Center for Studying Health System Change

Community Variation in Health Insurance Coverage: Where You Live Matters

News Releases
June 13, 2001

FURTHER INFORMATION, CONTACT:
Alwyn Cassil: (202) 264-3484

ASHINGTON, D.C. - While population, employment, state policy and market differences contribute to the wide variation in community health insurance rates, almost a third of the coverage variation is attributable to unexplained geographic factors, according to a comprehensive new study from the Center for Studying Health System Change (HSC).

The study’s findings have significant policy implications for local, state and federal efforts to increase coverage for the nearly 43 million uninsured Americans and suggest that one-size-fits-all approaches to coverage expansions may miss the mark, said Peter J. Cunningham, Ph.D., HSC senior health researcher. HSC is a nonpartisan policy research organization funded solely by The Robert Wood Johnson Foundation.

"Where you live plays a significant role in whether you will have health insurance coverage," Cunningham said. "Like so many other aspects of health care, geography for many people is destiny when it comes to health insurance coverage. And policy makers need to understand the unique characteristics and needs of communities, so they can target policies and outreach efforts at uninsured people in communities with the most intractable levels of uninsurance."

The study - "What Accounts for Differences in Uninsurance Rates Across Communities?" - appears in the spring edition of the journal INQUIRY and was coauthored by Cunningham and HSC President Paul B. Ginsburg, Ph.D. The article abstract is available online at www.inquiryjournal.org.

According to the study, communities with high rates of uninsured people have populations with more Hispanics, less education and lower incomes than communities with low uninsured rates. High uninsurance communities also have a greater proportion of self-employed people, workers in small firms, low-wage jobs, industries where health coverage typically is lacking, and lower levels of overall employment.

The study examined more than 30 variables, including race, ethnicity, education, income, health status, labor market and employment characteristics, health insurance costs, state Medicaid policies and market characteristics such as the level of managed care and the number of public hospitals, and found:

• Unexplained geographic factors account for 31.7 percent of the variation between communities with the highest and lowest rates of uninsured people.
• Employment-related factors account for 26 percent of the difference.
• Race/ethnicity accounts for 18.3 percent of the variation.
• State policy differences - due mostly to Medicaid eligibility requirements - account for 12.7 percent of the difference

The study used data from the 1996-97 Household Survey of the Community Tracking Study and included 47,900 people under age 65 in 60 nationally representative, randomly selected communities. The rate of uninsured people in the 60 communities ranged from a low of 4.7 percent in Rochester, N.Y., to a high of 28.9 percent in Miami, Fla. Communities with low rates of uninsured people are concentrated in the northeast and Midwest, while areas with high levels of uninsured people tend to be spread across the southern and western parts of the country.

"This is one of the most comprehensive pictures of why some communities have more uninsured people than others," Cunningham said. "But there are still a lot of unanswered questions. We need to delve deeper into the unexplained geographic factors affecting health insurance coverage."

Some of the geographic variation may be related to what is known as a strong "culture of offering" employer-sponsored insurance in certain areas and how regional economies developed historically, Cunningham said. For example, northeastern and Midwestern states - low uninsurance areas - typically have had large manufacturing and heavy industry employment bases, allowing efficient risk pooling for group plans, and heavily unionized workforces that demanded generous health benefits.

Among communities with high levels of uninsured people, there also are differences. For example, race/ethnicity plays a more important role in the rate of uninsured people in Miami and Orange County, Calif. - regions with large Hispanic populations - than in places like Little Rock, Ark., with few Hispanics.

Though the study found age, sex, health status, risk averseness and the cost of health insurance are important predictors of whether individuals have health coverage, these factors have virtually no effect on community-level variation in rates of the uninsured, suggesting lower health insurance costs may not be enough to significantly reduce the number of uninsured.

While state policies certainly are important in reducing the number of uninsured people, the study found that differences in state policy played a fairly minor role in explaining the wide community variation. This finding suggests that modest public program expansions may not be enough to significantly reduce the number of uninsured people in high uninsurance communities.

"What we don’t know about the unexplained community variation in coverage rates can hurt efforts to expand health insurance coverage because misdirected policies may be less effective in communities where the problem is most severe," Ginsburg said. "This is especially true in light of America’s rapidly changing population and labor market characteristics."

Noting that the number of uninsured Americans increased by nearly 10 million over the last decade during the longest economic expansion in American history, Ginsburg said the slowing economy, rising medical costs, changing demographics and labor market shifts are combining to create a formidable challenge to expanding - let alone maintaining - health insurance coverage in the future.

Declines in manufacturing jobs and growth in service and retail jobs that often don’t offer health insurance may contribute to higher levels of uninsured people in the future. Demographic changes may also factor into the equation. For example, 2000 census data show the number of Hispanics - one of the most challenging populations to insure because of cultural and economic barriers - grew by almost 60 percent in the 1990s to 35.3 million.