary, particularly in residency training.
On June 25-26, 2018, the AAMC sponsored the meeting “Population
Health Management in Primary Care Residency Training Programs.”
Held in the AAMC’s Washington, D.C., offices, the meeting included
epresentatives of the AAMC, University of California, San Francisco
Center for Excellence in Primary Care (UCSF CEPC), and Centers for
Disease Control and Prevention (CDC) and leaders from seven residency
programs in family medicine, internal medicine, and pediatrics that
were identified as exemplary in key aspects of PHM.
The meeting was funded in part by CDC’s Division of Scientific
Education and Professional Development, Centers for Surveillance,
Epidemiology, and Laboratory Services, through cooperative agreement
5 NU36OE XXXXXXXXXXThe AAMC asked the UCSF CEPC to assist with
meeting planning and facilitation, based in part on their efforts to
investigate high-functioning primary care in GME practice sites,
including PHM. The lessons from these prior visits were captured in two
AAMC reports, one in 2016 and one in 2018.1,2
The purpose of the meeting was to describe best practices in PHM
in primary care residency teaching practices, with an emphasis on
how residents are trained. Intended for residency program directors,
this report summarizes the components of PHM and offers vignettes
describing how several of the seven participating residency programs
TERMINOLOGY
The following terms represent
the consensus of the meeting
participants.
Clinician refers to physicians,
nurse practitioners (NPs), and
physician assistants (PAs) — those
professionals authorized to make
diagnoses, prescribe medications,
and bill for clinical services provided.
Population health management
(PHM) refers to a systematic
approach to ensuring that all
members of a defined population
(e.g., all patients receiving care at a
primary care practice or all patients
eceiving care from their personal
clinician) receive appropriate
preventive, chronic, and transitional
care.3 PHM also helps providers
identify and address health care
inequities among subgroups within
that population and includes
screening and assistance for those
clinic patients facing ba
iers
to optimal health such as food
insecurity, insufficient income,
precarious or unsafe housing,
and domestic or neighborhood
violence. PHM
oadens the
traditional approach of focusing
only on individual patients.
Population health is defined
as the health outcomes of a
group of individuals, including
the distribution of such outcomes
within the group.4 These
groups are often geographically
contiguous populations, such as
nations or communities, but can
also be groups such as employees
Association of
American Medical Colleges3
implemented these components. The vignettes may provide ideas to
esidency practices that wish to improve their PHM competency.
The seven programs described in this report vary significantly in the
practice of PHM. Similarly, programs that wish to develop a PHM
system may do so in different ways, applying either a na
ow, targeted
approach or a more comprehensive approach to adopting the elements
of PHM into their practices. The descriptions provided here can be
applied in a variety of combinations and illustrate how these elements
are operationalized in diverse clinical settings.
Meeting participants identified 10 inte
elated requirements that
comprise a comprehensive PHM system.
Figure 1. The 10 inte
elated requirements of a comprehensive
population health management system.
of a company, ethnic groups,
disabled persons, prisoners, or any
other defined group. Population
health includes the
oader
determinants of the health of
people within a defined population
and addresses health inequities
within subgroups. Population
health goes beyond specific
clinical metrics and considers
what interventions in addition
to medical care are needed to
keep a population healthy (e.g.,
access to jobs, healthy food, safe
environment, reasonable income).
Meeting participants often refe
ed
to PHM as focused on the “small
p” population and noted that it
addresses the needs of clinic and
clinician panels, while “large P”
population health is concerned
with populations not necessarily
associated with one clinic, clinician,
or health system. The distinction
etween PHM and population
health is analogous to the
distinction between health care
and health.
Population
Health
Management
Panel
management
Patient risk-
stratification
Care
management
Addressing
social
determinants
of health
Ensuring
health equity
Complex care
management
Self-
management
support
Data
infrastructure
Te
am
-
as
ed
c
a
e
C
o
m
m
u
nity engagement
Association of
American Medical Colleges4
Foundational Elements
1. Data infrastructure: Data infrastructure is the organizing, tracking, reporting, and making
transparent demographic and clinical data within the electronic health record (EHR) and
specific data registries.
2. Team-based care: Team-based care is the process of creating teams with a variety
of health professions represented, including physicians, nurse practitioners, medical
assistants, pharmacists, social workers, and others.
3. Community engagement: Community engagement requires an understanding of
local community needs and assets and the development of intentional partnerships with
community members near the clinic, including residents, community-based organizations,
and health departments.
Key Activities
4. Panel management: Panel management ensures that all patients in a population (the
entire clinic’s patients or the panel of one clinician or one team) have their routine evidence-
ased preventive and chronic care tasks performed in a timely manner, including recognizing
and addressing inequities. This process includes addressing “care gaps” and actively
engaging patients at risk of poor health outcomes while in the clinic and between visits.
5. Patient risk stratification: Risk stratification is the process by which patients are placed
into subgroups that are determined by the area of focus of the clinic.
6. Care management: Care management for patients with chronic conditions, often
done by nurses or pharmacists, assists patients with behavior change and medication
management.
7. Self-management support: Self-management support helps patients acquire the
knowledge, skills, and confidence to participate actively in the care of their chronic
condition or for preventive health measures.
8. Complex care management: Complex care management is a program to identify high-
needs and high-cost patients and systematically address these patient needs using a team-
ased interprofessional approach.
9. Addressing social determinants of health: Addressing social determinants of health
equires systematic efforts to meet patient’s social needs that may affect their well-being,
as identified by the care team and denoted in the EHR.
10. Ensuring health equity: PHM can be fully realized only if an equity lens is applied to
ensure a reduction of health disparities while improving health outcomes for the patient
population. Addressing health inequities among a population requires stratifying clinical
data by factors such as race/ethnicity, language, gender identity, sexual orientation, ZIP
code, and insurance status to identify any gaps in health outcomes within subgroups of
the patient population.
The rest of this report further describes the foundational elements and key activities of PHM and
concludes with an overview of how to integrate these principles into a primary care residency.
Association of
American Medical Colleges
Teaching Residents Population Health Management
November 2019
5
1. Data Infrastructure
Individual patient care requires data — from the history, physical exam, lab tests, and imaging
tests. Similarly, PHM relies on health systems and teaching clinics’ collecting, aggregating,
organizing, tracking, and reporting demographic and clinical data. Data are critical to tracking
quality measures that primary care practices are expected to collect. Increasingly, health system
eimbursement partly depends on quality performance metrics that in turn require data.
A teaching clinic’s data often derive from the data infrastructure of the health system in which
the clinic is embedded. Usually an information technology (IT) specialist or IT department at the
larger health system builds the data system. For teaching clinics, establishing a close relationship
with the health system’s IT specialist is essential. Smaller health systems, for example Federally
Qualified Health Centers (FQHCs), may have an IT specialist within the clinic. IT specialists are
not clinicians, so a data-savvy clinician within the teaching clinic must work with the IT specialist
to build a data system that serves the clinic’s PHM needs. If a clinic’s health system lacks an
engaged and accessible IT specialist, the teaching clinic’s leadership needs to make the case for
the resources that are needed to hire one.
A key component of a teaching clinic’s data system is the data registry. Registries are lists of
all patients enrolled in a clinic or empaneled to a clinician, with patient-specific clinical, social,
and demographic data. In the past, EHRs lacked a registry function and many practices used
stand-alone registries. Increasingly, EHR systems feature registries populated from patient charts
and laboratory results. Registries may be comprehensive, providing data on multiple chronic
conditions and preventive care. Alternatively, registries may be disease-specific, providing data
on diabetes, depression, or asthma, for example.
Once the registry is built, the IT specialist and data-savvy clinician(s) work together to create
daily, monthly, or quarterly reports for clinicians and staff at the teaching clinic. The reports
can provide data by clinician and team so that individual clinicians and teams can see and
discuss whether their preventive and chronic care metrics are improving and whether there are
disparities among subgroups. The reports may also aggregate the data longitudinally (run charts)
y metric so the clinic can judge its