Leveraging Epidemiology Skills in Population Health Outcome Measures

1
Leveraging Epidemiology Skills in Population
HealthOutcome Measures

• COL George J. Dydek
• Pharm.D.

Program Manager, Population Health Outcomes U.S.
Army Center for Health Promotion and Preventive
Medicine Aberdeen Proving Ground, Maryland
21010-5403
2
Measuring Clinical Outcomes

• Background
• Pilot Project
• Challenges/Future
• MHS Data Systems

3
Types of Outcome Measures and Selected Examples
4
Components of Quality
STRUCTURE
PROCESS
OUTCOME
Physical Exams Tests Treatments
Mortality Morbidity Disability
Hospital Beds Health Professionals
Reference Donabedian, A.1980. Exploration in
Qulatiy Assurance and Monitoring. Vol.1, The
Definition of Quality and Approaches to its
Measurement. Ann Arbor Health Administration
Press.
5
Performance MeasuresNational Level

• NCQA - HEDIS
• Immunization Status
• Breast CA Screening
• Beta-Blocker post AMI
• Cervical CA Screening
• Chlamydia Screening
• Prenatal Care
• Controlling HTN
• Diabetes Care
• Asthma Medications
• Mental Illness follow-up
• Antidepressant Medication Mgt
• Flu shots for older adults

• JCAHO - ORYX
• AMI
• Heart Failure
• Pneumonia (Community Acquired)
• Surgical Procedures
• Pregnancy

6
Pilot Project in Outcomes Measures

• Establish a Process
• Data Management
• Data Analysis
• Criteria for Selection
• Data availability
• Evidenced based
• National benchmark

7
Performance MeasuresNational Level

• NCQA - HEDIS
• Immunization Status
• Breast CA Screening
• Beta-Blocker post AMI
• Cervical CA Screening
• Chlamydia Screening
• Prenatal Care
• Controlling HTN
• Diabetes Care
• Asthma Medications
• Mental Illness follow-up
• Antidepressant Medication Mgt
• Flu shots for older adults

• JCAHO - ORYX
• AMI
• Heart Failure
• Pneumonia (Community Acquired)
• Surgical Procedures
• Pregnancy

8
Methodology

• Design
• Retrospective database analysis
• Timeframe of Oct 98 to Aug 00
• Population
• Active Duty personnel discharged from an MTF for
  an Acute Myocardial Infarction (primary ICD-9
  codes 410.xx)

9
MethodologyData Sources

• DMSS Defense Medical Surveillance System
• Active Duty Personnel
• SIDR, SADR, Demographic
• USPD Uniform Services Prescription Database
• All beneficiary CHCS Outpatient Prescription Data
• Timeframe Oct 1998 to present

10
MethodologyVariables

• Outcome of Interest
• The frequency of beta-blocker therapy prescribed
  within 30 days of discharge.
• Independent
• Age, Gender, Race, Rank, Service, Marital Status,
  Length of Stay, Comorbidities

11
Beta-Blocker Data Acquisition Overview
DMSS
Population Health Program
USPD
Demographic Prescription Hospitalizations Outpatie
nt
12
Identified AMI Population(N234)

• Age Group
• lt25 12
• 25-34 37
• 35-44 118
• 45-54 65
• 55-64 2
• Gender
• Male 225
• Female 9
• Race
• Black 52
• White 171
• Other 17

• Marital Status
• Married 203
• Single 20
• Other 9
• Unknown 2
• Rank
• E01-E04 17
• E05-E09 177
• O01-O03 10
• O04-O09 36

13
174
Yes
AMI Patient Population N 234
D/C on Beta 30 days
No
60
14
Potential Contraindicated Comorbidities

• Broad Examination
• Diabetes
• Conduction Disorders
• Cardiac Dysrhymthmias
• Heart failure
• Other Peripheral Vascular Disease
• Asthma
• Chronic Obstructive Pulmonary Disease

15
Potential Contraindicated Comorbidities
16
Potential Contraindicated Comorbidities- USPD Data

• Diabetes
• 2 patients were receiving insulin, only one coded
  for DM (ICD-9 code 250)
• Both received Beta-Blockers
• Asthma
• 11 patients identified - none coded for asthma
  (ICD-9 code 493)
• 3 bronchodilators
• 8 anti-inflammatory agents
• 9/11 received Beta-Blockers

17
Comparison of Subpopulations
18
Comparison of Subpopulations
19
Beta-Blocker SubpopulationPharmaceutical Data
20
Non Beta-Blocker SubpopulationPharmaceutical Data

• N 60
• 29 Patients received other cardiovascular type
  medications (e.g.,calcium channel blockers)
• 10 Patients eventually received beta-blocker
  therapy
• 21 Patients received no other cardiovascular type
  medications

21
MHS Performance Beta-Blocker Metric
 
HEDIS 1999 Data Average 85

• Raw Data

• Assumptions

PHOTO Criteria
22
Healthcare UtilizationReadmission Post AMI
23
Lipid Data Acquisition Overview
DMSS
Population Health Program
HL-7
24
HL-7 Data
Comparison to other Databases SIDR Data - 58
patients coded on admission
25
HL-7 Data

• Comparison to other Databases
• SADR Data - 60 patients coded Hyperlipidemia
• USPD Data - 170 patients on anti-hyperlipidemia
  drugs

26
Other Risk Factors

• Obesity (ICD-9 code 278) SIDR/SADR
• 13 patients (5.6 - all male)
• MSMR 5.4 female / 2.4 male
• Tobacco (ICD-9 code 350.1) SIDR/SADR
• 55 patients (23.5 )
• Hypertension (ICD-9 code 401) SIDR/SADR
• 122 patients (52.1 )

27
Challenges

• Data Availability
• Ancillary Clinical Data
• Timeliness
• Data Quality
• Reliability, Validity
• Actionable Items
• Risk Interventions

28
Outcomes Measures Future Direction

• Other (HEDIS-Like) Metrics
• Programs to reduce risk factor
• Condition/Disease Management
• CPG Measurements (Asthma, DM,)

29
PHOTO
Name Population Health Operational Tracking
Optimization DescriptionSet of MHS-related
metrics (e.g., clinical, customer service) with
various charting and graphing options Targeted
UsersServices, MTF Commanders, clinical
department heads Status 14 metrics available
now Next step Clinical metrics (e.g., cervical
cancer screening)
30
PHOTO Phase III Metrics

• Eye Exams for Diabetes
• Follow-up After Hospitalization for Mental
  Illness
• Check-Ups after delivery
• Beta-Blocker After AMI
• Asthma Management
• Breast Cancer Screening
• Prenatal Care in First Trimester
• Pharmacy Cost PMPM/PMPY

31
ARS Bridge
NameM2 (formerly, All Region Server
Bridge) Description MHS-wide summary
detailed data (e.g., enrollment, clinical,
pharmacy, population, workload, financial)
presented in customized or standard report
formats Targeted Users Decision makers at
all levels, MTF, TMA, Lead Agents, MAJCOM,
Services StatusCurrently being deployed
32
MDR
Name MHS Data Repository DescriptionCentral
data warehouse for integrating, managing,
storing raw processed MHS corporate data sets
to feed to user-defined data marts Targeted
UsersNo direct users feeds data
marts Status In development Phased implementation
33
Evolving Architecture
34
Summary

• Central database utilization
• Performance measures
• Outcomes
• Limitations
• Data Availability and Quality
• Capability to measure outcomes
• MHS
• Preventive Medicine/Health Promotion

35
QUESTIONS