Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 289
--> 13 Healthcare Utilization, Expenditures, and Insurance Coverage for American Indians and Alaska Natives Eligible for the Indian Health Service Peter J. Cunningham Introduction The Indian Health Service (IHS) was established in 1955 to raise the health status of American Indians and Alaska Natives who are members or descendants of federally recognized tribes and reside on or near federal reservations and other American Indian and Alaska Native communities. The Indian Health program became a primary responsibility of the federal government as a result of the Transfer Act of 1954 (P.L. 83-568). The establishment of federal Indian health services is consistent with the authority Congress has exercised to regulate commerce with American Indian nations as provided for in the Constitution. IHS operates a network of inpatient and ambulatory care facilities across the continental United States and Alaska, many of which are now managed by American Indian tribes and Alaska Native organizations. In addition, IHS directly subsidizes healthcare services through contracts with private providers, particularly for specialized services and other services not available in IHS direct care facilities (known as Contract Health Services). The Center for Studying Health System Change is supported in full by the Robert Wood Johnson Foundation. This study was conducted while the author was a researcher at the Agency for Health Care Policy and Research. The views expressed in this paper are those of the author, and no official endorsement by the U.S. Department of Health and Human Services, the Agency for Health Care Policy and Research, or the Indian Health Service is intended or should be inferred.
OCR for page 290
--> Persons eligible for IHS have several advantages with respect to healthcare that are generally unavailable to the U.S. population as a whole. Unlike most persons who have some form of private health insurance, IHS ''beneficiaries" do not pay premiums for IHS coverage, and there are no deductibles or copayments involved in receiving IHS-sponsored services, regardless of personal or family income level. Because IHS services are essentially free of charge to eligible persons, one might expect not to see significant differences in access to care by socioeconomic status, as is the case for the general U.S. population (Freeman et al., 1987; Rowland and Lyons, 1989). Also, while many in the general U.S. population live in medically underserved rural or inner-city areas where few private medical providers are available (Lee, 1991; Berk et al., 1983), IHS facilities and resources are targeted specifically in areas where IHS eligibles generally live, including many rural and sparsely populated areas. Thus, IHS resources ideally can be distributed to areas where need is highest, without regard to other factors that affect the location decisions of private physicians. Despite these advantages, access to care may still be limited for some IHS eligibles. First, many of the areas inhabited by IHS eligibles are among the most sparsely populated areas in the United States, and residential areas are frequently spread across vast distances. Thus, although many IHS facilities are located directly in these areas, it is difficult for I HS providers to reach all eligible persons. Transportation problems and long distances to medical providers are still a major barrier to care for many persons. Second, limitations in IHS-sponsored services often result in problems that affect access to care for some IHS beneficiaries. Unlike the Medicare and Medicaid programs, IHS is not an entitlement program, and its funds are obtained through an annual appropriation by the U.S. Congress. No additional funds are available for the year if additional resources for health services are needed. Also, IHS resources are not distributed evenly across all IHS service areas, since the previous method of distributing those resources was based on historical funding patterns, rather than need (U.S. General Accounting Office, 1991).1 Access to care for IHS eligibles may be inhibited to the extent that resource limitations in some IHS service areas result in staff and facility shortages. The result could be difficulties in obtaining health services in a timely manner even when IHS facilities are located in the area. Moreover, access may be particularly limited for Contract Health Services—including expensive diagnostic and treatment services. These services 1 A needs-based formula was recently included in the resource allocation models to achieve greater parity in funding across IHS service areas.
OCR for page 291
--> may be delayed or denied to patients if funds are unavailable. At times, such services have been restricted to emergency cases because of budget constraints (Office of Technology Assessment, 1986). Given these resource limitations, some IHS eligibles may be compelled to obtain additional healthcare services from private providers. In fact, by law IHS is required to be only a "residual" provider of health services (i.e., it provides only those services not available through other sources), even though it often serves as the primary or sole source of care for much of the eligible population. As with the general U.S. population, one would expect that access to other sources of care would be enhanced for persons having higher socioeconomic status, having other private or public health coverage, and living in closer proximity to medical providers (e.g., metropolitan areas) (Spillman, 1992; Davis and Rowland, 1983; Freeman et al., 1987; Rowland and Lyons, 1989). However, given the high proportion of IHS eligibles who are poor and low-income, lack other sources of healthcare coverage, and live in rural or "frontier" areas, access to non-IHS services is no doubt severely constrained for many individuals. As a result of resource constraints, it is likely that IHS will depend increasingly on more effective use of and coordination with other sources of healthcare, at least in areas where these other sources exist. Such measures might include contracting with private healthcare organizations, such as health maintenance organizations (HMOs), to provide all health services to IHS eligibles in a given area. In other words, IHS would provide the financing for the health services, but would not be directly involved in service delivery. However, given the substantial variations in geographic location and socioeconomic characteristics of the IHS population, it is doubtful whether complete privatization of IHS services could be implemented uniformly across all IHS service areas. While private providers could be used more effectively in some areas, it is likely that IHS direct care facilities would continue to be the sole or primary source of care for persons living in some of the most remote and sparsely populated areas in the United States, even were they to obtain other private or public health coverage. The present discussion uses data from the 1987 Survey of American Indians and Alaska Natives (SAIAN) to examine various aspects of healthcare access, utilization, and expenditures for persons eligible for IHS services. The SAIAN is unique in that data on healthcare use and expenditures were collected for all sources of care—IHS and otherwise—so that the analyses of healthcare use and expenditures are comprehensive. The focus here is specifically on key policy variables that affect health service utilization, and in particular the decision to use IHS or non-IHS care. These factors include healthcare coverage (e.g., private insurance, Medicaid,
OCR for page 292
--> Medicare), socioeconomic status, place of residence, and availability of IHS facilities. It is recognized that other factors are also important in explaining patterns of healthcare utilization, particularly cultural differences among tribes in interpreting and acting upon symptoms of illness, the value placed on seeking professional care, and trust in the efficacy of prescribed treatments (Susser et al., 1985). The effects of culture on health services are not a major focus of this discussion, and cultural measures (e.g., attitudes regarding health, healthcare seeking, and health services) are quite limited in the SAIAN data. However, there is some assessment of how involvement with native culture—including use of native language and involvement in tribal activities—affects the decision to use IHS or non-IHS healthcare. Data And Methods Sources of Data The SAIAN is part of the 1987 National Medical Expenditure Survey and was sponsored in part by IHS. It comprises a representative sample of American Indian and Alaska Native households in which at least one person was eligible to receive medical care from IHS. A multistage area probability design was used to select the sample (Harper et al., 1991). The sampling frame initially consisted of 482 U.S. counties served by IHS.2 A total of 274 primary sampling units—consisting of counties or groups of counties—were created out of the initial frame, and 20 primary sampling units were selected for the initial sample. Segments were identified and sampled within each primary sampling unit, and households were sampled within each segment.3 Altogether, about 2000 households and 7600 persons were included in the sample. About 6500 sampled persons were eligible for IHS services and are included in the present analysis (persons not eligible for IHS services are excluded). Field operations for the entire SAIAN component consisted of three core interviews conducted at 5- to 6-month intervals (for more detailed discussion of the questionnaires and data collection methods used, see Edwards and Berlin, 1989). In each round of data collection, detailed 2 For reasons of cost-efficiency, the frame was truncated to exclude counties with fewer than 400 American Indians or Alaska Natives. The truncated frame included 97.2 percent of the population of interest. 3 Segments were defined as 1980 census enumeration districts or individual blocks or block combinations. For cost-efficiency, the sample frame was further restricted by excluding segments with less than 0.5 percent population representation of American Indians and Alaska Natives.
OCR for page 293
--> information was collected on each individual's health insurance coverage, health status, and healthcare utilization and expenditures, as well as and socioeconomic characteristics. Each round of data collection also included supplemental questions on specific topics, including more detailed treatment of health status and access to care. The combined response rate for all three rounds of data collection was 86 percent. A Medical Provider Survey was also conducted to obtain expenditure information from non-IHS medical providers used by sample households during 1987. This information was used to verify and supplement incomplete or missing information on expenditures obtained from the household respondents (Tourangeau and Ward, 1992). Definition of Key Variables In this discussion, a distinction is made between the use of "IHS" and "non-IHS" services. IHS services include all those obtained at IHS-owned and-operated hospitals, clinics, and health stations, including those managed by American Indian tribes or Alaska Native organizations (distinctions between IHS-operated and tribally managed facilities could not be made, however). Non-IHS services include all other health services, including those obtained from contract care providers and those with no affiliation with IHS.4 By definition, all persons included in the analysis have IHS coverage (i.e., there are no "uninsured" persons as such). Thus, the healthcare coverage variable was constructed to reflect persons having (1) only IHS coverage all year; (2) only IHS coverage for part of the year (i.e., other health coverage for part of the year); (3) other coverage all year, including some private insurance; and (4) other coverage all year, with only public coverage (i.e., Medicare, Medicaid, other state or local programs). The key comparisons are made between persons with only IHS coverage all year and those with other private or public healthcare coverage all year. Additional data on the characteristics of each sample person's county of residence were also obtained and are used in this analysis. Urban vs. rural residence was determined based on whether the county of residence was part of a metropolitan statistical area or a nonmetropolitan area. The 4 Other analyses by the author have included IHS contract care providers along with IHS direct care providers (Cunningham and Altman, 1993; Cunningham, 1993; Cunningham and Cornelius, 1995). These studies were concerned primarily with the use of care outside of the IHS "system." However, since the focus here is on highlighting the use of private resources, regardless of whether these providers are reimbursed by IHS for services provided to eligible persons, contract care providers are included along with other private providers.
OCR for page 294
--> population density of nonmetro counties was also used to distinguish very sparsely populated nonmetro areas from other areas. Thus nonmetro areas were divided into those counties with 10 or more persons per square mile and those with fewer than 10 persons per square mile. Variables that indicate the availability of IHS or tribal healthcare facilities in the county of residence were also included. For the descriptive analyses, the classification included counties with (1) any IHS hospital, (2) IHS clinics or health stations but no hospitals, and (3) no IHS facilities. Data Limitations It should be noted that the SAIAN population is a subset of the total U.S. population of American Indians and Alaska Natives. The SAIAN sample was selected to be representative of American Indians and Alaska Natives who were members or descendants of federally recognized tribes and eligible to receive IHS services. Thus, the findings are not necessarily representative for all persons who identify themselves as American Indians or Alaska Natives, especially those who are not eligible for IHS services either because they reside outside of IHS service areas or because they are not members or descendants of federally recognized tribes. To avoid misinterpreting the results as being generalizable to all American Indians and Alaska Natives, findings from the SAIAN are discussed in terms of the "SAIAN population" or the "IHS eligible population." A second limitation with the SAIAN is that the effects of cultural factors on health service utilization cannot be thoroughly assessed because, as noted above, few direct measures of culture (e.g., health practices, attitudes regarding health and healthcare, use of traditional medicine) were included in the survey. In addition, resource constraints precluded sufficient subsampling within individual tribes or communities to allow assessment of differences among American Indian tribes or communities (which could be due to cultural differences). While cultural factors are not a focus here, it is possible that the effects of key variables of interest (e.g., health insurance coverage) on health service utilization are confounded by cultural factors. There is some control for cultural differences in the multivariate analysis through inclusion of the individual's primary language (i.e., English vs. a native tribal language) and participation in tribal activities. Although these measures probably do not capture all relevant dimensions of culture, they have significant effects on the use of healthcare services (as discussed in greater detail below). To the extent that other cultural factors not included here are correlated with socio-demographic characteristics and other control variables included in the analysis, the confounding effects of culture on key policy variables are minimized, although to what extent cannot be directly assessed.
OCR for page 295
--> Some estimates for the SAIAN population are compared with estimates for the general U.S. population. The latter estimates were derived from the National Medical Expenditure Survey's Household Survey (also conducted in 1987), which was designed to produce representative estimates of the civilian noninstitutionalized population. Field operations for this survey consisted of four core interviews conducted at 3- to 4-month intervals. Other than differences in the number of rounds of data collection, instruments and data collection procedures used in this survey and in the SAIAN were virtually identical, which facilitates making direct comparisons between the two populations. All estimates for the SAIAN and U.S. populations were weighted. Population weights were designed to yield representative estimates for the IHS eligible and general U.S. populations for 1987. Standard errors for the estimates were adjusted to account for the complex design of the SAIAN and the Household Survey. While standard errors are not included in the tables, differences between specific estimates are discussed only if they are statistically significant at the 0.05 level, unless stated otherwise. Findings Characteristics and Healthcare Coverage of the SAIAN Population Socioeconomic and geographic differences between the SAIAN population and the general U.S. population are striking (Table 13-1). Almost two-thirds of the SAIAN population resided in nonmetropolitan areas, and 30.9 percent resided in nonmetro areas with very low population density (i.e., fewer than 10 persons per square mile). By contrast, three-fourths of the general U.S. population resided in metropolitan areas, and less than 3 percent in areas with very low population density. Adults in the SAIAN population were less likely to be employed full-time and all year as compared with the general U.S. population (27 versus 43.9 percent) and more likely not to have been employed at all in 1987 (39 versus 29.7 percent). The SAIAN population had considerably higher rates of poverty and low income than the general U.S. population: 37.4 percent of the SAIAN population had incomes below the federal poverty line (compared with 13.5 percent for the general U.S. population), and more than two-thirds had family incomes below 200 percent of the poverty line (compared with about one-third for the general population). Since employment and income are highly correlated with having private insurance or public healthcare coverage, it is not surprising that only 24.9 percent of the SAIAN population had private insurance coverage, compared with 70.7 percent of the general U.S. population. The SAIAN
OCR for page 296
--> TABLE 13-1 Characteristics of the Total U.S. and SAIAN Populations, 1987 Percentage Distribution Characteristic Total U.S. Population (239,393,000) SAIAN Population (906,000) Age 0-5 9.2 14.6 6-17 17.6 27.0 18-44 42.7 38.8 45-64 18.8 13.2 65 and over 11.7 6.5 Sex Male 48.5 49.3 Female 51.5 50.7 Educational attainment Less than high school 26.1 40.1 Completed high school 36.3 34.6 Some college 36.3 20.9 Employmenta All year, full-time 43.9 27.0 Part of year or part-time 26.4 34.0 Not employed 29.7 39.0 Family incomeb Poor 13.5 37.4 Low-income 18.5 30.6 Middle-income 34.9 23.1 High-income 33.1 8.9 Healthcare coveragec Not covered all year (other than IHS) 10.3 42.5 population had somewhat higher rates of other public coverage all year (16 versus 9.2 percent), largely because of the disproportionately high number of poor and low-income persons eligible for Medicaid coverage. Almost 60 percent of the SAIAN population relied exclusively on IHS coverage for at least part of the year. Table 13-2 shows considerable variation in healthcare coverage by employment status, socioeconomic status, and place of residence. Of adults employed full-time and all year, 51.6 percent had private insurance coverage for all of 1987, compared with 21.9 percent of adults employed part-time or part of the year and about 15.2 percent of adults not employed. Those employed part-time or part of the year were more likely to rely exclusively on IHS coverage than were those not employed, largely because of greater availability of other public coverage for persons not
OCR for page 297
--> Percentage Distribution Characteristic Total U.S. Population (239,393,000) SAIAN Population (906,000) Not covered part of year 9.8 16.5 (other than IHS) Covered all year Any private 70.7 24.9 Public only 9.2 16.0 Place of residence Metropolitan statistical area 75.7 37.0 Nonmetro area At least 10 persons per square mile 21.6 32.1 Fewer than 10 persons per square mile 2.7 30.9 a Employed full-time includes those working at least 35 hours per week. Employed all year includes those employed at least 45 weeks during 1987. b Poor refers to individuals in families with incomes below the poverty line; low-income to those with incomes between the poverty line and 200 percent of the poverty line; middle-income to those with incomes over 200 to 400 percent of the poverty line; and high-income to those with incomes over 400 percent of the poverty line. c Private and public coverage is in addition to IHS coverage for persons in SAIAN. Public coverage includes Medicare, Medicaid, Civilian Health and Medical Programs of the Uniformed Services (CHAMPUS), Civilian Health and Medical Programs of the Veterans Administration (CHAMPVA), and other state or local public assistance. SOURCE: Agency for Health Care Policy and Research. National Medical Expenditure Survey—Survey of American Indians and Alaska Natives and Household Survey. employed (i.e., Medicare for elderly persons and Medicaid for unemployed nonelderly persons). All-year private insurance coverage increased sharply with income level: 44.8 percent of middle-income persons and almost 72.1 percent of high-income persons had private coverage all year, compared with only 6.9 percent of those with incomes below the poverty level. Of the SAIAN population living in poverty, 60 percent relied exclusively on IHS coverage for all of 1987, compared with 30 percent of middle-income persons and 12.9 percent of high-income persons. Scarce employment opportunities and high levels of poverty were prevalent for the majority of the SAIAN population living in nonmetro areas. Many of these areas have very low population densities and are far
OCR for page 298
--> TABLE 13-2 Healthcare Coverage of the SAIAN Population, 1987 IHS Coverage Only Other Coverage All Year Characteristic All year (%) Part of year (%) Any private(%) Public only(%) All persons 42.5 16.5 24.9 16.0 Age 0-17 43.3 18.1 22.1 16.6 18-64 46.2 16.3 27.0 10.5 65 years and older 8.1 7.9 27.3 56.7 Sex Male 44.7 16.7 24.5 14.3 Female 40.7 16.3 25.4 17.7 Employmenta Full-time, all year 32.8 14.7 51.6 0.9b Part-time or part of year 50.3 20.0 21.9 7.8 Not employed 41.3 12.0 15.2 31.5 Family incomec Poor 60.0 13.8 6.2 20.0 Low-income 39.2 19.5 19.2 22.1 Middle-income 30.0 18.4 44.8 6.9b High-income 12.9 13.2 72.1 1.7b Place of residence Metropolitan service area 24.4 21.4 35.4 18.9 Nonmetro area At least 10 persons per square mile 48.7 13.0 21.3 17.0 Fewer than 10 persons per square mile 57.7 14.4 16.3 11.6 IHS facilities Hospital 54.0 15.6 16.4 14.1 Clinics only 32.3 18.1 29.7 20.0 No facilities 21.7b 15.7b 51.8 10.9b a Employed full-time includes those working at least 35 hours per week. Employed all year includes those employed at least 45 weeks during 1987. b Standard error greater than 30 percent of the estimate. c Poor refers to individuals in families with incomes below the poverty line; low-income to those with incomes between the poverty line and 200 percent of the poverty line; middle-income to those with incomes over 200 to 400 percent of the poverty line; and high-income to those with incomes over 400 percent of the poverty line SOURCE: Agency for Health Care Policy and Research. National Medical Expenditure Survey—Survey of American Indians and Alaska Natives.
OCR for page 299
--> from urban areas, factors that combine to make private health insurance coverage difficult to obtain. Thus, IHS eligible persons residing in the most sparsely populated areas have the lowest rates of private health insurance coverage and are more likely to rely exclusively on IHS coverage. Regular Source of Healthcare The ability to identify a regular source of care—as well as the type of place—is strongly associated with the use of health services (Aday and Andersen, 1975; Aday et al., 1980). Since IHS facilities and services are targeted specifically to the IHS eligible population, it is perhaps not surprising that over 91 percent of the SAIAN population reported having a regular source of healthcare, compared with 81.6 percent of the general U.S. population (estimates not shown). For both populations, most persons who did not have a regular source of care said they did not need a doctor or had no need for healthcare (findings not shown). Only a very small percentage reported not having a regular source of care because of problems associated with the cost of care or the availability of providers. Of greater interest is the extent to which IHS eligible persons identify a non-IHS facility as their regular source of care and whether particular demographic and socioeconomic characteristics are associated with this response. Table 13-3 shows that almost one-third of the SAIAN population identified a non-IHS provider as their regular source of healthcare, although this response was strongly associated with living in a metro area, having a higher income, having other healthcare coverage, and living in an area with relatively few IHS facilities. In particular, the differences by urban/rural location are striking. While 63.2 percent of the SAIAN population in metro areas had a non-IHS regular source of care, this was the case for only 25.3 percent in nonmetro areas with relatively high population density and 6.3 percent in very low-density areas. While the percentage with a non-IHS regular source of care was generally higher for residents of metropolitan than nonmetro areas, regardless of healthcare coverage or income level, there were some important differences within metro and nonmetro areas. For example, the proportion of metropolitan residents with a non-IHS regular source of care was considerably smaller for persons with only IHS coverage all year (33.4 percent), persons with family incomes below the poverty line (34.4 percent), and persons living in areas with an IHS hospital (19.1 percent). By contrast, 87.5 percent of metro residents with private insurance, 74.5 percent of middle-income persons, 91.2 percent with high incomes, and 74.8 living in areas with no IHS hospital had a non-IHS regular source of care.
OCR for page 304
--> TABLE 13-4 Results of Logistic Regression Analysis for Probability of an Ambulatory Visit for SAIAN Population, 1987 Beta Coefficients Characteristic Likelihood of Any Ambulatory Use Likelihood of a Visit at an IHS Facilitya Likelihood of a Visit to a nonIHS Providerb Intercept 2.85** 1.64** 0.50** Perceived health status (Poor health is omitted category) Excellent -0.77** -0.84** -0.64** Good -0.48 -0.63** -0.43* Fair -0.29 -0.36 -0.22 Missing -0.70* -0.66** -0.57 One chronic condition 1.75** 1.14** 0.70** Two or more chronic conditions 2.25** 1.71** 1.07** Gender (1=male) -0.47** -0.37** -0.17** Age -0.07** -0.06** -0.03** Age squared × 10-2 0.07** 0.06** 0.02* Family size -0.08** -0.06** -0.06** Education (less than high school is omitted category) High school 0.06 -0.06 0.09 Some college 0.44** 0.11 0.51** Missing -0.09 -0.14 -0.28 Healthcare coverage (IHS only all year is omitted category) Other private coverage all year 0.55** -0.36** 1.27** Other public coverage all year 0.54 0.05 1.10 IHS only part of year 0.29 0.21** 0.70** Incomec (poor is omitted category) Low-income 0.18* 0.13 0.22** Middle-income 0.34** 0.23** 0.32** High-income -0.08 -0.10 0.21 Primary language other than English -0.19 0.05 -0.38** Participates in tribal activities 0.10 -0.02 -0.13 Place of residence (metro area is omitted category) Nonmetro (10 persons per square mile or more) -0.54** -0.12 -0.47** Nonmetro (fewer than 10 persons per square mile) -0.54** -0.01 -1.01**
OCR for page 305
--> Beta Coefficients Characteristic Likelihood of Any Ambulatory Use Likelihood of a Visit at an IHS Facilitya Likelihood of a Visit to a nonIHS Providerb IHS hospital in county of residence 0.15 1.05** -1.14** Log of number of IHS clinics in county per 10,000 persons Health professional shortage area designation (no shortage area is omitted category) 0.29** 0.42** -0.09 All of county is shortage area -0.34** -0.53** 0.19 Part of county is shortage area -0.43** -0.83** 0.38 N 6,473 6,473 6,473 NOTE: * p < .05, ** p < .01 a Includes visits at IHS direct care facilities and IHS facilities under tribal management. b Includes visits to IHS contract care providers and providers with no affiliation with IHS. c Poor refers to individuals in families with incomes below the poverty line; low-income to those with incomes between the poverty line and 200 percent of the poverty line; middle-income to those with incomes over 200 to 400 percent of the poverty line; and high-income to those with incomes over 400 percent of the poverty line. SOURCE: Agency for Health Care Policy and Research. National Medical Expenditure Survey—Survey of American Indians and Alaska Natives. tended to be the major factors associated with use, cultural factors were also significant. Participants in tribal cultural activities made a higher number of visits to IHS facilities, while those whose primary language was other than English made fewer visits to non-IHS providers. As in the previous analyses, these findings suggest greater barriers to the use of non-IHS providers for individuals who are less acculturated in the mainstream American culture. Out-Of-Pocket Expenditures Since IHS services—both direct care and contract care services—involve no deductibles or copayments for IHS eligible persons, one would expect out-of-pocket healthcare expenditures for this population to be quite low relative to the general U.S. population. Table 13-6 shows that
OCR for page 306
--> TABLE 13-5 Results of Weighted Least Squares Regression Analysis for Log of Number of Ambulatory Visits for SAIAN Population, 1987 Regression Coefficients Characteristic Log of Number of All Ambulatory Visits Log of Number of Visits at IHS Facilitiesa Log of Number of Visits to nonIHS Providersb Intercept 2.09 1.32 1.53 Perceived health status (Poor health is omitted category) Excellent -0.65** -0.33** -0.62** Good -0.43* -0.21 -0.31** Fair -0.28* -0.12 -0.22 Missing -0.49 -0.16 -0.59** One chronic condition 0.42** 0.32** 0.31** Two or more chronic conditions 0.86** 0.75** 0.33* Gender (1=male) -0.22** -0.18** -0.21 Age -0.03** -0.02** -0.02** Age squared × 10-3 0.03** 0.02** 0.02* Family size -0.02** -0.03* -0.02 Education (less than high school is omitted category) High school 0.06 0.06 0.03 Some college 0.12 0.16** 0.05 Missing -0.19 -0.05 -0.18 Healthcare coverage (IHS only all year is omitted category) Other private coverage all year 0.17** -0.06 0.38** Other public coverage all year 0.19** -0.01 0.47** IHS only part of year 0.10* -0.02 0.24* Incomec (poor is omitted category) Low-income 0.10* 0.10* 0.07 Middle-income 0.17* 0.07 0.19* High-income 0.16* -0.06 0.27* Primary language other than English -0.09 0.01 -0.19* Participates in tribal activities -0.06 0.11* -0.19 Place of residence (metro area is omitted category) Nonmetro (10 persons per square mile or more) -0.16** -0.05E-1 -0.23** Nonmetro (fewer than 10 persons per square mile) -0.08* 0.07 -0.23**
OCR for page 307
--> Regression Coefficients Characteristic Log of Number of All Ambulatory Visits Log of Number of Visits at IHS Facilitiesa Log of Number of Visits to nonIHS Providersb IHS hospital in county of residence 0.01 0.18** -0.11* Log of number of IHS clinics in county per 10,000 persons Health professional shortage area designation (no shortage area is omitted category) 0.02 0.09* -0.05 All of county is shortage area 0.03 -0.05 0.13 Part of county is shortage area 0.03 -0.06 0.14* N 4,446 3,687 1,598 NOTE: * p < .05; ** p < .01 a Includes visits at IHS direct care facilities and IHS facilities under tribal management. b Includes visits to IHS contract care providers and providers with no affiliation with IHS. c Poor refers to individuals in families with incomes below the poverty line; low-income to those with incomes between the poverty line and 200 percent of the poverty line; middle-income to those with incomes over 200 to 400 percent of the poverty line; and high-income to those with incomes over 400 percent of the poverty line while about three-fourths of the general U.S. population had some kind of out-of-pocket healthcare expenditure (premiums for health insurance are excluded), only one-third of the SAIAN population had such expenditures. The percentage with out-of-pocket expenditures increased with age and family income, and was also higher for persons with private insurance, those in metropolitan areas, and those living in areas with fewer IHS facilities. While one might also expect the percentage with out-of-pocket expenditures to be higher for persons in fair or poor health (since their use of healthcare is higher than for persons in good health), there were no statistically significant differences in the percentage with out-of-pocket expenditures between persons in excellent or good health and those in fair or poor health. There was also no difference in the percentage with out-of-pocket expenditures between persons with only IHS coverage all year and persons with other public coverage (about 22 percent). For persons who had out-of-pocket expenditures, the average expense
OCR for page 308
--> TABLE 13-6 Out-of-pocket healthcare Expenditures for Personal Health Services for U.S. and SAIAN Populations, 1987 Characteristic Percentage with Any Out-of Pocket Expense Average Out-of-Pocket Expenses for Persons with an Expense ($) Total U.S. population 75.7 476 SAIAN population 33.0 360 Age (SAIAN population) 0-5 18.1 178 6-17 26.7 241 18-44 36.1 258 45-64 46.3 548 65 and over 48.0 884 Perceived health status Good or excellent 31.8 222 Fair or poor 34.7 365 Family incomea Poor Low-income 19.3 415 Middle-income 30.0 264 High-income 48.8 419 Healthcare coverage 60.5 327 Not covered all year (other than IHS) 21.8 262 Covered part of year (other than IHS) 34.1 271 Covered all year Any private 58.6 374 Public only 21.9 704 in 1987 was more than $100 higher for the general U.S. population than for the SAIAN population. For the SAIAN population, average out-of-pocket expenditures increased considerable with age, and persons in fair or poor health had somewhat higher expenses than persons in good or excellent health. The relationship between family income and out-of- pocket expenditures was not linear: poor and middle-income persons had somewhat higher expenses than persons with low and high incomes, possible because of the confounding effects of differential health status and healthcare coverage by family income. Of the different types of healthcare coverage, persons with other public coverage had much higher expenditures ($704) than those with other coverage types, mainly as a result of elderly persons with Medicare coverage and nonelderly persons in poor health with Medicaid coverage. Average expenditures were also highest in areas with no IHS facilities.
OCR for page 309
--> Characteristic Percentage with Any Out-of Pocket Expense Average Out-of-Pocket Expenses for Persons with an Expense ($) IHS facilities Hospital 20.7 231 Clinics only 44.0 387 No facilities 55.5 531 Place of residence Metropolitan statistical area 49.7 382 Nonmetro area At least 10 persons per square mile 26.3 411 Fewer than 10 persons per square mile 20.1 224 NOTE: Includes expenditures for inpatient hospital and physician services and ambulatory physician and nonphysician services, including vision care and telephone calls with a charge, prescribed medicines, home healthcare services, dental services, and medical equipment purchases and rentals. a Poor refers to individuals in families with incomes below the poverty line; low-income to those with incomes between the poverty line and 200 percent of the poverty line; middle-income to those with incomes over 200 to 400 percent of the poverty line; and high-income to those with incomes over 400 percent of the poverty line. SOURCE: Agency for Health Care Policy and Research. National Medical Expenditure Survey—Survey of American Indians and Alaska Natives and Household Survey. Out-of-pocket expenditures are a concern to the extent that they impose a heavy financial burden on families and households with sick family members who require intensive healthcare use. A common way of assessing the burden of healthcare expenses is to compute the ratio of out-of-pocket expenditures to family income for households. Table 13-7 shows that SAIAN households generally had less of a financial burden due to healthcare expenses than did the general U.S. population. More than one-third of SAIAN households had no out-of-pocket healthcare expenditures in 1987, compared with only 11 percent of all U.S. households. Slightly fewer than half of SAIAN households had healthcare expenditures that comprised between 0 and 5 percent of family income (compared with about 68 percent of all U.S. households), while about 6 percent of SAIAN households had expenditures that amounted to 10 percent or more of family income (compared with about 11 percent of all U.S. households).
OCR for page 310
--> TABLE 13-7 Annual Family Out-of-Pocket Expenditures for Personal Health Services As a Percentage of Family Income, SAIAN and General U.S. Population, 198 Percentage Distribution of Families Percentage of Family Income Total U.S. population SAIAN population No expenditure 11.0 35.1 0.01-0.99% 30.4 23.9 1.00-1.99% 17.3 10.8 2.00-4.99% 20.7 13.3 5.00-9.99% 10.2 6.8 10.00-19.99% 5.6 3.1 20.00% or more 4.4 3.3 No income 0.4 3.8 NOTE: Includes expenditure for inpatient hospital and physician services and ambulatory physician and nonphysician services, including vision care and telephone calls with a charge, prescribed medicines, home healthcare services, dental services, and medical equipment purchases and rentals. SOURCE: Agency for Health Care Policy and Research. National Medical Expenditure Survey—Survey of American Indians and Alaska Natives and Household Survey. Despite the overall lower level of financial burden experienced by the SAIAN population relative to the general U.S. population, it is noteworthy that there was no statistically significant difference in the percentage of ''very high burden" families (about 3 percent for the SAIAN population and 4 percent for the general U.S. population), defined as having out-of-pocket expenditures that were 20 percent or more of family income. Also, differences between the SAIAN and general U.S. populations are not as large as one might expect, given that IHS eligibles by definition require no out-of-pocket healthcare expenditures. The smaller-than-expected differences are largely the result of the SAIAN population's having a considerably smaller denominator (i.e., family income) in the calculation of the ratios. Thus, even relatively modest out-of-pocket expenditures incurred by many IHS eligibles can be financially burdensome. Discussion Characteristics of the population eligible for IHS services indicate that many would be seriously underserved with respect to healthcare if not for the availability of IHS-supported services. The IHS eligible population
OCR for page 311
--> has lower rates of employment and much higher rates of poverty (37 percent) than the general U.S. population, most IHS eligibles lack other types of health insurance on a continuous basis, and a high proportion of the population lives in some of the most sparsely populated areas in the United States. IHS eligibles with these characteristics tend to rely quite heavily on IHS as their primary or sole source of healthcare. On the other hand, there is considerable diversity among the IHS eligible population, not only culturally, but also in demographic and socioeconomic characteristics. Among the SAIAN population, almost one-third were middle- or high-income individuals, over 40 percent had some other healthcare coverage, and more than one-third lived in or near a metropolitan area. This diversity is also reflected in the healthcare utilization and expenditures of IHS eligibles, even though all can receive IHS services free of charge. Many IHS eligibles do have other sources of healthcare, and the effects of other health insurance on patterns of utilization are quite profound. IHS eligibles in metropolitan areas with high income and other health coverage—especially private insurance—tend to have a non-IHS regular source of care. The findings also show that persons with other healthcare coverage are more likely to make use of any ambulatory care than persons with IHS coverage only, mainly as a result of higher use of non-IHS care. Moreover, the findings suggest that persons who use non-IHS services do not merely substitute non-IHS healthcare for services they would otherwise have received at IHS facilities. Because there are generally no differences in the use of IHS ambulatory care by type of healthcare coverage, it is possible that persons with other coverage supplement rather than substitute services received at IHS facilities. Of course, the ability to procure other healthcare services has much to do with geographic location. Almost two-thirds of the SAIAN population lived in nonmetropolitan areas, and almost one-third lived in areas with very low population density. Persons in these areas were much less likely to have other healthcare coverage (in part because of high rates of unemployment, which makes private insurance less available), and there were few other alternatives to IHS facilities. Persons who had other private and public coverage in nonmetro areas (especially low-density areas) were much less likely than persons with other coverage in metro areas to have a non-IHS regular source of care. The targeting of IHS facilities in very remote areas appears to have considerably enhanced access to care for persons in these areas. Among SAIAN respondents, travel times to IHS providers in low-density nonmetropolitan areas were actually shorter than in other areas (i.e., other nonmetro and metro areas), and there were no statistically significant differences in the use of ambulatory care at IHS
OCR for page 312
--> facilities between persons in metro and nonmetro areas. Nevertheless, persons in nonmetro areas still used less ambulatory care overall than persons in metro areas as a result of their more restricted access to non-IHS services. The multivariate results also suggest that some IHS eligibles experience language or cultural barriers to the use of non-IHS providers. These findings are consistent with those of other studies that have found disparities in access and health service utilization for ethnic groups that are less acculturated in American society (Wells et al., 1989). This is a difficult issue to address from a policy perspective because it suggests that merely extending health insurance coverage or enhancing physical access to non-IHS providers could still leave disparities in access to care and health service utilization. Thus, having "culturally competent" providers available to serve the local population is an important consideration in reforming healthcare for IHS eligibles, especially for communities that are more culturally isolated from mainstream American society. Even though the healthcare provided by IHS is comprehensive, and much of the IHS eligible population relies almost exclusively on IHS, the intent of IHS is not necessarily to be the sole or even primary provider of health services to the eligible population. As noted earlier, IHS was designed to be a residual provider of health services and is further restricted in providing all of the healthcare needed by its eligible population because it is not an entitlement program, and revenues are appropriated on an annual basis. Therefore, improvements in healthcare for the IHS eligible population will depend increasingly on utilizing other resources, particularly from the private sector. This is already occurring to a large extent through the IHS contract care system and IHS eligibles who rely on their own resources to use other healthcare. Portions of the service population already use the private-sector delivery system extensively, and this proportion could probably be expanded if there were greater subsidization for these services. IHS is also pursuing contracts with HMOs to provide all health services to eligible persons in a given area (e.g., the Pascua Yaqui tribe in Tucson), and tribal governments are increasingly taking over the management and operation of IHS facilities and services in their areas. The availability of other healthcare resources in some IHS areas results in inequities in access to and use of health services among IHS eligibles by income level, healthcare coverage, and residential location. As IHS relies increasingly on the private sector and individual tribes to provide health services to its eligible population, the task will be to distinguish between areas and individuals that can be served effectively by other health systems and individuals who have no recourse other than publicly provided healthcare.
OCR for page 313
--> References Aday, L.A., and R. Andersen 1975 Development of Indices of Access to Medical Care. Ann Arbor, MI: Health Administration Press. Aday, L.A., R. Anderson, and G. Fleming 1980 Healthcare in the U.S.: Equitable for Whom? Beverly Hills, CA: Sage Publications. Anderson, R. 1968 A Behavioral Model of Families' Use of Health Services. Chicago, IL: Center for Health Administration Studies. Beauregard, K., P. Cunningham, and L. Cornelius 1991 Access to Healthcare: Findings from the Survey of American Indians and Alaska Natives. (AHCPR Pub. No. 91-0028). National Medical Expenditure Survey Research Findings 9. Rockville, MD: Agency for Healthcare Policy and Research. Berk, M.L., A.B. Bernstein, and A.K. Taylor 1983 The use and availability of medical care in health manpower shortage areas. Inquiry 20(4):369-380. Cunningham, P.J., and B.M. Altman 1993 The use of ambulatory health care services by American Indians with disabilities. Medical Care 31(7):600-616. Cunningham, P.J., and L.J. Cornelius 1995 Access to ambulatory care for American Indians and Alaska Natives: The relative importance of personal and community resources. Social Science and Medicine 40(3):393-407. Cunningham, P.J. 1993 Access to care in the Indian Health Service. Health Affairs 12(3):224-233. Davis, K., and D. Rowland 1983 Uninsured and underserved: Inequities in healthcare in the United States. Milbank Memorial Fund Quarterly 61(2):149-176. Edwards, W., and M. Berlin 1989 Questionnaires and Data Collection Methods for the Household Survey and Survey of American Indians and Alaska Natives. DHHS Publication No. (PHS) 89-3450. Rockville, MD: Agency for Health Care Policy and Research. Freeman, H.E., R.J. Blendon, L.H. Aiken, S. Sudman, C.F. Mullinix, and C.R. Corey 1987 Americans report on their access to healthcare. Health Affairs 6(2):6-18. Harper, T., R. Apodaca, D. Northrup, and R. DiGaetano 1991 National Medical Expenditure Survey. Survey of American Indians and Alaska Natives: Final Methodology Report. Rockville, MD: Agency for Health Care Policy and Research. Lee, R.C. 1991 Current approaches to shortage area designation. Journal of Rural Health 7(4):437-450. Office of Technology Assessment, U.S. Congress 1986 Indian Health Care (OTA-H-290). Washington, D.C.: U.S. Government Printing Office. Rowland, D., and B. Lyons 1989 Triple jeopardy: Rural, poor, and uninsured. Health Services Research 23(6):975-1004. Spillman, B.C. 1992 The impact of being uninsured on utilization of basic health services. Inquiry 29(4):457-466.
OCR for page 314
--> Susser, M., W. Watson, and K. Hopper 1985 Culture and health. Chapter 4 in Sociology in Medicine. New York: Oxford University Press. Tourangeau, K., and E. Ward 1992 Questionnaires and Data Collection Methods for the Medical Provider Survey (AHCPR Pub. No. 92-0042). National Medical Expenditure Survey Methods 4, Rockville, MD: Agency for Health Care Policy and Research. U.S. General Accounting Office 1991 Indian Health Service: Funding Based on Historical Patterns, Not Need (GAO/HRD-91/5). Washington, D.C.: U.S. Government Printing Office. Wells, K.B., J.M. Golding, R.L. Hough, M.A. Burnam, and M. Karno 1989 Acculturation and the probability of use of health services by Mexican Americans. Health Services Research 24(2):237-257.
Representative terms from entire chapter: