Utility of a commercial claims database for studying in-vitro fertilization (ivf) covered by health insurance.

More employers are providing insurance coverage for IVF, but existing national IVF registries do not distinguish between covered and out-of-pocket IVF cycles. Our objective was to evaluate the suitability of using Clinformatics, a national database of over 80 million commercially insured individuals, for research about insurance coverage for IVF and clinical outcomes. We obtained insurance claims from Optum’s de-identified Clinformatics® Data Mart Database from 2014 to 2019. For each year, IVF cycles were identified using the Current Procedural Terminology (CPT) code for oocyte retrieval among women aged 20-44. Oocyte retrievals performed for fertility preservation were excluded. Pregnancies and live births were identified using relevant CPT, International Classification of Diseases (ICD), and Diagnosis Related Group (DRG) codes. We used cause-specific hazard models to estimate the annual rates and 95% confidence intervals of pregnancies and live births after first embryo transfers and the mean number of embryo transfer per oocyte retrieval. For each year, we compared results from Clinformatics to the analogous results reported by SART. From 2014 to 2019, the annual number of oocyte retrievals in Clinformatics ranged from 3,272 to 3,475. During this time period, there were no statistical differences in the pregnancy and live birth rates found in Clinformatics and the pregnancy rates and live birth rates reported by SART (Table). There was also no statistical difference in the mean number of embryo transfers per retrieval each year in Clinformatics and SART. In Clinformatics claims data, pregnancies and live births after IVF can be identified accurately using ICD, CPT, and DRG codes. The clinical outcomes after IVF in Clinformatics are nearly identical to those reported by SART. The Clinformatics database is a suitable data source for assessing clinical outcomes after IVF.