L26/P-369 Evaluating the landscape and themes of endometriosis randomised controlled trials, 1990-2024, a Meta-Epidemiological Study

Abstract Study question What are the primary foci of our published Endometriosis Randomised Controlled Trial literature and how effectively designed are they to investigate their primary aim? Summary answer Most RCTs focused on medical management of endometriosis, with over half likely unable to detect medication efficacy unless ‘large’ difference between trials groups is present. What is known already Endometriosis presents a significant challenge given its high prevalence and impact on patient quality of life. This is furthered by an ongoing dearth of evidence-based management options due to historical neglect of the condition. The historical limited funding of endometriosis demands that resources be directed towards the most impactful and effective science practices. Achieving this begins with effectively designed Randomised Controlled Trials. Studying the landscape and efficacy of published trials allows for improvement in trial design and optimisation of resources. For example, through tackling underpowered trials found to be widespread across biomedical literature. Study design, size, duration We conducted a meta-epidemiological study of Endometriosis Randomised Controlled Trials published between 1990-2024. We searched Dimensions.AI for articles pertaining to endometriosis and used text-mining to identify randomised controlled trials. We attempted to include all RCT abstracts published in English over this time – resulting in analysis of 540 abstracts. Participants/materials, setting, methods Identified RCTs were then examined for sample sizes, theme and primary outcomes. If full texts were unable to be accessed, this information was ascertained through the abstract. Sample sizes were used to calculate the minimum detectable difference required to achieve significance for studies at 80% and 90% power. We then plotted the distribution of study of publication of trials. Main results and the role of chance We identified 541 RCTs pertaining to endometriosis. We found that 98.0% (N = 529) of RCTs investigated management of endometriosis, with 55.2% (N = 298) of these investigating ‘Medical Management’. 58.7% (N = 175) of medical management RCTs used Gonadotrophin Releasing Hormone (GnRH) analogues in their treatment regime. There was a shift towards ‘Complementary Therapies’ in recent decades, with 28.7% (N = 31) of RCTs categorised under ‘Complementary Therapy’ in 2020-2024, compared to only one of this study type being published in the 1990s. We plotted RCTs by country of first author. The EU led overall with 147 RCTs, Individually, the US and China led with 70 and 67 RCTs respectively, followed by Iran with 34. Of 254 RCTs published from 2010 onwards we identified 60.2% (N = 153 RCTs) which reported registering their trial in a WHO primary or partner registry. Finally, we found that at 80% power, 37.4% of RCTs (N = 201) would not have been able to detect an effect of treatment unless the size of the standard mean difference (SMD) between trial groups was ‘large’ or greater. If aiming for 90% power, 52.3% (n = 281) of RCTs would only be able to detect an SMD ‘large’ or greater. Limitations, reasons for caution Power calculations require estimation of appropriate SMD, which varies by individual trial factors. RCTs requiring large SMDs between groups to achieve significance may be due to underpowering of trials (likely within context of wider literature), or, due to genuine lack of clinically significant results unless SMDs are large. Wider implications of the findings RCTs focusing on medical management are saturated with GnRH analogues, possibly to the detriment of alternatives with lesser adverse effects – potentially explaining increased patient use and research into complementary therapies. Over half of RCTs may miss clinically significant results due to underpowering – a potentially substantive waste of resources. Trial registration number Yes