Global inequities in the adoption of innovative clinical trial designs: a cross-sectional analysis of ClinicalTrials.gov.

Clinical research is currently undergoing a transformative shift as it embraces innovative trial designs. 1 This progressive trend is fuelled by the growing recognition of the benefits these designs offer in terms of ethical considerations, cost-effectiveness and a focus on patient needs. 2 Innovative clinical trial designs include methodological frameworks such as adaptive trials, 3 which allow prospectively planned modifications to aspects of the study based on interim data; Bayesian designs, 4 which formally incorporate prior information and update evidence as data accumulate; and master protocol approaches, including platform, basket and umbrella trials, 5 which enable the simultaneous evaluation of multiple interventions, diseases or patient subgroups within a single overarching protocol. These modifications are pre-specified and informed by observations gleaned during the trial, allowing for a more efficient and potentially more ethical approach to clinical research. 6 Innovative trials aim to expedite the identification of effective therapies, improve the safety profile for participants and tailor the study design to emergent data, thus enhancing the overall utility of the trial. 7 Their ability to adapt to emerging data mid-trial paves the way for potential cost savings, improved operational efficiency and, most importantly, a heightened emphasis on patient well-being. 8 Within this general framework, Bayesian trial designs are a subset of innovative clinical trials that employ Bayesian statistical methods to guide decision-making throughout the trial. 4 This method combines prior knowledge and data accumulated during the trial to continuously update the probability of treatment effectiveness. Unlike traditional frequentist statistics, Bayesian methods integrate prior evidence with current data to assess treatment efficacy and safety in a more nuanced manner. Bayesian designs are especially beneficial when dealing with small patient populations and rare diseases 9 or when ethical concerns call for minimising patient exposure to potentially ineffective or harmful treatments. 10 By adapting the trial design in response to the gathered evidence, Bayesian trials can more quickly identify beneficial treatments, discontinue ineffective ones and adjust dosing to optimal levels. 11 Indeed, economic considerations also play an important role, as the traditional ‘one-size-fits-all’ approach is often resource-intensive and may lead to inconclusive or suboptimal results. 12 Innovative designs promise more efficient use of resources by adapting to emerging data, potentially reducing the time and cost of bringing new therapies to market. 1 Regulatory agencies worldwide emphasise the importance of adaptable and responsive strategies in creating novel medical treatments. Adopting adaptive trial designs and Bayesian methods can improve clinical trial efficiency and patient focus while maintaining strict regulatory standards. 13 For example, the Food and Drug Administration (FDA) guides sponsors and applicants on the use of adaptive designs for clinical trials, highlighting the principles for designing, conducting and reporting results from such trials. These guidelines emphasise the importance of proper planning and transparency in implementing adaptive designs as well as the simulation and statistical analysis of these trials. 14 The FDA emphasises the need for these trials to be well planned, with adaptations based on pre-specified interim analyses to preserve the integrity of the trial. This approach allows for potential cost savings and ethical advantages by minimising patient exposure to ineffective treatments and adjusting the trial in real-time based on emerging data. 15 16 This guidance also advises on the types of information to submit for FDA evaluation, underscoring the importance of clear communication and robust decision-making processes. 14 The European Medicines Agency (EMA) has similarly shown support for adaptive design in clinical trials, recognising the advantages they offer in terms of ethical conduct and resource efficiency. 17 Despite their recognised advantages, the global adoption of innovative clinical trial designs remains uneven, with substantial variation observed across socioeconomic and gender-related contexts. 18 A major obstacle is operational complexity, requiring coordination of multidisciplinary experts, stakeholders and logistics, particularly in platform trials, a class of adaptive trial designs that allow the simultaneous evaluation of multiple interventions within a single disease area using a shared control group and prespecified adaptations. 19 Moreover, innovative trial designs are inherently dynamic and can evolve in ways that are difficult to predict at the outset, complicating the estimation of costs and securing funding. Therefore, funders may be hesitant to support such trials because of the unpredictability of costs and the potential for additional unforeseen expenses. 19 Although regulatory agencies such as the FDA and EMA have issued guidance supporting adaptive and innovative trial designs, challenges remain in their practical implementation within existing regulatory, ethics review and governance processes. 1 Furthermore, there are practical aspects related to trial implementation, such as securing ethics approval, obtaining and maintaining funding and communicating complex trial designs to stakeholders, including trial participants. 20 Finally, statistical complexities can arise during implementation, such as ensuring the control of type I error rates, addressing potential biases and interpreting trial data correctly. Confirmatory adaptive designs are particularly challenging because they require extensive planning and validation to demonstrate their lack of bias and ensure their advantageous operating characteristics. 20 While innovative clinical trial designs offer significant benefits and are increasingly advocated for their adaptability and patient-centric focus, the challenges associated with their implementation often limit their application to well-resourced and highly organised experimental settings. 19 From a global perspective, the literature consistently shows that developing countries, women and minority populations remain underrepresented in experimental clinical research. 21 22 In deprived contexts, barriers to conducting clinical trials include a lack of financial and human resources, ethical and regulatory obstacles, inadequate research environments, operational challenges and competing demands, which particularly impact innovative experimental research. 22 The documented underrepresentation of some categories and geographical locations in clinical research could be more pronounced for innovative experimental designs, given their peculiarities and issues; this aspect remains under-documented in the literature. Factors contributing to the underrepresentation of certain countries and regions in clinical trials include a lack of infrastructure, financial constraints, political instability, regulatory challenges and limited access to innovative medical technologies. 22 23 Additionally, the limited engagement of low- and middle-income countries in clinical research is compounded by ethical concerns, such as the lack of local ethical review processes and inadequate patient consent mechanisms. 24 As a result, these regions remain underrepresented in global clinical trials, especially those involving innovative experimental designs that demand high resource investment and logistical coordination. 25 This study aims to examine the global adoption of innovative clinical trial designs and to quantify their association with country-level human development and gender 1 inequality indicators, thereby providing empirical evidence on how structural and geographical contexts are related to the use of innovative methodologies in clinical research. These disparities were quantified using United Nations Human Development Indicators. In this framework, the ClinicalTrials.gov database provides a comprehensive snapshot of innovative trials worldwide, as the International Committee of Medical Journal Editors has mandated study registration on this platform since September 2005 as a condition for publishing trial results. 26 This rich dataset offers a global perspective to evaluate how innovative trial designs are implemented across varying socioeconomic landscapes, revealing possible insights into the equity and reach of modern clinical research methodologies.

A Comma-Separated Value export has been executed for all 499 740 studies registered on ClinicalTrials.gov on 27 June 2024. ClinicalTrials.gov does not include a structured variable explicitly indicating whether a study uses an innovative or adaptive design. Therefore, in addition to the standard downloadable registry dataset, we retrieved complementary information from the ClinicalTrials.gov Application Programming Interface, which provides access to extended free-text fields, including the Brief Summary and Detailed Description. Innovative trial designs were identified by applying a predefined keyword-based search strategy to these text fields; trials containing at least one keyword related to adaptive, Bayesian, platform or other innovative designs were classified as innovative. We merged the two databases, joining the rows on the National Clinical Trial (NCT) number, the field that uniquely identifies each clinical trial. A total of 383 235 interventional studies were identified. Adaptive trials expanded in the scientific literature since the 1990s 7 ; for this reason, and given the limited number of trials published on ClinicalTrials.gov before, 380 015 records published after 1990 are considered for the analysis ( figure 1 ). Our analysis performed a keyword search in the study description fields to identify innovative designs in the ClinicalTrials.gov database. A search string of 42 keywords identified in the study description field of ClinicalTrials.gov was considered. Keywords were derived from the literature. 7 27 Other details concerning the search strategy are reported in the online supplemental material . The global distribution of innovative design studies has been categorised into five continents based on the locations specified in the database’s ‘Locations’ column. This categorisation aims to examine the spread of these studies and their connection to achievements in human development across the countries conducting them and between genders. Information concerning gender inequality and development indices for each country was retrieved from the United Nations Human Development Report Data Center ( https://hdr.undp.org/data-center ), and the development indicators considered for the analyses were the Inequality-adjusted Human Development Index (IHDI) and Gender Inequality Index (GII). The IHDI, developed by the United Nations Development Programme, measures human development while accounting for inequality in the distribution of health, education, and income within a population. The indicator integrates income, education and health dimensions, providing a composite measure of socioeconomic context relevant to the capacity to conduct and participate in clinical research. Higher IHDI scores reflect not only greater economic wealth but also higher education levels and better health systems, which collectively contribute to a stronger research environment. Values range from 0 to 1, with higher values indicating higher levels of development after adjusting for inequality. 28 The GII quantifies gender-based disparities in reproductive health, empowerment, and labour market participation. The index ranges from 0 to 1, with higher values indicating greater gender inequality. 29 These indices were chosen because they provide internationally standardised, comparable measures of a country’s overall development and gender equity, both of which are hypothesised to influence the adoption of innovative clinical trial designs. Interpretation of the results is that higher IHDI values indicate a greater likelihood of innovative trial adoption; similarly, higher GII values may indicate structural barriers to adoption related to gender inequality. Other details concerning the indicators are reported in the online supplemental material . A descriptive table of the innovative and traditional trials has been reported in table 1 . Trial characteristics were summarised separately for innovative and traditional trial designs. Categorical variables, including publication period, trial status, participant sex eligibility, age group (paediatric or elderly), trial phase, funding source, continent of study conduct and categories of IHDI and GII, were reported as counts and percentages. The univariable logistic regression model ORs with 95% CIs and p values have also been reported. Percentages are calculated within columns. For descriptive and graphical purposes, IHDI and GII were categorised using data-driven cut-points (IHDI=0.8; GII=0.18), corresponding to inflection regions in the spline-based predicted probability curves; higher IHDI values indicate higher and more equally distributed human development, similarly, higher GII values indicate greater gender inequality. GII, Gender Inequality Index; IHDI, Inequality-adjusted Human Development Index. Trial status was classified according to the ClinicalTrials.gov ‘Overall Status’ field. For analytical purposes, studies were grouped into: (1) non-terminated trials, including recruiting, not yet recruiting, active but not recruiting and successfully completed studies and (2) terminated trials, defined as studies stopped prematurely before planned completion. Quantitative variables, including planned sample size, were summarised using the median and IQR. To describe differences between innovative and traditional trials, we fitted univariable logistic regression models with trial design (innovative vs traditional) as the outcome and each trial characteristic as the predictor, reporting ORs and 95% CIs. Predictor variables were selected a priori to reflect structural and contextual factors hypothesised to be associated with the adoption of innovative trial designs. These included the continent where trials were conducted, to account for broad regional differences in research infrastructure and regulatory environments; the year of trial registration, to capture temporal trends in methodological adoption; and country-level IHDI and GII, chosen as standardised, internationally comparable indicators of socioeconomic development and gender inequality. The relationship between IHDI and GII was explored descriptively using a scatterplot with a fitted linear regression line. The strength and direction of their linear association were quantified using the Pearson correlation coefficient. Because of the conceptual and empirical overlap between IHDI and GII prior to multivariable modelling. The association between the global spread of innovative trials and country-level development indicators was assessed by estimating a multivariable logistic regression model. The outcome was whether a trial used an innovative design, and predictors included the geographical continent of trial conduct (Africa, the Americas, Asia, Europe or Oceania, determined from the ‘Locations’ field in ClinicalTrials.gov), the year of study conduct, and the country-level IHDI and GII. The nonlinear effects were accounted for via Restricted Cubic Spline (RCS) estimation. RCS is a flexible regression approach that represents continuous variables as smooth piecewise polynomial functions joined at predefined knots, allowing non-linear relationships to be captured without assuming a specific parametric form. This approach enabled the identification of inflection points in the association between development indicators and innovative trial adoption while preserving model stability. To address potential multicollinearity between the IHDI and the GII, penalised logistic regression models were estimated using a ridge-type shrinkage approach. Penalisation reduces variance inflation by shrinking regression coefficients towards zero while retaining all predictors. The optimal penalty parameter was selected via bootstrap-based internal validation to maximise model discrimination (Somers’ Dxy). For descriptive and graphical purposes only, selected continuous variables were additionally categorised into low and high groups to facilitate interpretation of model-predicted probabilities and stratified summaries. These cut-points were identified post hoc by visually inspecting the spline-based predicted probability curves and identifying inflection regions where the slope of the association changed most markedly. This approach was used to derive data-driven interpretative thresholds rather than to dichotomise predictors for inferential modelling. Established cut-points from the literature were not imposed a priori because the aim was to describe how innovative trial adoption varied across the observed data distribution. In addition, descriptive plots were produced showing the proportion of innovative trials stratified by low versus high categories of development indices for continents and countries, based on post-hoc interpretative cut-points. These descriptive plots are intended to aid visual interpretation and are distinct from the spline-based model outputs. CIs for proportions were calculated using a mid-p version of the exact binomial method, which provides less conservative interval estimates than the standard Clopper-Pearson approach by adjusting the tail probabilities of the exact test. This method was used to obtain more informative interval estimates while maintaining appropriate coverage properties. 30 Country-level values of the IHDI and GII were visualised using coloured world maps, with countries shaded according to their index values to illustrate the global distribution of socioeconomic development and gender inequality. This manuscript has not been posted as a preprint on any server; the data and codes used for the analyses have been attached as an additional material ‘code.Rmd’. Analyses and data extraction were conducted using R V.3.4.2. 31 No patients or members of the public were involved in the design, conduct, reporting or dissemination of this research.

In total, 380 015 clinical trials were categorised into traditional and innovative trial designs ( table 1 ). Of these interventional trials included in the analysis, 1.7% were classified as using innovative trial designs based on the predefined keyword strategy. Although the absolute number of such trials has increased over time, their proportion relative to all registered trials remains relatively modest. From 1990 to 2024, the proportion of innovative trials increased. Innovative trials showed higher odds of occurrence in more recent periods, with an OR of 1.52 for the period 2020–2024. Active trials largely employed traditional designs, while terminated trials showed a higher propensity for innovative designs, with an OR of 1.28 compared with active trials. Moreover, studies involving female participants or male participants alone were less likely to use innovative designs than those including both males and females. Regarding patient age, paediatric trials showed a higher likelihood of using innovative designs; similarly, trials involving elderly participants showed a higher rate of novel design use. Moreover, industry-funded trials were less likely to employ innovative designs than trials funded by other sources. Regarding geographical assessment, innovative designs were more prevalent in trials conducted in the Americas and developed regions and were least common in Africa. The Americas had a higher proportion of innovative trials with an OR of 3.35 compared with Africa. The broader global relationship in which higher levels of IHDI tend to coincide with lower levels of gender inequality has been assessed. A strong inverse relationship was observed between the IHDI and the GII across countries ( figure 2 ). Higher IHDI values were associated with lower GII values, indicating that countries with higher levels of development adjusted for inequality tend to exhibit lower gender inequality (r=−0.87; p<0.001). The relationship appeared approximately linear over the observed range, with limited scatter about the fitted regression line. The strong inverse correlation suggests substantial shared variance between the two indices. The penalised multivariable analysis reports results from models with dichotomised variables based on the restricted cubic spline cut-off ( table 2 ). Confirmed that trials published after 2015 were more likely to employ innovative designs. Compared with Africa, trials conducted in the Americas, Europe and Oceania were more likely to use innovative designs. Regarding the development indicators, higher values in the IHDI (0.8–0.9 range) increased the likelihood of innovative trial adoption. Lower values in the GII (0.18–0.28 range) were associated with reduced use of innovative designs ( table 2 ). This table presents results from multivariable models using dichotomised predictors for comparison. GII, Gender Inequality Index; IHDI, Inequality-adjusted Human Development Index. Figure 3 presents the estimated probability of a trial adopting an innovative design for each predictor variable, IHDI, GII, geographical continent and year of trial conduct, derived from a multivariable logistic regression model. Each panel illustrates the effect of one predictor adjusted for all other predictors in the model, representing the analysis of associations. In Panel A, the curve shows how the probability of innovative trial publication increases with increasing IHDI values. The restricted cubic spline model estimation showed a marked change in the slope of the association between IHDI and the predicted probability of innovative trial adoption around an IHDI value of approximately 0.8. Below this threshold, the increase in innovative trial probability is gradual but accelerates once IHDI surpasses 0.8, indicating that higher development levels significantly encourage the adoption of novel trial designs. Panel B shows that as the GII increases, indicating higher levels of gender inequality, the probability of conducting novel trials decreases. The cut point at a GII of 0.18 emphasises that beyond this point, gender inequality may substantially worsen advancements in clinical trial methodologies. The trend depicted in Panel C highlights the evolution of trial designs over time, with an increase in the adoption of innovative designs, particularly evident after 2015. In the spline analyses, values around IHDI=0.8 and GII=0.18 represent data-driven inflection regions in the observed associations and should not be interpreted as normative policy thresholds. Figure 4 presents a forest plot summarising the observed proportions of innovative trial designs across continents. These estimates are not adjusted for other covariates and should therefore be interpreted as descriptive summaries rather than results from the regression models. The blue horizontal lines represent continents with a high IHDI score, while the red points and lines denote those with a low IHDI score. The results show that the Americas and Europe have higher proportions of innovative trial design adoption, as indicated by their high IHDI scores. In contrast, Asia and Africa have lower proportions, with Africa the lowest overall. The results show that the Americas and Europe have higher proportions of innovative trial design adoption, which correspond to their relatively high IHDI scores and, conversely, lower GII values. Online supplemental figure S1 in the online supplemental material provides a descriptive visualisation of the relationship between IHDI and the distribution of innovative trial designs. Countries with higher IHDI levels evidenced a higher proportion of novel trial designs, as demonstrated in Panel A of online supplemental figure S1 in the online supplemental material . Additionally, countries with a higher GII tend to have their estimates to the left, suggesting less frequent use of innovative trial designs, as shown in Panel B of online supplemental figure S1 . Online supplemental figure S2 shows the global distribution of IHDI (Panel A) and GII (Panel B). These maps provide an exploratory, descriptive overview of the geographical patterns of development and gender inequality across countries. Higher IHDI values are predominantly observed in North America, Europe and Oceania, while lower values are concentrated in sub-Saharan Africa and parts of South Asia. Conversely, higher GII values are mainly distributed across sub-Saharan Africa, the Middle East and parts of South Asia; similarly, lower values are observed in Europe, North America and Oceania.

This study highlights significant differences in the global adoption of innovative clinical trials across time, demographics and regions. Recognition of their benefits, especially since 2015, has led to increased implementation. The rise in innovative trial designs published on ClinicalTrials.gov indicates a shift toward more sophisticated methodologies, likely driven by enhanced regulatory support and global health initiatives. 15 This result is in line with the literature, which has evidenced an increase in clinical trial complexity both from the study design and clinical trial conduction perspective. 32 33 In this large registry-based analysis, only 1.7% of trials were identified as using innovative designs. This relatively small proportion is notable given the increasing methodological interest in adaptive, Bayesian and master-protocol approaches in the clinical trials literature. The limited representation of such designs in registry data may reflect barriers to implementation, including methodological complexity, regulatory uncertainty, limited statistical expertise or incomplete reporting of design features in trial registries. 20 Despite increasing interest in novel and flexible trial design solutions, this research highlights disparities in their adoption globally. The strong inverse correlation observed between the IHDI and the GII indicates that these indicators capture closely related structural dimensions of national contexts. As a result, their individual effects cannot be cleanly disentangled within the present analytical framework. Although penalised regression approaches were used to mitigate collinearity and stabilise estimation, the observed associations should be interpreted as reflecting broader contextual environments characterised by varying levels of socioeconomic development and gender equity rather than as independent or causal effects of either indicator. Higher IHDI values were associated with a greater likelihood of innovative trial adoption, suggesting that innovative designs tend to be more frequently observed in settings characterised by more developed and less unequal health, education and economic systems. However, given the strong correlation between IHDI and GII, this association should not be interpreted as an independent effect of socioeconomic development alone. Rather, it likely reflects broader structural environments in which multiple dimensions of development and equity coexist and jointly influence research capacity and the feasibility of implementing complex trial designs. The spline-based analysis indicated a change in the slope of the association between IHDI and the adoption of innovative trial designs around an IHDI value of approximately 0.8, reflecting non-linear variation across levels of socioeconomic development rather than a substantive threshold. In countries with higher socioeconomic status, there is better healthcare infrastructure, increased funding opportunities and greater access to innovative technology, all of which are important for implementing novel research methodologies. 3 7 This infrastructure supports the operational needs of complex trials and ensures that regulatory, ethical and scientific standards are maintained. The IHDI was selected because it integrates income, education and health dimensions, providing a composite measure of socioeconomic context relevant to the capacity to conduct and participate in clinical research. Higher IHDI scores reflect not only greater economic wealth but also higher education levels and better health systems, which collectively contribute to a stronger research environment. These conditions facilitate the training of skilled researchers, the development of sophisticated research facilities and the implementation of complex study designs. 6 The dependency of innovative trial adoption on higher IHDI values raises concerns regarding global health equity. Regions with lower IHDI scores might not only lack access to the benefits of innovative trial designs but also contribute less to the global body of clinical research, potentially leading to health solutions that are less tailored to their unique health challenges. 20 To address the challenges posed by lower IHDI scores, international collaboration and targeted funding are demanded. Improving healthcare infrastructure and providing substantial funding opportunities are essential to support innovative clinical trials in these regions. 7 Development agencies and global health initiatives can focus on building research capacities by investing in local healthcare systems, training local researchers and providing the necessary technological tools for innovative research methodologies. The international collaboration between countries with high and low IHDI values presents a viable strategy to bridge the disparities in innovative clinical trial adoption. Collaborative efforts can facilitate knowledge transfer, infrastructure development and capacity building in lower IHDI regions, fostering more equitable clinical research. Successful partnerships between high-income and lower-income countries have demonstrated the potential for shared expertise to improve clinical research outcomes. 34 Initiatives such as joint clinical trial networks, data-sharing agreements and co-sponsored funding mechanisms can help mitigate resource limitations and regulatory hurdles in lower IHDI settings. 35 Similarly, higher GII values were associated with lower adoption of innovative trial designs. Also in this case, this relationship should be interpreted with caution. Given the substantial overlap between GII and IHDI, gender inequality likely represents one dimension of a broader structural context rather than an isolated determinant. Instead, they suggest that environments with lower gender inequality are more conducive to methodological innovation in clinical research. This underscores how broader societal and structural inequalities are associated with differential adoption of innovative clinical trial designs, with implications for diversity and inclusivity in global clinical research. Gender inequality not only affects women’s participation rates but also influences health outcomes for the entire population. Studies show that diseases affecting women, especially sex-specific ones, often receive less attention and funding due to gender bias in research resource allocation. 36 Furthermore, when women are underrepresented in clinical trials, the findings may not effectively address or generalise to half of the population, potentially leading to suboptimal healthcare outcomes. 37 Gender disparities in research funding further contribute to the limited adoption of innovative trial designs, hindering innovation and equity in clinical research. Moreover, research on women’s health has historically been poorly funded, directly impacting the quality of innovative clinical trials in this field. 38 The disparities highlighted by higher GII values suggest that countries with significant gender inequality lack the policies or cultural support to encourage gender diversity in science and research. These settings face barriers such as limited women’s education access, fewer career opportunities in scientific fields and inadequate support for women pursuing research careers. 39 Gender-sensitive funding policies, such as the adoption of equitable criteria in grant allocation and dedicated programmes for female-led research, can help close the gap. 40 Greater transparency in funding distribution and inclusive research strategies can facilitate female participation in innovative research, increasing the diversity and relevance of clinical studies on a global scale. 41 Strengthening institutional support for gender equality in science and innovation is crucial to ensuring that opportunities for innovative clinical research are accessible to all, ultimately leading to more representative and impactful healthcare solutions. 42 The research findings also indicate an observed lower use of innovative designs in sex-specific trials overall, including both female-only and male-only studies, compared with trials enrolling both sexes. This pattern may reflect differences in therapeutic areas, trial scale, funding models, recruitment feasibility or regulatory pathways rather than disparities affecting one sex alone. 43 This includes creating mandates that require gender diversity among study participants and ensuring that studies adequately address gender-specific health issues. Additionally, policies should focus on removing barriers to women’s participation in research by supporting female scientists’ career development and promoting gender-sensitive research. 40 Furthermore, global reforms that promote a more equitable distribution of research funding are urgently needed. Ensuring that all researchers, regardless of gender or geographic location, have access to the resources necessary to conduct innovative clinical trials would foster greater inclusivity and diversity in the global research landscape. 44 45 In some cases, facilitating female participation in clinical research could not be the only solution; analytical methods tailored for profiling responses per gender and patient characteristics could be useful for improving the possibility of defining patient-specific effects. For example, the adoption of machine learning techniques could help to overcome some barriers. The use of algorithms for data analysis, trial logistics management and the personalisation of experimental designs enables more efficient and adaptive implementation, even in resource-limited settings. 46 For example, artificial intelligence facilitates the optimisation of patient recruitment strategies, the prediction of treatment responses and real-time data analysis, reducing operational burdens and improving the sustainability of trials. 47 48 Our analysis reveals global disparities in the distribution of innovative trials, with the highest proportions in the Americas and Europe and the lowest in Africa. These disparities underscore the need for international collaboration and policy initiatives to improve trial distribution in underrepresented regions. Remote monitoring and telemedicine can bridge geographical gaps, increase participant diversity and reduce logistical costs, promoting more inclusive and efficient research practices. 49 Beyond immediate inequities, the long-term impact of underrepresentation in innovative trials is particularly concerning for diseases that disproportionately affect neglected regions or marginalised populations. The lack of research on conditions prevalent in low-IHDI countries and among underrepresented genders leads to gaps in medical knowledge, delays in the development of targeted treatments and reduced effectiveness of global health interventions. For example, neglected tropical diseases remain underfunded and understudied due to limited research capacity in affected regions, exacerbating health disparities. 50 Similarly, conditions that primarily affect women, such as endometriosis or autoimmune diseases, often receive less attention and funding, leading to delayed diagnoses and suboptimal treatment options. 51 52 This research highlighted disparities in the conduct of innovative clinical trials across regions and demographics, underscoring global health equity issues. Strategic solutions in funding, policy reforms and technologies like telemedicine can foster more inclusive health solutions. Educational programmes are important in raising awareness about diversity and inclusivity in clinical trials, encouraging public participation and sensitising healthcare professionals to the need for gender and geographic inclusivity in recruitment and implementation. 53 Addressing these disparities is not only a matter of fairness but a critical factor in improving global health outcomes. Ensuring equitable representation in clinical trials will help develop treatments that are effective across diverse populations, ultimately leading to more comprehensive and impactful healthcare solutions worldwide.

This study documents the growth of innovative clinical trial designs since 2015 and highlights substantial global disparities in their adoption. Higher uptake was observed in regions such as the Americas and Europe and lower uptake in Africa, in association with country-level indicators of development and gender inequality. However, given the strong correlation between these indicators, their independent effects cannot be disentangled, and findings should be interpreted as reflecting broader structural contexts rather than causal relationships. The adoption of innovative trial designs appears closely linked to underlying research capacity and contextual environments. Efforts to promote more equitable implementation will likely require coordinated, system-level strategies rather than isolated interventions.

This study demonstrates global disparities in the use of innovative clinical trial designs, which are associated with differences in socioeconomic development. The data also lacks detail to distinguish between gender and sex in participation. Further research is needed to understand the impacts of sex as a biological variable and gender as a sociocultural factor and to address the underrepresentation of gender-diverse populations, such as the transgender community, in clinical trials. 54 Another limitation of this study is that participant-level sex distribution was not consistently available in registry records. Consequently, we were unable to assess the relative gender representation in innovative versus traditional trial designs, or to relate participant sex distributions to country-level GII. Future research should incorporate participant-level sex and gender data to better contextualise gender inequities in clinical trial participation. Moreover, trials conducted in the Americas were analysed as a single group; substantial heterogeneity likely exists across North, Central and South America, reflecting differences in socioeconomic development and research capacity; future work could explore subregional patterns in greater detail. While the study findings highlight important disparities, the present study does not permit causal inference about the effects of modifying specific structural factors, such as socioeconomic development or gender inequality. Therefore, the suggested policy implications should be interpreted cautiously. Efforts to improve research capacity, inclusivity and access to methodological expertise may contribute to more equitable environments for adopting innovative trial designs, but the relative contribution of individual structural components cannot be isolated in this analysis.