Methodological Integrity of Randomized Controlled Trials in Major Gynaecological Conditions: A Systematic Review

Methodological Integrity of Randomized Controlled Trials in Major Gynaecological Conditions: A Systematic Review | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Methodological Integrity of Randomized Controlled Trials in Major Gynaecological Conditions: A Systematic Review Neha Raghuraman, Nimesha Wijamuni, Sepide Ahmadi, Vindya Pathiraja, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9212126/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Randomized controlled trials (RCTs) provide the evidentiary foundation for clinical decision-making in obstetrics and gynaecology. However, concerns remain regarding methodological rigor, reporting transparency, and structural biases within women’s health research. Despite the global burden of gynaecological conditions such as endometriosis, polycystic ovary syndrome (PCOS), gestational diabetes mellitus (GDM), and premenstrual dysphoric disorder (PMDD), the methodological integrity of trials evaluating these conditions has not been comprehensively assessed. Methods This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (PRISMA) and prospectively registered in PROSPERO (CRD42025636543). Electronic searches were performed in PubMed, ScienceDirect, and Google Scholar to identify RCTs published between January 2005 and October 2025 evaluating interventions for endometriosis, PCOS, GDM, or PMDD. Eligible studies included randomized or cluster-randomized trials involving adult women. Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool. Reporting completeness was evaluated using the Consolidated Standards of Reporting Trials (CONSORT) 2010 checklist. Associations between trial-level characteristics and methodological quality were examined using logistic and linear regression analyses. Results A total of 695 RCTs were included in the review. Overall methodological quality was limited, with only 49 trials (7.0%) judged to be at low risk of bias. The most frequent concerns were related to outcome measurement and selective reporting. Sample sizes varied widely across conditions, with smaller trials predominating in PMDD and endometriosis research, while GDM trials tended to be larger. Prospectively registered trials demonstrated significantly higher reporting completeness, with an average increase of 8.6 percentage points in CONSORT adherence (p < 0.001), and were more likely to be classified as low risk of bias (OR 4.08, 95% CI 1.59–10.47). Reporting quality improved modestly over time but remained uneven across conditions and country income levels. Conclusions Trials with methodological limitations and variable reporting quality dominate the clinical evidence base for major gynaecological conditions. Prospective trial registration and transparency mechanisms were strongly associated with improved methodological standards. Strengthening trial design, enforcing reporting guidelines, and improving global equity in research infrastructure are essential to ensure that evidence guiding women’s health care is robust, transparent, and reliable. Systematic review Meta-research Research methodology Randomized controlled trials Risk of bias CONSORT Endometriosis Polycystic ovary syndrome Gestational diabetes mellitus Premenstrual dysphoric disorder Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 BACKGROUND Women’s Health, Structural Inequities, and the Reliability of Evidence Women’s health encompasses conditions that are unique to women, more prevalent in women, or that manifest differently across the life course( 1 ). These include gynaecological and reproductive disorders such as premenstrual dysphoric disorder (PMDD), gestational diabetes mellitus (GDM), endometriosis, and polycystic ovary syndrome (PCOS). They also cover more general issues related to physical and mental health throughout life ( 2 , 3 ). Obstetrics and gynaecology (OBGYN) is essential to addressing these medical requirements. It addresses menstrual diseases, fertility, menopause, pregnancy, and childbirth, and reproductive health ( 4 , 5 ). Many female-specific disorders remain underdiagnosed, undertreated, and underresearched despite the substantial global effect of gynaecological diseases ( 6 ). Endometriosis diagnosis delays, uneven PCOS treatment options, and heterogeneous GDM and PMDD care standards demonstrate persistent gaps in evidence-based clinical practice ( 7 – 10 ). Endometriosis affects an estimated 10% of women of reproductive age worldwide, representing approximately 190 million individuals globally ( 11 ). PCOS is similarly prevalent, affecting between 6–13% of women, depending on diagnostic criteria ( 12 ). GDM occurs in approximately 1–14% of pregnancies globally ( 13 ), while PMDD affects an estimated 3–8% of menstruating individuals ( 14 ). Women's health outcomes are still impacted by structural barriers in several contexts. Even though women are more likely than men to experience adverse medication responses, they are still underrepresented in clinical studies. There is inconsistent enforcement of policies requiring representative involvement in research. Furthermore, healthcare policy, training, and data systems frequently fail to appropriately incorporate sex- and gender-specific characteristics ( 15 ). These differences have obvious repercussions. Hundreds of thousands of pregnant women in Europe do not receive adequate care. Every day, over 800 women worldwide pass away from avoidable pregnancy and childbirth-related causes ( 16 ). Maternal mortality rates are greater among mothers from ethnic minorities and migrants. This circumstance demonstrates deep structural injustices ( 15 ). Maternal mortality and national income have a strong negative correlation, according to these global disparities. Maternal mortality rates are significantly higher in low-income nations. This is illustrated in Fig. 1. Figure 1. Maternal mortality ratio in relation to GDP per capita (2020) Maternal mortality ratio (deaths per 100,000 live births) plotted against GDP per capita (international dollars, purchasing power parity), both shown on logarithmic scales. The figure illustrates a strong inverse relationship between national income and maternal mortality, with the highest mortality ratios concentrated in low-income countries, particularly in sub-Saharan Africa. Data sources: UN Maternal Mortality Estimation Inter-Agency Group (MMEIG), World Bank, International Monetary Fund (IMF), OECD, and Eurostat. Figure reproduced from Our World in Data. While maternal mortality is a distinct clinical domain, it serves as a proxy for broader systemic inequalities ( 17 ). One of the main causes of teenage girls missing school is menstrual issues. This circumstance reinforces gender disparities in long-term economic stability and education. An increased risk of chronic illness in later life is also associated with reproductive health problems. Many of these illnesses, nonetheless, are mislabeled or remain untreated. This fact makes it difficult to monitor and plan interventions effectively ( 18 ). Women's health requires immediate and continuous care. Not only do women make up the majority of healthcare providers and consumers, but healthy women also translate into healthier families, communities, and economies, making this crucial. Reducing care gaps, saving lives, and promoting a healthier, more equitable future for all will result from addressing the full spectrum of women's health needs ( 19 , 20 ). Evolution of Women’s Health Research and Trial Exclusion Historically, women’s health research has been characterised by both neglect and exclusion. For much of modern medical history, women were underrepresented in clinical trials due to concerns regarding hormonal variability, reproductive capacity, and pregnancy-related risk( 21 ). New laws and institutional changes brought about a significant change in the 1990s. Among these were the establishment of the NIH Office of Research on Women's Health and the enactment of the NIH Revitalization Act. Research on women's health has advanced thanks to major projects like the Women's Health Initiative (WHI) ( 22 , 23 ). Still, persistent gaps remain in gynaecological and obstetric research. Pregnant individuals continue to be excluded from numerous interventional studies, and marginalised populations, including women from low and middle-income countries, ethnic minority communities, and socioeconomically disadvantaged groups, remain underrepresented in trial populations ( 24 – 28 ). These structural exclusions carry scientific consequences. When trial populations do not reflect the demographic and socioeconomic diversity of those most affected by the disease, external validity is compromised ( 29 ). Methodological Complexity in Obstetrics and Gynaecology Trials Randomized controlled trials are considered the gold standard for evaluating healthcare interventions ( 30 ). However, trial design in obstetrics and gynaecology presents unique methodological challenges. Many women's health issues, including menstruation, dysmenorrhea, menopause, incontinence, and sexual health, are still impacted by social stigma. These taboos limit open communication, restrict women from seeking medical assistance, and result in inadequate institutional attention and financing for research ( 15 , 16 ). As a result, these conditions are often left out of policy, education, and research agendas. Before 1993, women were not completely missing from clinical research, but they were often underrepresented ( 31 ). As seen during the COVID-19 vaccine launch, the lack of trial data has impacted clinical decision-making, especially in pregnancy ( 32 , 33 ). Women's involvement in trials is still restricted by other obstacles, including caregiving duties, mistrust of medical institutions, and practical limitations ( 34 ). These difficulties are made worse by inadequate statistics and reporting. The therapeutic significance of study findings is frequently diminished by small sample sizes, poor treatment of missing data, and reliance on surrogate outcomes. Commonly used outcomes in obstetric trials, such as those assessing labor induction, might not accurately represent significant patient experiences, which would limit their practical application. For example, the commonly used endpoint of a "live birth," which is typically defined as delivery within 24 hours, does not sufficiently convey the complex nature of the induction process or the experiences during labor and the time preceding delivery ( 35 ). Recent research indicates that, particularly in situations like PCOS, gynaecological studies may be especially susceptible to biased outcome reporting, poor handling of missing data, inadequate blinding, and an excessive dependence on surrogate biochemical endpoints. Moreover, outcome heterogeneity is still widespread, which hinders meaningful meta-analysis and cross-trial comparability ( 36 , 37 ). Although initiatives such as the Core Outcomes in Women’s Health (CROWN) project have sought to standardise outcome reporting ( 38 ), adoption has been inconsistent and poorly quantified across conditions. The Need for a Structural Audit of Trial Methodology Most systematic reviews in women’s health focus on intervention effectiveness rather than interrogating the methodological architecture of the trials themselves. Yet, the reliability of pooled estimates and subsequent guideline recommendations depends fundamentally on trial integrity. Evidence suggests that the methodological quality of both systematic reviews and underlying clinical trials in women’s health is often suboptimal, with frequent high risks of bias in key domains such as blinding and allocation concealment ( 39 ). This is consequential because empirical meta-epidemiological research demonstrates that such methodological shortcomings can systematically exaggerate intervention effect estimates and increase between-study heterogeneity ( 40 ). Although systematic reviews and meta-analyses are positioned at the apex of the evidence hierarchy, their conclusions are inherently constrained by the quality of the included studies, meaning that biased primary data can propagate uncertainty rather than resolve it ( 41 ). In addition, given the global disparities illustrated by income-linked health outcomes (Fig. 1), it is plausible that trial-level characteristics, including country income classification, registration practices, and publication era, may be associated with measurable differences in methodological quality. Understanding these patterns is essential for identifying structural weaknesses within the evidence base and informing strategies to strengthen future research design. Study Objectives In alignment with the PROSPERO registered review objectives, the present study aimed to systematically characterise the clinical trial methodologies used in randomized controlled trials evaluating interventions for endometriosis, PCOS, GDM, and PMDD. Drawing on 695 Randomized controlled (RCTs) published between 2005 and 2025, this review evaluates risk-of-bias profiles, reporting completeness, and trial-level predictors of methodological quality. By providing a large-scale structural assessment of trial integrity in contemporary obstetric and gynaecological research, this study seeks to inform improvements in rigor, transparency, and equity within the women’s health evidence ecosystem. METHODS Study Design and Reporting Standards This study was conducted as a systematic review of RCTs evaluating methodological characteristics in obstetric and gynaecological research. The review followed a pre-specified protocol and was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines ( 42 ). The PRISMA 2020 checklist is provided in Additional file 5. The review was prospectively registered in PROSPERO (CRD42025636543) ( 43 ). Eligibility Criteria Eligible studies were randomized controlled trials published between 1 January 2005 and 1 October 2025. The timeframe was selected to reflect contemporary methodological standards, including widespread implementation of trial registration and reporting guidelines ( 44 ). Studies were included if they evaluated interventions for one of four prespecified conditions: endometriosis, PCOS, GDM, or PMDD. Trials enrolling adult women or individuals with a uterus aged 18 years or older were eligible. Parallel-group, cluster-randomized, and crossover designs were included. Non-randomized studies, observational designs, conference abstracts, study protocols without results, retracted publications, and animal studies were excluded. Secondary analyses of previously published trials were excluded to ensure that assessed methodological characteristics reflected the primary trial design and conduct. Only peer-reviewed articles published in English were included. Information Sources and Search Strategy Comprehensive searches were conducted in PubMed, ScienceDirect, and Google Scholar. Search strategies combined controlled vocabulary terms and free-text keywords relating to the four target conditions and a randomized controlled trial design. The databases PubMed and ScienceDirect were last searched on 20 October 2025, and Google Scholar was last searched on 2 November 2025. Boolean operators were applied to optimise sensitivity. The PubMed search combined terms for endometriosis, PCOS, GDM, and PMDD with filters for randomized controlled trials. Searches were restricted to studies published between 1 January 2005 and 1 October 2025 and limited to English-language publications involving human adults. The full PubMed search strategy is provided in Additional file 1. The ScienceDirect search strategy was adapted from the PubMed query using equivalent keywords and Boolean operators: (“endometriosis” OR “PCOS” OR “gestational diabetes” OR “PMDD”) AND (“randomized controlled trial” OR “RCT”). Google Scholar searches were conducted using simplified keyword combinations: (“endometriosis” OR “PCOS” OR “gestational diabetes” OR “PMDD”) AND (“randomized controlled trial” OR “RCT”). Due to the large volume of results and limitations in reproducibility, screening was restricted to the first 500 results sorted by relevance, consistent with methodological recommendations. Reference lists of included studies were screened to identify additional eligible trials. Study Selection Study selection was conducted independently by three reviewers at both title/abstract and full-text screening stages. Discrepancies were resolved through discussion and consensus. All retrieved records were imported into Covidence software for screening and duplicate removal. Covidence automation tools were used to assist in identifying records that were likely ineligible based on predefined exclusion criteria (e.g., non-randomized designs, irrelevant populations). These records were reviewed to confirm exclusion, ensuring that no studies were excluded without human assessment. Screening was conducted in two stages: title and abstract screening followed by full-text review. Eligibility criteria were applied consistently at both stages using a pre-piloted screening framework. The study selection process was documented using a PRISMA flow diagram. A total of 695 trials were included in the final synthesis. Data Extraction A structured data extraction template was developed and piloted before full extraction. Extracted variables included publication year, trial registration status, funding source, country of conduct, World Bank income classification, sample size, trial design type, centre status (single vs multicentre), randomisation method, allocation concealment, blinding practices, outcome domains, intention-to-treat reporting, and handling of missing data. The primary outcomes of this review were methodological quality, assessed using the Cochrane Risk of Bias 2 (RoB 2) tool, and reporting quality, assessed using CONSORT 2010 adherence. Secondary outcomes included associations between trial-level characteristics (e.g., trial registration, publication period, and country income level) and these methodological outcomes. Outcome domains reported within included trials were categorised as patient-reported, clinical/physiological, biochemical, reproductive, or safety outcomes. Where multiple outcome measures or time points were reported within a domain, all relevant data were extracted. No selection based on statistical significance was performed. Equity-related variables were extracted but analysed narratively due to inconsistent reporting. Where data were missing or unclear, variables were coded as ‘not reported’, and no imputation was performed. Extracted variables were coded into predefined categories, with ‘not reported’ treated as a separate category. Trial registration status was categorised as prospectively registered versus not registered or retrospectively registered. Publication year was grouped into predefined periods (2005–2009, 2010–2014, 2015–2019, 2020–2025). Risk-of-bias judgments were dichotomised (low vs not low) for primary analyses, with alternative dichotomisation (high vs not high) used in sensitivity analyses. CONSORT checklist items were coded as binary variables and aggregated into percentage adherence scores. Data extraction was conducted independently by two reviewers, with discrepancies resolved through discussion and, when necessary, adjudication by a third reviewer. Risk of Bias Assessment Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool ( 45 ). Risk of bias assessment was conducted independently by two reviewers, with disagreements resolved through discussion or consultation with a third reviewer. The tool evaluates five domains: bias arising from the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data, bias in the measurement of outcomes, and bias in the selection of the reported result. Each domain was classified as low risk, some concerns, or high risk, resulting in an overall risk of bias judgment. For regression analyses, the overall risk of bias was dichotomised as low versus not low risk, consistent with methodological guidance when event counts are limited. Reporting Quality Assessment Reporting completeness was evaluated using the Consolidated Standards of Reporting Trials (CONSORT) 2010 checklist ( 46 ). CONSORT checklist items were operationalised as binary variables (reported vs not reported/unclear) and aggregated into a percentage adherence score, consistent with prior methodological studies evaluating reporting quality. Equal weighting was applied across items to provide an overall measure of reporting completeness. Sub-items were evaluated separately to allow granular assessment of reporting deficiencies. A total CONSORT score (maximum 37 items) and percentage adherence score were calculated for each trial. Percentage adherence was used as the primary reporting quality metric. Each study was evaluated across relevant checklist domains to generate an overall adherence score; however, item-level scoring data were not retained for all studies. Items deemed not applicable to specific study designs were excluded from scoring for that trial. Partial or ambiguous reporting was conservatively classified as not reported. Data Synthesis and Statistical Analysis All included studies contributed to all analyses. Variables that were missing or not reported were coded as ‘not reported’ and retained in the dataset, allowing all 695 trials to be included in descriptive and regression analyses. A descriptive synthesis was conducted to summarise trial characteristics and methodological features using counts, percentages, means, medians, and ranges. Findings were stratified by condition where appropriate. Data analyses were conducted using SPSS version 31 (IBM Corp., Armonk, NY, USA). Associations between trial-level characteristics and the likelihood of being classified as low risk of bias were explored using univariable binary logistic regression analyses. Effect measures included odds ratios (ORs) with 95% confidence intervals for logistic regression analyses. For regression analyses, the outcome variable was overall risk of bias, dichotomised as low risk versus not low risk (including some concerns and high risk), consistent with methodological guidance for analyses with limited event counts. This dichotomisation was applied to ensure sufficient statistical power and stability of estimates, given the relatively small number of trials classified as low risk of bias. Explanatory variables were selected a priori based on methodological relevance and included trial registration status, publication period, and World Bank country income classification. Variables directly contributing to the risk-of-bias assessment, such as blinding and allocation concealment, were not modelled as independent predictors to avoid circular inference. Associations between trial characteristics and reporting quality were examined using multivariable linear regression with CONSORT percentage adherence as the dependent variable and trial registration status, publication period, clinical condition, and country income level as prespecified predictors. For linear regression analyses, CONSORT adherence percentage was treated as a continuous outcome variable. Effect measures included β coefficients for linear regression models. Analyses were conducted using complete-case data, and variables with missing data were not imputed. Model assumptions for logistic and linear regression, including independence of observations and linearity of relationships, were considered during analysis, and no major violations were identified based on inspection of model outputs. All extracted variables and analytical data are available in Additional file 4. Due to substantial heterogeneity in interventions, outcome measures, and reporting practices, a meta-analysis was not undertaken. No formal statistical assessment of reporting bias was conducted due to the absence of a meta-analysis. Results were presented using structured tables and graphical displays, including summary tables of trial characteristics, regression outputs, and figures illustrating study selection, temporal trends, geographical distribution, and risk-of-bias assessments. Subgroup and Sensitivity Analyses Prespecified subgroup analyses examined differences by clinical condition, outcome type (subjective versus objective), and country income classification. Sensitivity analyses explored alternative dichotomisations of overall risk of bias, including comparisons of high-risk trials versus all other classifications. Use of Large Language Models No large language models or artificial intelligence tools were used in the design, analysis, or writing of this study. The conceptual framework underpinning the evaluation of methodological characteristics and risk of bias in gynaecological randomized controlled trials is presented in Fig. 2. Figure 2. Conceptual framework for evaluating methodological characteristics and risk of bias in gynaecological randomised controlled trials Results Overall, the included trials demonstrated substantial heterogeneity in sample size, reporting practices, and methodological design across conditions. Structural characteristics such as trial registration, funding transparency, and country income level varied considerably, with notable implications for methodological quality and reporting completeness. Study selection Database searches identified randomised controlled trials evaluating interventions for endometriosis, PCOS, GDM, and PMDD. A total of 6,937 records were identified through database searches. After removal of duplicates and records marked as ineligible by automation tools, 4,113 records were screened by title and abstract. Of these, 3,055 records were excluded. Covidence automation tools were used to assist screening; however, all eligibility decisions were confirmed through reviewer assessment. A total of 1,058 full-text articles were sought for retrieval, of which 22 could not be retrieved in full text despite attempts through institutional access, interlibrary requests, and online sources, and were therefore excluded. The remaining 1,036 articles were assessed for eligibility, with 341 studies excluded for predefined reasons. A total of 695 randomized controlled trials were included in the final synthesis. The study selection process is presented in the PRISMA flow diagram (Figure 3) Figure 3: PRISMA Overall characteristics of included trials The included trials demonstrated substantial heterogeneity in scale, design, and reporting practices. Sample sizes ranged from small single-centre studies enrolling fewer than 20 participants to large multicentre and population-based trials enrolling several thousand participants. The characteristics of included studies are presented in Additional file 2. Across the included randomised controlled trials, sample sizes varied substantially both within and across gynaecological conditions, with consistently right-skewed distributions reflecting the presence of a small number of large trials alongside many small studies. Among the 16 trials evaluating interventions for PMDD, sample sizes were generally small. The median sample size was 40 participants (range 27–252), with a mean of 84 (SD 80.8), indicating a markedly right-skewed distribution. Most PMDD trials (10/16) enrolled fewer than 50 participants, while only four trials included 100 or more participants, suggesting that the evidence base is largely derived from small-scale studies. In the 117 endometriosis trials, sample sizes were more broadly distributed but remained highly variable. The median sample size was 67 participants (range 18–1,689), with a mean of 151.49 (SD 233.29). While a substantial proportion of trials were small (34 studies enrolled fewer than 50 participants), a similar number recruited between 50 and 99 participants (43 trials), and 40 trials included 100 or more participants. This pattern reflects considerable heterogeneity in trial scale, with a minority of large trials contributing disproportionately to the overall sample size profile. The 364 trials evaluating interventions for PCOS also demonstrated wide variability in sample size. The median sample size was 69 participants (range 10–1,508), with a mean of 104.9 (SD 143.1), again indicating right skewness. Trials enrolling between 50 and 99 participants were most common (148 studies), while 104 trials enrolled fewer than 50 participants and 112 included 100 or more participants. In contrast, the 198 trials evaluating interventions for GDM tended to be substantially larger. The median sample size was 132 participants (range 12–47,080), while the mean was 693.96 (SD 3,933.06), reflecting extreme right skewness driven by a small number of very large trials. Most GDM trials (124 studies) enrolled 100 or more participants, with fewer trials recruiting between 50 and 99 participants (52 studies) or fewer than 50 participants (22 studies). Compared with the other conditions examined, the GDM evidence base included a higher proportion of large-scale trials. Table 1 shows the sample size characteristics across the four conditions. Table 1: Sample size characteristics Condition Trials (n) Median sample size Mean (SD) Range <50 n (%) 50–99 n (%) ≥100 n (%) PMDD 16 40 84 (80.8) 27–252 10 (62.5) 2 (12.5) 4 (25.0) Endometriosis 117 67 151.49 (233.29) 18–1,689 34 (29.1) 43 (36.8) 40 (34.2) PCOS 364 69 104.9 (143.1) 10–1,508 104 (28.6) 148 (40.7) 112 (30.8) GDM 198 132 693.96 (3,923.16) 12–47,080 22(11.11) 52 (26.26) 125 (62.62) Abbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus. Publication years spanned 2005 to 2025, with a marked increase in trial volume after 2010. Figure 4 compares the Publication years across the 4 different conditions. Figure 4. Publication timeline of randomised controlled trials across four gynaecological conditions. Bars represent the number of trials published within each time interval, stratified by condition. Across all conditions, trial volume increased markedly after 2010, with the greatest concentration of studies published between 2020 and 2025. Trial characteristics varied systematically by condition. GDM trials were generally larger and more frequently population-based, whereas PMDD trials were predominantly small. Endometriosis and PCOS trials showed the widest variability in sample size, reflecting a mixture of exploratory studies, mid-sized trials, and large multicentre investigations. Across all conditions, most trials reported the use of randomisation; however, the level of methodological detail varied considerably. Allocation concealment, blinding procedures, and analytical approaches were less consistently reported, contributing to heterogeneity in methodological quality. Descriptive methodological characteristics relevant to trial quality Trial registration status Prospective trial registration was reported in a substantial proportion of included trials, although reporting practices varied by condition. Among endometriosis trials, 78 of 117 studies (66.7%) were prospectively registered, while 39 trials (33.3%) did not report trial registration. No trials were retrospectively registered. For PCOS, 224 of 364 trials (61.5%) were prospectively registered, 4 trials (1.1%) were retrospectively registered, and 136 trials (37.4%) did not report registration. Trial registration was most consistently reported among GDM trials. Of 198 trials, 165 (83.3%) were prospectively registered, 2 (1.0%) were retrospectively registered, and 31 (15.7%) did not report registration. In PMDD, 11 of 16 trials (68.8%) were prospectively registered, while 5 trials (31.3%) did not report trial registration. No PMDD trials were retrospectively registered. Overall, while prospective registration was common, non-reporting of registration remained prevalent, particularly in endometriosis and PCOS trials. Funding sources Reporting of funding sources was heterogeneous across all conditions. Public or academic funding predominated across the evidence base, particularly in GDM (60.1%) and PCOS (51.1%) trials. Industry funding or support was reported in a substantial minority of trials across all conditions, most frequently in pharmacological studies, while mixed funding arrangements were also common. Non-reporting of funding source remained prevalent, particularly in endometriosis (12.8%) and PCOS (13.5%) trials, limiting full assessment of potential funding-related bias. Geographical distribution and country income level The geographical distribution of trials was uneven across conditions. Endometriosis and PMDD trials were predominantly conducted in high-income countries, with limited representation from lower- and lower-middle-income settings and few studies conducted across mixed-income contexts. PCOS trials demonstrated broader geographical spread, with substantial representation from upper-middle-income countries, although inclusion of lower- and lower-middle-income settings remained limited, and only a small number of studies were conducted in mixed-income settings. GDM trials were largely conducted in high-income and upper-middle-income countries, with very few trials spanning mixed-income contexts. Overall, evidence from low- and lower-middle-income countries remained sparse across all conditions, limiting generalisability to diverse healthcare settings. Figure 5 shows the income level of the settings across the four conditions. Figure 5. Geographical distribution of randomised controlled trials by World Bank income classification. Stacked bars represent the proportion of trials conducted in high-, upper-middle-, lower-/low-income, and mixed-income settings for each condition. Across all conditions, trials were predominantly conducted in high-income and upper-middle-income countries, with limited representation from lower-income settings. Overall risk of bias Across all 695 included trials, only 49 (7%) were judged to be at low risk of bias, whereas the majority were assessed as having some concerns or high risk of bias using the RoB 2 tool. A summary Plot of The Risk of Bias created using the Risk of Bias Visualization tool is represented in Figure 6. When stratified by condition, GDM trials demonstrated the most favourable overall risk of bias profile. Of 198 GDM trials, 15 (7.6%) were judged to be at low risk of bias, 148 (74.7%) had some concerns, and 35 (17.7%) were judged to be at high risk. In contrast, PMDD trials most frequently demonstrated elevated risk of bias. Of 16 PMDD trials, 1 (6.3%) was judged to be at low risk, 11 (68.8%) had some concerns, and 4 (25.0%) were classified as high risk. Among endometriosis trials (n = 117), 8 trials (6.84%) were judged to be at low risk of bias, while 66 (56.41%) had some concerns, and 43 (36.75%) were judged to be at high risk. For PCOS, only 25 of 364 trials (6.9%) were classified as low risk, while 172 (47.3%) had some concerns, and 167 (45.9%) were assessed as high risk of bias. Study-level risk of bias judgments across the five RoB 2 domains and overall risk are summarized in Additional file 3. Figure 6. Summary Plot of Cochrane Risk of Bias 2 As detailed in Table 2, multiple structural characteristics of trial design and reporting were associated with increased susceptibility to bias, particularly selective reporting, performance bias, and analytical bias. Table 2. Structural factors associated with elevated risk of bias and limited transparency in gynaecological randomised controlled trials (n = 695) Structural factor Evidence from the dataset The main type of bias introduced How does this affect methodological quality Impact on transparency Absence of prospective trial registration 33.3% of endometriosis trials, 37.4% of PCOS trials, and 31.3% of PMDD trials were unregistered, compared with 15.7% of GDM trials Selective outcome reporting bias Allows outcomes to be changed or selectively reported after results are known; prevents verification of prespecified analyses High Small sample size 62.5% of PMDD trials and ~30% of endometriosis and PCOS trials enrolled fewer than 50 participants Attrition bias, reporting bias, and performance bias Limits statistical power, increases sensitivity to missing data, reduces the feasibility of blinding, and encourages reporting of multiple exploratory outcomes High Reliance on subjective outcomes (e.g., pain, mood, quality of life) Dominant in endometriosis (pain in 88% of trials) and PMDD (symptom outcomes in 87.5%) Performance and detection bias Participant expectations and assessor knowledge can influence outcome measurement when blinding is incomplete High Use of multiple biochemical or surrogate outcomes 84.6% of PCOS trials reported biochemical outcomes, often multiple per study Selective reporting bias Increases flexibility to report only statistically significant results and obscures true clinical relevance High Non-pharmacological intervention designs (surgical, behavioural, lifestyle) Common in endometriosis and PCOS trials Performance and measurement bias Participant blinding is often infeasible; intervention adherence varies and is difficult to standardise Moderate–High Incomplete reporting of the funding source 12.8% of endometriosis trials and 13.5% of PCOS trials did not disclose funding Reporting bias; potential conflict-of-interest bias Obscures financial influences on outcome prioritisation, comparator choice, and interpretation Moderate Conduct in lower-resource research settings Trials from lower- and lower-middle-income countries were sparse but showed poorer reporting completeness Selection and reporting bias Limited regulatory oversight and research infrastructure reduce adherence to reporting standards Moderate Failure to report intention-to-treat analysis Reported in only 37.5–41.8% of PMDD, endometriosis, and PCOS trials (55% in GDM) Attrition and analytical bias Excluding participants post-randomisation distorts treatment effect estimates Moderate Abbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus. Bias arising from the randomisation process Across all conditions, most trials reported the use of random allocation; however, detailed reporting of sequence generation and allocation concealment was frequently absent, resulting in many trials being judged as having some concerns in this domain. This limitation was particularly evident in PCOS and endometriosis trials, where randomisation methods were often mentioned without further specification. In contrast, GDM trials more frequently reported computer-generated or centralised randomisation, contributing to a lower proportion of high-risk judgments in this domain. Bias due to deviations from intended interventions Bias related to deviations from intended interventions was one of the most prevalent sources of bias across all conditions. Incomplete reporting of participant and personnel blinding was common, particularly in behavioural, lifestyle, and surgical intervention trials. This domain was especially problematic in endometriosis and PMDD trials, where subjective symptom outcomes were primary endpoints, and blinding was frequently incomplete or absent. Across conditions, trials without participant blinding constituted a substantial proportion of the evidence base, contributing to the elevated risk of performance bias. Bias due to missing outcome data Missing outcome data contributed substantially to the elevated risk of bias across all conditions. Attrition was widespread in trials with longer durations and those assessing behavioral or lifestyle interventions. Reporting of reasons for dropout and methods for handling missing data was inconsistent, and intention-to-treat analyses were not uniformly applied. This domain was a notable source of bias in PCOS and PMDD trials, whereas GDM trials generally demonstrated lower attrition. Bias in the measurement of outcomes Risk of bias in outcome measurement varied systematically by outcome type. Trials relying on objective biochemical, metabolic, or obstetric outcomes, particularly among GDM trials, were generally assessed as having low risk of bias in this domain. In contrast, trials evaluating subjective outcomes, such as pain, mood, or quality of life, were frequently judged to be at higher risk, especially when blinding of outcome assessors was unclear or not reported. This issue was most prominent in endometriosis and PMDD trials, where patient-reported outcomes constituted primary endpoints. Bias in the selection of the reported result Selective outcome reporting was a major concern across the evidence base. In many trials, particularly those that were not prospectively registered, prespecified outcomes could not be verified. Discrepancies between methods and results sections were frequently observed, including selective reporting of statistically significant outcomes. This domain contributed substantially to overall risk of bias classifications, particularly among PCOS trials, which often assessed numerous biochemical and surrogate outcomes without clear prioritisation. Predictors of Overall Risk of Bias Univariable logistic regression analyses were conducted to examine associations between selected trial characteristics and the likelihood of being assessed as low risk of bias. Results are presented as odds ratios (ORs) with 95% confidence intervals. Trial registration Prospective trial registration was strongly associated with improved methodological quality, as described in Table 3. Prospectively registered trials had significantly higher odds of being assessed as low risk of bias compared with trials that were retrospectively registered or unregistered (OR 4.08, 95% CI 1.59–10.47; p = 0.003). Table 3. Association between trial registration and low risk of bias Predictor OR 95% CI p-value Prospectively Registered vs not registered/ Retrospectively registered 4.08 1.59–10.47 0.003 Abbreviations: OR, Odd ratio; CI, Confidence interval Publication period (year group) Compared with trials published in the earliest period, studies published in more recent periods demonstrated higher odds of being assessed as low risk of bias; however, these associations did not reach statistical significance, and confidence intervals were wide, indicating substantial uncertainty. This association is presented in Table 4. Table 4. Association between publication period and low risk of bias Publication period OR 95% CI p-value 2005-2009 (reference) 1.00 – – 2010-2014 3.58 0.77–16.57 0.103 2015-2019 2.71 0.59–12.38 0.200 2020-2025 3.10 0.71–13.46 0.131 Abbreviations: OR, Odd ratio; CI, Confidence interval Country income level Trials conducted in upper-middle-, lower-, and mixed-income countries did not demonstrate significantly different odds of being assessed as low risk of bias compared with trials conducted in high-income countries. Estimates for lower-/low- and mixed-income settings were imprecise, with wide confidence intervals reflecting sparse data. The results are presented in Table 5. Table 5. Association between country income level and low risk of bias Income level OR 95% CI p-value High income (reference) 1.00 – – Upper-middle income 1.12 0.61–2.07 0.716 Lower/Low income 0.28 0.04–2.09 0.213 Mixed income 3.13 0.64-15.34 0.159 Estimates for lower-/low-income settings were unstable due to sparse data. Abbreviations: OR, Odd ratio; CI, Confidence interval Sensitivity analysis of risk of bias classification A sensitivity analysis was conducted using an alternative dichotomisation of overall risk of bias, comparing trials classified as high risk of bias with all other trials. In this analysis, the direction of associations between trial characteristics and risk of bias was broadly consistent with the primary analysis. Prospective trial registration was associated with lower odds of being classified as high risk of bias, although this association did not reach conventional statistical significance (OR 0.70, 95% CI 0.48–1.04; p = 0.077). Associations with publication period, country income level, and clinical condition were weaker and not statistically significant. These findings suggest that the primary results were not driven by the specific definition of the low risk of bias category. Reporting quality (CONSORT adherence) Overall reporting quality, as assessed using CONSORT percentage adherence, was moderate across included trials and is presented in Table 6. The mean CONSORT adherence was 73.8%, with a median adherence of 75.7%. Adherence values ranged widely, from 24.3% to 97.3%, indicating substantial variability in the completeness of reporting across studies. Table 6. Overall CONSORT reporting adherence across included trials Statistic Value (%) Mean 73.7 Median 75.7 Minimum 24.3 Maximum 97.3 When stratified by publication period, mean CONSORT adherence demonstrated a modest upward trend over time. Trials published between 2005–2009 had a mean adherence of 62.6%, compared with 71.5% for trials published between 2010–2014, 74.7% for trials published between 2015–2019, and 77.3% for trials published between 2020–2025. When stratified by clinical condition, mean CONSORT adherence varied across women’s health conditions. Trials in GDM and PMDD demonstrated the highest mean adherence (77.55% and 77.21%, respectively), followed by endometriosis (75.67%), while trials in PCOS showed lower mean adherence (71.02%). Trials that were prospectively registered demonstrated higher reporting quality compared with unregistered trials. Mean CONSORT adherence was 77.6% among registered trials, compared with 65.4% among unregistered trials. Trials assessed as low risk of bias demonstrated higher reporting quality compared with trials assessed as high or unclear risk of bias. Mean CONSORT adherence was 81.3% among trials at low risk of bias, compared with 73.3% among trials at high or unclear risk of bias. Overall, these findings indicate moderate reporting completeness across trials, with higher adherence observed among registered studies, more recent publications, and trials assessed as low risk of bias. Reporting quality regression analysis In multivariable linear regression analysis, prospective trial registration was independently associated with higher CONSORT adherence, with registered trials demonstrating an average increase of in reporting completeness compared with unregistered trials (p < .001). Reporting completeness increased across successive publication periods, and modest differences were observed between clinical conditions. Trials conducted in upper-middle- and lower-/low-income settings demonstrated significantly lower reporting quality compared with those from high-income countries, while no difference was observed for trials conducted in mixed-income settings. The model demonstrated moderate explanatory power for variation in CONSORT adherence (R² = 0.23). The multivariable linear regression is presented in Table 7. Table 7. Multivariable linear regression of factors associated with CONSORT adherence (%) Predictor β (percentage points) 95% CI p-value Trial registration (registered vs not) 8.56 6.27 to 10.85 < .001 Publication period (per category increase) 2.53 1.50 to 3.56 < .001 Clinical condition — — .015 Upper-middle-income vs High-income −3.22 −5.33 to −1.11 0.003 Low-income vs High-income −6.35 −10.03 to −2.67 < .001 Mixed-income vs High-income −0.17 −7.54 to 7.21 0.965 Abbreviations: CI, Confidence interval Time-trend analysis of reporting quality Univariable linear regression demonstrated a significant improvement in reporting quality over time, indicating a clear temporal trend toward more complete CONSORT reporting in gynaecological randomised controlled trials. Reporting quality increased across successive publication periods. Publication period explained approximately 10% of the variance in reporting completeness (R² = 0.099). This temporal association remained evident, although attenuated, in multivariable analyses adjusting for trial registration, clinical condition, and country income level. The results are presented in Table 8. Table 8. Univariable linear regression of publication period and CONSORT adherence (%) Predictor β (percentage points) Standard Error p-value Publication period (ordered) 4.18 0.48 < .001 Constant 61.23 1.53 < .001 Model fit F-statistic 75.86 < .001 R² 0.099 N 695 Publication period was modelled as an ordered categorical variable, with higher values corresponding to more recent publication periods. The Key methodological characteristics across conditions are summarised in Table 9. Table 9. Methodological characteristics of included randomised controlled trials by condition Methodological characteristic PMDD (n = 16) Endometriosis (n = 117) PCOS (n = 364) GDM (n = 198) Participant blinding reported 12 (75.0%) 56 (47.9%) 142 (39.0%) 39 (19.7%) Outcome assessor blinding reported 7 (43.8%) 63 (53.8%) 113 (31.0%) 85 (42.9%) Random sequence generation reported 14 (87.5%) 103 (88.0%) 267 (73.4%) 170 (85.9%) Allocation concealment reported 9 (56.3%) 73 (62.4%) 209 (57.4%) 146 (73.7%) Intention-to-treat analysis reported 6 (37.5%) 47 (40.2%) 152 (41.8%) 109 (55.1%) Abbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus. Participant blinding and outcome assessor blinding were classified as present only when explicitly reported as fully blinded. Random sequence generation and allocation concealment were classified as present when clearly described. Intention-to-treat (ITT) analyses were classified as present when explicitly stated; partial or modified ITT and per-protocol or complete-case analyses were not classified as ITT. Patterns of outcome selection and heterogeneity across conditions Outcome reporting demonstrated substantial heterogeneity both within and across the four gynaecological conditions examined. Trials frequently assessed multiple outcome domains, and no uniform pattern of outcome prioritisation was observed across conditions. This heterogeneity encompassed not only the type of outcomes selected, but also the balance between objective and subjective measures, the clinical relevance of endpoints, and the degree of alignment with patient-centred priorities. Among endometriosis trials, pain-related outcomes were reported in 88% of studies, patient-reported outcomes in 70%, biochemical outcomes in 52%, safety outcomes in 55%, reproductive outcomes in 29%, and imaging or surgical outcomes in 22%. In PCOS trials, biochemical or physiological outcomes were reported in 85%, clinical symptom outcomes in 76%, reproductive outcomes in 59%, patient-reported outcomes in 36%, psychological or functional outcomes in 32%, and safety outcomes in 25%. Among GDM trials, maternal or neonatal clinical outcomes were reported in 86%, biochemical outcomes in 81%, behavioural or lifestyle outcomes in 48%, patient-reported outcomes in 32%, safety outcomes in 26%, and economic outcomes in 9%. PMDD trials predominantly reported clinical symptom outcomes 87%, patient-reported outcomes 75%, psychological or functional outcomes 69%, biochemical outcomes 56%, and behavioural outcomes 50%. Cross-condition heterogeneity Marked heterogeneity was observed across conditions in both the type and distribution of outcomes assessed: Endometriosis and PMDD trials prioritised subjective symptom-based and patient-reported outcomes. PCOS trials prioritised biochemical and reproductive surrogate endpoints. GDM trials prioritised objective maternal and neonatal clinical outcomes. The proportion of trials including patient-reported outcomes ranged from 31.6% in GDM to 75.0% in PMDD, while biochemical outcomes ranged from 56.3% in PMDD to 84.6% in PCOS. Reproductive outcomes were common in PCOS (58.8%) but uncommon in endometriosis (29.1%) and largely absent from PMDD and GDM trials. Structural features of outcome heterogeneity Outcome heterogeneity was not limited to domain selection. Substantial variation was also observed in: outcome definitions (e.g., differing pain scales and biochemical thresholds), measurement instruments, timing of outcome assessment, and prioritisation of primary versus secondary endpoints. Many trials reported multiple outcomes across several domains without a clear specification of primary outcomes, particularly in PCOS research, where numerous biochemical markers were often assessed concurrently. DISCUSSION Interpreting the overall methodological landscape This systematic review provides one of the largest structural assessments to date of methodological quality in randomized controlled trials addressing major gynaecological conditions. Only a small percentage of studies meet the low-risk-of-bias criteria. Despite the large number of trials identified, high methodological quality remains relatively uncommon across the evidence base. This discovery has substantial implications for the degree to which clinical practice, the creation of guidelines, and policy decisions pertaining to women's health can be confidently informed by existing research. These findings align with broader critiques of women’s health research, which emphasise that conventional methodologies often fail to capture complexity and limit the translation of evidence into practice, underscoring the need for methodological innovation and integration across research approaches (47). The majority of trials that were deemed to have some issues indicate that many research fall into a methodological "grey zone" where they are neither obviously defective nor rigorous enough to produce high-certainty evidence. This is especially troubling because these trials often serve as the foundation for clinical guidelines and systematic reviews. This implies that, even when they come from several randomized trials, treatment recommendations in gynaecology may frequently be informed by evidence with potential methodological limitations. The biggest predictor of low risk of bias was trial registration, highlighting the significance of prospective registration for methodological transparency. Clear temporal improvements in reporting quality were shown by regression analyses; however, these changes did not reflect consistent methodological improvement over time, but were instead largely mediated by structural factors, such as rising prospective trial registration rates. Although minimal data from lower-income settings limited the findings, differences by country income level were seen in the anticipated direction. Regression-based insights into reporting quality and bias The descriptive synthesis's indicated patterns are quantitatively confirmed by the regression analyses used in this review. Specifically, after adjusting for clinical condition, publication time, and World income level, prospective trial registration was found to be the most significant indicator of reporting quality and was independently linked to more thorough CONSORT adherence. This result highlights the crucial function of registration as a structural mechanism, rather than just a procedural or administrative necessity, for limiting analytical flexibility and preventing selective reporting (48,49). There were also noticeable improvements in reporting quality over time. While multivariable modeling revealed that rising trial registration rates and other structural changes over time contributed to this improvement, univariable time-trend analysis clearly showed an increase in CONSORT adherence throughout subsequent publishing periods. This attenuation suggests that increases in reporting are mediated by the progressive institutionalization of transparency rules rather than being only the result of temporal growth (50,51). After normalization, there were still differences in reporting quality by country income level, with trials carried out in non-high-income environments generally showing lower reporting completeness. Even though these variations were slight, they probably don't represent inherent variations in scientific rigor but rather institutional differences in research infrastructure, regulatory control, and availability to methodological support. Crucially, our results warn against assuming that gains in reporting quality are evenly spread throughout international research settings (52). Methodological quality varying by condition The continuously better methodological profile of GDM trials compared to those in endometriosis, PCOS, and PMDD is one of the review's most notable conclusions. It seems doubtful that this discrepancy is accidental; rather, it represents contextual and structural variations among circumstances. Because objectives including glycaemic control, birth weight, and neonatal morbidity are regularly assessed, scientifically defined, and clinically required, GDM trials benefit from being integrated into well-established antenatal care pathways. Larger, population-based trial designs may be made possible by these characteristics, which also naturally facilitate standardized outcome measurement and lower loss to follow-up (53,54). On the other hand, diseases like PMDD and endometriosis are characterized by chronicity, varying symptom profiles, delayed diagnosis, and dependence on subjective symptom reporting. These characteristics make trial design more difficult and more prone to bias, especially when blinding is insufficient or impractical (55–57). PCOS holds an intermediate position. The condition's heterogeneity has led to an excessive dependence on surrogate and laboratory endpoints, which are frequently evaluated in isolation from patient-centered outcomes, despite the fact that it permits objective biochemical measurement. Despite the measures' seeming neutrality, this has led to selective reporting and outcome multiplicity, hurting interpretability (36,58). The results show that structural features of trial design, conduct, and reporting, rather than specific technical flaws, are the main causes of the increased risk of bias and reduced transparency in gynecological randomized controlled trials. Specifically, selective outcome reporting, analytical flexibility, and poor treatment of missing data are consistently made possible by the small trial size and absence of prospective registration (49,59). These flaws are exacerbated by outcome selection: Research that primarily uses participants' self-reported symptoms is particularly vulnerable to expectation-driven bias when blinding is not done completely (60,61). At the same time, studies that measure numerous biochemical surrogate markers create ample opportunity for selective reporting, often without providing evidence of meaningful benefits for patients (62). Intervention modality also exerts an important structural influence, as non-pharmacological trials face intrinsic barriers to participant blinding and standardisation of treatment delivery (63,64). Transparency is further undermined by incomplete funding disclosure and variable adherence to reporting standards, particularly in lower-resource research settings (65). The comparatively stronger methodological profile observed in gestational diabetes trials illustrates that higher standards of transparency and outcome consistency are achievable when trials are embedded within regulated clinical pathways and employ objective, standardised endpoints. Collectively, these findings indicate that meaningful improvement in the credibility of gynaecological evidence will require systemic reforms, mandatory trial registration, standardised outcome frameworks, robust analytical protocols, and stronger enforcement of reporting requirements, rather than incremental adjustments to individual trial practices alone. Robustness of risk-of-bias findings Sensitivity analysis using an alternative dichotomisation of overall risk of bias, comparing trials classified as high risk with all other trials, demonstrated that the direction of associations between trial characteristics and risk of bias was broadly consistent with the primary analysis. Although effect estimates were reduced and did not consistently reach statistical significance, no associations reversed direction. This pattern implies that the main conclusions represent underlying structural factors of trial quality rather than being products of a specific risk-of-bias threshold. The sensitivity analysis's attenuation is in line with the high-risk classification's stricter criteria as well as the variety of ways bias might manifest itself. Together, our results highlight the limits imposed by few low-risk trials and the intricacy of bias processes in gynecological research, while also supporting the overall conclusions' robustness (66). Subjective outcomes, blinding, and the limits of traditional RCT paradigms The high levels of performance and detection bias in trials of endometriosis and PMDD raise important questions about whether standard randomised controlled trial designs are well-suited to gynaecological conditions that rely heavily on subjective symptom reporting. Although RCTs remain the dominant standard for causal inference in clinical research, their core methodological assumptions are poorly suited to conditions in which pain, mood disturbance, fatigue, and quality of life are the primary therapeutic targets. In such contexts, achieving effective blinding and stable outcome measurement can be challenging. These outcomes are inherently vulnerable to expectancy effects, reporting biases, and contextual influences that cannot be fully addressed through randomisation alone at the outcome assessment stage (63,67). In surgical, behavioural, and lifestyle interventions, where participant blinding is often impossible, these vulnerabilities are further intensified, blurring the conventional distinction between intervention effects and placebo responses (63,68). Treating these issues as mere implementation failures obscures a more fundamental structural mismatch between methodological ideals and clinical reality. Rather than continuing to evaluate such trials against an idealised pharmacological model of blinding, a more defensible methodological position is to acknowledge these constraints explicitly and recalibrate standards of internal validity accordingly. Enhanced emphasis on outcome assessor blinding, rigorous validation of patient-reported outcome instruments, pre-specification of minimal clinically important differences, and transparent handling of missing data may provide more meaningful safeguards against bias than formalistic adherence to blinding criteria that are, in many contexts, unattainable (69,70). Without such recalibration, continued reliance on orthodox RCT frameworks risks generating systematically biased estimates that may be influenced by bias despite adherence to conventional methodological frameworks. Trial registration and transparency: progress with persistent gaps In addition to being descriptive indicators of methodological quality, trial registration and transparency were found to be statistically significant predictors of reporting completeness. In multivariable analysis, there is a high correlation between prospective registration and CONSORT adherence, indicating that registration serves as a significant constraint on selective reporting and post-hoc outcome selection. This supports the body of research showing that transparency procedures work best when incorporated into trial design early on, rather than being added after the fact at the reporting stage (71). Prospective trial registration demonstrates that transparent research practices are institutionally achievable when supported by regulatory frameworks (72), with our findings showing substantially higher prospective registration among GDM trials. However, the continued prevalence of unregistered trials in endometriosis and PCOS indicates that adoption of these norms remains uneven and condition-dependent. Given the well-established association between non-registration and selective outcome reporting, this disparity carries direct implications for the credibility of published treatment effects (73). The evidentiary problem is not limited to the absence of registration. Even among registered trials, discrepancies between prespecified and reported outcomes are common, as demonstrated by multiple audits comparing trial registries and protocols with published reports, revealing persistent weaknesses in enforcement mechanisms and editorial oversight (74–76). These findings suggest that registration currently functions more as a procedural formality and not actually to avoid reporting bias. Strengthening transparency, therefore, requires institutional reform beyond voluntary compliance, including systematic verification of registry entries during peer review, harmonisation of journal reporting requirements, and proportionate sanctions for non-adherence. The generally accepted goal of trial registration runs the risk of becoming symbolic compliance rather than true methodological responsibility in the absence of such measures. Funding, industry involvement, and subtle influences on trial design The significant minority of trials reporting industry or mixed funding arrangements warrants ongoing critical attention, even though explicit industry dominance was not visible throughout the dataset. The choice of comparators, the framing of primary outcomes, and the criteria for clinical significance are just a few examples of how financing sources may affect trial architecture in pharmacologically directed investigations in ways that are both methodologically acceptable and epistemically significant. Rarely do these factors show up as explicit wrongdoing; instead, they work through the cumulative impacts of design optimization toward market positioning or regulatory acceptance (77). Another problem is the documented non-reporting of funding sources in a subset of trials within our dataset, which prevents informed assessment of potential conflicts of interest, because of which the interpretive transparency necessary for robust evidence appraisal is undermined. In clinical domains characterised by long-term therapeutic exposure, incomplete disclosure of financial relationships erodes confidence in individual findings and in the normal integrity of the research enterprise itself, as financial ties have been shown to influence research conduct and reporting (78). From a methodological perspective, funding transparency should be regarded not as an administrative accessory but as an important component of trial validity, enabling downstream users of evidence to situate reported effects within their institutional and economic context (79,80). Equity, geography, and the production of partial knowledge The high number of trials from our dataset in high-income and upper-middle-income countries highlights a persistent inequity in evidence generation. This imbalance has important consequences; for example, trial findings may not be transferable to settings with different healthcare infrastructure, cultural contexts, or disease prevalence (81). The limited number of trials from lower-middle-income countries suggests that the global burden of gynaecological disease is being addressed with geographically narrow evidence. This pattern reflects broader structural inequities in global health research funding and infrastructure (82). Without deliberate strategies to support inclusive trial design and multinational collaboration, these disparities are likely to persist, perpetuating a cycle in which evidence is generated primarily for populations already best served by healthcare systems. Outcome selection and the role of core outcome sets The extensive outcome heterogeneity identified across all conditions reinforces longstanding concerns about research waste in gynaecology. While core outcome initiatives have sought to address this issue, uptake remains inconsistent (83). Future progress will require not only broader adoption of core outcome sets but also cultural shifts in how outcomes are valued. Integrating patient voices into outcome prioritisation and aligning funding incentives with methodological quality rather than novelty may help address these challenges. Implications for clinical practice and guideline development The findings of this review suggest that clinicians and guideline developers should exercise cautious interpretation of gynaecological trial evidence, particularly when recommendations are based on small, unblinded trials with subjective outcomes. While RCTs remain the gold standard, their methodological limitations must be explicitly acknowledged when translating evidence into practice. For policymakers, the review highlights the need to invest not only in more trials but in better-designed trials, with emphasis on transparency, inclusivity, and patient relevance. Without such investment, the evidence base risks expanding in size without corresponding gains in reliability or impact. Key findings showcase that methodological weaknesses in gynaecological randomised controlled trials are systemic, patterned, and consequential. Differences in trial quality across conditions reflect underlying structural, clinical, and epistemic factors rather than isolated shortcomings. Addressing these challenges will require coordinated efforts from researchers, funders, journals, and clinicians to prioritise methodological rigour, transparency, and equity in women’s health research. Without improvements in methodological rigor, the expanding volume of gynaecological research risks generating an evidence base with important methodological limitations. The findings of this review have important implications for research governance and policy. Regulatory agencies, research funders, and journal editors play critical roles in strengthening methodological standards. Mandatory prospective trial registration, enforcement of CONSORT reporting guidelines, and greater investment in multicentre and multinational trials are essential steps toward improving the credibility and global applicability of women’s health research. Strengths This review analysed a large dataset of 695 randomized controlled trials across four major gynaecological conditions and applied contemporary methodological frameworks including the Cochrane Risk of Bias 2 tool and CONSORT reporting standards. Limitations Several limitations should be acknowledged. First, the review included only English-language publications, which may introduce language bias. Second, reliance on published reports may underestimate methodological rigor if trial methods were incompletely reported. Third, the underrepresentation of trials conducted in low-income settings limits conclusions regarding global equity in evidence generation. Conclusion This review demonstrates that methodological limitations in gynaecological randomized controlled trials are widespread and structurally patterned rather than isolated occurrences. Although reporting practices have improved over time, robust low-risk-of-bias evidence remains uncommon. Prospective registration and transparency mechanisms are strongly associated with improved methodological quality, suggesting that meaningful progress is achievable through systemic reinforcement of these practices. Strengthening the reliability, transparency, and inclusivity of gynaecological research will be essential to ensuring that clinical and policy decisions in women’s health are supported by evidence that is not only abundant but methodologically sound. Strengthening the evidence base for women’s health will require systemic improvements in trial design, transparency mechanisms, and global research equity. Abbreviations RCT Randomized Controlled Trial PCOS Polycystic Ovary Syndrome GDM Gestational Diabetes Mellitus PMDD Premenstrual Dysphoric Disorder OBGYN Obstetrics and Gynaecology CONSORT Consolidated Standards of Reporting Trials PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses RoB 2 Risk of Bias 2 CROWN Core Outcomes in Women’s Health WHI Women's Health Initiative OR Odds Ratio CI Confidence interval Declarations Ethics approval and consent to participate Not applicable. Consent for publication All authors consented to publication. Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information files. Competing interests The authors declare that they have no competing interests. Funding This research received no external funding. Authors’ contributions NR, NW, and SA contributed equally to this work and share first authorship. NR, NW, and SA were responsible for study design, study screening, data extraction, and manuscript drafting. SA acted as the third reviewer to resolve disagreements during study selection and data extraction. VP supervised the study and contributed to data interpretation and critical revision of the manuscript. GD contributed to the study's conceptualisation and methodological oversight. All authors read and approved the final manuscript. Acknowledgements Not applicable. References McKinney JL, Clinton SC, Keyser LE. Women’s health across the lifespan: A sex- and gender-focused perspective. Phys Ther. 2024;104(10). doi:10.1093/ptj/pzae121 Geller A, Salganicoff A, Burke SP. Nih.gov [Internet]. National Academies Press (US); 2025. Overview of Selected Women’s Health Conditions. Available from: https://www.ncbi.nlm.nih.gov/books/NBK612403 Vinter C, Linde D, Petersen L. Research in women - Women in research. Dan Med J. 2026;73(2):10250782. doi:10.61409/A10250782 Christopoulos P, Tsarna E. Gynecologic and Obstetric Pathologies from Birth to Menopause: Unveiling the Journey and Charting the Future. J Clin Med. 2024;13(20):6230. doi:10.3390/jcm13206230 La Marca-Ghaemmaghami P, Ehlert U. Gynecological Health: Psychosocial Aspects. Int Encycl Soc Behav Sci Second Ed Elsevier. 2015;462–8. doi:10.1016/B978-0-08-097086-8.14098-X Hoffman SR, Farland LV, Doll KM, Nicholson WK, Wright MA, Robinson WR. The epidemiology of gynaecologic health: contemporary opportunities and challenges. J Epidemiol Community Health. 2020;jech-2019. doi:10.1136/jech-2019-213149 Kirk UB, Bank-Mikkelsen AS, Rytter D, Hartwell D, Marschall H, Nyegaard M, et al. Understanding endometriosis underfunding and its detrimental impact on awareness and research. Npj Women Health. 2024;2(1). doi:10.1038/s44294-024-00048-6 Al Wattar BH, Fisher M, Bevington L, Talaulikar V, Davies M, Conway G, et al. Clinical Practice Guidelines on the Diagnosis and Management of Polycystic Ovary Syndrome: A Systematic Review and Quality Assessment Study. J Clin Endocrinol Metab. 2021;106(8). doi:10.1210/clinem/dgab232 Wexler DJ, Powe CE, Barbour LA, Buchanan T, Coustan DR, Corcoy R, et al. Research Gaps in Gestational Diabetes Mellitus. Obstet Gynecol. 2018;132(2):496–505. doi:10.1097/aog.0000000000002726 Carlini SV, Deligiannidis KM. Evidence-Based Treatment of Premenstrual Dysphoric Disorder. J Clin Psychiatry. 2020;81(2). doi:10.4088/jcp.19ac13071 World Health Organization. World Health Organization [Internet]. 2023. Endometriosis. Available from: https://www.who.int/news-room/fact-sheets/detail/endometriosis Satapathy P, Vadia N, Menon SV, Chennakesavulu K, Panigrahi R, Bushi G, et al. Prevalence of polycystic ovarian syndrome in South Asian countries and 21 global regions with Sociodemographic index: A systematic analysis of the Global Burden of Disease (1990–2021). J Gynecol Obstet Hum Reprod. 2025;54(9):103019. doi:10.1016/j.jogoh.2025.103019 Choudhury AA, Rajeswari VD. Gestational diabetes mellitus - A metabolic and reproductive disorder. Biomed Pharmacother. 2021;143(112183):112183. doi:10.1016/j.biopha.2021.112183 Pataki B, Bianka Luca Kiss, Sára Kálmán, Ildikó Kovács. Premenstrual dysphoric disorder—an undervalued diagnosis? A cross-sectional study in Hungarian women. Compr Psychoneuroendocrinology. 2024;20:100256–100256. doi:10.1016/j.cpnec.2024.100256 Eurohealth.ie [Internet]. 2024. 2024 – Women’s Health Manifesto – Eurohealth. Available from: https://eurohealth.ie/2024-womens-health-manifesto/ Tharmalingam H. Macmillan Cancer Support [Internet]. 2024. A greater society for all - why women’s health matters. Available from: https://www.macmillan.org.uk/healthcare-professionals/news-and-resources/blogs/iwd-why-womens-health-matters Ekinci G. The impact of maternal and child health on sustainable development goals: evidence from Turkic Republics (2000–2020). BMC Public Health. 2025;25(1). doi:10.1186/s12889-025-25172-z Wijeratne D, Gibson JFE, Fiander A, Rafii-Tabar E, Thakar R. The global burden of disease due to benign gynecological conditions: A call to action. Int J Gynaecol Obstet Off Organ Int Fed Gynaecol Obstet. 2023;164(3):1151–9. doi:10.1002/ijgo.15211 Davidson PM, McGrath SJ, Meleis AI, Stern P, DiGiacomo M, Dharmendra T, et al. The Health of Women and Girls Determines the Health and Well-Being of Our Modern World: A White Paper From the International Council on Women’s Health Issues. Health Care Women Int. 2011;32(10):870–86. doi:10.1080/07399332.2011.603872 Kleinman DV, Pronk N, Gómez CA, Gordon GLW, Ochiai E, Blakey C, et al. Addressing health equity and social determinants of health through healthy people 2030. J Public Health Manag Pract. 2021;27(6):249–57. doi:10.1097/PHH.0000000000001297 Lefton M. swhr.org [Internet]. 2022. The Evolution of Women’s Health Research - SWHR. Available from: https://swhr.org/the-evolution-of-womens-health-research/ Pinn VW. Women’s Health Research: Current State of the Art. Glob Adv Health Med. 2013;2(5):8–10. doi:10.7453/gahmj.2013.063 Price C. Nih.gov [Internet]. 2014. Women’s Health Initiative (WHI) | National Heart, Lung, and Blood Institute (NHLBI). Available from: https://www.nhlbi.nih.gov/science/womens-health-initiative-whi Research I of M (US) C on WH. www.ncbi.nlm.nih.gov [Internet]. National Academies Press (US); 2010. Methodologic Issues in Women’s Health Research. Available from: https://www.ncbi.nlm.nih.gov/books/NBK210137/ Steinberg JR, Turner BE, DiTosto JD, Weeks BT, Young AMP, Lu CF, et al. Race and Ethnicity Reporting and Representation in Obstetrics and Gynecology Clinical Trials and Publications From 2007-2020. JAMA Surg. 2023;158(2):181–90. doi:10.1001/jamasurg.2022.6600 Steinberg JR, Weeks BT, Reyes GA, Fitzgerald AC, Lyell DJ. The obstetric research landscape: a cross-sectional analysis of clinical trials from 2007-2020. Am J Obstet Gynecol MFM. 2020;3(1):100253. doi:10.1016/j.ajogmf.2020.100253 Lee SS, Dinicu AI, Arthurs L, Shields D, Pothuri B, Lightfoot MDS. Demographic reporting and language exclusion in gynecologic oncology clinical trials. Am J Obstet Gynecol. 2023;230(1):73.e1-73.e14. doi:10.1016/j.ajog.2023.09.019 Bierer BE, Meloney LG, Ahmed HR, White SA. Advancing the inclusion of underrepresented women in clinical research. Cell Rep Med. 2022;3(4):100553. doi:10.1016/j.xcrm.2022.100553 Collister D, Song C, Ruzycki SM. Fostering diversity in clinical trials: need for evidence and implementation to improve representation. BMJ Med. 2024;3(1):e000984–e000984. doi:10.1136/bmjmed-2024-000984 Gayathri De Lanerolle, Phiri P, Haroon A. Clinical Trials and Tribulations. Elsevier; 2023. Balch B. Association of American Medical Colleges [Internet]. 2024. Why we know so little about women’s health. Available from: https://www.aamc.org/news/why-we-know-so-little-about-women-s-health Waitt C, Astill D, Zavala E, Karron RA, Faden RR, Stratton P, et al. Clinical trials and pregnancy. Commun Med. 2022;2(1):1–5. doi:10.1038/s43856-022-00198-1 sanogenetics.com [Internet]. Pioneering progress: The history of women in clinical trials. Available from: https://sanogenetics.com/resources/blog/the-history-of-women-in-clinical-trials Steinberg JR, Turner BE, Weeks BT, Magnani CJ, Wong BO, Rodriguez F, et al. Analysis of Female Enrollment and Participant Sex by Burden of Disease in US Clinical Trials Between 2000 and 2020. JAMA Netw Open. 2021;4(6):e2113749. doi:10.1001/jamanetworkopen.2021.13749 van ’t Hooft J, van Dijk CE, Axfors C, Alfirevic Z, Oudijk MA, Khan KS, et al. Assessing the usefulness of randomised trials in obstetrics and gynaecology. BJOG Int J Obstet Gynaecol. 2023;130(7):695–701. doi:10.1111/1471-0528.17411 Al Wattar BH, Teede H, Garad R, Franks S, Balen A, Bhide P, et al. Harmonising research outcomes for polycystic ovary syndrome: an international multi-stakeholder core outcome set. Hum Reprod. 2020;35(2):404–12. doi:10.1093/humrep/dez272 Tellum T, Omtvedt M, Naftalin J, Hirsch M, Jurkovic D. A systematic review of outcome reporting and outcome measures in studies investigating uterine-sparing treatment for adenomyosis. Hum Reprod Open. 2021;2021(3). doi:10.1093/hropen/hoab030 The CROWN Initiative: journal editors invite researchers to develop core outcomes in women’s health. J Gynecol Oncol. 2014;25(3):166. doi:10.3802/jgo.2014.25.3.166 Macedo C, Riera R, Torloni M. Methodological quality of systematic reviews and clinical trials on women’s health published in a Brazilian evidence-based health journal. Clinics. 2013;68(4):563–7. doi:10.6061/clinics/2013(04)20 Savović J, Turner RM, Mawdsley D, Jones HE, Beynon R, Higgins JPT, et al. Association Between Risk-of-Bias Assessments and Results of Randomized Trials in Cochrane Reviews: The ROBES Meta-Epidemiologic Study. Am J Epidemiol. 2017;187(5):1113–22. doi:10.1093/aje/kwx344 Shrestha BM. Systematic Reviews and Meta-analysis: Principles and Practice. J Nepal Med Assoc. 2019;57(215). doi:10.31729/jnma.3986 Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Br Med J. 2021;372(71). doi:10.1136/bmj.n71 Raghuraman N, Delanerolle G. York.ac.uk [Internet]. 2025 [cited 2025 Jan 1]. PROSPERO. Available from: https://www.crd.york.ac.uk/PROSPERO/view/CRD42025636543 Garg R, Maurya I. Clinical trial registration: An essential step toward transparency in clinical research. Indian J Anaesth. 2023;67(4):321. doi:10.4103/ija.ija_234_23 Cochrane Methods. methods.cochrane.org [Internet]. 2017. Risk of Bias 2 (RoB 2) tool. Available from: https://methods.cochrane.org/risk-bias-2 CONSORT 2010 checklist of information to include when reporting a randomised trial [Internet]. 2010. Report No. Available from: https://legacyfileshare.elsevier.com/promis_misc/CONSORT-2010-Checklist.pdf Delanerolle G, Pathiraja V, Mudalige T, Mbwele B, Ikwuka D, Cavalini H, et al. A Perspective on Innovating Qualitative Research Methods to Optimise Women’s Health Research. Int J Qual Methods. 2025;24. doi:10.1177/16094069251406147 Simonsen E, Wang SV, Kildegaard H, Pottegård A. Preregistration: A Key to Credible Real‐World Evidence Generation. Pharmacoepidemiol Drug Saf. 2025;34(9). doi:10.1002/pds.70215 Taylor NJ, Gorman DM. Registration and primary outcome reporting in behavioral health trials. BMC Med Res Methodol. 2022;22(1). doi:10.1186/s12874-021-01500-w Freese J, King MM. Institutionalizing Transparency. Socius Sociol Res Dyn World. 2018;4:237802311773921. doi:10.1177/2378023117739216 Stefkovics Á, Eichhorst A, Skinnion D, Harrison CH. Are We Becoming More Transparent? Survey Reporting Trends in Top Journals of Social Sciences. Int J Public Opin Res. 2024;36(2). doi:10.1093/ijpor/edae013 Alemayehu C, Mitchell G, Nikles J. Barriers for conducting clinical trials in developing countries- a systematic review. Int J Equity Health. 2018;17(1). doi:10.1186/s12939-018-0748-6 Munda A, Mlinaric Z, Jakin PA, Lunder M, Pongrac Barlovic D. Effectiveness of a comprehensive telemedicine intervention replacing standard care in gestational diabetes: a randomized controlled trial. Acta Diabetol. 2023;60:1037–44. doi:10.1007/s00592-023-02099-8 Leblalta B, Kebaili H, Sim R, Lee SWH. Digital health interventions for gestational diabetes mellitus: A systematic review and meta-analysis of randomised controlled trials. Lichtner V, editor. PLOS Digit Health. 2022;1(2):e0000015. doi:10.1371/journal.pdig.0000015 Saunders PTK, Whitaker LHR, Horne AW. Endometriosis: Improvements and challenges in diagnosis and symptom management. Cell Rep Med. 2024;5(6):101596. doi:10.1016/j.xcrm.2024.101596 Nih.gov [Internet]. Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU); 2018. Endometriosis – Diagnosis, treatment and patient experiences. Available from: https://www.ncbi.nlm.nih.gov/books/NBK570473 Pearlstein T, Steiner M. Premenstrual dysphoric disorder: burden of illness and treatment update. J Psychiatry Neurosci JPN. 2008;33(4):291. Legro RS. Surrogate end-points or primary outcomes in clinical trials in women with polycystic ovary syndrome? Hum Reprod. 2004;19(8):1697–704. doi:10.1093/humrep/deh322 Zhang Z, Xu X, Ni H. Small studies may overestimate the effect sizes in critical care meta-analyses: a meta-epidemiological study. Crit Care. 2013;17(1):R2. doi:10.1186/cc11919 Pitre T, Kirsh S, Jassal T, Anderson M, Padoan A, Xiang A, et al. The Impact of Blinding on Trial results: a Systematic Review and Meta‐analysis. Cochrane Evid Synth Methods. 2023;1(4). doi:10.1002/cesm.12015 Hrobjartsson A, Thomsen ASS, Emanuelsson F, Tendal B, Hilden J, Boutron I, et al. Observer bias in randomized clinical trials with measurement scale outcomes: a systematic review of trials with both blinded and nonblinded assessors. Can Med Assoc J. 2013;185(4):E201–11. Manyara AM, Davies P, Stewart D, Weir CJ, Young AE, Blazeby J, et al. Reporting of surrogate endpoints in randomised controlled trial protocols (SPIRIT-Surrogate): extension checklist with explanation and elaboration. BMJ. 2024;386:e078525–e078525. doi:10.1136/bmj-2023-078525 Yonis A, Dean SG, Warren FC, Taylor RS, Levack W, Hay-Smith J. Navigating blinding challenges in complex intervention trials: insights from a UK researcher survey. Trials. 2025;26(1). doi:10.1186/s13063-025-09223-9 Boutron I, Tubach F, Giraudeau B, Ravaud P. Blinding was judged more difficult to achieve and maintain in nonpharmacologic than pharmacologic trials. J Clin Epidemiol. 2004;57(6):543–50. doi:10.1016/j.jclinepi.2003.12.010 Zhang W, DeVito NJ, Chan AW, Cunningham C, Pymento J, Karam G, et al. Enhancing global clinical trial transparency for better health outcomes for all. F1000Research. 2025;14:626. doi:10.12688/f1000research.166358.1 Marušić MF, Fidahić M, Cepeha CM, Farcaș LG, Tseke A, Puljak L. Methodological tools and sensitivity analysis for assessing quality or risk of bias used in systematic reviews published in the high-impact anesthesiology journals. BMC Med Res Methodol. 2020;20(1). doi:10.1186/s12874-020-00966-4 Moustgaard H, Bello S, Miller FG, Hróbjartsson A. Subjective and objective outcomes in randomized clinical trials: definitions differed in methods publications and were often absent from trial reports. J Clin Epidemiol. 2014;67(12):1327–34. doi:10.1016/j.jclinepi.2014.06.020 Wartolowska K, Feakins BG, Collins GS, Cook J, Judge A, Rombach I, et al. The magnitude and temporal changes of response in the placebo arm of surgical randomized controlled trials: a systematic review and meta-analysis. Trials. 2016;17(1). doi:10.1186/s13063-016-1720-7 Boutron I, Altman DG, Moher D, Schulz KF, Ravaud P. CONSORT Statement for Randomized Trials of Nonpharmacologic Treatments: A 2017 Update and a CONSORT Extension for Nonpharmacologic Trial Abstracts. Ann Intern Med. 2017;167(1):40. doi:10.7326/m17-0046 Revicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J Clin Epidemiol. 2008;61(2):102–9. doi:10.1016/j.jclinepi.2007.03.012 Vitali FC, Santos PS, Rocha A de O, Maia LC, Garcia L da FR, Teixeira C da S. Adherence to Registration and Selective Outcome Reporting in Randomized Clinical Trials Published in Endodontic Journals Over the Past 5 Years: A Meta-Research Study. J Endod. 2025;51(3):258-267.e7. doi:10.1016/j.joen.2024.12.001 De Angelis C, Drazen JM, Frizelle FA, Haug C, Hoey J, Horton R, et al. Clinical Trial Registration: A Statement from the International Committee of Medical Journal Editors. N Engl J Med. 2004;351(12):1250–1. doi:10.1056/nejme048225 Xu C, Fan S, Furuya-Kanamori L, Li S, Lin L, Chu H, et al. Harm effects in non-registered versus registered randomized controlled trials of medications: a retrospective cohort study of clinical trials. BMC Med. 2024;22(1). doi:10.1186/s12916-024-03621-7 Chan AW, Hróbjartsson A, Haahr MT, Gøtzsche PC, Altman DG. Empirical Evidence for Selective Reporting of Outcomes in Randomized Trials. JAMA. 2004;291(20):2457. doi:10.1001/jama.291.20.2457 Chan AW, Pello A, Kitchen J, Axentiev A, Virtanen JI, Liu A, et al. Association of Trial Registration With Reporting of Primary Outcomes in Protocols and Publications. JAMA. 2017;318(17):1709. doi:10.1001/jama.2017.13001 Dechartres A, Ravaud P, Atal I, Riveros C, Boutron I. Association between trial registration and treatment effect estimates: a meta-epidemiological study. BMC Med. 2016;14(1). doi:10.1186/s12916-016-0639-x Gazendam AM, Slawaska-Eng D, Nucci N, Bhatt O, Ghert M. The Impact of Industry Funding on Randomized Controlled Trials of Biologic Therapies. Medicines. 2022;9(3):18. doi:10.3390/medicines9030018 Bekelman JE, Li Y, Gross CP. Scope and Impact of Financial Conflicts of Interest in Biomedical Research. JAMA. 2003;289(4):454. doi:10.1001/jama.289.4.454 Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals [Internet]. 2025. Report No. Available from: https://www.icmje.org/icmje-recommendations.pdf Farrell A. Improving transparency in clinical trial reporting. PLOS Med. 2025;22(4):e1004588. doi:10.1371/journal.pmed.1004588 Rothwell PM. External validity of randomised controlled trials: “To whom do the results of this trial apply?” The Lancet. 2005;365(9453):82–93. doi:10.1016/s0140-6736(04)17670-8 Thiers FA, Sinskey AJ, Berndt ER. Trends in the globalization of clinical trials. Nat Rev Drug Discov. 2008;7(1):13–4. doi:10.1038/nrd2441 Duffy J, Rolph R, Gale C, Hirsch M, Khan K, Ziebland S, et al. Core outcome sets in women’s and newborn health: a systematic review. BJOG Int J Obstet Gynaecol. 2017;124(10):1481–9. doi:10.1111/1471-0528.14694 Additional Declarations No competing interests reported. Supplementary Files Additionalfile1.docx Additional file 1: Search strategy (DOCX) Detailed search strategies used for all databases. Additionalfile2.xlsx Additional file 2: Study characteristics (XLSX) Characteristics of all included studies. Additionalfile3.xlsx Additional file 3: Risk of bias assessment (XLSX) Risk of bias assessments using the RoB 2 tool. Additionalfile4.xlsx Additional file 4: Analysis dataset (XLSX) Extracted variables used for descriptive and analytical purposes. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9212126","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":616356265,"identity":"173eeda0-4ecf-409c-9cb1-df9bae30c8c9","order_by":0,"name":"Neha Raghuraman","email":"","orcid":"","institution":"University of Birmingham","correspondingAuthor":false,"prefix":"","firstName":"Neha","middleName":"","lastName":"Raghuraman","suffix":""},{"id":616356266,"identity":"423f6a98-d2e9-49f6-8364-39a672e66747","order_by":1,"name":"Nimesha Wijamuni","email":"","orcid":"","institution":"University of Colombo","correspondingAuthor":false,"prefix":"","firstName":"Nimesha","middleName":"","lastName":"Wijamuni","suffix":""},{"id":616356267,"identity":"286ef732-30a8-4d6f-9e6c-b2d2d1918afd","order_by":2,"name":"Sepide Ahmadi","email":"","orcid":"","institution":"Tehran University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Sepide","middleName":"","lastName":"Ahmadi","suffix":""},{"id":616356268,"identity":"c95bc870-442a-4c9d-8b77-ae8163c399fd","order_by":3,"name":"Vindya Pathiraja","email":"data:image/png;base64,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","orcid":"","institution":"University of Ruhuna","correspondingAuthor":true,"prefix":"","firstName":"Vindya","middleName":"","lastName":"Pathiraja","suffix":""},{"id":616356269,"identity":"96d7e915-7707-44fa-b8f3-cbe8be687667","order_by":4,"name":"Gayathri Delanerolle","email":"","orcid":"","institution":"University of Birmingham","correspondingAuthor":false,"prefix":"","firstName":"Gayathri","middleName":"","lastName":"Delanerolle","suffix":""}],"badges":[],"createdAt":"2026-03-24 12:53:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9212126/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9212126/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106341967,"identity":"fe7c2b81-8c22-4801-af55-b8dd4ab05b24","added_by":"auto","created_at":"2026-04-07 15:47:44","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":200276,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMaternal mortality ratio in relation to GDP per capita (2020)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMaternal mortality ratio (deaths per 100,000 live births) plotted against GDP per capita (international dollars, purchasing power parity), both shown on logarithmic scales. The figure illustrates a strong inverse relationship between national income and maternal mortality, with the highest mortality ratios concentrated in low-income countries, particularly in sub-Saharan Africa. Data sources: UN Maternal Mortality Estimation Inter-Agency Group (MMEIG), World Bank, International Monetary Fund (IMF), OECD, and Eurostat. Figure reproduced from Our World in Data.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/c57e51b120ea54a999d0ecf6.png"},{"id":106341969,"identity":"3abcb8eb-a622-4185-86a7-54f1f85f7f66","added_by":"auto","created_at":"2026-04-07 15:47:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":776379,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConceptual framework for evaluating methodological characteristics and risk of bias in gynaecological randomised controlled trials\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/c442a70ce30bad811a2811f2.png"},{"id":106404203,"identity":"3d69a152-8d33-44f5-bada-0c9f1cff34e7","added_by":"auto","created_at":"2026-04-08 09:15:38","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":62565,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePRISMA\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/9645c1ad480038a1127eb39d.png"},{"id":106341972,"identity":"df6a8284-3f91-4ed7-b910-2e9651e4a14b","added_by":"auto","created_at":"2026-04-07 15:47:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":27025,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePublication timeline of randomised controlled trials across four gynaecological conditions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBars represent the number of trials published within each time interval, stratified by condition. Across all conditions, trial volume increased markedly after 2010, with the greatest concentration of studies published between 2020 and 2025.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/6ef3b9fe34dd4ab1e6c3a91a.png"},{"id":106404006,"identity":"d0f32c4c-77f8-4114-b5c5-7120a8eef317","added_by":"auto","created_at":"2026-04-08 09:15:21","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":23981,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGeographical distribution of randomised controlled trials by World Bank income classification.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStacked bars represent the proportion of trials conducted in high-, upper-middle-, lower-/low-income, and mixed-income settings for each condition. Across all conditions, trials were predominantly conducted in high-income and upper-middle-income countries, with limited representation from lower-income settings.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/c33cf300a0277e114d3ee56c.png"},{"id":106341974,"identity":"5bffae5e-7f79-46bd-a5ea-c15cf34d973d","added_by":"auto","created_at":"2026-04-07 15:47:44","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":61113,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSummary Plot of Cochrane Risk of Bias 2\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/71fe76ccd2e39c127a30a787.png"},{"id":108627433,"identity":"8ccc1845-f377-45a9-9805-f8f3c4d44429","added_by":"auto","created_at":"2026-05-06 15:56:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1591882,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/3bdf6e15-6fb9-42a8-b06a-900ccdc13197.pdf"},{"id":106403547,"identity":"a812fdb7-8903-40dd-af6c-f943c1863b85","added_by":"auto","created_at":"2026-04-08 09:14:28","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":19046,"visible":true,"origin":"","legend":"\u003cp\u003eAdditional file 1: Search strategy (DOCX)\u003c/p\u003e\n\u003cp\u003eDetailed search strategies used for all databases.\u003c/p\u003e","description":"","filename":"Additionalfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/f68c943d530e56d9c244ba79.docx"},{"id":106403130,"identity":"1c398e4c-93a4-497b-8f86-6dbf27488d1a","added_by":"auto","created_at":"2026-04-08 09:13:38","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":273782,"visible":true,"origin":"","legend":"\u003cp\u003eAdditional file 2: Study characteristics (XLSX)\u003c/p\u003e\n\u003cp\u003eCharacteristics of all included studies.\u003c/p\u003e","description":"","filename":"Additionalfile2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/6ec9dfdfeddc9baddbbe0fde.xlsx"},{"id":106341976,"identity":"0c2d4b12-ec15-413c-918b-5115694312ba","added_by":"auto","created_at":"2026-04-07 15:47:45","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":33615,"visible":true,"origin":"","legend":"\u003cp\u003e\u0026nbsp;Additional file 3: Risk of bias assessment (XLSX)\u003c/p\u003e\n\u003cp\u003eRisk of bias assessments using the RoB 2 tool.\u003c/p\u003e","description":"","filename":"Additionalfile3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/4ec9d9ca5d6d2a03a9900f6e.xlsx"},{"id":106341975,"identity":"c3b4d3ff-1f77-46ff-b53d-d8cdf80d9f64","added_by":"auto","created_at":"2026-04-07 15:47:44","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":247618,"visible":true,"origin":"","legend":"\u003cp\u003eAdditional file 4: Analysis dataset (XLSX)\u003c/p\u003e\n\u003cp\u003eExtracted variables used for descriptive and analytical purposes.\u003c/p\u003e","description":"","filename":"Additionalfile4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9212126/v1/34bc8793edd9234fc2f0b195.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Methodological Integrity of Randomized Controlled Trials in Major Gynaecological Conditions: A Systematic Review","fulltext":[{"header":"BACKGROUND","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eWomen\u0026rsquo;s Health, Structural Inequities, and the Reliability of Evidence\u003c/h2\u003e \u003cp\u003eWomen\u0026rsquo;s health encompasses conditions that are unique to women, more prevalent in women, or that manifest differently across the life course(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). These include gynaecological and reproductive disorders such as premenstrual dysphoric disorder (PMDD), gestational diabetes mellitus (GDM), endometriosis, and polycystic ovary syndrome (PCOS). They also cover more general issues related to physical and mental health throughout life (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Obstetrics and gynaecology (OBGYN) is essential to addressing these medical requirements. It addresses menstrual diseases, fertility, menopause, pregnancy, and childbirth, and reproductive health (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Many female-specific disorders remain underdiagnosed, undertreated, and underresearched despite the substantial global effect of gynaecological diseases (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Endometriosis diagnosis delays, uneven PCOS treatment options, and heterogeneous GDM and PMDD care standards demonstrate persistent gaps in evidence-based clinical practice (\u003cspan additionalcitationids=\"CR8 CR9\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEndometriosis affects an estimated 10% of women of reproductive age worldwide, representing approximately 190\u0026nbsp;million individuals globally (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). PCOS is similarly prevalent, affecting between 6\u0026ndash;13% of women, depending on diagnostic criteria (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). GDM occurs in approximately 1\u0026ndash;14% of pregnancies globally (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e), while PMDD affects an estimated 3\u0026ndash;8% of menstruating individuals (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWomen's health outcomes are still impacted by structural barriers in several contexts. Even though women are more likely than men to experience adverse medication responses, they are still underrepresented in clinical studies. There is inconsistent enforcement of policies requiring representative involvement in research. Furthermore, healthcare policy, training, and data systems frequently fail to appropriately incorporate sex- and gender-specific characteristics (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese differences have obvious repercussions. Hundreds of thousands of pregnant women in Europe do not receive adequate care. Every day, over 800 women worldwide pass away from avoidable pregnancy and childbirth-related causes (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Maternal mortality rates are greater among mothers from ethnic minorities and migrants. This circumstance demonstrates deep structural injustices (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Maternal mortality and national income have a strong negative correlation, according to these global disparities. Maternal mortality rates are significantly higher in low-income nations. This is illustrated in Fig.\u0026nbsp;1.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 1. Maternal mortality ratio in relation to GDP per capita (2020)\u003c/b\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMaternal mortality ratio (deaths per 100,000 live births) plotted against GDP per capita (international dollars, purchasing power parity), both shown on logarithmic scales. The figure illustrates a strong inverse relationship between national income and maternal mortality, with the highest mortality ratios concentrated in low-income countries, particularly in sub-Saharan Africa. Data sources: UN Maternal Mortality Estimation Inter-Agency Group (MMEIG), World Bank, International Monetary Fund (IMF), OECD, and Eurostat. Figure reproduced from Our World in Data.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWhile maternal mortality is a distinct clinical domain, it serves as a proxy for broader systemic inequalities (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). One of the main causes of teenage girls missing school is menstrual issues. This circumstance reinforces gender disparities in long-term economic stability and education. An increased risk of chronic illness in later life is also associated with reproductive health problems. Many of these illnesses, nonetheless, are mislabeled or remain untreated. This fact makes it difficult to monitor and plan interventions effectively (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Women's health requires immediate and continuous care. Not only do women make up the majority of healthcare providers and consumers, but healthy women also translate into healthier families, communities, and economies, making this crucial. Reducing care gaps, saving lives, and promoting a healthier, more equitable future for all will result from addressing the full spectrum of women's health needs (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEvolution of Women\u0026rsquo;s Health Research and Trial Exclusion\u003c/h2\u003e \u003cp\u003eHistorically, women\u0026rsquo;s health research has been characterised by both neglect and exclusion. For much of modern medical history, women were underrepresented in clinical trials due to concerns regarding hormonal variability, reproductive capacity, and pregnancy-related risk(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). New laws and institutional changes brought about a significant change in the 1990s. Among these were the establishment of the NIH Office of Research on Women's Health and the enactment of the NIH Revitalization Act. Research on women's health has advanced thanks to major projects like the Women's Health Initiative (WHI) (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Still, persistent gaps remain in gynaecological and obstetric research. Pregnant individuals continue to be excluded from numerous interventional studies, and marginalised populations, including women from low and middle-income countries, ethnic minority communities, and socioeconomically disadvantaged groups, remain underrepresented in trial populations (\u003cspan additionalcitationids=\"CR25 CR26 CR27\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese structural exclusions carry scientific consequences. When trial populations do not reflect the demographic and socioeconomic diversity of those most affected by the disease, external validity is compromised (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMethodological Complexity in Obstetrics and Gynaecology Trials\u003c/h3\u003e\n\u003cp\u003eRandomized controlled trials are considered the gold standard for evaluating healthcare interventions (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). However, trial design in obstetrics and gynaecology presents unique methodological challenges. Many women's health issues, including menstruation, dysmenorrhea, menopause, incontinence, and sexual health, are still impacted by social stigma. These taboos limit open communication, restrict women from seeking medical assistance, and result in inadequate institutional attention and financing for research (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). As a result, these conditions are often left out of policy, education, and research agendas.\u003c/p\u003e \u003cp\u003eBefore 1993, women were not completely missing from clinical research, but they were often underrepresented (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). As seen during the COVID-19 vaccine launch, the lack of trial data has impacted clinical decision-making, especially in pregnancy (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). Women's involvement in trials is still restricted by other obstacles, including caregiving duties, mistrust of medical institutions, and practical limitations (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese difficulties are made worse by inadequate statistics and reporting. The therapeutic significance of study findings is frequently diminished by small sample sizes, poor treatment of missing data, and reliance on surrogate outcomes. Commonly used outcomes in obstetric trials, such as those assessing labor induction, might not accurately represent significant patient experiences, which would limit their practical application. For example, the commonly used endpoint of a \"live birth,\" which is typically defined as delivery within 24 hours, does not sufficiently convey the complex nature of the induction process or the experiences during labor and the time preceding delivery (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRecent research indicates that, particularly in situations like PCOS, gynaecological studies may be especially susceptible to biased outcome reporting, poor handling of missing data, inadequate blinding, and an excessive dependence on surrogate biochemical endpoints. Moreover, outcome heterogeneity is still widespread, which hinders meaningful meta-analysis and cross-trial comparability (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). Although initiatives such as the Core Outcomes in Women\u0026rsquo;s Health (CROWN) project have sought to standardise outcome reporting (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e), adoption has been inconsistent and poorly quantified across conditions.\u003c/p\u003e\n\u003ch3\u003eThe Need for a Structural Audit of Trial Methodology\u003c/h3\u003e\n\u003cp\u003eMost systematic reviews in women\u0026rsquo;s health focus on intervention effectiveness rather than interrogating the methodological architecture of the trials themselves. Yet, the reliability of pooled estimates and subsequent guideline recommendations depends fundamentally on trial integrity. Evidence suggests that the methodological quality of both systematic reviews and underlying clinical trials in women\u0026rsquo;s health is often suboptimal, with frequent high risks of bias in key domains such as blinding and allocation concealment (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). This is consequential because empirical meta-epidemiological research demonstrates that such methodological shortcomings can systematically exaggerate intervention effect estimates and increase between-study heterogeneity (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Although systematic reviews and meta-analyses are positioned at the apex of the evidence hierarchy, their conclusions are inherently constrained by the quality of the included studies, meaning that biased primary data can propagate uncertainty rather than resolve it (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn addition, given the global disparities illustrated by income-linked health outcomes (Fig.\u0026nbsp;1), it is plausible that trial-level characteristics, including country income classification, registration practices, and publication era, may be associated with measurable differences in methodological quality. Understanding these patterns is essential for identifying structural weaknesses within the evidence base and informing strategies to strengthen future research design.\u003c/p\u003e\n\u003ch3\u003eStudy Objectives\u003c/h3\u003e\n\u003cp\u003e In alignment with the PROSPERO registered review objectives, the present study aimed to systematically characterise the clinical trial methodologies used in randomized controlled trials evaluating interventions for endometriosis, PCOS, GDM, and PMDD. Drawing on 695 Randomized controlled (RCTs) published between 2005 and 2025, this review evaluates risk-of-bias profiles, reporting completeness, and trial-level predictors of methodological quality. By providing a large-scale structural assessment of trial integrity in contemporary obstetric and gynaecological research, this study seeks to inform improvements in rigor, transparency, and equity within the women\u0026rsquo;s health evidence ecosystem.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Reporting Standards\u003c/h2\u003e \u003cp\u003eThis study was conducted as a systematic review of RCTs evaluating methodological characteristics in obstetric and gynaecological research. The review followed a pre-specified protocol and was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). The PRISMA 2020 checklist is provided in Additional file 5. The review was prospectively registered in PROSPERO (CRD42025636543) (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEligibility Criteria\u003c/h3\u003e\n\u003cp\u003eEligible studies were randomized controlled trials published between 1 January 2005 and 1 October 2025. The timeframe was selected to reflect contemporary methodological standards, including widespread implementation of trial registration and reporting guidelines (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStudies were included if they evaluated interventions for one of four prespecified conditions: endometriosis, PCOS, GDM, or PMDD. Trials enrolling adult women or individuals with a uterus aged 18 years or older were eligible.\u003c/p\u003e \u003cp\u003eParallel-group, cluster-randomized, and crossover designs were included. Non-randomized studies, observational designs, conference abstracts, study protocols without results, retracted publications, and animal studies were excluded. Secondary analyses of previously published trials were excluded to ensure that assessed methodological characteristics reflected the primary trial design and conduct. Only peer-reviewed articles published in English were included.\u003c/p\u003e\n\u003ch3\u003eInformation Sources and Search Strategy\u003c/h3\u003e\n\u003cp\u003eComprehensive searches were conducted in PubMed, ScienceDirect, and Google Scholar. Search strategies combined controlled vocabulary terms and free-text keywords relating to the four target conditions and a randomized controlled trial design. The databases PubMed and ScienceDirect were last searched on 20 October 2025, and Google Scholar was last searched on 2 November 2025. Boolean operators were applied to optimise sensitivity. The PubMed search combined terms for endometriosis, PCOS, GDM, and PMDD with filters for randomized controlled trials. Searches were restricted to studies published between 1 January 2005 and 1 October 2025 and limited to English-language publications involving human adults. The full PubMed search strategy is provided in Additional file 1.\u003c/p\u003e \u003cp\u003eThe ScienceDirect search strategy was adapted from the PubMed query using equivalent keywords and Boolean operators:\u003c/p\u003e \u003cp\u003e(\u0026ldquo;endometriosis\u0026rdquo; OR \u0026ldquo;PCOS\u0026rdquo; OR \u0026ldquo;gestational diabetes\u0026rdquo; OR \u0026ldquo;PMDD\u0026rdquo;)\u003c/p\u003e \u003cp\u003eAND (\u0026ldquo;randomized controlled trial\u0026rdquo; OR \u0026ldquo;RCT\u0026rdquo;).\u003c/p\u003e \u003cp\u003eGoogle Scholar searches were conducted using simplified keyword combinations: (\u0026ldquo;endometriosis\u0026rdquo; OR \u0026ldquo;PCOS\u0026rdquo; OR \u0026ldquo;gestational diabetes\u0026rdquo; OR \u0026ldquo;PMDD\u0026rdquo;) AND (\u0026ldquo;randomized controlled trial\u0026rdquo; OR \u0026ldquo;RCT\u0026rdquo;). Due to the large volume of results and limitations in reproducibility, screening was restricted to the first 500 results sorted by relevance, consistent with methodological recommendations.\u003c/p\u003e \u003cp\u003eReference lists of included studies were screened to identify additional eligible trials.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStudy Selection\u003c/h2\u003e \u003cp\u003eStudy selection was conducted independently by three reviewers at both title/abstract and full-text screening stages. Discrepancies were resolved through discussion and consensus. All retrieved records were imported into Covidence software for screening and duplicate removal. Covidence automation tools were used to assist in identifying records that were likely ineligible based on predefined exclusion criteria (e.g., non-randomized designs, irrelevant populations). These records were reviewed to confirm exclusion, ensuring that no studies were excluded without human assessment. Screening was conducted in two stages: title and abstract screening followed by full-text review. Eligibility criteria were applied consistently at both stages using a pre-piloted screening framework.\u003c/p\u003e \u003cp\u003eThe study selection process was documented using a PRISMA flow diagram. A total of 695 trials were included in the final synthesis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eData Extraction\u003c/h2\u003e \u003cp\u003eA structured data extraction template was developed and piloted before full extraction. Extracted variables included publication year, trial registration status, funding source, country of conduct, World Bank income classification, sample size, trial design type, centre status (single vs multicentre), randomisation method, allocation concealment, blinding practices, outcome domains, intention-to-treat reporting, and handling of missing data.\u003c/p\u003e \u003cp\u003eThe primary outcomes of this review were methodological quality, assessed using the Cochrane Risk of Bias 2 (RoB 2) tool, and reporting quality, assessed using CONSORT 2010 adherence. Secondary outcomes included associations between trial-level characteristics (e.g., trial registration, publication period, and country income level) and these methodological outcomes. Outcome domains reported within included trials were categorised as patient-reported, clinical/physiological, biochemical, reproductive, or safety outcomes. Where multiple outcome measures or time points were reported within a domain, all relevant data were extracted. No selection based on statistical significance was performed. Equity-related variables were extracted but analysed narratively due to inconsistent reporting.\u003c/p\u003e \u003cp\u003eWhere data were missing or unclear, variables were coded as \u0026lsquo;not reported\u0026rsquo;, and no imputation was performed. Extracted variables were coded into predefined categories, with \u0026lsquo;not reported\u0026rsquo; treated as a separate category. Trial registration status was categorised as prospectively registered versus not registered or retrospectively registered. Publication year was grouped into predefined periods (2005\u0026ndash;2009, 2010\u0026ndash;2014, 2015\u0026ndash;2019, 2020\u0026ndash;2025). Risk-of-bias judgments were dichotomised (low vs not low) for primary analyses, with alternative dichotomisation (high vs not high) used in sensitivity analyses. CONSORT checklist items were coded as binary variables and aggregated into percentage adherence scores.\u003c/p\u003e \u003cp\u003eData extraction was conducted independently by two reviewers, with discrepancies resolved through discussion and, when necessary, adjudication by a third reviewer.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eRisk of Bias Assessment\u003c/h2\u003e \u003cp\u003eRisk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). Risk of bias assessment was conducted independently by two reviewers, with disagreements resolved through discussion or consultation with a third reviewer. The tool evaluates five domains: bias arising from the randomisation process, bias due to deviations from the intended interventions, bias due to missing outcome data, bias in the measurement of outcomes, and bias in the selection of the reported result. Each domain was classified as low risk, some concerns, or high risk, resulting in an overall risk of bias judgment.\u003c/p\u003e \u003cp\u003eFor regression analyses, the overall risk of bias was dichotomised as low versus not low risk, consistent with methodological guidance when event counts are limited.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eReporting Quality Assessment\u003c/h2\u003e \u003cp\u003eReporting completeness was evaluated using the Consolidated Standards of Reporting Trials (CONSORT) 2010 checklist (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). CONSORT checklist items were operationalised as binary variables (reported vs not reported/unclear) and aggregated into a percentage adherence score, consistent with prior methodological studies evaluating reporting quality. Equal weighting was applied across items to provide an overall measure of reporting completeness. Sub-items were evaluated separately to allow granular assessment of reporting deficiencies.\u003c/p\u003e \u003cp\u003eA total CONSORT score (maximum 37 items) and percentage adherence score were calculated for each trial. Percentage adherence was used as the primary reporting quality metric. Each study was evaluated across relevant checklist domains to generate an overall adherence score; however, item-level scoring data were not retained for all studies.\u003c/p\u003e \u003cp\u003eItems deemed not applicable to specific study designs were excluded from scoring for that trial. Partial or ambiguous reporting was conservatively classified as not reported.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eData Synthesis and Statistical Analysis\u003c/h2\u003e \u003cp\u003eAll included studies contributed to all analyses. Variables that were missing or not reported were coded as \u0026lsquo;not reported\u0026rsquo; and retained in the dataset, allowing all 695 trials to be included in descriptive and regression analyses.\u003c/p\u003e \u003cp\u003eA descriptive synthesis was conducted to summarise trial characteristics and methodological features using counts, percentages, means, medians, and ranges. Findings were stratified by condition where appropriate.\u003c/p\u003e \u003cp\u003eData analyses were conducted using SPSS version 31 (IBM Corp., Armonk, NY, USA). Associations between trial-level characteristics and the likelihood of being classified as low risk of bias were explored using univariable binary logistic regression analyses. Effect measures included odds ratios (ORs) with 95% confidence intervals for logistic regression analyses. For regression analyses, the outcome variable was overall risk of bias, dichotomised as low risk versus not low risk (including some concerns and high risk), consistent with methodological guidance for analyses with limited event counts. This dichotomisation was applied to ensure sufficient statistical power and stability of estimates, given the relatively small number of trials classified as low risk of bias. Explanatory variables were selected a priori based on methodological relevance and included trial registration status, publication period, and World Bank country income classification.\u003c/p\u003e \u003cp\u003eVariables directly contributing to the risk-of-bias assessment, such as blinding and allocation concealment, were not modelled as independent predictors to avoid circular inference.\u003c/p\u003e \u003cp\u003eAssociations between trial characteristics and reporting quality were examined using multivariable linear regression with CONSORT percentage adherence as the dependent variable and trial registration status, publication period, clinical condition, and country income level as prespecified predictors. For linear regression analyses, CONSORT adherence percentage was treated as a continuous outcome variable. Effect measures included β coefficients for linear regression models.\u003c/p\u003e \u003cp\u003eAnalyses were conducted using complete-case data, and variables with missing data were not imputed. Model assumptions for logistic and linear regression, including independence of observations and linearity of relationships, were considered during analysis, and no major violations were identified based on inspection of model outputs. All extracted variables and analytical data are available in Additional file 4.\u003c/p\u003e \u003cp\u003eDue to substantial heterogeneity in interventions, outcome measures, and reporting practices, a meta-analysis was not undertaken. No formal statistical assessment of reporting bias was conducted due to the absence of a meta-analysis.\u003c/p\u003e \u003cp\u003eResults were presented using structured tables and graphical displays, including summary tables of trial characteristics, regression outputs, and figures illustrating study selection, temporal trends, geographical distribution, and risk-of-bias assessments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eSubgroup and Sensitivity Analyses\u003c/h2\u003e \u003cp\u003ePrespecified subgroup analyses examined differences by clinical condition, outcome type (subjective versus objective), and country income classification. Sensitivity analyses explored alternative dichotomisations of overall risk of bias, including comparisons of high-risk trials versus all other classifications.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eUse of Large Language Models\u003c/h2\u003e \u003cp\u003eNo large language models or artificial intelligence tools were used in the design, analysis, or writing of this study.\u003c/p\u003e \u003cp\u003eThe conceptual framework underpinning the evaluation of methodological characteristics and risk of bias in gynaecological randomized controlled trials is presented in Fig.\u0026nbsp;2.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 2. Conceptual framework for evaluating methodological characteristics and risk of bias in gynaecological randomised controlled trials\u003c/b\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eOverall, the included trials demonstrated substantial heterogeneity in sample size, reporting practices, and methodological design across conditions. Structural characteristics such as trial registration, funding transparency, and country income level varied considerably, with notable implications for methodological quality and reporting completeness.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDatabase searches identified randomised controlled trials evaluating interventions for endometriosis, PCOS, GDM, and PMDD. A total of 6,937 records were identified through database searches. After removal of duplicates and records marked as ineligible by automation tools, 4,113 records were screened by title and abstract. Of these, 3,055 records were excluded. Covidence automation tools were used to assist screening; however, all eligibility decisions were confirmed through reviewer assessment.\u003c/p\u003e\n\u003cp\u003eA total of 1,058 full-text articles were sought for retrieval, of which 22 could not be retrieved in full text despite attempts through institutional access, interlibrary requests, and online sources, and were therefore excluded. The remaining 1,036 articles were assessed for eligibility, with 341 studies excluded for predefined reasons. A total of 695 randomized controlled trials were included in the final synthesis.\u003c/p\u003e\n\u003cp\u003eThe study selection process is presented in the PRISMA flow diagram (Figure 3)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 3: PRISMA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOverall characteristics of included trials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe included trials demonstrated substantial heterogeneity in scale, design, and reporting practices. Sample sizes ranged from small single-centre studies enrolling fewer than 20 participants to large multicentre and population-based trials enrolling several thousand participants. The characteristics of included studies are presented in Additional file 2.\u003c/p\u003e\n\u003cp\u003eAcross the included randomised controlled trials, sample sizes varied substantially both within and across gynaecological conditions, with consistently right-skewed distributions reflecting the presence of a small number of large trials alongside many small studies.\u003c/p\u003e\n\u003cp\u003eAmong the 16 trials evaluating interventions for PMDD, sample sizes were generally small. The median sample size was 40 participants (range 27\u0026ndash;252), with a mean of 84 (SD 80.8), indicating a markedly right-skewed distribution. Most PMDD trials (10/16) enrolled fewer than 50 participants, while only four trials included 100 or more participants, suggesting that the evidence base is largely derived from small-scale studies.\u003c/p\u003e\n\u003cp\u003eIn the 117 endometriosis trials, sample sizes were more broadly distributed but remained highly variable. The median sample size was 67 participants (range 18\u0026ndash;1,689), with a mean of 151.49 (SD 233.29). While a substantial proportion of trials were small (34 studies enrolled fewer than 50 participants), a similar number recruited between 50 and 99 participants (43 trials), and 40 trials included 100 or more participants. This pattern reflects considerable heterogeneity in trial scale, with a minority of large trials contributing disproportionately to the overall sample size profile.\u003c/p\u003e\n\u003cp\u003eThe 364 trials evaluating interventions for PCOS also demonstrated wide variability in sample size. The median sample size was 69 participants (range 10\u0026ndash;1,508), with a mean of 104.9 (SD 143.1), again indicating right skewness. Trials enrolling between 50 and 99 participants were most common (148 studies), while 104 trials enrolled fewer than 50 participants and 112 included 100 or more participants.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn contrast, the 198 trials evaluating interventions for GDM tended to be substantially larger. The median sample size was 132 participants (range 12\u0026ndash;47,080), while the mean was 693.96 (SD 3,933.06), reflecting extreme right skewness driven by a small number of very large trials. Most GDM trials (124 studies) enrolled 100 or more participants, with fewer trials recruiting between 50 and 99 participants (52 studies) or fewer than 50 participants (22 studies). Compared with the other conditions examined, the GDM evidence base included a higher proportion of large-scale trials. Table 1 shows the sample size characteristics across the four conditions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1: Sample size characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCondition\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrials (n)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian sample size\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean (SD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRange\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;50 n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e50\u0026ndash;99 n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026ge;100 n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePMDD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e84 (80.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e27\u0026ndash;252\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10 (62.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (12.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (25.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEndometriosis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e117\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e151.49 (233.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e18\u0026ndash;1,689\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e34 (29.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e43 (36.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e40 (34.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePCOS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e364\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e104.9 (143.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10\u0026ndash;1,508\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e104 (28.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e148 (40.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e112 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGDM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e198\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e132\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e693.96\u003c/p\u003e\n \u003cp\u003e(3,923.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12\u0026ndash;47,080\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e22(11.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e52 (26.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e125 (62.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePublication years spanned 2005 to 2025, with a marked increase in trial volume after 2010. Figure 4 compares the Publication years across the 4 different conditions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 4. Publication timeline of randomised controlled trials across four gynaecological conditions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBars represent the number of trials published within each time interval, stratified by condition. Across all conditions, trial volume increased markedly after 2010, with the greatest concentration of studies published between 2020 and 2025.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTrial characteristics varied systematically by condition. GDM trials were generally larger and more frequently population-based, whereas PMDD trials were predominantly small. Endometriosis and PCOS trials showed the widest variability in sample size, reflecting a mixture of exploratory studies, mid-sized trials, and large multicentre investigations.\u003c/p\u003e\n\u003cp\u003eAcross all conditions, most trials reported the use of randomisation; however, the level of methodological detail varied considerably. Allocation concealment, blinding procedures, and analytical approaches were less consistently reported, contributing to heterogeneity in methodological quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDescriptive methodological characteristics relevant to trial quality\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTrial registration status\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eProspective trial registration was reported in a substantial proportion of included trials, although reporting practices varied by condition. Among endometriosis trials, 78 of 117 studies (66.7%) were prospectively registered, while 39 trials (33.3%) did not report trial registration. No trials were retrospectively registered. For PCOS, 224 of 364 trials (61.5%) were prospectively registered, 4 trials (1.1%) were retrospectively registered, and 136 trials (37.4%) did not report registration. Trial registration was most consistently reported among GDM trials. Of 198 trials, 165 (83.3%) were prospectively registered, 2 (1.0%) were retrospectively registered, and 31 (15.7%) did not report registration. In PMDD, 11 of 16 trials (68.8%) were prospectively registered, while 5 trials (31.3%) did not report trial registration. No PMDD trials were retrospectively registered. Overall, while prospective registration was common, non-reporting of registration remained prevalent, particularly in endometriosis and PCOS trials.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFunding sources\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eReporting of funding sources was heterogeneous across all conditions. Public or academic funding predominated across the evidence base, particularly in GDM (60.1%) and PCOS (51.1%) trials. Industry funding or support was reported in a substantial minority of trials across all conditions, most frequently in pharmacological studies, while mixed funding arrangements were also common. Non-reporting of funding source remained prevalent, particularly in endometriosis (12.8%) and PCOS (13.5%) trials, limiting full assessment of potential funding-related bias.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGeographical distribution and country income level\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe geographical distribution of trials was uneven across conditions. Endometriosis and PMDD trials were predominantly conducted in high-income countries, with limited representation from lower- and lower-middle-income settings and few studies conducted across mixed-income contexts. PCOS trials demonstrated broader geographical spread, with substantial representation from upper-middle-income countries, although inclusion of lower- and lower-middle-income settings remained limited, and only a small number of studies were conducted in mixed-income settings. GDM trials were largely conducted in high-income and upper-middle-income countries, with very few trials spanning mixed-income contexts. Overall, evidence from low- and lower-middle-income countries remained sparse across all conditions, limiting generalisability to diverse healthcare settings. Figure 5 shows the income level of the settings across the four conditions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 5. Geographical distribution of randomised controlled trials by World Bank income classification.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStacked bars represent the proportion of trials conducted in high-, upper-middle-, lower-/low-income, and mixed-income settings for each condition. Across all conditions, trials were predominantly conducted in high-income and upper-middle-income countries, with limited representation from lower-income settings.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOverall risk of bias\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAcross all 695 included trials, only 49 (7%) were judged to be at low risk of bias, whereas the majority were assessed as having some concerns or high risk of bias using the RoB 2 tool. A summary Plot of The Risk of Bias created using the Risk of Bias Visualization tool is represented in Figure 6. When stratified by condition, GDM trials demonstrated the most favourable overall risk of bias profile. Of 198 GDM trials, 15 (7.6%) were judged to be at low risk of bias, 148 (74.7%) had some concerns, and 35 (17.7%) were judged to be at high risk. In contrast, PMDD trials most frequently demonstrated elevated risk of bias. Of 16 PMDD trials, 1 (6.3%) was judged to be at low risk, 11 (68.8%) had some concerns, and 4 (25.0%) were classified as high risk. Among endometriosis trials (n = 117), 8 trials (6.84%) were judged to be at low risk of bias, while 66 (56.41%) had some concerns, and 43 (36.75%) were judged to be at high risk. For PCOS, only 25 of 364 trials (6.9%) were classified as low risk, while 172 (47.3%) had some concerns, and 167 (45.9%) were assessed as high risk of bias. Study-level risk of bias judgments across the five RoB 2 domains and overall risk are summarized in Additional file 3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 6. Summary Plot of Cochrane Risk of Bias 2\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs detailed in Table 2, multiple structural characteristics of trial design and reporting were associated with increased susceptibility to bias, particularly selective reporting, performance bias, and analytical bias.\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eTable 2. Structural factors associated with elevated risk of bias and limited transparency in gynaecological randomised controlled trials (n = 695)\u003c/strong\u003e\u003c/h3\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStructural factor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEvidence from the dataset\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eThe main type of bias introduced\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHow does this affect methodological quality\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eImpact on transparency\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbsence of prospective trial registration\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e33.3% of endometriosis trials, 37.4% of PCOS trials, and 31.3% of PMDD trials were unregistered, compared with 15.7% of GDM trials\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSelective outcome reporting bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAllows outcomes to be changed or selectively reported after results are known; prevents verification of prespecified analyses\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigh\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSmall sample size\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e62.5% of PMDD trials and ~30% of endometriosis and PCOS trials enrolled fewer than 50 participants\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAttrition bias, reporting bias, and performance bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLimits statistical power, increases sensitivity to missing data, reduces the feasibility of blinding, and encourages reporting of multiple exploratory outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigh\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eReliance on subjective outcomes (e.g., pain, mood, quality of life)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDominant in endometriosis (pain in 88% of trials) and PMDD (symptom outcomes in 87.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePerformance and detection bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eParticipant expectations and assessor knowledge can influence outcome measurement when blinding is incomplete\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigh\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eUse of multiple biochemical or surrogate outcomes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e84.6% of PCOS trials reported biochemical outcomes, often multiple per study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSelective reporting bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIncreases flexibility to report only statistically significant results and obscures true clinical relevance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigh\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-pharmacological intervention designs (surgical, behavioural, lifestyle)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCommon in endometriosis and PCOS trials\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePerformance and measurement bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eParticipant blinding is often infeasible; intervention adherence varies and is difficult to standardise\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eModerate\u0026ndash;High\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncomplete reporting of the funding source\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12.8% of endometriosis trials and 13.5% of PCOS trials did not disclose funding\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eReporting bias; potential conflict-of-interest bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eObscures financial influences on outcome prioritisation, comparator choice, and interpretation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eModerate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eConduct in lower-resource research settings\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTrials from lower- and lower-middle-income countries were sparse but showed poorer reporting completeness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSelection and reporting bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLimited regulatory oversight and research infrastructure reduce adherence to reporting standards\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eModerate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFailure to report intention-to-treat analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eReported in only 37.5\u0026ndash;41.8% of PMDD, endometriosis, and PCOS trials (55% in GDM)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAttrition and analytical bias\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eExcluding participants post-randomisation distorts treatment effect estimates\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eModerate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBias arising from the randomisation process\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAcross all conditions, most trials reported the use of random allocation; however, detailed reporting of sequence generation and allocation concealment was frequently absent, resulting in many trials being judged as having some concerns in this domain. This limitation was particularly evident in PCOS and endometriosis trials, where randomisation methods were often mentioned without further specification. In contrast, GDM trials more frequently reported computer-generated or centralised randomisation, contributing to a lower proportion of high-risk judgments in this domain.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBias due to deviations from intended interventions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBias related to deviations from intended interventions was one of the most prevalent sources of bias across all conditions. Incomplete reporting of participant and personnel blinding was common, particularly in behavioural, lifestyle, and surgical intervention trials. This domain was especially problematic in endometriosis and PMDD trials, where subjective symptom outcomes were primary endpoints, and blinding was frequently incomplete or absent. Across conditions, trials without participant blinding constituted a substantial proportion of the evidence base, contributing to the elevated risk of performance bias.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBias due to missing outcome data\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMissing outcome data contributed substantially to the elevated risk of bias across all conditions. Attrition was widespread in trials with longer durations and those assessing behavioral or lifestyle interventions. Reporting of reasons for dropout and methods for handling missing data was inconsistent, and intention-to-treat analyses were not uniformly applied. This domain was a notable source of bias in PCOS and PMDD trials, whereas GDM trials generally demonstrated lower attrition.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBias in the measurement of outcomes\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRisk of bias in outcome measurement varied systematically by outcome type. Trials relying on objective biochemical, metabolic, or obstetric outcomes, particularly among GDM trials, were generally assessed as having low risk of bias in this domain. In contrast, trials evaluating subjective outcomes, such as pain, mood, or quality of life, were frequently judged to be at higher risk, especially when blinding of outcome assessors was unclear or not reported. This issue was most prominent in endometriosis and PMDD trials, where patient-reported outcomes constituted primary endpoints.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBias in the selection of the reported result\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSelective outcome reporting was a major concern across the evidence base. In many trials, particularly those that were not prospectively registered, prespecified outcomes could not be verified. Discrepancies between methods and results sections were frequently observed, including selective reporting of statistically significant outcomes.\u003c/p\u003e\n\u003cp\u003eThis domain contributed substantially to overall risk of bias classifications, particularly among PCOS trials, which often assessed numerous biochemical and surrogate outcomes without clear prioritisation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePredictors of Overall Risk of Bias\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnivariable logistic regression analyses were conducted to examine associations between selected trial characteristics and the likelihood of being assessed as low risk of bias. Results are presented as odds ratios (ORs) with 95% confidence intervals.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTrial registration\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eProspective trial registration was strongly associated with improved methodological quality, as described in Table 3. Prospectively registered trials had significantly higher odds of being assessed as low risk of bias compared with trials that were retrospectively registered or unregistered (OR 4.08, 95% CI 1.59\u0026ndash;10.47; p = 0.003).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Association between trial registration and low risk of bias\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePredictor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eProspectively Registered vs not registered/ Retrospectively registered\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.59\u0026ndash;10.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: OR, Odd ratio; CI, Confidence interval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePublication period (year group)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCompared with trials published in the earliest period, studies published in more recent periods demonstrated higher odds of being assessed as low risk of bias; however, these associations did not reach statistical significance, and confidence intervals were wide, indicating substantial uncertainty. This association is presented in Table 4.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4. Association between publication period and low risk of bias\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePublication period\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2005-2009 (reference)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026ndash;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026ndash;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2010-2014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.77\u0026ndash;16.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.103\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2015-2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.59\u0026ndash;12.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.200\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2020-2025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.71\u0026ndash;13.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.131\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: OR, Odd ratio; CI, Confidence interval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCountry income level\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTrials conducted in upper-middle-, lower-, and mixed-income countries did not demonstrate significantly different odds of being assessed as low risk of bias compared with trials conducted in high-income countries. Estimates for lower-/low- and mixed-income settings were imprecise, with wide confidence intervals reflecting sparse data. The results are presented in Table 5.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 5. Association between country income level and low risk of bias\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncome level\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHigh income (reference)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026ndash;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026ndash;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eUpper-middle income\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.61\u0026ndash;2.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.716\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLower/Low income\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.04\u0026ndash;2.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.213\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMixed income\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.64-15.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.159\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eEstimates for lower-/low-income settings were unstable due to sparse data. Abbreviations: OR, Odd ratio; CI, Confidence interval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSensitivity analysis of risk of bias classification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA sensitivity analysis was conducted using an alternative dichotomisation of overall risk of bias, comparing trials classified as high risk of bias with all other trials. In this analysis, the direction of associations between trial characteristics and risk of bias was broadly consistent with the primary analysis. Prospective trial registration was associated with lower odds of being classified as high risk of bias, although this association did not reach conventional statistical significance (OR 0.70, 95% CI 0.48\u0026ndash;1.04; p = 0.077). Associations with publication period, country income level, and clinical condition were weaker and not statistically significant. These findings suggest that the primary results were not driven by the specific definition of the low risk of bias category.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eReporting quality (CONSORT adherence)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOverall reporting quality, as assessed using CONSORT percentage adherence, was moderate across included trials and is presented in Table 6. The mean CONSORT adherence was 73.8%, with a median adherence of 75.7%. Adherence values ranged widely, from 24.3% to 97.3%, indicating substantial variability in the completeness of reporting across studies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 6. Overall CONSORT reporting adherence across included trials\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatistic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eValue (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e73.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMedian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e75.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMinimum\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e24.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMaximum\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e97.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eWhen stratified by publication period, mean CONSORT adherence demonstrated a modest upward trend over time. Trials published between 2005\u0026ndash;2009 had a mean adherence of 62.6%, compared with 71.5% for trials published between 2010\u0026ndash;2014, 74.7% for trials published between 2015\u0026ndash;2019, and 77.3% for trials published between 2020\u0026ndash;2025.\u003c/p\u003e\n\u003cp\u003eWhen stratified by clinical condition, mean CONSORT adherence varied across women\u0026rsquo;s health conditions. Trials in GDM and PMDD demonstrated the highest mean adherence (77.55% and 77.21%, respectively), followed by endometriosis (75.67%), while trials in PCOS showed lower mean adherence (71.02%).\u003c/p\u003e\n\u003cp\u003eTrials that were prospectively registered demonstrated higher reporting quality compared with unregistered trials. Mean CONSORT adherence was 77.6% among registered trials, compared with 65.4% among unregistered trials.\u003c/p\u003e\n\u003cp\u003eTrials assessed as low risk of bias demonstrated higher reporting quality compared with trials assessed as high or unclear risk of bias. Mean CONSORT adherence was 81.3% among trials at low risk of bias, compared with 73.3% among trials at high or unclear risk of bias.\u003c/p\u003e\n\u003cp\u003eOverall, these findings indicate moderate reporting completeness across trials, with higher adherence observed among registered studies, more recent publications, and trials assessed as low risk of bias.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eReporting quality regression analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn multivariable linear regression analysis, prospective trial registration was independently associated with higher CONSORT adherence, with registered trials demonstrating an average increase of \u0026nbsp;in reporting completeness compared with unregistered trials (p \u0026lt; .001). Reporting completeness increased across successive publication periods, and modest differences were observed between clinical conditions. Trials conducted in upper-middle- and lower-/low-income settings demonstrated significantly lower reporting quality compared with those from high-income countries, while no difference was observed for trials conducted in mixed-income settings. The model demonstrated moderate explanatory power for variation in CONSORT adherence (R\u0026sup2; = 0.23). The multivariable linear regression is presented in Table 7.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 7. Multivariable linear regression of factors associated with CONSORT adherence (%)\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePredictor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026beta; (percentage points)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e95% CI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTrial registration (registered vs not)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6.27 to 10.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePublication period (per category increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.50 to 3.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eClinical condition\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026mdash;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026mdash;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e.015\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eUpper-middle-income vs High-income\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;3.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;5.33 to \u0026minus;1.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLow-income vs High-income\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;6.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;10.03 to \u0026minus;2.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMixed-income vs High-income\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026minus;7.54 to 7.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.965\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: CI, Confidence interval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTime-trend analysis of reporting quality\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnivariable linear regression demonstrated a significant improvement in reporting quality over time, indicating a clear temporal trend toward more complete CONSORT reporting in gynaecological randomised controlled trials. Reporting quality increased across successive publication periods. Publication period explained approximately 10% of the variance in reporting completeness (R\u0026sup2; = 0.099). This temporal association remained evident, although attenuated, in multivariable analyses adjusting for trial registration, clinical condition, and country income level. The results are presented in Table 8.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 8. Univariable linear regression of publication period and CONSORT adherence (%)\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePredictor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026beta; (percentage points)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStandard Error\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePublication period (ordered)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eConstant\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e61.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eModel fit\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eF-statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e75.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eR\u0026sup2;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.099\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e695\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003ePublication period was modelled as an ordered categorical variable, with higher values corresponding to more recent publication periods.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Key methodological characteristics across conditions are summarised in Table 9.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 9. Methodological characteristics of included randomised controlled trials by condition\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMethodological characteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePMDD (n = 16)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEndometriosis (n = 117)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePCOS (n = 364)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGDM (n = 198)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eParticipant blinding reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12 (75.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e56 (47.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e142 (39.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e39 (19.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eOutcome assessor blinding reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (43.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e63 (53.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e113 (31.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e85 (42.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRandom sequence generation reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14 (87.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e103 (88.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e267 (73.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e170 (85.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAllocation concealment reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9 (56.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e73 (62.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e209 (57.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e146 (73.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIntention-to-treat analysis reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6 (37.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e47 (40.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e152 (41.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e109 (55.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations: PMDD, premenstrual dysphoric disorder; PCOS, polycystic ovary syndrome; GDM, gestational diabetes mellitus. Participant blinding and outcome assessor blinding were classified as present only when explicitly reported as fully blinded. Random sequence generation and allocation concealment were classified as present when clearly described. Intention-to-treat (ITT) analyses were classified as present when explicitly stated; partial or modified ITT and per-protocol or complete-case analyses were not classified as ITT.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePatterns of outcome selection and heterogeneity across conditions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOutcome reporting demonstrated substantial heterogeneity both within and across the four gynaecological conditions examined. Trials frequently assessed multiple outcome domains, and no uniform pattern of outcome prioritisation was observed across conditions. This heterogeneity encompassed not only the type of outcomes selected, but also the balance between objective and subjective measures, the clinical relevance of endpoints, and the degree of alignment with patient-centred priorities.\u003c/p\u003e\n\u003cp\u003eAmong endometriosis trials, pain-related outcomes were reported in 88% of studies, patient-reported outcomes in 70%, biochemical outcomes in 52%, safety outcomes in 55%, reproductive outcomes in 29%, and imaging or surgical outcomes in 22%.\u003c/p\u003e\n\u003cp\u003eIn PCOS trials, biochemical or physiological outcomes were reported in 85%, clinical symptom outcomes in 76%, reproductive outcomes in 59%, patient-reported outcomes in 36%, psychological or functional outcomes in 32%, and safety outcomes in 25%.\u003c/p\u003e\n\u003cp\u003eAmong GDM trials, maternal or neonatal clinical outcomes were reported in 86%, biochemical outcomes in 81%, behavioural or lifestyle outcomes in 48%, patient-reported outcomes in 32%, safety outcomes in 26%, and economic outcomes in 9%.\u003c/p\u003e\n\u003cp\u003ePMDD trials predominantly reported clinical symptom outcomes 87%, \u0026nbsp;patient-reported outcomes 75%, psychological or functional outcomes 69%, biochemical outcomes 56%, and behavioural outcomes 50%.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eCross-condition heterogeneity\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eMarked heterogeneity was observed across conditions in both the \u003cstrong\u003etype\u003c/strong\u003e and \u003cstrong\u003edistribution\u003c/strong\u003e of outcomes assessed:\u003c/p\u003e\n\u003cul class=\"decimal_type\"\u003e\n \u003cli\u003eEndometriosis and PMDD trials prioritised subjective symptom-based and patient-reported outcomes.\u003c/li\u003e\n \u003cli\u003ePCOS trials prioritised biochemical and reproductive surrogate endpoints.\u003c/li\u003e\n \u003cli\u003eGDM trials prioritised objective maternal and neonatal clinical outcomes.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe proportion of trials including patient-reported outcomes ranged from 31.6% in GDM to 75.0% in PMDD, while biochemical outcomes ranged from 56.3% in PMDD to 84.6% in PCOS. Reproductive outcomes were common in PCOS (58.8%) but uncommon in endometriosis (29.1%) and largely absent from PMDD and GDM trials.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eStructural features of outcome heterogeneity\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eOutcome heterogeneity was not limited to domain selection. Substantial variation was also observed in:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eoutcome definitions (e.g., differing pain scales and biochemical thresholds),\u003c/li\u003e\n \u003cli\u003emeasurement instruments,\u003c/li\u003e\n \u003cli\u003etiming of outcome assessment,\u003c/li\u003e\n \u003cli\u003eand prioritisation of primary versus secondary endpoints.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eMany trials reported multiple outcomes across several domains without a clear specification of primary outcomes, particularly in PCOS research, where numerous biochemical markers were often assessed concurrently.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003e\u003cem\u003eInterpreting the overall methodological landscape\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis systematic review provides one of the largest structural assessments to date of methodological quality in randomized controlled trials addressing major gynaecological conditions. Only a small percentage of studies meet the low-risk-of-bias criteria. Despite the large number of trials identified, high methodological quality remains relatively uncommon across the evidence base. This discovery has substantial implications for the degree to which clinical practice, the creation of guidelines, and policy decisions pertaining to women\u0026apos;s health can be confidently informed by existing research.\u003c/p\u003e\n\u003cp\u003eThese findings align with broader critiques of women\u0026rsquo;s health research, which emphasise that conventional methodologies often fail to capture complexity and limit the translation of evidence into practice, underscoring the need for methodological innovation and integration across research approaches (47). The majority of trials that were deemed to have some issues indicate that many research fall into a methodological \u0026quot;grey zone\u0026quot; where they are neither obviously defective nor rigorous enough to produce high-certainty evidence. This is especially troubling because these trials often serve as the foundation for clinical guidelines and systematic reviews. This implies that, even when they come from several randomized trials, treatment recommendations in gynaecology may frequently be informed by evidence with potential methodological limitations.\u003c/p\u003e\n\u003cp\u003eThe biggest predictor of low risk of bias was trial registration, highlighting the significance of prospective registration for methodological transparency. Clear temporal improvements in reporting quality were shown by regression analyses; however, these changes did not reflect consistent methodological improvement over time, but were instead largely mediated by structural factors, such as rising prospective trial registration rates. Although minimal data from lower-income settings limited the findings, differences by country income level were seen in the anticipated direction.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRegression-based insights into reporting quality and bias\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe descriptive synthesis\u0026apos;s indicated patterns are quantitatively confirmed by the regression analyses used in this review. Specifically, after adjusting for clinical condition, publication time, and World income level, prospective trial registration was found to be the most significant indicator of reporting quality and was independently linked to more thorough CONSORT adherence. This result highlights the crucial function of registration as a structural mechanism, rather than just a procedural or administrative necessity, for limiting analytical flexibility and preventing selective reporting (48,49).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThere were also noticeable improvements in reporting quality over time. While multivariable modeling revealed that rising trial registration rates and other structural changes over time contributed to this improvement, univariable time-trend analysis clearly showed an increase in CONSORT adherence throughout subsequent publishing periods. This attenuation suggests that increases in reporting are mediated by the progressive institutionalization of transparency rules rather than being only the result of temporal growth (50,51).\u003c/p\u003e\n\u003cp\u003eAfter normalization, there were still differences in reporting quality by country income level, with trials carried out in non-high-income environments generally showing lower reporting completeness. Even though these variations were slight, they probably don\u0026apos;t represent inherent variations in scientific rigor but rather institutional differences in research infrastructure, regulatory control, and availability to methodological support. Crucially, our results warn against assuming that gains in reporting quality are evenly spread throughout international research settings (52).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMethodological quality varying by condition\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe continuously better methodological profile of GDM trials compared to those in endometriosis, PCOS, and PMDD is one of the review\u0026apos;s most notable conclusions. It seems doubtful that this discrepancy is accidental; rather, it represents contextual and structural variations among circumstances.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cbr\u003eBecause objectives including glycaemic control, birth weight, and neonatal morbidity are regularly assessed, scientifically defined, and clinically required, GDM trials benefit from being integrated into well-established antenatal care pathways. Larger, population-based trial designs may be made possible by these characteristics, which also naturally facilitate standardized outcome measurement and lower loss to follow-up (53,54). On the other hand, diseases like PMDD and endometriosis are characterized by chronicity, varying symptom profiles, delayed diagnosis, and dependence on subjective symptom reporting. These characteristics make trial design more difficult and more prone to bias, especially when blinding is insufficient or impractical (55\u0026ndash;57).\u003c/p\u003e\n\u003cp\u003ePCOS holds an intermediate position. The condition\u0026apos;s heterogeneity has led to an excessive dependence on surrogate and laboratory endpoints, which are frequently evaluated in isolation from patient-centered outcomes, despite the fact that it permits objective biochemical measurement. Despite the measures\u0026apos; seeming neutrality, this has led to selective reporting and outcome multiplicity, hurting interpretability (36,58).\u003c/p\u003e\n\u003cp\u003eThe results show that structural features of trial design, conduct, and reporting, rather than specific technical flaws, are the main causes of the increased risk of bias and reduced transparency in gynecological randomized controlled trials. Specifically, selective outcome reporting, analytical flexibility, and poor treatment of missing data are consistently made possible by the small trial size and absence of prospective registration\u0026nbsp;(49,59). These flaws are exacerbated by outcome selection: Research that primarily uses participants\u0026apos; self-reported symptoms is particularly vulnerable to expectation-driven bias when blinding is not done completely\u0026nbsp;(60,61). At the same time, studies that measure numerous biochemical surrogate markers create ample opportunity for selective reporting, often without providing evidence of meaningful benefits for patients\u0026nbsp;(62). Intervention modality also exerts an important structural influence, as non-pharmacological trials face intrinsic barriers to participant blinding and standardisation of treatment delivery\u0026nbsp;(63,64). Transparency is further undermined by incomplete funding disclosure and variable adherence to reporting standards, particularly in lower-resource research settings\u0026nbsp;(65). The comparatively stronger methodological profile observed in gestational diabetes trials illustrates that higher standards of transparency and outcome consistency are achievable when trials are embedded within regulated clinical pathways and employ objective, standardised endpoints. Collectively, these findings indicate that meaningful improvement in the credibility of gynaecological evidence will require systemic reforms, mandatory trial registration, standardised outcome frameworks, robust analytical protocols, and stronger enforcement of reporting requirements, rather than incremental adjustments to individual trial practices alone.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRobustness of risk-of-bias findings\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSensitivity analysis using an alternative dichotomisation of overall risk of bias, comparing trials classified as high risk with all other trials, demonstrated that the direction of associations between trial characteristics and risk of bias was broadly consistent with the primary analysis. Although effect estimates were reduced and did not consistently reach statistical significance, no associations reversed direction.\u003c/p\u003e\n\u003cp\u003eThis pattern implies that the main conclusions represent underlying structural factors of trial quality rather than being products of a specific risk-of-bias threshold. The sensitivity analysis\u0026apos;s attenuation is in line with the high-risk classification\u0026apos;s stricter criteria as well as the variety of ways bias might manifest itself. Together, our results highlight the limits imposed by few low-risk trials and the intricacy of bias processes in gynecological research, while also supporting the overall conclusions\u0026apos; robustness\u0026nbsp;(66).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSubjective outcomes, blinding, and the limits of traditional RCT paradigms\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe high levels of performance and detection bias in trials of endometriosis and PMDD raise important questions about whether standard randomised controlled trial designs are well-suited to gynaecological conditions that rely heavily on subjective symptom reporting. Although RCTs remain the dominant standard for causal inference in clinical research, their core methodological assumptions are poorly suited to conditions in which pain, mood disturbance, fatigue, and quality of life are the primary therapeutic targets. In such contexts, achieving effective blinding and stable outcome measurement can be challenging. These outcomes are inherently vulnerable to expectancy effects, reporting biases, and contextual influences that cannot be fully addressed through randomisation alone at the outcome assessment stage\u0026nbsp;(63,67). In surgical, behavioural, and lifestyle interventions, where participant blinding is often impossible, these vulnerabilities are further intensified, blurring the conventional distinction between intervention effects and placebo responses\u0026nbsp;(63,68). Treating these issues as mere implementation failures obscures a more fundamental structural mismatch between methodological ideals and clinical reality.\u003c/p\u003e\n\u003cp\u003eRather than continuing to evaluate such trials against an idealised pharmacological model of blinding, a more defensible methodological position is to acknowledge these constraints explicitly and recalibrate standards of internal validity accordingly. Enhanced emphasis on outcome assessor blinding, rigorous validation of patient-reported outcome instruments, pre-specification of minimal clinically important differences, and transparent handling of missing data may provide more meaningful safeguards against bias than formalistic adherence to blinding criteria that are, in many contexts, unattainable\u0026nbsp;(69,70). Without such recalibration, continued reliance on orthodox RCT frameworks risks generating systematically biased estimates that may be influenced by bias despite adherence to conventional methodological frameworks.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTrial registration and transparency: progress with persistent gaps\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn addition to being descriptive indicators of methodological quality, trial registration and transparency were found to be statistically significant predictors of reporting completeness. In multivariable analysis, there is a high correlation between prospective registration and CONSORT adherence, indicating that registration serves as a significant constraint on selective reporting and post-hoc outcome selection. This supports the body of research showing that transparency procedures work best when incorporated into trial design early on, rather than being added after the fact at the reporting stage\u0026nbsp;(71).\u003c/p\u003e\n\u003cp\u003eProspective trial registration demonstrates that transparent research practices are institutionally achievable when supported by regulatory frameworks\u0026nbsp;(72), with our findings showing substantially higher prospective registration among GDM trials. However, the continued prevalence of unregistered trials in endometriosis and PCOS indicates that adoption of these norms remains uneven and condition-dependent. Given the well-established association between non-registration and selective outcome reporting, this disparity carries direct implications for the credibility of published treatment effects\u0026nbsp;(73). The evidentiary problem is not limited to the absence of registration. Even among registered trials, discrepancies between prespecified and reported outcomes are common, as demonstrated by multiple audits comparing trial registries and protocols with published reports, revealing persistent weaknesses in enforcement mechanisms and editorial oversight\u0026nbsp;(74\u0026ndash;76). These findings suggest that registration currently functions more as a procedural formality and not actually to avoid reporting bias. Strengthening transparency, therefore, requires institutional reform beyond voluntary compliance, including systematic verification of registry entries during peer review, harmonisation of journal reporting requirements, and proportionate sanctions for non-adherence. The generally accepted goal of trial registration runs the risk of becoming symbolic compliance rather than true methodological responsibility in the absence of such measures.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFunding, industry involvement, and subtle influences on trial design\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe significant minority of trials reporting industry or mixed funding arrangements warrants ongoing critical attention, even though explicit industry dominance was not visible throughout the dataset. The choice of comparators, the framing of primary outcomes, and the criteria for clinical significance are just a few examples of how financing sources may affect trial architecture in pharmacologically directed investigations in ways that are both methodologically acceptable and epistemically significant. Rarely do these factors show up as explicit wrongdoing; instead, they work through the cumulative impacts of design optimization toward market positioning or regulatory acceptance\u0026nbsp;(77). Another problem is the documented non-reporting of funding sources in a subset of trials within our dataset, which prevents informed assessment of potential conflicts of interest, because of which the interpretive transparency necessary for robust evidence appraisal is undermined. In clinical domains characterised by long-term therapeutic exposure, incomplete disclosure of financial relationships erodes confidence in individual findings and in the normal integrity of the research enterprise itself, as financial ties have been shown to influence research conduct and reporting (78). From a methodological perspective, funding transparency should be regarded not as an administrative accessory but as an important component of trial validity, enabling downstream users of evidence to situate reported effects within their institutional and economic context (79,80).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEquity, geography, and the production of partial knowledge\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe high number of trials from our dataset in high-income and upper-middle-income countries highlights a persistent inequity in evidence generation. This imbalance has important consequences; for example, trial findings may not be transferable to settings with different healthcare infrastructure, cultural contexts, or disease prevalence (81). The limited number of trials from lower-middle-income countries suggests that the global burden of gynaecological disease is being addressed with geographically narrow evidence. This pattern reflects broader structural inequities in global health research funding and infrastructure (82). Without deliberate strategies to support inclusive trial design and multinational collaboration, these disparities are likely to persist, perpetuating a cycle in which evidence is generated primarily for populations already best served by healthcare systems.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOutcome selection and the role of core outcome sets\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe extensive outcome heterogeneity identified across all conditions reinforces longstanding concerns about research waste in gynaecology. While core outcome initiatives have sought to address this issue, uptake remains inconsistent (83). Future progress will require not only broader adoption of core outcome sets but also cultural shifts in how outcomes are valued. Integrating patient voices into outcome prioritisation and aligning funding incentives with methodological quality rather than novelty may help address these challenges.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImplications for clinical practice and guideline development\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe findings of this review suggest that clinicians and guideline developers should exercise cautious interpretation of gynaecological trial evidence, particularly when recommendations are based on small, unblinded trials with subjective outcomes. While RCTs remain the gold standard, their methodological limitations must be explicitly acknowledged when translating evidence into practice. For policymakers, the review highlights the need to invest not only in more trials but in better-designed trials, with emphasis on transparency, inclusivity, and patient relevance. Without such investment, the evidence base risks expanding in size without corresponding gains in reliability or impact.\u003c/p\u003e\n\u003cp\u003eKey findings showcase that methodological weaknesses in gynaecological randomised controlled trials are systemic, patterned, and consequential. Differences in trial quality across conditions reflect underlying structural, clinical, and epistemic factors rather than isolated shortcomings. Addressing these challenges will require coordinated efforts from researchers, funders, journals, and clinicians to prioritise methodological rigour, transparency, and equity in women\u0026rsquo;s health research. Without improvements in methodological rigor, the expanding volume of gynaecological research risks generating an evidence base with important methodological limitations.\u003c/p\u003e\n\u003cp\u003eThe findings of this review have important implications for research governance and policy. Regulatory agencies, research funders, and journal editors play critical roles in strengthening methodological standards. Mandatory prospective trial registration, enforcement of CONSORT reporting guidelines, and greater investment in multicentre and multinational trials are essential steps toward improving the credibility and global applicability of women\u0026rsquo;s health research.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eStrengths\u003c/p\u003e\n\u003cp\u003eThis review analysed a large dataset of 695 randomized controlled trials across four major gynaecological conditions and applied contemporary methodological frameworks including the Cochrane Risk of Bias 2 tool and CONSORT reporting standards.\u003c/p\u003e\n\u003cp\u003eLimitations\u003c/p\u003e\n\u003cp\u003eSeveral limitations should be acknowledged. First, the review included only English-language publications, which may introduce language bias. Second, reliance on published reports may underestimate methodological rigor if trial methods were incompletely reported. Third, the underrepresentation of trials conducted in low-income settings limits conclusions regarding global equity in evidence generation.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis review demonstrates that methodological limitations in gynaecological randomized controlled trials are widespread and structurally patterned rather than isolated occurrences. Although reporting practices have improved over time, robust low-risk-of-bias evidence remains uncommon. Prospective registration and transparency mechanisms are strongly associated with improved methodological quality, suggesting that meaningful progress is achievable through systemic reinforcement of these practices. Strengthening the reliability, transparency, and inclusivity of gynaecological research will be essential to ensuring that clinical and policy decisions in women\u0026rsquo;s health are supported by evidence that is not only abundant but methodologically sound. Strengthening the evidence base for women\u0026rsquo;s health will require systemic improvements in trial design, transparency mechanisms, and global research equity.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRCT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRandomized Controlled Trial\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePCOS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePolycystic Ovary Syndrome\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGDM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGestational Diabetes Mellitus\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePMDD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePremenstrual Dysphoric Disorder\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOBGYN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eObstetrics and Gynaecology\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCONSORT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eConsolidated Standards of Reporting Trials\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePRISMA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePreferred Reporting Items for Systematic Reviews and Meta-Analyses\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRoB 2\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRisk of Bias 2\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCROWN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCore Outcomes in Women\u0026rsquo;s Health\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eWHI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eWomen's Health Initiative\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eOdds Ratio\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eConfidence interval\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors consented to publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data generated or analysed during this study are included in this published article and its supplementary information files.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no external funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNR, NW, and SA contributed equally to this work and share first authorship. NR, NW, and SA were responsible for study design, study screening, data extraction, and manuscript drafting. SA acted as the third reviewer to resolve disagreements during study selection and data extraction. VP supervised the study and contributed to data interpretation and critical revision of the manuscript. GD contributed to the study's conceptualisation and methodological oversight. \u0026nbsp; All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMcKinney JL, Clinton SC, Keyser LE. Women\u0026rsquo;s health across the lifespan: A sex- and gender-focused perspective. Phys Ther. 2024;104(10). doi:10.1093/ptj/pzae121\u003c/li\u003e\n\u003cli\u003eGeller A, Salganicoff A, Burke SP. Nih.gov [Internet]. National Academies Press (US); 2025. Overview of Selected Women\u0026rsquo;s Health Conditions. Available from: https://www.ncbi.nlm.nih.gov/books/NBK612403\u003c/li\u003e\n\u003cli\u003eVinter C, Linde D, Petersen L. Research in women - Women in research. Dan Med J. 2026;73(2):10250782. doi:10.61409/A10250782\u003c/li\u003e\n\u003cli\u003eChristopoulos P, Tsarna E. Gynecologic and Obstetric Pathologies from Birth to Menopause: Unveiling the Journey and Charting the Future. J Clin Med. 2024;13(20):6230. doi:10.3390/jcm13206230\u003c/li\u003e\n\u003cli\u003eLa Marca-Ghaemmaghami P, Ehlert U. Gynecological Health: Psychosocial Aspects. Int Encycl Soc Behav Sci Second Ed Elsevier. 2015;462\u0026ndash;8. doi:10.1016/B978-0-08-097086-8.14098-X\u003c/li\u003e\n\u003cli\u003eHoffman SR, Farland LV, Doll KM, Nicholson WK, Wright MA, Robinson WR. The epidemiology of gynaecologic health: contemporary opportunities and challenges. J Epidemiol Community Health. 2020;jech-2019. doi:10.1136/jech-2019-213149\u003c/li\u003e\n\u003cli\u003eKirk UB, Bank-Mikkelsen AS, Rytter D, Hartwell D, Marschall H, Nyegaard M, et al. Understanding endometriosis underfunding and its detrimental impact on awareness and research. Npj Women Health. 2024;2(1). doi:10.1038/s44294-024-00048-6\u003c/li\u003e\n\u003cli\u003eAl Wattar BH, Fisher M, Bevington L, Talaulikar V, Davies M, Conway G, et al. Clinical Practice Guidelines on the Diagnosis and Management of Polycystic Ovary Syndrome: A Systematic Review and Quality Assessment Study. J Clin Endocrinol Metab. 2021;106(8). doi:10.1210/clinem/dgab232\u003c/li\u003e\n\u003cli\u003eWexler DJ, Powe CE, Barbour LA, Buchanan T, Coustan DR, Corcoy R, et al. Research Gaps in Gestational Diabetes Mellitus. Obstet Gynecol. 2018;132(2):496\u0026ndash;505. doi:10.1097/aog.0000000000002726\u003c/li\u003e\n\u003cli\u003eCarlini SV, Deligiannidis KM. Evidence-Based Treatment of Premenstrual Dysphoric Disorder. J Clin Psychiatry. 2020;81(2). doi:10.4088/jcp.19ac13071\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. World Health Organization [Internet]. 2023. Endometriosis. Available from: https://www.who.int/news-room/fact-sheets/detail/endometriosis\u003c/li\u003e\n\u003cli\u003eSatapathy P, Vadia N, Menon SV, Chennakesavulu K, Panigrahi R, Bushi G, et al. Prevalence of polycystic ovarian syndrome in South Asian countries and 21 global regions with Sociodemographic index: A systematic analysis of the Global Burden of Disease (1990\u0026ndash;2021). J Gynecol Obstet Hum Reprod. 2025;54(9):103019. doi:10.1016/j.jogoh.2025.103019\u003c/li\u003e\n\u003cli\u003eChoudhury AA, Rajeswari VD. Gestational diabetes mellitus - A metabolic and reproductive disorder. Biomed Pharmacother. 2021;143(112183):112183. doi:10.1016/j.biopha.2021.112183\u003c/li\u003e\n\u003cli\u003ePataki B, Bianka Luca Kiss, S\u0026aacute;ra K\u0026aacute;lm\u0026aacute;n, Ildik\u0026oacute; Kov\u0026aacute;cs. Premenstrual dysphoric disorder\u0026mdash;an undervalued diagnosis? A cross-sectional study in Hungarian women. Compr Psychoneuroendocrinology. 2024;20:100256\u0026ndash;100256. doi:10.1016/j.cpnec.2024.100256\u003c/li\u003e\n\u003cli\u003eEurohealth.ie [Internet]. 2024. 2024 \u0026ndash; Women\u0026rsquo;s Health Manifesto \u0026ndash; Eurohealth. Available from: https://eurohealth.ie/2024-womens-health-manifesto/\u003c/li\u003e\n\u003cli\u003eTharmalingam H. Macmillan Cancer Support [Internet]. 2024. A greater society for all - why women\u0026rsquo;s health matters. Available from: https://www.macmillan.org.uk/healthcare-professionals/news-and-resources/blogs/iwd-why-womens-health-matters\u003c/li\u003e\n\u003cli\u003eEkinci G. The impact of maternal and child health on sustainable development goals: evidence from Turkic Republics (2000\u0026ndash;2020). BMC Public Health. 2025;25(1). doi:10.1186/s12889-025-25172-z\u003c/li\u003e\n\u003cli\u003eWijeratne D, Gibson JFE, Fiander A, Rafii-Tabar E, Thakar R. The global burden of disease due to benign gynecological conditions: A call to action. Int J Gynaecol Obstet Off Organ Int Fed Gynaecol Obstet. 2023;164(3):1151\u0026ndash;9. doi:10.1002/ijgo.15211\u003c/li\u003e\n\u003cli\u003eDavidson PM, McGrath SJ, Meleis AI, Stern P, DiGiacomo M, Dharmendra T, et al. The Health of Women and Girls Determines the Health and Well-Being of Our Modern World: A White Paper From the International Council on Women\u0026rsquo;s Health Issues. Health Care Women Int. 2011;32(10):870\u0026ndash;86. doi:10.1080/07399332.2011.603872\u003c/li\u003e\n\u003cli\u003eKleinman DV, Pronk N, G\u0026oacute;mez CA, Gordon GLW, Ochiai E, Blakey C, et al. Addressing health equity and social determinants of health through healthy people 2030. J Public Health Manag Pract. 2021;27(6):249\u0026ndash;57. doi:10.1097/PHH.0000000000001297\u003c/li\u003e\n\u003cli\u003eLefton M. swhr.org [Internet]. 2022. The Evolution of Women\u0026rsquo;s Health Research - SWHR. Available from: https://swhr.org/the-evolution-of-womens-health-research/\u003c/li\u003e\n\u003cli\u003ePinn VW. Women\u0026rsquo;s Health Research: Current State of the Art. Glob Adv Health Med. 2013;2(5):8\u0026ndash;10. doi:10.7453/gahmj.2013.063\u003c/li\u003e\n\u003cli\u003ePrice C. Nih.gov [Internet]. 2014. Women\u0026rsquo;s Health Initiative (WHI) | National Heart, Lung, and Blood Institute (NHLBI). Available from: https://www.nhlbi.nih.gov/science/womens-health-initiative-whi\u003c/li\u003e\n\u003cli\u003eResearch I of M (US) C on WH. www.ncbi.nlm.nih.gov [Internet]. National Academies Press (US); 2010. Methodologic Issues in Women\u0026rsquo;s Health Research. Available from: https://www.ncbi.nlm.nih.gov/books/NBK210137/\u003c/li\u003e\n\u003cli\u003eSteinberg JR, Turner BE, DiTosto JD, Weeks BT, Young AMP, Lu CF, et al. Race and Ethnicity Reporting and Representation in Obstetrics and Gynecology Clinical Trials and Publications From 2007-2020. JAMA Surg. 2023;158(2):181\u0026ndash;90. doi:10.1001/jamasurg.2022.6600\u003c/li\u003e\n\u003cli\u003eSteinberg JR, Weeks BT, Reyes GA, Fitzgerald AC, Lyell DJ. The obstetric research landscape: a cross-sectional analysis of clinical trials from 2007-2020. Am J Obstet Gynecol MFM. 2020;3(1):100253. doi:10.1016/j.ajogmf.2020.100253\u003c/li\u003e\n\u003cli\u003eLee SS, Dinicu AI, Arthurs L, Shields D, Pothuri B, Lightfoot MDS. Demographic reporting and language exclusion in gynecologic oncology clinical trials. Am J Obstet Gynecol. 2023;230(1):73.e1-73.e14. doi:10.1016/j.ajog.2023.09.019\u003c/li\u003e\n\u003cli\u003eBierer BE, Meloney LG, Ahmed HR, White SA. Advancing the inclusion of underrepresented women in clinical research. Cell Rep Med. 2022;3(4):100553. doi:10.1016/j.xcrm.2022.100553\u003c/li\u003e\n\u003cli\u003eCollister D, Song C, Ruzycki SM. Fostering diversity in clinical trials: need for evidence and implementation to improve representation. BMJ Med. 2024;3(1):e000984\u0026ndash;e000984. doi:10.1136/bmjmed-2024-000984\u003c/li\u003e\n\u003cli\u003eGayathri De Lanerolle, Phiri P, Haroon A. Clinical Trials and Tribulations. Elsevier; 2023.\u003c/li\u003e\n\u003cli\u003eBalch B. Association of American Medical Colleges [Internet]. 2024. Why we know so little about women\u0026rsquo;s health. Available from: https://www.aamc.org/news/why-we-know-so-little-about-women-s-health\u003c/li\u003e\n\u003cli\u003eWaitt C, Astill D, Zavala E, Karron RA, Faden RR, Stratton P, et al. Clinical trials and pregnancy. Commun Med. 2022;2(1):1\u0026ndash;5. doi:10.1038/s43856-022-00198-1\u003c/li\u003e\n\u003cli\u003esanogenetics.com [Internet]. Pioneering progress: The history of women in clinical trials. Available from: https://sanogenetics.com/resources/blog/the-history-of-women-in-clinical-trials\u003c/li\u003e\n\u003cli\u003eSteinberg JR, Turner BE, Weeks BT, Magnani CJ, Wong BO, Rodriguez F, et al. Analysis of Female Enrollment and Participant Sex by Burden of Disease in US Clinical Trials Between 2000 and 2020. JAMA Netw Open. 2021;4(6):e2113749. doi:10.1001/jamanetworkopen.2021.13749\u003c/li\u003e\n\u003cli\u003evan \u0026rsquo;t Hooft J, van Dijk CE, Axfors C, Alfirevic Z, Oudijk MA, Khan KS, et al. Assessing the usefulness of randomised trials in obstetrics and gynaecology. BJOG Int J Obstet Gynaecol. 2023;130(7):695\u0026ndash;701. doi:10.1111/1471-0528.17411\u003c/li\u003e\n\u003cli\u003eAl Wattar BH, Teede H, Garad R, Franks S, Balen A, Bhide P, et al. Harmonising research outcomes for polycystic ovary syndrome: an international multi-stakeholder core outcome set. Hum Reprod. 2020;35(2):404\u0026ndash;12. doi:10.1093/humrep/dez272\u003c/li\u003e\n\u003cli\u003eTellum T, Omtvedt M, Naftalin J, Hirsch M, Jurkovic D. A systematic review of outcome reporting and outcome measures in studies investigating uterine-sparing treatment for adenomyosis. Hum Reprod Open. 2021;2021(3). doi:10.1093/hropen/hoab030\u003c/li\u003e\n\u003cli\u003eThe CROWN Initiative: journal editors invite researchers to develop core outcomes in women\u0026rsquo;s health. J Gynecol Oncol. 2014;25(3):166. doi:10.3802/jgo.2014.25.3.166\u003c/li\u003e\n\u003cli\u003eMacedo C, Riera R, Torloni M. Methodological quality of systematic reviews and clinical trials on women\u0026rsquo;s health published in a Brazilian evidence-based health journal. Clinics. 2013;68(4):563\u0026ndash;7. doi:10.6061/clinics/2013(04)20\u003c/li\u003e\n\u003cli\u003eSavović J, Turner RM, Mawdsley D, Jones HE, Beynon R, Higgins JPT, et al. Association Between Risk-of-Bias Assessments and Results of Randomized Trials in Cochrane Reviews: The ROBES Meta-Epidemiologic Study. Am J Epidemiol. 2017;187(5):1113\u0026ndash;22. doi:10.1093/aje/kwx344\u003c/li\u003e\n\u003cli\u003eShrestha BM. Systematic Reviews and Meta-analysis: Principles and Practice. J Nepal Med Assoc. 2019;57(215). doi:10.31729/jnma.3986\u003c/li\u003e\n\u003cli\u003ePage MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Br Med J. 2021;372(71). doi:10.1136/bmj.n71\u003c/li\u003e\n\u003cli\u003eRaghuraman N, Delanerolle G. York.ac.uk [Internet]. 2025 [cited 2025 Jan 1]. PROSPERO. Available from: https://www.crd.york.ac.uk/PROSPERO/view/CRD42025636543\u003c/li\u003e\n\u003cli\u003eGarg R, Maurya I. Clinical trial registration: An essential step toward transparency in clinical research. Indian J Anaesth. 2023;67(4):321. doi:10.4103/ija.ija_234_23\u003c/li\u003e\n\u003cli\u003eCochrane Methods. methods.cochrane.org [Internet]. 2017. Risk of Bias 2 (RoB 2) tool. Available from: https://methods.cochrane.org/risk-bias-2\u003c/li\u003e\n\u003cli\u003eCONSORT 2010 checklist of information to include when reporting a randomised trial [Internet]. 2010. Report No. Available from: https://legacyfileshare.elsevier.com/promis_misc/CONSORT-2010-Checklist.pdf\u003c/li\u003e\n\u003cli\u003eDelanerolle G, Pathiraja V, Mudalige T, Mbwele B, Ikwuka D, Cavalini H, et al. A Perspective on Innovating Qualitative Research Methods to Optimise Women\u0026rsquo;s Health Research. Int J Qual Methods. 2025;24. doi:10.1177/16094069251406147\u003c/li\u003e\n\u003cli\u003eSimonsen E, Wang SV, Kildegaard H, Potteg\u0026aring;rd A. Preregistration: A Key to Credible Real‐World Evidence Generation. Pharmacoepidemiol Drug Saf. 2025;34(9). doi:10.1002/pds.70215\u003c/li\u003e\n\u003cli\u003eTaylor NJ, Gorman DM. Registration and primary outcome reporting in behavioral health trials. BMC Med Res Methodol. 2022;22(1). doi:10.1186/s12874-021-01500-w\u003c/li\u003e\n\u003cli\u003eFreese J, King MM. Institutionalizing Transparency. Socius Sociol Res Dyn World. 2018;4:237802311773921. doi:10.1177/2378023117739216\u003c/li\u003e\n\u003cli\u003eStefkovics \u0026Aacute;, Eichhorst A, Skinnion D, Harrison CH. Are We Becoming More Transparent? Survey Reporting Trends in Top Journals of Social Sciences. Int J Public Opin Res. 2024;36(2). doi:10.1093/ijpor/edae013\u003c/li\u003e\n\u003cli\u003eAlemayehu C, Mitchell G, Nikles J. Barriers for conducting clinical trials in developing countries- a systematic review. Int J Equity Health. 2018;17(1). doi:10.1186/s12939-018-0748-6\u003c/li\u003e\n\u003cli\u003eMunda A, Mlinaric Z, Jakin PA, Lunder M, Pongrac Barlovic D. Effectiveness of a comprehensive telemedicine intervention replacing standard care in gestational diabetes: a randomized controlled trial. Acta Diabetol. 2023;60:1037\u0026ndash;44. doi:10.1007/s00592-023-02099-8\u003c/li\u003e\n\u003cli\u003eLeblalta B, Kebaili H, Sim R, Lee SWH. Digital health interventions for gestational diabetes mellitus: A systematic review and meta-analysis of randomised controlled trials. Lichtner V, editor. PLOS Digit Health. 2022;1(2):e0000015. doi:10.1371/journal.pdig.0000015\u003c/li\u003e\n\u003cli\u003eSaunders PTK, Whitaker LHR, Horne AW. Endometriosis: Improvements and challenges in diagnosis and symptom management. Cell Rep Med. 2024;5(6):101596. doi:10.1016/j.xcrm.2024.101596\u003c/li\u003e\n\u003cli\u003eNih.gov [Internet]. Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU); 2018. Endometriosis \u0026ndash; Diagnosis, treatment and patient experiences. Available from: https://www.ncbi.nlm.nih.gov/books/NBK570473\u003c/li\u003e\n\u003cli\u003ePearlstein T, Steiner M. Premenstrual dysphoric disorder: burden of illness and treatment update. J Psychiatry Neurosci JPN. 2008;33(4):291.\u003c/li\u003e\n\u003cli\u003eLegro RS. Surrogate end-points or primary outcomes in clinical trials in women with polycystic ovary syndrome? Hum Reprod. 2004;19(8):1697\u0026ndash;704. doi:10.1093/humrep/deh322\u003c/li\u003e\n\u003cli\u003eZhang Z, Xu X, Ni H. Small studies may overestimate the effect sizes in critical care meta-analyses: a meta-epidemiological study. Crit Care. 2013;17(1):R2. doi:10.1186/cc11919\u003c/li\u003e\n\u003cli\u003ePitre T, Kirsh S, Jassal T, Anderson M, Padoan A, Xiang A, et al. The Impact of Blinding on Trial results: a Systematic Review and Meta‐analysis. Cochrane Evid Synth Methods. 2023;1(4). doi:10.1002/cesm.12015\u003c/li\u003e\n\u003cli\u003eHrobjartsson A, Thomsen ASS, Emanuelsson F, Tendal B, Hilden J, Boutron I, et al. Observer bias in randomized clinical trials with measurement scale outcomes: a systematic review of trials with both blinded and nonblinded assessors. Can Med Assoc J. 2013;185(4):E201\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003eManyara AM, Davies P, Stewart D, Weir CJ, Young AE, Blazeby J, et al. Reporting of surrogate endpoints in randomised controlled trial protocols (SPIRIT-Surrogate): extension checklist with explanation and elaboration. BMJ. 2024;386:e078525\u0026ndash;e078525. doi:10.1136/bmj-2023-078525\u003c/li\u003e\n\u003cli\u003eYonis A, Dean SG, Warren FC, Taylor RS, Levack W, Hay-Smith J. Navigating blinding challenges in complex intervention trials: insights from a UK researcher survey. Trials. 2025;26(1). doi:10.1186/s13063-025-09223-9\u003c/li\u003e\n\u003cli\u003eBoutron I, Tubach F, Giraudeau B, Ravaud P. Blinding was judged more difficult to achieve and maintain in nonpharmacologic than pharmacologic trials. J Clin Epidemiol. 2004;57(6):543\u0026ndash;50. doi:10.1016/j.jclinepi.2003.12.010\u003c/li\u003e\n\u003cli\u003eZhang W, DeVito NJ, Chan AW, Cunningham C, Pymento J, Karam G, et al. Enhancing global clinical trial transparency for better health outcomes for all. F1000Research. 2025;14:626. doi:10.12688/f1000research.166358.1\u003c/li\u003e\n\u003cli\u003eMaru\u0026scaron;ić MF, Fidahić M, Cepeha CM, Farcaș LG, Tseke A, Puljak L. Methodological tools and sensitivity analysis for assessing quality or risk of bias used in systematic reviews published in the high-impact anesthesiology journals. BMC Med Res Methodol. 2020;20(1). doi:10.1186/s12874-020-00966-4\u003c/li\u003e\n\u003cli\u003eMoustgaard H, Bello S, Miller FG, Hr\u0026oacute;bjartsson A. Subjective and objective outcomes in randomized clinical trials: definitions differed in methods publications and were often absent from trial reports. J Clin Epidemiol. 2014;67(12):1327\u0026ndash;34. doi:10.1016/j.jclinepi.2014.06.020\u003c/li\u003e\n\u003cli\u003eWartolowska K, Feakins BG, Collins GS, Cook J, Judge A, Rombach I, et al. The magnitude and temporal changes of response in the placebo arm of surgical randomized controlled trials: a systematic review and meta-analysis. Trials. 2016;17(1). doi:10.1186/s13063-016-1720-7\u003c/li\u003e\n\u003cli\u003eBoutron I, Altman DG, Moher D, Schulz KF, Ravaud P. CONSORT Statement for Randomized Trials of Nonpharmacologic Treatments: A 2017 Update and a CONSORT Extension for Nonpharmacologic Trial Abstracts. Ann Intern Med. 2017;167(1):40. doi:10.7326/m17-0046\u003c/li\u003e\n\u003cli\u003eRevicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J Clin Epidemiol. 2008;61(2):102\u0026ndash;9. doi:10.1016/j.jclinepi.2007.03.012\u003c/li\u003e\n\u003cli\u003eVitali FC, Santos PS, Rocha A de O, Maia LC, Garcia L da FR, Teixeira C da S. Adherence to Registration and Selective Outcome Reporting in Randomized Clinical Trials Published in Endodontic Journals Over the Past 5 Years: A Meta-Research Study. J Endod. 2025;51(3):258-267.e7. doi:10.1016/j.joen.2024.12.001\u003c/li\u003e\n\u003cli\u003eDe Angelis C, Drazen JM, Frizelle FA, Haug C, Hoey J, Horton R, et al. Clinical Trial Registration: A Statement from the International Committee of Medical Journal Editors. N Engl J Med. 2004;351(12):1250\u0026ndash;1. doi:10.1056/nejme048225\u003c/li\u003e\n\u003cli\u003eXu C, Fan S, Furuya-Kanamori L, Li S, Lin L, Chu H, et al. Harm effects in non-registered versus registered randomized controlled trials of medications: a retrospective cohort study of clinical trials. BMC Med. 2024;22(1). doi:10.1186/s12916-024-03621-7\u003c/li\u003e\n\u003cli\u003eChan AW, Hr\u0026oacute;bjartsson A, Haahr MT, G\u0026oslash;tzsche PC, Altman DG. Empirical Evidence for Selective Reporting of Outcomes in Randomized Trials. JAMA. 2004;291(20):2457. doi:10.1001/jama.291.20.2457\u003c/li\u003e\n\u003cli\u003eChan AW, Pello A, Kitchen J, Axentiev A, Virtanen JI, Liu A, et al. Association of Trial Registration With Reporting of Primary Outcomes in Protocols and Publications. JAMA. 2017;318(17):1709. doi:10.1001/jama.2017.13001\u003c/li\u003e\n\u003cli\u003eDechartres A, Ravaud P, Atal I, Riveros C, Boutron I. Association between trial registration and treatment effect estimates: a meta-epidemiological study. BMC Med. 2016;14(1). doi:10.1186/s12916-016-0639-x\u003c/li\u003e\n\u003cli\u003eGazendam AM, Slawaska-Eng D, Nucci N, Bhatt O, Ghert M. The Impact of Industry Funding on Randomized Controlled Trials of Biologic Therapies. Medicines. 2022;9(3):18. doi:10.3390/medicines9030018\u003c/li\u003e\n\u003cli\u003eBekelman JE, Li Y, Gross CP. Scope and Impact of Financial Conflicts of Interest in Biomedical Research. JAMA. 2003;289(4):454. doi:10.1001/jama.289.4.454\u003c/li\u003e\n\u003cli\u003eRecommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals [Internet]. 2025. Report No. Available from: https://www.icmje.org/icmje-recommendations.pdf\u003c/li\u003e\n\u003cli\u003eFarrell A. Improving transparency in clinical trial reporting. PLOS Med. 2025;22(4):e1004588. doi:10.1371/journal.pmed.1004588\u003c/li\u003e\n\u003cli\u003eRothwell PM. External validity of randomised controlled trials: \u0026ldquo;To whom do the results of this trial apply?\u0026rdquo; The Lancet. 2005;365(9453):82\u0026ndash;93. doi:10.1016/s0140-6736(04)17670-8\u003c/li\u003e\n\u003cli\u003eThiers FA, Sinskey AJ, Berndt ER. Trends in the globalization of clinical trials. Nat Rev Drug Discov. 2008;7(1):13\u0026ndash;4. doi:10.1038/nrd2441\u003c/li\u003e\n\u003cli\u003eDuffy J, Rolph R, Gale C, Hirsch M, Khan K, Ziebland S, et al. Core outcome sets in women\u0026rsquo;s and newborn health: a systematic review. BJOG Int J Obstet Gynaecol. 2017;124(10):1481\u0026ndash;9. doi:10.1111/1471-0528.14694\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Systematic review, Meta-research, Research methodology, Randomized controlled trials, Risk of bias, CONSORT, Endometriosis, Polycystic ovary syndrome, Gestational diabetes mellitus, Premenstrual dysphoric disorder","lastPublishedDoi":"10.21203/rs.3.rs-9212126/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9212126/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003cbr\u003e\nRandomized controlled trials (RCTs) provide the evidentiary foundation for clinical decision-making in obstetrics and gynaecology. However, concerns remain regarding methodological rigor, reporting transparency, and structural biases within women’s health research. Despite the global burden of gynaecological conditions such as endometriosis, polycystic ovary syndrome (PCOS), gestational diabetes mellitus (GDM), and premenstrual dysphoric disorder (PMDD), the methodological integrity of trials evaluating these conditions has not been comprehensively assessed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003cbr\u003e\nThis systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (PRISMA) and prospectively registered in PROSPERO (CRD42025636543). Electronic searches were performed in PubMed, ScienceDirect, and Google Scholar to identify RCTs published between January 2005 and October 2025 evaluating interventions for endometriosis, PCOS, GDM, or PMDD. Eligible studies included randomized or cluster-randomized trials involving adult women. Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool. Reporting completeness was evaluated using the Consolidated Standards of Reporting Trials (CONSORT) 2010 checklist. Associations between trial-level characteristics and methodological quality were examined using logistic and linear regression analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003cbr\u003e\nA total of 695 RCTs were included in the review. Overall methodological quality was limited, with only 49 trials (7.0%) judged to be at low risk of bias. The most frequent concerns were related to outcome measurement and selective reporting. Sample sizes varied widely across conditions, with smaller trials predominating in PMDD and endometriosis research, while GDM trials tended to be larger. Prospectively registered trials demonstrated significantly higher reporting completeness, with an average increase of 8.6 percentage points in CONSORT adherence (p \u0026lt; 0.001), and were more likely to be classified as low risk of bias (OR 4.08, 95% CI 1.59–10.47). Reporting quality improved modestly over time but remained uneven across conditions and country income levels.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e\u003cbr\u003e\nTrials with methodological limitations and variable reporting quality dominate the clinical evidence base for major gynaecological conditions. Prospective trial registration and transparency mechanisms were strongly associated with improved methodological standards. Strengthening trial design, enforcing reporting guidelines, and improving global equity in research infrastructure are essential to ensure that evidence guiding women’s health care is robust, transparent, and reliable.\u003c/p\u003e","manuscriptTitle":"Methodological Integrity of Randomized Controlled Trials in Major Gynaecological Conditions: A Systematic Review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-07 15:47:39","doi":"10.21203/rs.3.rs-9212126/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ce937bfe-4cc1-4775-b4c2-dc29770bdc0f","owner":[],"postedDate":"April 7th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Withdrawn","date":"2026-05-06T15:39:06+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-06T15:54:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-07 15:47:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9212126","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9212126","identity":"rs-9212126","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}