Estimation of Treatment Policy Estimands for Continuous Outcomes Using Off-Treatment Sequential Multiple Imputation.

The estimands framework outlined in ICH E9 (R1) describes the components needed to precisely define the effects to be estimated in clinical trials, which includes how post-baseline 'intercurrent' events (IEs) are to be handled. In late-stage clinical trials, it is common to handle IEs like 'treatment discontinuation' using the treatment policy strategy and target the treatment effect on outcomes regardless of treatment discontinuation. For continuous repeated measures, this type of effect is often estimated using all observed data before and after discontinuation using either a mixed model for repeated measures (MMRM) or multiple imputation (MI) to handle any missing data. In basic form, both these estimation methods ignore treatment discontinuation in the analysis and therefore may be biased if there are differences in patient outcomes after treatment discontinuation compared with patients still assigned to treatment, and missing data being more common for patients who have discontinued treatment. We therefore propose and evaluate a set of MI models that can accommodate differences between outcomes before and after treatment discontinuation. The models are evaluated in the context of planning a Phase 3 trial for a respiratory disease. We show that analyses ignoring treatment discontinuation can introduce substantial bias and can sometimes underestimate variability. We also show that some of the MI models proposed can successfully correct the bias, but inevitably lead to increases in variance. We conclude that some of the proposed MI models are preferable to the traditional analysis ignoring treatment discontinuation, but the precise choice of MI model will likely depend on the trial design, disease of interest and amount of observed and missing data following treatment discontinuation.

Proactive integrated consultation-liaison psychiatry and time spent in hospital by older medical inpatients in England (The HOME Study): a multicentre, parallel-group, randomised controlled trial.

BACKGROUND: Older people admitted to hospital in an emergency often have prolonged inpatient stays that worsen their outcomes, increase health-care costs, and reduce bed availability. Growing evidence suggests that the biopsychosocial complexity of their problems, which include cognitive impairment, depression, anxiety, multiple medical illnesses, and care needs resulting from functional dependency, prolongs hospital stays by making medical treatment less efficient and the planning of post-discharge care more difficult. We aimed to assess the effects of enhancing older inpatients' care with Proactive Integrated Consultation-Liaison Psychiatry (PICLP) in The HOME Study. We have previously described the benefits of PICLP reported by patients and clinicians. In this Article, we report the effectiveness and cost-effectiveness of PICLP-enhanced care, compared with usual care alone, in reducing time in hospital. METHODS: We did a parallel-group, multicentre, randomised controlled trial in 24 medical wards of three English acute general hospitals. Patients were eligible to take part if they were 65 years or older, had been admitted in an emergency, and were expected to remain in hospital for at least 2 days from the time of enrolment. Participants were randomly allocated to PICLP or usual care in a 1:1 ratio by a database software algorithm that used stratification by hospital, sex, and age, and randomly selected block sizes to ensure allocation concealment. PICLP clinicians (consultation-liaison psychiatrists supported by assisting clinicians) made proactive biopsychosocial assessments of patients' problems, then delivered discharge-focused care as integrated members of ward teams. The primary outcome was time spent as an inpatient (during the index admission and any emergency readmissions) in the 30 days post-randomisation. Secondary outcomes were the rate of discharge from hospital for the total length of the index admission; discharge destination; the length of the index admission after random allocation truncated at 30 days; the number of emergency readmissions to hospital, the number of days spent as an inpatient in an acute general hospital, and the rate of death in the year after random allocation; the patient's experience of the hospital stay; their view on the length of the hospital stay; anxiety (Generalized Anxiety Disorder-2); depression (Patient Health Questionnaire-2); cognitive function (Montreal Cognitive Assessment-Telephone version); independent functioning (Barthel Index of Activities of Daily Living); health-related quality of life (five-level EuroQol five-dimension questionnaire); and overall quality of life. Statisticians and data collectors were masked to treatment allocation; participants and ward staff could not be. Analyses were intention-to-treat. The trial had a patient and public involvement panel and was registered with ISRTCN (ISRCTN86120296). FINDINGS: 2744 participants (1399 [51·0%] male and 1345 [49·0%] female) were enrolled between May 2, 2018, and March 5, 2020; 1373 were allocated to PICLP and 1371 to usual care. Participants' mean age was 82·3 years (SD 8·2) and 2565 (93·5%) participants were White. The mean time spent in hospital in the 30 days post-randomisation (analysed for 2710 [98·8%] participants) was 11·37 days (SD 8·74) with PICLP and 11·85 days (SD 9·00) with usual care; adjusted mean difference -0·45 (95% CI -1·11 to 0·21; p=0·18). The only statistically and clinically significant difference in secondary outcomes was the rate of discharge, which was 8.5% higher (rate ratio 1·09 [95% CI 1·00 to 1·17]; p=0·042) with PICLP-a difference most apparent in patients who stayed for more than 2 weeks. Compared with usual care, PICLP was estimated to be modestly cost-saving and cost-effective over 1 and 3, but not 12, months. No intervention-related serious adverse events occurred. INTERPRETATION: This is the first randomised controlled trial of PICLP. PICLP is experienced by older medical inpatients and ward staff as enhancing medical care. It is also likely to be cost-saving in the short-term. Although the trial does not provide strong evidence that PICLP reduces time in hospital, it does support and inform its future development and evaluation. FUNDING: UK National Institute for Health and Care Research.

Risk Factors for Rectal Cancer Recurrence after Local Excision of T1 Lesions from a Decade-Long Multicenter Retrospective Study.

Background: Local surgical excision of T1 rectal adenocarcinoma is a well-established approach. Yet, there are still open questions regarding the recurrence rates and its risk factors. Methods: A retrospective multicenter study including all patients who underwent local excision of early rectal cancer with an open or MIS approach and had a T1 lesion from 2010 to 2020 in six academic centers. Data included demographics, preoperative studies, surgical findings, postoperative outcomes, and local and systemic recurrence. A univariable and multivariable logistic regression analysis was performed to identify risk factors for recurrence. Results: Overall, 274 patients underwent local excision of rectal lesions. Of them, 97 (35.4%) patients with a T1 lesion were included in the cohort. The mean age was 69 ± 10.5 years, and 42 (43.3%) were female. The mean distance of the lesions from the anal verge was 7.8 ± 3.2 cm, and the average tumor size was 2.7 ± 1.6 cm. Eighty-two patients (85%) had a full-thickness resection. Eight patients (8%) had postoperative complications. Kikuchi classification of submucosal (SM) involvement was reported in 29 (30%) patients. Twelve patients had SM1, two SM2, and fifteen SM3. Following pathology, 24 patients (24.7%) returned for additional surgery or treatment. The overall recurrence rate was 14.4% (14 patients), with 11 patients having a local recurrence and 6 having a systemic metastatic recurrence, 3 of which had both. The mean time for recurrence was 2.78 ± 2.8 years and the overall mortality rate was 11%. On univariable and multivariable logistic regression analysis of recurrence vs. non-recurrence groups, the strongest and most significant association and possible risk factors for recurrence were larger lesions (4.3 vs. 2.5 cm, p < 0.001) with an OR of 6.67 (CI-1.82-24.36), especially for tumors larger than 3.5 cm, mucinous histology (14.3% vs. 1.2%, p = 0.004, OR of 14.02, CI-1.13-173.85), and involved margins (41.7% vs. 16.2%, p = 0.003, OR of 9.59, CI-2.14-43.07). The open transanal excision (TAE) approach was also identified as a possible significant risk factor in univariant analysis, while SM3 level penetration showed only a trend. Conclusion: Surgical local excision of T1 rectal malignancy is a safe and viable option. Still, one in four patients received additional treatment. There is an almost 15% chance for recurrence, especially in large tumors, mucinous histology, or involved margin cases. These high-risk patients might warrant additional intervention and stricter surveillance protocols.

Sample-to-Answer Detection of SARS-CoV-2 Viremia Using Thermally Responsive Alkane Partitions.

The diagnosis of bloodborne viral infections (viremia) is currently relegated to central laboratories because of the complex procedures required to detect viruses in blood samples. The development of point-of-care diagnostics for viremia would enable patients to receive a diagnosis and begin treatment immediately instead of waiting days for results. Point-of-care systems for viremia have been limited by the challenges of integrating multiple precise steps into a fully automated (i.e., sample-to-answer), compact, low-cost system. We recently reported the development of thermally responsive alkane partitions (TRAPs), which enable the complete automation of diagnostic assays with complex samples. Here we report the use of TRAPs for the sample-to-answer detection of viruses in blood using a low-cost portable device and easily manufacturable cassettes. Specifically, we demonstrate the detection of SARS-CoV-2 in spiked blood samples, and we show that our system detects viremia in COVID-19 patient samples with good agreement to conventional RT-qPCR. We anticipate that our sample-to-answer system can be used to rapidly diagnose SARS-CoV-2 viremia at the point of care, leading to better health outcomes for patients with severe COVID-19 disease, and that our system can be applied to the diagnosis of other life-threatening bloodborne viral diseases, including Hepatitis C and HIV.

An evaluation of sample size requirements for developing risk prediction models with binary outcomes.

BACKGROUND: Risk prediction models are routinely used to assist in clinical decision making. A small sample size for model development can compromise model performance when the model is applied to new patients. For binary outcomes, the calibration slope (CS) and the mean absolute prediction error (MAPE) are two key measures on which sample size calculations for the development of risk models have been based. CS quantifies the degree of model overfitting while MAPE assesses the accuracy of individual predictions. METHODS: Recently, two formulae were proposed to calculate the sample size required, given anticipated features of the development data such as the outcome prevalence and c-statistic, to ensure that the expectation of the CS and MAPE (over repeated samples) in models fitted using MLE will meet prespecified target values. In this article, we use a simulation study to evaluate the performance of these formulae. RESULTS: We found that both formulae work reasonably well when the anticipated model strength is not too high (c-statistic < 0.8), regardless of the outcome prevalence. However, for higher model strengths the CS formula underestimates the sample size substantially. For example, for c-statistic = 0.85 and 0.9, the sample size needed to be increased by at least 50% and 100%, respectively, to meet the target expected CS. On the other hand, the MAPE formula tends to overestimate the sample size for high model strengths. These conclusions were more pronounced for higher prevalence than for lower prevalence. Similar results were drawn when the outcome was time to event with censoring. Given these findings, we propose a simulation-based approach, implemented in the new R package 'samplesizedev', to correctly estimate the sample size even for high model strengths. The software can also calculate the variability in CS and MAPE, thus allowing for assessment of model stability. CONCLUSIONS: The calibration and MAPE formulae suggest sample sizes that are generally appropriate for use when the model strength is not too high. However, they tend to be biased for higher model strengths, which are not uncommon in clinical risk prediction studies. On those occasions, our proposed adjustments to the sample size calculations will be relevant.

Determining sample size in a personalized randomized controlled (PRACTical) trial.

In clinical settings with no commonly accepted standard-of-care, multiple treatment regimens are potentially useful, but some treatments may not be appropriate for some patients. A personalized randomized controlled trial (PRACTical) design has been proposed for this setting. For a network of treatments, each patient is randomized only among treatments which are appropriate for them. The aim is to produce treatment rankings that can inform clinical decisions about treatment choices for individual patients. Here we propose methods for determining sample size in a PRACTical design, since standard power-based methods are not applicable. We derive a sample size by evaluating information gained from trials of varying sizes. For a binary outcome, we quantify how many adverse outcomes would be prevented by choosing the top-ranked treatment for each patient based on trial results rather than choosing a random treatment from the appropriate personalized randomization list. In simulations, we evaluate three performance measures: mean reduction in adverse outcomes using sample information, proportion of simulated patients for whom the top-ranked treatment performed as well or almost as well as the best appropriate treatment, and proportion of simulated trials in which the top-ranked treatment performed better than a randomly chosen treatment. We apply the methods to a trial evaluating eight different combination antibiotic regimens for neonatal sepsis (NeoSep1), in which a PRACTical design addresses varying patterns of antibiotic choice based on disease characteristics and resistance. Our proposed approach produces results that are more relevant to complex decision making by clinicians and policy makers.

Changes in the prevalence of mental health problems during the first year of the pandemic: a systematic review and dose-response meta-analysis.

AIM: To describe the pattern of the prevalence of mental health problems during the first year of the COVID-19 pandemic and examine the impact of containment measures on these trends. METHODS: We identified articles published until 30 August 2021 that reported the prevalence of mental health problems in the general population at two or more time points. A crowd of 114 reviewers extracted data on prevalence, study and participant characteristics. We collected information on the number of days since the first SARS-CoV-2 infection in the study country, the stringency of containment measures and the number of cases and deaths. We synthesised changes in prevalence during the pandemic using a random-effects model. We used dose-response meta-analysis to evaluate the trajectory of the changes in mental health problems. RESULTS: We included 41 studies for 7 mental health conditions. The average odds of symptoms increased during the pandemic (mean OR ranging from 1.23 to 2.08). Heterogeneity was very large and could not be explained by differences in participants or study characteristics. Average odds of psychological distress, depression and anxiety increased during the first 2 months of the pandemic, with increased stringency of the measures, reported infections and deaths. The confidence in the evidence was low to very low. CONCLUSIONS: We observed an initial increase in the average risk of psychological distress, depression-related and anxiety-related problems during the first 2 months of the pandemic. However, large heterogeneity suggests that different populations had different responses to the challenges imposed by the pandemic.

Severely compromised supply of patch test allergens in Europe hampers adequate diagnosis of occupational and non-occupational contact allergy. A European Society of Contact Dermatitis (ESCD), European Academy of Allergy and Clinical Immunology (EAACI), European Academy of Dermatology and Venereology (EADV) task forces 'Contact Dermatitis' and 'Occupational Skin Disease' position paper.

Patch testing is the only clinically applicable diagnostic method for Type IV allergy. The availability of Type IV patch test (PT) allergens in Europe, however, is currently scarce. This severely compromises adequate diagnostics of contact allergy, leading to serious consequences for the affected patients. Against this background, the European Society of Contact Dermatitis (ESCD) has created a task force (TF) (i) to explore the current availability of PT substances in different member states, (ii) to highlight some of the unique characteristics of Type IV vs. other allergens and (iii) to suggest ways forward to promote and ensure availability of high-quality patch testing substances for the diagnosis of Type IV allergies throughout Europe. The suggestions of the TF on how to improve the availability of PT allergens are supported by the ESCD, the European Academy of Allergy and Clinical Immunology, and the European Academy of Dermatology and Venereology and intend to provide potential means to resolve the present medical crisis.

Comparison of Post-Operative Outcomes of Right Colectomy between Crohn's Disease and Adenocarcinoma of the Right Colon: A Retrospective Cohort Study.

(1) Background: Crohn's disease (CD) and right-sided colorectal-carcinoma (CRC) are two common indications for right colectomies. Many studies have tried to identify risk factors associated with post-operative complications for both CD and CRC. However, data directly comparing the outcomes of the two are sparse. This study aims to compare the 30-day post-operative outcome after a right colectomy for CD versus CRC. Its secondary aim is to identify the factors associated with these outcomes for each group. (2) Methods: A retrospective cohort study of 123 patients who underwent a right colectomy for CD or CRC in a single institution between March 2011 and March 2016. (3) Results: There were no significant differences between the groups when comparing the overall complication rate, the median Clavien-Dindo score, reoperation rates and the length of hospitalization. The leak rate was higher in the Crohn's group (13.95% (6/43) vs. 3.75% (3/80)), p = 0.049), although the stoma rate was the same (4/43 9.5%; 7/80 9.9%). (4) Conclusions: This study has shown that post-operative complication rate is similar for right colectomy in CD and CRC. However, Crohn's disease patients did have a higher leak rate.

Insights into treatment of complex Crohn's perianal fistulas.

Complex perianal fistula is a common complication of Crohn's disease (CD) which leads to negative impact on patient's quality of life. Successful management of the disease requires a multidisciplinary approach, including a gastroenterologist and a colorectal surgeon, applying combined surgical and medical therapy. One of frequently practiced surgical procedures is seton placement in the fistula tract, which is used to control perianal sepsis and drain the fistula, while preventing recurrent abscess formation.Darvadstrocel, a suspension of expanded, allogeneic, adipose-derived, mesenchymal stem cells, is safe and effective for treatment-refractory complex perianal fistulas in patients with Crohn's disease. Following approval of darvadstrocel, the INSPIRE registry is being conducted in order to evaluate long-term safety and effectiveness of the drug on a large, heterogenous population.An online expert meeting was held from March 20 to March 30, 2023, which provided relevant insights into the decision-making process regarding seton use and obtained feedback on the first experiences with darvadstrocel. The aim of this article is to present the perspectives from gastroenterologists and colorectal surgeons practicing in Czechia, Hungary, Israel, Lithuania, Serbia, and Slovenia in topics such as diagnosis and treatment options for patients with complex Crohn's perianal fistulas (CPF), specifically focusing on the use of setons and darvadstrocel.During this virtual session, unavailability of comprehensive data on safety and efficacy of available treatment procedures was emphasized as an important obstacle towards development of standardized recommendations and improvement of outcomes in treatment of (CPF). Furthermore, achieving consensus in seton use, duration of its placement, and frequency of change is recognized as one of CPF treatments major challenges. Despite these issues, it is important to promote better understanding and treatment of complex perianal fistulas in order to improve the quality of life of those affected by this condition.

Cardiac glycosides inhibit early and late vaccinia virus protein expression.

Cardiac glycosides (CGs) are natural steroid glycosides, which act as inhibitors of the cellular sodium-potassium ATPase pump. Although traditionally considered toxic to human cells, CGs are widely used as drugs for the treatment of cardiovascular-related medical conditions. More recently, CGs have been explored as potential anti-viral drugs and inhibit replication of a range of RNA and DNA viruses. Previously, a compound screen identified CGs that inhibited vaccinia virus (VACV) infection. However, no further investigation of the inhibitory potential of these compounds was performed, nor was there investigation of the stage(s) of the poxvirus lifecycle they impacted. Here, we investigated the anti-poxvirus activity of a broad panel of CGs. We found that all CGs tested were potent inhibitors of VACV replication. Our virological experiments showed that CGs did not impact virus infectivity, binding, or entry. Rather, experiments using recombinant viruses expressing reporter proteins controlled by VACV promoters and arabinoside release assays demonstrated that CGs inhibited early and late VACV protein expression at different concentrations. Lack of virus assembly in the presence of CGs was confirmed using electron microscopy. Thus, we expand our understanding of compounds with anti-poxvirus activity and highlight a yet unrecognized mechanism by which poxvirus replication can be inhibited.

Handling missing data when estimating causal effects with targeted maximum likelihood estimation.

Targeted maximum likelihood estimation (TMLE) is increasingly used for doubly robust causal inference, but how missing data should be handled when using TMLE with data-adaptive approaches is unclear. Based on data (1992-1998) from the Victorian Adolescent Health Cohort Study, we conducted a simulation study to evaluate 8 missing-data methods in this context: complete-case analysis, extended TMLE incorporating an outcome-missingness model, the missing covariate missing indicator method, and 5 multiple imputation (MI) approaches using parametric or machine-learning models. We considered 6 scenarios that varied in terms of exposure/outcome generation models (presence of confounder-confounder interactions) and missingness mechanisms (whether outcome influenced missingness in other variables and presence of interaction/nonlinear terms in missingness models). Complete-case analysis and extended TMLE had small biases when outcome did not influence missingness in other variables. Parametric MI without interactions had large bias when exposure/outcome generation models included interactions. Parametric MI including interactions performed best in bias and variance reduction across all settings, except when missingness models included a nonlinear term. When choosing a method for handling missing data in the context of TMLE, researchers must consider the missingness mechanism and, for MI, compatibility with the analysis method. In many settings, a parametric MI approach that incorporates interactions and nonlinearities is expected to perform well.

Increasing non-cosmetic exposure and sensitization to isothiazolinones require action for prevention: Review.

Frequent use of methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) and MI in cosmetic products has been the main cause of widespread sensitization and allergic contact dermatitis to these preservatives (biocides). Their use in non-cosmetic products is also an important source of sensitization. Less is known about sensitization rates and use of benzisothiazolinone (BIT), octylisothiazolinone (OIT), and dichlorooctylisothiazolinone (DCOIT), which have never been permitted in cosmetic products in Europe. BIT and OIT have occasionally been routinely patch-tested. These preservatives are often used together in chemical products and articles. In this study, we review the occurrence of contact allergy to MI, BIT, OIT, and DCOIT over time, based on concomitant patch testing in large studies, and case reports. We review EU legislations, and we discuss the role of industry, regulators, and dermatology in prevention of sensitization and protection of health. The frequency of contact allergy to MI, BIT, and OIT has increased. The frequency of contact allergy to DCOIT is not known because it has seldom been patch-tested. Label information on isothiazolinones in chemical products and articles, irrespective of concentration, is required for assessment of relevance, information to patients, and avoidance of exposure and allergic contact dermatitis.

Methodological review of NMA bias concepts provides groundwork for the development of a list of concepts for potential inclusion in a new risk of bias tool for network meta-analysis (RoB NMA Tool).

INTRODUCTION: Network meta-analyses (NMAs) have gained popularity and grown in number due to their ability to provide estimates of the comparative effectiveness of multiple treatments for the same condition. The aim of this study is to conduct a methodological review to compile a preliminary list of concepts related to bias in NMAs. METHODS AND ANALYSIS: We included papers that present items related to bias, reporting or methodological quality, papers assessing the quality of NMAs, or method papers. We searched MEDLINE, the Cochrane Library and unpublished literature (up to July 2020). We extracted items related to bias in NMAs. An item was excluded if it related to general systematic review quality or bias and was included in currently available tools such as ROBIS or AMSTAR 2. We reworded items, typically structured as questions, into concepts (i.e. general notions). RESULTS: One hundred eighty-one articles were assessed in full text and 58 were included. Of these articles, 12 were tools, checklists or journal standards; 13 were guidance documents for NMAs; 27 were studies related to bias or NMA methods; and 6 were papers assessing the quality of NMAs. These studies yielded 99 items of which the majority related to general systematic review quality and biases and were therefore excluded. The 22 items we included were reworded into concepts specific to bias in NMAs. CONCLUSIONS: A list of 22 concepts was included. This list is not intended to be used to assess biases in NMAs, but to inform the development of items to be included in our tool.

HIGHLIGHTS: • Our research aimed to develop a preliminary list of concepts related to bias with the goal of developing the first tool for assessing the risk of bias in the results and conclusions of a network meta-analysis (NMA).• We followed the methodology proposed by Whiting (2017) and Sanderson (2007) for creating systematically developed lists of quality items, as a first step in the development of a risk of bias tool for network meta-analysis (RoB NMA Tool).• We included items related to biases in NMAs and excluded items that are equally applicable to all systematic reviews as they are covered by other tools (e.g. ROBIS, AMSTAR 2).• Fifty-seven studies were included generating 99 items, which when screened, yielded 22 included items. These items were then reworded into concepts in preparation for a Delphi process for further vetting by external experts.• A limitation of our study is the challenge in retrieving methods studies as methods collections are not regularly updated.

A comparison of strategies for selecting auxiliary variables for multiple imputation.

Multiple imputation (MI) is a popular method for handling missing data. Auxiliary variables can be added to the imputation model(s) to improve MI estimates. However, the choice of which auxiliary variables to include is not always straightforward. Several data-driven auxiliary variable selection strategies have been proposed, but there has been limited evaluation of their performance. Using a simulation study we evaluated the performance of eight auxiliary variable selection strategies: (1, 2) two versions of selection based on correlations in the observed data; (3) selection using hypothesis tests of the "missing completely at random" assumption; (4) replacing auxiliary variables with their principal components; (5, 6) forward and forward stepwise selection; (7) forward selection based on the estimated fraction of missing information; and (8) selection via the least absolute shrinkage and selection operator (LASSO). A complete case analysis and an MI analysis using all auxiliary variables (the "full model") were included for comparison. We also applied all strategies to a motivating case study. The full model outperformed all auxiliary variable selection strategies in the simulation study, with the LASSO strategy the best performing auxiliary variable selection strategy overall. All MI analysis strategies that we were able to apply to the case study led to similar estimates, although computational time was substantially reduced when variable selection was employed. This study provides further support for adopting an inclusive auxiliary variable strategy where possible. Auxiliary variable selection using the LASSO may be a promising alternative when the full model fails or is too burdensome.

Phases of methodological research in biostatistics-Building the evidence base for new methods.

Although new biostatistical methods are published at a very high rate, many of these developments are not trustworthy enough to be adopted by the scientific community. We propose a framework to think about how a piece of methodological work contributes to the evidence base for a method. Similar to the well-known phases of clinical research in drug development, we propose to define four phases of methodological research. These four phases cover (I) proposing a new methodological idea while providing, for example, logical reasoning or proofs, (II) providing empirical evidence, first in a narrow target setting, then (III) in an extended range of settings and for various outcomes, accompanied by appropriate application examples, and (IV) investigations that establish a method as sufficiently well-understood to know when it is preferred over others and when it is not; that is, its pitfalls. We suggest basic definitions of the four phases to provoke thought and discussion rather than devising an unambiguous classification of studies into phases. Too many methodological developments finish before phase III/IV, but we give two examples with references. Our concept rebalances the emphasis to studies in phases III and IV, that is, carefully planned method comparison studies and studies that explore the empirical properties of existing methods in a wider range of problems.

When should factorial designs be used for late-phase randomised controlled trials?

BACKGROUND: A 2×2 factorial design evaluates two interventions (A versus control and B versus control) by randomising to control, A-only, B-only or both A and B together. Extended factorial designs are also possible (e.g. 3×3 or 2×2×2). Factorial designs often require fewer resources and participants than alternative randomised controlled trials, but they are not widely used. We identified several issues that investigators considering this design need to address, before they use it in a late-phase setting. METHODS: We surveyed journal articles published in 2000-2022 relating to designing factorial randomised controlled trials. We identified issues to consider based on these and our personal experiences. RESULTS: We identified clinical, practical, statistical and external issues that make factorial randomised controlled trials more desirable. Clinical issues are (1) interventions can be easily co-administered; (2) risk of safety issues from co-administration above individual risks of the separate interventions is low; (3) safety or efficacy data are wanted on the combination intervention; (4) potential for interaction (e.g. effect of A differing when B administered) is low; (5) it is important to compare interventions with other interventions balanced, rather than allowing randomised interventions to affect the choice of other interventions; (6) eligibility criteria for different interventions are similar. Practical issues are (7) recruitment is not harmed by testing many interventions; (8) each intervention and associated toxicities is unlikely to reduce either adherence to the other intervention or overall follow-up; (9) blinding is easy to implement or not required. Statistical issues are (10) a suitable scale of analysis can be identified; (11) adjustment for multiplicity is not required; (12) early stopping for efficacy or lack of benefit can be done effectively. External issues are (13) adequate funding is available and (14) the trial is not intended for licensing purposes. An overarching issue (15) is that factorial design should give a lower sample size requirement than alternative designs. Across designs with varying non-adherence, retention, intervention effects and interaction effects, 2×2 factorial designs require lower sample size than a three-arm alternative when one intervention effect is reduced by no more than 24%-48% in the presence of the other intervention compared with in the absence of the other intervention. CONCLUSIONS: Factorial designs are not widely used and should be considered more often using our issues to consider. Low potential for at most small to modest interaction is key, for example, where the interventions have different mechanisms of action or target different aspects of the disease being studied.