Women’s Health and the Affordable Care Act: High Hopes Versus Harsh Realities?

Abstract

Our population-based survey of 1078 randomly sampled US women, aged 18 to 55 years, sought to characterize their understanding of and attitudes toward the Affordable Care Act (ACA). Most women, especially socially disadvantaged groups, had negative or uncertain attitudes toward the ACA and limited understanding of its health benefits, including its relevance for their own health service coverage and utilization. Our findings are important for continued research, policy, and practice, with implications for whether, when, and how improved coverage will translate to improved access and outcomes for US women.

Women’s health clinicians, researchers, and policymakers are hopeful that expanding health care coverage under the Patient Protection and Affordable Care Act (ACA)¹ will improve the health of US women. By requiring coverage, increasing access to affordable health plans, incentivizing utilization of high-value services, establishing benefit mandates, and reducing cost sharing, the ACA is expected to improve health outcomes and reduce health disparities for women. Since ACA implementation began, however, it has become clear that the public’s participation in its programs and benefits is compromised by widespread confusion.^2–6 Recognizing that the ACA can only have an impact on women’s health (individual and population) if women are aware of available benefits and act upon them,^7–9 we conducted a study to examine women’s understanding of and attitudes toward the ACA. Specifically, we sought to determine (1) whether women were aware and approved of the ACA and the women’s health benefits attributable to it, (2) whether women expected their coverage of women’s health services and subsequent service utilization to change as a result of the ACA, and (3) whether women’s awareness and attitudes differed across sociodemographic groups.

METHODS

In September 2013, we conducted a cross-sectional, population-based, Internet survey of US women aged 18 to 55 years, called The Women’s Health Care Experiences and Preferences Study. The survey was fielded by GfK, formerly Knowledge Networks, using their existing nationally representative, random probability Internet panel of 50 000 residents from across all US states, with oversampling of African American and Hispanic households (the GfK panel sample and design information are available at http://www.gfk.com/us/Solutions/consumer-panels/Pages/GfK-KnowledgePanel.aspx). Of the GfK panelists who met our inclusion criteria (English-speaking women aged 18–55 years) who were randomly sampled and emailed an invitation to participate (n = 2520), 1078 completed the survey (response rate = 43%). Sampling weights were applied to adjust for the complex, stratified sampling design of the survey and to reach national demographic characteristic benchmarks for the sample.

The 29-item survey measured women’s health care experiences, preferences, intentions, and behaviors. A series of items assessed ACA awareness and attitudes, including

• 1. whether women had ever heard of the ACA (sometimes referred to as “Obamacare”),

• 2. whether they believed their insurance coverage would change as a result of the ACA for different types of women’s health services,

• 3. how they expected those changes would affect their personal use of those health services,

• 4. what effect the ACA would have on their ability to get preferred care (i.e., care they would most like to have), and

• 5. overall, whether they agreed or disagreed with the passage of the ACA.

All data were weighted and analyzed with Stata version 12.0 (StataCorp, College Station, TX) using svy commands. We used weighted proportions and χ² analysis to describe and compare women’s sociodemographic characteristics and ACA awareness and attitudes. We used multivariable logistic regression to identify sociodemographic determinants of ACA awareness and attitudes.

RESULTS

Table 1 highlights the sociodemographic and ACA-related characteristics of the sample (n = 1078). The majority (80%) had heard of the ACA, but only 24% of these women expected their insurance coverage to change as a result of the ACA. “Do not know” was a common response for specific coverage changes (not shown in Table 1), including preventive health (61%), women’s health examination (62%), breast examination (66%), contraception (65%), and mental health (76%) services. Among women who expected coverage changes and reported that those changes would affect their service use (n = 90), 52% did not intend to use more services and 17% did not know how coverage changes would affect their service use. Similar proportions were noted for changes in women’s health examinations and contraception coverage on expected service utilization (not shown). Few women (14%) believed the ACA would improve their ability to get preferred care (44% reported it would not improve their ability, and 42% did not know). Moreover, 32% of women disagreed with the ACA overall, and 44% did not know how they felt.

TABLE 1—

Proportions of Women’s Knowledge of and Attitudes Toward the Affordable Care Act, Among All Women and by Sociodemographic Groups: The Women’s Health Care Experiences and Preferences Study, United States, September 2013

		Ever Heard of ACA (n = 1078)				Expect Changes in Coverage as a Result of ACA (n = 955)				Expect Increase in Service Use as a Result of ACA Changes (n = 249)				ACA Will Improve Ability to Get Preferred Care (n = 955)				Agree With the ACA Overall (n = 955)
Sociodemographic Characteristics	%	Yes, No.	No, No.	Don't Know, No.	P	Yes, No.	No, No.	Don't Know, No.	P	Yes, No.	No, No.	Don't Know, No.	P	Yes, No.	No, No.	Don't Know, No.	P	Yes, No.	No, No.	Don't Know, No.	P
All women		80	12	8		24	8	68		31	52	17		14	44	42		24	32	44
Age group, y					< .001				.73				.13				.14				.84
18–24	15	70	23	7		23	11	66		56	37	7		22	37	41		28	26	45
25–34	28	75	12	13		25	8	67		38	47	15		16	41	43		23	29	48
35–44	27	83	11	6		28	5	67		22	58	20		11	49	40		24	34	42
45–55	30	86	7	8		22	8	70		23	58	19		10	46	44		23	34	43
Educational attainment					< .001				.03				.19				.001				< .001
< high school	9	54	31	15		17	6	77		64	27	10		19	23	58		21	20	59
High school	27	67	17	16		17	6	77		20	50	30		10	37	54		19	26	55
Some college	32	85	8	6		23	8	69		32	57	11		16	48	36		20	35	45
≥ bachelor’s	32	92	5	3		32	10	58		32	52	16		12	51	36		32	35	34
Race/ethnicity					< .001				.74				.02				.01				< .001
Non-Hispanic White	60	85	8	6		26	7	67		26	61	13		10	50	40		21	40	39
Non-Hispanic Black	14	66	20	14		23	10	67		40	51	9		19	31	50		41	5	55
Non-Hispanic Other	9	72	21	7		18	13	70		50	19	31		20	38	42		22	26	52
Hispanic	17	73	13	14		21	8	71		38	26	36		19	35	46		23	23	54
Income, $					< .001				.03				.19				.002				.04
< 25 000	18	61	23	16		16	8	77		59	30	11		18	32	49		19	23	58
25 000–49 999	22	76	14	10		22	3	75		33	46	21		18	33	49		25	27	48
50 000–74 999	19	85	9	6		29	8	63		30	60	10		13	53	34		23	34	48
≥ 75 000	41	87	7	6		27	10	63		25	55	20		10	50	40		25	36	39
Marital status					< .001				.37				.14				.21				< .001
Married	53	85	8	7		28	7	65		24	57	19		11	49	39		19	39	42
Previously married	9	85	9	6		26	10	65		40	47	13		15	39	45		27	30	44
Never married	27	67	22	11		19	8	73		50	44	6		17	37	46		32	21	47
Cohabitating	11	78	10	12		17	9	75		31	37	32		14	39	48		28	19	53
Residence					.08				.14				.47				.12				.11
Metropolitan	86	81	12	8		26	8	67		31	56	18		15	44	42		25	30	45
Nonmetropolitan	14	74	12	14		16	9	75		37	52	8		7	47	46		17	39	44
Employment status					< .001				.18				.12				.008				.03
Employed	61	85	7	8		27	8	66		33	55	12		13	49	38		24	35	41
Not employed	39	71	19	11		20	8	72		28	45	28		14	36	50		35	26	51
Religious affiliation					.04				.17				.03				.17				< .001
Yes	79	81	12	7		26	7	67		27	55	18		14	46	40		22	36	42
No	21	75	11	14		19	11	71		52	38	11		11	39	50		30	17	53
Religious service attendance					.39				.65				.52				.19				< .001
≥ weekly	28	78	12	10		26	7	68		34	50	16		12	50	38		16	44	39
< weekly	48	81	13	7		26	9	66		25	55	19		15	45	41		27	30	43
Never	24	78	10	12		21	7	72		41	48	12		13	38	50		27	20	53
Political party					< .001				.003				.14				< .001				< .001
Democrat	35	82	11	8		26	11	63		41	48	12		22	37	41		47	9	44
Republican	23	90	6	4		26	5	69		14	65	22		4	60	36		3	70	27
Independent/other	12	89	9	3		37	8	55		27	52	21		9	51	41		21	41	39
None	30	69	18	13		16	7	77		41	43	16		12	37	51		15	22	63
Type of insurance					< .001				.001				.008				< .001				< .001
Private	59	88	6	6		28	9	63		24	56	20		10	52	38		25	37	38
Medicaid/care	12	66	24	11		16	9	75		39	58	3		15	40	45		21	16	62
Other	8	82	8	10		26	14	60		56	40	5		19	26	54		22	21	58
None	17	72	20	8		13	< 1	87		66	20	15		25	27	48		26	28	47
Don’t know	4	25	27	48		18	< 1	82		< 1	< 1	< 1		5	33	62		12	17	72
Women’s health service use past 5 y					< .001				.29				.006				.59				.23
> once per year	44	83	10	8		27	10	64		40	48	11		15	45	40		28	31	41
About once per year	32	85	9	6		25	7	69		12	64	24		13	45	42		24	30	47
< once per year	14	79	12	10		20	5	75		45	35	20		14	42	44		16	39	45
Never	10	50	29	21		12	7	81		51	46	3		5	42	54		16	28	56
Childbirth (ever)a					.43				.35				.35				.05				< .001
Yes	56	82	10	8		26	7	68		27	55	18		12	49	40		18	37	45
No	44	79	13	7		22	9	68		39	47	15		16	39	44		31	26	42
Prescription contraception (ever)a					< .001				.12				.29				.15				.03
Yes	70	87	7	6		26	8	66		34	53	13		14	47	40		22	35	43
No	30	69	20	10		18	8	74		24	51	24		13	38	48		30	25	45

Note. ACA = Affordable Care Act. Results are presented as weighted proportions (%). P values are from unadjusted χ² analysis comparing ACA knowledge and attitudes across sociodemographic and reproductive history characteristics. P values considered significant at < .05, < .01, and < .001.

^aReproductive history characteristics among women who reported sexual intercourse experience (n = 938).

Women’s ACA awareness and attitudes varied by nearly all sociodemographic characteristics (Table 1). For example, compared with their counterparts, the proportion of women who had ever heard of the ACA was higher among older, White, college-educated, higher income, politically affiliated, married, employed, and privately insured women, and women with histories of recent health services and lifetime prescription contraception use. Women with college education, higher incomes, and private insurance also had a higher proportion of expecting ACA-related coverage changes and agreeing with the ACA.

Factors associated with ACA awareness and attitudes were similar in multivariable models (not shown). Sociodemographic characteristic determinants included age, race/ethnicity, educational attainment, income level, insurance status, political affiliation, and women’s recent health service use. Political affiliation was the strongest determinant of ACA awareness and attitudes across models.

DISCUSSION

Most women in our study had negative or uncertain attitudes about the ACA and lacked understanding of its health benefits for women. Women did not know how the ACA would affect their coverage and utilization of women’s health services, including well-woman and contraception services—the types of care emphasized by the ACA.

We also found sociodemographic characteristics disparities in ACA awareness, with less educated, poor, and uninsured women reporting less understanding of the ACA than socially advantaged women. We also noted this trend among infrequent users of women’s health services. These are the same groups who experience disproportionately high rates of unintended pregnancy, adverse birth outcomes, cancer-related mortality, and inequities in access to care.^10,11 In other words, women who might benefit the greatest from the ACA were particularly uninformed.

Our findings are consistent with others’ that have documented the public’s limited understanding of the ACA and its benefits.^2–4 Collectively, we could not assume that ACA implementation alone would result in desired health outcomes.^7–9 If women are unaware of their covered benefits, do not know how to enroll in or use a health plan, or have limited access to women’s health providers, then the ACA is insufficient to improve women’s health care delivery. Even if women’s rates of service utilization and insurance program participation increase as a result of the ACA, it is unclear whether specific benefits are incentive enough to improve health behaviors and outcomes.⁷ Our recent work that described a statistically modest and clinically questionable effect of cost-sharing on long-acting reversible contraception uptake would support this.¹²

The social and political context of women’s health care in the United States also determines whether improved coverage leads to service access and utilization. For instance, ongoing legislative challenges to mandated contraception coverage continue despite clear evidence of the individual and societal health benefits of family planning.^6,13 Moreover, the landscape of access to women’s health services, including cancer screening, pregnancy care, and family planning, for poor and minority women will likely change as a result of the ACA.^10,11,14 Whether and how these concurrent sociopolitical and structural processes will influence the ACA’s implementation and success are unknown.

In conclusion, our findings suggested that perhaps we should temper our expectations about the ACA’s potential for an immediate, broad impact on women’s health for 2 important reasons. First, women’s understanding of the ACA is a prerequisite of improved women’s health care utilization and outcomes. Furthermore, the ACA cannot reduce health inequities if the groups of women who would benefit the most have the least understanding of it. Research, practice, and policy efforts are urgently needed to identify and address the most critical ACA knowledge deficits and attitudes, especially for women who experience disparities in health care access and outcomes. Second, complex, macrolevel factors will continue to play a major role in the ACA’s implementation and impact. Delivery of women’s health services is uniquely influenced by broader social norms and political agendas, which also deserve greater attention from women’s health professionals in the future.