Abstract
Objective:
To evaluate the effect of the 2013–2014 Affordable Care Act (ACA) Medicaid Primary Care Rate Increase on Medicaid-insured women’s prenatal care utilization, overall and by race and ethnicity.
Methods:
We employed a difference-in-differences design, using births data from the 2010–2014 National Vital Statistics System. Our study population included approximately 6.2 million births to Medicaid insured mothers conceived between April 2009 and March 2014. Our treatment group was births in states with a large (relative to small) fee bump, defined as having a Medicaid-to-Medicare fee ratio below the median of all states (0.7) in 2012. Our control group was births in states with a small fee bump. Prenatal care utilization measures included initiation of prenatal care in the first trimester and number of prenatal care visits.
Results:
Non-Hispanic Black women giving births in large fee bump states had 9% higher odds (95% CI 1.02, 1.17) of initiating prenatal care in the first trimester during the fee bump period, compared to small fee bump states. Prenatal care visits in this group also increased by 0.24 (95% CI 0.10, 0.39), 2.4% of the mean. A smaller increase in prenatal care visits of 0.17 (95% CI 0.00, 0.33) was found among non-Hispanic Whites. The fee bump had no impact among Hispanics or non-Hispanic women of other races.
Conclusions for practice:
The Medicaid “fee bump” improved prenatal care utilization for non-Hispanic Black and White women. Policymakers may consider reinstating higher Medicaid reimbursements to improve access to care for disadvantaged populations.
Keywords: Medicaid fee bump, prenatal care, racial and ethnic disparities, difference-in-differences
INTRODUCTION
In this study, we examined the effect of the largest increase in Medicaid reimbursement rates for primary care services in the history of the program on prenatal care utilization. The reimbursement increases were a result of the Medicaid Primary Care Rate Increase (the “fee bump”) enacted through the Affordable Care Act (ACA). The fee bump used federal funds to increase Medicaid reimbursements for primary care services to Medicare levels during a two-year period, 2013–2014. The goal of the intiative was to increase access to primary care providers for Medicaid beneficiaries. Studies have shown that the Medicaid fee bump increased availability of primary care appointments (Polsky et al., 2015) and overall access to primary care among Medicaid patients ( D. Alexander & Schnell, 2017). For these reasons, pregnant Medicaid beneficiaries may have improved access to primary care, leading to increased utilization of prenatal care.
States’ Medicaid programs generally reimburse health care providers at much lower rates than Medicare and private insurers (Zuckerman & Goin, 2012; Zuckerman, Williams, & Stockley, 2009). To rectify this disparity, the federal government spent between $7 and $12 billion on the fee bump in 2013 and 2014 (Medicaid and CHIP Payment and Access Commission, 2015). As an example of how the fee bump increased the incentive to provide more primary care to Medicaid beneficiaries, the mean Medicaid reimbursement to a physician for a 30-minute office visit was approximately $65 in 2012, but increased to approximately $110 when the intiative was implemented in 2013 (Zuckerman & Goin, 2012). The fee increases covered 146 primary care services for physicians specializing in family medicine, general internal medicine, or pediatric medicine (Medicare & Medicaid Services 2012; Zuckerman & Goin, 2012). Importantly, they included a range of primary care services that are important in Medicaid, but that are not reimbursed by Medicare, such as new and established patient comprehensive preventive medicine, counseling for risk factor reduction and behavioral change interventions (Kaiser Family Foundation, 2012).
Although prenatal care was not covered by the initiative, and obstetrician gynecologists (OB-GYNs) did not qualify for the higher primary care reimbursements, increased access to primary care resulting from the fee bump may have led to more use of prenatal care among Medicaid-insured women before and during pregnancy through two mechanisms. First, pregnant women may have been more likely to be referred to prenatal care by primary care physicians. Second, increased access to primary care may improve low-income women’s awareness of the importance of prenatal care. Either mechanism could lead to earlier initiation of prenatal care and better adherence to prenatal care. These mechanisms may be especially important among Medicaid-insured women as barriers to prenatal care associated with income and race have been well-documented (G. R. Alexander, Kogan, & Nabukera, 2002; Currie & Grogger, 2002; Dubay, Joyce, Kaestner, & Kenney, 2001). Hispanic and African American women are known to enter prenatal care later than white women, and have a higher likelihood of non-adherence to prenatal care (G. R. Alexander et al., 2002; Mayberry, Mili, & Ofili, 2000).
We exploited variation in the size of the fee bump across states to examine the relationships between higher reimbursements for primary care services and prenatal care utilization of women insured by Medicaid. Further, we study the relationship separately by race and ethnicity to examine the impact of the fee bump on disparities in prenatal care utilization. We hypothesized that prenatal care utilization among Medicaid-insured women improved in states with larger fee bump relative to women in states with lower fee bump.
METHODS
Study Design
We used a difference-in-differences (DD) multivariable regression design to compare changes in prenatal care utilization among Medicaid women in states with larger Medicaid fee bump increases (the treatment group) to women in states with smaller fee bump increases (the control group) between 2010 and 2014. This approach compares two changes: the change in prenatal care utilization in the treatment group after versus before the implementation of the Medicaid fee bump, and such change after versus before the implementation of the fee bump in the control group (Dimick & Ryan, 2014). By doing so, our approach accounts for any confounding caused by differences between treatment and control groups that stay constant throughout the study period (e.g. prenatal care infrastructure), and any time trends that would have occurred to all states in the absence of the fee bump. We also controlled for an extensive set of individual, county and state characteristics (detailed below).
Study Data and Sample
Our primary data source is Birth Data of the National Vital Statistics System from the National Center for Health Statistics for years 2010 through 2014. The Standard Certificate of Live Birth was revised in 2003 and used in 38 states by 2010. The Certificate provides demographic information on mothers (age, education, race, and state and county of residence), payment source (at the time of birth), order of birth, and prenatal care utilization (Centers for Disease Control and Prevention, 2017). We additionally used data from the Area Health Resource File (AHRF) to construct county characteristics as control variables.
We examined births from women insured by Medicaid. We only included births that were conceived between April 2009 through March 2014 to avoid including outliers with unusually long or short gestation at the beginning and end of our study period. This inclusion is important, as our treatment period was defined with respect to the timing of conception (more detail below). We excluded births in 12 states that were not using the revised birth record form by 2010 because their records do not contain information on payment source (see Panel B of Table A1 in Supplemental Appendix). We also excluded births in Tennessee from our analysis as it does not have a fee-for-service component in its Medicaid program (Zuckerman & Goin, 2012) and therefore the size of its Medicaid fee bump could not be identified. Our final analysis sample consists of 6,214,618 births in 37 states.
Variables
Dependent Variables
We examined two established measures of prenatal care utilization: (1) an indicator for whether prenatal care was initiated during the first trimester and (2) the number of prenatal care visits. Both measures have been used to indicate adequacy of prenatal care which has been linked to improved birth outcomes (G. R. Alexander & Comely, 1987; Kotelchuck, 1994; Krueger & Scholl, 2000; Partridge, Balayla, Holcroft, & Abenhaim, 2012).
Exposure: State Fee Bump Size
The classification of large versus small fee bump states was based on the state-specific 2012 Medicaid-to-Medicare Fee Index for the Affordable Care Act (ACA), the ratio of the state’s Medicaid fees to Medicare fees for primary care services covered under the fee bump (Zuckerman & Goin, 2012). The initial (pre-fee bump) difference in payment rates between Medicaid and Medicare providers effectively determined the size of the fee bump, as the policy required payment of primary care services to Medicaid providers to be raised to the same level as Medicare. Large-fee bump states were defined as states with a median Medicaid-to-Medicare fee index below the national median (70%) in 2012 (the year prior to the fee bump), and low-fee bump states were states above the median. High fee bump states had an average Medicaid-to-Medicare fee ratio of 57% in 2012, the year prior to the fee bump, and therefore the Medicaid primary care fees increased by an average of 43 percentage points in these states, compared to an average increase of 15 percentage points in low fee bump states (Table 1, Panel A). The geographic locations of the states in each category are shown in Figure A1. Such variation has been documented by previous literature (Wilk & Jones, 2014).
Table 1.
2012 Medicaid-to-Medicare fee index for affordable care act primary care services, categorization of large fee bump states and revised birth records status
| Revised Birth | ||||
|---|---|---|---|---|
| State | 2012 Medicaid-to- | Large Fee | Records Since | |
| Medicare Fee Index | Bump State | 2010 | ||
| Panel A: States Included in Analytical Sample | ||||
| NY | 0.39 | Y | Y | |
| CA | 0.42 | Y | Y | |
| MI | 0.44 | Y | Y | |
| FL | 0.49 | Y | Y | |
| PA | 0.51 | Y | Y | |
| IL | 0.52 | Y | Y | |
| IN | 0.54 | Y | Y | |
| WI | 0.56 | Y | Y | |
| MO | 0.57 | Y | Y | |
| OH | 0.57 | Y | Y | |
| NH | 0.59 | Y | Y | |
| TX | 0.60 | Y | Y | |
| NV | 0.66 | Y | Y | |
| WA | 0.66 | Y | Y | |
| GA | 0.67 | Y | Y | |
| SD | 0.67 | Y | Y | |
| MD | 0.69 | Y | Y | |
| KY | 0.70 | Y | Y | |
| NE | 0.72 | N | Y | |
| OR | 0.72 | N | Y | |
| SC | 0.74 | N | Y | |
| UT | 0.74 | N | Y | |
| VA | 0.74 | N | Y | |
| CO | 0.75 | N | Y | |
| IA | 0.75 | N | Y | |
| LA | 0.75 | N | Y | |
| KS | 0.77 | N | Y | |
| DC | 0.80 | N | Y | |
| NM | 0.82 | N | Y | |
| VT | 0.82 | N | Y | |
| NC | 0.85 | N | Y | |
| ID | 0.89 | N | Y | |
| MT | 0.94 | N | Y | |
| WY | 0.96 | N | Y | |
| OK | 0.97 | N | Y | |
| DE | 0.98 | N | Y | |
| ND | 1.37 | N | Y | |
| Panel B:States Excluded from Analytical Sample | ||||
| TN | N/A | N/A | Y | |
| AL | 0.68 | Y | N | |
| AR | 0.68 | Y | N | |
| AZ | 0.75 | N | N | |
| HI | 0.56 | Y | N | |
| MA | 0.68 | Y | N | |
| ME | 0.62 | Y | N | |
| MN | 0.73 | N | N | |
| MS | 0.90 | N | N | |
| NJ | 0.48 | Y | N | |
| WV | 0.74 | N | N | |
| CT | 0.71 | N | N | |
| RI | 0.34 | Y | N | |
Notes: Data on 2012 Medicaid-to-Medicare Medicaid fee index are from Zuckerman & Goin (2012).
Although we relied on Medicaid-to-Medicare fee index to classify states into large versus small fee bump categories, the primary care rate increase also applied to Medicaid managed care plans (Centers for Medicare and Medicaid Services, 2012). In addition, managed care plans often receive monthly capitation payments based on what states would have paid for care on a fee-for-service basis (Zuckerman & Goin, 2012). Thus, the fee-for-service reimbursement ratio that we use should proxy managed care payment increases associated with the fee bump.
Treatment Period
We defined a birth as occurring during the treatment period if the conception occurred between January 2013 and March 2014. This definition ensured that the entire pregnancy occurred during the fee bump period and that mothers could have benefited during their full pregnancy. Our difference-in-differences strategy involves interacting the indicator for a large fee bump state and the indicator for the treatment period.
Control Variables
We controlled for an extensive set of the mother’s demographic characteristics, state-level and county-level time-varying characeristics, indicators for states, and indicators for year-month of conception. The mother’s demographic characteristics included age indicators, education level (less than high school, high school, some college and bachelor’s degree or more), race/ethnicity (non-Hispanic white, non-Hispanic black and non-Hispanic other and Hispanic), and indicators for birth order. County-year level controls included the number of primary care physicians per 1,000 child-bearing age woman, number of OB-GYNs per 1,000 child-bearing age woman, median household income (<=$25,000, $25,001–$50,000, $50,001–$75,000, $75,001–$100,000 and >$100,000) and the percentage of the population in poverty (defined as income below 135% of the federal poverty line). We also control for a state’s Medicaid expansion status (see eTable 1).
Statistical Analysis
For summary statistics, t-tests were used to assess differences in continuous variables between births in high and low fee bump states. Chi-squared tests were used to assess differences in categorical variables. For multivariable regression models using the initiation of prenatal care in the first trimester as the outcome, odds ratios were estimates with logistic regression. We estimated ordinary least squares (OLS) regressions with the number of prenatal care visits as the outcome. Each of our regressions included an indicator for whether the birth occurred in one of the large fee bump states (relative to births in small fee bump states), an indicator for whether the birth was conceived during the treatment period of years 2013–2014 (relative to pre-treatment period of years 2010–2012), an interaction term for the two indicators (which is our primary coefficient of interest, showing the effect of the fee bump in large fee bump states), and the control variables described above. We estimated the regressions for all births to Medicaid-insured women, as well as stratified by racial and ethnic group. We clustered standard errors at the state level.
Robustness Checks
We conducted two additional sets of analyses to further examine the robustness of our study design. First, we repeated the analyses on births from Medicaid-insured women only in states that did not expand Medicaid during our study period. This restriction addressed the concern that our results in Panel A may be confounded by other changes associated with the Medicaid expansion such as changes in racial and ethnic composition of the Medicaid population. For these analyses, we clustered the standard errors at the county level as there were not enough states to allow for state-level clustering (Cameron & Miller, 2015).
Second, we repeated the analyses using births from privately insured women who should not have been affected by the Medicaid fee bump as a placebo test. Although our statistical analysis employed DD regression models and controled for a variety of individual, county and state characteristics, it is still possible that differential changes in state characteristics (such as socioeconomic conditions and availability of prenatal medical care) overtime may confound our results. A null result from estimating the same DD models among privately insured women not subject to the fee bump would strengthen the credibility of our analytical methods. Stata MP version 14 (StataCorp) was used for all analyses. Weill Cornell Medical College Institutional Review Board approved this research. Informed consent was waived for the study as it was retrospective and involved no more than minimum risk to subjects.
RESULTS
After our restrictions, our sample included birth records for approximately 6 million pregnant women. Table 2 shows descriptive statistics for the sample, overall and by state fee bump size (large or small). Medicaid mothers in large and small fee bump states were reasonably similar in age and education, although large fee bump states had a larger proportion of Hispanic mothers and a smaller proportion of White non-Hispanic mothers than small fee bump states. The distribution of birth order and the timing of birth were almost identical in the two groups. County characteristics were similar between large and small fee bump states, except that Medicaid mothers in large fee bump states were more likely to live in counties with somewhat higher median household income (greater than $50,000). Finally, large fee bump states were also more likely to have expanded Medicaid. All characteristics were significantly different (p<0.05) between large and fee bump states because of large sample size, except birth order. Table 3 shows our difference-in-difference results for initiation of prenatal care during the first trimester, for all Medicaid mothers as well as by race and ethnicity. Panel A presents results using births in all states included in our sample. Being in a large fee bump state did not have a significant impact on Medicaid mothers overall or among most racial and ethnic groups, with the exception of non-Hispanic Black women. Non-Hispanic Black women in large fee bump states had 9% higher odds (OR = 1.09; 95% CI: 1.02, 1.17; P = 0.015) of initiating prenatal care in the first trimester after the fee bump was implemented relative to Non-Hispanic Black women in small fee bump states.
Table 2.
Summary statistics on births of Medicaid-insured women by state fee bump size in 37 states (mean (sd), n=6,214,618)
| By State Fee Bump Sizea | |||
|---|---|---|---|
| All | Small | Large | |
| (37 states, N=6,214,618) |
(18 states, N=l,249,138) |
(19 states, N=4,965,480) |
|
| Dependent Variables | |||
| Initiation of prenatal care in the first trimester (%) | 65 | 64 | 66 |
| Number of prenatal care visits | 10.78 (4.15) | 10.86 (4.19) | 10.74 (4.13) |
| Mother Characteristics | |||
| Age | 25.73 (5.82) | 25.23 (5.58) | 25.85 (5.88) |
| Education (%) | |||
| Less than high school | 30 | 28 | 30 |
| High school | 36 | 35 | 37 |
| Some college | 28 | 31 | 27 |
| Bachelor or more | 6 | 6 | 6 |
| Race/Ethnicity (%) | |||
| White, non-Hispanic | 38 | 49 | 35 |
| Black, non-Hispanic | 21 | 23 | 21 |
| Other, Non-white | 6 | 7 | 6 |
| Hispanic | 35 | 21 | 39 |
| Birth order (%) | |||
| 1st | 31 | 31 | 31 |
| 2nd | 26 | 26 | 26 |
| 3rd | 19 | 19 | 19 |
| 4th | 11 | 11 | 11 |
| 5th | 6 | 6 | 6 |
| 6th | 3 | 3 | 3 |
| 7th | 2 | 2 | 2 |
| 8th or higher | 2 | 2 | 2 |
| Conceived during fee bump period (%) | 25 | 26 | 25 |
| County Characteristics | |||
| # PCP per 1,000 childbearing age woman | 3.72 (1.35) | 3.86 (1.25) | 3.68 (1.37) |
| # OBGYN per 1,000 childbearing age woman | 0.63 (0.33) | 0.63 (0.34) | 0.63 (0.33) |
| Median household income (%) | |||
| <= $25,000 | 0.03 | 0.00 | 0.03 |
| $25,001–$50,000 | 52 | 63 | 49 |
| $50,001–$75,000 | 44 | 37 | 46 |
| $75,001–$100,000 | 5 | 1 | 6 |
| >$100,000 | 0.1 | 0.1 | 0.1 |
| Pct population in poverty | 17.56(5.22) | 16.96(4.62) | 17.70(5.35) |
| State Characteristics | |||
| Expanded Medicaid by time of birth (%) | 11 | 5 | 12 |
All characteristics were significant different (p<0.05) between large and fee bump states because of large sample size, except birth order.
Table 3.
Adjusted estimates of the relationship between the fee bump and intiation of prenatal care in the first trimester among Medicaid-insured women in 37 states (n=6,214,618)
| By Race and Ethnicity | |||||
|---|---|---|---|---|---|
| All | White, Non-Hispanic | Black, Non-Hispanic | Other, Non-Hispanic | Hispanic | |
| (1) | (2) | (3) | (4) | (5) | |
| Panel A. All States in Analytical Sample | |||||
| Large fee bump | |||||
| states × Post | |||||
| OR | 1.02 | 1.02 | 1.09* | 1.03 | 0.94 |
| 95% CI | (0.94, 1.11) | (0.93, 1.11) | (1.02, 1.17) | (0.95, 1.13) | (0.86, 1.02) |
| Mean of | |||||
| Dependent Var | 0.65 | 0.67 | 0.60 | 0.63 | 0.67 |
| N | 6,214,618 | 2,383,501 | 1,267,105 | 367,107 | 2,196,905 |
| Panel B. States without Medicaid Expansion by 2014 | |||||
| Large fee bump | |||||
| states × Post | |||||
| OR | 1.04 | 1.04 | 1.13* | 0.93 | 0.93 |
| 95% CI | (0.97, 1.11) | (0.98, 1.10) | (1.03, 1.24) | (0.83,1.04) | (0.85, 1.03) |
| Mean of | |||||
| Dependent Var | 0.61 | 0.65 | 0.58 | 0.56 | 0.59 |
| N | 2,783,603 | 1,142,940 | 697,871 | 115,837 | 826,955 |
Notes: Unit of observation is individual birth. Cluster-robust standard errors in parentheses, clustered by state in Panel A and by county in Panel B. All regressions control for mother, county and state characteristics in Table 1, as well as year-month and state fixed effects.
p<0.05,
p<0.01.
Panel B includes results using only births in the states that did not expand Medicaid by 2014. These results were very similar to the results in Panel A. In states without Medicaid expansion, non-Hispanic Black women in large fee bump states had 13% higher odds (OR = 1.13; 95% CI: 1.03, 1.24; P = 0.013) of initiating prenatal care in the first trimester. None of the other racial and ethnic groups experienced statistically significant changes in the odds of initiating primary care in the first trimester. Table 4 shows analogous results using the number of prenatal care visits as the dependent variable. The effect of the fee bump was heterogeneous across racial and ethnic groups: the fee bump led to an increase of 0.17 (95% CI: 0.00, 0.33; P = 0.045) visits among non-Hispanic White women and 0.24 (95% CI: 0.10, 0.39; P= 0.002) visits among non-Hispanic Black women, corresponding to 1.5% and 2.4% of the mean number of prenatal care visits, respectively, in each group. No signifant changes in prenatal care visits were observed among Hispanic women and non-Hispanic women of other races. Accordingly, the overall number of prenatal care visits increased by 0.13 (95% CI: −0,02, 0.29; P = 0.093). Similarly, Panel B shows that among Medicaid mothers giving births in states that did not expand Medicaid by 2014, the fee bump led to an increase of 0.20 (95% CI: 0.03, 0.37; P = 0.024) visits overall, 0.18 (95% CI: 0.02, 0.34; P = 0.029) visits among the non-Hispanic White women, and 0.33 (95% CI: 0.07, 0.59; P = 0.012) visits among non-Hispanic Black women. These results correspond to relative increases of 1.9%, 1.7% and 3.2%, respectively, compared to the group means.
Table 4.
Adjusted estimates of the relationship between the fee bump and the number of prenatal care visits among Medicaid-insured women in 37 states (n=6,214,618)
| By Race and Ethnicity | |||||
|---|---|---|---|---|---|
| All | White, Non-Hispanic | Black, Non-Hispanic | Other, Non-Hispanic | Hispanic | |
| (1) | (2) | (3) | (4) | (5) | |
| Panel A. All States in Analytical Sample | |||||
| Large fee bump | |||||
| states × Post | |||||
| Coefficient | 0.13 | 0.17* | 0.24** | 0.05 | −0.06 |
| 95% CI | (−0.02, 0.29) | (0.00, 0.33) | (0.10, 0.39) | (−0.14, 0.23) | (−0.23, 0.10) |
| Mean of | |||||
| Dependent Var | 10.78 | 11.08 | 10.17 | 10.56 | 10.83 |
| N | 6,214,618 | 2,383,501 | 1,267,105 | 367,107 | 2,196,905 |
| Panel B. States vithout Medicaid Expansion by 2014 | |||||
| Large fee bump | |||||
| states × Post | |||||
| Coefficient | 0.20* | 0.18* | 0.33* | 0.02 | 0.01 |
| 95% CI | (0.03, 0.37) | (0.02, 0.34) | (0.07, 0.59) | (−0.26, 0.29) | (−0.22, 0.24) |
| Mean of | |||||
| Dependent Var | 10.45 | 10.88 | 10.18 | 9.93 | 10.16 |
| N | 2,783,603 | 1,142,940 | 697,871 | 115,837 | 826,955 |
Notes: Unit of observation is individual birth. Cluster-robust standard errors in parentheses, clustered by state in Panel A and by county in Panel B. All regressions control for mother, county and state characteristics in Table 1, as well as year-month and state fixed effects.
p<0.05,
p<0.01.
We did not find any significant effect of being in a large fee bump state on prenatal care utilization among privately insured patients in the overall sample nor by race and ethnicity (Table A2).
DISCUSSION
We explored the effect of increases in Medicaid reimbursement rates for primary care services that may have led to changes in prenatal care utilization among Medicaid mothers. We found that the fee bump increased the odds of initating prenatal care in the first trimester by 9 to 13 percentage points among non-Hispanic Black women. In addition, the Medicaid fee bump significantly increased the number of prenatal care visits among non-Hispanic White and Black women by 1.5 to 3.2%. The results were not driven by confounding factors related to the Medicaid expansion. The fact that we did not find any effect of the fee bump on privately insured women adds to the validity of our findings.
While the fee bump did not directly incentivize the provision of more prenatal care, it could have benefited women either before or during pregnancy through improved access to primary care. For example, Medicaid-insured women may have had more visits with primary care physicians, more time with physicians during visits, and more communication with physicians between visits. Improved access to primary care could have led to more referrals to prenatal care or increased awareness of the importance of prenatal care. The fact that we found small but significant improvement in prenatal care utilization among non-Hispanic Black women (and to some extent among non-Hispanic White women) suggests that the fee increases narrowed disparities associated with race. On the other hand, we did not find any change in prenatal care utilization among Hispanic women or non-Hispanic women of other races, possibly because language and other barriers prevented them from taking advantage of improved access to primary and prenatal care (Avila & Bramlett, 2013; Clayman, Manganello, Viswanath, Hesse, & Arora, 2010; Fernandez et al., 2011; Fiscella, Franks, Doescher, & Saver, 2002; Timmins, 2002).
Our results have important implications for Medicaid reimbursement policy. Prior studies evaluating the effect of the fee bump may have underestimated the overall impact of the policy by solely focusing on primary care quality and outcomes. Following the expiration of federal funding for the fee bump at the end of 2014, only 19 states have used their own funds to continue the higher Medicaid reimbursements for primary care (Zuckerman, Skopec, & Epstein, 2017). Further, a recent study found that decreases in the Medicaid primary care rate following the expiration of the fee bump was associated with decreases in primary care appointment availability, raising concerns about the sustainability of any gains in care access brought by the fee bump policy (Candon et al., 2018). This study adds to the evidence base that supports reinstating federal subsidies to all states to continue the higher reimbursement levels of the Medicaid program. The federal government may also consider expanding the subsidies to support prenatal care services. Additional measures are also needed to improve prenatal care access specifically in the minority population, including Hispanics and non-Hispanics of other races. Innovative interventions such as prenatal care coordination programs and group prenatal care may be promising in this regard, as supported by recent literature (Gareau et al., 2016; Hillemeier et al., 2018; Mazul, Ward, & Ngui, 2017; Meghea, You, Raffo, Leach, & Roman, 2015).
Limitations
Our study has several limitations. First, we were not able to directly examine the impact of the fee bump on primary care utilization among Medicaid mothers to conclusively determine the mechniam through which the fee bump affected prenatal care utilization. Second, we lacked data on the content of prenatal care received, such as specific services provided and counseling on health behavior. Such information would provide a more comprehensive picture of the quality of prenatal care among Medicaid mothers. Third, we excluded 12 states because of data limitations. However, our sample included 37 states, covering the majority of births of Medicaid-insured women occurring during the study period.
CONCLUSION
To our knowledge, this study is the first that examines the effect of the Medicaid fee bump on prenatal care. We found that being in a large fee bump state led to small but significant increases in the odds of initiating prenatal care in the first trimester and in the number of prenatal care visits for non-Hispanic Black women. To a lesser extent, the number of prenatal care visits also increased among non-Hispanic White women. We did not find any effect of the fee bump on Hispanic women or women of other races. Our findings highlight the potential of higher Medicaid reimbursements to improve access to care and reduce income and racial disparities.
Supplementary Material
SIGNIFICANCE.
What is already known on this subject? Inadequacy and racial disparities in prenatal care are long-standing problems among low-income women in the United States.
What this study adds? This study is the first examining the effect of the Medicaid fee bump on prenatal care utilization, both overall and by the mother’s race and ethnicity. We found that the fee bump improved prenatal care utilization in non-Hispanic Black and White women. Our findings highlight the potential for higher Medicaid reimbursement rates to improve access to prenatal care for disadvantaged populations.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
REFERENCES
- Alexander D, & Schnell M (2017). Closing the Gap: The Impact of the Medicaid Primary Care Rate Increase on Access and Health. [Google Scholar]
- Alexander GR, & Comely DA (1987). Prenatal care utilization: its measurement and relationship to pregnancy outcome. American journal of preventive medicine, 3(5), 243–253. [PubMed] [Google Scholar]
- Alexander GR, Kogan MD, & Nabukera S (2002). Racial differences in prenatal care use in the United States: Are disparities decreasing?? American Journal of Public Health, 92(12), 1970–1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Avila RM, & Bramlett MD (2013). Language and immigrant status effects on disparities in Hispanic children’s health status and access to health care. Maternal and child health journal, 17(3), 415–423. [DOI] [PubMed] [Google Scholar]
- Cameron AC, & Miller DL (2015). A practitioner’s guide to cluster-robust inference. Journal of Human Resources, 50(2), 317–372. [Google Scholar]
- Candon M, Zuckerman S, Wissoker D, Saloner B, Kenney GM, Rhodes K, & Polsky D (2018). Declining Medicaid fees and primary care appointment availability for new Medicaid patients. JAMA internal medicine, 178(1), 145–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2017). Revisions of the U.S. Standard Certificates and Reports. Retrieved November 1, 2018, from https://www.cdc.gov/nchs/nvss/revisions-of-the-us-standard-certificates-and-reports.htm [Google Scholar]
- Centers for Medicare and Medicaid Services. (2012). Medicaid program; payments for services furnished by certain primary care physicians and charges for vaccine administration under the Vaccines for Children program. Final rule. Federal register, 77(215), 66669. [PubMed] [Google Scholar]
- Clayman ML, Manganello JA, Viswanath K, Hesse BW, & Arora NK (2010). Providing health messages to Hispanics/Latinos: understanding the importance of language, trust in health information sources, and media use. Journal of health communication, 15(sup3), 252–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Currie J, & Grogger J (2002). Medicaid expansions and welfare contractions: offsetting effects on prenatal care and infant health? Journal of Health Economics, 21(2), 313–335. [DOI] [PubMed] [Google Scholar]
- Dimick JB, & Ryan AM (2014). Methods for evaluating changes in health care policy: the difference-in-differences approach. Jama, 312(22), 2401–2402. [DOI] [PubMed] [Google Scholar]
- Dubay L, Joyce T, Kaestner R, & Kenney GM (2001). Changes in prenatal care timing and low birth weight by race and socioeconomic status: implications for the Medicaid expansions for pregnant women. Health Services Research, 36(2), 373. [PMC free article] [PubMed] [Google Scholar]
- Fernandez A, Schillinger D, Warton EM, Adler N, Moffet HH, Schenker Y,… Karter AJ (2011). Language barriers, physician-patient language concordance, and glycemic control among insured Latinos with diabetes: the Diabetes Study of Northern California (DISTANCE). Journal of general internal medicine, 26(2), 170–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fiscella K, Franks P, Doescher MP, & Saver BG (2002). Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Medical care, 52–59. [DOI] [PubMed] [Google Scholar]
- Gareau S, Lòpez-De Fede A, Loudermilk BL, Cummings TH, Hardin JW, Picklesimer AH,… Covington-Kolb S (2016). Group prenatal care results in Medicaid savings with better outcomes: a propensity score analysis of Centering Pregnancy participation in South Carolina. Maternal and child health journal, 20(7), 1384–1393. [DOI] [PubMed] [Google Scholar]
- Hillemeier MM, Domino ME, Wells R, Goyal RK, Kum HC, Cilenti D, & Basu A (2018). Does maternity care coordination influence perinatal health care utilization? Evidence from North Carolina. Health Services Research, 53(4), 2368–2383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaiser Family Foundation. (2012). Increasing Medicaid Primary Care Fees for Certain Physicians in 2013 and 2014: A Primer on the Health Reform Provision and Final Rule. The Kaiser Family Foundation Issue Brief. from https://kaiserfamilyfoundation.files.wordpress.com/2013/01/8397.pdf [Google Scholar]
- Kotelchuck M (1994). An evaluation of the Kessner adequacy of prenatal care index and a proposed adequacy of prenatal care utilization index. American Journal of Public Health, 84(9), 1414–1420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Krueger PM, & Scholl TO (2000). Adequacy of prenatal care and pregnancy outcome. The Journal of the American Osteopathic Association, 100(8), 485–492. [PubMed] [Google Scholar]
- Mayberry RM, Mili F, & Ofili E (2000). Racial and ethnic differences in access to medical care. Medical Care Research and Review, 57(1_suppl), 108–145. [DOI] [PubMed] [Google Scholar]
- Mazul MC, Ward TCS, & Ngui EM (2017). Anatomy of good prenatal care: Perspectives of low income African-American women on barriers and facilitators to prenatal care. Journal of racial and ethnic health disparities, 4(1), 79–86. [DOI] [PubMed] [Google Scholar]
- Medicaid and CHIP Payment and Access Commission. (2015). Report to Congress on Medicaid and CHIP - March 2015. Retrieved August 21, , 2017, from https://www.macpac.gov/wp-content/uploads/2015/03/March-2015-Report-to-Congress-on-Medicaid-and-CHIP.pdf [Google Scholar]
- Meghea CI, You Z, Raffo J, Leach RE, & Roman LA (2015). Statewide medicaid enhanced prenatal care programs and infant mortality. Pediatrics, 136(2), 334–342. [DOI] [PubMed] [Google Scholar]
- Partridge S, Balayla J, Holcroft CA, & Abenhaim HA (2012). Inadequate prenatal care utilization and risks of infant mortality and poor birth outcome: a retrospective analysis of 28,729,765 US deliveries over 8 years. American journal of perinatology, 29(10), 787. [DOI] [PubMed] [Google Scholar]
- Polsky D, Richards M, Basseyn S, Wissoker D, Kenney GM, Zuckerman S, & Rhodes KV (2015). Appointment availability after increases in Medicaid payments for primary care. New England Journal of Medicine, 372(6), 537–545. [DOI] [PubMed] [Google Scholar]
- Timmins CL (2002). The impact of language barriers on the health care of Latinos in the United States: a review of the literature and guidelines for practice. The Journal of Midwifery & Women’s Health, 47(2), 80–96. [DOI] [PubMed] [Google Scholar]
- Wilk AS, & Jones DK (2014). To extend or not to extend the primary care “fee bump” in Medicaid? Journal of health politics, policy and law, 39(6), 1263–1275. [DOI] [PubMed] [Google Scholar]
- Zuckerman S, & Goin D (2012). How much will Medicaid physician fees for primary care rise in 2013? Evidence from a 2012 survey of Medicaid physician fees. The Kaiser Family Foundation Issue Paper. [Google Scholar]
- Zuckerman S, Skopec L, & Epstein M (2017). Medicaid Physician Fees after the ACA Primary Care Fee Bump: Washington: Urban Institute. [Google Scholar]
- Zuckerman S, Williams AF, & Stockley KE (2009). Trends in Medicaid physician fees, 2003–2008. Health Affairs, 28(3), w510–w519. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.