PEDIATRICS Vol. 104 No. 1 Supplement July 1999, pp. 151-157

GENERAL PEDIATRIC RESEARCH:
Black and White Middle Class Children Who Have Private Health Insurance in the United States

Michael Weitzman, MD^*, Robert S. Byrd, MD, MPH, and Peggy Auinger, MS^*

From the ^* Department of Pediatrics, University of Rochester School of Medicine and Dentistry, Rochester, New York, and the  Department of Pediatrics, University of California School of Medicine at Davis, Davis, California.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  To compare the health, behavior and school problems, and use of medical, mental health, and special education services of privately insured, middle class black and white children in the United States.

Design/Methods.  Analyses of the Child Health Supplement to the 1988 National Health Interview Survey, with a nationally representative sample of 17 110 children age 0-17 years.

Results.  Privately insured middle class black children had fewer chronic health conditions, but were less likely to be reported to be in excellent health (46.2% vs 57.3%) and more likely to have had asthma (8.5% vs 5.8%) or to have been of low birth weight (10.7% vs 5.6%). There were no differences in rates of having a usual source of routine care (92.2% vs 93.8%) or of being up to date with well-child care (79.3% vs 78.2%), but black children made fewer physician visits, were less likely to use physicians' offices, were more likely to lack continuity of care, and were twice as likely to use emergency departments. These differences in use of medical services persisted in multivariate analyses and analyses restricted to more affluent children. Despite similar rates of behavior problems, black children were more likely to repeat a grade (20.0% vs 12.3%) and to have been suspended from school (11.3% vs 5.0%). Although significantly fewer black middle class children received mental health or special education services in bivariate analyses, no differences in receipt of these services were noted in multivariate analyses. All differences reported were significant.

Conclusions.  Among middle class children in the United States, black and white children have similar rates of health and behavior problems, but black children experience substantially increased rates of asthma, low birth weight, and school difficulties. Although not differing in the receipt of mental health or special education services, middle class black children, even in the presence of private health insurance, have markedly different sources and patterns of use of medical services.  Key words:  black and white children, middle class, private health insurance.

Substantial time and resources have been devoted to identifying, attempting to understand the causal mechanisms behind, and rectifying racial disparities in health in the United States.^1-10 Much of our understanding of racial differences in children's health and access to health services has been hindered by black race being confounded and often confused with low socioeconomic status,^4,9,11,12 and we often fail to recognize that >50% of all black children in the United States do not live in poverty.

In contrast to the extensive literature comparing the health and use of health services of poor and nonpoor children^13-18 or of uninsured and insured children,¹⁹ virtually no literature has focused on these issues for middle class black and white children. The objectives of this study were to investigate and compare the health, behavior and school problems, sources of ambulatory care for, and use of medical, mental health, and special education services of middle class black and white children with private health insurance in the United States. These children were the focus because private insurance is the primary means of financing health care for children in this country,²⁰ and we wished to examine the health and patterns of service use among children for whom there are no, or minimal, financial barriers to care.

	METHODS

Top Abstract Methods Results Discussion References

Data were from the Child Health Supplement to the 1988 National Health Interview Survey. This is a cross-sectional, randomized household survey conducted by the Bureau of the Census for the National Center for Health Statistics.^21,22 In selected years, the most recent being 1988, a Child Health Supplement is added to obtain detailed information about the health of and use of health services by the nation's children. Information was obtained by face-to-face interviews with adult family members on 1 randomly chosen child per household surveyed, resulting in a sample of 17 110 children 0 to 17 years of age, representative of all noninstitutionalized children and youth in the United States in 1988.

The Census Bureau does not have an official definition of the "middle class," but it does rank households from lowest to highest based on income and then divides them into equal population groups, typically quintiles. The lowest quintile often is used as a proxy measure of those in poverty and the highest quintile as those who are most affluent.²³ For the purposes of these analyses, children were categorized as being middle class if their families' reported income was in the second, third, or fourth quintile. In 1988, family incomes in the second quintile ranged from $15 103 to $26 182, in the third quintile from $26 183 to $38 500, and in the fourth quintile from $38 501 to $55 906. Therefore, children in middle class families, as defined in these analyses, were in families whose incomes ranged from $15 103 to $55 906 in 1988. The comparable range in 1995 was $19 071 to $72 260. Because of concern that comparisons using children in families whose incomes are in the second through fourth quintiles might be biased because of the unequal distribution of black and white children across this income range, all bivariate analyses were repeated after analyses were restricted only to children in families whose incomes were in the third and fourth quintiles.

Children were categorized as being covered by private health insurance if the respondent answered yes to the question, "Is  now covered by a health insurance plan that pays any part of a hospital, doctor's office or surgeon's bill?" Although this question may be answered affirmatively by some children covered by Medicaid, such misclassification is likely to be minimized with the analyses limited to children whose family income is above the poverty level. Demographic and family characteristics, health status, behavior and school problems, and sources of ambulatory care and utilization of health services of privately insured, middle class black and white children and youth were compared, as presented below.

Demographic and Family Characteristics

Characteristics investigated included family income and size; region and degree of urbanization of residence^24,25; maternal educational level and age at the time of child's birth; single- or two-parent family; and index child's age and gender. As is true of all other variables in this dataset, children's race was based on parental report. Race and Hispanic ethnicity were asked separately. Hispanic children were not excluded from analyses, and racial comparisons in this study are likely to be less pronounced than in analyses confined to non-Hispanic children.

Health Status

The following aspects of children's physical health were investigated: parental rating of children's overall health; the percentage of children whose birth weights were low and very low (2500 and 1500 g, respectively) or who were reported as having asthma; the prevalence of chronic physical health conditions (ascertained from affirmative responses to a checklist of 76 childhood health conditions that respondents were asked whether the child had in the past 12 months68 of which were considered chronic, consistent with earlier work using the survey)^19,21,26,27; limitations of activities among children reported as having any of these chronic conditions; and the number of days spent in bed.

Behavior and School Problems

Extreme scores (top 10th percentile) on the 32-item Behavior Problem Index (BPI) developed by Zill for children 4 years of age and older and modeled after the Child Behavior Checklist of Achenbach and Edelbrock^28,29 were used to identify the percent of children who had behavior problems. This index has been used in several earlier studies,^30-35 and extreme scores on it have been shown to correlate substantially with referrals to mental health professionals.³⁰ Whether the respondent believed that the child needed psychological help in the past year also was determined, as was whether the child had ever repeated a grade or been suspended from school.

Utilization of Medical, Mental Health, and Special Education Services

Sources of routine and sick care and whether the child utilized the same source for routine and sick care (continuity of routine and sick care) were assessed. Sources of care were categorized as 1) physicians' office (doctors' offices, private clinics, and health maintenance organization/prepaid group plans); 2) outpatient clinics (hospital, school, and migrant clinics); 3) health centers (neighborhood and rural centers); and 4) emergency departments (hospital emergency departments and walk-in emergency care centers). Also investigated were the number of visits made to a physician, the number of hospitalizations in the past 12 months, and whether the child was up to date for well-child care,¹⁹ estimated by comparing the reported number of routine physician visits with age-specific guidelines established by the American Academy of Pediatrics.³⁶ Also assessed was whether the child had ever seen a psychiatrist or counselor or received special education services. No detailed information regarding the specific nature of special education services received was available, thus, it was not possible to determine whether parents were referring to educational counseling, tutoring, or other school-based support services, or to their children being placed in totally separate classes or educational tracks.

Statistical Analyses

² Tests were used to test for differences in weighted proportions. Logistic and linear regression modeling was conducted to investigate the independent association of children's race and aspects of their physical health, behavior and school problems, sources of ambulatory care, and utilization of services.

All analyses were conducted on an IBM-compatible PC. PC SAS software was used for the initial analyses.³⁷ SUDAAN software was used to obtain precise confidence intervals, accounting for the complex, two-stage survey design.³⁸

	RESULTS

Top Abstract Methods Results Discussion References

Of the ~64 million children in the United States in 1988, 48.3% (4.3 million) of black and 67% (32.5 million) of white children lived in families whose incomes were in the second, third, and fourth quintiles of family incomes (P < .001). Of these middle class children, 75.9% of black and 85.3% of white youth had private health insurance (P < .001). All subsequent results and all tabular presentations of data refer to middle class children who were reported to have had private health insurance.

Demographic and Family Characteristics

Among middle class children who had private health insurance, black children were less affluent and more likely to live in single-parent households (31.3% vs 12.4%; P < .001) of larger size (3.62 vs 3.34 family members per household; P = .003) with mothers whose age at their birth was 16 years (3.6% vs 1.4%; P < .01) (Table 1). Maternal educational status did not differ between groups.

Physical Health, Behavior, and School Problems

Black children were reported as having fewer chronic health conditions, fewer days in bed because of illness, and comparable rates of limitations of activity. However, they were less likely to be reported to be in excellent health (46.2% vs 57.3%; P < .001) (Table 2). They also were more likely to have been born with a low birth weight (10.7% vs 5.6%; P < .001) and to have asthma (8.5% vs 5.8%; P < .01).

Whereas there were no differences in rates of scoring in the top 10th percentile on the BPI or in being perceived by parents as needing psychological help in the past 12 months, black children were almost twice as likely to have ever repeated any grade (20.0% vs 12.3%; P < .001), and they were more than twice as likely to have ever been suspended from school (11.3% vs 5.0%; P < .001).

Utilization of Medical, Mental Health, and Special Education Services

Although comparable percentages were reported as having a usual source of routine care, and there were no differences in rates of being up to date with well-child care or in the number of hospital episodes in the past 12 months, on every other measure, the two groups differed (Table 3). Black children lacked a source for sick care (6.6% vs 3.8%; P < .01) or continuity of routine and sick care more frequently (13.8% vs 7.5%; P < .001), used a physician's office for routine (79.4% vs 92.8%; P < .001) or sick care (81.8% vs 92.0%; P < .001) less frequently, and were more than twice as likely to have used emergency departments for sick care (5.4% vs 2.4%; P < .01). They also made fewer ambulatory care visits, were less likely to receive mental health services (3.6% vs 5.8%; P < .05), and among children who repeated a grade or were suspended from school, they were substantially less likely to have received special education (11.8% vs 18.8%; P < .05) or mental health services (4.0% vs 13.2%; P < .001).

Children Whose Family Incomes Were in the Third and Fourth Quintiles

Analyses restricted to children whose family incomes were in the third and fourth quintiles revealed findings similar to those from analyses that also included children of less affluent families having incomes in the second quintile. Although the findings in the more restricted and smaller sample did not reach statistical significance for some measures, the trend remained similar to that found in the larger middle class sample. For example, black children still were more likely to lack the same source of care for routine and sick care (10.1% vs 6.5%; P = .06), more likely to use emergency departments as their source of sick care (4.2% vs 2.2%; P = .07); and less likely to receive mental health (4.4% vs 5.9%; P = .22) or special education services (2.8% vs 4.1%; P = .13). Among children who experienced educational difficulties, black children in these quintiles also were less likely to receive both special education (9.6% vs 19.4%; P < .05) and mental health (6.4% vs 14.5%; P < .06) services.

Multivariate Analyses

Table 5 demonstrates adjusted ORs for selected measures of health, behavior and school problems, and sources of care and use of services, controlling for family income and size, source of routine care, region, urbanization, gender, single- vs two-parent household, and maternal age at child's birth. Black children in these quintiles still were less likely to be reported as being in excellent health (OR: 0.7, 95% CI: 0.6-0.8), and they still were more likely to have been born with a low birth weight (OR: 1.9, 95% CI: 1.3-2.8) and to have asthma (OR: 1.4, 95% CI: 1.01-2.0; P < .05). Black children also remained at increased risk for having ever been suspended from school (OR: 2.2, 95% CI: 1.5-3.3) and tended to have higher rates of repeating a grade (OR: 1.4, 95% CI: 0.98-1.9; P = .07) in adjusted analyses.

In multivariate analyses, race no longer was associated with disparities in lack of a usual source of sick care, but black children still were more likely to lack continuity of routine and sick care. The likelihood of black children overall ever having received mental health or special education services, although still less than that of white children, was now of marginal statistical significance. Similarly, among children who had ever repeated a grade or had ever been suspended from school, although no longer statistically significant, the trend was still for black children to be less likely to have received special education services or to have seen a psychiatrist or counselor. Black children still were found to use physicians' offices for routine care half as often (OR: 0.4, 95% CI: 0.3-0.5) and emergency departments for sick care three times as often as white children (OR: 3.2, 95% CI: 1.7-5.8). Not shown are data from a linear regression model indicating that black children still had fewer physician visits than did white children (P < .001).

	DISCUSSION

Top Abstract Methods Results Discussion References

This study differs from other studies of children's health and use of health services in that it focuses on middle class black and white children who have private health insurance. Black and white children were found to have comparable rates of being up to date with well-child care. In contrast, they differed substantially on many measures of utilization of services, and most of these differences persisted in multivariate analyses and analyses restricted to more economically privileged middle class children, as well as across the entire middle class income range. Black children were more likely to lack a usual source for sick care and continuity of routine and sick care, get their care from outpatient clinics and health centers, and use emergency rooms, and they made fewer physician visits.

Privately insured middle class black children were more likely to have been born with low birth weight, a finding consistent with other studies showing this same difference among infants born to nonpoor parents.^39,40 They also were more likely to have asthma, a finding compatible with results of other studies^3241-43 that show that higher rates of this condition among black children cannot be explained entirely by social or economic factors. In contrast, black children were reported as having fewer chronic health conditions overall and similar rates of limitations of activity because of these conditions, yet they were less likely to be perceived by their parents as being in excellent health. Although black children and white children had similar rates of behavior problems, black children were twice as likely to have repeated a grade or to have been suspended from school. In multivariate analyses, the association of children's race and repeating a grade was no longer statistically significant (OR: 1.4; P = .09), but black children were still more than twice as likely to have been suspended from school compared with their white peers. We are unaware of any previous studies indicating this heightened risk for black middle class children.

Other studies, although not focusing on middle class children, also have found increased rates of discontinuity of routine and sick care, increased use of hospital outpatient departments and community health centers,^1,2,18,44 and decreased numbers of ambulatory care visits by black children.^13,45,46 Although it appears that the increased rates of lacking continuity of care were attributable to their increased use of community health centers and hospital outpatient departments, this does not appear to account for black children's diminished use of private physicians for routine care, their increased reliance on emergency departments for sick care, or their fewer ambulatory care visits, because these differences persisted in multivariate analyses that controlled for their source of routine care. The data available provide no insights into whether these differences are attributable to nonfinancial barriers to care,^46,47 such as long waits and diminished evening and weekend availability at hospital outpatient departments and community health centers; longer distance to or more difficult access to private physicians' offices; families feeling out of place in some practice settings; or overt or covert discrimination.

There are other limitations to these data. All data were from parent report, with limited information available about many important aspects of children's insurance and health care. Potentially important unassessed factors include the comprehensiveness, cost-sharing (ie, copays or deductibles), and limits of insurance coverage; content, comprehensiveness, and cultural appropriateness of care received; variability of any of these characteristics with degree of urbanization, specific city, state, or region of the United States; and factors influencing choice of sources of care. There also was no information available regarding family wealth (ie, financial assets), and resources such as home ownership, cars, other material possessions and liquid savings might influence the patterns of service use noted. Also, the data are 10 years old, predating the rapid introduction of managed care into many communities.

Despite the limitations, these data offer what we believe is the first detailed picture of the health, behavior and school problems, and use of medical, mental health, and special education services of black and white middle class children in the United States. They demonstrate similar rates of health and behavior problems, but substantially increased rates of school difficulties among black children. They also demonstrate no differences in rates of having a usual source of routine care being current with well-child care, or receiving mental health or special education services. In contrast, even in the presence of private health insurance, middle class black children had markedly different sources and patterns of ambulatory medical service use. Although the findings do not identify the mechanisms behind the differences, they clearly demonstrate the increased risk and relative disadvantage that black race appears to portend for middle class children in the United States. These and similar findings, we believe, are crucial to our efforts to ensure equal access to medical services for all our children, irrespective of their income, insurance, or race.

	FOOTNOTES

Received for publication Dec 26, 1998; accepted Feb 4, 1999.

This work was presented in part at the 37th Annual Meeting of the Ambulatory Pediatric Association; May 1997; Washington, DC.

Address correspondence to Michael Weitzman, MD, Department of Pediatrics, Rochester General Hospital, 1425 Portland Ave, Rochester, NY 14621. E-mail: michael.weitzman{at}viahealth.org

	ABBREVIATIONS

BPI, Behavior Problem Index.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Blendon RJ, Aiken LH, Freeman HE, Corey CR Access to medical care for black and white Americans. JAMA. 1989; 261:278-281 [Abstract/Free Full Text]
2. Council on Ethical and Judicial Affairs Black-white disparities in health care. JAMA. 1990; 263:2344-2346 [Abstract/Free Full Text]
3. Strogatz D Use of medical care for chest pain: differences between blacks and whites. Am J Public Health. 1990; 80:290-294 [Abstract/Free Full Text]
4. Navarro V Race or class versus race and class: mortality differentials in the United States. Lancet. 1990; 336:1238-1240 [CrossRef][Medline]
5. Council on Scientific Affairs of the American Medical Association Hispanic health in the United States. JAMA. 1991; 265:248-252 [Abstract/Free Full Text]
6. Public Health Service. Healthy People 2000: National Health Promotion and Disease Prevention Objectives. Washington, DC: US Dept of Health and Human Services, Public Health Service. DHHS publication no. (PHS)91-50212; 1991
7. Lieu TA, Newacheck PW, McManus MA Race, ethnicity, and access to ambulatory care among US adolescents. Am J Public Health. 1993; 83:960-965 [Abstract/Free Full Text]
8. Gornick ME, Eggers PW, Reilly TW, Effects of race and income on mortality and use of services among Medicare Beneficiaries. N Engl J Med. 1996; 335:791-799 [Abstract/Free Full Text]
9. Geiger HJ Race and health carean American dilemma? N Engl J Med. 1996; 335:815-816 [Free Full Text]
10. Peterson ED, Shaw BS, DeLong ER, Pryor DB, Califf RM, Mark DB Racial variation in the use of coronary revascularization procedures. N Engl J Med. 1997; 336:480-486 [Abstract/Free Full Text]
11. Adler NE, Boyce T, Chesney MA, Folkman S, Syme SL Socioeconomic inequalities in health: no easy solution. JAMA 1993; 269:3140-3145 [Abstract/Free Full Text]
12. Pappas G Elucidating the relationships between race, socioeconomic status, and health. Am J Public Health. 1994; 84:892-893 [Free Full Text]
13. Kleinman JC, Gold M, Makuc D Use of ambulatory care by the poor: another look at equity. Med Care. 1981; 19:1011-1029 [CrossRef][Medline]
14. Egbuonu BS, Starfield B Child health and social status. Pediatrics. 1982; 69:550-557 [Abstract/Free Full Text]
15. Starfield B Family income, ill health, and medical care of US children. J Public Health Policy. 1982; 3:244-259 [CrossRef][Medline]
16. Newacheck PW, Halfon N Preventive care use by school-aged children. Differences by socioeconomic status. Pediatrics. 1988; 82:462-468 [Abstract/Free Full Text]
17. Wise PH, Meyers A Poverty and child health. Pediatr Clin North Am. 1988; 35:1169-1186 [Medline]
18. St. Peter RF, Newacheck PW, Halfon N Access to care for poor children: separate and unequal? JAMA. 1992; 267:2760-2764 [Abstract/Free Full Text]
19. Holl JL, Szilagyi PG, Rodewald LE, Byrd RS, Weitzman M Profile of uninsured children in the United States. Arch Pediatr Adolesc Med. 1995; 149:398-406 [Abstract/Free Full Text]
20. Sheils JF, Wolfe PR The role of private health insurance in children's health care. Future Child. 1992; 2:115-133
21. Adams PF, Hardy AM. Current estimates from the National Health Interview Survey: United States, 1988. Vital Health Stat. 1989;173:1-11, 133-144
22. Bloom B. Health Insurance and Medical Care. Health of Our Nation's Children, United States, 1988. Hyattsville, MD: National Center for Health Statistics. Advance data from Vital Health Stat. No. 188; 1990
23. US Bureau of the Census. Middle Class and Income Inequality. Current Population Reports, Series P60. Washington, DC: US Bureau of the Census; 1997. http://www.census.gov/hhes/income/midclass/midclsan.html
24. Massey JT. Design and Estimation for the National Health Interview Survey, 1985-94. Data Evaluation and Methods Research. Vital Health Stat-Series 2;1989;110:1-33
25. Collins JG, Hendershot GE. Health Characteristics of Large Metropolitan Statistical Areas: United States, 1988-89. Vital Health Stat-Series 10; 1993;187:1-97. Data from the National Health Survey. US Bureau of the Census. Statistical Abstracts of the United States: 1993 (113th ed). Washington, DC: Government Printing Office; 1993
26. Newacheck PW, Stoddard JJ, McManus P Ethnocultural variations in the prevalence and impact of childhood chronic conditions. Pediatrics. 1993; 91:1031-1039 [Abstract/Free Full Text]
27. Weitzman M, Byrd RS, Auinger P Children in big cities in the United States: health and related needs and services. Ambulatory Child Health. 1996; 1:347-359
28. Peterson JL, Zill N Marital disruption, parent-child relationships, and behavior problems in children. J Marriage Fam. 1986; 48:295-307 [CrossRef]
29. Achenbach TM, Edelbrock C. Manual for the Child Behavior Checklist and Revised Child Behavior Profile. Burlington, VT: Queen City Printers; 1983
30. Gortmaker SL, Walker DK, Weitzman M, Sobol AM Chronic conditions, socioeconomic risks, and behavioral problems in children and adolescents. Pediatrics. 1990; 85:267-276 [Abstract/Free Full Text]
31. Byrd RS, Weitzman M Predictors of early grade retention among children in the United States. Pediatrics. 1994; 93:481-487 [Abstract/Free Full Text]
32. Weitzman M, Gortmaker SL, Sobol AM, Perrin JM Recent trends in the prevalence and severity of childhood asthma. JAMA. 1992; 268:2673-2677 [Abstract/Free Full Text]
33. McCormick MC, Gortmaker SL, Sobol AM Very low birthweight children: behavior problems and school difficulty in a national sample. J Pediatr. 1990; 117:687-693 [Medline]
34. Weitzman M, Gortmaker SL, Sobol AM Maternal smoking and behavior problems of children. Pediatrics. 1992; 90:342-349 [Abstract/Free Full Text]
35. Byrd RS, Weitzman M, Auinger P. Behavior problems, age-for-grade and grade retention among children in the United States. Pediatrics. In press
36. American Academy of Pediatrics. Guidelines for Health Supervision II. Elk Grove Village, IL American Academy of Pediatrics; 1988
37. SAS Institute Inc. SAS/STAT User's Guide: Statistics. Release 6.03 ed. Cary, NC: SAS Institute Inc; 1988
38. Shah BV, Barnwell BG, Hunt N, LaVange LM. SUDAAN User's Manual Release 6.40. Research Triangle Park, NC: Research Triangle Institute; 1995
39. Barfield WD, Wise PH, Rust FP, Rust KJ, Gould JB, Gortmaker SL Racial disparities in outcomes of military and civilian births in California. Arch Pediatr Adolesc Med. 1996; 150:1062-1067 [Abstract/Free Full Text]
40. Schoendorf KC, Hogue CJR, Kleinman JC, Rowley D Mortality among infants of black as compared with white college-educated parents. N Engl J Med. 1992; 326:1522-1526 [Abstract]
41. Weitzman M, Gortmaker S, Sobol A Racial, social, and environmental risks for childhood asthma. Am J Dis Child. 1990; 144:1189-1194 [Abstract/Free Full Text]
42. Cunningham J, Dockery DW, Speizer FE Race, asthma, and persistent wheeze in Philadelphia school children. Am J Public Health. 1996; 86:1406-1409 [Abstract/Free Full Text]
43. Gergen P Social class and asthmadistinguishing between the disease and the diagnosis. Am J Public Health. 1996; 86:1361-1362 [Free Full Text]
44. Hahn BA Children's health. Racial and ethnic differences in the use of prescription medications. Pediatrics. 1995; 95:727-732 [Abstract/Free Full Text]
45. Wood DL, Hayward RA, Corey CR, Freeman HE, Shapiro MF Access to medical care for children and adolescents in the United States. Pediatrics. 1990; 86:666-673 [Abstract/Free Full Text]
46. Orr ST, Miller CA, James SA Differences in use of health services by children according to race: relative importance of cultural and system-related factors. Med Care. 1984; 22:848-853 [CrossRef][Medline]
47. Margolis PA, Carey T, Lannon CM, Earp JL, Leninger L The rest of the access-to-care puzzle: addressing structural and personal barriers to health care for socially disadvantaged children. Arch Pediatr Adolesc Med. 1995; 149:541-545 [Abstract/Free Full Text]