Abstract
Objectives:
We aimed to compare differences in receipt of any and specific types of fertility services between people with Medicaid and private insurance.
Methods:
We used National Survey of Family Growth (2002-2019) data and linear probability regression models to examine the association between insurance type (Medicaid or private) and fertility service use. The primary outcome was use of fertility services in the past 12 months, and secondary outcomes were use of specific types of fertility services at any time: (1) testing, (2) common medical treatment, and (3) use of any fertility treatment type (testing, medical treatment, or surgical treatment of infertility). We additionally calculated time-to-pregnancy using a method that estimates the unobserved total amount of time the respondent spent trying to become pregnant using their current duration of pregnancy attempt at the time of the survey. We calculated time-to-pregnancy ratios across respondent characteristics to examine if insurance type was associated with differential time-to-pregnancy.
Results:
In adjusted models, Medicaid coverage was associated with an 11.2 percentage point (95% CI: −22.3, −0.0) lower use of fertility services in the past 12 months compared to private coverage. Relative to private coverage, Medicaid insurance was also associated with large and statistically significantly lower rates of ever having used infertility testing or any fertility services. Insurance type was not associated with differences in time-to-pregnancy.
Conclusions:
People covered by Medicaid were less likely to have used fertility services compared with people with private insurance. Differences in coverage of fertility services between Medicaid and private payers may represent a barrier to fertility treatment for Medicaid recipients.
Keywords: Medicaid, Fertility, Infertility, Health Care Utilization, Reproductive Health Services, Health Disparities, Health Insurance Coverage
Introduction
In the United States, approximately 10-15% of heterosexual couples experience infertility, defined as the inability to conceive after 12 months of regular unprotected sex (American Society for Reproductive Medicine, 2022). In addition to helping heterosexual couples who experience infertility to become pregnant, fertility treatment is needed for same-sex couples and single individuals to have biological children. Use of Assisted Reproductive Technology (most commonly in-vitro fertilization, or IVF) has increased over time in the United States, from 1.0% of all infants born in 2005 (Wright, Chang, Jeng, & Macaluso, 2008) to 2.0% of all infants born in 2018 (Sunderam et al., 2022). During the period from 2015 to 2019, approximately 12% of reproductive-age women had previously used of some form of fertility services, including counseling, diagnosis, and treatment services (National Center for Health Statistics, 2021).
Despite the high need for and growing use of infertility services, insurance coverage of infertility diagnosis and treatment is limited in the US (Weigel et al., 2020). Over the past several decades, many states have mandated that private insurers cover fertility treatment, and these policy changes have been associated with increased fertility treatment in those states, primarily among more educated women (Bitler & Schmidt, 2012). West Virginia was the first state to pass a fertility treatment mandate in 1977 (Schmidt, 2005), and since then an additional 15 states have mandated private insurers to cover fertility treatment (Weigel et al., 2020). However, even after these policy changes, coverage for more expensive fertility treatment services such intrauterine insemination (IUI) and IVF remains limited for people with private insurance. As of 2017, 69% of large employers offered coverage for fertility evaluation, 49% covered fertility drugs, 35% covered IUI, and 37% covered IVF (Ferreira, 2018).
While important gaps in coverage exist for people with private insurance — in particular for single individuals, women and men in same-sex relationships, and transgender individuals (Kawwass, Penzias, & Adashi, 2021) — coverage of infertility treatment is nearly non-existent for most persons covered by Medicaid (The Ethics Committee of the American Society for Reproductive Medicine, 2021). Medicaid is a public health insurance program in the US that is operated by states within federal guidelines to cover people with low incomes, with a median eligibility level for adults without dependent children at 138% of the federal poverty level (FPL) in 2022 (Brooks et al., 2022). As of 2021, Medicaid covered 18% of nonelderly adult women overall and 50% of women with incomes below the federal poverty level (Kaiser Family Foundation, 2022a), but most state Medicaid programs (42) did not cover infertility diagnosis or treatment services (Weigel et al., 2020). Eight states covered education, counseling, screening, or diagnosis of infertility only, without offering coverage for any form of fertility treatment. Two states, New Hampshire and Nebraska, cover fertility treatment only when infertility is due to a medical condition. Only New York’s state Medicaid program offers fertility treatment regardless of the cause of infertility. However, coverage is limited to up to three cycles of fertility medication, without coverage for more expensive services such as IUI and IVF. Basic fertility services (e.g., reproductive history, physical exam, diagnostic services, and referrals as appropriate) are recommended by US guidelines for quality family planning provision (Gavin et al., 2017). In 2013-2014, 65% of publicly funded clinics offered some form of basic fertility services, while only approximately 16% of publicly funded clinics provided infertility treatment services (Loyola Briceno et al., 2019).
For people whose insurance coverage does not cover fertility treatment, out-of-pocket costs for treatment range from $2,623 for IUI to $19,232 for IVF (Wu et al., 2014). For low-income people with Medicaid insurance coverage, the cost of treatment may be prohibitively expensive. Inequity in access to fertility services, and treatment specifically, has important implications for public policy and reproductive justice, which includes the right to have a child as well as the right to not have child (Perritt & Eugene, 2021). Differences in coverage of fertility services, even basic fertility care as recommended in the quality family planning guidelines, may threaten reproductive justice (Gavin et al., 2017). However, little evidence exists documenting the association between Medicaid versus private insurance coverage on use of fertility services (Kessler et al., 2013; Thakker et al., 2021).
In this study we examined the association between type of insurance coverage (Medicaid versus private) and use of fertility services in the past 12 months among people trying to conceive. We also examined ever use of fertility services and ever use of specific types of services (i.e., diagnostic services, treatment types). We hypothesize that compared to people with private coverage, people with Medicaid coverage trying to conceive were less likely to have used fertility services during the past 12 months.
Material and Methods
Data and study population
We used 2002-2019 data from the National Survey of Family Growth (NSFG) Female Respondent Files. The NSFG includes nationally representative data on pregnancy, births, infertility, and reproductive health among a sample of respondents aged 15-49, with a response rate among female respondents ranging from 80% in 2002 to 65% in 2017-2019 (National Center for Health Statistics, 2020a). The NSFG offers a small financial incentive to participants, and was designed to gather information on pregnancy and births, marriage and cohabitation, infertility, use of contraception, family life, and general and reproductive health. Use of NSFG survey weights allows for national estimates, accounting for nonresponse, the sampling stratum, variance units, and complex survey design, detailed in Appendix Table A1 (National Center for Health Statistics, 2020b).
As the NSFG originally surveyed respondents aged 15-44 and then expanded to include individuals aged 45-49 years in 2015, we limited the study analyses to respondents aged 15-44 to maintain a consistent age range over the survey years (Centers for Disease Control and Prevention, 2020). To identify a study population who may have wanted fertility services, the main study population was restricted to include non-pregnant women who had not undergone surgical sterilization, who had sex in the month of the interview, who were not using contraception, and who reported that they were currently trying to become pregnant at the time of the survey (n=1,134; 4% of survey respondents).
Outcome, exposure, and covariate variables
The study’s primary outcome was use of fertility services in the past 12 months, measured as a binary variable indicating that the respondent reported at least one visit within the past 12 months to a doctor or other medical care provider to help the respondent become pregnant. As the type of visit was not specified, these fertility services could potentially include general advice or counseling related to fertility, fertility testing, or fertility treatment. We also examined different binary measures to assess whether the respondent ever used specific types of fertility services measured as: 1) infertility testing (for respondent and/or partner), 2) common medical treatment (drugs to improve ovulation, artificial insemination, IVF or other assisted reproductive technology), or 3) any fertility services, including advice or treatment: infertility testing, drugs to improve ovulation, artificial insemination, IVF or other assisted reproductive technology, surgery to correct blocked tubes, or surgery or drug treatment for endometriosis or uterine fibroids.
The study exposure was insurance type, measured as a binary variable indicating whether the respondent was covered by Medicaid or a private payer at the time of the survey. The covariates were age at the time of the survey, categorized as 15-24, 25-29, 30-34, 35-39, or 40-44; income measured as percent of the federal poverty level (<100, 100-199, 200-299, 300-399, 400-499, 500+); education categorized as middle school or less, high school, college, or more than college; and survey wave measured as an indicator for NSFG questionnaire cycle (2002, 2006-2010, 2011-2013, 2013-2015, 2015-2017, and 2017-2019).
Main analyses
The study’s main analysis examined the association between insurance coverage type (Medicaid compared to private) and use of fertility services within the past 12 months among respondents who reported trying to become pregnant at the time of the survey using multiple linear regression models. To obtain coefficient estimates with a direct interpretation of the magnitude of the difference in the outcome between groups, we used linear regression models to estimate absolute differences in probability between groups. In adjusted models, we included covariates for age, education level, and poverty level, and survey wave to control for different factors that could be driving differences in use of fertility services, such as age or increases in availability of fertility services over time and ability to pay high out-of-pocket costs for fertility services.
An additional consideration for understanding differences in receipt of fertility services is that the experience of infertility may differ depending on insurance status. Therefore, we examined whether coverage type was associated with the total duration of pregnancy attempt, or time-to-pregnancy (TTP), as estimated using a previously developed current duration approach (Keiding, Kvist, Hartvig, Tvede, & Juul, 2002; Thoma et al., 2013). Using backward recurrence time survival methods, we are able to estimate a total (unobserved) duration of pregnancy from the observed current duration of pregnancy attempt; this approach has been described in detail in previous publications using NSFG data (Keiding et al., 2002; Thoma et al., 2013). To compare TTP by coverage type, we applied an accelerated failure time parametric survival model that calculates time ratios representing the ratio of median values of the estimated TTP across characteristics. Consistent with main analyses, these models included covariates for age, education level, poverty level, and survey wave. NSFG survey data does not contain individual identifiers; therefore, this study was not considered human subjects research by [Blinded] University.
Additional analyses
We conducted several supplemental analyses, available in the Appendix. First, to better understand the extent to which there are differences in demand for fertility services between persons covered by Medicaid and private insurance, we compared the percent of surveyed individuals who were trying to become pregnant by payer overall and by age group.
Previous research on access to fertility treatment has restricted the analysis to respondents meeting the “constructed measure of infertility” (Thoma et al., 2013), a measure created by NSFG that measures infertility based on 12 months of unprotected sexual intercourse without pregnancy among respondents who are not surgically sterile and are married or cohabitating; however, the measure is not limited to respondents who reported wanting to conceive. As a sensitivity analysis, we also examined the association between insurance type and recent use of fertility services among the population of respondents who met the constructed measure of infertility using the same linear regression model and control variables described above. Finally, we conducted the same analysis in a third population of individuals who met the NSFG constructed measure of infertility and who also reported trying to conceive at the time of the survey.
As selection into the study population by insurance type is also a potential concern for people who met NSFG’s constructed measure of infertility, we examined whether insurance type was associated with infertility as measured by the constructed measure of infertility using linear regression adjusted for the same covariates as the main analysis.
Finally, Medicaid expansion in 2014 under the Affordable Care Act (ACA) greatly increased the share of reproductive age women in the US who were eligible and enrolled in Medicaid (Bullinger et al., 2022). Therefore, the population of people with Medicaid before 2014 differed from people with Medicaid coverage after Medicaid expansion. To examine whether our results remained consistent among the expanded Medicaid population, we repeated the main analysis between insurance type and fertility services only in the 2015-2019 period, after the majority of the coverage expansions has occurred.
Results
The combined NSFG survey years included a total of 31,085 respondents with health insurance who were not pregnant, of whom 1,134 were trying to become pregnant at the time of the survey. Approximately 87% of women trying to become pregnant had private insurance coverage, and 13% had Medicaid coverage at the time of the survey (Table 1). The percentage of respondents trying to become pregnant who were 15 to 24 years old was 11%; 27% were ages 25 to 29, 26% were ages 30 to 34, 24% were 35 to 39, and 12.4% were ages 40 to 44.
Table 1.
Sample demographics, National Survey of Family Growth 2002-2019
| Characteristics | Respondents Currently Trying to Become Pregnant (n=1,134) | All Non-Pregnant Respondents (n=31,085) | ||
|---|---|---|---|---|
| N | (%) | N | (%) | |
| Insurance | ||||
| Private | 932 | (86.9) | 23.156 | (80.1) |
| Medicaid | 202 | (13.1) | 7,929 | (19.9) |
| Age | ||||
| 15-24 | 144 | (11.2) | 10.610 | (31.8) |
| 25-29 | 298 | (27.0) | 5,525 | (16.4) |
| 30-34 | 307 | (25.8) | 5,406 | (16.6) |
| 35-39 | 248 | (23.6) | 4,956 | (17.3) |
| 40-44 | 137 | (12.4) | 4,588 | (18.0) |
| Education | ||||
| Middle School or Less | 60 | (3.6) | 3,025 | (8.3) |
| High School | 312 | (23.4) | 10,999 | (31.5) |
| College | 547 | (50.5) | 12,908 | (4.5) |
| More than College | 215 | (22.5) | 4,152 | (15.4) |
| Poverty Income Level | ||||
| <100 | 185 | (12.3) | 7,697 | (20.5) |
| 100-199 | 187 | (14.2) | 6,707 | (20.4) |
| 200-299 | 185 | (15.2) | 5,346 | (17.7) |
| 300-399 | 129 | (12.9) | 4,176 | (14.1) |
| 400-499 | 162 | (14.6) | 3,210 | (11.4) |
| 500 or Greater | 286 | (30.8) | 3,949 | (15.9) |
Notes: Unweighted sample sizes and weighted percentages are presented. Data are weighted using NSFG complex survey weights. Sample includes respondents age 15-44 with private insurance or Medicaid coverage.
Approximately 12% and 14% of respondents lived in households with income levels less than 100% and 100-199% of the federal poverty level, respectively (Table 1). Respondents trying to become pregnant were relatively evenly distributed between NSFG survey rounds.
Appendix Figure A1 presents the percentage of NSFG respondents who were trying to become pregnant at the time of the survey by insurance payer overall and by age group. Overall 4.3 percent of women with private coverage and 2.6 percent of women with Medicaid were trying to become pregnant at the time of the survey, with women with private coverage 1.7 percentage points (95% CI: 1.0, 2.4) more likely to be trying to become pregnant at the time of the survey compared to women with Medicaid. In the 25 to 29 and 30 to 34 year age groups, women who were covered by private coverage were 3.5 percentage points (95% CI: 1.4, 5.6) and 3.2 percentage points (95% CI: 1.3, 5.1) respectively more likely to be trying to become pregnant at the time of the survey. In the 35-39 year age group, women with private coverage were 4.5 percentage points (95% CI: 2.9, 6.0) more likely to be trying to become pregnant. There was no difference in the percent of respondents trying to become pregnant by insurance payer in the 15 to 24 year and 40 to 44 year age groups.
Comparing Medicaid with private insurance among those trying to become pregnant, insurance coverage type was not associated with TTP (adjusted time ratio 0.8 [95% CI: 0.6, 1.1]) (Table 2). Similarly, there was no statistically significant association between insurance coverage type and meeting the constructed measure of infertility (adjusted percent −1.5 [95% CI: −3.3, 0.3]) (Appendix Table A2).
Table 2.
Time ratios and 95% confidence intervals (CIs) for the association between descriptive characteristics and time-to-pregnancy using the current duration approach, National Survey of Family Growth 2002-2019
| Characteristics | Unweighted N (n=1,134) | Unadjusted Time Ratio (95% CI) | Adjusted Time Ratio (95% CI) |
|---|---|---|---|
| Insurance | |||
| Private | 932 | 1.0 | 1.0 |
| Medicaid | 202 | 0.9 (0.7, 1.1) | 0.8 (0.6, 1.1) |
| Age | |||
| 15-24 | 144 | 1.0 | 1.0 |
| 25-29 | 298 | 1.4 (1.1, 1.8)** | 1.5 (1.2, 2.0)** |
| 30-34 | 307 | 1.6 (1.2, 2.1)** | 1.9 (1.4, 2.6)*** |
| 35-39 | 248 | 2.5 (1.8, 3.6)*** | 3.1 (2.1, 4.5)*** |
| 40-44 | 137 | 4.0 (2.8, 5.8)*** | 4.8 (3.3, 7.1)*** |
| Education | |||
| Middle School or Less | 60 | 1.0 | 1.0 |
| High School | 312 | 1.0 (0.6, 1.6) | 1.2 (0.7, 1.9) |
| College | 547 | 0.9 (0.5, 1.4) | 0.8 (0.5, 1.3) |
| More than College | 215 | 0.8 (0.4, 1.3) | 0.5 (0.3, 0.9)* |
| Poverty Income Level | |||
| <100 | 185 | 1.0 | 1.0 |
| 100-199 | 187 | 1.0 (0.7, 1.4) | 1.0 (0.7, 1.3) |
| 200-299 | 185 | 1.0 (0.7, 1.4) | 1.1 (0.8, 1.7) |
| 300-399 | 129 | 0.7 (0.5, 1.0)* | 0.7 (0.5, 1.0) |
| 400-499 | 162 | 1.0 (0.7, 1.4) | 0.9 (0.6, 1.4) |
| 500 or Greater | 286 | 1.1 (0.8, 1.4) | 1.0 (0.7, 1.5) |
Notes: Unweighted sample sizes and weighted estimates are presented. Data are weighted using NSFG complex survey weights. Sample includes respondents age 15-44 with private insurance or Medicaid coverage.
p<0.05
p<0.01
p<0.001
Among those trying to become pregnant with private insurance coverage, 24.1% (95% CI: 19.6, 29.2) of respondents had used fertility services in the 12 months before the survey after adjustment (Table 3). Medicaid coverage was associated with an 11.2 percentage point (95% CI: −22.3, −0.0, p<0.05) lower use of fertility services in the past 12 months compared to private coverage after adjustment (Table 3). When limiting the survey period to 2015-2019 after the majority of the ACA Medicaid expansions, the association between Medicaid and use of fertility services was similar in magnitude (−13.9 percentage points, 95% CI: −32.7, 3.8, p=0.121), but was not statistically significant at p<0.05 (Appendix Table A3).
Table 3.
Association between insurance type and use of fertility services within the past 12 months, National Survey of Family Growth 2002-2019
| Unadjusted | Adjusted | |
|---|---|---|
| Percent (95% CI) | Percent (95% CI) | |
| Insurance | ||
| Private | 1.0 | 1.0 |
| Medicaid | −14.3*** (−21.5, −7.2) | −11.2* (−22.3, −0.0) |
| Age | ||
| 15-24 | 1.0 | |
| 25-29 | 2.0 (−13.4, 17.4) | |
| 30-34 | −3.6 (−17.3, 10.1) | |
| 35-39 | −3.8 (−19.8, 12.2) | |
| 40-44 | −3.7 (−18.7, 11.2) | |
| Education | ||
| Middle School or Less | 1.0 | |
| High School | 9.4 (−0.1, 19.0) | |
| College | 10.3 (−0.6, 21.2) | |
| More than College | 11.0 (−1.7, 23.7) | |
| Poverty Level | ||
| <100 | 1.0 | |
| 100-199 | −2.9 (−15.6, 9.7) | |
| 200-299 | 1.8 (−12.0, 15.6) | |
| 300-399 | −4.4 (−19.7, 10.9) | |
| 400-499 | −1.9 (−16.9, 13.1) | |
| 500 or Greater | 8.0 (−6.4, 22.4) | |
| Mean Among Private Insurance | 24.1 (19.6, 29.2) | 24.1 (19.6, 29.2) |
Notes: n=1,134. Coefficient estimate for insurance type represents percentage point difference in the outcome among people with Medicaid coverage compared to people with private coverage. Data are weighted using NSFG complex survey weights. Sample includes respondents age 15-44 trying to become pregnant at the time of the survey with private insurance or Medicaid coverage, 2002-2019. Adjusted model includes age, poverty level, and survey wave fixed effects.
p<0.05
p<0.01
p<0.001
Approximately 16% of all respondents who met the constructed definition of infertility had used infertility services in the past 12 months (Appendix Table A4). In this population, the association between insurance coverage type and use of fertility services in the past twelve months was smaller than the effect among respondents trying to become pregnant at the time of the survey and was not statistically significant after adjusting for covariates (−6.0 percentage point, 95% CI: −14.7, 2.6). Among individuals who met the constructed definition of infertility and were trying to become pregnant at the time of the survey, the coefficient was similar in magnitude to the estimate in the main study population (i.e., people trying to become pregnancy at the time of the survey), but the result was not statistically significant (−13.3 percentage points, 95% CI: −28.8, 2.1, p=0.090).
Among individuals with private insurance, 31.9% (95% CI: 27.0, 37.3) reported ever using infertility testing, 22.2% (95% CI: 18.0, 27.1) reported ever using common fertility treatment, and 35.5% (95% CI: 30.2, 41.2) reported ever using any fertility services. In adjusted analyses of the association between insurance type and ever having used each specific type of fertility services, Medicaid insurance was associated with a 14.8 percentage point (95% CI: −24.0, −5.6) lower rate of ever having received infertility testing, and a 16.5 percentage point (95% CI: −27.4, −5.5) lower rate of receiving any fertility services relative to private insurance (Table 4). However, we find no evidence of a statistically significant relationship between insurance type and previous use of drugs to improve ovulation, artificial insemination, IVF, or other assisted reproductive technology (−6.8 percentage points, 95% CI: −15.4, 1.7, p=0.118).
Table 4.
Association between insurance type and types of fertility services, National survey of Growth 2002-2019
| (1) Infertility Testing | (2) Common Fertility Treatment | (3) Any Fertility Services | ||||
|---|---|---|---|---|---|---|
| Unadjusted | Adjusted | Unadjusted | Adjusted | Unadjusted | Adjusted | |
| Percent (95% CI) | Percent (95% CI) | Percent (95% CI) | Percent (95% CI) | Percent (95% CI) | Percent (95% CI) | |
| Insurance | ||||||
| Private | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
| Medicaid | −25.7*** | −14.8** | −17.5*** | −6.8 | −27.0*** | −16.5** |
| (−32.3, −19.0) | (−24.0, −5.6) | (−23.9, −11.1) | (−15.4, 1.7) | (−34.6, −19.4) | (−27.4, −5.5) | |
| Age | ||||||
| 15-24 | 1.0 | 1.0 | 1.0 | |||
| 25-29 | 10.1 | 3.2 | 10.1 | |||
| (−0.5, 20.8) | (−6.0, 12.4) | (−1.9, 22.1) | ||||
| 30-34 | 16.2** | 9.1* | 15.7** | |||
| (7.3, 25.1) | (0.5, 17.7) | (5.6, 25.8) | ||||
| 35-39 | 21.9** | 11.2* | 22.5** | |||
| (9.2, 34.5) | (0.6, 21.7) | (9.0, 36.0) | ||||
| 40-44 | 35.8** | 24.7** | 33.8** | |||
| (21.7, 49.9) | (11.7, 37.6) | (19.2, 48.3) | ||||
| Education | ||||||
| Middle School or Less | 1.0 | 1.0 | 1.0 | |||
| High School | 16.4* | 3.2 | 17.0* | |||
| (2.2, 30.5) | (−8.9, 15.3) | (2.2, 31.8) | ||||
| College | 14.4* | 10.9 | 16.3* | |||
| (1.6, 27.3) | (−1.7, 23.6) | (2.4, 30.3) | ||||
| More than College | 20.0* | 7.3 | 18.6* | |||
| (3.3, 36.6) | (−6.8, 21.4) | (1.5, 35.8) | ||||
| Poverty Level | ||||||
| <100 | 1.0 | 1.0 | 1.0 | |||
| 100-199 | 7.1 | 3.2 | 7.5 | |||
| (−4.5, 18.6) | (−7.3, 13.8) | (−4.5, 19.6) | ||||
| 200-299 | 11.6 | 12.3* | 14.3* | |||
| (−0.4, 23.5) | (0.9, 23.7) | (0.9, 27.8) | ||||
| 300-399 | 2.2 | −0.7 | 0.8 | |||
| (−11.2, 15.6) | (−12.3, 10.8) | (−13.4, 15.0) | ||||
| 400-499 | 5.1 | 7.0 | 7.0 | |||
| (−9.0, 19.1) | (−6.4, 20.5) | (−8.4, 22.3) | ||||
| 500 or Greater | 8.2 | 6.0 | 6.2 | |||
| (−5.7, 22.1) | (−6.1, 18.1) | (−8.0, 20.4) | ||||
| Mean Among Private Insurance | 31.9 | 31.9 | 22.2 | 22.2 | 35.5 | 35.5 |
| (27.0, 37.3) | (27.0, 37.3) | (18.0, 27.1) | (18.0, 27.1) | (30.2, 41.2) | (30.2, 41.2) | |
Notes: n=1,134. Coefficient estimate for insurance type represents percentage point difference in the outcome among people with Medicaid coverage compared to people with private coverage. Data are weighted using NSFG complex survey weights. Sample includes respondents age 15-44 trying to become pregnant at the time of the survey with private insurance or Medicaid coverage, 2002-2019. Adjusted model includes age, poverty level, and survey wave fixed effects.
p<0.05
p<0.01
p<0.001
Discussion
This study found that among individuals who were trying to become pregnant at the time of the survey, persons with Medicaid coverage were 11.2 percentage points, or approximately 50%, less likely to have used fertility services in the past 12 months compared to persons with private insurance. These associations do not appear to be driven by differences in household income, education, or differences in age between persons with Medicaid and private insurance coverage who were trying to become pregnant.
There was little difference between insurance coverage group in time-to-pregnancy or infertility among people trying to conceive. This indicates that our choice of study population of people trying to conceive did not result in selection bias, which may have occurred if the length of time trying differed by payer. Had we found a difference in time trying by payer, we would have been concerned that any difference in use of services was driven by differential need by payer, since people with a longer period of trying could be more likely to be experiencing fertility problems and need fertility services.
When partitioning out the types of services received, we found that the difference between Medicaid and private insurance was largely related to infertility diagnosis and testing, rather than treatment. Together, these findings suggest that the experience of infertility was similar regardless of insurance status, yet people covered by Medicaid were less likely to have sought or accessed fertility care services compared with people with private insurance. When partitioning out the types of services received, we found that respondents with private insurance were twice as likely to have received fertility testing and diagnosis services compared to respondents with Medicaid, but the difference between Medicaid and privately insured respondents in use of fertility treatment was small and not statistically significant. These results suggest that high out-of-pocket cost for infertility treatment may have been a barrier for respondents irrespective of their insurance type, though persons covered by private insurance may have had better access to testing and diagnosis services than people with Medicaid. As testing and diagnosis services are less expensive to provide, achieving more equal coverage of these services between insurance types is feasible without dramatically increasing Medicaid spending. Increasing access to basic fertility diagnosis, testing services, and referrals would be consistent with goals from the Centers for Disease Control and Prevention and the Office of Population Affairs for the provision of quality family planning services (Gavin et al., 2017).
This study adds evidence to a very small literature on the difference in access to fertility treatment by coverage type. Previous studies, using 1995 NSFG data and a sample of respondents who were infertile or reported difficulty getting pregnant, found that women with private coverage had higher rates of ever having used infertility services compared to respondents with other insurance, but these studies did not specifically examine Medicaid coverage (Farley Ordovensky Staniec & Webb, 2007; Stephen & Chandra, 2000). More recent research found no significant difference in ever having used infertility services among respondents with private and Medicaid coverage among women who reported difficulty conceiving (Thakker et al., 2021). However, the study population included in these previous studies limits their policy relevance. Sixteen percent of women who met the criteria for the NSFG definition of infertility were neither trying to become pregnant nor desired a child, and therefore may not have wanted or needed fertility services (Greil, Slauson-Blevins, Tiemeyer, McQuillan, & Shreffler, 2016). Additionally, self-reported infertility likely differentially samples people with commercial coverage, because they are more likely to have access to diagnostic services. To our knowledge, this study is the first to overcome these limitations by examining recent use of fertility services among a sample of women who were trying to become pregnant at the time of the survey.
Limitations
This study has several limitations. First, the results of this study do not provide a causal estimate of the effect of insurance type on use of fertility services. While we adjusted for income and other demographic factors that are likely to be important confounding variables in the relationship between insurance type and fertility treatment, experimental or quasi-experimental research is needed to establish a causal relationship. An additional limitation is that the income eligibility limit for Medicaid for non-pregnant adults has changed considerably during the last decade, from an average of 64% of the FPL in 2011 to 138% of FPL in 2022 (Kaiser Family Foundation, 2022b). Therefore, the estimates we present in this study may not represent the present day relationship between Medicaid coverage and fertility treatment use. However, we found that the magnitude of the association between Medicaid and recent use of fertility services was similar when we restricted the sample to only survey years after the ACA’s coverage expansions (2015-2019), when most states had raised Medicaid eligibility for adults to at least 138% of FPL. In addition, NSFG survey questions on specific types of fertility services do not specify a time period of use, and therefore some women may report services used prior to the episode of trying to become pregnant and being covered by Medicaid captured by the survey. However, the magnitudes of effect when examining current and ever use of services were relatively comparable in our analysis. Finally, insurance type could have a more limited influence on care-seeking decisions if patients lack the knowledge of which services are covered. Increasing awareness among both healthcare providers and patients regarding the range of services available within coverage types could potentially improve access to fertility services.
Implications for Practice and/or Policy
The Centers for Disease Control and Prevention considers the detection and management of infertility to be a public health priority, and has specifically called for improved access to infertility screening, diagnosis, and treatment services (Centers for Disease Control and Prevention, 2014). Although increasing coverage of fertility services for people with Medicaid insurance would increase public healthcare expenditures at the state and national levels, increasing coverage to the level of people with private insurance would not generate limitless costs. Most state mandates allow insurance companies to limit fertility service coverage in some way, often by allowing insurers to limit the number of IVF cycles that will be paid for by their coverage. Furthermore, services such as basic testing, counseling, and medication are relatively low-cost services, and therefore could be covered without adding significant Medicaid program costs. As demonstrated from prior research, some publicly funded family planning clinics provide basic infertility services (Loyola Briceno et al., 2019), in keeping with quality family planning recommendations (Gavin et al., 2017). Existing service availability in publicly funded family planning clinics (Loyola Briceno et al., 2019) and recent state-level policy changes expanding Medicaid coverage of fertility services (Weigel et al., 2020) challenge arguments that increasing access is not feasible. Finally, the cost of infertility services is not fixed, and could potentially be decreased with expansion of available providers and insurance coverage.
While increasing coverage of fertility services is necessary to achieve reproductive justice, ensuring equitable access would also require addressing other barriers that disproportionately affect low-income individuals (e.g., inability to take time off and negative provider-patient experiences) (Bell, 2010). Although increasing equitable fertility coverage alone is not sufficient to resolve reproductive inequalities (Bell, 2010), policy efforts should include advocating for public insurance coverage of the full spectrum of reproductive health services, including diagnosis and treatment of infertility, to increase the ability of all persons in the United States to choose whether and when to have children.
Conclusions
These findings suggest that Medicaid’s lack of coverage for fertility diagnosis and treatment services may be a barrier to fertility service access for people with Medicaid coverage who want to become pregnant. Medicaid coverage of fertility testing, diagnosis, and treatment services may be necessary to ensure that people with Medicaid are able to access fertility services and have the information and means to do so.
Supplementary Material
Funding statement:
Dr. Eliason reports research support from the Agency for Healthcare Research and Quality under grant award T32 HS000011. Dr. Steenland was supported by the National Institute of Child Health and Human Development (P2C HD041020) and by the Agency for Healthcare Research and Quality (K01 HS027464). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Biographies
Dr. Erica Eliason is a postdoctoral research fellow at Brown University School of Public Health. Her research evaluates the effects of health policies on maternal, child, and reproductive health, with a particular focus on Medicaid
Dr. Marie Thoma is an Associate Professor in the Department of Family Science at the University of Maryland, School of Public Health. Her research focuses on population-based approaches to improve reproductive and maternal and child health.
Dr. Maria Steenland is a health services and health policy researcher focused on maternal and reproductive health policy in the United States. She uses econometric methods to evaluate maternal and reproductive health programs and policies.
Footnotes
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 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
- American Society for Reproductive Medicine. (2022). FAQs About Infertility. Retrieved from https://www.reproductivefacts.org/faqs/frequently-asked-questions-about-infertility/q01-what-is-infertility/
- Bell AV (2010). Beyond (financial) accessibility: Inequalities within the medicalisation of infertility: Inequalities within the medicalisation of infertility. Sociology of Health & Illness, 32(4), 631–646. 10.1111/j.1467-9566.2009.01235.x [DOI] [PubMed] [Google Scholar]
- Bitler MP, & Schmidt L (2012). Utilization of infertility treatments: the effects of insurance mandates. Demography, 49(1), 125–149. doi: 10.1007/s13524-011-0078-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brooks T, Gardner A, Osorio A, Tolbert J, Corallo B, Ammula M, & Moreno S (2022, March). Medicaid and CHIP Eligibility and Enrollment Policies as of January 2022: Findings from a 50-State Survey. Kaiser Family Foundation. https://files.kff.org/attachment/REPORT-Medicaid-and-CHIP-Eligibility-and-Enrollment-Policies-as-of-January-2022.pdf [Google Scholar]
- Bullinger LR, Simon K, & Edmonds BT (2022). Coverage Effects of the ACA’s Medicaid Expansion on Adult Reproductive-Aged Women, Postpartum Mothers, and Mothers with Older Children. Matern Child Health J, 26(5), 1104–1114. doi: 10.1007/s10995-022-03384-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2020). About the National Survey of Family Growth. Retrieved from https://www.cdc.gov/nchs/nsfg/about_nsfg.htm
- Centers for Disease Control Prevention. (2014). National public health action plan for the detection, prevention, and management of infertility. Retrieved from Atlanta, Georgia: Centers for Disease Control and Prevention: https://www.cdc.gov/reproductivehealth/infertility/pdf/drh_nap_final_508.pdf [Google Scholar]
- Farley Ordovensky Staniec J, & Webb NJ (2007). Utilization of infertility services: how much does money matter? Health Serv Res, 42(3 Pt 1), 971–989. doi: 10.1111/j.1475-6773.2006.00640.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ferreira E (2018). The Birth Rate is Rising Among Older Women. Got IVF Coverage? [Google Scholar]
- Gavin L, Pazol K, & Ahrens K (2017). Update: Providing Quality Family Planning Services - Recommendations from CDC and the U.S. Office of Population Affairs, 2017. MMWR Morb Mortal Wkly Rep, 66(50), 1383–1385. doi: 10.15585/mmwr.mm6650a4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Greil AL, Slauson-Blevins KS, Tiemeyer S, McQuillan J, & Shreffler KM (2016). A New Way to Estimate the Potential Unmet Need for Infertility Services Among Women in the United States. Journal of women’s health (2002), 25(2), 133–138. doi: 10.1089/jwh.2015.5390 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaiser Family Foundation. (2022a). Women’s Health Insurance Coverage. Women’s Health Policy, https://www.kff.org/womens-health-policy/fact-sheet/womens-health-insurance-coverage/
- Kaiser Family Foundation. (2022b). Medicaid Income Eligibility Limits for Parents, 2002-2022. State Health Facts. Retrieved from https://www.kff.org/medicaid/state-indicator/medicaid-income-eligibility-limits-for-parents/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D
- Kawwass JF, Penzias AS, & Adashi EY (2021). Fertility-a human right worthy of mandated insurance coverage: the evolution, limitations, and future of access to care. Fertil Steril, 115(1), 29–42. doi: 10.1016/j.fertnstert.2020.09.155 [DOI] [PubMed] [Google Scholar]
- Keiding N, Kvist K, Hartvig H, Tvede M, & Juul S (2002). Estimating time to pregnancy from current durations in a cross-sectional sample. Biostatistics, 3(4), 565–578. doi: 10.1093/biostatistics/3.4.565 [DOI] [PubMed] [Google Scholar]
- Kessler LM, Craig BM, Plosker SM, Reed DR, & Quinn GP (2013). Infertility evaluation and treatment among women in the United States. Fertil Steril, 100(4), 1025–1032. doi: 10.1016/j.fertnstert.2013.05.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Loyola Briceno AC, Ahrens KA, Thoma ME, & Moskosky S (2019). Availability of Services Related to Achieving Pregnancy in U.S. Publicly Funded Family Planning Clinics. Womens Health Issues, 29(6), 447–454. doi: 10.1016/j.whi.2019.07.005 [DOI] [PubMed] [Google Scholar]
- National Center for Health Statistics. (2021). Key Statistics from the National Survey of Family Growth-I Listing Infertility services. Retrieved from https://www.cdc.gov/nchs/nsfg/key_statistics/i-keystat.htm#infertilityservices
- National Center for Health Statistics. (2020a). Public-use data file documentation 2017-2019 National Survey of Family Growth user’s guide. Retrieved from Hyattsville, Maryland: https://www.cdc.gov/nchs/data/nsfg/NSFG-2017-2019-UG-MainText-508.pdf [Google Scholar]
- National Center for Health Statistics. (2020b). 2017-2019 National Survey of Family Growth (NSFG): Weighting Design Documentation. Centers for Disease Control and Prevention. https://www.cdc.gov/nchs/data/nsfg/nsfg-2017-2019-weighting-design-documentation-508.pdf [Google Scholar]
- Perritt J, & Eugene N (2021). Inequity and injustice: recognizing infertility as a reproductive justice issue. F&S Reports. 10.1016/j.xfre.2021.08.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schmidt L (2005). Infertility Insurance Mandates and Fertility. Am Econ Rev, 95(2), 204–208. doi: 10.1257/000282805774670086 [DOI] [PubMed] [Google Scholar]
- Stephen EH, & Chandra A (2000). Use of infertility services in the United States: 1995. Fam Plann Perspect, 32(3), 132–137. [PubMed] [Google Scholar]
- Sunderam S, Kissin DM, Zhang Y, Jewett A, Boulet SL, Warner L, … Barfield WD (2022). Assisted Reproductive Technology Surveillance - United States, 2018. MMWR Surveill Summ, 71(4), 1–19. doi: 10.15585/mmwr.ss7104a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thakker S, Persily J, Voigt P, Blakemore J, Licciardi F, & Najari BB (2021). Evaluating the unevaluated: a secondary analysis of the National Survey for Family Growth (NSFG) examining infertile women who did not access care. J Assist Reprod Genet, 38(5), 1071–1076. doi: 10.1007/s10815-021-02149-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- The Ethics Committee of the American Society for Reproductive Medicine. (2021). Disparities in access to effective treatment for infertility in the United States: an Ethics Committee opinion. Fertil Steril, 116(1), 54–63. doi: 10.1016/j.fertnstert.2021.02.019 [DOI] [PubMed] [Google Scholar]
- Thoma ME, McLain AC, Louis JF, King RB, Trumble AC, Sundaram R, & Buck Louis GM (2013). Prevalence of infertility in the United States as estimated by the current duration approach and a traditional constructed approach. Fertil Steril, 99(5), 1324–1331.e1321. doi: 10.1016/j.fertnstert.2012.11.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weigel G, Ranji U, Long M, & Salganicoff A (2020). Coverage and use of fertility services in the US. Retrieved from https://www.kff.org/womens-health-policy/issue-brief/coverage-and-use-of-fertility-services-in-the-u-s/
- Wright VC, Chang J, Jeng G, & Macaluso M (2008). Assisted reproductive technology surveillance--United States, 2005. MMWR Surveill Summ, 57(5), 1–23. [PubMed] [Google Scholar]
- Wu AK, Odisho AY, Washington SL 3rd, Katz PP, & Smith JF (2014). Out-of-pocket fertility patient expense: data from a multicenter prospective infertility cohort. J Urol, 191(2), 427–432. doi: 10.1016/j.juro.2013.08.083 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.