Health Affairs, 22, no. 1 (2003): 219-229 doi: 10.1377/hlthaff.22.1.219 © 2003 by Project HOPE	New Online   New Issue: China & India   Obesity In China   Pay Cuts For Medicare Docs   Access To Care Woes

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Adams, E. K. Articles by Raskind-Hood, C. Search for Related Content PubMed Articles by Adams, E. K. Articles by Raskind-Hood, C. Related Collections Maternal And Child Health Medicaid State/Local Issues

DataWatch

Transitions In Insurance Coverage From Before Pregnancy Through Delivery In Nine States, 1996–1999

Abstract

 
Efforts to extend coverage to pregnant women, along with an expanding economy, did not prevent increases in the uninsured in the latter 1990s. Welfare reform may have led to declining Medicaid enrollments and caseloads. Data representative of live births in nine states show that in some states more than one-third of all pregnant women and almost two-thirds of low-income pregnant women lacked insurance before their pregnancy in 1996 and 1999. More than one-third of all pregnant women made some change in coverage by the time they delivered their baby. Among low-income women, the largest change was from uninsured status before pregnancy to Medicaid at delivery.

Medicaid, an important safety net for low-income women, saw a series of expansions that began in the mid-1980s; the percentage of live births paid for by Medicaid increased markedly. Yet studies indicate that Medicaid recipients, likely eligible at the beginning of pregnancy, delay the beginning of prenatal care.² Less is known about the insurance status of women before pregnancy or about how they make the transition into Medicaid. If they were insured before Medicaid, for example, they are less likely to experience access barriers than if they were previously uninsured.

Women’s experiences with insurance coverage also may be affected by the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). PRWORA replaced Aid to Families with Dependent Children (AFDC) with Temporary Assistance for Needy Families (TANF), restricted welfare tenure, imposed new work requirements, and uncoupled welfare and Medicaid eligibility. An unintended effect of PRWORA might have been higher levels of uninsurance among low-income women in their childbearing years. In turn, PRWORA might affect access to early and adequate prenatal care.³ Disruption in coverage related to PRWORA also might inhibit women’s ability to space the births of their children.

We used data from states participating in the Pregnancy Risk Assessment Monitoring System (PRAMS) through the Centers for Disease Control and Prevention (CDC) to investigate insurance coverage of pregnant women before and after the passage of PRWORA. Use of these data over a period in which national welfare policies might have led to changes in employment and insurance for low-income women provides new insights into the experiences of pregnant women.

Pregnancy And Insurance

Top Pregnancy And Insurance Study Data And Methods Study Results Discussion NOTES

 
Women are generally more vulnerable than men are to becoming uninsured, because they are less likely to be insured through a job and more likely to be covered as dependents.⁴ The most vulnerable are younger, low-income, and Latinas.⁵

Being pregnant has the potential to further complicate insurance options. Privately insured women who work in small firms, who have nongroup policies, or who change or leave jobs are most vulnerable. Although the Health Insurance Portability and Accountability Act (HIPAA) of 1996 prevents plans from considering pregnancy a preexisting condition, there may be a waiting period for coverage of any benefits, pregnancy included, for new employees. Nongroup policies will be costly and subject to variable state mandates regarding pregnancy coverage.

The Medicaid expansions were aimed at improving access and outcomes for low-income women and their families. These expansions clearly altered the financing of childbirth, as the proportion paid by Medicaid increased from 17 percent in 1985 to 35 percent in 1998.⁶ Medicaid payments for deliveries, however, does not mean that eligible women confront the sometimes difficult enrollment process early on. Women with Medicaid-covered deliveries often do not enroll until their second or third trimester and hence enter prenatal care late, have fewer visits, and perhaps receive inadequate care.⁷ However, women in Medicaid and uninsured women receive similar numbers of visits and are as likely to receive adequate prenatal care as are privately insured women who start care in the same trimester.⁸

Delayed Medicaid enrollment may have confounded studies of the Medicaid expansions on prenatal care and outcomes. Key multivariate studies of the expansions found mixed results of Medicaid on the timing and use of prenatal care, while others found that they increased use and improved birth outcomes.⁹ Some data showed that the expansions reduced the number of women who began prenatal care late, a key policy impact.¹⁰

A recent study of Medi-Cal (California Medicaid) enrollees found that nearly half (45 percent) were uninsured before pregnancy and one-fifth (21 percent) were uninsured through their first trimester.¹¹ For both Medi-Cal and privately insured (at delivery) women, rates of delayed prenatal care were highest among those without, and lowest among those with, coverage during the first trimester.

Although informative, the California data predate PRWORA. After PRWORA, Medicaid enrollment has declined with caseloads.¹² Moreover, the proportion of uninsured women of childbearing age rose as the proportion covered by Medicaid fell during 1994–1998.¹³ This trend may apply even to pregnant women. Data from the Current Population Survey (CPS) indicate that the percentage of pregnant women with Medicaid coverage sometime during the year fell from almost 24 percent in 1993 to only 15 percent in 1997.¹⁴ Although economic growth could increase private coverage among low-income women, only 23 percent of mothers who left welfare during 1995–1997 had private coverage.¹⁵

Earlier studies either do not distinguish timing of Medicaid coverage or are limited to one state. The PRAMS data are well suited to examining insurance coverage among pregnant women, since they are representative of live births in participating states and since the stratified random design oversamples high-risk pregnancies.¹⁶ In this paper we use these data for 1996–1999 to ask the following questions: (1) What is the rate of uninsurance among pregnant women prior to pregnancy? (2) What transitions in insurance coverage do pregnant women make from the period prior to pregnancy to time of delivery? (3) Do women in different states exhibit different transition patterns? (4) Were there changes in these patterns from 1996 to 1999 that are consistent with the 1996 national welfare reform?

Study Data And Methods

Top Pregnancy And Insurance Study Data And Methods Study Results Discussion NOTES

 
Study states include Alaska, Florida, Georgia, Maine, Oklahoma, New York (excluding New York city), South Carolina, West Virginia, and Washington. Errors in Georgia’s PRAMS sampling during 1998–1999 prevent its use, but 2000 data were to have been available in the fall of 2002. This analysis is part of a larger study, funded by the Robert Wood Johnson Foundation’s Health Care Financing and Organization (HCFO) initiative, which analyzes economic factors and PRWORA in relation to insurance coverage of pregnant women, 1996 to 2000.

State background. A number of state characteristics, including their overall economy and insurance markets, as well as Medicaid and welfare eligibility policies affect the insurance coverage of pregnant women. Data on study states are shown in Exhibit 1.

Mothers receive a self-administered questionnaire two to six months after delivery; an attempt is made to contact nonrespondents by telephone. Most states achieve a 70 percent or higher overall response rate; sample sizes range from 1,300 to 3,000 women per state each year. Birth certificate data are appended to each survey record; survey data include Medicaid and insurance status before pregnancy, during the prenatal period, and at delivery. Survey weights are provided, and measures of variance are adjusted for the complex sampling design.

Insurance measures. We used three survey questions to measure insurance. The first two asked women to recall their insurance status before pregnancy (whether or not they had insurance, and whether or not it was Medicaid). The third question asked about insurance at delivery; this question asked respondents to check all that applied (Medicaid, other insurance, self-pay, other). We then placed each woman into Medicaid, private/other insurance, and uninsured categories.

Since the question about coverage at delivery had multiple answers and state-specific response categories (for example state-specific Medicaid managed care, Indian Health Service [IHS], military), we made coding decisions to (1) include Medicaid managed care in Medicaid; (2) include military with private insurance; (3) report those with private coverage and Medicaid as private since Medicaid is the payer of last resort; (4) include IHS with the uninsured; and (5) include those reporting other insurance or self-pay as uninsured. Although women who use the IHS often do not have coverage outside this specific provider system, the IHS is important in two study states—Alaska and Oklahoma—providing both prenatal and maternity care for a sizable portion of women. We present data both with and without these observations.

We analyze two measures: uninsured before pregnancy, and a categorical measure of change in insurance coverage from the prepregnancy period through delivery. The transition categories include (1) private to Medicaid; (2) private to uninsured; (3) private throughout; (4) Medicaid/uninsured to private; (5) Medicaid throughout; (6) uninsured to Medicaid; and (7) uninsured throughout.

Study Results

Top Pregnancy And Insurance Study Data And Methods Study Results Discussion NOTES

 
Insurance prior to pregnancy. The percentage of all women uninsured before pregnancy was sizable but varied across states (Exhibit 2). In 1996 this percentage ranged from 15 percent in upstate New York to 44 percent in Oklahoma. If IHS women are not included, the percentage uninsured before pregnancy in Oklahoma was 39 percent and in Alaska, 27 percent. Several other states (Florida, South Carolina, and West Virginia) had rates above 30 percent in 1996.

Being uninsured poses more of a burden on low-income women, who may not be able to pay out-of-pocket costs. The data in the lower half of Exhibit 2 are for women reporting annual incomes below $16,000 ($17,000 in West Virginia). As expected, the percentage uninsured before pregnancy was higher, varying from 40 percent in upstate New York to 60 percent in Oklahoma in 1996. These percentages are well above the 32 percent of low-income mothers reported by Rachel Gold to be uninsured.¹⁸ Thus, the subset of women who become pregnant during the year appears to be particularly prone to becoming uninsured. The PRAMS data are consistent, however, with the 45 percent uninsured before pregnancy reported in California.¹⁹

The percentage of low-income women uninsured before pregnancy increased in five states from 1996 to 1999. Low-income women in South Carolina experienced a significant increase (p = .05), as did low-income pregnant women in Washington (p = .10). In 1999 almost two-thirds of low-income pregnant women in South Carolina and West Virginia were uninsured before pregnancy.

Insurance transitions. Lack of insurance prior to pregnancy does not mean that women remain uninsured throughout pregnancy. Indeed, pooled state data (not shown) indicate that 32–35 percent of all pregnant women experienced some kind of transition. Although the pooled distribution of women across the transition categories was significantly different (p < .001) in 1999 versus 1996, there was marked variation in patterns across states.

Transitions within study states. We show data on the seven transition categories for each of the nine states during 1996–1999 for all women (Exhibit 3) and low-income women (Exhibit 4). For all women the change in the distribution across categories was significant at least at the .10 level in all states. For low-income women, the 1999 distribution was significantly different only in five states, perhaps as a result of small samples within some transition categories.

The percentage of all women with Medicaid throughout their pregnancy in 1996 ranged from 6 percent in Oklahoma to 14 percent in West Virginia and remained fairly stable or dropped in the study states during 1996–1999. Of the women with Medicaid at delivery (Exhibit 1), those who moved from uninsured status before pregnancy accounted for 48 percent to 72 percent in 1996 and from 50 percent to 69 percent in 1999.

There is only some evidence that privately insured women either lose or drop this coverage by the time they deliver. In 1996 from 4 percent to 11 percent of women across all income groups moved from private coverage to Medicaid; in 1999 this group ranged from 3 percent to 10 percent. The highest percentage of women experiencing this transition was in South Carolina.

There were also small percentages who reported being uninsured throughout pregnancy. Based on data that omit IHS in Oklahoma and Alaska, these ranged from 1 percent in several states to 7 percent in Oklahoma in 1996. By 1999 this percentage was lower in Oklahoma (4 percent) and three other study states but slightly higher in South Carolina, Washington, and West Virginia.

Low-income women. As expected, the percentage of low-income women with private insurance throughout their pregnancy was lower in 1996, ranging from 7 percent in West Virginia to 21 percent in Washington. If the combination of welfare reform and a strong economy led women into jobs that paid little but offered health insurance, this percentage could have increased by 1999. In most study states the percentage of low-income women who were privately insured throughout pregnancy was either stable or rising during 1996–1999. Increases were highest in New York and Oklahoma (moving from 12 percent to 23 percent and 11 percent to 19 percent, respectively). Washington and South Carolina also experienced sizable increases.

The group of low-income women who lost or dropped private coverage was larger than for all women: 5 percent to 15 percent in 1996 and 6 percent to 13 percent in 1999. Low-income women may be only marginally attached to the labor force and lose this connection when pregnant; their employers may also recognize available Medicaid coverage and not offer benefits. Although the highest percentage of low-income women in this category was in South Carolina in 1996, by 1999 the highest observations were in New York and Washington.

There were generally larger percentages of low-income women uninsured throughout pregnancy compared with all women, varying from a low of 1 percent in New York to a high of 7 percent in Florida in 1996. This percentage did not increase in most study states during 1996–1999. One exception is Washington, where it increased from 3 percent to 11 percent. A recent study noted declines in Medicaid enrollment and a collapsing of the individual and nonsubsidized Basic Health Plan (BHP) in this state during 1995–1998.²⁰ Given that enrollments in the state-subsidized BHP were capped, this left low-income women with few insurance alternatives. It is possible that uninsured women in this and other states received prenatal care through physicians’ charity, indigent care, or safety-net providers; it is also possible they went without needed care.

The largest single group of low-income women made the transition from uninsured status before pregnancy to Medicaid at delivery in all study states. In 1996 this group represented more than 35 percent in each state; in South Carolina and West Virginia, 52–53 percent. By 1999 this percentage increased in these two states and Alaska while remaining stable or declining in the others.

Of the low-income women with Medicaid at delivery in 1996 (ranging from 57 percent to 87 percent), the group making the transition from uninsured status before pregnancy constituted from 46 percent to 68 percent. By 1999 this transition group represented 45 percent to 70 percent of the low-income women with Medicaid-covered births in the study states. Thus, while the percentage of low-income women with Medicaid at delivery declined, the portion of this group who were uninsured before pregnancy remained fairly stable.

Discussion

Top Pregnancy And Insurance Study Data And Methods Study Results Discussion NOTES

 
These data provide a picture of the transitions in insurance coverage made by pregnant women and highlight the plight of low-income women regarding coverage. Findings indicate that up to 40 percent of all women and two-thirds of low-income pregnant women in some states lack insurance just prior to pregnancy. Further, while the largest group of insured women in all income categories was privately insured throughout, the largest category for low-income women was uninsured before pregnancy and Medicaid at delivery. Changes seen here from 1996 to 1999 are consistent with high economic growth and welfare reform but indicate that in some states these two forces may have left a larger percentage of women uninsured before pregnancy.

These patterns occur despite the fact that Medicaid expansion policies for pregnancy remain in place. Given that pregnancy is the criterion for eligibility and that coverage generally ends sixty days after childbirth, a large number of women are apparently left uninsured between pregnancies or prior to their first. Studies before the enactment of PRWORA indicated that the expansion policies did not guarantee early entry; our data show that large percentages of women with Medicaid-covered births made the transition from uninsured status. There also was a general decline in the percentage of low-income pregnant women covered by Medicaid at delivery from 1996 to 1999. Some decline is attributable to increased private insurance for low-income women, but this was not seen in all study states.

Long-standing differences in the generosity of Medicaid eligibility as well as geographic differences in economic growth and state-specific welfare reform policies are behind the state variations seen here. However, sizable numbers of low-income women remain uninsured before pregnancy in each study state. The health benefits expected from the Medicaid expansions will not likely occur if Medicaid coverage is nonexistent immediately before or early in pregnancy.

Although this analysis is limited to nine states, our findings highlight the importance of the prepregnancy period when examining insurance coverage. A recent study concluded that unplanned pregnancies and lack of a regular source of care before pregnancy were more important in explaining delays in beginning prenatal care than were noninsurance logistic barriers during pregnancy.²¹ Both of these could stem from lack of insurance coverage before pregnancy. Many states recognized the importance of the prepregnancy period and implemented Medicaid family planning waivers in the mid- to late 1990s.²² Evaluating the effect of these waivers will be of great importance.

Editor's Notes

 
Kathleen Adams is an associate professor in the Department of Health Policy and Management, Emory University, in Atlanta. Norma Gavin is director of the Maternal, Child, and Reproductive Health Program at the Research Triangle Institute (RTI), Research Triangle Park, North Carolina. Arden Handler is a professor at the University of Illinois at Chicago, Community Health Sciences, School of Public Health. Will Manning is a professor in the Department of Health Studies, University of Chicago. Cheryl Raskind-Hood is a research professor at the Rollins School of Public Health, Emory University.

This work was funded by the Robert Wood Johnson Foundation under its Changes in Health Care Financing and Organization (HCFO) initiative. All opinions expressed in this paper represent those of the authors and not the funding agency. The authors also acknowledge the support of staff in the Centers for Disease Control and Prevention (CDC) Division of Reproductive Health regarding the content and interpretation of the PRAMS data.

NOTES

Top Pregnancy And Insurance Study Data And Methods Study Results Discussion NOTES

 

1. J. Guyer, M. Broaddus, and A. Dude, Millions of Mothers Lack Insurance Coverage (Washington: Center on Budget and Policy Priorities, 10 May 2001), 1–10.
2. R. Kaestner, "Health Insurance, the Quantity and Quality of Prenatal Care and Infant Health," Inquiry (Summer 1999): 162–175.
3. P. Wise, W. Chavkin, and D. Romero, "Assessing the Effects of Welfare Reform Policies on Reproductive and Infant Health," American Journal of Public Health 89, no. 10 (1999): 1514–1521.[Abstract/Free Full Text]
4. Henry J. Kaiser Family Foundation, Fact Sheet: Women’s Health Insurance Coverage (Washington: Kaiser Family Foundation, July 2001).
5. Kaiser Family Foundation, Fact Sheet: Health Insurance Coverage of Low-Income Women (Washington: Kaiser Family Foundation, February 2001).
6. National Governors’ Association, Maternal and Child Health(MCH)Update: States Have Expanded Eligibility and Increased Access to Health Care for Pregnant Women and Children (Washington: NGA,, 2001); and Alan Guttmacher Institute, Blessed Events and the Bottom Line: Financing Maternity Care in the United States (New York: Alan Guttmacher Institute, 1987).
7. P. Braveman et al., "Access to Prenatal Care following Major Medicaid Eligibility Expansions," Journal of the American Medical Association (10 March 1993): 1285–1289. Also see Kaestner, "Health Insurance"; and S.J. Katz, R.W. Armstrong, and J.P. LoGerfo, "The Adequacy of Prenatal Care and the Incidence of Low Birthweight among the Poor in Washington State and British Columbia," American Journal of Public Health 84, no. 6 (1994): 986–991.[Abstract/Free Full Text]
8. Kaestner, "Health Insurance."
9. See for example, J. Currie and J. Gruber, "Saving Babies: The Efficacy and Cost of Recent Changes in the Medicaid Eligibility of Pregnant Women," Journal of Political Economy (December 1996): 1263–1296; G.M. Kenney and L.C. Dubay, "A National Study of the Impacts of Medicaid Expansions for Pregnant Women," Working Paper (Washington: Urban Institute, 1995); N. Cole, Increasing Access to Health Care: The Effects of Medicaid Expansions to Pregnant Women, Final report prepared by Abt Associates for the Health Care Financing Administration (Cambridge, Mass.: Abt Associates, 1995); and L.M. Baldwin et al., "The Effect of Expanding Medicaid Prenatal Services on Birth Outcomes," American Journal of Public Health 88, no. 11 (1998): 1623–1629.[Abstract/Free Full Text]
10. J. Piper, W. Ray, and M. Griffin, "Effects of Medicaid Eligibility Expansion on Prenatal Care and Pregnancy Outcome in Tennessee," Journal of the American Medical Association 264, no. 17 (1990): 2219–2223.[Abstract]
11. S. Egerter, P. Braveman, and M. Kristen, "Timing of Insurance Coverage and Use of Prenatal Care among Low-Income Women," American Journal of Public Health 92, no. 3 (2002): 423–427.[Abstract/Free Full Text]
12. M. Ellwood, The Medicaid Eligibility Maze: Coverage Expands, but Enrollment Problems Persist: Findings from a Five State Study (Cambridge, Mass.: Mathematica Policy Research, September 1999).
13. R.B. Gold, "Implications for Family Planning of Post–Welfare Reform Insurance Trends," Guttmacher Report 2, no. 6 (1999): 1–7.
14. K.E. Thorpe, "The Distribution of Health Insurance Coverage among Pregnant Women, 1990–1997," Paper prepared for the March of Dimes (Atlanta: Emory University, March 1999)
15. B. Garrett and J. Holahan, "Health Insurance Coverage after Welfare," Health Affairs (Jan/Feb 2000): 175–184.
16. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, PRAMS Pregnancy Risk Assessment Monitoring System 1997 Surveillance Report (Atlanta: CDC, 1999).
17. B.C. Gilbert et al., "The Pregnancy Risk Assessment Monitoring System (PRAMS): Methods and 1996 Response Rates from 11 States," Maternal and Child Health Journal 3, no. 4 (1999): 199–209; and [Medline]CDC, PRAMS Pregnancy Risk Assessment Monitoring System.
18. See Gold, "Implications for Family Planning."
19. See Egerter et al., "Timing of Insurance Coverage."
20. J. Holahan and M.B. Pohl, "Recent Changes in Health Policy for Low-Income People in Washington," Assessing the New Federalism State Update no. 24 (Washington: Urban Institute, February 2002).
21. P. Braveman et al., "Barriers to Timely Prenatal Care among Women with Insurance: The Importance of Prepregnancy Factors," Obstetrics and Gynecology (June 2000): 874–880.
22. According to the Centers for Medicare and Medicaid Services Family Planning Division, 2002.

                      What's this?

This article has been cited by other articles:

				E. A. Finkelstein and J. G. Trogdon Public Health Interventions for Addressing Childhood Overweight: Analysis of the Business Case Am J Public Health, March 1, 2008; 98(3): 411 - 415. [Abstract] [Full Text] [PDF]

				D. M Dudzinski and S. E Shannon Competent Patients' Refusal of Nursing Care Nursing Ethics, November 1, 2006; 13(6): 608 - 621. [Abstract] [PDF]