Insuring the children: a decade of change.

Cite this article as: P J Cunningham and A C Monheit Insuring the children: a decade of change Health Affairs 9, no.4 (1990):76-90 doi: 10.1377/hlthaff.9.4.76

The online version of this article, along with updated information and services, is available at: http://content.healthaffairs.org/content/9/4/76.c itation

For Reprints, Links & http://content.healthaff Permissions : airs.org/1340_reprints .php Email Alertings :

http://content.healthaffairs.or g/subscriptions/etoc.dtl

Not for commercial use or unauthorized distribution

Downloaded from http://content.healthaffairs.org/ by Health Affairs on June 21, 2016 by HW Team

At the Intersection of Health, Health Care and Policy

To Subscribe :

Health Affairs is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600, Bethesda, MD 20814-6133. Copyright © by Project HOPE - The People-to-People Health Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical, including photocopying or by information storage or retrieval systems, without prior written permission from the Publisher. All rights reserved.

Not for commercial use or unauthorized distribution


https://fulfillment.healthaffair s.org

by Peter J. Cunningham and Alan C. Monheit Prologue: The hard-fought budget legislation passed at the end of the 101st Congress’s second session included a provision to expand Medicaid coverage amidst cuts in other social entitlement programs. The expansion would require states and the federal government to phase in coverage of all poor children through age eighteen by the end of the decade. According to The New York Times (4 November 1990) the Children’s Medicaid Coalition, whose members include the Health Insurance Association of America, the U.S . Chamber of Commerce , and children’s interest groups, were “motivated by a combination of altruism and financial self-interest” to exert sufficient political pressure to get the expansion through Congress. Medicaid is but one factor in the insurance status of America’s children, however, as Peter Cunningham and Alan Monheit of the Agency for Health Care Policy and Research (AHCPR) report in this article. In their analysis, they focus on the role different family characteristics play in determining whether children’s health care is covered adequately. They use data from over a decade of national medical surveys to truce changes in children’s insurance coverage. Cunningham, a service fellow at AHCPR, received a doctoral degree in sociology from Purdue University in West Lafayette, Indiana. During his four-year tenure at the agency, he has worked on the 1987 National Medical Expenditure Survey. Monheit, a senior economist at AHCPR, received his doctorate in economics from the Graduate Center, City University of New York. In his years at AHCPR and its predecessor agency, the National Center for Health Service Research and Health Care Technology Assessment, he has been deeply involved in research into expenditures for medical care and the causes for uneven insurance coverage in this country.


INSURING THE CHILDREN: A DECADE OF CHANGE

INSURING THE CHILDREN 77

D


uring the past three decades, profound changes have affected the composition and economic well-being of the American family. The rising divorce rates of the 1960s and 1970s, growth in the labor-force participation of women, stagnation in male earnings, and changing social mores have all contributed to significant growth in the 1 proportion of American families headed by single mothers. Since the mid-1970s, economic resources available to the family, as measured by median family income adjusted for inflation, have risen little and at times 2 actually declined. These changes in the American family have caused growing concern that the social and economic well-being of American children is deteriorating. For example, one-fifth of all children live in households with incomes below the federal poverty line, which was $12,675 for a family of four in 1989. Poverty is especially prevalent among children in families headed by single mothers, which now include almost one-fourth of all children. In fact, the majority of today’s children can expect to spend part of their childhood in a single-parent family, many of which are female3 headed, low-income households. The economic circumstances of children are directly related to the types of families in which they live, and both have important implications for their patterns of health insurance coverage. For instance, poverty is especially prevalent among the quarter of all children in families headed by single mothers. Single mothers often find it difficult to work full time while caring for their children. When they do work, they often lack the skills and experience necessary for higher-paying jobs that provide health insurance. Additional sources of income, including child support, alimony, and welfare, are usually insufficient to free single parents and their children from poverty, let alone to directly purchase private insurance. In contrast, children in two-parent families benefit economically from high rates of employment among their parents, from the higher wages typically commanded by male workers, and, increasingly, from both parents participating in the labor force. Indeed, the latter has played an important role in stemming the economic decline of the two-parent 4 family over the past decade. Although children in two-parent families traditionally have relied on private insurance, usually obtained as a fringe benefit at the father’s place of employment, the growth in two-worker households also enhances the likelihood that at least one parent will be offered employment-related health insurance coverage. In addition, it provides a level of family income sufficient to pay for the rising costs of 5 employment-related family coverage. At the same time that private family coverage has become more expensive for working parents, public insurance has become less acces-

78 HEALTH AFFAIRS | Winter 1990

Sources Of Data Both the 1987 NMES and 1977 NMCES are year-long panel surveys of the medical care use, expenditures, and health insurance coverage of the U.S. population. Information on insurance coverage, employment, and income was derived from the household components of both surveys. The NMES household component surveyed approximately 15,000 households consisting of 36,000 individuals, while the NMCES household component included 16,000 households consisting of 40,000 persons. NMCES insurance information was also supplemented with information on employment-related coverage derived from the Health 9 Insurance Employer Survey (HIES). Estimates of children’s health insurance coverage for both 1977 and 1987 are point-in-time estimates for 10 approximately the first quarter of each year. Data from NMES and NMCES are a particularly rich and accurate source of information on the health insurance coverage of children. In


sible, for at least two reasons. First, states have failed to increase their income eligibility standards for Aid to Families with Dependent Children (AFDC) with the rate of inflation. This has effectively reduced Medicaid eligibility for persons with small amounts of earned income. Second, the early 1980s witnessed a significant retrenchment in eligibility rules for Medicaid. The Omnibus Budget Reconciliation Act of 1981 (OBRA 1981) restricted the Medicaid eligibility of children with working parents 6 by limiting the gross income of families applying for AFDC. Since then, beginning with the Deficit Reduction Act of 1984 (DEFRA 1984), incremental expansions of Medicaid have restored eligibility for some children, especially for poor and very young children. Despite these expansions, eligibility levels have not returned to their pre-1981 levels, and children of the working poor-including those in single-mother 7 families-are still largely excluded. Thus, the likelihood that the growing number of poor children can obtain public medical insurance has also diminished over the past decade. In this article, we examine the health insurance status of children according to family type, parents’ employment status, and family income. We do so to determine the characteristics of families in which children are most at risk for no health insurance coverage and, therefore, at a disadvantage regarding access to, quality of, and continuity of health 8 care. Our analysis documents the changes in children’s health insurance coverage over a ten-year period using data from the 1987 National Medical Expenditure Survey (NMES) and the 1977 National Medical Care Expenditure Survey (NMCES).


Family Characteristics The vast majority of children in 1987 lived in two-parent families (Exhibit 1). However, the percentage of children in single-parent families

Exhibit 1 Distribution Of Children Across Family Types, 1977-1987 Family structure, 1987

Population (in thousands)

Percent of children

Percent poora

Low incomeb

Middle and upper incomec

Total

63,073

100.0%

18.1%

21.1%

60.8%

46,663 16,345 14,880

74.0 25.9 23.6

11.0 38.1 40.3

18.5 28.5 29.2

70.5 .33.4 30.4

63,776

100.0%

13.6%

22.0%

64.4%

52,332 10,742 9,201

82.1 16.8 14.4

7.7 41.2 45.3

21.0 26.6 26.8

71.3 32.2 27.9

27,069 17,995 1.599

42.9% 28.5 2.5

5.6% 16.1 43.8

12.8% 25.3 38.3

81.6% 55.6 17.9

9,021 7,324

14.3 11.6

24.4 54.9

25.2 32.6

50.4 12.5

7,764 7,090

12.3 11.2

26.4 55.7

26.4 32.3

47.2 12.0

Two parents Single parent d Single parent female Family structure, 1977 Total Two parents Single parent d Single parent female

Employment status of parents, 1987 Two parents Both employed One employed Neither employed All single parent Employed Not employed d

Single parent female Employed Not employed

Sources: Agency for Health Care Policy and Research, National Medical Expenditure Survey, 1987; and National Medical Care Expenditure Survey, 1977. a Below federal poverty level. b Includcs 100 to 199 percent of the federal poverty level. c Includes 200 percent and above of the federal poverty level. d Children in single-parenr, female-headed families comprise a subgroup of all children in single-parent families.


contrast to the determination of health insurance status in the Current Population Survey (CPS) between 1980 and 1987, questions regarding the presence and type of health insurance coverage (both public and private) in NMES and NMCES are asked of each person in the household (including children), so that children covered by persons outside the household (for example, by a divorced parent) are included among the insured. Since CPS has only recently adopted this procedure, it has been argued that CPS estimates of the uninsured prior to 1988 may seriously 11 overstate the number of uninsured children.


Health Insurance Coverage Of Children The percentage of children without private or public health insurance coverage grew sharply (40 percent) between 1977 and 1987 (Exhibit 2). In 1987, 17.8 percent of children age seventeen or younger were uninsured (11.2 million uninsured children), compared to 12.7 percent of all children (8.1 million uninsured children) a decade earlier. This growth in the number of uninsured children reflects the corresponding decline in the private health insurance coverage of children since 1977. At that time, almost three-quarters of all children had private coverage, primarily through their parents’ employment-related coverage. However, by 1987, just over two-thirds of children retained such coverage. This loss of private coverage was concentrated almost exclusively among children in two-parent, single-worker households. Private coverage of children in these households declined from 82.6 percent in 1987 to just over 70 percent in 1987, with the employment-related coverage of these children declining from three-quarters to two-thirds. The decline in coverage reflects the fact that children in two-parent families with one worker were far less likely to have a working parent who was offered employment-related coverage than were children in two-


increased from 16.8 percent in 1977 to 25.9 percent in 1987. The economic status of children varies dramatically across family types (Exhibit 1). In 1987, over 70 percent of children in two-parent families lived in middle- or upper-income households, while two-thirds of children in single-parent families lived in poor or low-income households. However, between 1977 and 1987 there was also a significant increase in the percentage of children in two-parent families who were poor or near-poor (from 7.7 percent to almost 11 percent). The economic status of children also varies significantly according to their parents’ employment status (Exhibit 1). Children living in twoparent families with both parents employed have the highest economic status of all, with over 80 percent in middle- or upper-income families. By comparison, over 40 percent of children in two-parent families where only one parent is employed are poor or low-income. Over half of all children from single-parent families where the parent is employed are poor or low-income. Thus, even where there is only one wage earner in the family, children in two-parent families still have an economic advantage over children in single-parent families. This reflects the disparity between the earnings of males, who are most likely to be the employed parent in two-parent, one-worker households, and females, who primarily head single-parent households.


Exhibit 2 Health Insurance Of Children, By Family Type, 1977-1987 Private insurance

Employmentrelated, private

Public insurance only

Uninsured

1987 total

67.6%

62.9%

14.6%

17.8%

Two parents Both employed One employed Neither employed

77.6 85.9 70.4 17.1

72.8 81.0 66.2 7.1

6.2 3.1 7.7 41.3

16.2 11.0 21.9 41.6

All single parent Employed Not employed

39.2 63.3 9.6

34.8 57.3 6.7

38.6 13.4 69.7

22.1 23.3 20.8

a

36.2 61.0 9.1

31.5 54.6 6.2

41.7 15.1 70.9

22.1 23.9 19.9

1977 total

73.6%

67.7%

13.6%

12.7%


82.4 86.6 82.6 15.3

76.3 82.7 74.9 9.7

5.6 4.0 5.2 51.2

12.0 9.5 12.2 33.6


34.5 7.8 .

28.9 45.7 2.6

51.3 33.9 78.0

14.2 14.1 14.1

32.7 50.7 7.6

27.6 45.9 1.7

53.9 35.4 79.3

13.4 13.9 13.2


a


Sources: Agency for Health Care Policy and Research, National Medical Expenditure Survey, 1987; and National Medical Care Expenditure Survey, 1977. a Children in single-parent, female-headed families comprise a subgroup of all children in single-parent families.

parent households where both parents were employed. Moreover, the difficulty two-parent, single-worker households experienced in maintaining their standard of living during a period in which real earnings of full-time, full-year male workers were declining was likely to preclude the 12 purchase of private insurance coverage for their children. In contrast, children with two working parents had the highest rates of private coverage in both 1977 and 1987, were least likely to be uninsured, and experienced little change in their health insurance status over the decade. Perhaps an even more significant change in children’s health insurance coverage between 1977 and 1987 was the dramatic decline in the percentage of children covered by public insurance in households with employed single parents. In 1977, one-third of such children were covered by public insurance (primarily Medicaid), but by 1987, only 13.4 percent of children in these families had such coverage. This decline in public


Children’s characteri stics



coverage occurred at the same time that children in single-parent families experienced an increase in private insurance coverage, largely because more working single parents were able to obtain employment-related coverage. However, this increase in private coverage was not sufficient to offset the decline in public coverage of children in single-parent families with a working parent (or in single-parent families generally) over this period. The corresponding rise in the percentage of uninsured children in single-parent families-from 14.2 percent in 1977 to 22.1 percent in 1987-reflects the significant retrenchment in Medicaid eligibility that affected the working poor under OBRA 1981 and the failure of state governments to adjust income eligibility standards for AFDC to keep pace with inflation. Subsequent expansions of the Medicaid program, beginning with DEFRA 1984 and culminating with OBRA 1989, have required states to extend coverage to pregnant women and young children, but as of 1987 had not significantly altered the health insurance status of many children with employed single-parents. To examine how changes in the rates of private and public health insurance coverage since 1977 contributed to the number of uninsured children in 1987, we applied 1977 rates to the 1987 population (according to family type and insurance status, as shown in Exhibit 2). We then compared these estimates to the actual number of children with coverage in 1987 to compute the change in health insurance coverage over the decade. The sum of these changes in coverage across family types yields the net change in children’s health insurance due to coverage shifts 13 alone. Our computations reveal that had coverage rates remained at their 1977 levels, 7.6 million children would have lacked health insurance in 1987. Compared to the actual number of uninsured children in 1987 (11.2 million), the shift in rates of health insurance coverage between 1977 and 1987 was responsible for the net addition of 3.6 million 14 uninsured children. Two-thirds of this net increase resulted from a decline in the public health insurance coverage of children, primarily those in households with an employed single parent. The remaining third resulted from declining private insurance coverage, largely affecting children in two-parent, one-worker households. Our computations also indicate that overall, 48 percent of the increase in the number of uninsured children occurred in two-parent, single-worker households, while 23 percent of the increase was borne by children in employed single-parent households. Children in nonworking single-parent households represented 13 percent of the increase in uninsured children; those in two-parent, two-worker families represented 12 percent of the increase;

INSURING THE CHILDREN 83 and those in two-parent households where neither parent worked accounted for 4 percent of the increase. Health Insurance Of Poor And Low-Income Children

Exhibit 3 Health Insurance Coverage Of Children, By Income Level, 1977-1987 Income level


Private insurance

Employments related, private

Public insurance only Uninsured

11,356 13,239 23,515 14,632

26.3% 49.3 83.0 91.9

22.2% 44.4 78.9 86.2

37.6% 23.7 5.8 3.1

36.0% 27.0 11.3 5.0

8,681 14,020 27,210 13,065

27.6 62.0 85.0 90.5

22.6 56.7 79.7 83.2

50.3 16.9 5.9 2.7

21.8 20.9 8.8 6.5

1987 a

Po or b Low income c Middle income d Upper income 1977 a

Poor b Low income c Middle income d Upper income

Sources: Agency for Health Care Policy and Research, National Medical Expenditure Survey, 1987; and National Medical Care Expenditure Survey, 1977. Note: Population subtotals may not sum to total population due to rounding. a Below federal poverty level. b Includes 100 to 199 percent of federal poverty level. c Includes 200 to 399 percent of federal poverty level. d Includes 400 percent and above of federal poverty level.


Many of the gaps in children’s health insurance coverage are exacerbated for children from poor and low-income families. Their parents are less likely to have access to private employment-related insurance than are the working parents of children in higher-income families. Changes in eligibility standards over the past decade make dependency on public programs somewhat tenuous for poor and low-income children. Over four million poor children (36 percent) were without health insurance in 1987 (Exhibit 3). This represents a substantial increase from the 21.8 percent reported in 1977. Most of this increase can be attributed to a substantial loss in public insurance-from 50.3 percent in 1977 to 37.6 percent in 1987. The likelihood of being without health insurance for low-income children (100-199 percent of the federal poverty level) increased from 20.9 percent in 1977 to 27 percent in 1987. In contrast to poor children, low-income children experienced a substantial loss in private coveragefrom 62 percent in 1977 to 49.3 percent in 1987. While the risk of being without health insurance has worsened for all poor and low-income children, the risk has grown substantially for those in certain family types (Exhibit 4). Despite having much higher rates of

84 HEALTH AFFAIRS | Winter 1990 Exhibit 4 Health Insurance Of Poor And Low-Income Children, By Family Type, 1977-1987 Private insurance

Employmentrelated, private

Public insurance only

Uninsured

1987 total

38.7%

34.1%

30.1%

31.2%


50.5 64.3 47.3 16.3a

45.2 57.2 44.4 4.9a

13.1 6.3 12.7 40.9

36.4 29.4 40.0 42.8


23.9 46.7 8.0

20.2 41.0 5.6

51.6 22.6 71.8

24.5 30.6 20.2


22.3 44.5 7.9

18.5 38.4 5.4

53.4 24.1 72.6

24.3 31.4 19.6

1977 total

49.1%

43.7%

29.7%

21.2%

Two parents Both employed One employed Neither employed All single parent Employed Not employed

65.3 67.3 71.0 9.5a

59.2 61.9 64.2 6.8a

10.8 7.3 9.6 54.8

23.9 25.4 19.4 35.7

18.5 34.1 4.0a

14.3 29.1 0.4”

67.7 51.4 83.8

13.8 14.5 12.2

18.7 33.5 4.5a

14.4 29.2 0.4’

68.6 52.7 84.0

12.8 13.8 11.5

b

b


Sources: Agency for Health Care Policy and Research, National Medical Expenditure Survey, 1987; and National Medical Care Expenditure Survey, 1977. a Standard error is greater than 30 percent of the estimate. b Children in single-parent, female-headed families comprise a subgroup of all children in single-parent families.

private insurance, poor and low-income children in two-parent households are more likely to be without health insurance than are poor and low-income children in single-parent households (36.4 percent versus 24.5 percent). This is due largely to the greater accessibility of Medicaid to children of single parents. Poor and low-income children in two-parent families experienced a sharp decrease in private coverage, from 65.3 percent in 1977 to 50.5 percent in 1987. Even worse, children in two-parent families with only one working parent saw a precipitous decline in coverage-from 71 percent in 1977 to 47.3 percent in 1987. This dramatic decline in coverage reflects the growth in the proportion of these children in families of the working poor (less than 125 percent of the federal poverty line). The representation of children in two-parent, one-worker families considered working poor doubled between 1977 and 1987, from 10.7 percent of children to 21.8 percent. The likelihood that such children would be


Children’s characteristics


Availability Of Employment-Related Private Coverage The findings thus far show significant differences in the rates of private insurance coverage-primarily employment-related coverage-for children across family type and income categories, even when controlling for parents’ employment status. These disparities may be a result of differences in (1) whether parents are offered health insurance at their jobs; (2) whether parents accept insurance when offered; or (3) a combination of the two. The findings from Exhibit 5 strongly support the first hypothesis. Children with two working parents have the highest rates of available work-related coverage (85.3 percent), followed by children in two-parent, one-worker families (69.6 percent) and children with an employed single parent (61.8 percent). Having both parents active in the labor force is especially important for poor children. The likelihood that at least one of their parents will be offered health insurance is much higher than when only one parent is working (59.4 percent versus 35.6 percent). The rate of parents’ accepting insurance coverage when offered is consistently high, even across family type and income categories (Exhibit 5). Despite some small differences, the rate of acceptance is over 90 percent for all groups, with the exception of poor children in single-parent families, where the rate of acceptance is 82 percent. Conclusions And Policy Implications There are profound differences in the types of health insurance coverage available to children from different family types; such coverage has indeed changed over the past decade. Children living in a variety of


uninsured also doubled over this period from 19.4 percent in 1977 to 40 percent in 1987. Poor and low-income children in two-parent families with one working parent accounted for nearly 60 percent of all such children in two-parent families who were without health insurance. For poor and low-income children in single-parent families, the decline in coverage largely resulted from a loss in public coverage-from 67.7 percent in 1977 to 51.4 percent in 1987. This decline was even more dramatic for poor and low-income children with a working single parent-from 53.9 percent in 1977 to 22.6 percent in 1987. To some extent, substantial gains in private coverage compensated for the loss in public coverage for poor and low-income children with a single parent. However, even for poor and low-income children with an employed single parent, the likelihood of being uninsured increased dramatically-from 14.5 percent in 1977 to 30.6 percent in 1987.

86 HEALTH AFFAIRS | Winter 1990 Exhibit 5 Availability Of Health Insurance To Children With Employed Parents, 1987 Two parents

Insurance offered to parents b Poor c Other low income Middle and d upper income Accepted insurance when offered b Poor Other low income’ Middle and upper income” Insurance not offered b Poor Other low income’ Middle and d upper income

Al l

All

53,654

44,629

27,013

17,616

9,021

7,764

76.2% 42.5 60.9

79.1% 43.8 63.3

85.3% 59.4 67.6

69.6% 35.6 60.0

61.8% 39.9 52.4

58.5% 37.2 49.4

86.9

88.3

90.2

84.2

77.3

75.7

96.7 89.0 95.8

97.3 92.3 96.0

97.3 92.9 96.7

97.3 91.7 95.4

92.9 82.0 95.0

92.1 80.2 94.8

97.6

97.9

97.7

98.4

94.9

94.4

23.8 57.5 39.1

20.9 56.2 36.7

14.7 40.6 32.4

30.4 64.4 40.0

38.2 60.1 47.7

41.5 62.8 50.6

13.1

11.8

9.9

15.8

22.7

24.3

Sources: Agency for Health Care Policy and Research, National Medical Expenditure Survey, 1987. a Children in single-parent. female-headed families comprise a subgroup of all children in single-parent families. b Below federal poverty level. c Includes 100 percent to 199 percent of the federal poverty level. d Includes 200 percent and above of the federal poverty level.

different family circumstances face substantial risk of being uninsured. What is most disturbing about the changes in children’s health insurance coverage over the past decade-particularly with respect to changes in Medicaid eligibility-is that they have occurred at a time when the economic and demographic circumstances of children have worsened. Poor children and those from single-parent families have traditionally depended on Medicaid for their health insurance coverage. Yet, while both the number and percentage of children who were poor and living in single-parent families increased between 1977 and 1987, the percentage with public insurance sharply decreased, and the percentage who were without any health insurance increased. On the other hand, children in two-parent families have traditionally relied on coverage through private health insurance obtained at their parents’ (usually the father’s) place of employment. However, children in two-parent families-especially those in low-income families-exper ienced a decrease in the rate of private insurance and an increase in the percentage who were uninsured. Possible reasons for the decrease in



Single parent

Both One All employed employed single parent Femalea



private coverage include an apparent decline in the real earnings of full-time male workers and the loss of jobs in the industrial sector to the service sector, where employers are less likely to offer health insurance fringe benefits. Children in poor and low-income two-parent families are far less likely to have a working parent who is offered employment-related health insurance. The failure of these working parents to be offered health insurance appears to be a more important reason for their children’s lack of coverage than is the rising costs of family coverage. The effects of changes in Medicaid eligibility are most apparent for poor and low-income children with working parents. Medicaid is more likely to benefit poor children with a single, nonworking parent, so that these children are less likely to be uninsured. The changes in Medicaid imply that single parents face a substantial risk of losing coverage for their families if they decide to enter the labor force. States are now required by the federal Family Support Act of 1988 to continue Medicaid coverage for twelve months to families who received Medicaid but became ineligible because of increased earnings. However, after the grace period for Medicaid expires, many single parents who were previously not employed may be unable to find jobs where private health insurance is offered. Enabling such employees to buy into Medicaid through a combination of employer/ employee contributions and public subsidies would be one way to avoid such an outcome. Medicaid has never been widely accessible to most poor and lowincome children in working two-parent families. To compound the problem, these children’s reliance on private insurance has become more precarious. Apparently, there is now considerable economic pressure on both parents to be active in the labor force, not only to enable them to meet basic needs, but also to ensure that at least one parent can obtain health insurance at the workplace. These economic imperatives for two-worker families also affect other aspects of family life, especially the ability of low-income families to obtain appropriate and affordable day care for their younger children. Recent expansions of the Medicaid program will help to loosen the eligibility requirements for certain groups of children. OBRA 1987 gave states the option of covering pregnant women and infants with incomes up to 185 percent of the federal poverty line. Under provisions retained from the recently repealed Medicare Catastrophic Coverage Act of 1988, states will be required to cover all pregnant women and infants with incomes below poverty. Effective in April 1990, OBRA 1989 mandates Medicaid coverage for children under age six with family incomes at or below 133 percent of the federal poverty level and gives states the option of extending coverage to children born after 30 September 1983.

Such provisions will certainly benefit children who fall within these expanded Medicaid eligibility criteria. However, it is clear that such changes will not reach all children at risk. Expanding Medicaid coverage to older children could significantly reduce the number of uninsured children. For example, a recent study on adolescents’ health insurance coverage by the congressional Office of Technology Assessment found that if Medicaid were expanded to cover all children with family incomes less than 200 percent of the poverty level, three-quarters of currently 15 uninsured adolescents would obtain coverage. It is encouraging to note that OBRA 1990 expands Medicaid coverage to include ail children age eighteen and under in families with incomes below the poverty line. This expansion will be phased in during the 1990s. In sum, continued expansions of Medicaid eligibility beyond AFDC standards and the federal poverty level can make an important contribution to resolving the problem of uninsured children and reducing many of the disparities in coverage between children in two-parent families versus single-parent families. However, our findings also indicate that children at high risk for being uninsured come from a multitude of family circumstances. To ensure that health insurance is available to all children, an expansion of employer-based coverage in addition to Medicaid is needed. Otherwise, a new series of initiatives will be required that are more broadly aimed at all children currently without coverage. The views expressed in this article are those of the authors, and no official endorsement by the Agency for Health Care Policy and Research or the Department of Health and Human Services is intended or should be inferred. The authors thank Joel W. Cohen, Steven B. Cohen, Pamela Farley Short, and John Moeller for their comments. Karen Davis’s thoughtful review provided a number of important suggestions. Suzanne Worth of Social and Scientific Systems provided excellent programming assistance.

rom http://content.healthaffairs.org/ by Health Affairs on June 21, 2016 by HW Team


INSURING THE CHILDREN 89 NOTES


1. D.T. Ellwood, Poor Support: Poverty in the American Family (New York: Basic Books, 1988). 2. See, for example, F. Levy, Dollars and Dreams: The Changing American Income Distribution (New York: W.W. Norton and Company, 1987). 3. See Ellwood, Poor Support, 45-46. Between 1970 and 1986, the proportion of children in households headed by a single female doubled (from 10.8 percent to 21 percent). See U.S. Bureau of the Census, Statistical Abstract of the United States: 1988, 108th ed. (Washington, D.C., 1987), 50, Table 69. 4. M.J. Bane and D.T. Ellwood, “One-Fifth of the Nation’s Children: Why Are They Poor?” Science 245 (1989): 1047-1053. 5. Between 1981 and 1985, employee contributions for family coverage (adjusted for inflation) rose from $25.38 to $36.93, or by 31 percent. See G.A. Jensen, M.A. Morrisey, and J.W. Marcus, “Cost-Sharing and the Changing Pattern of Employer-sponsored Health Benefits,” The Milbank Quarterly 65, no. 4 (1987): 521-550. Average monthly employee premiums for family coverage increased from $28 in 1982 to $41 in 1986, or by 46 percent. See P.F. Short, “Trends in Employee Health Benefits,” Health Affairs (Summer 1988): 186-198. 6. For the most significant provisions of OBRA 1981 affecting AFDC and hence Medicaid eligibility, see J.E. Holahan and J.W. Cohen, Medicaid: The Trade-off between Cost Containment and Access to Care (Washington, DC.: The Urban Institute Press, 1986). Holahan and Cohen also discuss the effect of inflation on Medicaid eligibility. See also “Through the Looking Glass: An Overview of Medicaid Today,” Issue Brief 457 (The George Washington University, National Health Policy Forum, January 1987); and J. Cromwell, S. Hurdle, and G. Wedig, “Impacts of Economic and Programmatic Changes on Medicaid Enrollments,” Review of Economics and Statistics (May 1986): 232-240. 7. For a description of DEFRA, COBRA, and OBRA 1986, 1987, and 1989 provisions, see “Stretching the Limits of Medicaid: Are We Willing to Pay the Price?” Issue Brief 27 (The George Washington University, National Health Policy Forum, June 1989). 8. See, for example, J.D. Kasper, “The Importance of Type of Usual Source of Care for Children’s Physician Access and Expenditures,” Medical Care 25 (1987): 386-398; S.L. Gortmaker, “Medicaid and the Health Care of Children in Poverty and Near Poverty,” Medical Care 19 (1981): 567-582; and J.D. Kasper and G. Barrish, “Usual Source of Medical Care and Their Characteristics,” Data Preview 12, DHHS Pub. no. (PHS) 82-3324 (Hyattsville, Md.: National Center for Health Services Research, October 1982). 9. In contrast to the 1987 NMES household survey, respondents to the 1977 NMCES household component were not asked to specify whether their private health insurance was employment-related or purchased directly from an insurance company. This information was collected in the Health Insurance Employer Survey (HIES), which supplemented the 1977 NMCES household component. HIES was administered to a subsample of household respondents and includes information on all private health insurance coverage of household members in 1977. Since our estimates of children’s health insurance coverage are for the first quarter of 1977, we designated a child’s private insurance as employment-related if HIES reported that the child had such coverage at any time in 1977. This assumption appears to be quite reasonable, since in 1977, less than 3 percent of the general population with private insurance in the first quarter were reported by HIES to have both employment-related and other private coverage during 1977. Since many of these persons are likely to bc retirees with both employmentrelated and private Medicare supplemental insurance, the likelihood of error in assigning a child’s private coverage as employment-related is relatively small. Finally, es-


10.

11.

12. 13.

14. 15.

timates of children’s health insurance coverage in 1977 were made by merging household and HIES data and applying HIES population weights. The resulting estimates for the first quarter of 1977 are quite consistent with full-year and part-year estimates of coverage derived from NMCES household data. Parents counted as employed in 1977 were employed all year, while in 1987, the employment status of parents is currently available only from the first round of data collection. Thus, in 1977, we exclude the relatively small number of parents who were employed during only part of 1977. Since temporary workers have low rates of private health insurance coverage [see AC. Monheit et al., “The Employed Uninsured and the Role of Public Policy,” Inquiry 22 (1985): 348-364], it can be expected that the estimates of private and employment-related health insurance may be slightly inflated for children with employed parents. In fact, when we combine parents who were employed part year with those employed all year, rates of children’s health insurance coverage are only a few percentage points different from when we focus only on parents employed all year. Thus, large differences observed between 1977 and 1987 are not substantively altered when excluding those employed part year. For discussions of the problems involved in using CPS to estimate the size of the uninsured population in general and the health insurance coverage of children, see R. Kronick, Adolescent Health Insurance Status: Analyses of Trends in Coverage and Preliminary Estimates of the Effect of an Employer Mandate and Medicaid Expansion on the Uninsured (Washington, D.C.: U.S. Congress, Office of Technology Assessment, 1989); M.E. Moyer, “A Revised Look at the Number of Uninsured Americans,” Health Affairs (Summer 1989): 102-110; and K. Swartz and P. Purcell, “Counting Uninsured Americans,” Health Affairs (Winter 1989): 193-197. Bane and Ellwood, in “One-Fifth of the Nation’s Children,” observe that after adjusting for inflation, the median income of full-year, full-time workers is lower today than in 1973. This is a net change in insurance coverage, because some children gained insurance as a result of coverage shifts while other children lost coverage. On balance, however, children were net losers of health insurance over this period, yielding an increase in the number of uninsured children. The actual difference in the number of uninsured children between 1987 and 1977 (the 3.1 million children reported in the text) reflects the actual rates of coverage, number, and distribution of children prevailing in each of these years. Kronick. Adolescent Health Insurance Status.

Paediatric pituitary adenomas: a decade of change.

Parent Mentors and Insuring Uninsured Children: A Randomized Controlled Trial.

Over-the-Counter Hearing Aids: A Lost Decade for Change.

A decade of international change in abortion law: 1967-1977.

Advance directive completion by elderly Americans: a decade of change.

Insuring intensive psychotherapy.

Insuring against AIDS.

For Better or Worse? Change in Service Use by Children Investigated by Child Welfare Over a Decade.

A Decade of Change: Measuring the Extent, Depth and Severity of Food Insecurity.

Hospitalization due to drug use did not change after a decade of the Psychiatric Reform.

Reform issues for insuring mental health care.

Juvenile dermatomyositis at a tertiary care hospital: is there any change in the last decade?

Metals in tern eggs in a New Jersey estuary: A decade of change.

A Decade of Change: Training and Career Paths of Cardiothoracic Surgery Residents 2003 to 2014.

Resilience of institutional culture: mental nursing in a decade of radical change.

A decade of euroleague basketball: an analysis of trends and recent rule change effects.

The full spectrum of handlebar injuries in children: a decade of experience.

Systems-based content in medical morbidity and mortality conferences: a decade of change.

Lending to Parents and Insuring Children: Is There a Role for Microcredit in Complementing Health Insurance in Rural China?

The Difference a Decade Makes.

A device for insuring the congruency of opposed irregularly blocked fields.

A DeCade of science.

A decade of uncertainty.

A decade of MEDICAID.