Customize your JAMA Network experience by selecting one or more topics from the list below.
Jungheim ES, Leung MY, Macones GA, Odem RR, Pollack LM, Hamilton BH. In Vitro Fertilization Insurance Coverage and Chances of a Live Birth. JAMA. 2017;317(12):1273–1275. doi:10.1001/jama.2017.0727
Because in vitro fertilization (IVF) is expensive and often cost-prohibitive, some states mandate IVF insurance coverage.1 Previous analyses of aggregate, state-level data suggest these mandates are associated with increased IVF utilization but lower live birth rates per IVF cycle.2 These analyses did not account for important confounders or for the fact that women may require multiple IVF cycles to achieve a live birth.3 This study compared the cumulative probability of live birth among women with and without IVF insurance coverage at the Fertility and Reproductive Medicine Center at Washington University—a center located near the border between Illinois, which mandates IVF coverage, and Missouri, which does not.4
This retrospective study and waiver of consent was approved by the Washington University Human Research Protection Office. Women initiating IVF from 2001 through 2010 were included and observed through 2014. Women residing more than 160 kilometers from the center or using donor oocytes or gestational carriers were excluded. Data extracted from medical and billing records are listed in Table 1. The primary outcome was cumulative probability of live birth according to IVF insurance status over 4 IVF cycles, calculated as a function of live birth rates and return probabilities. Four IVF cycles are covered by the Illinois mandate. Cycles were defined as controlled ovarian hyperstimulation started with intent of fresh or subsequent frozen embryo transfer. To account for differences in patient characteristics by insurance status, logistic regression was used to estimate risk-adjusted live birth probabilities after the first, second, third, and fourth cycles and risk-adjusted return probabilities if a woman was unsuccessful after a given cycle. Covariates included IVF cycle number, interactions between insurance status and IVF cycle number, age at oocyte retrieval, race, prior live birth, income, fresh vs frozen embryo transfer, and linear time trend. For fresh embryo transfers, covariates also included antral follicle count, total amount of gonadotropin used, peak estradiol, and number of oocytes retrieved. Return probabilities included the above covariates, cycle cancellation, and the woman’s distance from the clinic by residence zip code. Women were defined as returning if they did so within 365 days of an unsuccessful cycle. Standard errors were bootstrapped using 1000 replications.5 Two-sided tests of significance were conducted with a threshold significance level of .05. Analyses were performed in STATA (StataCorp), version 13.1.
Of the 1572 women in the sample (230 excluded), 875 (55.7%) had IVF insurance coverage (40% mandated, 60% nonmandated) and 697 (44.3%) were self-pay. The 2 groups did not differ medically, but patients with coverage were younger (Table 1). IVF coverage status was not associated with probability of live birth in individual cycles (Table 2). However, the proportion returning for a second cycle if unsuccessful in the first cycle was 0.703 among women with coverage compared with 0.516 among self-paying women (difference, 0.187 [95% CI, 0.127-0.248]; P < .001) (Table 2). The mean cumulative live birth probability after 4 cycles for women with coverage 0.585, was significantly higher than that for self-paying women, 0.505 (difference, 0.081 [95% CI, 0.030-0.131]; P = .001). The difference in cumulative live birth rates adjusting for patient risk factors between insured and self-pay patients after 4 cycles narrowed to 0.054, but was still significant (95% CI, 0.008-0.099; P = .01).
Women with insurance coverage for IVF were more likely to attempt IVF again, and they had a higher cumulative probability of live birth than women who self-paid for IVF. This study was limited because the data were obtained from 1 center and the findings may not apply to other centers. Also, out-of-pocket costs and information about women who may have returned elsewhere after a failed cycle were not available, but access to patient-level data allowed control for other important confounders. These findings demonstrate legislation mandating IVF insurance coverage may improve the delivery and outcomes of fertility treatments.
Corresponding Author: Emily S. Jungheim, MD, MSCI, 4444 Forest Park, Ste 3100, St Louis, MO 63108 (email@example.com).
Author Contributions: Dr Jungheim had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Jungheim, Macones, Odem, Pollack, Hamilton.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Jungheim, Leung, Pollack, Hamilton.
Critical revision of the manuscript for important intellectual content: Jungheim, Leung, Macones, Odem.
Statistical analysis: Jungheim, Leung, Hamilton.
Obtained funding: Jungheim.
Administrative, technical, or material support: Jungheim, Macones, Odem.
Supervision: Jungheim, Macones, Hamilton.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This work was supported by grant NIH K12HD063086 from the Women’s Reproductive Health Research Program of the National Institutes of Health (Dr Jungheim).
Role of the Funder/Sponsor: The National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: