Objective
In 2013, 1.7% of births in the US were the result of in vitro fertilization (IVF) (1, 2). Identifying children born from infertility treatments using vital records would help clarify the etiology of adverse perinatal outcomes on a population basis (3). Data from the National Survey of Family Growth indicate that infertility services used include ovulation drugs (5.8%), artificial insemination (1.7%), and IVF (0.7%) (4). The 2003 revision of the US Certificate of Live Birth includes three questions regarding the use of infertility treatments: Q1) Pregnancy resulted from infertility treatment; Q2) Fertility-enhancing drugs, artificial insemination (AI) or intrauterine insemination (IUI); Q3) Assisted reproductive technology (e.g., IVF, gamete intrafallopian transfer (GIFT)). As part of a larger study evaluating IVF and the risk of childhood cancer (NIH grant, R01 CA151973), we evaluated the accuracy of infertility treatment and IVF reported on the birth certificate.
Study Design
IVF cycles from the Society for Assisted Reproductive Technology Clinic Online Reporting System (SART CORS), which includes more than 95% of all IVF cycles performed in the US, were linked to certificates of live birth in Florida, Massachusetts, New York, and Pennsylvania (2004-09 births), Texas (2005-09 births), California and Ohio (2006-09 births), and Colorado (2007-09 births) (IVF children). Redshift Technologies, Inc. (who maintains the SART CORS for SART) sent a file to each of the eight study States that included: woman’s name, social security number, date of birth, zip code, date of delivery, plurality, gender(s), and birthweight(s); using these identifiers, the linkage rate was >95%. All other live births to the same woman were also identified (IVF siblings). As part of the primary study, a 10:1 sample of control deliveries (deliveries where the mother was not in the SART CORS database) were selected by the same States and in the same years as the IVF children. Controls were selected as all infants in the next ten deliveries; if this was not possible, they were chosen as a random sample of ten deliveries from the same month and year as the IVF births. Once linked, the data was de-identified before being sent to the investigators. Since not all items were included by each State, we created a summary item: Any infertility question checked ‘Yes’. Information on the birth certificate was evaluated for each of the three groups of children, overall and by plurality (singleton vs multiple birth). IRB approval was obtained from each State and each University of the investigators.
Results
The study population included 716,103 live births (69,969 IVF children, 9,489 IVF siblings, and 636,645 control children). Sensitivity and specificity were calculated to measure the accuracy of IVF use reported on the birth certificate compared to IVF use recorded in the SART CORS. The sensitivity of Q3 was 28.2% and the specificity was 99.7%. Only 36.5% of births of IVF children were identified by any checkbox on the birth certificate; multiple pregnancies were more likely to be indicated as the result of infertility treatment than singletons (43.4% vs 33.3%). If this undercount is applied to the IVF siblings, about one-third of the singletons and nearly all of the multiple births also resulted from some type of infertility treatment.
Conclusions
These results, based on multi-year data from eight States, suggest that infertility treatment and IVF are greatly under-reported on the birth certificate, accurately identifying only about one-third (36.5%) of children conceived with IVF, confirming the percentage reported by prior studies (5, 6). If this under-reporting estimate is applied to the controls, about 1.5% of singletons and about 25% of multiples were conceived with some type of infertility treatment.
Table.
The distribution (%) by study group, plurality, and checkbox item on the birth certificate*
| IVF Children |
IVF Siblings |
Control Children |
||
|---|---|---|---|---|
| Overall | 69,969 | 9,489 | 636,645 | |
| Singleton births | 47,737 | 8,890 | 623,030 | |
| Multiple births | 22,232 | 599 | 13,615 | |
| Q1: Pregnancy resulted from infertility treatment | Overall | 36.8 | 14.1 | 0.8 |
| Singleton births | 33.8 | 12.3 | 0.6 | |
| Multiple births | 43.0 | 38.4 | 8.9 | |
| Q2: Fertility-enhancing drugs, AI, or IUI | Overall | 11.5 | 6.2 | 0.4 |
| Singleton births | 10.1 | 5.2 | 0.3 | |
| Multiple births | 14.4 | 20.6 | 5.3 | |
| Q3: Assisted reproductive technology, IVF, or GIFT | Overall | 28.2 | 5.5 | 0.3 |
| Singleton births | 26.0 | 4.9 | 0.2 | |
| Multiple births | 33.0 | 15.0 | 4.0 | |
| Summary: any infertility item checked ‘Yes’** | Overall | 36.5 | 12.8 | 0.7 |
| Singleton births | 33.3 | 11.1 | 0.5 | |
| Multiple births | 43.4 | 37.1 | 8.7 |
Percents were computed from the years that the item was present; States may not have included all items on the birth certificate.
The response is included in the summary when at least one of the three items was present on the birth certificate.
Acknowledgments
Supported by the National Cancer Institute, National Institutes of Health (grant R01 CA151973). The views expressed in this paper are those of the authors and do not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.
Barbara Luke is a research consultant to the Society for Assisted Reproductive Technologies;
SART wishes to thank all of its members for providing clinical information to the SART CORS database for use by patients and researchers. Without the efforts of our members, this research would not have been possible.
Footnotes
Presented at the 71st Annual Meeting of the American Society for Reproductive Medicine; Baltimore, Maryland; October 17–21, 2015
all other authors report no conflicts of interest.
Contributor Information
Barbara Luke, Department of Obstetrics, Gynecology, and Reproductive Biology, College of Human Medicine, Michigan State University, East Lansing, Michigan 48824.
Morton B. Brown, Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, Michigan 48109.
Logan G. Spector, Department of Pediatrics, University of Minnesota, Minneapolis, Minnesota 55455.
References
- 1.Martin JA, Hamilton BE, Osterman MJK, Curtin SC, Mathews TJ. National Vital Statistics Reports. 1. Vol. 64. Hyattsville, MD: National Center for Health Statistics; 2015. Births: Final Data for 2013. [PubMed] [Google Scholar]
- 2.Centers for Disease Control and Prevention. 2013 Assisted Reproductive Technology National Summary Report. Atlanta (GA): US Department of Health and Human Services; 2015. American Society for Reproductive Medicine, Society for Assisted Reproductive Technology. [Google Scholar]
- 3.Sutcliffe AG, Ludwig M. Outcome of assisted reproduction. Lancet. 2007;370:351–359. doi: 10.1016/S0140-6736(07)60456-5. [DOI] [PubMed] [Google Scholar]
- 4.Chandra A, Copen CE, Stephen EH. National Health Statistics Reports. No. 73. Hyattsville, MD: National Center for Health Statistics; 2014. Infertility service use in the United States: Data from the National Survey of Family Growth, 1982–2010. [PubMed] [Google Scholar]
- 5.Zhang Z, Macaluso M, Cohen B, Schieve L, Nannini A, Chen M, Wright V. Accuracy of assisted reproductive technology information on the Massachusetts birth certificate, 1997–2000. Fertility and Sterility. 2010;94:1657–1661. doi: 10.1016/j.fertnstert.2009.10.059. [DOI] [PubMed] [Google Scholar]
- 6.Cohen B, Bernson D, Sappenfield W, Kirby RS, Kissin D, Zhang Y, Copeland G, Zhang Z, Macaluso M. Accuracy of assisted reproductive technology information on birth certificates: Florida and Massachusetts, 2004-06. Paediatric and Perinatal Epidemiology. 2014;28:181–190. doi: 10.1111/ppe.12110. [DOI] [PubMed] [Google Scholar]