Search for a command to run...
Complexities in contemporary clinical trials are frequently self-defeating relative to the goals of clinical research -namely to inform and improve care for as many patients as possible. Trials need to be rigorous, relevant, ethical, inclusive, feasible and affordable. Trials require broad access to research participation to optimize the generalizability of results and influence practice and improve outcomes for diverse populations. The degree of complexity of many clinical trials today too often restricts their conduct to academic centers or specialized research organizations where only sufficiently resourced investigators can conduct them, narrowing the populations of participants having access to trials. Limiting who conducts, delivers, and participates in trials decreases the representation of participants and threatens the generalizability of trial results to underrepresented persons. Both internal and external validity may be negatively affected.Unnecessary complexity risks recruitment and retention challenges, slows time to trial completion, and requires additional research resources, driving up costs often with unclear benefit. Delayed knowledge acquisition creates lags in evidence generation and application in practice. The benefits of health care advances are particularly slow to reach marginalized populations and those in low-and middle-income countries. Experiences during the COVID-19 pandemic provided a window into ways that trials could become simpler and more accessible. However, much of the regulatory guidance that allowed these adjustments has reverted to 'business as usual', foregoing these useful simplifications.We outline how the trials enterprise has lostthe proportionality that should inform trial complexity. Humans consistently default to additive changes and are less likely to identify opportunities for subtractive change -that is, removing components and complexity. 1 However, the 'more is more' approach runs counter to our ethical imperatives to represent relevant, diverse populations in our research, limit participant burdens, and to efficiently and economically generate impactful results. Although leading trialists have been calling for the design of large, simple trials for decades, the trend has been toward more complex procedures, rather than simplicity since then 2 . Large, simple trials are an obvious starting point for the issues we highlight here. Proportional simplicity requires retaining more stringent approaches for trials that are 'high risk' (first-in-human, vulnerable populations, high-risk interventions (e.g. chimeric antigen receptor T cells), registrational trials, or those focused on mechanistic exploration. However, even the complexity of these 'higher risk' trials needs careful review and reconsideration.The creep in trial complexity is multifactorial. Necessary protections have been added to address structural research harms that have disproportionately impacted vulnerable populations, including Black and Indigenous communities and incarcerated people, and populations in formerly colonized countries. Complexities were introduced to satisfy regulatory authorities (Food and Drug Administration (FDA), European Medicines Agency (EMA), etc.) for new drugs and devices, but are often applied across the board, including in trials of lower-risk and post-approval interventions. A vicious cycle of exhaustive data collection and monitoring, and query generation and feedback, can penalize lack of exact adherence to processes for research personnel and adds unnecessary burden for participants. This 'regulatory-industrial complex' 3 is enforced by contract research organizations (CROs) -generally for-profit companies -whose role is to execute and/or monitor trials on behalf of sponsors. There is a tipping point at which the unintended consequences of complexity jeopardize trial feasibility and affordability, especially as future research funding is likely to be further constricted in many jurisdictions.Returning to the premise that trials should be rigorous, relevant, ethical, inclusive, feasible and affordable, means that judicious subtractive change must come from funders and regulators but involve all impacted parties, especially affected communities (strategies in Table ). Trial simplification could initially focus on late-phase and post-approval studies, but be addressed for most designs, retaining more stringent approaches for trials that are 'high risk'.The 2025 E6(R3) revision to The International Council on Harmonisation (ICH) Guideline for Good Clinical Practice 4 emphasizes proportionality in trials. If implemented correctly, monitoring proportionality increases flexibility of trial designs while retaining focus on participant safety and data accuracy.Decreasing the stringency of trial procedures and processes could allow for greater participation of community and research-naïve sites, and lead to inclusion of participants outside of major medical centers. Revisions to ICH decrease emphasis on perfection, instead favoring innovations that allow flexibility to bring trials to people via remote or hybrid designs, and enable patient-collected data. Regulatory and ethics authorities, such as the FDA and EMA, national offices of human research protections, sponsors, and monitors should begin, where feasible, to integrate clinically-obtained data, allowing measurements from self-performed home vital sign monitoring, wearable physiologic devices, and blood self-collection kits.Mechanistic substudies or more complex procedures, including collection and banking of samples that require specialized processing or cryopreservation, could be limited to selected, well-resourced research sites and be optional for other sites. Protocols should allow for accredited, non-research laboratories or other standard-of-care procedures to report data without extensive re-validation. Allowing procedures to be performed locally, or remotely, or with more flexible hours facilitates participation by those with shift-work jobs, those who require childcare coverage (a common reason women underenroll in research 5 ), or those who live outside metropolitan areas.Lengthy informed consent documents, often further complicated by each institutions' risk management and legal departments, can create a façade of integrity with no face validity. Lengthy consent documents may deter potential participants and discriminate against others, particularly those with low (health) literacy or earned distrust in the healthcare system. Re-consenting participants for administrative changes that do not impact risk should be minimized, as reconsentingmay only increase participant burden and jeopardize their retention. One way to include impacted communities and decrease the research burden is to include Community Advisory Board (CAB) feedback on study design, with special focus on the informed consent documents and schedule of activities 6,7 . More than requiring that the informed consent document meets a specified reading level, community imput can make sure that the document places appropriate emphasis on the components that are the most salient in their view after learning about the proposed intervention in the context of their disease 8 . This feedback will also inform whether potential participants find the planned research activities feasible (e.g., regarding number of questionnaires) and affordable (e.g., personal travel to research centers in follow-up). Clear, simple informed consent processes must extend beyond the written informed consent document to facilitate participation by people who speak various languages, considering synchronous medical translation services, visual abstracts, or video-assisted informed consent encounters to illustrate trial processes 9 .Extensive trial eligibility criteria should be simplified to enroll the broadest participant population that maintains safety and responsiveness to the research question. Trials attempting to evaluate a "clean" population with only the disease under study will naturally exclude older people and those from marginalized backgrounds who are more likely to have accumulated comorbidities. Women, children, the elderly, and those with common medical conditions are frequently excluded from multicenter trials, which should be transparently justified in protocols, or better yet, reconsidered 10 .The introduction of a single institutional review board has the potential to simplify and streamline ethics review. This has been realized in many situations, but requires tracking to be sure that an approach doesn't duplicate work or increase the regulatory and reporting requirements to accommodate all parties. The US National Cancer Institute launched their Central Institutional Review Board (CIRB) in 2001, and has demonstrated time savings and faster approvals in regulatory submissions made by partner institutions. One report demonstrated cost-savings, except that as used at the time of this study, many institutions were not fully delegating review to CIRB, and this analysis was limited in studying only time for local approval after the study was centrally approved 11 .Intensity of data collection should mirror the trial's aim. For some trials, a very small proportion of data collected on lengthy case report forms is ever used, creatingwaste and opportunity cost. 12 Further, data is increasingly being leveraged from electronic health records and registries for trials. For example, the TASTE study, which evaluated thrombus aspiration during ST-elevation myocardial infarctions, embedded study screening into clinical practice, relied on clinical data, and cost a fraction of most trials. 13 Advances in electronic health record interoperability and data standardization could support re-use of health data for multiple purposes, and augment clinical trial data infrastructure in the future 14 .Adverse Event (AE) reporting, similarly, can be contextualized to avoid wasting resources on detailed documention that has no impact on either the trial's internal validity or participant safety. For trials testing interventions already in clinical practice, events that are possible consequences of that intervention could be pre-determined as trial outcomes rather than AEs; events that are part of the natural history of a disease process should similarly not trigger AEs. 15 The scope and frequency of data monitoring should be proportionate to risk and adapted to early trial findings. One approach, supported by ICH E6(R3), is 'risk-based monitoring', whereby intensive monitoring is reserved for ethics assurances (eg. informed consent), outcomes that would change study results if missing or erroneous, and key safety data. The over-used label 'pragmatic trial' is not a synonym for casually designed and implemented; proportional but robust approaches to monitor key details are also critical for pragmatic trials.Changes to data collection and data monitoring will require concerted collaborative efforts from regulatory authorities and sponsors, including direction to CROs and investigators to focus on subtractive change and avoid 'one size fits all' approaches 16 .The pandemic underscored the vital role of research for society, highlighting the need to optimize and streamline procedures and processes, while maintaining ethics and rigor. Some trials for emerging infections innovated within this broken research infrastructure, 17 but the present maximalistapproach continues to challenge principals of making research accessible, equitable, timely, inclusive, and applicable. 18 What if Marie Kondo wrote a protocol? We must carefully consider which parts of clinical trials do serve participants and uphold research integrity, and which are needless clutter. The art of subtractive change, popularized by Kondo 16 and long-championed in the Toyota Way of continuous improvement 19 , can be applied to clinical trials. Subtractive change can facilitate our collective aims without exhausting the limited resources of the research enterprise, the most important of which is our study participants.