Journal of Community Health, Vol. 30, No. 3, June 2005 (Ó 2005) DOI: 10.1007/s10900-004-1957-z

EMPLOYMENT AND HEALTH INSURANCE COVERAGE FOR RURAL LATINO POPULATIONS Lynn A. Blewett, PhD; Michael Davern, PhD; Holly Rodin, MPA

ABSTRACT: Rural Latino populations continue to grow in part due to relocation of food processing industries to rural America along with other manufacturing and large retail stores. We use data from the Current Population Survey to examine the labor force participation of rural Latino population and the role rural employers play in providing health insurance coverage. We found that while rural Latinos are more likely to be uninsured, the meat packing industry has higher health insurance coverage rates than other rural employers such as construction and retail. Local communities recruiting new businesses to their rural communities need to explore the role that employers will play in providing health insurance coverage. Lack of adequate coverage will have an impact on the income, resources, and day-to-day activities of physicians, hospitals and traditional safety net providers. KEY WORDS: rural Latinos; health insurance coverage; rural employers; safety net.

INTRODUCTION Health insurance coverage is one of the most important determinants of access to health care and positive health outcomes. Research has shown that working age people without coverage: (1) receive too little medical care and the care they receive is often too late; (2) are generally sicker and die earlier than insured working age adults; and (3) tend to receive poorer quality care.1 The lack of health insurance coverage is an especially pressing problem for Latinos in the United States. According to the most recent US Census Bureau report on health insurance coverage 32.4% of Latinos in the US do not have health insurance coverage. Lynn A. Blewett, PhD is Associate Professor in the Division of Health Services Research and Policy, School of Public Health, University of Minnesota and Director of the State Health Access Data Assistance Center (SHADAC), a research policy center supported by a grant from the Robert Wood Johnson Foundation. Michael Davern, PhD is Assistant Professor in the Division of Health Services Research and Policy, School of Public Health, University of Minnesota and Research Associate with SHADAC. Holly Rodin, MPA is a Research Assistant with SHADAC. Requests for reprints should be addressed to Lynn A. Blewett, PhD, School of Public Health, University of Minnesota, 420 Delaware Street SE, MMC 729, Minneapolis, MN 55455; e-mail: [email protected].

181 0094-5145/05/0600-0181/0 Ó 2005 Springer Science+Business Media, Inc.

182

JOURNAL OF COMMUNITY HEALTH

This is the highest rate of uninsurance across the various racial and ethnic groups that the US Census Bureau reports on, with the next highest rate being 20.2% for Blacks.2 Latinos are the fastest growing minority population in the US. In 1980, Latinos made up 6.4% of the population in the US and this number nearly doubled to 12.5% in 2000.3 Furthermore, a substantial number of the growing Latino population live in rural areas. According to the 2002 Current Population Survey’s Annual Demographic Supplement (CPS), 3.3 million of the 37.4 million Latinos live in non-metropolitan areas in the United States (see Table 1).4 Lack of health insurance coverage for the growing Latino population in the US is a significant health policy issue. In this paper, we examine differences in demographic and employment characteristics between the rural and urban Latino populations to understand whether the two populations face different circumstances in obtaining health insurance coverage. We further examine the health insurance coverage rates for urban and rural Latinos by various demographic and employment characteristics. As with the population in general, the characteristics of Latinos’ employers have significant effects on their health insurance coverage status. Because employer-sponsored health insurance is the most common form of coverage for working age adults, we examine how employment sponsored health insurance coverage rates vary for urban and rural Latinos with different demographic and employment related characteristics. Our analysis provides the background to understand the increasingly diverse population in rural America and the changing need for health care services. Again, as in the general population, having health insurance coverage significantly increases access to care and the use of primary services, resulting in improved health status. This analysis will provide needed information on whether different policy interventions are warranted for rural communities to increase coverage and access for Latinos. The influx of Latino immigrants to rural America introduces new challenges to the capacity of rural health care systems to ensure access to care for a population that has low rates of health insurance coverage, limited financial resources, considerable language and cultural differences, and a variety of special health care needs.5,6,7 Substantial numbers of Latinos are moving to rural America, in part to work in one of the fastest growing rural businesses—food processing. The food processing industry has shifted many of its plants from metropolitan areas to rural America, especially in meatpacking, where rural areas now account for 52% of the jobs.8

Lynn A. Blewett, Michael Davern, and Holly Rodin 183

TABLE 1 Characteristics of the US Latino Population, 2001 Latino Characteristics All Ages

Urban

Rural

N = 26,075 (34,190,000) N = 3,344 (3,250,000)

Gender Male Female

50.9% 49.1

52.7% 47.3

Age Less than 18 years 18–24 years 25–44 years 45–64 years 65 years and over

34.3 13.1 33.4* 14.2 5.0

35.8 11.9 31.0 15.5 5.8

Citizenship Born US citizen Not a US citizen or foreign born

59.2*** 40.8***

66.0 34.0

Health Insurance Status Covered by any health insurance Covered by private insurance Covered by public insurance Uninsured

66.8 43.8 23 33.2

67.4 43.9 23.5 32.6

Health Status Excellent/Very Good Good Fair/Poor

61.9 28.1 10.0

58.9 29.7 11.4

Family Income ( percent of FPL) <50% 50–99% 100–199% 200–299% 300 +%

8.2* 13.1*** 29.9*** 19.1* 29.7***

9.7 16.1 33.8 21.5 18.9

Adults 18 years and older Marital Status Married Widowed

N = 16,993 (22,460,000) N = 2,110 (2,088,000) 53.3%*** 3.6

62.2 3.7

184

JOURNAL OF COMMUNITY HEALTH

TABLE 1 (Continued) Latino Characteristics

Urban

Rural

Divorced/Separated Never Married

10.8 32.4***

9.3 24.9

Level of Education Less than High School High school graduate Some College College Graduate/Postgraduate

42.4* 28.1* 14.8** 14.7***

46.8 31.7 12.1 9.5

Work Force Status Working Unemployed Not in Labor Force1

N = 16,928 (22,380,000) N = 2,100 (2,080,000) 65.1 5.6 29.2

64.3 4.8 31.0

*p < .05, **p < .01, ***p < .001, N = observations (weighted counts). 1 Not in labor force due to retirement, disability or not looking for work. Source: 2002 Annual Demographic Supplement to the Current Population Survey.4

Using non-public data from the Census Bureau’s Longitudinal Research Database (LRD), Drabenstott documented this shift to rural America. The jobs in rural areas tend to be physically difficult, as well as low wage ($7–9 an hour). The new growth of food processing (particularly the meat industry) is linked to the increase in the number of Latinos in rural areas who are recruited to take these low-paying jobs. Drabenstott’s research also showed that the meatpacking industry has moved to the Midwest and the South; rural regions have captured most of the shift with the more remote areas getting more plant locations; the Midwest is home to over 58% of meatpacking plants; the South dominates the poultry industry, but there is a significant cluster of turkey production in rural Minnesota and Iowa; the poultry industry has co-located production, processing and marketing functions (vertical integration); and the trend in food processing is toward larger, more integrated plants, with new jobs concentrated in the largest (450+ employees) rural plants. Rural employment is key to health insurance coverage as employers continue to be the foundation of health insurance coverage in this country. Over 65% of Americans receive health insurance coverage through their employers.9 National employer survey data demonstrate that large employers are more likely to offer health insurance coverage

Lynn A. Blewett, Michael Davern, and Holly Rodin 185

and to have higher take-up rates due to several factors. First, large employers are more likely to have the resources to provide coverage. In addition, they have a large enough pool of employees over which to spread the costs associated with health insurance risk, thus keeping coverage affordable. And finally, their pool of employees is sufficiently large to provide leverage in the marketplace for negotiating health insurance rates. However, in rural food processing plants, preliminary focus group results indicated that this might not be the case.10 We know that jobs in these food-processing plants tend to be low-wage. What is not known is the level or extent of health insurance offerings and type of coverage. Past research has demonstrated that even when these employers offer coverage, there are likely barriers to take-up, such as waiting periods for new employees, exclusions for part-time workers, and high premiums or employee benefit contributions that prevent or discourage low-wage workers from signing up. Among those employers that offer insurance, 11% of employees have waiting periods of 4 months or more.11 In addition, people in low-wage jobs are less likely to be offered health insurance coverage regardless of the size of their employer. The growing Latino population in the rural Midwest has created new challenges for public insurance programs and providers to keep up with demand for health care services. Latinos have the highest rates of uninsurance in the US (32.4%), almost twice the national average.2 The reasons often cited are that Latinos work in low-wage jobs that typically do not offer health insurance coverage, or they can not afford the premiums if coverage is offered. There is also well-documented evidence on racial disparities in access and health status for Latino populations.6,7 The growth of new food processing plants appears to parallel the influx of Latinos in rural areas as documented by Census 2000. Yet there is limited information on the role these new industries play in providing health insurance coverage and the impact these changes have on local communities. In this paper, we examine the personal and employment characteristics of the rural Latino population compared to their urban Latino counterparts, by conducting both descriptive and multivariate analyses to assess the role of employer characteristics (size and wages) on health insurance coverage rates.

METHODS Using data from the 2002 Current Population Survey Annual Demographic Supplement (CPS), we examine the basic demographics,

186

JOURNAL OF COMMUNITY HEALTH

employment patterns and health insurance coverage rates for rural and urban Latinos.4 The CPS is a monthly survey sponsored by the Bureau of Labor Statistics to track labor force dynamics in the US. Throughout the year, supplements are added to the CPS to collect additional information. The Annual Demographic Supplement is conducted from February through April each year, with the majority of the interviews taking place in March. The 2002 CPS is a nationally representative survey of 78,265 interviewed households with a Latino over-sample. As a result, the data from the CPS allow us to describe in detail the employment characteristics of rural Latino residents including age, gender, employment status, wages, occupation, type of industry, and usual hours worked per week. Because the CPS also contains information regarding health insurance coverage, we can examine the relationship between the employment status and demographic characteristics and health insurance coverage. These data have enough statistical power to allow us to examine differences between rural and urban Latinos with respect to the employment, demographic and health insurance coverage characteristics. The CPS respondents were weighted to represent the non-institutionalized population of the US. The weights allow researchers to correct for the differential probability of selection and differential non-response. Some respondents are more likely to be selected into the CPS based on geography and household characteristics. For example, Latino households were more likely to be selected than non-Latino households. The CPS has a 16.2 non-response rate; and, because certain types of people are more likely than others not to respond, the weights of others in the data set are adjusted to correct for differential non-response.4 The CPS consists of a complex multi-stage cluster sample design. Because the respondents in the CPS are sampled through such a complex design, it is important to use the appropriate statistical techniques to derive standard errors for our estimates. We implement a survey design based estimator in STATA 7.0 that treats the household as the sampling cluster and accounts for the differential probabilities of selection.12 All of our significance tests in Tables 1 and 2 are done using these standard errors. In addition to describing the characteristics of rural Latinos, we also conducted multivariate analysis to assess the role of size of employer, wages, and rural residency on health insurance coverage. We conducted a logistic regression for working-age Latino adults (age 18–64) on health insurance coverage. Covariates in the model include age (18–24 year olds were the reference group), gender, marital status, health status, education (highest degree attained), citizenship, full time worker, unemployed, employer size (employers larger than 100 as the reference group), wages

N=13,541 (17,340,000) 34.1 30.9 16.2 14.0 49.7** 55.1

0.3 #

Size of Employers## Less than 25 employees 25–99 employees 100+ employees

4.8 3.3

3.8

N = 17,565 (22,650,00) 40.7* 43.9 32.6** 37.0 19.9*** 15.1 5.9*** 3.7 .9 #

5.1

0.5% 10.7 8.3

Rural N = 13,747 (17,640,000) 10.1% 7.9 5.3

Urban

Annual Wages## <$10,000 $10,000 <$25,000 $25,000 < $50,000 $50,000 <100,000 $100,000 and above

Top Industries Percent Employed Meat Products Construction Eating & drinking establishments Elementary & secondary schools Agricultural crops Agricultural stock

Characteristic

Latino

12.5 35.4

87.6

N = 663 (665,062) 68.9% 41.4 32.7

Rural

N=13,541 (17,340,000) 36.6 40.8 52.9 52.6 69.0 66.4

N = 17,565 (22,650,000) 29.4 31.7 48.2* 55.5 75.6 81.4 89.0 88.2 80.8 #

28.4* #

80

N = 4,492 (5,956,000) 65.8% 36.2 27.9

Urban

Latino Private Insurance Rates Urban

78.8 65.7

10.4

N=13,541 (17,340,000) 56.9 56.0 42.3 41.4 26.0 28.0

N = 17,565 (22,650,000) 49.9 48.3 46.9 41.6 22.4 # 10.5 # # #

60.8* #

16.8

N = 663 (665,062) 29.5% 56.7 53.5

Rural

Latino Uninsurance Rates

N = 4,492 (5,956,000) 31.4% 60.7 65.1

Characteristics of Working-Aged Latino Adults (18–64), 2001

TABLE 2

Lynn A. Blewett, Michael Davern, and Holly Rodin 187

Rural

N = 13,541 (17,340,000) 57.0 58.1 45.9 45.7

Urban

Latino Private Insurance Rates

* p < .05 ** p < .01 ***<.001 N = Observations (weighted counts). # The sample size is too small to be reliable with less than 50 respondents. ## Both of these variables include both full-time and part-time workers. Source: 2002 Annual Demographic Supplement to the Current Population Survey.4

N = 13,541 (17,340,000) 85.4 87.2 14.6 12.8

Employment Status Full-time Part-Time

Rural

Urban

Characteristic

Latino

TABLE 2 (Continued)

Rural

N = 13,541 (17,340,000) 38.4 37.7 43.6 44.2

Urban

Latino Uninsurance Rates

188 JOURNAL OF COMMUNITY HEALTH

Lynn A. Blewett, Michael Davern, and Holly Rodin 189

(income over 300% of poverty as the reference group), and the major employers for both rural and urban Latinos.The industries represented in the model include construction, eating/drinking establishments, meatpacking, and K-12 education (all other industries serve as the reference group). Measurement of Concepts In the following analyses, urban was defined as being in a county that is part of a Metropolitan Statistical Area (MSA) and rural was defined as living in a county that is not part of an MSA. The CPS, although rich in health insurance coverage, employment, and ethnicity data, is rather limited with respect to geographic areas, thus limiting our analysis to an MSA/non-MSA comparison. Latino was defined as a person of Hispanic ethnic origin as determined by the question that asked for self-identification of the person’s ethnic origin on the CPS survey. Respondents are asked to select their origin (or the origin of some other household member) from a ‘‘flash card’’ listing ethnic origins. Persons of Hispanic ethnic origin are those who indicated that their origin was Mexican-American, Chicano, Mexican, Mexicano, Puerto Rican, Cuban, Central or South American, or other Hispanic.13 Health insurance status has two components.The first is whether the respondent is covered by any type of health insurance or is uninsured.The second, for respondents with health insurance coverage, is whether the main source of coverage is a public program (i.e., Medicare, Medicaid, or a state-specific program) or an employer/union or privately purchased plan. If a respondent reported having both a public and a private source of coverage, the person is coded as having public coverage (e.g., a person with Medicare and a privately purchased Medicare wraparound policy). The citizenship variable was coded ‘‘1’’ if the person was born a US citizen and ‘‘0’’ if the person was foreign-born or naturalized US citizen.The health status variable was taken from the health status item on the CPS and coded as excellent health, very good health, or good health, versus fair health or poor health.The percent of poverty was determined by adding the income amounts for everyone living in a family and comparing the aggregated total family income to the federal poverty threshold for a family of that size. The remaining variables were constructed only for people 18 and over. The level of education was coded into four categories: less than

190

JOURNAL OF COMMUNITY HEALTH

high school, high school graduate, at least some college, college graduate or postgraduate. The four-category marital status variable was constructed from the CPS: married, widowed, divorced or separated, and never been married. Employment status was based on the current employment status of each respondent. We recoded the CPS employment status into three categories:employed, unemployed, or out of the labor force. Out of the labor force includes people who are discouraged workers (long-term unemployed) as well as people who are out of work by choice (e.g., fulltime student, retired, and non-paid family workers). Size of employer, wages, and full-time employment were based on the employment status of respondents during the previous calendar year (2001). We broke down the employer size variable into the following categories: less than 25 employees, 25–99, and greater than 100 employees. Wages represent the total amount of earned income from a job working for an employer and/or self-employment over the past year. The wage variable was categorized as follows: under $10,000, $10,000 but less than $25,000, $25,000 but less than $50,000, $50,000 but less than $100,000 and $100,000 and over. Full-time employment was determined by asking respondents how many hours in a typical week they worked in the past year. If a respondent answered 35 hours or more per week, they were considered to be working full-time.

RESULTS Table 1 provides an overview of the differences in socio-demographics of rural and urban Latino populations.There are significant differences in the characteristics of Latinos living in rural areas compared to their urban counterparts.There are fewer younger-aged Latino workers in rural areas (31% between age 25–44 compared to 33.4% in urban areas); rural Latinos are more likely to be born in the US (66.0% vs. 59.2%); rural Latinos are more likely to be married than urban Latinos (62.2% vs. 53.3%); and there are fewer respondents who were never married among rural Latinos (24.9% vs. 32.4%). Rural Latinos have lower levels of education than their urban counterparts, with more rural Latinos reporting a high school degree or less and more urban Latinos reporting some college or a college or postgraduate degree. Health status is similar between the two groups. Urban and rural Latinos are virtually identical when comparing levels of uninsurance by demographic and employment characteristics.In contrast to the differences in the population characteristics, there were

Lynn A. Blewett, Michael Davern, and Holly Rodin 191

no statistical differences between the characteristics of the uninsured for urban and rural Latinos: 33.2% of Latinos in urban areas were uninsured compared to 32.6% for rural Latinos. In addition, there were similar employment rates for urban and rural Latinos (65.1% vs. 64.3%).However, income levels were significantly lower for rural Latinos with over one quarter (25.8%) of rural Latinos with incomes at or below the Federal Poverty Level (FPL). Table 2 provides information on the employment and wage levels for both urban and rural Latinos, both privately insured and uninsured. The meat products industry employs the largest number of rural Latinos—10.1% of the employed rural Latinos; the next largest employer of rural Latinos is the construction industry, employing 7.9% of rural Latinos. Among urban Latinos, the construction industry is the largest employer, employing 10.7% of urban Latinos; the next largest employer of Latinos are eating and drinking establishments, employing 8.3% of the urban Latinos and 5.3% of the rural Latinos. In terms of employer size, rural Latinos were significantly more likely to be employed by large employers with more than 100 employees (55.1% vs. 49.7%). However, being employed by a large employer did not result in higher coverage rates for rural Latinos (Table 1). Annual earnings were significantly lower for rural Latinos than for urban which may in part explain why working for a large employer does not improve the insurance rates for rural Latinos. The role of employment-based health insurance coverage is mixed. Rural Latinos employed in the agricultural crops area had the highest uninsurance rates (78.8%), compared to workers in the other industry sectors. Interestingly, however, no statistical differences were found between urban and rural Latinos in uninsurance levels in the meat products industry (where uninsurance rates are quite high) and in elementary and secondary school employment (where uninsurance rates are quite low). As expected, employers with 100+ employees had higher coverage rates than smaller employers; and full-time employees had higher coverage than part-time employees. The general pattern for wages and insurance status suggests that as earnings increase, private coverage increases and uninsurance rates decrease. This is true for both rural and urban Latinos. The results of our logistic regression on predictors of not having health insurance coverage are presented in Table 3. We control for rural location of residence (non-MSA) and other individual characteristics to assess the factors associated with health insurance coverage for Latinos. The logistic regression confirms that rural residency in and of itself is

192

JOURNAL OF COMMUNITY HEALTH

TABLE 3 Odds Ratios for Logistic Regression on Uninsurance for Latinos (18–64), 2001 Variables

Odds Ratio

(Standard Errors)

Rural 25–34 years 35–44 years 45–54 years 55–64 years Male Married Good Health Less than HS HS Graduate Some College Born Citizen Not Working Full-time Work Employer Employer 25–99 Meatpacking Construction Eating/Drinking Est. K-12 Education < 50% FPL 50–99% FPL 100–199% FPL 200–299% FPL

1.0052 0.9336 0.7146*** 0.6618*** 0.6989** 1.1827** 0.5371*** 1.0312 3.0696*** 1.8832*** 1.4760*** 0.5124*** 1.4842** 0.9571 2.7273*** 1.5007*** 0.6119 1.7092*** 2.2711*** 0.8925 4.1290*** 3.9965*** 2.8020*** 1.3976***

(0.1180) (0.0740) (0.0610) (0.0623) (0.0848) (0.0598) (0.0327) (0.1084) (0.3214) (0.1938) (0.1558) (0.0324) (0.1690) (0.0759) (0.1730) (0.1104) (0.1924) (0.1494) (0.2327) (0.1341) (0.7510) (0.4409) (0.2269) (0.1167)

*** p < .001. ** p < .01. * p < .05. Source: 2002 Annual Demographic Supplement to the Current Population Survey.4

not a predictor of health insurance coverage rates for Latinos. Variables significantly associated with being uninsured include the following: being male; having less than high school education; working for a small employer; working for construction or eating/drinking establishment; being unemployed; and having low incomes (less than 300% of the FPL). Significant predictors that increase the likelihood of having health insurance coverage include: being married; being born a US citizen; working for a large employer; and having higher incomes (above 300% of the

Lynn A. Blewett, Michael Davern, and Holly Rodin 193

FPL). Of the industries examined, Latinos in construction and the eating/drinking establishments are less likely to have health insurance coverage than the Latinos working in all the other industries. There were no significant differences in the likelihood of being insured given that a Latino worker was employed in K-12 education or meatpacking vs. all the other industries.

DISCUSSION Latinos continue to migrate to rural America, in part to take the jobs created by the movement of the food processing industry to rural settings. Latinos also work in other industries as well, providing an economic foundation for rural America. As the largest employers in rural America, the new food processing plants attract a large Latino workforce willing to take low-wage jobs that are physically demanding and conducive to injuries. Participants in previous focus groups with rural Latinos voiced concerns about the lack of health insurance coverage, long waiting periods, and underinsurance in the food processing industry.10 Our analysis highlights unique demographic and employment characteristics of the growing rural Latino population. Based on a few characteristics (more likely to be born in US, more likely to be married), we would expect higher health insurance coverage rates for rural Latinos than their urban counterparts. In this analysis, we found that the largest employers—meatpacking plants—had the highest rates of health insurance coverage but still almost one third (29.5%) of their employees were uninsured. Yet other industries had even lower coverage rates; 53.5% rural Latinos working in eating and drinking establishments and 78.8% working in agricultural crops were uninsured. Wages as well as firm size play an important role in health insurance coverage. Based on CPS wage data, the average wage for a meatpacking employee was $7.52 an hour, representing an annual income of $15,648. This wage is 13.5% below the Federal Poverty Level (FPL) for a family of four. The low wages paid by employers in the meatpacking and other food processing industries often make the offered health insurance coverage unaffordable. This might explain why the rates of uninsurance are high for these workers, despite the size of their employers. These results were also consistent with the multivariate analysis, which controlled for wages and size of firm. Knowing the basic demographics of the growing rural Latino population can be useful to community planners and policy makers interested

194

JOURNAL OF COMMUNITY HEALTH

in increasing health insurance coverage rates. As food processing, and specifically, meatpacking, has moved to rural America, they have brought both economic benefits as well as significant challenges. For the meat industry, livestock production increases the income of farmers and boosts local payrolls. However, these jobs are frequently unattractive to local residents due to the low wages, which leads, in turn, to changing rural demographics as Latinos move to rural areas to fill these positions. Consequently, the plants do not increase the per capita incomes of local communities and bring additional pressures on local services, particularly in the health care sector. While the meatpacking firms have higher take-up rates of health insurance coverage compared to agriculture employers, there are still significant numbers of Latino workers in meatpacking plants (29.5%) who are uninsured. Our findings are consistent with other recent research that has raised concerns about the increasing number of uninsured working for large firms.14 These uninsured workers are more likely to have low wages similar to the firms where rural Latinos are employed. Policies targeted to increasing health insurance coverage for the growing Latino population will need to address the ‘‘growing gaps in employee health coverage at large firms’’ particularly for those with low wages.14 Rural economic development policy may be one tool to consider to increase health insurance coverage for Latinos in rural areas. State and local communities that provide tax breaks or tax-free zones for business development may consider adding requirements for tax-exempt employers in terms of wage levels and/or mandated levels of health insurance coverage.Another policy option to consider is the use of state premium subsidy programs to provide financial assistance to low-wage employees in purchasing health insurance coverage.15 Rural America is changing. Efforts are needed to understand the changing socio-demographics and the impact on local communities. The high rates of uninsurance for rural Latinos have introduced new pressures to local providers and safety net support services to meet the needs of a growing number of uninsured in their communities. Efforts to increase coverage for rural Latinos will need to address the coverage offered by those who employ them—mainly the large food processing and meatpacking plants, as well as agriculture employers.

REFERENCES 1.

Institute of Medicine. Care Without Coverage: Too Little, Too late. Washington, DC: National Academy Press; 2001.

Lynn A. Blewett, Michael Davern, and Holly Rodin 195

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Mills R, Bhandari S. Health Insurance Coverage in the United States: 2002. Washington, DC: US Census Bureau; 2003. Hobbs F, Stoops N. Demographic Trends in the 20th Century.Washington, DC:US Census Bureau; 2002. US Census Bureau. Current Population Survey: Annual Demographic File; 2002: Technical Documentation. [Distributed by Inter-university Consortium for Political and Social Research. Ann Arbor, Mich.] Washington, DC: US Census Bureau; 2002. Trevin ˜ o R, Trevin ˜ o F, Medina R, Ramirez G, Ramirez R. Health care access among Mexican Americans with different health insurance coverage. J Health Care Poor Underserved 1996; 7: 112–121. Mueller K, Ortega S, Parker K, Patil K, Askenazi A.Health status and access to care among rural minorities.J Health Care Poor Underserved 1999; 10:230–249. Flores G, Abreu M, Olivar MA, Kastner B. Access barriers to health care for Latino children. Arch Pediatr Adolesc Med 1998; 152: 1119–1125. Drabenstott M, Henry M, Mitchell K. Where have all the packing plants gone?The new meat geography in rural America. Economic Review 1999; Issue Q III: 65–82. Kansas City: Federal Reserve Bank of Kansas City. Cooper P, Vistnes J. Workers’ decisions to take-up offered health insurance coverage: assessing the importance of out-of-pocket premium costs.Med Care 2003; 41: III 35–III 47. Blewett LA, Smaida S, Fuentes C, Zuehlke E.Health care needs of the growing Latino population in rural America: focus group findings in one midwestern state. J Rural Health 2003; 19: 33–41. Gabel J, Pickreign J, Whitmore H, Schoen C. Embraceable you: how employers influence health plan enrollment. Health Aff 2001; 20: 196–208. Stata Reference Manual Extract, Release 7.0. College Station, Tex.: StataCorp LP. US Census Bureau. Current Population Survey Glossary (Attachment 4). Available at: http://www.census.gov/apsd/techdoc/cps/sep97/glossary.html.Accessed December 31, 2003. Glied S, Lambrew J, Little S.The Growing Share of Uninsured Workers Employed by Large Firms. New York, NY: The Commonwealth Fund; 2003. Engquist G, Burns P. Health Insurance Flexibility and Accountability Initiative: Opportunities and Issues for States. Washington, DC: State Coverage Initiatives Program Issue Brief. 2002; Vol. III(2).