California’s Undocumented Latino Immigrants: A Report on Access to Health Care Services Prepared for the Henry J. Kaiser Family Foundation by: Claudia L. Schur, Ph.D. Marc L. Berk, Ph.D. Cynthia D. Good, M.A. Eric N. Gardner, M.P.P. The Project HOPE Center for Health Affairs May 1999 Table of Contents Table of Contents Section I. Purpose and Overview ..........................................................................................1 Section II. Summary of Findings ...........................................................................................5 Key Findings ........................................................................................................................... 6 Implications for Policy .......................................................................................................................8 Section III. Surveying California's Undocumented Latino Immigrants ............................9 Setting the Context — Undocumented Immigrants in the U.S. ......................................................10 The Study Sites — Fresno and Los Angeles Counties ..................................................................11 The Survey and its Implementation ................................................................................................12 Section IV. Characteristics of the Study Population ........................................................13 Age and Gender ..............................................................................................................................14 Family Income .................................................................................................................................16 Labor Force Participation ................................................................................................................18 Section V. Circumstances of Immigration .........................................................................21 Reasons for Immigration .................................................................................................................22 Country of Origin .............................................................................................................................24 Length of Time in the U.S. ..............................................................................................................26 Family Composition — Marital Status............................................................................................28 Family Composition — Children .....................................................................................................30 Section VI. Access to Health Care Coverage and Services.............................................33 Health Insurance Status..................................................................................................................34 Health Status...................................................................................................................................36 Use of Health Care Services — Hospitalizations and Physician Visits ..........................................38 Site of Care for Most Recent Visit...................................................................................................40 Inability to Obtain Health Care ........................................................................................................42 Reasons for Inability to Obtain Health Care ...................................................................................44 English Language Proficiency.........................................................................................................46 Section VII. Participation in Government Programs ......................................................................49 Income Support Programs ..............................................................................................................50 Health and Nutrition Programs........................................................................................................52 Other Government Services ...........................................................................................................54 Section VIII. Appendices. .....................................................................................................57 Appendix A: Sample Design for the Undocumented Immigrant Health Care Access Survey........58 Appendix B: Field Operations for the Undocumented Immigrant Health Care Access Survey .....60 Section IX. References .........................................................................................................63 Table of Contents List of Figures Figure 1: Age of Undocumented Latino Immigrants by County and All Latinos by County.....................................................................................................15 Figure 2: Annual Family Income of Undocumented Latinos by County and of Latinos Nationally .........................................................................................17 Figure 3: Most Important Reason for Immigration, Ages 18 and Older by County ...............23 Figure 4: Country of Origin of Undocumented Latino Immigrants by County .......................25 Figure 5: Length of Time in U.S. for Undocumented Latino Immigrants Ages 18 and Older by County ....................................................................................27 Figure 6: Marital Status of Undocumented Latinos Ages 18 and Older by County .............29 Figure 7: Children of Undocumented Latinos by County ....................................................31 Figure 8: Health Insurance Status of Undocumented Latinos by County, Latinos Nationally, and All Persons Nationally .........................................................35 Figure 9: Self Reported Health Status of Nonelderly Undocumented Latinos by County, and of Nonelderly Latinos of Mexican Origin Nationally..................37 Figure 10: Site of Care for Most Recent Visit: Undocumented Latinos Ages 16 and Older by County ..........................................................................................41 Figure 11: Inability to Obtain Care in the Past Twelve Months, Undocumented Latinos by County and All Persons Nationally .............................................43 Figure 12: Reasons for Inability to Obtain Medical Care: Undocumented Latinos by County and Latinos Nationally .....................................................................45 Figure 13: English Proficiency of Undocumented Latinos Ages 16 and Older by County: Ability to Communicate with a Medical Professional .......................47 Figure 14a: Participation in Income Support Programs by Undocumented Latinos or Family Member .......................................................................................51 Figure 14b: Participation in Health and Nutrition Programs by Undocumented Latinos or Family Member .......................................................................................53 Figure 14c: Participation in Other Government Services by Undocumented Latinos or Family Member .......................................................................................55 List of Tables Table 1: Labor Force Participation (%) During the Last Two Weeks, Persons Ages 18-64, by Sex .....................................................................................19 Table 2: Medical Utilization in the U.S. During the Previous 12 Months, Persons Ages 16 and Older ......................................................................................39 Section I. Purpose and Overview 1 Section I. Purpose and Overview The use of health care services by undocumented persons living in the U.S. and their eligibility for public financing of these and other services has been the subject of heated national debate over the past five years. Much of the controversy began with the introduction and passage of Proposition 187 in California in 1994, an initiative that would have denied undocumented immigrants access to certain state-funded health care services. Proposition 187 was never implemented because it was ruled unconstitutional in Federal district court. At the request of California Governor Gray Davis, a Federal appeals court has agreed to mediate the issue. Services for immigrants have also been limited by the federal welfare reforms contained in the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (along with its subsequent amendments). This legislation limits the use of federal funds for the provision of certain services to legal non-citizen immigrants and, by implication, undocumented persons as well. Those states that wish to provide benefits to undocumented immigrants must pass specific laws to do so. The debate over the provision of publicly-funded services to undocumented immigrants has focused on several issues. Proponents of limiting access to services have argued that undocumented immigrants use large amounts of publicly-funded health care services – burdening state and local governments financially and reducing resources available to other populations. They also contend that government-provided benefits serve as an incentive for immigrants to come to this country. Those opposing limitations on access to health care services argue that immigration is economically motivated so that reducing service availability will not stem immigration. It is also noted that immigrants pay taxes that support publicly-funded services, and that the denial of needed health care will cause unnecessary suffering for the immigrant population and could affect the public health of the broader community. Throughout this debate, empirical evidence has been limited. A number of studies have estimated the costs of providing services to undocumented immigrants.1, 2, 3, 4, 5 Other studies have examined access to health care services for undocumented immigrants, but primarily in one locality or at one institution.6, 7 The purpose of this study was to expand the current level of knowledge about undocumented populations by collecting and analyzing information on the use of health care services by undocumented immigrants from Latin American countries. The study focuses on undocumented Latino immigrants because they represent the largest portion of the undocumented population -- nearly three-quarters of all undocumented immigrants are of Latino origin -- and because of the complexities that would have been involved in interviewing in more than one language. The study addresses such question as: Ø What are the demographic and socio-economic characteristics of undocumented Latinos? Ø To what extent do undocumented Latinos come to the United States for the specific purpose of obtaining more or better medical care? 2 Section I. Purpose and Overview Ø How much health care is used by undocumented Latino immigrants? What barriers to care are faced? Is language an obstacle in seeking care? To what extent is there unmet need for services? Ø To what extent do the undocumented participate in government health and social welfare programs? This information should be useful to policymakers at the state level and has implications for the broader national debate on the provision of services to and the health care needs of non-citizen populations, providing a factual basis for the ongoing debate on these issues. The findings described in this report are based on a 1996-1997 survey of 533 undocumented Latino immigrants living in Los Angeles and Fresno Counties in California. A companion study of undocumented Latinos in Texas, funded by the Robert Wood Johnson Foundation, was conducted in El Paso and Houston in 1996. Respondents in all four sites were identified using a probability sample of residential units within Census tracts comprising a relatively high proportion of foreign-born Latinos. A 25-minute interview was administered in Spanish by professional interviewers in the respondent’s home. The interview collected information on reasons for immigrating and length of time in the U.S., use of health care services, use of other public services, labor force participation, and other relevant factors. The overall response rate obtained in the two California sites was 78 percent. A more detailed description of the study methods is found in Appendices A and B to this report. 3 4 Section II. Summary of Findings Ø Key Findings Ø Implications for Policy 5 Section II. Summary of Findings Key Findings Key findings from the study are: Ø The population of undocumented Latino immigrants in Fresno and Los Angeles Counties was relatively young and divided almost evenly between males and females. Over half of undocumented Latinos were in the 18 to 34 age group, about one- quarter were children under 18, and one percent or fewer were 65 years of age and over. Half of all adults were married and just under half of adults had children who were documented, though the adults themselves were undocumented. Ø Over 90 percent of undocumented Latino immigrants in Fresno and 80 percent of those in Los Angeles immigrated from Mexico, and the majority had lived in the U.S. for more than five years. Ø Although undocumented Latino immigrants have come to the U.S. for economic opportunity, their employment rates were lower than for Latinos nationally. And despite their coming in search of work, the vast majority of undocumented Latinos had poverty-level incomes. Approximately two-thirds of undocumented adult males were working during the two weeks prior to being interviewed, compared to 80 percent of Latinos nationally. Of undocumented Latinos in Fresno and Los Angeles Counties, over three-quarters reported annual family incomes of $10,000 or less. Ø Finding work was reported as the most important reason for entering the U.S.; few indicated that they immigrated to obtain U.S. social services. In Fresno, 63 percent of undocumented Latino immigrants cited the search for employment as their primary reason for coming to the U.S. In Los Angeles, that figure was 56 percent. One percent or fewer of the undocumented Latinos in each site said that they immigrated to the U.S. primarily to take advantage of the social services available. 6 Section II. Summary of Findings Ø While self-reported health status of undocumented Latino immigrants in Los Angeles was comparable to that of Latinos nationally (Fresno reports lower health status), use of physician services was low in both sites, especially Los Angeles. Fewer than half (38%) of undocumented Latino adults in both sites combined had an annual visit to a physician in the U.S., compared to two-thirds of all Latinos in the country and three-quarters of all adults nationwide. In contrast, hospitalization for childbirth was somewhat more common among undocumented Latino immigrants than for Latinos as a whole, which is not surprising given the large proportion of the study population aged 18-34. Ø Most undocumented Latinos were uninsured. Financial barriers were paramount as reasons for inability to obtain care. Sixty-eight percent of undocumented Latinos in Fresno and 84 percent in Los Angeles were uninsured, compared to 35 percent of Latinos nationally and 19 percent of all persons nationally. Ø Communication with non-Spanish speaking medical professionals is a concern. Almost 60 percent of undocumented Latino immigrants in Fresno County and almost 40 percent of those in Los Angeles reported that they were unable to communicate with a medical professional in English. However, affordability and lack of insurance – rather than language barriers – were cited as the primary reasons for inability to obtain health care. Ø Participation in government programs was limited in both Fresno and Los Angeles. One-half of undocumented Latino adults had children (many of whom are themselves U.S. citizens) in public schools. Most of the children were receiving free/reduced price lunches. Other use of government services included the Supplemental Nutrition Program for Women, Infants, and Children (WIC -- for which undocumented immigrants are eligible), Medicaid (for which there was limited eligibility for undocumented immigrants, primarily for non-emergency pregnancy related care and emergency care), and food stamps (for which undocumented persons are not eligible). Again, both Medicaid and food stamps would be legally available to U.S. citizen children of undocumented immigrants. 7 Section II. Summary of Findings Implications for Policy Many of the assumptions that appear to underlie the passage of Proposition 187 in California, as well as subsequent and similar policy directions, are not supported by the findings from this study. The following results may help to inform policy. Ø Obtaining health care does not appear to be an important reason for immigration by Latinos. Regardless of the financial burden imposed by the public provision of health care services for undocumented immigrants, it is unlikely that restrictions on access to health care will lower immigration as long as there are major differences in economic opportunity between the U.S. and other countries. No more than one percent of undocumented Latino immigrants reported that they came to the U.S. to receive social services. Ø Use of discretionary medical care by undocumented Latino immigrants is low. Limiting the use of medical care by undocumented immigrants could have significant health impacts since the care being used is largely childbirth-related or is already at such a minimal level that the care is likely to be truly necessary. Ø A large proportion of undocumented Latino immigrants has children who are legal residents, most of whom are likely to be U.S. citizens. While policymakers may be interested in reducing the financial burden of funding health care services for undocumented immigrants, denying health care to parents of U.S. citizen children has implications for the well-being of those families and children. Ø Spanish-speaking medical personnel are important in ensuring access to health care services for many Latinos. Because a substantial proportion of undocumented Latinos reported being unable to communicate with a medical professional in English, Spanish-speaking medical personnel are critical to ensuring the delivery of appropriate medical care. 8 Section III. Surveying California’s Undocumented Latino Immigrants Ø Setting the Context – Undocumented Immigrants in the U.S. Ø The Study Sites – Fresno and Los Angeles Counties Ø The Survey and its Implementation 9 Section III. Surveying California’s Undocumented Latino Immigrants Setting the Context — Undocumented Immigrants in the U.S. According to the U.S. Immigration and Naturalization Service (INS), about five million undocumented immigrants—1.9 percent of the total U.S. population—were residing in the U.S. as of October 1996. Of this five million, at least 72 percent, or 3.6 million, were estimated to be of Latino origin, representing 12.7 percent of the total U.S. Latino population in 1996.8, 9 The total undocumented population is highly concentrated, with 83 percent living in California, Texas, New York, Florida, Illinois, New Jersey, and Arizona. An estimated two million undocumented immigrants (40 percent of the U.S. total) live in California alone. Mexico is by far the largest country of origin, comprising 54 percent (2.7 million persons) of the total U.S. undocumented population.8 This study focuses on undocumented Latino immigrants living in Fresno and Los Angeles Counties in California. Due to the enormous level of resources that would have been required to identify a statewide target population and the interest in variation across different communities in California, we selected these two distinct geographic areas. The study sample was limited to Latino immigrants because of the size and importance of this population and the complications inherent in interviewing persons in more than one language. Latino designation was determined at the time of household screening, based on self-reporting by respondents. The survey was designed to gather information on the socio-economic characteristics and patterns of health care access and utilization of undocumented Latino immigrants living in these two communities, as well as factors related to their decisions to enter the U.S. 10 Section III. Surveying California’s Undocumented Latino Immigrants The Study Sites — Fresno and Los Angeles Counties The two California sites were selected to represent some of the diversity within the undocumented population; Fresno County was chosen because of its large agricultural sector and Los Angeles County was selected because it contains the largest concentration of undocumented immigrants in California. The most recent estimates available, although somewhat old today, show that approximately 70 percent of California’s undocumented population lived in these two areas in 1980.10 While more recent estimates of the undocumented population at the county or metropolitan area level are not available, it is believed that the percentages of undocumented persons residing in Fresno and Los Angeles Counties are approximately the same today as they were in 1980. In terms of population, Los Angeles County is the largest in the nation. As of January 1997, Los Angeles County had a population of about 9.5 million persons, of which its largest city, Los Angeles, comprised approximately 3.7 million. As of January 1997, Fresno County had a population of about 776,000 persons, with 407,000 in the city of Fresno.11 Los Angeles County has, by far, the largest Latino population of any county in the United States, with about four million persons, accounting for 44 percent of its overall population. Fresno County’s Latino population, about 42 percent of its total population, places it 19th among counties in the United States.12 In 1996, per capita personal income in Los Angeles County was $24,945, and in Fresno County it was $19,012.13 Fresno is the largest agricultural producing county in California. In 1992, the county had 7,021 farms encompassing nearly 47 percent of the county's land area. In 1996, the value of its agricultural production was $3,313.4 million, 13.3 percent of the state’s total agricultural production.13 11 Section III. Surveying California’s Undocumented Latino Immigrants The Survey and its Implementation The study represents a major contribution in terms of the scientific methods used to implement the survey. This was the first attempt to use probability sampling of the undocumented population along with in-person interviewing in a multi-site survey. Probability sampling is defined by each member of the group of interest having a known probability of being selected for an interview. In practical terms, this means that the persons interviewed are representative of all undocumented Latino immigrants in the geographic sites; thus, we have reached a more diverse population than in other studies, including people who use more than one health care facility as well as those who use no services at all. The use of in-person interviewing is significant because approximately one-third of undocumented immigrants interviewed did not have telephones (and thus would have been excluded from a telephone survey). Due to the sensitive nature of the information being obtained, in-person interviewing is also critical to establishing trust between interviewer and respondent. The survey sample design was designed to produce a representative probability sample of approximately 400 undocumented Latino immigrants living in Fresno County and Los Angeles County—200 completed interviews in each site. A total of 533 interviews were completed, 256 in Fresno County and 277 in Los Angeles County. Housing units were sampled from comprehensive address listings within Census blocks containing a relatively high proportion of foreign-born Latinos. Sampled housing units were then screened for eligible respondents. To conduct the highly sensitive task of determining respondent eligibility, a two-step in-person screening process was used. First, interviewers were carefully trained to quickly establish a good rapport with a household member. During this initial informal conversation, the interviewer determined whether at least one household member did not have “papers” to legally reside in this country. If this appeared to be the case, the interviewer then used a carefully designed screening form and statistical formula to enumerate all household members and to select one or more eligible respondents to complete the questionnaire. Eligible household members were defined as Latino persons who were not born in the U.S. were not naturalized U.S. citizens, or did not hold a green card. Immigration status was elicited through a process of elimination (based on answers to a series of questions designed to identify undocumented persons), rather than by directly asking the screener respondent if each member of the household was undocumented. Once respondents were selected from the eligible household members, the field staff immediately attempted to interview those persons. Data collection occurred between October 1996 and July 1997, although the length of the field period varied by site. Data were collected through a 25-minute in-person interview conducted in Spanish by trained professional interviewers. The overall response rate was 78 percent. Field supervisors worked very closely with the interviewing staff to ensure that all survey procedures were carried out as required. In both sites, 15 to 20 percent of each interviewer’s work was verified. A more detailed description of the study methods can be found in Appendices A and B to this report. 12 Section IV. Characteristics of the Study Population Ø Age and Gender Ø Family Income Ø Labor Force Participation 13 Section IV. Characteristics of the Study Population Age and Gender Undocumented Latino immigrants in Fresno and Los Angeles Counties were younger than all Latinos in those counties, with the majority in the 18 to 34 age group and few who were elderly. Males and females were evenly represented. Ø The majority of undocumented Latino immigrants in Fresno and Los Angeles were between the ages of 18 and 34 — 57 percent in Fresno and 54 percent in Los Angeles (Figure 1). This compares to only 25 and 26 percent, respectively, of all Latinos in that age group in the two counties. Ø About one-quarter of undocumented Latino immigrants were children under age 18 (27% in both sites). Ø One percent or fewer of the undocumented Latinos in each of the study sites were 65 years of age or older, while 10 percent of all Latinos in the two sites were elderly. Ø Males and females were represented about evenly in the population of undocumented Latino immigrants in both Fresno and Los Angeles. Among all Latinos in the two study sites, males and females were also represented relatively evenly (data not shown). 14 14 Section IV. Characteristics of the Study Population Figure 1. Age of Undocumented Latino Immigrants by County and All Latinos by County Fresno County 70% 60% 57% 50% 40% 33% 32% 30% 27% 25% 20% 15% 10% 10% 1% 0% 0-17 Yrs 18-34 Yrs 35-64 Yrs 65+ Yrs Undocumented Latinos All Latinos in Fresno County* Los Angeles County 70% 60% 54% 50% 40% 35% 27% 29% 30% 26% 19% 20% 10% 10% <1% 0% 0-17 Yrs 18-34 Yrs 35-64 Yrs 65+ Yrs Undocumented Latinos All Latinos in Los Angeles County* SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTE: *State of California, Department of Finance, Race/Ethnic Population Estimates with Age and Sex Detail, 1996 . Sacramento, CA, January 1998. 15 Section IV. Characteristics of the Study Population Family Income Although undocumented Latino immigrants reported work as the most important reason for entering the U.S. illegally (see Figure 3), most still have poverty level family incomes. Ø A substantial majority of undocumented Latino immigrants in each site reported annual family incomes of $10,000 or less -- 83 percent in Fresno County and 78 percent in Los Angeles County. This compares to only 15 percent of Latinos nationwide (Figure 2). Ø Only 4 percent of the study population in Fresno and 10 percent in Los Angeles reported an annual family income greater than $15,000. Nationally, nearly 75 percent of all Latinos have a family income of $15,000 or more. 16 Section IV. Characteristics of the Study Population Figure 2. Annual Family Income of Undocumented Latinos by County and of Latinos Nationally Fresno County Los Angeles County 4% 10% 14% 27% 37% 13% 46% 51% $5,000 or less $5,000 or less $5,001 to $10,000 $5,001 to $10,000 $10,001 to $15,000 $10,001 to $15,000 $15,001 or more $15,001 or more U.S. Latinos* 5% 10% 12% 74% less than $5,000 $5,000 to $9,999 $10,000 to $14,999 $15,000 or more SOURCES: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. *U.S. Bureau of the Census, Ethnic and Hispanic Statistics Branch, Current Population Survey, 1996. NOTE: The definition of family income in the Project HOPE survey and the Current Population Survey vary somewhat; however, this is not expected to substantially affect the distributions. In addition, due to differences in the reporting of the data, the income categories vary slightly. 17 Section IV. Characteristics of the Study Population Labor Force Participation Although they have come to the U.S. to find work, many undocumented Latino immigrants have been unsuccessful in these efforts. In both study sites, labor force participation rates are below those of Latinos nationally. Ø A substantial percentage of undocumented Latinos in Fresno and Los Angeles reported having no job in the two weeks prior to being interviewed. In Fresno, 36 percent of males and 85 percent of females had no job; 25 percent of males and 82 percent of females in Los Angeles had no job in the 2-week period. Nationally, a much lower proportion of Latinos were without work -- 18 percent of Latino males and 44 percent of Latino females reported not having a job (Table 1). Ø Sixty-one percent of undocumented Latino males in Fresno worked at least some of the time during the two weeks prior to the interview. Similarly, 71 percent of undocumented Latino males in Los Angeles worked during the previous two weeks. On the national level, 80 percent of Latino males worked during a comparable two-week period. Ø Of those not working in the 2-week period prior to being interviewed, seventy-five percent of undocumented Latino men in Fresno and Los Angeles reported that they were looking for work (data not shown). 18 Section IV. Characteristics of the Study Population Table 1. Labor Force Participation (%) During the Last Two Weeks, Persons Ages 18-64, by Sex Employment Status* Worked Anywhere No Job Fresno Male 61% 36% Female 12% 85% Los Angeles Male 71% 25% Female 17% 82% Latinos Nationally** Male 80% 18% Female 54% 44% SOURCES: Project HOPE tabulations from the 1996-1997 Undocumented Immigrant Health Care Access Survey. **1994 National Health Interview Survey. NOTE: *Rows do not sum to 100% because of a small number of persons who had a job but did not work during the 2-week period. This may have been due to sickness, vacation, temporary layoff, or some other reason. 19 20 Section V. Circumstances Of Immigration Ø Reasons for Immigration Ø Country of Origin Ø Length of Time in the U.S. Ø Family Composition – Marital Status Ø Family Composition – Children 21 Section V. Circumstances Of Immigration Reasons for Immigration Economic betterment appears to be the main motivating factor for immigration -- undocumented Latino immigrants in both Fresno and Los Angeles Counties reported the search for employment as their primary reason for coming to the U.S. Ø Among undocumented Latinos ages 18 and older in Fresno County, 63 percent reported that the search for employment was the most important reason they entered the U.S. Slightly fewer, 56 percent, of those living in Los Angeles County said the most important reason they entered the U.S. was to find work (Figure 3). Ø Approximately one-third of the undocumented Latino adult population from Fresno and from Los Angeles cited the desire to be with family or friends as their primary motivation for entering the country. Ø One percent or fewer of the undocumented Latino adults in each site immigrated to the U.S. primarily to take advantage of the social services available. 22 Section V. Circumstances Of Immigration Figure 3. Most Important Reason for Immigration, Ages 18 and Older by County Fresno County 70% 63% 60% 50% 40% 30% 30% 20% 10% 3% 2% 0% 1% 0% Work Family/Friends Education Political Social Services Other Los Angeles County 70% 60% 56% 50% 40% 33% 30% 20% 10% 4% 4% 2% 1% 0% Work Family/Friends Education Political Social Services Other SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 23 Section V. Circumstances Of Immigration Country of Origin The vast majority of undocumented Latino immigrants in both survey sites were from Mexico; a somewhat more diverse population in terms of country of origin was represented in Los Angeles than in Fresno. Ø About 94 percent of the undocumented Latinos in Fresno County reported Mexico as their country of origin, compared to 80 percent in Los Angeles County (Figure 4). Ø In Fresno, persons from El Salvador comprised 4 percent of the study population and those from Chile represented 1 percent. In Los Angeles, 10 percent of the undocumented Latino immigrants identified El Salvador as their country of origin, 4 percent Nicaragua, and 3 percent Chile. Ø Data at the national level on all undocumented immigrants (not limited to Latinos) show a similar country of origin distribution. Mexico contributes the majority of all undocumented immigrants (54 percent), followed by El Salvador (7 percent), and Guatemala (3 percent) (data not shown).8 24 Section V. Circumstances Of Immigration Figure 4. Country of Origin of Undocumented Latino Immigrants by County Fresno County El Salvador 4% Chile 1% Mexico Other 94% 1% Los Angeles County Nicaragua Other 4% Chile 3% 3% El Salvador 10% Mexico 80% SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 25 Section V. Circumstances Of Immigration Length of Time in the U.S.* The majority of the undocumented Latino adult population in both Fresno and Los Angeles Counties has resided in the U.S. more than five years (60 percent in Fresno and 65 percent in Los Angeles) (Figure 5). * Survey eligibility was limited to persons who had resided in the U.S. for 6 months or longer. 26 Section V. Circumstances Of Immigration Figure 5. Length of Time in U.S. for Undocumented Latino Immigrants Ages 18 and Older by County Fresno County Less than 1 year 7% More than 1 to 5 years 5 years 34% 60% Los Angeles County Less than 1 year 4% More than 1 to 5 years 5 years 31% 65% SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 27 Section V. Circumstances Of Immigration Family Composition – Marital Status About half of the undocumented Latino adult population in both Fresno and Los Angeles Counties were married, most with spouses living in the U.S. About one-third at both sites were single. Ø In Fresno County, 52 percent of undocumented Latino immigrants ages 18 and older were married. Of those, over four-fifths had spouses living in the U.S. Almost half of Los Angeles County undocumented Latinos ages 18 and older reported they were married, with almost all spouses living in the U.S. (Figure 6). On the national level, a somewhat higher proportion of Latinos ages 18 and older were married — 62 percent (data not shown). * Ø Approximately one-third of the adults in each site indicated they were single (Figure 6), compared to 24 percent of all Latinos on the national level (data not shown). * 1994 National Health Interview Survey. 28 Section V. Circumstances Of Immigration Figure 6. Marital Status of Undocumented Latinos Ages 18 and Older by County Fresno County 8% Married/ 3% Divorced or Spouse Separated not in U.S. 1% Widowed 32% Single 44% Married/ Spouse in U.S. 12% Living with Partner Los Angeles County 5% Divorced or 2% Married/ Separated Spouse not in U.S. 1% Widowed 35% Single 45% Married/ Spouse in U.S. 12% Living with Partner SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 29 Section V. Circumstances Of Immigration Family Composition – Children Almost half of undocumented Latino adults in both Fresno and Los Angeles reported at least one documented child in the U.S. Many undocumented Latino adults reported having children with differing documentation statuses, and sometimes indicated they had children living both inside and outside the U.S. Ø Forty-nine percent of Fresno’s undocumented adults had at least one documented child in the U.S. In Los Angeles County, this proportion was similar — 42 percent (Figure 7). Ø Thirty-four percent of the undocumented Latino population ages 18 and older in Fresno County reported they did not have any children, compared to 41 percent in Los Angeles County. Ø Ten percent of the undocumented Latino immigrants in Fresno County and 7 percent of those in Los Angeles County reported that all of their children were living outside the U.S. 30 Section V. Circumstances Of Immigration Figure 7. Children of Undocumented Latinos by County Fresno County Other** 7% No children 34% Documented children in U.S.* 49% All children not in U.S. 10% Los Angeles County Other** 10% No children 41% Documented children in U.S.* 42% All children not in U.S. 7% SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTES: *Of those respondents who reported a documented child in the U.S., 29 percent in Fresno and 22 percent in Los Angeles reported they also had an undocumented child in the U.S. and/or children not residing in the U.S. **Includes only respondents who reported ALL their children residing in the U.S. were undocumented or respondents who reported they had undocumented children residing in the U.S. AND one or more children residing outside the U.S. 31 32 Section VI. Access to Health Care Coverage and Services Ø Health Insurance Status Ø Health Status Ø Use of Health Care Services – Hospitalizations and Physician Visits Ø Site of Care for Most Recent Visit Ø Inability to Obtain Care Ø Reasons for Inability to Obtain Care Ø English Language Proficiency 33 Section VI. Access to Health Care Coverage and Services Health Insurance Status Extremely low rates of health insurance coverage were found among undocumented Latino immigrants in Fresno and Los Angeles Counties. Coverage is related to immigration status through the types of jobs available to the undocumented (i.e., those less likely to offer benefits) and the lack of eligibility for public insurance. Ø Sixty-eight percent of undocumented Latino immigrants in Fresno under the age of 65 and 84 percent in Los Angeles reported they had no health insurance. In comparison, on a national basis, 35 percent of Latinos under age 65 and 19 percent of all persons under age 65 had no health insurance (Figure 8). Ø In both counties, very few undocumented Latino immigrants had private health insurance (5 percent in Fresno, 4 percent in Los Angeles) compared with 45 percent of Latinos nationally (mostly employment-based) and 69 percent of all persons nationally. Ø Approximately one-quarter (27%) of undocumented Latinos in Fresno and 12 percent of those in Los Angeles reported coverage through a publicly-financed program; most were covered through Medi-Cal, California’s Medicaid program (26% in Fresno and 10% in Los Angeles).∗ ∗ See Section VII for a discussion of eligibility and participation. 34 Section VI. Access to Health Care Coverage and Services Figure 8. Health Insurance Status of Undocumented Latinos by County, Latinos Nationally, and All Persons Nationally Persons Under 65 Years of Age 100% 90% 84% 80% 68% 69% 70% 60% 50% 45% 40% 35% 30% 27% 20% 19% 20% 12% 12% 10% 5% 4% 0% Fresno Los Angeles Latinos Nationally* All Persons Nationally* Any Private Public Only Uninsured SOURCES: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. *Center for Cost and Financing Studies, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1996 (Round 1) in Vistnes J, Monheit A. Health Insurance Status of the Civilian Noninstitutionalized Population: 1996 . Agency for Health Care Policy and Research Pub. No. 97-0030, Table 2, 1997. NOTE: Of those with Public Only coverage, 26% in Fresno and 10% in Los Angeles have coverage through Medi-Cal, California's Medicaid program. 35 Section VI. Access to Health Care Coverage and Services Health Status In Los Angeles, self-reported health status among undocumented immigrants younger than 65 years old appears to be fairly comparable to that of Latinos of Mexican origin nationally;* undocumented immigrants in Fresno, however, report being less healthy. Ø Nationally, 41 percent of Latinos† under the age of 65 described themselves as being in excellent or very good health. Similarly, 34 percent of undocumented Latinos in Los Angeles described their health as excellent or very good. In Fresno, however, only 12 percent reported themselves as being in excellent or very good health (Figure 9). Ø The percentage of undocumented Latinos in both study sites who reported themselves as being in good health was virtually the same—40 percent in Fresno and 41 percent in Los Angeles. However, those in Fresno were much more likely (48 percent) to describe their health status as fair or poor compared to those in Los Angeles (25 percent) or Latinos nationally (23 percent). * Because the Project HOPE survey of undocumented immigrants was conducted entirely in Spanish, comparisons to estimates from other surveys, which may have been conducted in English or with the aid of Spanish-speaking translators, should be interpreted with caution. † National data are for Latinos of Mexican origin only. 36 Section VI. Access to Health Care Coverage and Services Figure 9. Self Reported Health Status of Nonelderly Undocumented Latinos By County, and ofand of Latinos of Mexican OriginOrigin Nationally by County, Nonelderly Latinos of Mexican Nationally Persons Under 65 Years of Age 60% 50% 48% 41% 41% 40% 40% 36% 34% 30% 25% 23% 20% 12% 10% 0% Excellent / Very Good Good Fair / Poor Fresno Los Angeles All U.S. of Mexican Origin* SOURCES: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. *National Health and Nutrition Examination Survey III, 1988-1994. 37 Section VI. Access to Health Care Coverage and Services Use of Health Care Services – Hospitalizations and Physician Visits Hospitalizations. Overall rates of hospitalization were roughly comparable for undocumented Latinos in the study sites and all U.S. residents. Over one-third of all hospitalizations for undocumented Latinos were for childbirth. Ø Twelve percent of undocumented Latino adults* in Fresno County and 6.8 percent in Los Angeles County reported they had been hospitalized at least once overnight or longer in the U.S. during the last 12 months. Nationally, 8.5 percent of adult Latinos were hospitalized in the previous 12-month period (Table 2). Ø The hospitalization rate for childbirth (including normal births) among the undocumented Latino immigrants in the two study sites was twice as high as the rate for all women in the U.S. This high rate is not surprising given the high proportion of the study population aged 18-34. Physician Visits, All Ambulatory Settings. Undocumented Latino adults used substantially fewer physician services than Latino adults nationally or all adults nationally. Ø Twenty-one percent of undocumented Latino adults in Fresno and 32 percent of those in Los Angeles reported their most recent visit to a medical practitioner occurred more than two years prior to being interviewed (data not shown). Ø In Fresno County, half of undocumented Latinos ages 16 and older had a visit to a physician in the U.S. during the last 12 months. In contrast, about 27 percent of the undocumented Latino adults in Los Angeles County reported a physician visit in the U.S. during that time. For the adult Latino population nationally, 66 percent reported a physician visit in the U.S. during the prior 12-month period and that proportion was 75 percent for all U.S. adults (Table 2). Ø For those undocumented Latino adults who reported a physician visit during the past 12 months, the mean number of visits in the U.S. was 4.3 in Fresno and 3.2 in Los Angeles. The national average for all Latino adults was 6.2 visits per person during a 12-month period; for all U.S. adults it was also 6.2 visits. Ø In the twelve months prior to being surveyed, 13 percent of undocumented Latino adults in Fresno County and 3 percent of those in Los Angeles County had visited a curandero (faith healer or shaman) for medical care (data not shown). * The adult version of the questionnaire was administered to persons ages 16 and older. A comparable age cutoff was used for national estimates. 38 Section VI. Access to Health Care Coverage and Services Table 2. Medical Utilization in the U.S. During the Previous 12 Months, Persons Ages 16 and Older Combined All U.S. Total U.S. Fresno Los Angeles Sites Latinos* Population* Hospitalized (%) Any Reason 9.3% 12.0% 6.8% 8.5% 8.9% †,‡ † † Childbirth 3.5% 3.4% 3.5% 2.6% 1.7% ‡,§ All Other 6.1% 9.2% 3.3% 6.0% 7.4% Physician Visits Percent with ‡, § ‡,§ ‡,§ 38.2% 49.9% 27.2% 65.8% 74.8% Physician Visit Mean Number of ‡, § ‡, § ‡, § 3.9 4.3 3.2 6.2 6.2 Physician Visits for those with at least one visit SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTES: *1994 National Health Interview Survey. † Standard error for these estimates is high. ‡ Different from Total U.S. Population at .05 § Different from All U.S. Latinos at .05 Hospitalizations for childbirth and all other reasons are not mutually exclusive and therefore may not sum to the overall percent hospitalized. 39 Section VI. Access to Health Care Coverage and Services Site of Care for Most Recent Visit* Most undocumented Latino adults visited a clinic or health center for their most recent medical visit. The next most likely site of care was a hospital emergency room, followed by a private physician’s office. Ø Among undocumented Latino adults,** 61 percent in Fresno County and 82 percent in Los Angeles County reported that their most recent visit was to a clinic or health center (Figure 10). Ø In Fresno, 21 percent of those with a visit indicated they went to an emergency room for treatment; 11 percent in Los Angeles used a hospital emergency room. Fewer went to a private physician’s office (18% in Fresno and 8% in Los Angeles). * Data shown for those with visit during prior 12 months. ** Includes persons 16 years of age and older. 40 Section VI. Access to Health Care Coverage and Services Figure 10. Site of Care for Most Recent Visit: Undocumented Latinos Ages 16 and Older by County 100% 90% 82% 80% 70% 61% 60% 50% 40% 30% 21% 20% 18% 11% 10% 8% 0% Hospital Clinic or Health Private Physician Emergency Room Center Office Fresno Los Angeles SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 41 Section VI. Access to Health Care Coverage and Services Inability to Obtain Health Care* Study results suggest undocumented Latino immigrants do not report high levels of unmet health care needs. However, given the relatively low use of health care services and the high poverty rates, it is likely that low reports of inability to obtain care are influenced by lower expectations about obtaining care and by different assessments about what constitutes need. 15 Ø In Fresno County, 9 percent of the undocumented Latino immigrants reported they were unable to obtain medical care in the past 12 months, compared to 1 percent in Los Angeles (Figure 11). Ø In Fresno County, 8 percent of undocumented Latinos reported they were unable to obtain dental care and 7 percent were unable to get eyeglasses within the last 12 months. In Los Angeles County, 6 percent indicated they were unable to obtain dental care while only 2 percent were unable to get eyeglasses. Ø Three percent of the study population in Fresno and two percent in Los Angeles reported a time when they were unable to get a drug prescription filled. Ø Reports on inability to obtain care at the national level are similar to those for undocumented Latinos in Fresno. Undocumented persons in Los Angeles reported somewhat less unmet health care needs. * Based on responses to the following question: “In the past 12 months, was there a time when you were unable to obtain the medical care you thought you needed?” Similar questions were asked with respect to dental care, eyeglasses, and prescription drugs. 42 Section VI. Access to Health Care Coverage and Services Figure 11. Inability to Obtain Health Care in the Past Twelve Months, Undocumented Latinos by County and All Persons Nationally 15% 10% 9% 9% 8% 7% 6% 6% 5% 5% 5% 3% 2% 2% 1% 0% Medical Care Dental Care Prescription Filled Eyeglasses Fresno Los Angeles National* SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTE: *Project HOPE tabulations from the 1994 Robert Wood Johnson Foundation National Access to Care Survey. 43 Section VI. Access to Health Care Coverage and Services Reasons for Inability to Obtain Health Care Financial hardship (including affordability and lack of insurance) represented the most commonly cited barrier to health care for undocumented Latino immigrants. These reasons were similar for all Latinos in the U.S. Ø In both Fresno (86 percent) and Los Angeles (100 percent), undocumented Latino immigrants cited financial reasons (affordability and lack of insurance) as the primary barriers to obtaining medical attention (Figure 12). Ø More specifically, in Fresno County, 63 percent of undocumented Latino immigrants indicated they could not afford the needed care and 23 percent cited lack of health insurance as the main barrier to obtaining care. In Los Angeles County, 86 percent of the undocumented Latino immigrants said they could not afford the care and 14 percent reported lack of health insurance as an obstacle. Ø In Fresno, 14 percent of the study population specified other, non-financial reasons for not seeking care, including fear due to immigration status, lack of transportation, lack of time, lack of availability of the needed service, or long clinic waiting times. Fewer than one percent of undocumented Latino immigrants in Los Angeles cited other non-financial reasons. No one in Fresno or Los Angeles reported language as a barrier to obtaining medical care. 44 Section VI. Access to Health Care Coverage and Services Figure 12. Reasons for Inability to Obtain Health Care: Undocumented Latinos by County and Latinos Nationally Fresno County Los Angeles County 100% 100% 86% 80% 80% 63% 60% 60% 40% 40% 23% 20% 14% 20% 14% <1% 0% 0% Could Not No Other Could Not No Other Afford Insurance Reasons Afford Insurance Reasons Latinos Nationally* 100% 80% 69% 60% 40% 20% 16% 15% 0% Insurance-related Other Inability to 1 2 reasons problems afford care SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTES: *Medical Expenditure Panel Survey Household Component, 1996 (Rounds 1 and 2) in Weinick RM, Zuvekas SH, Drilea S. Access to Health Care--Sources and Barriers, 1996. Agency for Health Care Policy and Research, Table 4, 1996. 1. Includes: insurance company wouldn't approve, cover, or pay for care; pre-existing condition; insurance required a referral but couldn't get one; doctor refused to accept family's insurance plan. 2. Includes: transportation problems; physical problems (e.g., building access, medical equipment in office); communication problems (e.g. work-related, waiting time, did not know where to go). 45 Section VI. Access to Health Care Coverage and Services English Language Proficiency While language problems are not cited by undocumented Latino immigrants as reasons for their inability to obtain care, communication with non-Spanish speaking medical professionals is a concern. Ø A substantial percentage of the undocumented Latino adults* in both sites, 59 percent in Fresno and 39 percent in Los Angeles, reported they were unable to communicate with a medical professional in English (Figure 13). Ø Only 11 percent of the undocumented Latino adults in Fresno County indicated they could speak English very well or well when communicating with a medical professional. In Los Angeles, 15 percent indicated they could communicate very well or well in English with a medical professional. * Includes persons ages 16 and older. 46 Section VI. Access to Health Care Coverage and Services Figure 13. English Proficiency of Undocumented Latinos Ages 16 and Older by County: Ability to Communicate with a Medical Professional Fresno County 70% 59% 60% 50% 40% 31% 30% 20% 9% 10% 2% 0% Very well Well Not very well Not at all Los Angeles County 70% 60% 50% 46% 39% 40% 30% 20% 12% 10% 3% 0% Very well Well Not very well Not at all SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. 47 48 Section VII. Participation in Government Programs∗ Ø Income Support Programs Ø Health and Nutrition Programs Ø Other Government Services ∗ Medicaid participation was asked of the respondent only, so estimates are for the undocumented population and therefore do not reflect participation by other family members (e.g., children) who may be U.S. citizens or documented residents. All other services were asked in reference to the respondent or family members (including spouse or partner and their children under age 18 residing in the household). These latter estimates may include family members who are documented residents or U.S. citizens. 49 Section VII. Participation in Government Programs Income Support Programs Use of income support programs by undocumented Latino immigrants or their family members was limited. Eligibility for such income support programs as Aid to Families with Dependent Children (AFDC), Social Security, and Supplemental Security Income (SSI): Undocumented immigrants are not eligible for any of these income support programs. Children of undocumented immigrants who are themselves U.S. citizens may be eligible. Ø Nine percent of undocumented Latinos residing in Fresno and 18 percent of those in Los Angeles indicated they or a member of their family currently received Aid to Families with Dependent Children (AFDC) benefits* (Figure 14a). These families may include children who are U.S. citizens. Ø For other income support programs, including Social Security and Supplemental Security Income (SSI), reported participation rates among undocumented Latino immigrants ranged from 1 to 4 percent in Fresno and Los Angeles. * The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 consolidated three federal-state match-grant programs, Aid to Families with Dependent Children (AFDC), Emergency Assistance (EA), and the Job Opportunities and Basic Skills (JOBS) training program, into one block grant program. The new program, Temporary Assistance to Needy Families (TANF), gives states considerable spending flexibility, but also imposes new work requirements and time limits for welfare recipients. As was the case with AFDC, undocumented immigrants are not eligible for TANF. All interviews were conducted prior to TANF’s implementation on July 1, 1997. 50 Section VII. Participation in Government Programs Figure 14a. Participation in Income Support Programs by Undocumented Latinos or Family Member* Fresno County 60% 50% 40% 30% 20% 9% 10% 4% 2% 1% 0% AFDC Social SSI Other Security Los Angeles County 60% 50% 40% 30% 18% 20% 10% <1% <1% <1% 0% AFDC Social SSI Other Security SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTE: * Family is defined as the respondent, the respondent’s spouse or partner, and their children under age 18 living in one household. Medicaid participation was asked of the respondent only. All other services were asked in reference to the respondent or family members residing in the household. Family may include persons who are lawful permanent residents or U.S. citizens. 51 Section VII. Participation in Government Programs Health and Nutrition Programs No more than about one-quarter of undocumented Latino immigrants or their family members used certain health and nutrition programs. Eligibility for such health and nutrition programs as Medicaid, Food Stamps, and Supplemental Nutrition Program for Women, Infants, and Children (WIC): Undocumented immigrants are not eligible for participation in the Medicaid or Food Stamp programs (exceptions noted below),* although U.S. citizen children may be. The WIC program is available regardless of immigration status. Ø In Fresno County, 26 percent of the undocumented Latinos indicated they were covered by Medicaid. A considerably lower proportion in Los Angeles County, 10 percent, indicated they were a Medicaid beneficiary (Figure 14b).* Ø Eighteen percent of the undocumented Latinos in Fresno County and 11 percent in Los Angeles County reported they or a member of their family were currently receiving Food Stamps. Participation in the WIC program was more prevalent; 25 percent of the study population in both Fresno and Los Angeles reported that they or a family member were currently receiving WIC benefits. Ø In both Fresno and Los Angeles, the study population reported greater participation in health and nutrition programs than in income support programs. * With the passage of the Federal Personal Responsibility and Work Opportunity Reconciliation Act of 1996 and amended sections of the legislation, undocumented immigrants are ineligible for federally-funded health care with the exception of emergency services (including labor and delivery), public health immunizations, and testing and treatment of communicable diseases. State legislatures must pass specific legislation to provide additional health care services with state funds, which California has not done. Prior to passage of this law (during the fielding of this survey), California provided non-emergency pregnancy-related care, including prenatal care, labor, delivery and postpartum care, as a state-only funded benefit under the Medi-Cal program. 52 Section VII. Participation in Government Programs Figure 14b. Participation in Health and Nutrition Programs by Undocumented Latinos or Family Member* Fresno County 60% 50% 40% 30% 26% 25% 18% 20% 10% 1% 0% Medicaid* Food Stamps WIC Other Los Angeles County 60% 50% 40% 30% 25% 20% 10% 11% 10% <1% 0% Medicaid* Food Stamps WIC Other SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTE: * Family is defined as the respondent, the respondent’s spouse or partner, and their children under age 18 living in one household. Medicaid participation was asked of the respondent only. All other services were asked in reference to the respondent or family members residing in the household. Family members may include persons who are lawful permanent residents or U.S. citizens. 53 Section VII. Participation in Government Programs Other Government Services About half of undocumented Latino immigrants or their family members used public schools and free/reduced price lunches. Eligibility for other government programs such as public schools, the free/reduced lunch program, and subsidized housing: A public school education is available to all persons residing in the U.S., irrespective of documentation status. Any low-income child attending a school participating in the National School Lunch Program may also be eligible for free or reduced priced meals at school. Undocumented immigrants are not eligible for federal public housing assistance. Ø Undocumented Latinos in both study sites utilized the public school system more than any other type of government program. Fifty percent of the undocumented Latino immigrants* in both Fresno and Los Angeles reported they had at least one child currently enrolled in a public school (Figure 14c). Many of these children may be U.S. citizens. Ø Overall, 46 percent of the undocumented Latinos* in both Fresno County and Los Angeles County reported they had children receiving free or reduced-price meals at school. Ø Finally, 4 percent of undocumented Latinos in Fresno and 2 percent in Los Angeles reported living in government-subsidized housing. * Includes undocumented Latino immigrants with and without children. 54 Section VII. Participation in Government Programs Figure 14c. Participation in Other Government Services by Undocumented Latinos or Family Member* Fresno County 60% 50% 50% 46% 40% 30% 20% 10% 4% 0% Public Free/Reduced Subsidized Schools** Price Lunches** Housing Los Angeles County 60% 50% 50% 46% 40% 30% 20% 10% 2% 0% Public Free/Reduced Subsidized Schools** Price Lunches** Housing SOURCE: Project HOPE tabulations from the 1996-97 Undocumented Immigrant Health Care Access Survey. NOTE: * Family is defined as the respondent, the respondent’s spouse or partner, and their children under age 18 living in one household. Medicaid participation was asked of the respondent only. All other services were asked in reference to the respondent or family members residing in the household. Family members may include persons who are lawful permanent residents or U.S. citizens. ** Percentages are of the total surveyed population, not just those with children. 55 56 Section VIII. Appendices Ø Appendix A. Sample Design for the Undocumented Immigrant Health Care Access Survey Ø Appendix B. Field Operations for the Undocumented Immigrant Health Care Access Survey 57 Appendix A Sample Design for the Undocumented Immigrant Health Care Access Survey The sample design was developed to produce a representative probability sample of approximately 400 undocumented Latino immigrants living in two separate geographic areas in California — Fresno County and Los Angeles County. The two sites were selected based on their likelihood of containing a high proportion of the target population, while also being somewhat demographically dissimilar to each other. Sampling was conducted in more than one locality within the state in order to find differing socio-economic and labor market conditions. Los Angeles was selected to cover the largest concentration of undocumented immigrants in the state, and Fresno was chosen for its agricultural sector and recent rapid growth. Within each site, Census block groups were stratified based on population and household characteristics from 1990 Census data. We made the assumption that block groups which contained at least 20 percent foreign-born persons and at least 20 percent Spanish linguistically isolated households* were likely to yield the highest concentrations of our target population, and thus these block groups were defined as eligible for block group sampling. While this decision excludes from the sample a presumably small but unknown number of undocumented persons living in areas with low proportions of foreign-born persons and Spanish-speaking households, the approach had dramatic cost implications since the majority of neighborhoods could be excluded from the survey. Two-stage probability proportional to size (PPS) sampling was used to make block group and housing unit selections, using the number of housing units in each block group (from 1990 Census data) as the measure of size. First, a sample of 50 qualifying block groups from each site was selected. Second, listings of all housing units in each of the selected block groups were compiled both from address directories and by in-person canvassing of the blocks. From these address listings, housing units were sampled and screened for eligible respondents. Eligibility was established based on the following criteria (both self-reported): Ø Of Latino origin, and Ø Residing in the U.S. illegally for six months or longer (not born in the U.S., not naturalized U.S. citizens, or not green cardholders). * For a household to be counted as Spanish linguistically isolated, Spanish must be spoken in the household and there can be no one living in the household age 14 years or older who speaks only English or who speaks English very well. 58 Appendix A To complete the highly sensitive task of establishing eligibility, a two-step in-person screening process was employed. First, interviewers were carefully trained to establish rapport with a household member. Through this rapport-building informal conversation, interviewers determined if at least one household member did not have “papers” to legally reside in this country. Next, if it appeared that one or more household members were thus eligible to participate in the survey, the interviewer immediately used the household screener to enumerate all household members and to select one or more eligible respondents to complete the questionnaire. Households were enumerated in such a way that multiple family units within the household could be determined and respondents were listed in age order within family units. (A family unit was defined as an adult household member, his/her spouse or unmarried partner, and their children or dependents under age 18 living in the household.) Within each family unit in the household, one eligible person was selected to be interviewed using the Kish method of respondent selection. Once eligible respondents were selected, the field staff immediately attempted to interview those persons. Field operations are described in Appendix B. Reliability of Estimates With a sample size of 533 persons, 256 in Fresno and 277 in Los Angeles, the sampling error is between 4 and 7 percentage points (at a confidence level of 95 percent.) The sampling error varies depending on the responses to a particular question. For example, if 10 percent of those surveyed gave a particular response to a question, the sampling error would be about plus or minus 4 percentage points, while a response by 50 percent of those surveyed would have a sampling error of about 7 percentage points. 59 Appendix B Field Operations for the Undocumented Immigrant Health Care Access Survey Questionnaire Design and Translation The content of the questionnaire focused on the following: use of hospital and physician services; self-reported health status; insurance coverage; inability to obtain care; a symptom- response sequence; medication use; use of prenatal care and preventative care; information on the most recent medical visit; reason for immigration to the U.S.; and socio-demographic characteristics and family composition. Two versions of the questionnaire were developed—one for adults (age 16 and older) and the other for children under age 16. Both questionnaires were translated into Spanish for administration. Average administration time for the adult questionnaire was 27 minutes; the child questionnaire was approximately one-third shorter. Pretest Prior to the main survey, a small pre-test was conducted in Austin, Texas. The pre-test was carefully designed to test 1) the household sampling and respondent selection methods; 2) the adequacy of the sampling frame; 3) the effect of the advance mailing and the monetary incentive on respondent participation; and 4) the questionnaire design, including appropriateness of the Spanish translation, comprehension, perceived relevance, perceived respondent burden, absolute question order, and coding schemes. Questionnaires were administered to 25 undocumented Latino immigrants residing in selected block groups in June 1996. Block groups were selected using the same sampling procedures as planned for the main study. Following the pretest, interviewers were debriefed by project staff members to identify problem areas in the questionnaire and field procedures. Interviewer Recruiting, Training and Supervision In both sites, well-qualified interviewing teams of both males and females were hired, six interviewers and one field supervisor for each survey site. Most interviewers had previous survey interviewing experience, and all were Latino and fully bilingual (English/Spanish). Separate training sessions were held for each site, which allowed trainers to spend time addressing concerns specific to the fieldwork in each area. The training sessions were standardized and consisted of the usual interviewer training components. Unique to this training, however, was the special attention given to the introductory section of the screening 60 Appendix B interview as the key to gaining respondent trust and cooperation. To successfully screen the household, interviewers were trained to follow the introduction as worded and to avoid the use of intimidating words.* However, an informal and friendly introduction was critical for building the rapport needed to determine eligibility. Thus, interviewers were asked to adhere to the suggested introduction during their first few cases and then were allowed to deviate from the script and develop their own style of building rapport as they became more comfortable with the study. Data collection occurred between October 1996 and July 1997, although the length of the field period varied by site. Strategies for assigning and supervising fieldwork also differed by site. In Fresno, groups of interviewers were accompanied by their supervisor to sampled block groups, working as a team to complete the assignments in one block group before moving to another. In Los Angeles, however, on-site supervision was difficult because distances between block groups did not allow the supervisor to accompany a group of interviewers to the field; thus interviewers worked their assignments individually. Each week, the supervisor met with each interviewer at least once to collect completed questionnaires and paperwork, and to distribute new work assignments. In both sites, supervisors verified 15 to 20 percent of each interviewer’s work by re-contacting respondents. Gaining Respondent Cooperation Several strategies were used to obtain a high level of respondent cooperation. First, a letter and brochure (printed in both English and Spanish) were mailed to all sampled households in advance of interviewer contact. These explained the purpose of the study and informed the household that a representative would be visiting their home in the next few days. Second, interviewers were instructed to wear casual clothes, as opposed to professional attire, to look approachable and not out of place in the neighborhoods. Third, great care was taken to train the interviewers in building rapport and gaining trust during their initial contact with the respondent. It was believed this trust would elicit honest answers to the screening questions on documentation status and minimize item non-response in the questionnaire. Fourth, a $15 incentive was used to encourage participation. Interviewers also quickly learned vocabulary used by neighbors and residents of each area to describe the undocumented population, which was extremely useful for eliciting cooperation and accurately determining eligibility status.† While in the field, each interviewer carried an ID badge for identification purposes, as well as additional copies of the survey advance letter and informational brochures for community members and respondents who requested more information about the survey. * During the pretest, interviewers noticed that the word Immigrant had a negative connotation and suggested that it be replaced with another word or completely deleted from all field materials. Thus, for the main survey, the word Immigrant was deleted from all field materials. † Some of the words used to describe the target population were no están arreglados or not fixed, and recién llegados or newcomers. 61 Appendix B Response Rates Across both sites, 3,525 housing units (HUs) were screened successfully, resulting in an overall screener response rate of 90 percent. Screener response rates were calculated by dividing the number of fielded housing units, minus addresses that were vacant, did not exist, or did not meet the definition of a housing unit, by the number of housing units screened successfully. The screener response rate in Los Angeles was 98 percent compared to 81 percent in Fresno. Of the 3,525 HUs screened, 492 contained at least one eligible respondent. These 492 HUs yielded 629 respondents, from which a total of 533 interviews were completed, resulting in a combined interview response rate of 85 percent. Interview response rates were 88 percent and 81 percent for Los Angeles and Fresno, respectively. The combined overall survey response rate was 78 percent, with 87 percent in Los Angeles and 69 percent in Fresno. The overall response rate is calculated at the household level as the number of complete interviews divided by the sum of eligible households plus a portion of non- screened households for whom eligibility status is unknown. The proportion of non-screened households included in the denominator is based on the proportion of eligible households found among all screened households; it was assumed that the proportion of screened households that were eligible is the same as the proportion of non-screened households that would have been eligible. 62 Section IX. References 63 Section IX. References 1. Clark RL, Passel JS, Zimmerman WN, and Fix ME. Fiscal Impacts of Undocumented Aliens: Selected Estimates for Seven States. Washington, DC: The Urban Institute. September 1994. 2. United States General Accounting Office. “Illegal Aliens: National Net Cost Estimates Vary Widely.” Report to Congressional Requesters. GAO/HEHS-95- 133. July 1995. 3. United States General Accounting Office. “Illegal Aliens: Assessing Estimates of Financial Burden on California.” Report to the Honorable Barbara Boxer, U.S. Senate. GAO-HEHS-95-22. November 1994. 4. Norton SA, Kenney GM, and Ellwood MR. “Medicaid Coverage of Maternity Care for Aliens in California.” Family Planning Perspectives. 1996;28:108-112. 5. Ku L, Kessler B. The Number and Cost of Immigrants on Medicaid: National and State Estimates. The Urban Institute. December 1997. 6. Hubbell FA, Waitzkin H, Mishra SI, Dombrink J, Chavez LR. “Access to Medical Care for Documented and Undocumented Latinos in a Southern California County.” Western Journal of Medicine. 1991 April; 154:414-417. 7. Chavez LR, Hubbell FA, Mishra SI, Valdez RB. “Undocumented Latina Immigrants in Orange County, California: A Comparative Analysis.” International Migration Review. Volume XXXI, Number 1. Spring 1997, p. 88-107. 8. U.S. Department of Justice, Immigration and Naturalization Service. “INS Releases Updated Estimates of U.S. Illegal Population.” News Release. February 7, 1997. 9. U.S. Bureau of the Census, Population Division, release PPL-91. “United States Population Estimates, by Age, Sex, Race, and Hispanic Origin, 1990 to 1997.” February 6, 1998. 10. Passel JS. The Urban Institute. Unpublished estimates. August 1985. 11. California Department of Finance, Economic Research. February 1998. 12. Population Estimates Program, Population Division, U.S. Bureau of the Census: Washington, DC. 1997. 13. U.S. Bureau of Economic Analysis, Regional Economic Information System. May 1998. 64 Section IX. References 14. State of California, Department of Finance. “Race/Ethnic Population Estimates with Age and Sex Detail, 1996.” Sacramento, CA. January 1998. 15. Studies have shown that persons not experienced in the use of the health care system report lower needs. For example, see Wilson RW ,White EL. “Changes in Morbidity, Disability, and Utilization Differentials between the Poor and the Nonpoor: Data from the Health Interview Survey: 1964 and 1973.” Medical Care. August 1977;15(8):636-646. Stewart WH and Enterline PE. “Effects of the National Health Service on Physician Utilization and Health in England and Wales.” New England Journal of Medicine. 1961; 265:1187-1194. Vargas CM, Burt VL et al. “Validity of Self-reported Hypertension in the National Health and Nutrition Survey III, 1988-1991.” Preventive Medicine. Sep-Oct 1997; 26(5 Pt 1):678-85. 65