Population Health Management
The Population Health Sciences Perspective
How it helps us address health needs
Roger T. Ande...
Population Health Management
 Improving the systems and policies that affect
health care quality, access, and outcomes,
u...
Individual-focused Patient Care Policies
 HbA1C < 7
 BP < 130/80
 LDL cholesterol < 100
 Aspirin, ACE, statin use
Individual-focused Patient Care Policies
 e.g., Breast cancer.
Use of adjuvant chemotherapy is defined as appropriate in ...
Population-focused policies
what do they add?
 Goal: Improve Access or use of services:
8.4
12.3
4.6
High education area ...
Population-focused policies
what do they add?
 Goal: Overcome non-medical barriers to maximize health
outcomes
Percent of...
Adherence: Recognition of the Problem
12-Month Persistence by Drug Class
64.9
62.1
60.7
56.5
54.8
50*
83
0 20 40 60 80 100...
Burden of
Treatment
Theme 1
Work
patients
must do to
care for
their
health
Theme 3
Challenges
/ stressors
that
exacerbate
...
Population-focused policies
what do they add?
 Goal: Coordinate care and provide meaningful
integration
Population-focused policies
what do they add?
 Goal: Monitor and address health disparities
69.0
94.8
53.3
7.0
35.1
17.9
...
“RISK” Assessment to Ascertain level
of Cultural Influence
 Resources of patient and family –
 Tangible resources, e.g.,...
Broad view – Geographic
Patterns
Breast cancer in Appalachia
Population Health Management
An Application
Tracking patient outcomes from risk factor
identification through receipt of c...
Stroke Risk Screening Development of
a Sustainable and Effective Program in
the „Stroke Belt‟ of NC
Roger T. Anderson, Ph....
Geography
Smoothed Age-Adjusted Stroke Mortality Rates
by US County
WM 1996
Importance
2001-2005 Stroke Risk Screening Regions
2000-2006 NCSA Stroke Screening Regions
Health care supply
NCSA + Community Hospital Partners
2001-2005 Stroke Risk Screening Regions
2000-2006 NCSA Stroke Screening Regions
Community Stroke Risk Screening
Program
Challenge - Build a community-based stroke risk screening
program to reduce stroke...
Work with Community Partners to:
 Develop brief, sensitive screening tool.
 Provide health education and stroke risk
awa...
Stroke Risk Assessment Tool
 Easy to collect by trained volunteers
 Age
 Height/ Weight (BMI)
 Blood pressure
 Lipids...
Predictors of an Ischemic Stroke Event (the ARIC study)
Variables
Units of
Measurement
Odds Ratios Odds Ratios
Male Gender...
NCSA Central
Site
 Protocol/Manual
 Survey
 Training
 Funding Sources
 Database/Reports
Local Employer
Local Health
C...
NCSA Stroke Risk Screening Program
 Promote stroke risk screening
 Make available standard protocol
 Regional health ce...
Risk factor # with
Risk Factor
% with Risk
Factor
High Cholesterol > 200 3028 36%
High Glucose (non fasting >130) 1921 23%...
Levels of Stroke Risk in Population
# of Risk
factors
Number of
Individuals
% of Total
0 1626 18.95%
1 2,842 33.13%
2 2,40...
Estimate potential effectiveness
Expected Ischemic Stroke Events in Screened Population
(N=8,579)
ARIC Stroke Risk Score
N...
Proportion of persons found with risk who were unaware of this
status, or who did not seek help for this risk factor durin...
Barrier: low access and receipt of care
following screening:
Of those with high risk, what proportion sought or received
c...
Barrier: Brief risk factor communication does not
result in knowledge and awareness
 90 days after screening After Screen...
Participant support intervention
 Randomized design:
 Usual care vs. telephonic coaching intervention
 N- 227, followed...
Follow-up at 3 months
Attention Control Motivational
Intervention
p-value
MD visit after screening date (recorded
at 3 mon...
% with Medication Treatment for
BP and Cholesterol at Follow-up.
66.67
35.19
15.15
23.26
0
10
20
30
40
50
60
70
80
90
100
...
Conclusions
 Low-cost follow-up phone call with counseling
is effective in improving access in a high-risk
community
 Ba...
Overall Summary
Starting point – Define the Problem
 Identify health needs of a community to facilitate
prioritization and strategies to ...
Challenges
 Collect and analyze data on patient/community health
needs
 ‘Emic’ vs ‘etic’ the right balance?
 Select hea...
Thank you
Population health management anderson
of 40

Population health management anderson

Published on: Mar 4, 2016
Source: www.slideshare.net


Transcripts - Population health management anderson

  • 1. Population Health Management The Population Health Sciences Perspective How it helps us address health needs Roger T. Anderson, Ph.D. Professor of Public Health Sciences Chief, Division of Health Services and Behavioral Research Associate Director for Population Sciences, Penn State Hershey Cancer Institute Director, Penn State MPH Program Penn State College of Medicine, Hershey PA
  • 2. Population Health Management  Improving the systems and policies that affect health care quality, access, and outcomes, ultimately improving the health of an entire population.
  • 3. Individual-focused Patient Care Policies  HbA1C < 7  BP < 130/80  LDL cholesterol < 100  Aspirin, ACE, statin use
  • 4. Individual-focused Patient Care Policies  e.g., Breast cancer. Use of adjuvant chemotherapy is defined as appropriate in the following settings:  node negative invasive cancer with tubular or colloid histology that is 3 cm or larger in size  node negative invasive cancer with ductal, lobular, or mixed histology that is larger than 1 cm in size  any node positive (N1) tumor
  • 5. Population-focused policies what do they add?  Goal: Improve Access or use of services: 8.4 12.3 4.6 High education area Medium education area Low education area Percent decrease in mammography due to copays
  • 6. Population-focused policies what do they add?  Goal: Overcome non-medical barriers to maximize health outcomes Percent of patients with diabetes with self-management barriers 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Limiting Food Mindful of Schedule Meal Planning Organize day around Tx Some Hassle Major Hassle
  • 7. Adherence: Recognition of the Problem 12-Month Persistence by Drug Class 64.9 62.1 60.7 56.5 54.8 50* 83 0 20 40 60 80 100 Proportion of Days Covered (%) ACE=angiotensin-converting enzyme. Shrank WH et al. Arch Intern Med. 2006;166:332-337; Partridge AH et al. J Clin Oncol. 2003;21:602-606. DrugClass Oral contraceptives Calcium-channel blockers Angiotensin-receptor blockers Statins ACE inhibitors Adjuvant endocrine therapy for breast cancer Adjuvant endocrine therapy - 48 month data
  • 8. Burden of Treatment Theme 1 Work patients must do to care for their health Theme 3 Challenges / stressors that exacerbate felt burden Learn about conditions & care Medications Medical appointments Monitoring health status Health behaviors Other treatments (non-medication) Enlisting support Routinizing self-care Planning for future Using technology Preserving autonomy / independence Exerting control over providers Theme 2 Problem- focused strategies to facilitate self-care Challenges with taking medication Interpersonal challenges Financial challenges Confusion about medical information Healthcare provider obstacles – individual provider Role & social activity limitations Physical & mental exhaustion of self-care Barriers to self-care Healthcare provider obstacles – system issues Conceptual Model D e m a n d s E n a b l e r s Theme 4 Impacts of burden
  • 9. Population-focused policies what do they add?  Goal: Coordinate care and provide meaningful integration
  • 10. Population-focused policies what do they add?  Goal: Monitor and address health disparities 69.0 94.8 53.3 7.0 35.1 17.9 64.6 88.9 41.9 14.5 40.2 24.1 0 10 20 30 40 50 60 70 80 90 100 Received Dr. Instructions Full Insurance Married Cost Problems Activity Limitation Equipment Needed White Non-white
  • 11. “RISK” Assessment to Ascertain level of Cultural Influence  Resources of patient and family –  Tangible resources, e.g., level of education, SES, social support, past experiences  Identity and Ability to navigate health care services –  Individual circumstances, place of birth, language  Skills available –  Ability to navigate the HCS and cope w/ disease emotionally, physically, socially and spiritually  Knowledge about ethnic group –  Family structure, decision-making, truth-telling, preferences, modes of providing social support
  • 12. Broad view – Geographic Patterns
  • 13. Breast cancer in Appalachia
  • 14. Population Health Management An Application Tracking patient outcomes from risk factor identification through receipt of care
  • 15. Stroke Risk Screening Development of a Sustainable and Effective Program in the „Stroke Belt‟ of NC Roger T. Anderson, Ph.D1., James Toole, M.D2., Rong Tang, M.D.2 Beth Parks3 1 Pennsylvania State University College of Medicine 2 Wake Forest University, Department of Neurology 3North Carolina Stroke Association, Winston-Salem, NC. USA
  • 16. Geography Smoothed Age-Adjusted Stroke Mortality Rates by US County WM 1996
  • 17. Importance 2001-2005 Stroke Risk Screening Regions 2000-2006 NCSA Stroke Screening Regions
  • 18. Health care supply
  • 19. NCSA + Community Hospital Partners 2001-2005 Stroke Risk Screening Regions 2000-2006 NCSA Stroke Screening Regions
  • 20. Community Stroke Risk Screening Program Challenge - Build a community-based stroke risk screening program to reduce stroke risk in the community.  Community lead: North Carolina Stroke Association  Stakeholders: Community Hospitals  Reach 1000’s of high risk persons across all regions of North Carolina.  Sustainable (1999 – ongoing).  Must have low operational cost: use existing outreach resources and volunteer staffing.
  • 21. Work with Community Partners to:  Develop brief, sensitive screening tool.  Provide health education and stroke risk awareness to participants.  Identify patient barriers to access.  Quantify program effectiveness. Operationally Feasible
  • 22. Stroke Risk Assessment Tool  Easy to collect by trained volunteers  Age  Height/ Weight (BMI)  Blood pressure  Lipids (LDL, HDL)  Glucose (non fasting) and patient history of diabetes  Irregular heart beat  Carotid Bruits
  • 23. Predictors of an Ischemic Stroke Event (the ARIC study) Variables Units of Measurement Odds Ratios Odds Ratios Male Gender Binary(1 = Male) 1.63 (1.16,2.29) Age Year 1.07 (1.04,1.11) No Smoking History Binary ( 1= No History) 1.10 (ns) (0.72,1.70) Current Smoker Binary ( 1 = Current) 2.20 (1.45,3.32) Diabetes Binary 3.78 (2.61,5.46) High Blood Pressure Binary ( 1 if > 140/90 mmHG) 2.64 (1.87,3.71) High Cholesterol Binary (1 if >200 mg/dL) 2.02 (1.23, 3.31) TIA Stroke Algorithm Discrete 0 (low risk) – 8 (high risk) 1.26 (1.02,1.55) Body Mass Index Kg / M*M 1.00 (ns) (0.96,1.03)
  • 24. NCSA Central Site  Protocol/Manual  Survey  Training  Funding Sources  Database/Reports Local Employer Local Health Center Local Facility Regional Sites Stroke Risk Screening Program
  • 25. NCSA Stroke Risk Screening Program  Promote stroke risk screening  Make available standard protocol  Regional health centers in high risk areas.  Provide technical assistance.  Train volunteers  Provide all materials needed  Lab tests paid by small grant and donations  Methods to encourage patients to seek health care  All data analysis centralized  Quality Improvement  Study screening effectiveness  Research on optimal screening effectiveness
  • 26. Risk factor # with Risk Factor % with Risk Factor High Cholesterol > 200 3028 36% High Glucose (non fasting >130) 1921 23% High Blood Pressure >140/90 2954 35% Carotid Bruits 137 2% Current Smoker 2237 27% H/P: Irregular Heart Beat TIA symptoms detected1 1118 628 14% 7% 1 Unilateral numbness, paralysis Goal: Identify mosifiable stroke risk in needy communities Screened 8,579 Persons
  • 27. Levels of Stroke Risk in Population # of Risk factors Number of Individuals % of Total 0 1626 18.95% 1 2,842 33.13% 2 2,408 28.07% 3 1220 14.22% 4 403 4.70% 5 6 70 10 0.82% 0.12% 8579 100%
  • 28. Estimate potential effectiveness Expected Ischemic Stroke Events in Screened Population (N=8,579) ARIC Stroke Risk Score N Predicted no. of Subjects to have Stroke (20 year period) 0 - .0090 Lowest 1429 7 0.0090-.015 1843 19 .015-0.026 1968 38 .026 - .068 2143 102 .068-.14 563 48 .14+ Highest 187 29 If empirical model is correct: 74% of future stroke events in screened sample would occur among 33% of individuals
  • 29. Proportion of persons found with risk who were unaware of this status, or who did not seek help for this risk factor during last six months (i.e.,TIA symptom or overweight): High cholesterol 55.5% High blood glucose 65.4% Hypertension 44.3% Irregular heart beat 72.2% TIA Symptoms 83.3% Overweight 78.1% Goal: Identify Barriers to Population Health Goal of RF Lowering Barrier: Not aware of stroke risk
  • 30. Barrier: low access and receipt of care following screening: Of those with high risk, what proportion sought or received care after the NCSA screen? _______________________________ n=461 with high risk followed 267 visited health provider (57%) 112 discussed results ( 26%) 63 were offered tests (9.5%) _______________________________ DeLemos and colleagues (2002) found that after 3 months 73% of high risk individuals did not report any lifestyle changes to lower risk, and only 9% had visited a doctor.
  • 31. Barrier: Brief risk factor communication does not result in knowledge and awareness  90 days after screening After Screening:  40% could not correctly report the risk factors found in screening.  30% talked to doctor/nurse about risk.
  • 32. Participant support intervention  Randomized design:  Usual care vs. telephonic coaching intervention  N- 227, followed 1 and 3 months post-screening  Intervention: review risk, assess patient contemplation and readiness to seek care, offer assistance with healthcare access. Send copy of stroke risk finding to patient physician by clinician co- investigator.  Outcome: proportion seeking medical care/advice, treatment.  Low cost to administer~ $10 per patient.
  • 33. Follow-up at 3 months Attention Control Motivational Intervention p-value MD visit after screening date (recorded at 3 months) %52.9 70.1 0.0156 % Concern over risk factors 59.0 72.0 0.0422 % Very Confident about lowering stroke risk 58.0 63.2 0.5466 % Planning to modify risk “on your own” 48.48 48.36 0.9853 % “With your doctor’s help” 52.58 68.60 0.0157
  • 34. % with Medication Treatment for BP and Cholesterol at Follow-up. 66.67 35.19 15.15 23.26 0 10 20 30 40 50 60 70 80 90 100 BP Chol %PrescribedMedications Visit to Dr. after screening No Visit 95% Confidence Intervals shown
  • 35. Conclusions  Low-cost follow-up phone call with counseling is effective in improving access in a high-risk community  Barrier needing more attention is self-help with risk factor lowering and support of healthy behaviors.  Requires community input and support
  • 36. Overall Summary
  • 37. Starting point – Define the Problem  Identify health needs of a community to facilitate prioritization and strategies to address them.  Morbidity and Mortality  ‘Health’ = physical, social and emotional  Treatment burden
  • 38. Challenges  Collect and analyze data on patient/community health needs  ‘Emic’ vs ‘etic’ the right balance?  Select health priorities for improvement  Benchmark (e.g., Health People 2020)  Notable high need or disparity in community  Leading cause of health damage (i.e. smoking, obesity, etc.)  Basic Priority Rating System (BPSR) or variations
  • 39. Thank you

Related Documents