Evaluating the efficacy, safety, and clinical effectiveness of IVF add-ons: Methodological challenges and future solutions

Abstract Assisted reproductive technology (ART) has expanded rapidly into a complex, highly regulated, and innovative field, with in vitro fertilisation (IVF) now accounting for millions of treatment cycles globally each year. Alongside these advances, numerous supplementary interventions, commonly referred to as "IVF add-ons," have been introduced into routine clinical practice with the aim of improving pregnancy or live birth rates, reducing miscarriage risk, or shortening time to conception. Despite their widespread adoption and substantial additional costs to patients, most IVF add-ons lack robust evidence of safety, efficacy, and cost-effectiveness. Regulatory and policy efforts to guide their use are constrained by significant methodological weaknesses in the existing evidence base, including heterogeneous definitions, suboptimal trial design, inconsistent outcome reporting, and limited translation of research findings into clinical practice. This article explores the principal methodological challenges that currently impede rigorous health technology assessment of IVF add-ons. These challenges include the absence of a clear, validated taxonomy to define and classify add-ons; lack of consensus on appropriate comparators and clinically meaningful outcomes; and failure to establish agreed thresholds for clinical utility and futility that incorporate economic considerations and patient perspectives. A major limitation arises from reliance on conventional parallel-group randomised controlled trials, which are often poorly suited to evaluating complex, multi-stage ART interventions in heterogeneous populations. We discuss the potential value of innovative trial designs—such as platform, basket, sequential multiple assignment randomised trials, hybrid pragmatic–explanatory approaches, and decentralised digital trials—to strengthen evidence generation. Collectively, these methods may enhance efficiency, improve interpretability, and better align research with real-world reproductive care.