1. State Variation in High Burden
Spending for Health Care: Preliminary
findings from new data in the Current
Population Survey
Lacey Hartman
State Health Access Data Assistance Center/SHADAC
University of Minnesota
www.shadac.org
May 4, 2012
Funded by a grant from the Robert Wood Johnson Foundation
Funded by a grant from the Robert Wood Johnson Foundation

2. Acknowledgements
• Co-Authors
– Gilbert Gonzales, SHADAC
– Sharon Long, Urban Institute & SHADAC
• Funding
– Robert Wood Johnson Foundation
2

3. Background
• Health care costs outpacing growth in
income
• Key goal of ACA is to address affordability
– Medicaid expansion
– Premium and cost-sharing subsidies in the
Exchange
• State variation in ACA implementation and
health care costs/markets
3

4. Goals
• Assess variation in high burden spending
across states
• Estimate potential for ACA to alleviate high
burden spending
4

5. Data: Annual Social and Economic Supplement to the
Current Population Survey (CPS ASEC)
• New questions related to OOP spending added to
the CPS in 2010 (reference CY 2009)
• CPS is a monthly labor survey
– ASEC fielded in Feb-April
– Questions on work, income, migration and health
insurance
– Supports state estimates
5

6. OOP Spending in the CPS
• Spending related to: Medical, dental, vision, medical
supplies, and prescription drugs
• Includes:
– Premiums (except Medicare Part B)
– Non-premium
• Co-pays, deductibles, other cost sharing
• Over the counter expenses (separate in 2011)

7. Data Quality
• Compares well to MEPS and SIPP (Caswell et.
al 2011)
– Compared statistics by age, race, income, etc.
– Tested differences in distribution of OOP
spending across data sources
SKL1
– Small expenditures underreported in CPS
• Overall, data performs well for capturing high
burden spending

8. Slide 7
SKL1 Highlight focus on high costs. Aren't concerned exact estimate of OOP, but on whether OOP are high
SKL, 2/27/2012

9. Measures & Methods
• OOP spending as a share of family income
– High burden: >10% of income
– Very high burden: >20% of income
• Unit of analysis=individuals in families
• Premium and non-premium high burden
spending (2011 only)

10. Methods, Potential Impacts of ACA
• Potentially Medicaid eligible
– Non-elderly citizens below 138% FPG
• Potentially subsidy eligible
– Non-elderly citizens I39-399% FPG
– Uninsured or with nongroup coverage
• Assign potential savings at individual level,
recalculate family spending and burden
9

11. Methods, Potential Impacts of ACA
Family Income as Premium Cap as Out-of-Pocket Maximum
% of FPG % of Income
Individuals Families
<=138 0% ---------- -------------
138-149 3-4% $1,983 $3967
150-199 4-6.3% $1,983 $3967
200-249 6.3-8.05% $2,975 $5950
250-299 8.05-9.5% $2,975 $5950
300-399 9.5% $3,967 $7933
10

12. Limitations
• Only 1 year of data for estimates of ACA
impacts
– Rerun results with 2012 data in the fall
• Conservative estimates of ACA impacts
– CPS coverage questions don’t assess full year
coverage
– Citizenship (some non-citizens would be eligible
for Medicaid/subsidies)
11

13. High Burden Spending, National Results
• Impacts many Americans
– Nearly 20% or 56 million high burden (>10%)
– 8% very high burden (>20%)
• Compared to total population, more likely
– Income below 250% FPG (59% vs. 42%)
– Fair/poor health (19% vs. 12%)
– Disabled person in family (25% vs. 17%)
– Elderly (22% vs. 13%)
– Nongroup coverage (21% vs. 9%)
12

14. Percent of Individuals in Families with High Burden
(>10%) Out-of-Pocket Spending, by State
Source 2010-2011 CPS ASEC
13

15. Percent of Individuals in Families with High Burden (>10%) Health Care
Premium Spending, by State
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
South Dakota
North Dakota
Maine
Nebraska
Tennessee
Montana
Wisconsin
Kansas
Mississippi
Indiana
Louisiana
Rhode Island
Oregon
Oklahoma
Minnesota
New Hampshire
Arkansas
Connecticut
North Carolina
Georgia
Colorado
Vermont
Wyoming
Florida
Iowa
Pennsylvania
Washington
Alabama
Maryland
Missouri
Kentucky
National Average
Nevada
Michigan
Massachusetts
New Mexico
Virginia
Delaware
Idaho
Arizona
Utah
West Virginia
Ohio
Illinois
Texas
South Carolina
New Jersey
New York
California
Hawaii
Alaska
District of Columbia
Source 2011 CPS ASEC
14

16. Percent of Individuals in Families with High Burden (>10%) Health Care Non-
Premium Spending, by State
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Montana
Wyoming
Mississippi
Idaho
Utah
West Virginia
Nevada
North Dakota
Washington
Tennessee
Kentucky
Oregon
Colorado
Arkansas
Indiana
North Carolina
Alaska
Louisiana
Michigan
Arizona
Oklahoma
Georgia
Ohio
Florida
Maine
National Average
Missouri
South Dakota
South Carolina
Texas
Alabama
Nebraska
Kansas
Pennsylvania
Delaware
Iowa
Illinois
Wisconsin
New Mexico
Minnesota
Maryland
New Jersey
California
Vermont
Virginia
Rhode Island
Connecticut
New Hampshire
Massachusetts
Hawaii
New York
District of Columbia
Source 2011 CPS ASEC
15

17. Potential Impact of ACA on High (>10%) Burden
Spending
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Montana
Nebraska
South Dakota
Idaho
Wyoming
Utah
Oregon
Maine
North Dakota
Mississippi
Tennessee
Kentucky
Colorado
Arkansas
Kansas
Wisconsin
Minnesota
Alabama
Indiana
Louisiana
Nevada
Washington
Missouri Post‐ACA
Florida
West Virginia Pre‐ACA
North Carolina
South Carolina
Iowa
Georgia
Connecticut
Arizona
Vermont
Illinois
Pennsylvania
Ohio
New Hampshire
Oklahoma
New Mexico
Rhode Island
Michigan
Texas
Virginia
Delaware
Maryland
Alaska
Massachusetts
New Jersey
California
Hawaii
New York
District of Columbia
Source 2011 CPS ASEC
16

18. Potential Impact of ACA on Very High
(>20%) Burden Spending
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Montana
Wyoming
Utah
Mississippi
Idaho
South Dakota
Tennessee
Nevada
Arkansas
Oregon
Colorado
Kentucky
Florida
North Dakota
Washington
Maine
Kansas
Nebraska
North Carolina
Alabama
Louisiana
Wisconsin
Oklahoma pre‐ACA
Georgia
Pennsylvania
Indiana post‐ACA
Missouri
Ohio
West Virginia
New Mexico
Minnesota
Total
Illinois
South Carolina
Michigan
Arizona
New Hampshire
Vermont
Connecticut
Delaware
Alaska
Virginia
Rhode Island
Texas
New Jersey
Iowa
Maryland
Massachusetts
Hawaii
California
District of Columbia
New York
Source 2011 CPS ASEC
17

19. Characteristics of People with High Burden
Spending After ACA
• Compared to total population
– Income below 250% FPG (48% vs. 42%)
– Fair/poor health (18% vs. 12%)
– Elderly (22% vs. 13%)
• Compared to high burden before
– Income above 250% FPG (51% vs. 40%)
– Employer based coverage (59% vs. 53%)
– Elderly (28% vs. 22%)
18

20. Conclusions, Policy Implications
• High burden spending issue for many
Americans, varies across states
• Estimate ACA will help many, 40 million
remain high burden
• Policy solutions for people with ESI and
elderly
• CPS useful new data source for
– Monitoring
– Informing policy solutions

21. Lacey Hartman
hartm042@umn.edu
Sign up to receive our
State Health Access Data Assistance Center
newsletter and updates at
University of Minnesota, Minneapolis, MN
www.shadac.org
612-624-4802
@shadac
Sign up to receive our newsletter and updates at www.shadac.org
www.facebook.com/shadac4states
@shadac

22. Sign up to receive our
newsletter and updates at
www.shadac.org
@shadac