Julia Phillipson

Background: Adaptive designs (ADs) are increasingly used, whereby outcome data accumulated during the trial may inform a trial’s course in accordance with pre-specified rules at interim analyses. The benefits however will only be realised if the approaches to conducting interim analyses are optimised. This study aims to identify existing literature highlighting best practice (and existing evidence gaps) with regard to conducting high-quality, timely interim analyses. Methods: Medline and Embase databases were searched for studies published between 2005 and July 2025 to identify best practice models and lessons learned for interim analyses in the delivery of adaptive trials. Specifically, we considered papers that discussed adaptive trial methodological approaches (Phases II–IV) in any clinical population and condition. A narrative synthesis focused on the following outcomes was conducted: design considerations, project/trial management, data management, statistical processes, trial committee processes, the implementation of interim decisions and patient and public involvement and engagement (PPIE). Results: We screened 6720 articles from databases and citation chaining and assessed 329 articles at full text. One hundred and one articles representing 92 unique studies were deemed eligible for inclusion. Key issues included the following: (a) Detailed planning of interim analyses set out in the protocol, and all patient information sheets for possible changes prepared in advance; (b) Effective communication and collaborative decision-making processes across stakeholders, effective staff training, and prompt site query resolutions; (c) Use of electronic data capture with automated data flow processes with integrated query processes to improve data quality; and (d) Use of secure databases with data transfer procedures to maintain data integrity. Conclusion: The currently available data suggest that although there is a considerable volume of evidence with regards to the conduct of adaptive trials and good trial management, the existing literature base offers limited specific guidance for interim analyses. There are significant gaps in the literature regarding best practice guidance related to PPIE and statistical considerations in adaptive trials. While there are examples of innovative methods speeding up the collation and analysis of data used for interim analyses, we advocate more robust research exploring the operational opportunities and challenges with regards to undertaking adaptive trials.

Background: Orthostatic hypotension is a significant drop in blood pressure upon standing upright. It is very common and can result in symptoms such as postural dizziness, fainting and falls. Within the United Kingdom National Health Service, there are three principal treatments: non-drug therapies, and two medications – fludrocortisone or midodrine. Despite this we do not know which treatments are the most effective, nor whether they are cost-effective. Objective: Evaluate the feasibility of a randomised trial to evaluate the clinical and cost-effectiveness of fludrocortisone and midodrine in comparison to non-drug therapies for the treatment of orthostatic hypotension. Design: A 10-month pilot of a pragmatic, open-label, randomised, prospective, superiority, multiarm, multistage clinical trial. The pilot evaluated recruitment, attrition, crossover and quality of outcomes. Setting: Falls and Syncope, Movement Disorder, Geriatrics, and Cardiology clinics in United Kingdom National Health Service secondary care. Participants: Adults with symptomatic orthostatic hypotension. Interventions: Control: Non-drug therapies (conservative management). Interventions: Conservative management plus fludrocortisone (50–400 mcg daily) or conservative management plus midodrine (5–30 mg daily). Main outcome measures: Recruitment: Target was 40–64 participants from 14 sites. Attrition target: ≤ 15% participants withdraw before the primary end point. Crossover: To determine the rates of crossover between intervention arms. Outcome data: Assess the quality and completeness of outcomes. Other important outcomes included feedback via questionnaire and interview from sites and participants. Results: Two hundred and eighty-two patients were screened for eligibility during the pilot; of these, 13 were recruited from 4 of 9 open sites. Current or recent use of one of the study medications accounted for 120 (52%) exclusions due to ineligibility. At the primary end point of 6 months, 10 of the 13 participants (77%) remained in the study. Of those, completion rates for primary and secondary outcomes were 100%, except for the falls diaries, which was 60%. Feedback from sites revealed that redeployment of clinical and research staff due to COVID-19 negatively impacted on site opening and screening for eligible participants. Adapting the protocol to make it more flexible for remote clinics and more pragmatic for clinical use did not improve recruitment. Limitations: The sample size is too small to provide a reliable estimate of attrition and crossover rates for future studies. Conclusions: This study was not feasible in its current design. COVID-19 had an impact on staffing and site opening, while the exclusion criteria limited recruitment. Future work: To answer the research question, novel trial designs are needed. Funding: This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number NIHR127385.

Background: Adaptive designs are increasingly being used in clinical trials within diverse clinical areas. They can offer advantages over traditional non-adaptive approaches, including improved efficiency and patient benefit. The level of improvement observed in practice depends to a large degree on conducting interim analyses (at which adaptations can be made to the trial based on collected data) rapidly and to a high standard. Methods: The ROBust INterims for adaptive designs (ROBIN) project aimed to identify best practice for conducting high-quality and rapid interim analyses. This was done through evidence synthesis of published work, qualitative research with trial stakeholders working at public sector clinical trials units, engagement with patients and the public, and a meeting of trial stakeholders to discuss findings and agree recommendations. Results: This paper provides recommendations for teams that conduct adaptive trials about how to ensure interim analyses are done rapidly and to a high standard. We break down recommendations by stage of the trial. We also identify a lack of methodology on how best to involve patients in adaptive trials and related decision-making. A limitation of our recommendations is that the research was mostly focused on UK academic settings, although we believe much of the recommendations are relevant in other countries and to industry-sponsored trials. Conclusions: When following the recommendations outlined in this paper, the process of planning and executing interim analyses will be smoother; in turn, this will lead to more benefits from using adaptive designs.

Health Technology Assessment Volume: 25, Issue: 63, Published in November 2021