Improving IVF Utilization with Patient-Centric Artificial Intelligence-Machine Learning (AI/ML): A Retrospective Multicenter Experience

Abstract

Objectives: In vitro fertilization (IVF) has the potential to give babies to millions more people globally, yet it continues to be underutilized. We established a globally applicable and locally adaptable IVF prognostics report and framework to support patient-provider counseling and enable validated, data-driven treatment decisions. This study investigates the IVF utilization rates associated with the usage of machine learning, center-specific (MLCS) prognostic reports (the Univfy^® report) in provider-patient pre-treatment and IVF counseling. Methods: We used a retrospective cohort comprising 24,238 patients with new patient visits (NPV) from 2016 to 2022 across seven fertility centers in 17 locations in seven US states and Ontario, Canada. We tested the association of Univfy report usage and first intra-uterine insemination (IUI) and/or first IVF usage (a.k.a. conversion) within 180 days, 360 days, and "Ever" of NPV as primary outcomes. Results: Univfy report usage was associated with higher direct IVF conversion (without prior IUI), with odds ratios (OR) 3.13 (95% CI 2.83, 3.46), 2.89 (95% CI 2.63, 3.17), and 2.04 (95% CI 1.90, 2.20) and total IVF conversion (with or without prior IUI), OR 3.41 (95% CI 3.09, 3.75), 3.81 (95% CI 3.49, 4.16), and 2.78 (95% CI 2.59, 2.98) in 180-day, 360-day, and Ever analyses, respectively; p < 0.05. Among patients with Univfy report usage, after accounting for center as a factor, older age was a small yet independent predictor of IVF conversion. Conclusions: Usage of a patient-centric, MLCS-based prognostics report was associated with increased IVF conversion among new fertility patients. Further research to study factors influencing treatment decision making and real-world optimization of patient-centric workflows utilizing the MLCS reports is warranted.

Keywords: AI/ML; IVF; IVF access; IVF conversion; IVF outcomes; IVF utilization; artificial intelligence (AI); fertility; fertility access; live birth outcomes; machine learning (ML); prediction model; prognostics tool.

Figure 1

Summary of raw data sources, data processing steps, and the resulting processed and merged data for analysis.

Figure 2

Event History mapping for 3 timed analyses: (A) 180 days post-NPV (180D), 360 days post-NPV (360D), and “Ever”. (B) Examples showing the mapping of Event History to Patient Groups and Event Groups as per Methods 2.4.

Figure 3

Patient care navigation to treatment conversion by Patient Groups and Event Groups. (A) Retrospective patient cohort aggregated from 7 centers and their Event Groups (bolded) based on the Ever timed analysis: new patient visit (NPV)-IUI conversion (IUI Group), NPV-IVF conversion (Direct IVF Group), usage of 1st IVF cycle with or without one or more preceding IUIs (Total IVF Group), and “No IUI or IVF” Group. (B) The patient cohort from (A) separated into 2 Patient Groups—the Univfy Group (green) and the No Univfy Group (gray).

Figure 4

Comparison of IUI, Direct IVF, and Total IVF conversion rates between Univfy Group and No Univfy Group in 3 timed analyses. (A) The conversion rates across 7 centers were compared between Univfy Group and No Univfy Group for IUI, Direct IVF, and Total IVF conversions in each timed analyses, 180D, 360D, and Ever, using Wilcoxon signed-rank test. * p < 0.05. (B) The per-center IUI, Direct IVF, and Total IVF conversion rates are compared using chi-square tests * p < 0.001, NS not significant.

Figure 4

Comparison of IUI, Direct IVF, and Total IVF conversion rates between Univfy Group and No Univfy Group in 3 timed analyses. (A) The conversion rates across 7 centers were compared between Univfy Group and No Univfy Group for IUI, Direct IVF, and Total IVF conversions in each timed analyses, 180D, 360D, and Ever, using Wilcoxon signed-rank test. * p < 0.05. (B) The per-center IUI, Direct IVF, and Total IVF conversion rates are compared using chi-square tests * p < 0.001, NS not significant.