CostBenefit Analysis and Evaluation

1
Cost-Benefit Analysis and Evaluation

• William S. Custer, Ph.D
• Georgia State University

2
Cost-Benefit Analysis

• To compare the public benefit of actions to their
  costs, benefits (and sometimes costs) are
  estimated in dollars.
• Objective
• Determine the public action that produces the
  largest net gain to society, not in private
  sector profit but as an estimated surplus of
  monetized benefits over estimated costs
• Said another way
• Cost-benefit analysis attempts seeks to identify
  the most economically efficient way to meet a
  public objective.
• Adapted from CRS Report-95-760 ENR

3
Estimating Costs Benefits

• Identify outcomes relevant to project and
  audience
• Describe outcomes in dollar terms
• Include all benefits
• Incidence of benefits (who reaps the benefits)
• Identify costs of achieving goals
• Include all costs
• Incidence of costs (who bears costs)

4
Estimating Costs Benefits

• Identify sources of information
• Primary data
• Internal data
• Secondary Data
• Existing research literature

5
Measuring Change

• StructureProcessOutcomes
• Structure measures
• Capacity-infrastructure-people
• Process measures
• Methods--utilization
• Outcome Measures
• Results

6
Measuring Change

• StructureProcessOutcomes
• Structure measures
• Descriptive
• Process measures
• Flowss
• Outcome Measures
• Requires control groups

7
Outcomes

• Outcomes Measures
• ER Visits
• Inpatient Admissions
• Self-reported Patient Health Status
• Clinical Health Status Measures
• Provider Satisfaction
• Total Cost of Care

8
Structure

• Structural Measures
• Staff per patient
• IT system
• Case management system

9
Process

• Process Measures
• ER triage
• Program enrollment
• Eligibility assessment
• Disease management

10
Data Sources and Resources

• Medical Expenditure Panel Data
• http//www.ahrq.gov/data/mepsix.htm
• Hospital Discharge Data
• National data HCUP http//www.ahrq.gov/data/hcup
  /
• That site has individual state contacts listed
• Center for Studying Health System Change
• http//www.hschange.com/
• National Association of Health Data Organizations
• http//www.nahdo.org/
• Behavioral Risk Factor Surveillance System
• CDC survey administered by many states
• http//www.cdc.gov/brfss/sitemap.htm

11
Evaluation Tools

• MIS System
• Case Managers Assessment and Screening
• BRFSS-based initial assessment
• Follow up every 6 months
• Encounter Data
• Physician services
• Prescription Drug programs
• Hospital Care
• Inpatient
• Emergency Room Visits

12
Evaluation Tools

• BRFSS-based Assessment
• Much of Core Sections
• Modules on
• Diabetes
• Hypertension
• Heart Attack and Stroke
• And others
• Built into MIS system

13
Process Measures Utilization
14
Process Measures Utilization

• Encounter Data provides utilization by
• Type of provider
• Type of Service (CPT codes)
• Diagnosis (ICD-9 codes)
• BRFSS-based Assessment
• Patient understanding of care
• Patient adherence

15
Process Measures Utilization

• Encounter Data Examples
• 12 patients given mammograms
• 9 patients treated for Psychosocial Dysfunction
• Monetized benefits would assign value to
  mammogram and treatment
• BRFSS Assessment Example
• 60 of Diabetes patients did not know if they had
  ever had a Hemoglobin test for blood sugar at
  initial assessment
• Of those 2/3s said they had had their Hemoglobin
  checked at the six month reassessment
• Monetized benefits would assign value to
  Hemoglobin check and increased awareness

16
Process MeasuresOutcomes

• Encounter Data
• Reduced Hospitalizations
• Reduced Emergency Room Visits
• BRFSS Assessment
• Increased Health Status
• Increased Health Care Knowledge

17
Process MeasuresOutcomes-Strategies

• Encounter Data Control Groups
• Short run
• Use MEPS to model utilization, costs and outcomes
  of similar patients
• Long Run
• Use community rates as measure control groups
• Requires accurate denominators

18
Process MeasuresOutcomes-Strategies

• BRFSS-assessment Control Groups
• Initial Assessment is base-line
• BRFSS re-assessments integrated with encounter
  data to provide utilization and health status
  trends.

19
Self-Reported Health Status at Enrollment-BRFSS
20
Change in Self-Reported Health Status
21
MEPS Control Group- Diabetes

• Identified those who had received care for
  diabetes
• Uninsured adults under 65
• Family incomes below 235 of Federal Poverty
  Level
• 54 such individuals in latest round of MEPS

22
MEPS Control Group- Diabetes

• Average annual expenditure-2,372
• ER Visits per person year-0.26
• ER Charges per person year-104
• Hospital Stays per person year-0.19
• Hospital Charges per person year-1,287

23
Summary

• Cost-Benefit Analysis requires pricing of all
  costs and benefits
• May not be possible or desirable in all
  cases-especially when valuing quality of
  care/health outcomes
• Need to quantify change (benefits)
• Need to identify control group
• Optimal control group determined by intervention,
  data availability