Prepregnancy Health Care Engagement Among American Indian and Alaska Native People Before and After the Affordable Care Act

Abstract

Background:

Despite including several provisions focused on American Indian and Alaska Native (AI/AN) people, few studies look at the Affordable Care Act (ACA)'s potential impact among AI/AN birth givers, a group that experiences a disproportionate burden of adverse pregnancy outcomes. While ACA repeal conversations are ongoing, our objective was to examine changes, and equity in changes, in health care interactions before and after the ACA between AI/AN birth givers and white birth givers.

Materials and Methods:

We used the 2009–2015 data from the Pregnancy Risk Assessment Monitoring System, a state-level representative sample of individuals with live births. We included those identifying as AI/AN people or white people to estimate change over time in five health measures. We used adjusted linear probability regression models to compare prepregnancy recall periods before, during, and after the implementation of the ACA with the period just before implementation.

Results:

Among AI/AN birth givers, prepregnancy Medicaid coverage, multivitamin/prenatal use, and teeth cleaning increased by 10 percentage points (ppts), 5 ppts, and 9 ppts, respectively, and were larger than increases experienced by white birth givers (5, 3, and 2–4 ppts, respectively). Increases in preventive health care screening (12–18 ppts) and provider conversation (8–10 ppts) were similar for AI/AN birth givers and white birth givers.

Discussion:

While some ACA-associated increases in health care coverage and care were quantitatively larger among AI/AN birth givers compared with white birth givers, the existence of pre-ACA disparities suggests that the ACA may have reduced, but may not have eliminated, health inequities between AI/AN birth givers and white birth givers.

Introduction

Birth ceremonies help usher new life into the physical world and create space for a community to honor this sacred process. Unfortunately, American Indian and Alaska Native birth givers bear a disproportionate burden of poor pregnancy health outcomes (e.g., preterm birth, hypertensive disorders, gestational diabetes mellitus, severe maternal morbidity, and maternal and infant mortality), compared with white birth givers,^1–8 that can disrupt these important ceremonies.⁹ Furthermore, poor pregnancy outcomes portend future health concerns experienced by families, birth givers, and children. While essential for many reasons, prenatal care does little to improve disparities in birth outcomes,^10–12 partly because some proximal causes of poor pregnancy health originate before pregnancy (e.g., overweight or obesity, chronic hypertension, chronic diabetes mellitus, and smoking^13–16). As such, interventions during pregnancy may be ineffective or potentially harmful.^17–19

Therefore, robust health care use, including regular screenings, conversations with providers, and teeth cleanings for everyone, regardless of their pregnancy status, may provide one avenue to improve the overall health, reduce persistent racial disparities, and achieve health equity.²⁰ For decades the United States has supported safety net health coverage plans during pregnancy and limited periods postpartum. However, outside that time frame, when there is greater potential to intervene on factors that contribute to poor health, lack of health insurance may limit care.

The Patient Protection and Affordable Care Act (ACA) was enacted on March 23, 2010, and is a complex set of policies. Within its 10 titles, the ACA included several provisions to improve health care access and use, such as the Medicaid expansions, individual and employer mandates, policy coverage requirements, insurance exchanges, insurance subsidies, delivery system reforms, and dependent coverage. Because of American Indian and Alaska Native people's political status as (often) members of sovereign nations and the ongoing Trust Responsibilities of the US Federal Government to Native nations, the ACA also included several provisions specific to tribally enrolled American Indian and Alaska Native people that may lead to differential impact and ultimately has the potential to reduce health inequities.

For example, Title 10 reauthorized the Indian Health Care Improvement Act, which aims, among other things, to reduce the incidence and prevalence of comorbidities affecting American Indian and Alaska Native people through increased federal funding, expanding availability of health care facilities, and recruitment of culturally congruent providers at the Indian Health Service (IHS). Furthermore, Title 2 set the IHS, Tribal, and Urban Indian Organizations as “express lane agencies” and payers of last resort, and furthermore put aside funding to increase tribal home visiting.²¹

Little work has monitored the overall impact, or equity in impact, of the ACA on American Indian and Alaska Native people. Furthermore, to our knowledge, none has focused on birth givers. Therefore, our objective was to examine changes in health care interactions before and after the ACA between American Indian and Alaska Native birth givers and white birth givers. To do this, we used responses regarding prepregnancy health care interactions among a sample of individuals who experienced a live birth between 2009 and 2015. We estimated changes in the prevalence of outcomes following the enactment of the ACA to determine if changes in these measures were (1) significantly different from zero and (2) equitable between American Indian and Alaska Native birth givers and white birth givers. We hypothesized that both American Indian and Alaska Native birth givers and white birth givers would demonstrate increases in health care interactions, although were uncertain if American Indian and Alaska Native birth givers would experience greater gains.

On the one hand, lower baseline prevalence before the passing of the ACA would mean that there would be more room for increases among American Indian and Alaska Native birth givers, but on the other hand, it is often privileged people who benefit from policy changes.^22,23 A secondary objective was to investigate if there were differences in trends between American Indian and Alaska Native birth givers in urban or rural locations. This objective was motivated by knowledge that severe maternal morbidity and mortality are twice as high for American Indian and Alaska Native birth givers than white birth givers (and even higher for American Indian and Alaska Native birth givers living in rural settings⁵) and that federal policies have the potential to limit or exacerbate inequities within American Indian and Alaska Native communities.

Materials and Methods

Data

We used the only multistate survey on prepregnancy health behaviors, the Centers for Disease Control and Prevention (CDC)'s Pregnancy Risk Assessment Monitoring System (PRAMS) Phases 6 (births 2009–2011) and 7 (births 2012–2015), which provides a representative sample of people experiencing live births from 2009 to 2015 in participating states (N=265,814). This study was approved by the Michigan State University's Institutional Review Board (#00002645).

Sample and comparison groups

We limited our analytic sample to adults identified as or identifying as American Indian and Alaska Native people, either as their primary or secondary race (data provided through linked birth certificates). This operationalization strives to recognize the American Indian and Alaska Native categorization as both a racial identity and political status.²⁴ As a comparison group, we also constructed a sample of non-Hispanic white people, which includes anyone whose primary race is reported as white and does not identify as a Hispanic person or an American Indian and Alaska Native person.

For the between-group analyses, white people were used as the referent because Federal Trust Responsibility, as stated in section 1601 of U.S. Code 25, declares “A major national goal of the United States is to provide the resources, processes, and structure that will enable Indian tribes and tribal members to obtain the quantity and quality of health care services and opportunities that will eradicate the health disparities between Indians and the general population of the United States.”²⁵ We recognize there are multiple ways to define health equity, although think it is important to consider the goal of Trust Responsibility as one particularly relevant metric of health equity for American Indian and Alaska Native people. For all analyses, we excluded observations with missing data on covariates. After applying the exclusion criteria, our analytic sample included 11,060 American Indian and Alaska Native birth givers and 131,119 non-Hispanic white birth givers.

There are residents of 35 states from across the United States included in the sample. While 15 states and DC do not participate in the survey during this study period, there do not appear to be any clear regional patterns in participation (Supplementary Table S1). States with a larger share of American Indian and Alaska Native people are more likely to report urban/rural classification data, meaning 85% versus 61% of the American Indian and Alaska Native sample and white sample, respectively, were not missing urban/rural classifications. We believe this difference is driven by states with a larger urban/rural variation being more likely to collect these data and American Indian and Alaska Native people being more likely to reside in these states. We used urban/rural classification at the time of birth to investigate equity within American Indian and Alaska Native communities over time.

Time allocation

We grouped the total study period into seven smaller periods (Supplementary Figure S1) relative to two key dates of the ACA (03/23/2010 when the ACA was signed; 01/01/2014 when the Medicaid expansions began) and assigned births to respective periods by estimating conception date with birth certificate provided month of birth and estimated gestational age. Estimated conception dates were used to allocate birth givers into the following time periods: 03/01/2008–03/22/2009 (pre-ACA); 03/23/2009–03/22/2010 (reference); 03/23/2010–03/22/2011 (transition); 03/23/2011–03/22/2012 (post-ACA 1); 03/23/2012–03/22/2013 (post-ACA 2); 03/23/2013–12/31/2013 (post-ACA 3); and 01/01/2014–06/30/2015 (post-ACA 4, post-Medicaid expansions) (Supplementary Table S2). While we aimed to include time periods of the same length and relative start time in each year, this was not possible given the ACA rollout dates used to divide the time periods.

Outcome measures

Prepregnancy outcome measures included one insurance measure (enrolled in Medicaid), three preventive health screening measures (diabetes, blood pressure, and anxiety/depression) that were combined to indicate a single measure of health care interaction and screening, one provider conversation measure (talk about family medical history), one dental measure (teeth cleaning), and one supplement measure (multivitamin/prenatal use). Medicaid coverage and multivitamin/prenatal use had a referent time of 1 month prepregnancy, while the other measures had a referent time of 12 months prepregnancy. Survey questions are included in Supplementary Table S3. We use the imperfect language of prepregnancy to indicate that the outcome measures refer to the period just before pregnancy, among this sample of people who experienced a live birth.²⁰

Statistical analyses

We stratified by race/ethnicity and used linear probability regression models with time period indicators to estimate period-specific changes in the prevalence of each outcome, relative to the time just before the ACA implementation (reference). Beta coefficients were multiplied by 100 and interpreted as percentage point changes. Adjusted models included respondents' years of education, age, previous live birth, marital status, and household income as a percentage of the federal poverty level as covariates. Output from these models was used to create figures that convey (1) the baseline prevalence of each outcome by race/ethnicity and (2) the change in prevalence for each outcome over time, by race/ethnicity. To determine if there were differences in trends by urban or rural setting among American Indian and Alaska Native birth givers, we also stratified the American Indian and Alaska Native sample by residence. All statistical analyses used the PRAMS-provided survey weights to account for nonresponse, noncoverage, and stratification by state.

Robustness checks

First, because states participating in PRAMS change over time, we conducted a robustness check to determine the impact on estimates of restricting to a balanced panel of participating states. Second, to determine potential impact on estimates of seasonality, and thus the composition of birth givers in each time period, we adjusted for month of conception and reran the main analyses.

Results

Compared with the white analytic sample, the American Indian and Alaska Native sample was younger (72% vs. 55% younger than the age of 29), had fewer years of education (53% vs. 29% with high school diploma or less), was more likely to be unmarried (58% vs. 28%), was more likely to have a household income less than 200% of the federal poverty level (79% vs. 47%), and was more likely to report rural (vs. urban) residence (60% vs. 40%) (Table 1).

Table 1.

Prevalence (%) of Maternal Characteristics and Outcomes by Racial/Ethnic Identity and Conception Date, Pregnancy Risk Assessment Monitoring System, 2008–2015 (n=142,179)

	Conception dates and time period
			Pre-ACA	Reference	Transition	Post-ACA 1	Post-ACA 2	Post-ACA 3	Post-ACA 4
	Racial/ethnic group	Overall	03/01/08–03/22/09	03/23/09–03/22/10	03/23/10–03/22/11	03/23/11–03/22/12	03/23/12–03/22/13	03/23/13–12/31/13	01/01/14–06/30/15
Sample size	nH AIAN	11,060	1630	1566	1594	1322	1939	1057	1952
	nH W	13,1119	19,901	20,128	19,207	16,269	18,978	12,787	23,849
	Total	14,2179	21,531	21,694	20,801	17,591	20,917	13,844	25,801
Maternal characteristics
Age
<20	nH AIAN	11.6	14.7	12.5	14.6	13.0	9.1	8.2	9.5
<20	nH W	4.8	6.9	5.7	5.1	4.4	4.0	3.4	3.6
20–29	nH AIAN	60.4	61.8	59.6	60.4	62.2	64.8	58.3	56.6
20–29	nH W	49.8	52.9	50.7	49.9	49.5	49.1	48.4	48.2
30–39	nH AIAN	26.6	22.4	26.7	23.8	23.5	24.6	31.4	32.6
30–39	nH W	42.6	37.4	40.9	42.1	43.0	44.1	45.4	45.6
≥40	nH AIAN	1.4	1.1	1.1	1.2	1.4	1.5	2.2	1.4
≥40	nH W	2.8	2.8	2.8	3.0	3.1	2.8	2.7	2.6
Years of education
0–11 Years	nH AIAN	18.5	23.5	19.1	21.0	17.3	17.9	15.8	15.7
0–11 Years	nH W	7.5	9.4	8.9	7.5	7.2	6.9	6.6	6.0
12 Years	nH AIAN	34.5	36.2	35.7	32.3	32.5	35.5	31.8	36.3
12 Years	nH W	21.4	24.6	22.3	20.9	20.6	20.2	20.1	20.3
13–15 Years	nH AIAN	33.6	30.2	31.5	34.8	36.9	32.7	42.3	30.9
13–15 Years	nH W	29.0	27.7	28.7	29.3	29.4	29.2	29.3	29.6
≥16 Years	nH AIAN	13.4	10.1	13.8	11.9	13.3	13.9	10.1	17.2
≥16 Years	nH W	42.1	28.3	40.1	42.3	42.8	43.7	44.0	44.2
Marital status
Unmarried	nH AIAN	58.0	59.3	58.1	60.7	61.4	58.8	59.7	51.8
Unmarried	nH W	27.8	28.5	27.9	27.0	27.2	27.6	28.6	27.7
Married	nH AIAN	42.0	40.7	41.9	39.3	38.6	41.2	40.3	48.2
Married	nH W	72.2	71.5	72.1	73.0	72.8	72.4	71.4	72.3
Previous live birth
Yes	nH AIAN	39.0	35.7	40.7	41.5	42.5	39.0	34.6	37.4
Yes	nH W	41.8	42.8	42.4	43.2	41.9	41.8	40.6	40.5
No	nH AIAN	61.0	64.3	59.3	58.5	57.5	61.0	65.4	62.6
No	nH W	58.2	57.2	57.6	56.8	58.2	58.3	59.4	59.5
Household income (as % of FPL)
<100%	nH AIAN	57.2	62.1	58.4	62.7	59.2	53.8	55.8	51.4
<100%	nH W	26.2	27.6	27.1	26.4	25.9	25.6	24.1	25.8
100–199%	nH AIAN	22.0	19.8	21.2	20.8	19.9	25.7	24.9	22.4
100–199%	nH W	20.3	21.7	20.4	20.0	21.1	19.7	19.7	19.4
200–399%	nH AIAN	12.4	6.8	9.2	5.7	15.9	14.6	15.1	16.9
200–399%	nH W	19.9	15.8	16.8	16.6	23.3	23.5	23.7	21.0
≥400%	nH AIAN	8.4	11.3	11.2	10.8	5.0	5.9	4.1	9.4
≥400%	nH W	33.7	34.8	35.7	37.1	29.8	31.2	32.6	33.9
Urban/rural residence
Rural	nH AIAN	45.9	50.2	38.0	45.1	48.0	47.6	49.7	45.0
Rural	nH W	20.4	17.8	16.8	21.2	25.1	19.8	22.1	21.3
Urban	nH AIAN	38.7	32.9	41.1	40.0	37.7	33.1	36.3	45.4
Urban	nH W	40.5	40.9	42.8	34.5	43.3	35.7	38.2	44.8
Missing	nH AIAN	15.4	16.9	20.9	14.9	14.3	19.3	14.0	9.6
Missing	nH W	39.1	41.3	40.5	44.4	31.7	44.5	39.6	33.8
Outcomes
Medicaid covered month before pregnancy	nH AIAN	32.0	29.1	27.6	32.7	32.0	33.4	33.8	34.5
Medicaid covered month before pregnancy	nH W	14.0	11.2	12.2	14.1	14.5	13.8	15.3	16.6
Prepregnancy multivitamin/prenatal use	nH AIAN	26.0	23.6	23.9	27.6	26.9	26.2	26.5	26.7
Prepregnancy multivitamin/prenatal use	nH W	45.2	41.0	44.0	45.5	46.9	46.6	46.0	46.6
Prepregnancy talk family history	nH AIAN	32.2	24.9	27.2	26.7	37.4	33.6	36.0	37.1
Prepregnancy talk family history	nH W	31.1	24.9	26.3	28.1	34.0	34.5	34.8	35.5
Prepregnancy teeth cleaned	nH AIAN	48.9	42.4	44.1	46.1	51.0	52.4	51.1	52.7
Prepregnancy teeth cleaned	nH W	62.2	58.8	60.1	62.1	63.8	64.7	64.2	62.8
Prepregnancy health care interaction and screening (MH, BP, Diabetes)	nH AIAN	35.9	22.7	27.6	30.5	41.4	44.2	42.3	40.1
	nH W	26.9	19.3	18.2	19.7	32.7	33.0	32.6	33.5
Prepregnancy check diabetes^a	nH AIAN	19.3	11.3	10.4	15.5	24.0	25.7	23.6	22.8
Prepregnancy check diabetes^a	nH W	10.8	5.4	5.1	5.9	14.7	14.9	14.7	15.3
Prepregnancy check high blood pressure^a	nH AIAN	22.1	9.9	8.7	13.4	28.7	30.3	31.7	29.3
Prepregnancy check high blood pressure^a	nH W	15.9	6.5	5.9	6.3	23.2	23.8	23.1	23.5
Prepregnancy check depression/anxiety^a	nH AIAN	21.2	14.4	17.5	17.5	21.5	26.3	23.8	25.2
Prepregnancy check depression/anxiety^a	nH W	16.6	13.4	12.9	14.2	18.6	18.3	18.7	20.0

Measures combined to create one indicator of health care interaction and screening.

ACA, Affordable Care Act; AIAN, American Indian and Alaska Native; BP, blood pressure; FPL, federal poverty level; MH, mental health, nH, non-Hispanic; W, white.

Medicaid coverage

Before implementation of the ACA, 28% of American Indian and Alaska Native birth givers had Medicaid coverage in the month before pregnancy (Fig. 1A; Table 1). Medicaid coverage increased steadily after implementation, but the largest time-specific increase (10.1 percentage points [ppts], 95% confidence interval [CI]: 3.1–17.2), compared with the pre-ACA (reference) estimate, occurred after some states expanded Medicaid (2014–2015) (Supplementary Table S4). There were slightly smaller increases in Medicaid coverage among white birth givers following the implementation of the ACA, although Medicaid coverage in the month before pregnancy was lower than among American Indian and Alaska Native birth givers, at 12% (Table 1). The greatest increase in coverage among white birth givers was observed after the Medicaid expansions in the ACA-Post 4 time period (5.4 ppts, 95% CI: 4.5–6.3) (Supplementary Table S5).

FIG. 1.

Figures include prevalence estimates and 95% CIs (vertical lines) for study outcomes among American Indian and Alaska Native birth givers (triangles) and non-Hispanic white birth givers (circles) by conception date (March 2008 to July 2015). The five study outcomes include (A) Medicaid coverage the month before pregnancy, (B) prepregnancy multivitamin or prenatal use, (C) prepregnancy health care interaction and disease screenings, (D) prepregnancy talk about family history with health care worker, and (E) prepregnancy teeth cleaning. The dashed vertical lines mark the start date of the ACA. Linear probability regression models estimated time period-specific percentage point changes in prevalence, relative to the time period just before ACA implementation (the estimate with no 95% CI, the referent period) and adjusted for respondents' years of education, age, previous live birth, marital status, and household income as a percentage of the federal poverty level. The survey weighted mean for the March 2009–2010 conception period (referent period) was used to plot the baseline outcome prevalence. Then, each year-specific outcome prevalence was calculated by adding or subtracting from the referent mean, each year-specific beta estimate (multiplied by 100) provided by each linear probability regression model. Filled/shaded triangles and squares indicate estimates for which the 95% CIs do not contain zero, while unshaded markers indicate those where there is no significant change in estimate. In some cases, the 95% CIs are narrow and therefore obscured from view by the circular estimate markers. ACA, Affordable Care Act; CI, confidence interval.

Multivitamin or prenatal use

Just before implementation of the ACA, multivitamin/prenatal use in the month before pregnancy was lower among American Indian and Alaska Native birth givers than among white birth givers (24% vs. 44%) (Fig. 1B; Table 1). While there were increases of 1–5 ppts in multivitamin/prenatal use among American Indian and Alaska Native birth givers in the post-ACA time periods, the 95% CIs of these increases contained the null value of 0 (Supplementary Table S4). Among white birth givers, similarly small increases of 1–3 ppts were observed (Supplementary Table S5), although in some post-ACA time periods, increases were significantly different than the referent period, presumably because of the larger sample size.

Health care interaction and disease screening

The referent period prevalence of the measure for health care interaction, including screenings for diabetes, blood pressure, or depression/anxiety, was 28% among American Indian and Alaska Native birth givers and 18% among white birth givers (Fig. 1C; Table 1). The percentage of people reporting a health care interaction was 12–18 ppts higher in each post-ACA period relative to the referent period for both groups starting in March 2011 (Supplementary Tables S4 and S5).

Talk about family history

A similar percent of American Indian and Alaska Native birth givers and white birth givers reported that they spoke with a health care worker about their family medical history in the year before their pregnancy (27% vs. 26%) during the referent period (Fig. 1D; Table 1). Time period-specific increases in this measure were of similar magnitude among both groups (Supplementary Tables S4 and S5).

Teeth cleaning

Just before implementation of the ACA, 44% of American Indian and Alaska Native birth givers reported a prepregnancy teeth cleaning, compared with 60% of white birth givers (Fig. 1E; Table 1). Teeth cleaning prevalence increased for both groups following the passing of the ACA, although the increases were sustained at 2–4 ppts and were significantly different than the referent period only among white birth givers (Supplementary Table S5). While there were increases in teeth cleaning among American Indian and Alaska Native birth givers following the ACA, only one period (Post-ACA 2: 03/2012–2013) differed significantly from the referent (8.6 ppts, 95% CI: 0.2–17.0) (Supplementary Table S4).

Urban/rural modification among American Indian and Alaska Native birth givers

Referent period outcome prevalences were similar between rural and urban American Indian and Alaska Native birth givers, except for discussing family history and multivitamin/prenatal use, for which rural prevalence was 9–15 ppts lower, respectively, than urban prevalence (Table 2). While there were increases in all outcome measures after the implementation of the ACA, the rural-specific estimates were often larger in magnitude and more often statistically significant than the urban estimates (Table 2; Supplementary Table S6). For example, rural-specific estimates of discussing family history increased by 12.7 ppts (95% CI: 4.1–21.4), 13.7 ppts (95% CI: 4.9–22.5), 6.8 ppts (95% CI: −2.6 to 16.2), and 11.7 ppts (95% CI: 2.9–20.4), respectively, in each period after the passing of the ACA relative to the referent period. Urban-specific increases over the same periods were 10.1, 0.3, 7.1, and 6.6 ppts.

Table 2.

Beta Coefficients and 95% Confidence Intervals from Adjusted Linear Probability Models Estimating Change in Outcome by Conception Date Among American Indian and Alaska Native Birth Givers, by Urban/Rural Residential Location, Pregnancy Risk Assessment Monitoring System, 2008–2015 (n=1259)

	Medicaid coverage		Health care interaction and screening		Discussed family history		Multivitamin/Prenatal use		Teeth cleaning
	Rural	Urban	Rural	Urban	Rural	Urban	Rural	Urban	Rural	Urban
Ref period prevalence (%)	26.6	23.7	24.1	27.1	21.1	30.1	16.0	30.6	43.3	44.1
	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)	Beta (95% CI)
Conception date
March 1, 2008, to March 22, 2009 (pre-ACA)	0.010 (−0.075 to 0.095)	−0.040 (−0.172 to 0.091)	0.010 (−0.077 to 0.097)	−0.056 (−0.150 to 0.038)	0.047 (−0.041 to 0.135)	−0.067 (−0.171 to 0.036)	0.044 (−0.043 to 0.131)	−0.032 (−0.149 to 0.085)	0.019 (−0.080 to 0.118)	0.007 (−0.110 to 0.124)
March 23, 2009, to March 22, 2010 (ref)	—	—	—	—	—	—	—	—	—	—
March 23, 2010, to March 22, 2011 (transition)	0.031 (−0.056 to 0.118)	0.045 (−0.077 to 0.166)	0.065 (−0.033 to 0.163)	0.043 (−0.058 to 0.144)	0.028 (−0.059 to 0.116)	0.012 (−0.100 to 0.124)	0.085 (0.011 to 0.160)	0.007 (−0.125 to 0.139)	0.010 (−0.090 to 0.109)	0.054 (−0.067 to 0.174)
March 23, 2011, to March 22, 2012 (post-ACA 1)	0.067 (−0.023 to 0.157)	0.068 (−0.053 to 0.190)	0.162 (0.070 to 0.254)	0.144 (0.028 to 0.259)	0.127 (0.041 to 0.214)	0.101 (−0.021 to 0.223)	0.096 (0.018 to 0.173)	0.023 (−0.122 to 0.168)	0.072 (−0.029 to 0.172)	0.051 (−0.086 to 0.187)
March 23, 2012, to March 22, 2013 (post-ACA 2)	0.081 (−0.003 to 0.165)	0.110 (−0.008 to 0.229)	0.228 (0.135 to 0.320)	0.072 (−0.025 to 0.169)	0.137 (0.049 to 0.225)	0.003 (−0.106 to 0.111)	0.109 (0.031 to 0.188)	0.009 (−0.116 to 0.134)	0.059 (−0.039 to 0.157)	0.121 (−0.004 to 0.246)
March 23, 2013, to December 31, 2013 (post-ACA 3)	0.100 (0.002 to 0.198)	0.077 (−0.054 to 0.207)	0.138 (0.037 to 0.238)	0.117 (−0.015 to 0.249)	0.068 (−0.026 to 0.162)	0.071 (−0.048 to 0.189)	0.057 (−0.025 to 0.139)	0.004 (−0.130 to 0.139)	0.079 (−0.027 to 0.185)	0.093 (−0.037 to 0.223)
January 1, 2014, to June 30, 2015 (post-ACA 4)	0.121 (0.040 to 0.203)	0.106 (−0.009 to 0.222)	0.144 (0.057 to 0.230)	0.161 (0.061 to 0.262)	0.117 (0.029 to 0.204)	0.066 (−0.043 to 0.175)	0.091 (0.013 to 0.168)	−0.071 (−0.194 to 0.051)	0.098 (0.003 to 0.192)	0.071 (−0.044 to 0.187)

Bolded values indicate p<0.05.

Linear probability regression models estimated time period-specific percentage point changes in prevalence, relative to the period just before ACA implementation. Beta estimates can be multiplied by 100 to be interpreted as percentage point changes relative to the referent period prevalence. Estimates adjust for respondents' years of education, age, previous live birth, marital status, and household income as a percentage of the federal poverty level.

ACA, Patient Protection & Affordable Care Act; CI, confidence interval; Ref, Reference.

Similar patterns among rural and urban American Indian and Alaska Native people were observed for Medicaid coverage, disease screenings, and multivitamin/prenatal use. Trends for teeth cleaning were not as clear.

Robustness checks

With use of a balanced panel among American Indian and Alaska Native birth givers (11 states included), compared with the main analysis, post-ACA estimates for (1) Medicaid coverage, disease screenings, and discussed family history were attenuated; (2) multivitamin/prenatal use strengthened but remained nonsignificant; and (3) teeth cleaning strengthened (Supplementary Table S7). Among the white sample's balanced panel (16 states included), impacts of the balanced panel were nearly identical to those described for the American Indian and Alaska Native sample's balanced panel, apart from teeth cleaning where the post-ACA estimates were attenuated among white birth givers in the balanced panel (Supplementary Table S8). Including additional adjustment for the calendar month of conception resulted in identical results as those reported in the main analyses, apart from slightly wider CIs (Supplementary Table S9).

Discussion

Despite the several provisions included in the ACA that focus on American Indian and Alaska Native people, few studies emphasize the ACA's potential impacts among American Indian and Alaska Native birth givers, a group that experiences the burden of adverse pregnancy outcomes. We first sought to estimate changes in five prepregnancy outcomes after, compared with before, the ACA among American Indian and Alaska Native birth givers and white birth givers. While both groups experienced increases in outcomes over time, American Indian and Alaska Native birth giver-specific increases tended to be larger for the outcomes of Medicaid coverage, multivitamin/prenatal use, and teeth cleaning. White birth giver-specific increases tended to be larger for the screening measure. Increases were similar for talking with a health care provider about family history.

While there are no studies looking at ways the ACA might impact American Indian and Alaska Native birth givers, some research exists regarding insurance expansions and their association with prepregnancy health outcomes among other populations. For example, the Ohio Medicaid expansion was associated with increases in prepregnancy insurance enrollment.²⁶ Frerichs et al. document increases in public and private insurance among American Indian and Alaska Native people following the ACA.²⁷ The Medicaid expansions were also associated with increases in insurance coverage among American Indian and Alaska Native women of reproductive age.^28,29 Our results are also consistent with recent studies focused on the ACA's impact on the prepregnancy period. The Medicaid expansions, for example, are associated with increases in insurance coverage and health care use among reproductive-aged women, as well as prepregnancy folic acid intake and health counseling.^30,31

Our second objective was to explore equity in outcome changes between American Indian and Alaska Native birth givers and white birth givers. To answer this question, we considered both changes in estimate following the ACA and the baseline prevalence of each outcome in the referent period. For example, focusing only on the change estimates for teeth cleaning, we might conclude that American Indian and Alaska Native birth givers experienced slightly larger increases than white birth givers (3–9 ppts vs. 1–4 ppts). However, before ACA implementation, teeth cleaning prevalence was 44% for American Indian and Alaska Native birth givers (vs. 60% for white birth givers). While American Indian and Alaska Native birth givers experienced larger increases in teeth cleaning, it was not enough to offset the initial health inequity. So, while the ACA is associated with increases in health care use for both groups, the American Indian and Alaska Native white inequity did not close, particularly for multivitamin/prenatal use or teeth cleaning, two outcomes with clear pre-ACA inequities.

For later time periods, change estimates for disease screenings are compatible for American Indian and Alaska Native birth givers and white birth givers, suggesting that the ACA may be associated with health equity for this outcome. We estimated conversations about family health to be similar before and after the ACA for both samples.

The ACA alone may not be enough to overcome legacies and contemporary realities of colonization experienced by American Indian and Alaska Native people. First, health care systems delivered through the IHS and Tribal nations face woefully insufficient federal funding which impacts staffing and capacity, including for informing American Indian and Alaska Native people about the various forms of health insurance coverage available and how to enroll.³² Even if insurance coverage improves, infrastructure may not exist to support increased use. Further, communication challenges, lack of continuity in care, and transportation barriers could remain.^33–34 Also, American Indian and Alaska Native birth givers use of healthcare services, and experiences therein, cannot be separated from contexts that include histories of forced sterilization and child removal, significant intimate partner violence, substance use, and in some cases, substantial life stressors, as well as experiences that have led to distrust of medical practitioners. Indeed, American Indian and Alaska Native people report more experiences of discrimination in healthcare, compared to their peers.^36,37

Mortality and severe maternal morbidity are twice as high for American Indian and Alaska Native birth givers than white birth givers, but even higher for American Indian and Alaska Native people living in rural settings,⁵ where uninsurance tends to be higher.³⁸ Therefore, it is especially important to pay attention to potential differential impact among American Indian and Alaska Native birth givers that might exacerbate existing inequities. The ACA was often associated with larger increases in rural spaces (vs. urban), where the American Indian and Alaska Native sample size was larger in these data. While there were increases in prepregnancy health care engagement following the implementation of the ACA, the differences in context experienced by rural and urban American Indian and Alaska Native birth givers may necessitate additional action to reimagine health care delivery in a way that addresses health disparities within American Indian and Alaska Native communities and also between American Indian and Alaska Native people and other racial/ethnic groups.³⁹

In this article, we explore a topic that is not well understood using data from the largest data set regarding perinatal health. Whereas most studies aggregate multiracial people or drop them from analyses, we included multiracial American Indian and Alaska Native people in the American Indian and Alaska Native sample. This approach allows us to better capture the diversity of experiences of American Indian and Alaska Native birth givers and prioritizes American Indian and Alaska Native identity, recognizing its several conceptualizations: a racialized identity and a political status. Our analytic approach was agnostic to any time lag that might occur following the implementation of ACA provisions. However, the increases we observed between March 2010 to March 2011 and March 2011 to March 2012 for some outcomes suggest there might be a time lag between the implementation of the ACA and when changes are observable. Future work might focus on specific mechanisms or provisions of the ACA that are associated with observed increases.

This work includes several limitations. First, limitations inherent to our statistical approach mean we cannot rule out causes, outside the ACA, for the observed trends.

Second, there were minor changes in wording, and therefore implication, of some screening questions across survey phases (Supplementary Table S3), although we suspect that most of the observed prevalence increase is unrelated to question wording because (1) we inserted an indicator for phase in the models and observed the same trends (data not shown) and (2) we observed time trends in outcomes without question wording changes. While there were updates to professional organization recommendations during the study period,⁴⁰ there was none in 2010 to coincide with our findings, although the 2008 blood pressure screening recommendation update could have taken time to be implemented. Relatedly, we did not include PRAMS Phase 8 data in this analysis because of large question wording changes that ultimately changed the meaning of questions for some outcomes.

Third, we were unable to account for heterogeneity among American Indian and Alaska Native birth givers, which may be masking important geographic or tribe-specific inequities between American Indian and Alaska Native communities.⁴¹ Moreover, the urban/rural classification we used is likely an oversimplification of a nuanced phenomenon. Relatedly, we do not know the tribal enrollment status of included American Indian and Alaska Native people, so it is possible that our intention-to-treat analysis is underestimating associations. Fourth, we were unable to estimate changes in the quality of care that people accessed, just whether a service was used. And lastly, we were limited to standardized measures that reflect the priorities of the CDC, rather than measures deemed important by American Indian and Alaska Native birth givers.⁴²

Conclusion

In this article, we begin to investigate equity in the ACA's impact across the Indian Country. This work is important because conversations to repeal the ACA are ongoing and often do not include the specific needs or contexts of American Indian and Alaska Native people, including the acknowledgment of the federal government's Trust Responsibility. We estimate that the ACA is associated with changes (for the better) in health care interactions and use among people who soon after experience a live birth. While the ACA is associated with increases in prepregnancy health care for American Indian and Alaska Native birth givers and white birth givers, and therefore potentially reductions in maternal and child health inequities, the ACA is not associated with elimination of inequities between American Indian and Alaska Native and white people. However, the ACA may have narrowed the gap between American Indian and Alaska Native birth givers living in different contexts.

Footnotes

Acknowledgments

We thank Florence Kizza, Nupur Huria, and Harsna Chahal for formatting tables for publication as well as the PRAMS Working Group and the Centers for Disease Control and Prevention (CDC) for the stewardship of these data. We also thank members of our advisory board, and audiences at ASHEcon and Academy Health, for feedback on an earlier version of this article.

Authors' Contributions

D.R.G.: Conceptualization, methodology, writing—original draft, writing— review and editing, visualization, and funding acquisition. K.H.: Conceptualization, methodology, formal analysis, writing— original draft, writing—review and editing, and visualization. H.H.-B.: Conceptualization, writing—original draft, and writing—review and editing. C.M.: Conceptualization, methodology, writing—review and editing, and funding acquisition.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This project was funded by the Robert Wood Johnson Foundation (78812) and the National Institute of Child Health & Human Development (R01HD095951). The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders. Funders played no role in the study design; collection, analysis, and interpretation of data; writing the article; or in the decision to submit the article for publication. No financial disclosures were reported by the authors of this article.

Supplementary Material

Abbreviations Used

References

Danielson

, Wallenborn

, Warne

, et al. Disparities in risk factors and birth outcomes among American Indians in North Dakota. Matern Child Health J, 2018; 22(10):1519–1525; doi: 10.1007/s10995-018-2551-9

Raglan

, Lannon

, Jones

, et al. Racial and ethnic disparities in preterm birth among American Indian and Alaska Native Women. Matern Child Health J, 2016; 20(1):16–24; doi: 10.1007/s10995-015-1803-1

Hwang

, Shrestha

, Yazzie

, et al. Preterm birth among American Indian/Alaskan Natives in Washington and Montana: Comparison with non-Hispanic Whites. Matern Child Health J, 2013; 17(10):1908–1912; doi: 10.1007/s10995-012-1215-4

Heck

, Jones

, Bohn

, et al. Maternal mortality among American Indian/Alaska Native women: A scoping review. J Womens Health, 2021; 30(2):220–229; doi: 10.1089/jwh.2020.8890

Kozhimannil

, Interrante

, Tofte

, et al. Severe maternal morbidity and mortality among indigenous women in the United States. Obstet Gynecol, 2020; 135(2):294–300; doi: 10.1097/AOG.0000000000003647

Wingate

, Barfield

, Smith

, et al. Perinatal disparities between American Indians and Alaska Natives and other US Populations: Comparative changes in fetal and first day mortality, 1995–2008. Matern Child Health J, 2015; 19(8):1802–1812; doi: 10.1007/s10995-015-1694-1

Creanga

, Bateman

, Kuklina

, et al. Racial and ethnic disparities in severe maternal morbidity: A multistate analysis, 2008–2010. Am J Obstet Gynecol, 2014; 210(5):435.e1–435.e8; doi: 10.1016/j.ajog.2013.11.039

Burns

, DeAtley

, Short

. The maternal health of American Indian and Alaska Native people: A scoping review. Soc Sci Med, 2023; 317:115584; doi: 10.1016/j.socscimed.2022.115584

Hayward

, Cidro

. Indigenous birth as ceremony and a human right. Health Hum Rights, 2021; 23(1):213–224.

10.

Thurston

, Fields

, White

. Does increasing access to prenatal care reduce racial disparities in birth outcomes?. J Pediatr Nurs, 2021; 59:96–102; doi: 10.1016/j.pedn.2021.01.012

11.

Fiscella

. Does prenatal care improve birth outcomes? A critical review. Obstet Gynecol, 1995; 85(3):468–479; doi: 10.1016/0029-7844(94)00408-6

12.

Alexander

, Kotelchuck

. Assessing the role and effectiveness of prenatal care: History, challenges, and directions for future research. Public Health Rep Wash DC 1974, 2001; 116(4):306–316; doi: 10.1093/phr/116.4.306

13.

Muchira

, Heelan-Fancher

, Hayman

. Maternal cardiovascular health and adverse childbirth outcomes in the United States: Analysis of the Pregnancy Risk Assessment Monitoring System. J Cardiovasc Nurs, 2021; 36(4):E51–E59; doi: 10.1097/JCN.0000000000000813

14.

Anderson

, Spicer

, Peercy

. Obesity, diabetes, and birth outcomes among American Indians and Alaska Natives. Matern Child Health J, 2016; 20(12):2548–2556; doi: 10.1007/s10995-016-2080-3

15.

Dorfman

, Srinath

, Rockhill

, et al. The association between diabetes mellitus among American Indian/Alaska Native populations with preterm birth in eight US States from 2004–2011. Matern Child Health J, 2015; 19(11):2419–2428; doi: 10.1007/s10995-015-1761-7

16.

Tiwari

, Enquobahrie

, Wander

, et al. A retrospective cohort study of race/ethnicity, pre-pregnancy weight, and pregnancy complications. J Matern Fetal Neonatal Med, 2022; 35(25):6388–6395; doi: 10.1080/14767058.2021.1914573

17.

Wisborg

, Henriksen

, Jespersen

, et al. Nicotine patches for pregnant smokers: A randomized controlled study. Obstet Gynecol, 2000; 96(6):967–971; doi: 10.1016/s0029-7844(00)01071-1

18.

Gardiner

, Nelson

, Shellhaas

, et al. The clinical content of preconception care: Nutrition and dietary supplements. Am J Obstet Gynecol, 2008; 199(6, Supplement B):S345–S356; doi: 10.1016/j.ajog.2008.10.049

19.

Floyd

, Jack

, Cefalo

, et al. The clinical content of preconception care: Alcohol, tobacco, and illicit drug exposures. Am J Obstet Gynecol, 2008; 199(6 Suppl 2):S333–S339; doi: 10.1016/j.ajog.2008.09.018

20.

Dehlendorf

, Akers

, Borrero

, et al. Evolving the preconception health framework: A call for reproductive and sexual health equity. Obstet Gynecol, 2021; 137(2):234; doi: 10.1097/AOG.0000000000004255

21.

Warne

, Delrow

, Angus-Hornbuckle

, et al. Impact of ACA Repeal on American Indians and Alaska Natives. Issue Brief. National Indian Health Board: Washington, DC; 2017.

22.

Han

, Zhu

, Jemal

. Characteristics of young adults enrolled through the Affordable Care Act–dependent coverage expansion. J Adolesc Health, 2016; 59(6):648–653; doi: 10.1016/j.jadohealth.2016.07.027

23.

Phelan

, Link

, Tehranifar

. Social conditions as fundamental causes of health inequalities: Theory, evidence, and policy implications. J Health Soc Behav, 2010; 51 Suppl:S28–S40; doi: 10.1177/0022146510383498

24.

Gartner

, Wilbur

, McCoy

. “American Indian” as a racial category in public health: Implications for communities and practice. Am J Public Health, 2021; 111(11):1969–1975; doi: 10.2105/AJPH.2021.306465

25.

Indian Health Care Improvement Act, 15 U.S.C. §§ 1601.

26.

Adams

, Dunlop

, Strahan

, et al. Prepregnancy insurance and timely prenatal care for medicaid births: Before and after the affordable care act in Ohio. J Womens Health, 2019; 28(5):654–664; doi: 10.1089/jwh.2017.6871

27.

Frerichs

, Bell

, Lich

, et al. Health insurance coverage among American Indians and Alaska Natives in the context of the Affordable Care Act. Ethnicity Health, 2022; 27(1):174–189; doi: 10.1080/13557858.2019.1625873

28.

Strully

, Chatterji

, Liu

, et al. Effects of medicaid expansions on coverage, prenatal care, and health among American Indian/Alaska Native Women. Health Aff (Millwood), 2024; 43(3):344–353; doi: 10.1377/hlthaff.2023.00672

29.

Gartner

, Islam

, Margerison

. Medicaid expansions and differences in guideline-adherent cervical cancer screening between American Indian and White women. Cancer Med, 2023; 12(7):8700–8709; doi: 10.1002/cam4.5593

30.

Myerson

, Crawford

, Wherry

. Medicaid expansion increased preconception health counseling, folic acid intake, and postpartum contraception: Study examines the impact of ACA Medicaid expansion on health behaviors including birth control use and pregnancy intention, and receipt of preconception health services. Health Aff (Millwood), 2020; 39(11):1883–1890; doi: 10.1377/hlthaff.2020.00106

31.

Margerison

, MacCallum

, Chen

, et al. Impacts of Medicaid expansion on health among women of reproductive age. Am J Prev Med, 2020; 58(1):1–11; doi: 10.1016/j.amepre.2019.08.019

32.

Liddell

. “Treat Me like Your Family”: Positive factors that influence patient-provider relationships for Native American women. Soc Work Public Health, 2023; 38(3):221–234; doi: 10.1080/19371918.2022.2127434

33.

Cromer

, Wofford

, Wyant

. Barriers to healthcare access facing American Indian and Alaska Natives in Rural America. J Community Health Nurs, 2019; 36(4):165–187; doi: 10.1080/07370016.2019.1665320

34.

Marrone

. Understanding barriers to health care: A review of disparities in health care services among indigenous populations. Int J Circumpolar Health, 2007; 66(3):188–198; doi: 10.3402/ijch.v66i3.18254

35.

Hanson

. Understanding prenatal health care for American Indian women in a Northern Plains tribe. J Transcult Nurs, 2012; 23(1):29–37; doi: 10.1177/1043659611423826

36.

Johansson

, Jacobsen

, Buchwald

. Perceived discrimination in health care among American Indians/Alaska natives. Ethn Dis, 2006; 16(4):766–771.

37.

Vedam

, Stoll

, Taiwo

, et al. The Giving Voice to Mothers study: Inequity and mistreatment during pregnancy and childbirth in the United States. Reprod Health, 2019; 16(1):77; doi: 10.1186/s12978-019-0729-2

38.

Admon

, Daw

, Interrante

, et al. Rural and urban differences in insurance coverage at prepregnancy, birth, and postpartum. Obstet Gynecol, 2023; 141(3):570; doi: 10.1097/AOG.0000000000005081

39.

Thorsen

, Harris

, McGarvey

, et al. Evaluating disparities in access to obstetric services for American Indian women across Montana. J Rural Health, 2022; 38(1):151–160; doi: 10.1111/jrh.12572

40.

Hunt

, Arndt

, Bell

, et al. Are corporations re-defining illness and health? The diabetes epidemic, goal numbers, and blockbuster drugs. J Bioethical Inq, 2021; 18(3):477–497; doi: 10.1007/s11673-021-10119-x

41.

Johnson

, Call

, Blewett

. The importance of geographic data aggregation in assessing disparities in American Indian prenatal care. Am J Public Health, 2010; 100(1):122–128; doi: 10.2105/AJPH.2008.148908

42.

Rainie

, Schultz

, Briggs

, et al. Data as a strategic resource: Self-determination, governance, and the data challenge for Indigenous Nations in the United States. Int Indig Policy J, 2017; 8(2);1–9; doi: 10.18584/iipj.2017.8.2.1

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

0.03 MB

0.01 MB

0.02 MB

0.05 MB

0.02 MB