Customize your JAMA Network experience by selecting one or more topics from the list below.
Dupree JM, Levinson Z, Kelley AS, et al. Provision of Insurance Coverage for IVF by a Large Employer and Changes in IVF Rates Among Health Plan Enrollees. JAMA. 2019;322(19):1920–1921. doi:10.1001/jama.2019.16055
In 2010, 12.1% of US couples reported problems becoming pregnant or carrying a pregnancy to term,1 and many required in vitro fertilization (IVF) to conceive. However, IVF is expensive2 and typically not covered by insurance.3 Some employers now offer insurance plans with IVF coverage, but the association of employer-sponsored coverage with use of IVF has not been studied.
Beginning January 1, 2015, self-insured plans at the University of Michigan provided IVF coverage for women aged 42 years or younger with an infertility diagnosis, including a 20% coinsurance and requirement of single-embryo transfer for women younger than 35 years. Claims data from October 2012 through January 2017 were obtained for women aged 22 to 42 years enrolled in the self-insured plans. Because IVF claims were not available before coverage began, a claims-based algorithm was used to identify IVF cycles before and after January 1, 2015, by leveraging specific patterns of covered fertility medications and paired pelvic ultrasounds associated with oocyte retrievals. The algorithm was validated with primary chart review compared with IVF claims after January 1, 2015, with a sensitivity of 93.0% and a specificity greater than 99.9%.
The primary outcome was IVF rate, measured as oocyte retrievals per 10 000 women aged 22 to 42 years. Rates of IVF before and after the benefit change and absolute and proportional IVF rate changes were calculated for all women and by employee salary, employee position, age, and race.4 First, a single multivariable logit regression model with robust standard errors clustered at the individual level was estimated. Second, average monthly predicted probabilities were generated after assigning each individual in the study to the relevant subgroup variable (eg, high-salary group) and time period (eg, 2013-2014). Third, the monthly probabilities were scaled to annual IVF rates. We excluded 15 581 women with missing data on explanatory variables; the results were consistent when a “missing” group was added to the model. Two-sided significance tests were conducted with a threshold of P < .05, using Stata version 15.1 (StataCorp). The University of Michigan Medical School Institutional Review Board approved a waiver of consent.
There were 18 282 women included (mean age, 32 years; 70% white; 57% regular staff). Regression-adjusted IVF use increased among all women from 34.3 cycles per 10 000 women before 2015 to 92.6 cycles afterward—a ratio of 2.7 (95% CI, 1.9-3.8; P < .001) (Table). Use of IVF remained highest among women in the high-salary group (132.3 cycles per 10 000 women in 2015-2016). However, women in the low-salary group had the largest proportional increase in IVF use, from 8.2 to 78.8 cycles per 10 000 (ratio, 9.6; 95% CI, 4.3-21.6; P < .001), which was statistically significantly higher than the proportional increase for the high-salary group. Women in all employee positions experienced increases in absolute and proportional IVF use after the benefit change, but the increases were not statistically significantly different. Women aged 30 to 34 years and 35 to 42 years and white, black, and Asian/Pacific Islander women had statistically significant increases in absolute IVF use after the benefit change; however, only women aged 30 to 34 years had a proportional increase in IVF use that was statistically significantly different from their reference group (Table).
Employer-sponsored IVF coverage was associated with increased use of IVF among all women, with a large proportional increase among low-salary women. However, absolute IVF use remained highest among high-salary women, which may suggest that the 20% coinsurance discouraged use for some women. Limitations include data from a single employer and no control population to evaluate for causality or secular trends. Additionally, employee salary may not reflect total household income, and data are not available on pregnancy or birth outcomes or total health plan costs associated with the benefit change. Further research is needed for a comprehensive understanding of the effect of employer-sponsored IVF benefits and to complement studies demonstrating improvement in IVF outcomes with state-level coverage mandates.5
Accepted for Publication: September 12, 2019.
Corresponding Author: James M. Dupree, MD, MPH, Departments of Urology and Obstetrics and Gynecology, University of Michigan, 2800 Plymouth Rd, NCRC, Bldg 16, 100S-20, Ann Arbor, MI 48109 (email@example.com).
Published Online: November 13, 2019. doi:10.1001/jama.2019.16055
Author Contributions: Dr Dupree and Mr Levinson had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Dupree, Levinson, Dalton, Levy, Hirth.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Dupree, Levinson, Hirth.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Dupree, Levinson, Dalton, Hirth.
Obtained funding: Dupree, Dalton, Hirth.
Administrative, technical, or material support: Dupree, Manning, Levy.
Conflict of Interest Disclosures: Dr Dupree reported receiving grant funding from Blue Cross Blue Shield of Michigan for quality improvement work unrelated to this research. Dr Dalton reported receiving consulting fees from Bayer; receiving grant funding from the Agency for Healthcare Research and Quality, the Blue Cross Blue Shield of Michigan Foundation, the Arnold Foundation, an anonymous donor, and ECOG-ACRIN; and being a contributing editor for The Medical Letter. No other disclosures were reported.
Funding/Support: This work was partially supported by a grant from the University of Michigan’s Mcubed research stimulator program.
Role of the Funder/Sponsor: The funder 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; or decision to submit the manuscript for publication.
Additional Contributions: We thank Edward Norton, PhD, University of Michigan School of Public Health, for statistical consulting and Michael Lanham, MD, Department of Obstetrics and Gynecology, University of Michigan Medical School, for clinical consulting. Drs Norton and Lanham received no compensation for their roles in the study.
Create a personal account or sign in to: