Exhibit A1. Per Capita Personal Health Care Spending by State Model: Independent Variables Descriptive Statistics, and Rationale.
| Variable | Source | Expected sign | N | Mean | Std. Dev. | Rationale for Inclusion and Interpretation |
|---|---|---|---|---|---|---|
| Natural log of per capita income by state, adjusted by the PCE deflator ($) | Bureau of Economic Analysis | + | 950 | 10.42 | 0.18 | Controls for differences in means to pay for health care and a state's ability to support a certain level of health care supply; transformation as natural log can be interpreted as an income elasticity.1 |
| Community hospital beds per 1,000 population, by state | Health, US | + | 950 | 3.25 | 1.02 | Controls for health care supply/capacity |
| Share (%) of state population associated with women of childbearing age (20–44) | U.S. Census Bureau | + | 950 | 18.13 | 1.27 | Controls for a key demographic group (gender-related) |
| Share (%) of state population that is African American | U.S. Census Bureau | - | 950 | 9.96 | 9.40 | Controls for a key demographic group (race-related) |
| Share (%) of state population that is age 65 or older | U.S. Census Bureau | + | 950 | 12.65 | 1.89 | Controls for a key demographic group (age-related) |
| Bad health index by state (smoking rate*obesity rate)*100 (out of a maximum of 100) | U.S. Centers for Disease Control, Behavioral Risk Factor Surveillance System | + | 950 | 4.42 | 1.34 | Controls for health status, health behaviors |
| Share (%) of state population that is uninsured | Enhanced Current Population Survey2 | - | 950 | 12.54 | 3.74 | Controls for Insurance Status |
| HMO penetration by state (%) | Health, US | - | 950 | 17.70 | 12.15 | Controls for more tightly managed networks |
| Time Trend (linear) | + | 950 | 9.00 | 5.48 | Controls for unmeasured or unaccounted for factors associated with the passage of time (e.g., changes in policy, economic conditions, health care practice patterns, health care technology, tastes and preferences of patients, etc.) |
1
As illustrated in Getzen (2000), the interpretation of an income elasticity depends on the unit of analysis. In this case, the unit of analysis is the state, and the income elasticity reflects the relative impact of changes in income on health spending between the states.
2
The uninsured population was adjusted for observed underreporting of Medicaid coverage in the Census Population Survey, based on research by Davern et al. (2009) and updated adjustment estimates from J. Ziegenfuss (personal communication, September 6, 2011).
SOURCE: Centers for Medicare & Medicaid Services, Office of the Actuary.