Abstract
Structured Abstract
Background
Supreme Court cases challenging the Affordable Care Act (ACA) mandate for employer-provided reproductive health care have focused on religiously based opposition to coverage. Little is known about women’s perspectives on such reproductive health policies.
Study Design
Data were drawn from the Women’s Health Care Experiences and Preferences survey, a randomly selected, nationally representative sample of 1078 US women age 18–55. We examined associations between religious affiliation and attitudes toward employer-provided insurance coverage of contraception and abortion services, and the exclusion of religious institutions from this coverage. We used chi-square and multivariable logistic regression for analysis.
Results
Respondents self-identified as Baptist (18%), Protestant (Other Mainline, 17%), Catholic (17%), Other Christian (20%), Religious, Non-Christian (7%) or no affiliation (21%). Religious affiliation was associated with proportions of agreement for contraception (p = 0.03), abortion (p <0.01), and religious exclusion (p <0.01) policies. In multivariable models, differences in the odds of agreement varied across religious affiliations and frequency of service attendance. For example, compared to non-affiliated women, Baptists and Other Nondenominational Christians (but not Catholics) had lower odds of agreement with employer coverage of contraception (OR 0.63, 95% CI 0.4-0.1 and OR 0.57, CI 0.4–0.9, respectively); women who attended services weekly or more than weekly had lower odds of agreement (OR 0.53, 95% CI 0.3–0.8 and OR 0.33, CI 0.2–0.6, respectively), compared to less frequent attenders.
Conclusions
Recent religiously motivated legal challenges to employer-provided reproductive health care coverage may not represent the attitudes of many religious women.
Keywords: Contraception, Religion, Affordable Care Act, Policy
1. Introduction
Public and policy discourse around religious beliefs, women’s health, and the Affordable Care Act (ACA) is polarized. Common narratives, which have often featured a limited number of voices from within religious groups, have generalized opposition to the ACA’s provisions on reproductive health care across religious affiliations [1]. Recent legal challenges to the ACA’s contraceptive mandate based on claims of religious freedom, including the Supreme Court decision in Hobby Lobby v. Burwell, have fueled the debate and framed the overall sociopolitical picture [2]. While many religiously affiliated organizations (including such groups as Catholics for Choice, Hadassah, and the Religious Coalition for Reproductive Justice) did advocate for the contraceptive mandate [3], overall these counter narratives received less attention than the more common narrative frame of conflict between religious freedom and the contraceptive mandate. Furthermore, although the high utilization of contraception by American women across the religious spectrum has been well documented [4], the views of women, specifically religiously affiliated women, a demographic that includes the majority of women in the US, have been neglected in the greater dialogue on religious freedom and reproductive health coverage.
Recent research suggests that the public’s support for some ACA-related women’s health benefits is generally high (69% of men and women in a recent national survey agreed with contraceptive coverage)[5]. Yet many women are uncertain (or have negative impressions) of how the ACA will affect their ability to get preferred care and their use of women’s health services [6]. However, women’s religious beliefs, which may shape their views on reproductive health services and health policies, have been notably neglected in research to date. For instance, how religiously affiliated women perceive the role of employers, including religious-based institutions, in health care coverage, and especially reproductive health coverage, has not been well documented. Furthermore, while the ACA contraceptive coverage mandate does not cover abortion services, some opponents (including some of the litigants) of the contraceptive coverage guarantee have claimed otherwise, asserting that certain methods of covered contraceptives are abortifacients, despite a lack of supporting scientific evidence [7]. Thus, although not part of the ACA coverage guarantee, women’s views on abortion may also influence their views on contraceptive policy.
Failing to comprehensively account for women’s perspectives, the group arguably at the heart of the current dialogue on religion and reproductive health care, impoverishes discourse, and misinforms policy on this major public health issue. We investigated the relationships between religious affiliation and attitudes toward reproductive health policies among a nationally representative sample of U.S. women.
2. Methods
2.1 Study design and sample
We drew data from the Women’s Health Experiences and Preferences Study, a survey we conducted in September 2013 among a nationally representative sample of US women aged 18–55. Our sample was randomly selected from the GfK probability-based Internet panel of 50,000 US residents (formerly Knowledge Networks, Menlo Park, CA). GfK Panel members are sampled using address-based methods, which includes cell-phone only households that are often excluded from Random Digit Dialing sample frames. Individuals solicited to participate in the GfK panel, but who do not have Internet access are provided with a laptop and Internet access at no cost. All panelists provide individual and household demographic data. Each member of the panel has a unique login to allow them to access online surveys and survey invitations are sent by email. GfK offers modest incentives to encourage ongoing participation among panelists. Survey weights are provided and applied to account for non-response bias and the complex survey design and to bring the final sample in line with national demographic benchmarks.
Our 29-item survey assessed women’s health care experiences and preferences across the reproductive life course and in the context of the ongoing implementation of the ACA. The survey included a series of items measuring women’s attitudes toward three policy-relevant statements about employer coverage of reproductive health care: 1) “Employers should provide health plans that cover the costs of contraception for their employees;” 2) “Employers should provide health plans that cover the costs of abortion care for their employees;” and 3) “Religious affiliated hospitals and colleges should be excluded from having to cover the costs of contraception for their employees.” Response choices included “Agree,” “Disagree,” or “Don’t know.” Our outcomes were modeled as agreement with each of these policy statements.
Our primary independent variable was self-identified religious affiliation. Women were given a choice of 13 religious affiliations: Baptist, Protestant: Other Mainline Denomination (e.g., Methodist, Lutheran, Presbyterian, Episcopal), Catholic, Pentecostal, Eastern Orthodox, Non-denominational Christian, Mormon, Jewish, Muslim, Hindu, Buddhist, Other Non-Christian Religion, or None (No Affiliation). Given the small number of respondents selecting certain affiliations, we grouped religious affiliation responses into six categories for analytic purposes: Baptist, Protestant: Other Mainline Denomination (including Lutheran, Presbyterian, Methodist, Episcopal), Catholic, Other Christian Affiliation (Non-denominational Christian, Pentecostal, Eastern Orthodox, Mormon), Religious Non-Christian (Jewish, Muslim, Buddhist, Hindu, Other) and No Affiliation.
Among women who identified a religious affiliation, we further asked about their frequency of religious service attendance, which we examined as a secondary independent variable. Response options included more than once a week, weekly, once or twice a month, a few times per year, yearly or never.
2.2 Data Analysis
We used descriptive statistics (weighted proportions, un-weighted frequencies) and bivariate chi-square tests to describe and compare reproductive health policy attitudes across religious and socio-demographic groups. We used multivariable logistic regression to evaluate associations between women’s religious affiliation, frequency of religious service attendance, and reproductive health policy attitudes while controlling for socio-demographic and reproductive history characteristics. Our covariate selection was based upon our prior work, as well as data from prior national surveys of religious life in the United States [6, 8]. These covariates included age, income, race/ethnicity, educational attainment and region of residence. We also included employment status and childbirth, as we hypothesized they would potentially influence a woman’s policy views regarding employer-based coverage of reproductive health services (our outcomes). Socio-demographic covariates were included in regression models if their p-values were <0.10 in bivariate tests.
We modeled the effects of religious affiliation and frequency of religious service attendance first separately on the three reproductive health policy attitudes and then together in a combined model, controlling for socio-demographic characteristics. We then added political party (which we hypothesized would be the strongest predictor of women’s health policy attitudes) in separate models. This approach allowed us to explore the independent as well as combined effects of religious affiliation, service attendance, and political party on the outcomes. Results are presented as weighted proportions and adjusted odds ratios with 95% confidence intervals. We considered p-values of <0.05 significant. Results are from weighted analyses using STATA 13 (College Station, TX).
3. Results
3.1 Sample characteristics
Of the 2,520 randomly sampled eligible GfK panelists (English-speaking women aged 18–55) who received the survey invitations, 43% (n=1078) opened the electronic link and completed the study. Compared to respondents, non-respondents were more likely to be younger than age 30, identify as Black or Hispanic ethnicity, have less than a high school education, and have annual incomes of less than $25,000 (all p<0.01). GfK does not collect data on religious characteristics of non-responders. The mean age of respondents was 37 years old (Table 1). The majorities were white (61%), had attended at least some college (64%), had annual incomes of $50,000 or more (60%), were employed (62%) and had had at least one child (54%). Thirty-four percent of respondents self-identified as Democrats and 23% as Republicans. Women identified as Baptist (18%), Protestant: Other Mainline Denomination (17%), Catholic (17%), Other Christian Affiliation (20%), Religious, Non-Christian (7%) and No Affiliation (21%). Amongst religiously affiliated women, service attendance varied: 26% reported attending once per week or more frequently while 24% never attended. Baptist or Other Christian Affiliation respondents reported the most frequent attendance (Table 2).
Table 1.
Sample Characteristics
| % | n | |
|---|---|---|
| 100 | 1078 | |
| Age (yrs) | ||
| 18–29 | 31 | 244 |
| 30–44 | 38 | 431 |
| 45–55 | 31 | 403 |
|
| ||
| Race/Ethnicity | ||
| White | 61 | 754 |
| Black | 14 | 108 |
| Hispanic | 17 | 133 |
| Other | 9 | 83 |
|
| ||
| Education | ||
| Less than high school | 10 | 81 |
| High school | 27 | 324 |
| Some college | 32 | 311 |
| Bachelor or higher degree | 32 | 362 |
|
| ||
| Household Income | ||
| < 25K | 18 | 200 |
| 25–<50K | 22 | 241 |
| 50–<75K | 19 | 194 |
| >=75K | 41 | 443 |
|
| ||
| Employee Status | ||
| Not working | 39 | 398 |
| Working | 62 | 680 |
|
| ||
| Region of Residence | ||
| Northeast | 18 | 190 |
| Midwest | 21 | 245 |
| South | 38 | 388 |
| West | 23 | 255 |
|
| ||
| Religious Affiliation | ||
| Refused | 0.9 | 8 |
| Baptist | 18 | 185 |
| Protestant:Other Mainline Denomination | 17 | 195 |
| Catholic | 17 | 200 |
| Other Christian Affiliation | 20 | 224 |
| Religious, Non-Christian | 7 | 63 |
| None | 21 | 203 |
|
| ||
| Religious Attendance | ||
| Refused | 0.1 | 1 |
| More than once a week | 8 | 88 |
| Once a week | 18 | 207 |
| Once or twice a month | 10 | 114 |
| A few times a year | 17 | 177 |
| Once a year or less | 21 | 218 |
| Never | 24 | 247 |
| Did not answer | 3 | 26 |
|
| ||
| Childbirth History | ||
| Never given birth | 43 | 406 |
| Ever given birth | 54 | 646 |
| Did not answer | 3 | 26 |
|
| ||
| Political Party | ||
| Democrat | 34 | 362 |
| Republican | 23 | 263 |
| Independent/Other | 12 | 136 |
| I do not affiliate with a political party | 29 | 294 |
| Did not answer | 2 | 23 |
Table 2.
Religious Affiliation and Religious Service Attendance
| % (n) p<0.01 |
more than once a week | once a week | once or twice a month | a few times a year | once a year or less | never |
|---|---|---|---|---|---|---|
| Baptist (n=184) | 30% | 27% | 23% | 22% | 14% | 6% |
| Protestant-Other Mainline Denomination (n=193) | 8% | 24% | 30% | 21% | 20% | 5% |
| Catholic (n=198) | 4% | 0% | 17% | 25% | 18% | 16% |
| Other Christian Affiliation (n=198) | 48% | 21% | 19% | 17% | 19% | 7% |
| Religious, Non-Christian (n=87) | 0.1 | 9% | 9% | 10% | 11% | 6% |
| No Affiliation (n=190) | 0% | 2% | 1% | 4% | 18% | 61% |
| Total (n=1050) | 100% (88) | 100% (207) | 100% (114) | 100% (177) | 100% (217) | 100% (247) |
Columns indicate frequency of religious service attendance broken down by religious affiliation. 27 respondents did not complete both questions, including 7 that did not complete either, 1 that answered attendance only, but not affiliation, and 20 that indicated an affiliation but no attendance response (of these, 13 indicated “no affiliation”).
3.2 Unadjusted results
Overall, 56% of women agreed that employer health plans should cover contraception, 23% agreed that abortion should be covered, and 22% agreed that religious hospitals and colleges should be excluded from contraceptive coverage requirements. While high levels of agreement with employer-provided contraceptive coverage were noted across all religious affiliations, there were notable differences among groups. Protestants: Other Mainline Denomination (66%) and Catholics (63%) had the highest agreement (even higher than No Affiliation, 59%), while Baptists (48%) and Other Christians (45%) had the lowest (p=0.03). For agreement with employer-provided coverage of abortion, more women disagreed overall, but Baptists (17%) and Other Christians (15%) had the lowest agreement and Religious Non-Christian (35%) and non-affiliated women (30%) had the highest (p<0.01). For agreement with exclusion of religious institutions from contraceptive coverage Baptists (29%) and Other Christians (27%) had the highest agreement and non-affiliated women (10%) had the lowest (p<0.01, data not shown)
For religious service attendance, proportions of agreement with contraception and abortion coverage were higher among women with less frequent religious service attendance; proportions of agreement with the exclusion of religious colleges and hospitals from contraceptive coverage requirements were higher among women with more frequent attendance (p-values <0.01, data not shown).
3.2.2 Adjusted results
In multivariable models (Tables 3a–c), we examined the impact of religious affiliation and attendance individually, and then in combined models. In models of religious affiliation alone, Baptists and Other Christians had lower odds of agreement with employer-provided coverage of contraception (OR 0.63, 95% CI 0.6–1 and OR 0.55, CI 0.4–0.8, respectively) than women with no affiliation (Table 3a, Model 1). For agreement of employer coverage of abortion (Table 3b), Baptist, Protestant: Other Mainline Denomination, Catholic, and Other Christian Affiliation women had lower odds of agreement than non-affiliated women. For agreement with the exclusion of religious hospitals and colleges from contraceptive coverage requirements (Table 3c), Baptists, Protestants: Other Mainline Denomination), Catholics and Other Christian Affiliation women had higher odds of agreement than non-affiliated women.
Table 3a.
Adjusted multivariable analysis (OR & 95%CI)
| Employers should provide health plans that cover the costs of contraception for their employees
|
|||||
|---|---|---|---|---|---|
| Model 1: affiliation | Model 2: attendance | Model 3: combined model | Model 4: combined plus political party | ||
|
| |||||
| Religious affiliation | No Affiliation | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Baptist | 0.63 (0.4–1)* | 0.93 (0.6–1.5) | 0.99 (0.6–1.7) | ||
| Protestant:Other Mainline Denomination | 0.92 (0.6–1.4) | 1.3 (0.8–2.1) | 1.3 (0.8–2.2) | ||
| Catholic | 0.79 (0.5–1.2) | 1 (0.6–1.7) | 1 (0.6–1.7) | ||
| Other Christian Affiliation | 0.55 (0.4–0.8)* | 0.9 (0.6–1.5) | 1 (0.6–1.7) | ||
| Religious, Non-Christian | 0.9 (0.5–1.7) | 1.1 (0.6–2.1) | 1 (0.5–1.9) | ||
|
| |||||
| Religious attendance | Never | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Yearly | 1.2 (0.8–1.8) | 1.2 (0.8–1.9) | 1.25 (0.8–1.9) | ||
| A Few Times per Year | 1 (0.7–1.5) | 1 (0.6–1.6) | 1 (0.6–1.6) | ||
| Once or Twice a month | 0.71 (0.4–1.1) | 0.68 (0.4–1.2) | 0.7 (0.4–1.2) | ||
| Weekly | 0.52 (0.4–0.8)* | 0.51 (0.3–0.8)* | 0.61 (0.4–1)* | ||
| More than Once per Week | 0.33 (0.2–0.6)* | 0.35 (0.2–0.6)* | 0.47 (0.3–0.9)* | ||
Model 1 (affiliation): Adjusted for socio demographic characteristics including age, race, income, education, childbirth history, employment status, region of residence. Socio demographic characteristics included in the model if p value was less than 0.1 in bivariate analyses.
Model 2 (attendance): Adjusted for the same socio-demographic characteristics as Model 1
Model 3: (combined) affiliation plus attendance, adjusted for socio-demographics as in Model 1
Model 4: combined model plus political party affiliation
indicates p<0.05
Table 3c.
Adjusted multivariable analysis (OR & 95%CI)
| Religious affiliated hospitals and colleges should be excluded from having to cover the costs of contraception for their employees
|
|||||
|---|---|---|---|---|---|
| Model 1: affiliation | Model 2: attendance | Model 3: combined model | Model 4: combined plus political party | ||
|
| |||||
| Religious affiliation | No Affiliation | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Baptist | 3.3 (1.8–6)* | 1.7 (0.9–3.4) | 1.5 (0.7–3) | ||
| Protestant: Other Mainline Denomination | 2.6 (1.5–4.6)* | 1.5 (0.7–2.9) | 1.3 (0.7–2.6) | ||
| Catholic | 3.2 (1.8–5.7)* | 2 (1–3.9)* | 1.8 (0.9–3.6) | ||
| Other Christian Affiliation | 4 (2.3–6.9)* | 1.9 (1–3.6) | 1.6 (0.8–3.1) | ||
| Religious, Non-Christian | 1.9 (0.8–4.3) | 1.4 (0.6–3.3) | 1.5 (0.6–3.6) | ||
|
| |||||
| Religious attendance | Never | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Yearly | 1.8 (1.1–3.1) | 1.6 (0.9–2.7) | 1.5 (0.8–2.6) | ||
| A Few Times per Year | 1.7 (0.9–2.9) | 1.3 (0.7–2.4) | 1.3 (0.7–2.4) | ||
| Once or Twice a month | 2.4 (1.3–4.4)* | 1.9 (1–3.6) | 1.7 (0.9–3.5) | ||
| Weekly | 5.4 (3.2–9)* | 4.1 (2.3–7.4)* | 3.5 (1.9–6.5)* | ||
| More than Once per Week | 7 (3.8–12.9)* | 5.2 (2.6–10.3)* | 4.2 (2.1–8.6)* | ||
Model 1 (affiliation): Adjusted for socio demographic characteristics including age, race, income, education, childbirth history, employment status, region of residence. Socio demographic characteristics included in the model if p value was less than 0.1 in bivariate analyses.
Model 2 (attendance): Adjusted for the same socio-demographic characteristics as Model 1
Model 3: (combined) affiliation plus attendance, adjusted for socio-demographics as in Model 1
Model 4: combined model plus political party affiliation
indicates p<0.05
Table 3b.
Adjusted multivariable analysis (OR & 95%CI)
| Employers should provide health plans that cover the costs of abortion care for their employees
|
|||||
|---|---|---|---|---|---|
| Model 1: affiliation | Model 2: attendance | Model 3: combined model | Model 4: combined plus political party | ||
|
| |||||
| Religious affiliation | No Affiliation | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Baptist | 0.31(0.2–0.5)* | 0.46 (0.3–0.9)* | 0.52 (0.3–1)* | ||
| Protestant: Other Mainline Denomination | 0.6 (0.4–1)* | 0.9 (0.5–1.6) | 0.96 (0.5–1.7) | ||
| Catholic | 0.51 (0.3–0.8)* | 0.69 (0.4–1.2) | 0.72 (0.4–1.3) | ||
| Other Christian Affiliation | 0.39 (0.2–0.6)* | 0.65 (0.4–0.1.2) | 0.77 (0.4–1.4) | ||
| Religious, Non-Christian | 1.4 (0.7–2.5) | 1.8 (1–3.5) | 1.6 (0.8–3.1) | ||
|
| |||||
| Religious attendance | Never | 1 (Referent) | 1 (Referent) | 1 (Referent) | |
| Yearly | 1.1 (0.7–1.6) | 1.2 (0.8–1.9) | 1.3 (0.8–2) | ||
| A Few Times per Year | 0.63 (0.4–1) | 0.77 (0.4–1.3) | 0.79 (0.5–1.4) | ||
| Once or Twice a month | 0.43 (0.2–0.8)* | 0.52 (0.3–1)* | 0.51 (0.3–1) | ||
| Weekly | 0.35 (0.2–0.6)* | 0.46 (0.3–0.8)* | 0.58 (0.3–1.1) | ||
| More than Once per Week | 0.14 (0.1–0.4)* | 0.18 (0.1–0.5)* | 0.28 (0.1–0.8)* | ||
Model 1 (affiliation): Adjusted for socio demographic characteristics including age, race, income, education, childbirth history, employment status, region of residence. Socio demographic characteristics included in the model if p value was less than 0.1 in bivariate analyses.
Model 2 (attendance): Adjusted for the same socio-demographic characteristics as Model 1
Model 3: (combined) affiliation plus attendance, adjusted for socio-demographics as in Model 1
Model 4: combined model plus political party affiliation
indicates p<0.05
In the models of religious service attendance alone, women who attended weekly or more than weekly had lower odds of agreement with employer-provided contraception (OR 0.52, 95% CI 0.4–0.8 and OR 0.33, CI 0.2–0.6, respectively; Table 3a) and abortion care (OR 0.35, CI 0.2–0.6 and OR 0.14, CI 0.1–0.4, respectively; Table 3b) and higher odds of agreement for exclusion of religious institutions from contraceptive coverage requirements, compared to their counterparts (OR 4.1, CI 2.3–7.4 and OR 5.2, CI 2.6–10.3, respectively; Table 3c). In combined models of religious affiliation and attendance, women who attended weekly or more than weekly had lower odds of agreement with employer-provided contraception (Table 3a) and abortion care (Table 3b) and higher odds of agreement with religious exclusion from contraceptive coverage, compared to their counterparts (Table 3c). Additionally, Baptists had lower odds of agreement with abortion care than non-affiliated women (Table 3b). Controlling for political party, the effects of religious affiliation and attendance across all outcomes became statistically insignificant, although point estimates remained stable (Tables 3a–c).
4. Discussion
In contrast to recent media narratives highlighting religious opposition toward reproductive health policies, our data argue for a more nuanced perspective of this major public health issue. Positive attitudes toward employer-based reproductive health coverage were generally high among these women. Women had overwhelmingly positive attitudes toward employer coverage of contraception, regardless of religious affiliation, and in multivariable models controlling for religious service attendance and political party, both religiously affiliated and unaffiliated women were statistically similar. Notably, the groups of religious women who have been targeted in the policy debate (i.e., Catholics) had amongst the highest rates of agreement with contraceptive coverage, as well as moderate views on abortion coverage and religious exemption. Although reproductive health issues like contraception and abortion are sometimes conflated in public and policy discourse [7], women in our study appeared to distinguish between them regardless of religious affiliation. Attitudes toward abortion coverage were the most negative, and indeed abortion remains the most polarizing reproductive health policy issue in the U.S. across socio-demographic groups [9].
We further considered the influence of religious service attendance on women’s reproductive health policy attitudes, as another proxy for religiosity in the context of family planning [10]. More frequent attendance among these women was associated with lower levels of agreement for employer provided contraceptive coverage and abortion care coverage, and higher levels of agreement with exclusion of religious institutions from contraceptive coverage requirements. Yet even among women attending services weekly, 46% agreed with employer provided insurance coverage for contraception, suggesting a significant proportion of “highly religious” women support employer provided insurance coverage of contraception. The views of religious women, which have rarely been heard in recent narratives around reproductive health policy and religion, are likely even more complex than our data permit. Further research is needed to disentangle the multiple influences of religiosity on women’s reproductive health policy attitudes.
As we hypothesized, political affiliation was strongly associated with women’s reproductive health policy attitudes but the addition of political party to models had only modest effects on the relationships between religious affiliation, service attendance and our outcomes (as evidenced by relatively stable point estimates). Furthermore, other socio-demographic and reproductive history characteristics including geographic region of residence, racial and ethnic background, income, and level of education, also appeared to shape these women’s attitudes. Overall, the complex and interrelated social, political, economic, and individual level factors that contribute to women’s understanding and acceptance of reproductive health policy in the United States require continued investigation.
Several limitations of our study are notable. We had small numbers of respondents from some religious affiliation groups (i.e. Muslim, Buddhist, Hindu) and grouping of these non-Christian religions for statistical purposes may have conceptually biased results.. Religious service attendance and political party measures may not fully reflect the intensity of religious or political interests or their impact on women’s beliefs and behaviors. Self-reported frequency of religious attendance may be overestimated [11], or may not fit the ritual practice of some particular faiths. Employer-provided reproductive health care coverage may not be solely an issue of religious freedom for some women but rather a broader issue of employer-provided health care coverage generally. Our survey did not assess this, nor did it differentiate between employer and employee views or between for- and not-for-profit employers.
While our sample was drawn from a national probability panel, our respondents appeared to be of higher age, income, education, employment, and insurance levels than the general population. Our responders were also older, more White, higher educated, and higher income than GfK non-responders. Our moderate response rate of 43% may reflect response bias. More non-respondents were racial/ethnic minority, less educated, lower income, and younger than were respondents. Gfk does not collect baseline religious demographic data on panel participants, but the religious demographic distribution of women in our study was similar to that of other large population-based religious surveys of Americans [8]. Nonetheless, our findings may not be generalizable to all U.S. women, especially socially disadvantaged women.
Despite these limitations, our findings have important implications for reproductive health policy and practice. Broadly, a paradigm shift is needed – one that more accurately reflects all women’s, including religious women’s, perspectives on contraception and abortion and debunks the notion that religious participation is a marker of opposition to reproductive health care coverage. Given that the vast majority of U.S. women identify with a religion [8], reproductive health professionals have a responsibility to correct the dominant, inaccurate public narrative around religion and reproductive health care. Continued research to provide a more complex understanding of religion and women’s health policy attitudes and their influence on women’s reproductive health care needs and service experiences is warranted.
Implications.
Recent challenges to the ACA contraceptive mandate appear to equate religious belief with opposition to employer-sponsored reproductive health coverage, but women’s views are more complex.
Acknowledgments
Hwa-Jung Choi, PhD, for assistance with statistical analysis
Funding/Support: Our work was supported by the Robert Wood Johnson Clinical Scholars Program/Department of Veterans Affairs (E. W. Patton), a NICHD Building Interdisciplinary Research Careers in Women’s Health K-12 Career Development award #K12HD001-438 (awarded to K. S. Hall; P.I. T.R.B. Johnson), and by the Program on Women’s Health Care Effectiveness Research at the Department of Obstetrics and Gynecology at the University of Michigan (Director V. K. Dalton).
Role of Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis and interpretation of data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Author Contributions: K. S. Hall and V. K. Dalton collected the data. All three authors conceptualized the project and designed the study together. E. W. Patton and K. S. Hall managed the data, conducted the analysis and interpreted the data, with guidance from V. K. Dalton. All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. E. W. Patton drafted the first version of the article, with K. S. Hall and V. K. Dalton providing substantive and editorial feedback on drafts, with E. W. Patton making subsequent revisions. All of the authors approved the final revision prior to submission.
Prior Presentation of Data: The data in this study was presented in an oral presentation at the AcademyHealth 2014 Annual Research Meeting Gender and Health Interest Group June 7,2014 in San Diego, California.
Conflict of Interest Disclosures: E. W. Patton and K. S. Hall have no conflicts of interest to disclose. V. K. Dalton received funding from McNeil Consumer Healthcare to participate on an advisory committee held in December 2012, has served as a paid consultant for Bayer (2014) and received payments from the University of California San Francisco to provide content expertise during the development of a web-based educational program on early pregnancy failure. An anonymous donor funded this project.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.See for example, Bronner E. A Flood of Suits on the Coverage of Birth Control. [accessed September 30,2014];New York Times. 2013 Jan 27;:A1. http://www.nytimes.com/2013/01/27/health/religious-groups-and-employers-battle-contraception-mandate.html?smid=pl-share.
- 2.Cohen IG, Lynch HF, Curfman GD. When Religious Freedom Clashes with Access to Care. N Engl J Med. 2014 Aug 14;361(7):596–9. doi: 10.1056/NEJMp1407965. [DOI] [PubMed] [Google Scholar]
- 3.See for example [Accessed 12/20/2014];Brief of Religious Organizations as Amici Curiae supporting the Government. at http://www.americanbar.org/content/dam/aba/publications/supreme_court_preview/briefs-v3/13-354-13-356_amcu-ro.authcheckdam.pdf.
- 4.See for example Jones RK, Dreweke J. Countering Conventional Wisdom: New Evidence on Religion and Contraceptive Use. New York: Guttmacher Institute; 2011. [Accessed 12/20/2014]. http://www.guttmacher.org/pubs/Religion-and-Contraceptive-Use.pdf.
- 5.Moniz MH, Davis MM, Chang T. Attitudes About Mandated Coverage of Birth Control Medication and Other Health Benefits in a US National Sample. JAMA. 2014 Jun 25;311(24):2539–41. doi: 10.1001/jama.2014.4766. [DOI] [PubMed] [Google Scholar]
- 6.Hall KS, Fendrick AM, Zochowski M, Dalton VK. Women’s Health and the Affordable Care Act: High Hopes Versus Harsh Realities? Am J Pub Health. 2014 Aug;104(8):e10–3. doi: 10.2105/AJPH.2014.302045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.See for example Belluck P, Echkholm E. Religious Groups Equate Some Contraceptives with Abortion. [accessed September 30, 2014];New York Times. 2012 Feb; http://www.nytimes.com/2012/02/17/health/religious-groups-equate-some-contraceptives-with-abortion.html.
- 8.Lugo L, et al. US Religious Landscape Survey: Religion, Diverse and Dynamic. Pew Research Center; Washington, D.C: Feb, 2008. [accessed on May 17, 2014]. http://www.pewforum.org/2008/02/01/chapter-3-religious-affiliation-and-demographic-groups/ [Google Scholar]
- 9.McCarthy M. Opposition to abortion is growing in Southern and Midwestern US states. BMJ. 2013 Aug 2;347:f4915. doi: 10.1136/bmj.f4915. [DOI] [PubMed] [Google Scholar]
- 10.Kramer MR, Hogue CJ, Gaydos LM. Noncontracepting behavior in women at risk for unintended pregnancy: what’s religion got to do with it? Ann Epidemiol. 2007 May;17(5):327–34. doi: 10.1016/j.annepidem.2006.10.016. [DOI] [PubMed] [Google Scholar]
- 11.Cox D, Jones R, Navarro-Rivera J. I Know What You Did Last Sunday: Measuring Social Desirability Bias in Self-Reported Religious Behavior, Belief, and Identity. Public Religion Research Institute; Washington, D. C: May, 2014. [accessed September 30, 2014]. http://publicreligion.org/research/2014/05/aapor-2014/ [Google Scholar]