Medicaid Eligibility Expansion in Florida: Effects On Maternity Care Financing and the Delivery System
Context: In July 1989, the income limit on Medicaid eligibility for pregnant women in Florida was increased from 100% to 150% of the poverty level. This change may have led to substantial shifts in the financing of pregnancy-related care, and also may have had distinct effects on different providers in the health care delivery system.
Methods: Matched birth and death certificates, hospital discharge abstracts, Medicaid eligibility records and encounter records from county public health departments were used to estimate changes in the flows of funds and services by major payer groups during the period preceding the expansion (July 1988-June 1989) and for calendar year 1991. A total of 188,793 births in the first period and 193,292 in the second were examined.
Results: The number of births financed annually by Medicaid in Florida increased by 47% following the eligibility expansion, from 47,400 in 1988-1989 to 69,600 in 1991. This increase stemmed largely from covered births to women who otherwise would have been uninsured. Seventy-three percent of the additional 22,200 deliveries funded through Medicaid in 1991 are attributed to women who were eligible as a result of the expansions. The additional prenatal care financed by Medicaid was delivered almost entirely by county public health departments, which increased their capacity by more than 100%, from 177,000 visits in 1988-1989 to 433,000 in 1991. Medicaid payments for maternity care increased 39%, from $135 million to $187 million, while payments made by the uninsured dropped by 29%. These changes resulted in a 5% rise in hospital revenues, despite little change in the number of admissions.
Conclusions: The Medicaid expansion benefited low-income pregnant women and hospitals in Florida. It is unknown whether the private delivery system would have accommodated the increased demand in the absence of the public health system response.
Family Planning Perspectives, 1999, 31(3):112-116 & 121
During the past decade, Congress enacted a series of laws that extended Medicaid eligibility to many low-income pregnant women. These expansions mandated that Medicaid be made available to all pregnant women with an income below 133% of the federally designated poverty level, and gave states the option to establish higher income thresholds for coverage. The expansions also stimulated states to simplify Medicaid eligibility processing, implement outreach programs and introduce enhanced prenatal care benefit programs, in an effort to improve low-income women's access to prenatal care and their birth outcomes and infant's health.
These eligibility expansions were the most important policy changes in the Medicaid program during the 1980s. The expansions led to a significant increase in the number of deliveries financed by Medicaid, but evidence is mixed as to whether they produced an improvement in prenatal care and birth outcomes.1Moreover, to comprehend the full effect of these expansions, it is essential to understand how changes in the Medicaid program affected other government programs that deliver prenatal care, and how the expansions affected private payers and providers.
The effects of the program's expansion on prenatal care access and birth outcomes are likely to be quite different if Medicaid- financed care substitutes for care that was previously financed and provided under other programs, such as Title V (the maternal and child health block grant) or by private insurance, rather than providing new coverage for those who previously lacked either insurance or access to other public programs.2 Previous studies, however, have not addressed these substitutions.
Our study focuses on the substitution that occurred following a Medicaid eligibility expansion in Florida. We chose Florida for a number of reasons: It is the nation's fourth most populous state; women in the state have some 200,000 births each year; and in July 1989, the state significantly expanded Medicaid eligibility for pregnant women. In addition, Florida relies heavily on county health departments to provide prenatal care to its low-income female population, and hence it is a good place to study interactions between the Medicaid financing changes and the publicly financed direct delivery system.
Florida also aggressively implemented other strategies to ensure that women who became eligible because of the expansions actually gained coverage under the program.3 For example, the state created and deployed technical assistance teams that worked with district and local public health staff to develop procedures for stationing Medicaid workers at health facilities, for determining eligibility and assisting with billing. These procedures were intended to help county health units maximize Medicaid revenues when treating eligible patients, in order to provide financing to expand services.4
Our objective in this article is to investigate the interactions between the Medicaid program and other sources of financing and providing pregnancy-related health care. Among the central questions that we address are: How did the expansions affect the sources of payment for deliveries? Were there changes in where women obtained their prenatal care and delivery services? Finally, how did the expansions affect payments to providers?
We studied all births in Florida occurring from July 1988 through June 1989 and in calendar year 1991. The earlier period covers the 12 months just before Medicaid eligibility was expanded to include women with an income between 100% and 150% of poverty. The later period is the second full calendar year after the expansion was implemented. We chose the earlier period as the "baseline" year, so that women with an income below poverty who delivered during this period would have been eligible for Medicaid throughout their pregnancy. (An initial Medicaid eligibility expansion for pregnant women in Florida had occurred in October 1987, when the program was opened to those whose income was at or below the poverty level. The 1989 expansion we study here was thus the second such expansion of Medicaid eligibility for pregnant women in Florida.) We chose 1991 as the postexpansion period to allow time for the new eligibility policy to be implemented.
Data come from a number of sources. We used Florida birth and fetal death certificates to define our study universe and provide information on the quantity of prenatal care. Hospital discharge abstracts provided information on the primary payer for the delivery, as well as a measure of hospital charges. We used Medicaid eligibility and claims files to confirm Medicaid payer status. In addition, we used individual encounter records for personal health services provided through each county health department in the state to construct summary records for each episode of prenatal care provided by county health departments during the study period. These four data sets were linked using specially constructed computer algorithms.5 Finally, we used American Hospital Association annual survey files for Florida hospitals to identify the type of hospital that provided care, and we added this information to our analytic file.
In this article, we examine aggregate changes in the primary source of financing for deliveries in the two study periods, as well as changes in the quantity of maternity-related services, in the type of provider delivering care, in the financing of these services and in the flow of payments for care. We use the vital statistics data to measure the aggregate number of deliveries in each of the two study periods. The observations in the linked data base are used to determine the distribution of services and source of payment for deliveries. These distributions are multiplied by the total count of deliveries from the vital statistics to measure the aggregate services and payments.
Our method is modeled after the National Health Expenditure accounts, a continuing series of annual estimates compiled by the Health Care Financing Administration, which estimates the flow of funds financing all health care in the United States.6 This methodology also has been applied to study the costs and financing of perinatal care nationwide.7
•General measures. We present a series of matrices categorizing deliveries, services and payments according to financing source. Payment categories include private insurance, Medicaid and "other payer." The last category includes other third-party payers, such as the Civilian Health and Medical Program of the Uniformed Services (CHAMPUS) or other state and federal programs that make payments on behalf of a patient receiving care, as well as the uninsured. The hospital discharge data—our primary source for payer information—do not allow us to further classify these other payers. We also are unable to separately identify a woman's source of insurance, if any, for prenatal care.
The quantity of prenatal care and the number of prenatal and delivery admissions are categorized by both payer and site of service. For ambulatory care, we distinguish between care provided in county health departments and care provided at all other sites, including physicians' offices, hospital clinics and hospital outpatient departments. Limitations of the data prevent us from further subdividing the site of ambulatory care. Hospital admissions are categorized by the type of hospital—public, voluntary or proprietary. Our measures of quantity of care and of payment cover care received by women who delivered during the study period, irrespective of whether the care was provided during that period.
•Measuring deliveries. The vital statistics records for all births and fetal deaths registered to Florida residents measure the total number of deliveries in each year. We distributed these deliveries among the three payer categories based on the distribution of primary payer at delivery for births included in the matched hospital discharge file and vital statistics file for each year.
•Measuring use of services. The linked data file also provides us with an estimate of the average number of prenatal care visits made by women in the various payer subgroups. We multiplied these estimates by the number of deliveries to measure aggregate prenatal care visits. The encounter data from the county health department system yield a count of the total number of prenatal care visits provided by county health departments. We distributed this total among the different payers based on the distribution of county health department visits that we were able to match to the vital statistics-hospital discharge file.
The total number of hospitalizations for deliveries was counted from the vital statistics data based on the location of the delivery. We allocated this total among the types of hospitals and payers based on the distribution observed in our linked analysis file. We estimated the number of prenatal admissions for women delivering in each study period from the number of admissions in the period with diagnostic codes related to prenatal or maternity care that did not result in delivery. Because our hospital discharge data set lacks individual identifiers, we were not able to track the prenatal hospital admissions of women who gave birth in the study period. We therefore approximated these by looking at all prenatal admissions in a period, regardless of whether the woman actually gave birth in that period. Because the number of births changes little from year to year, this method provides a good estimate of the number of prenatal admissions for women delivering in the study period.
•Measuring payments. A matrix showing the flow of payments for pregnancy-related care during the two periods measures the direct payments for care by patients and on account of patients by third-party payers. That is, it measures what was actually collected by the provider for the care of a particular patient. It does not include contributions that are not tied to particular patients, such as federal block grants to states for Title V programs and general contributions by local governments to public hospitals for charity care. Thus, Medicaid reimbursements to county health units were included in the payment flows; state general revenues, county funds and Title V block grant funds that support these units were not included.
To measure the payment flows, we started with an estimate of the total charges for inpatient hospital services and for physicians' services and related services (such as laboratory tests and x-rays), categorized by the payer for that delivery. To derive the total charge estimates for each payer, we multiplied estimates of the average charge per quantity of service by our estimates of the units of service. For hospital admissions, we estimated the average charge for women with different payers and for different hospital types using the hospital discharge data. For physicians' services and other services, we used as the basis for assigning total charges per payer an estimate of the total charges for prenatal care and the physician's delivery charge, divided by the number of prenatal visits, taken from the Florida Medicaid claims file for 1991 and from the claims data for two large employers in Florida.
Our primary data sources do not provide the information needed to separate the charges for pregnancy-related care obtained by women who are uninsured from those for women with "other" third-party sources of payment. Therefore, we used data on the distribution of payers for delivery services for Florida residents sampled in the 1988 National Maternal and Infant Health Survey (NMIHS) to develop estimates. We calculated from the NMIHS the deliveries and visits for uninsured women and for women with third-party payers other than Medicaid or private insurance. We used an estimate of the proportions of service use between these two groups to allocate our aggregate estimate of hospital and physician charges for women in both groups to the uninsured and to other third-party payers. These proportions from the NMIHS were used to make this allocation in both of our analysis periods. Because the Medicaid expansion would be expected to decrease the share attributable to the uninsured, our procedure will somewhat understate any increase in payments over the period.
Finally, we converted the resulting matrix of charges to payments using estimates of the ratio of payments to charges for different payers. Payment-to-charge ratios for hospital care were provided by the Agency for Health Care Administration in Florida. A Medicaid payment-to-charge ratio for physicians' services and related services was derived from the Florida Medicaid claims files for maternity care. Medicaid payments to county health departments for prenatal care were measured directly from State Health Office budget and revenue statistics for the county health department system. The ratio for private insurance payers was based on the claims data for maternity care from two large Florida employers.
In the absence of other data on payment-to-charge ratios for physicians' services in Florida for the uninsured and for those covered by third-party payers other than private insurance and Medicaid, we applied the hospital payment-to-charge ratios for patients classified as "direct payment" and "other third-party payment" to calculate physicians' payments for the uninsured and for other third-party payers. (This seemed to be a reasonable assumption given the similarity of the private insurance and Medicaid ratios for hospital and physician services.) We also divided the estimated hospital and physicians' payments for privately insured patients into insurance payments and cost-sharing payments by using a typical coinsurance rate for these services for insured pregnant women.
Changes in Source of Financing
There were substantial shifts in the source of payment for deliveries in Florida between the baseline period and the 1991 calendar year. In just two and one-half years, the number of births covered annually by Medicaid rose from 47,400 to 69,600, a 47% increase (Table 1). The proportion of all deliveries paid for by Medicaid during the study period rose from 25% to 36%. Since the total number of births per year increased by only 2% over this period, nearly all of the growth in Medicaid coverage represented shifts among payment sources.
|Table 1. Number and percentage distribution of deliveries, and percentage change between years, by primary payment source, Florida, 1988-1989 and 1991|
|Source||July 1988-June 1989||1991||% change|
|*Includes medically needy and other eligibility categories. In this and all subsequent tables, the "other payer" category includes other nonprivate third-party payers and the uninsured.|
Most of the Medicaid growth can be attributed to the eligibility expansion. Of the 22,000 additional deliveries covered by Medicaid during the study period, 16,000 (73%) were to women eligible under the income expansions. This includes women who were entitled by the expansion of eligibility to the poverty level in October 1987. Some of the increase, therefore, may have stemmed from an increase in the number of women below the poverty level. However, it is likely that the vast majority of the increase represents those who were made newly eligible by the July 1989 expansion. Of the 6,000 additional deliveries among women entitled to Medicaid because they received payments from the Aid to Families with Dependent Children program, some likely resulted from women's increased knowledge of Medicaid coverage owing to the increased outreach efforts that accompanied the expansion. The rest may have resulted from an increase in the number of women eligible for coverage as a result of the effects of economic recession.
In contrast to the sharp increase in Medicaid's role in financing births, the proportion of births covered by private insurance remained nearly the same over the study period. This finding may allay concerns that the Medicaid expansion served to "crowd out" private insurance.8 An upper-bound estimate of the substitution of public coverage for private coverage is the total decline in private deliveries (2,800) divided by the total increase in Medicaid deliveries (22,000)—a difference of about 13%. (This is an upper bound because some of the decline in private insurance coverage might have occurred in the absence of the expansion, as a result of the downward trend in private coverage over this time period.9 In addition, the recession might have increased the number of women eligible for Medicaid by drawing from those previously relying on private insurance.)
Therefore, the Medicaid expansion primarily covered either the uninsured or replaced other sources of public third-party coverage (e.g., other Federal and state programs that pay for patients' care). Although our data do not permit us to distinguish between these two groups of "others," data from the NMIHS in Florida show that during the baseline period, two-thirds of those without Medicaid or private insurance (about 33,000 women) were uninsured. Thus, it is likely that the Medicaid expansion served largely to cover women who otherwise would have been uninsured.
CHANGES IN PRENATAL CARE VISITS
The total number of ambulatory prenatal visits rose by 150,000, or about 7%, between the baseline period and the 1991 calendar year (Table 2). Striking changes occurred in the quantity of prenatal care obtained through different parts of the delivery system. The total number of visits made to county health departments rose by 250,000 (a 100% increase). Over the study period, the proportion of all prenatal care visits made at health departments rose from 12% to 22%. In contrast, the proportion of care obtained at all other ambulatory care sites decreased from 88% to 78%.
|Table 2. Number and percentage distribution of prenatal visits, by site of care, according to primary payment source, and percentage distribution of prenatal visits, by payment source, according to site of care, 1988-1989 and 1991|
|Payment source||July 1988-June 1989||1991|
|Total||County health department||Other||Total||County health department||Other|
|No. of visits (in 000s)|
|% distribution by site of care|
|% distribution by payment source|
The Medicaid eligibility expansion was almost fully accommodated by the huge growth in care provided by county health departments. Of the 263,000 additional prenatal visits by women receiving Medicaid, 256,000 took place at these sites, indicating that the expansion was accompanied by a significant shift in where Medicaid beneficiaries obtained their prenatal care. By 1991, county health departments provided 59% of ambulatory prenatal care for women covered by Medicaid, compared with 38% only two and one-half years earlier. As a result, the share of prenatal care delivered to Medicaid beneficiaries at other sites decreased: The proportion of prenatal care visits obtained in physicians' offices decreased from 23% to 15% (not shown), and the proportion of prenatal care visits obtained in other settings (including community health centers, hospital outpatient departments and emergency rooms) declined from 39% to 26% of visits.
The number of prenatal visits made by "other payers"—that is, uninsured patients and those covered by other nonprivate third-party payers—decreased by 121,000, or about 26%, subsequent to the Medicaid expansion. According to the NMIHS, in 1988 about 63% of these visits were made by uninsured patients; therefore, a large proportion of the decrease could be attributed to care that previously was delivered as charity care by hospitals or physicians. Visits to county health departments by women covered by other payers decreased by 12,000, or by 10% of the total decrease in visits among all women in this group. Most of the drop in this group, however, stemmed from decreases in visits to other sites, which declined by 109,000 visits, or by almost 90% of the total.
On average, there was a 5% increase in the number of prenatal visits made per person between 1988-1989 and 1991 (from 11.0 to 11.5), and this increase occurred among women in all three coverage groups. The number of prenatal care visits made per person among women receiving Medicaid increased by 7% (from 9.8 to 10.5), while the number among women who were either uninsured or covered by other payers increased by 6% over the study period (from 9.6 to 10.2). In contrast, the increase among the privately insured was 4% (from 12.3 to 12.8). Thus, the Medicaid expansion appears to have led to a small increase in access to care among women without private insurance, since growth rates for these women exceeded those for privately insured women. The additional Medicaid financing for care provided by county health departments may have allowed these sites to expand services both to the Medicaid population and to those remaining uninsured.
CHANGES IN HOSPITAL ADMISSIONS
Compared with the shifts in access to prenatal care, the Medicaid expansion had a much smaller effect on the types of hospitals at which women obtained pregnancy-related care. Across all categories of payers, women's use of public hospitals declined by 3%: Public hospitals comprised 29% of admissions during the baseline period and 27% during the 1991 calendar year; use of voluntary hospitals declined by a similar percentage (Table 3). While the share of maternity care provided by proprietary hospitals rose from 14% to 17%, most of this increase resulted from a general trend among all payer groups and is probably not a consequence of the expansion.
|Table 3. Number and percentage distribution of maternity-related hospital admissions, by site of care, according to primary payment source, and percentage distribution of maternity-related admissions, by primary payment source, according to hospital type, 1988-1989 and 1991|
|Payment source||July 1988-June 1989||1991|
|Total||Public hospital||Voluntary hospital||Proprietary hospital||Total||Public hospital||Voluntary hospital||Proprietary hospital|
|No. of admissions (in 000s)|
|% distribution by hospital type|
|% distribution by payment source|
Medicaid admissions rose from 25% to 36% of all maternity admissions, while those for "other payers" (including self-pay and the uninsured) fell from 27% to 18%. Although there was little change in the number of public and voluntary hospital admissions for pregnant women, the share of these admissions that were financed by Medicaid increased sizably, while the share that were patient-financed or financed by some other government program declined. For public hospitals, the Medicaid share of pregnancy-related care increased from 32% to 47%, whereas among voluntary hospitals this proportion increased from 25% to 36%. In public hospitals, the share of admissions financed by other payers fell from 42% to 28%, while for voluntary hospitals the share decreased from 22% to 16%. The share of admissions to proprietary hospitals financed by Medicaid also increased, while the share financed by other payers declined. However, the magnitude of these changes was not large, because the vast majority of admissions to proprietary hospitals were among the privately insured (70% in 1991, a proportion essentially unchanged from that in the baseline period).
During both periods, Medicaid beneficiaries used a very different mix of hospitals than did women who were privately insured. For example, in 1991, public hospitals provided 36% of maternity-related admissions for women relying on Medicaid, versus 15% among the privately insured. In the same year, voluntary hospitals accounted for 55% of Medicaid admissions and 58% of privately insured admissions. Finally, proprietary hospitals accounted for 9% of Medicaid admissions, compared with 27% among the privately insured. The pattern of hospital usage among the uninsured and among those covered by other payers was similar to that of Medicaid beneficiaries, but with greater use of public hospitals (42%) and less use of voluntary hospitals (48%).
CHANGES IN PAYMENTS FOR MATERNITY CARE
Payments to providers for pregnancy-related care in 1991 totaled $928 million dollars, an increase of 3% from the total of $902 million (in 1991 dollars) during the baseline period (Table 4). Medicaid payments for pregnancy-related care (which, in Florida, are financed 55% by federal funds and 45% by state funds) increased 39% over the study period, from $135 million to $187 million. Medicaid payments for care obtained by women eligible for benefits due to the expansions in October 1987 and July 1989 increased by 140%, having risen from $43 million in the baseline period to $106 million in 1991 (not shown). Medicaid payments for other beneficiaries declined by 11% over this period, from $92 million during the baseline year to $82 million in 1991. In contrast, payments by other third parties decreased by 27%. Payments by the uninsured also fell, by 29%, suggesting that the financing shifts reduced substantially the cost burden for uninsured families.
|Table 4. Payments for maternity-related care and percentage distribution of payments, by type of service and source of payment, 1988-1989 and 1991|
|Payment Source||July 1988-June 1989||1991||% change|
|Amount (in millions)*||%||Amount (in millions)*||%|
|Other third-party payer||91.5||10.1||66.9||7.2||-26.9|
|Other third-party payer||35.0||7.9||25.9||5.8||-26.0|
|Other third-party payer||56.5||12.4||41.0||8.5||-27.4|
|*In 1991 dollars|
Hospitals also benefited from the expansions. Their maternity-related revenues grew by 5% (from $457 million to $481 million), although admissions of maternity patients remained fairly constant. This extra revenue largely resulted from Medicaid payments for admissions that previously were financed through uncompensated care funds.
In contrast, total payments to physicians increased by less than 1% in constant 1991 dollars, although the overall number of ambulatory prenatal visits increased by 7%. Two main reasons account for this difference. First, total physician payments are a combination of payments for ambulatory prenatal care and inpatient care, mostly for deliveries, and the number of deliveries and the total amount of physician payments for them remained relatively constant over the study period. Second, the increase in ambulatory prenatal care was accommodated almost entirely by county health departments. These new visits were paid at the Medicaid reimbursement rate, which is lower than that of other payers.*
Medicaid accounted for 20% of total payments for maternity-related care in 1991, although 36% of deliveries in that year were to women with Medicaid coverage. This difference arises because Medicaid pays a smaller proportion of what it is charged for pregnancy-related health services than do private insurers. Private insurers accounted for 63% of payments in 1991, although only 46% of deliveries were to women with private insurance coverage. Direct patient payments—including cost sharing by insured patients and payments made by uninsured patients—accounted for 9% of funds. This distribution of payments by source in Florida is similar to national patterns.10
The expansion of Medicaid eligibility in Florida from 100% of poverty to 150% of poverty led to a large increase in Medicaid enrollment by pregnant women. The new enrollees were women who otherwise would have lacked insurance coverage to pay for their prenatal care and delivery. That is, the expansion did not substitute for private insurance. The increased insurance coverage was associated with improved access to prenatal care for low-income pregnant women, a finding confirmed by multivariate analysis of the effects of Medicaid eligibility on individuals' access to prenatal care in Florida.11
The public health care system played a significant role in the Florida experience. Most of the additional prenatal care financed by Medicaid was accommodated through county health departments, which resulted in a doubling in the amount of prenatal care provided by these public clinics. Thus, the Medicaid expansion did not substitute for care already provided by county health departments; rather, it appears to have provided the resources needed to extend the counties' capacity to deliver prenatal care to low-income women.
Moreover, initiatives were developed to help counties collect Medicaid payments for previously eligible patients whose care had not been reimbursed. Such efforts, coupled with the infusion of new funds to county health departments stemming from the eligibility expansion, may have provided the resources to expand coverage to other populations: We found some evidence of improved access for pregnant women who remained uninsured; other public health system clients may have benefited as well.
The expansion had little effect on which types of hospitals women used for their maternity care. However, hospitals benefited financially from the expansion because Medicaid paid for many deliveries that previously had been financed as bad-debt or charity care. Although we do not have evidence about the use of these resources, they may have provided spillover benefits for other populations if the resources previously directed to providing uncompensated treatment to poor pregnant women were used to extend charity services to others.
In sum, the Medicaid expansion benefited low-income pregnant women and providers of maternity services in Florida. Much of the success of the Florida experience, however, may be due to the ability of the public health system to expand its provision of clinical services in response to increased demand for prenatal care. Whether the private delivery system would accommodate the increased demand and provide improved access for newly entitled women in the absence of a strong public health tradition of providing maternal health care will require study of the expansions in other states.
*Our examination of payment flows includes only specific payments for this patient population; it does not include any expansions in county health department budgets that may have been used to extend care to this population.
1. Alpha Center, The Medicaid Expansions for Pregnant Women and Children, Washington, DC: Alpha Center, 1995.
2. Cutler DM and Gruber J, Medicaid and private insurance: evidence and implications, Health Affairs, 1997, 16(1):194-200.
3. Hill IT, Implementing the Medicaid Expansions for Pregnant Women: The Experience in Florida, Washington, DC: Health Systems Research, Inc., 1995.
5. Harrison ER, Long SH and Marquis MS, Evaluation of a Medicaid Eligibility Expansion in Florida: Developing the Database, Santa Monica, CA: RAND, 1996.
6. Levit KR et al., National health expenditures, 1994, Health Care Financing Review, 1996, 17(3):205-242.
7. Long SH, Marquis MS and Harrison ER, The costs and financing of perinatal care in the United States, American Journal of Public Health, 1994, 84(9):1473-1478.
8. Cutler DM and Gruber J, 1997, op. cit. (see reference 2).
9. Cooper PF and Schone BS, More offers, fewer takers for employment-based health insurance: 1987 and 1996, Health Affairs, 1997, 16(6):142-149.
10. Long SH, Marquis MS and Harrison ER, 1994, op. cit. (see reference 7).
11. Long SH and Marquis MS, The effects of Florida's Medicaid eligibility expansion for pregnant women, American Journal Public Health, 1998, 88(3):371-376.