Search

BackgroundEnsuring rapid access to high quality sexual health services is a key public health objective, both in the United Kingdom and internationally. Internet-based testing services for sexually transmitted infections (STIs) are considered to be a promising way to achieve this goal. This study will evaluate a nascent online STI testing and results service in South East London, delivered alongside standard face-to-face STI testing services. ObjectiveThe aim of this study is to establish whether an online testing and results services can (1) increase diagnoses of STIs and (2) increase uptake of STI testing, when delivered alongside standard face-to-face STI testing services. MethodsThis is a single-blind randomized controlled trial. We will recruit 3000 participants who meet the following eligibility criteria: 16-30 years of age, resident in the London boroughs of Lambeth and Southwark, having at least one sexual partner in the last 12 months, having access to the Internet and willing to take an STI test. People unable to provide informed consent and unable to read and understand English (the websites will be in English) will be excluded. Baseline data will be collected at enrolment. This includes participant contact details, demographic data (date of birth, gender, ethnicity, and sexual orientation), and sexual health behaviors (last STI test, service used at last STI test and number of sexual partners in the last 12 months). Once enrolled, participants will be randomly allocated either (1) to an online STI testing and results service (Sexual Health 24) offering postal self-administered STI kits for chlamydia, gonorrhoea, syphilis, and HIV; results via text message (short message service, SMS), except positive results for HIV, which will be delivered by phone; and direct referrals to local clinics for treatment or (2) to a conventional sexual health information website with signposting to local clinic-based sexual health services. Participants will be free to use any other interventions or services during the trial period. At 6 weeks from randomization we will collect self-reported follow-up data on service use, STI tests and results, treatment prescribed, and acceptability of STI testing services. We will also collect objective data from participating STI testing services on uptake of STI testing, STI diagnoses and treatment. We hypothesise that uptake of STI testing and STI diagnoses will be higher in the intervention arm. Our hypothesis is based on the assumption that the intervention is less time-consuming, more convenient, more private, and incur less stigma and embarrassment than face-to-face STI testing pathways. The primary outcome measure is diagnosis of any STI at 6 weeks from randomization and our co-primary outcome is completion of any STI test at 6 weeks from randomization. We define completion of a test, as samples returned, processed, and results delivered to the intervention and/or clinic settings. We will use risk ratios to calculate the effect of the intervention on our primary outcomes with 95% confidence intervals. All analyses will be based on the intention-to-treat (ITT) principle. ResultsThis study is funded by Guy’s and St Thomas’ Charity and it has received ethical approval from NRES Committee London-Camberwell St Giles (Ref 14/LO/1477). Research and Development approval has been obtained from Kings College Hospital NHS Foundation Trust and Guy’s and St Thomas’ NHS Foundation Trust. Results are expected in June 2016. ConclusionsThis study will provide evidence on the effectiveness of an online STI testing and results service in South East London. Our findings may also be generalizable to similar populations in the United Kingdom. Trial RegistrationInternational Standard Randomized Controlled Trial Number (ISRCTN): 13354298; http://www.isrctn.com/ISRCTN13354298 (Archived by WebCite at http://www.webcitation.org/6d9xT2bPj)

Estimates from genome-wide association studies (GWAS) of unrelated individuals capture effects of inherited variation (direct effects), demography (population stratification, assortative mating) and relatives (indirect genetic effects). Family-based GWAS designs can control for demographic and indirect genetic effects, but large-scale family datasets have been lacking. We combined data from 178,086 siblings from 19 cohorts to generate population (between-family) and within-sibship (within-family) GWAS estimates for 25 phenotypes. Within-sibship GWAS estimates were smaller than population estimates for height, educational attainment, age at first birth, number of children, cognitive ability, depressive symptoms and smoking. Some differences were observed in downstream SNP heritability, genetic correlations and Mendelian randomization analyses. For example, the within-sibship genetic correlation between educational attainment and body mass index attenuated towards zero. In contrast, analyses of most molecular phenotypes (for example, low-density lipoprotein-cholesterol) were generally consistent. We also found within-sibship evidence of polygenic adaptation on taller height. Here, we illustrate the importance of family-based GWAS data for phenotypes influenced by demographic and indirect genetic effects.

Individual participant data meta-analysis (IPDMA) projects obtain, check, harmonise and synthesise raw data from multiple studies. When undertaking the meta-analysis, researchers must decide between a two-stage or a one-stage approach. In a two-stage approach, the IPD are first analysed separately within each study to obtain aggregate data (e.g., treatment effect estimates and standard errors); then, in the second stage, these aggregate data are combined in a standard meta-analysis model (e.g., common-effect or random-effects). In a one-stage approach, the IPD from all studies are analysed in a single step using an appropriate model that accounts for clustering of participants within studies and, potentially, between-study heterogeneity (e.g., a general or generalised linear mixed model). The best approach to take is debated in the literature, and so here we provide clearer guidance for a broad audience. Both approaches are important tools for IPDMA researchers and neither are a panacea. If most studies in the IPDMA are small (few participants or events), a one-stage approach is recommended due to using a more exact likelihood. However, in other situations, researchers can choose either approach, carefully following best practice. Some previous claims recommending to always use a one-stage approach are misleading, and the two-stage approach will often suffice for most researchers. When differences do arise between the two approaches, often it is caused by researchers using different modelling assumptions or estimation methods, rather than using one or two stages per se.

BackgroundUC San Diego Health System (UCSDHS) is an academic medical center and integrated care network in the US-Mexico border area of California contiguous to the Mexican Northern Baja region. The COVID-19 pandemic deeply influenced UCSDHS activities as new public health challenges increasingly related to high population density, cross-border traffic, economic disparities, and interconnectedness between cross-border communities, which accelerated development of clinical collaborations between UCSDHS and several border community hospitals - one in the US, two in Mexico - as high volumes of severely ill patients overwhelmed hospitals.ObjectiveWe describe the development, implementation, feasibility, and acceptance of a novel critical care support program in three community hospitals along the US-Mexico border.MethodsWe created and instituted a hybrid critical care program involving: 1) in-person activities to perform needs assessments of equipment and supplies and hands-on training and education, and 2) creation of a telemedicine-based (Tele-ICU) service for direct patient management and/or consultative, education-based experiences. We collected performance metrics surrounding adherence to evidence-based practices and staff perceptions of critical care delivery.FindingsIn-person intervention phase identified and filled gaps in equipment and supplies, and Tele-ICU program promoted adherence to evidence-based practices and improved staff confidence in caring for critically ill COVID-19 patients at each hospital.ConclusionA collaborative, hybrid critical care program across academic and community centers is feasible and effective to address cross-cultural public health emergencies.

ObjectivesTo examine reactions to the proposed improvements to standard Kaplan-Meier plots, the standard way to present time-to-event data, and to understand which (if any) facilitated better depiction of (1) the state of patients over time, and (2) uncertainty over time in the estimates of survival.DesignA survey of stakeholders' opinions on the proposals.SettingA web-based survey, open to international participation, for those with an interest in visualisation of time-to-event data.Participants1174 people participated in the survey over a 6-week period. Participation was global (although primarily Europe and North America) and represented a wide range of researchers (primarily statisticians and clinicians).Main outcome measuresTwo outcome measures were of principal importance: (1) participants' opinions of each proposal compared with a 'standard' Kaplan-Meier plot; and (2) participants' overall ranking of the proposals (including the standard).ResultsMost proposals were more popular than the standard Kaplan-Meier plot. The most popular proposals in the two categories, respectively, were an extended table beneath the plot depicting the numbers at risk, censored and having experienced an event at periodic timepoints, and CIs around each Kaplan-Meier curve.ConclusionsThis study produced a high response number, reflecting the importance of graphics for time-to-event data. Those producing and publishing Kaplan-Meier plots-both authors and journals-should, as a starting point, consider using the combination of the two favoured proposals.

Accurate frequentist performance of a method is desirable in confirmatory clinical trials, but is not sufficient on its own to justify the use of a missing data method. Reference‐based conditional mean imputation, with variance estimation justified solely by its frequentist performance, has the surprising and undesirable property that the estimated variance becomes smaller the greater the number of missing observations; as explained under jump‐to‐reference it effectively forces the true treatment effect to be exactly zero for patients with missing data. [ABSTRACT FROM AUTHOR]

ObjectiveTo assess the adequacy of reporting of non-inferiority trials alongside the consistency and utility of current recommended analyses and guidelines.DesignReview of randomised clinical trials that used a non-inferiority design published between January 2010 and May 2015 in medical journals that had an impact factor >10 (JAMA Internal Medicine, Archives Internal Medicine, PLOS Medicine, Annals of Internal Medicine, BMJ, JAMA, Lancet and New England Journal of Medicine).Data sourcesOvid (MEDLINE).MethodsWe searched for non-inferiority trials and assessed the following: choice of non-inferiority margin and justification of margin; power and significance level for sample size; patient population used and how this was defined; any missing data methods used and assumptions declared and any sensitivity analyses used.ResultsA total of 168 trial publications were included. Most trials concluded non-inferiority (132; 79%). The non-inferiority margin was reported for 98% (164), but less than half reported any justification for the margin (77; 46%). While most chose two different analyses (91; 54%) the most common being intention-to-treat (ITT) or modified ITT and per-protocol, a large number of articles only chose to conduct and report one analysis (65; 39%), most commonly the ITT analysis. There was lack of clarity or inconsistency between the type I error rate and corresponding CIs for 73 (43%) articles. Missing data were rarely considered with (99; 59%) not declaring whether imputation techniques were used.ConclusionsReporting and conduct of non-inferiority trials is inconsistent and does not follow the recommendations in available statistical guidelines, which are not wholly consistent themselves. Authors should clearly describe the methods used and provide clear descriptions of and justifications for their design and primary analysis. Failure to do this risks misleading conclusions being drawn, with consequent effects on clinical practice.

Value-added measures of educational progress have been used by education researchers and policy-makers to assess the performance of teachers and schools, contributing to performance-related pay and position in school league tables. They are designed to control for all underlying differences between pupils and should therefore provide unbiased measures of school and teacher influence on pupil progress, however, their effectiveness has been questioned. We exploit genetic data from a UK birth cohort to investigate how successfully value-added measures control for genetic differences between pupils. We use raw value-added, contextual value-added (which additionally controls for background characteristics) and teacher-reported value-added measures built from data at ages 11, 14 and 16. Sample sizes for analyses range from 4,600 to 6,518. Our findings demonstrate that genetic differences between pupils explain little variation in raw value-added measures but explain up to 20% of the variation in contextual value-added measures (95% CI = 6.06% to 35.71%). Value-added measures built from teacher-rated ability have a greater proportion of variance explained by genetic differences between pupils, with 36.3% of their cross-sectional variation being statistically accounted for by genetics (95% CI = 22.8% to 49.8%). By contrast, a far greater proportion of variance is explained by genetic differences for raw test scores at each age of at least 47.3% (95% CI: 35.9 to 58.7). These findings provide evidence that value-added measures of educational progress can be influenced by genetic differences between pupils, and therefore may provide a biased measure of school and teacher performance. We include a glossary of genetic terms for educational researchers interested in the use of genetic data in educational research.

To date, the OCAM2 system has demonstrated to be the fastest and lowest noise production ready wavefront sensor, achieving 2067 full frames per second with subelectron readout noise. This makes OCAM2 the ideal system for natural as well as continuous wave laser guide star wavefront sensing. In this paper we present the new gated version of OCAM2 named OCAM2-S, using E2V’s CCD219 sensor with integral shutter. This new camera offers the same superb characteristics than OCAM2 both in terms of speed and readout noise but also offers a shutter function that makes the sensor only sensitive to light for very short periods, at will. We will report on gating time and extinction ratio performances of this new camera. This device opens new possibilities for Rayleigh pulsed lasers adaptive optics systems. With a shutter time constant well below 1 microsecond, this camera opens new solutions for pulsed sodium lasers with backscatter suppression or even spot elongation minimization for ELT LGS. This camera has been successfully used for the first time on the CANARY instrument on the William Herschel Telescope. The final phase of CANARY is a demonstrator of a single MOAO channel for the proposed EAGLE instrument on the future European ELT. We will report on-sky performance with this instrument.

MOSAIC is the proposed multiple object spectrograph for the E-ELT that will eventually combine two AO observing modes within a single instrument. MOSAIC will contain up to 20 open-loop multiple object AO channels feeding NIR IFUs in addition to up to 200 seeing-limited (or GLAO corrected) VIS – NIR fibre pickoffs. Wavefront tomography will be implemented using a combination of LGS and a few high-order NGS distributed across the field with the wavefront correction applied in a split open/closed loop configuration. MOSAIC will be the only E-ELT instrument planned that can utilize the full 10 arcminute diameter field of view, enabling highly efficient observing modes for this workhorse instrument. Use of the full E-ELT field inevitably requires a closer integration between the telescope control system and the instrument AO systems, however this can bring several potential benefits to overall system performance. Here we present the initial design concept and baseline performance of the MOSAIC instrument and AO system(s) taking advantage of the CANARY on-sky results and inheriting from the previous Phase A study of EAGLE. Finally, we will highlight areas of system performance and calibration that will require further analysis and trade-off during the course of the upcoming Phase A study.

We present a concept for visible near-diffraction limited lucky imaging with full-sky laser assisted adaptive optics.Simulation results are given, along with first on-sky measurements taken with the CANARY AO system on theWilliam Herschel Telescope, and extension to ELT scales.We also present a study of readout noise in EMCCD and sCMOS detectors, with findings that are perhapssurprising when compared with naive models of sCMOS noise. We discuss the relative trade-offs of these detectorsfor ELT-scale AO and lucky imaging.Finally we introduce the concept of real-time simulation, necessary for integration and verification of ELT-scale AO systems, and provide details of the system developed for CANARY.

Current Laser Guide Star (LGS) Adaptive Optics (AO) systems disregard all tip-tilt wavefront information received from the LGS Wavefront Sensors (WFS), as it is considered irretrievably entangled with the up-link turbulence that the laser encounters as it passes through the atmosphere to form a guide star. Consequently, they must still observe a Natural Guide Star (NGS) in order to correct for tip-tilt aberrations.A method has recently been presented that use the tomographic capabilities of centre-launched, multi-LGS AO systems to predict the LGS uplink turbulence and hence allow correction with a reduced requirement on the NGS, or potentially no NGS requirement at all for some scientific applications. This method is summarised here, and its limitations discussed. Due to the increased separation of the laser beams at higher altitude, the method is more effective for correction of tip-tilt from high altitude turbulence, with performance approaching that of tomographic LGS AO with a tip-tilt NGS. The method is less successful for correcting tip-tilt contributions from low altitudes, though potential mitigation of this is considered.We finally discuss the methods potential for ELT scale operation. Due to the large aperture size, and large LGS separation, it is expected that the method would be more effective for larger telescopes.

Membership of overlapping or intersecting sets may be recorded in a bundle of (0, 1) indicator variables. Annotated Euler or Venn diagrams may be used to show graphically the frequencies of subsets so defined, but beyond just a few sets such diagrams can be hard to draw and use effectively. This column presents two new commands for graphical alternatives: upsetplot and vennbar. Each command produces a bar chart by default, but there is scope to recast to different graphical forms. The differences between the new commands reflect the divide in Stata between twoway commands and other graph commands. They also provide some flexibility in graph design to match tastes and circumstances. The discussion includes many historical details and references. [ABSTRACT FROM AUTHOR]

To obtain valid inference following stratified randomisation, treatment effects should be estimated with adjustment for stratification variables. Stratification sometimes requires categorisation of a continuous prognostic variable (eg, age), which raises the question: should adjustment be based on randomisation categories or underlying continuous values? In practice, adjustment for randomisation categories is more common. We reviewed trials published in general medical journals and found none of the 32 trials that stratified randomisation based on a continuous variable adjusted for continuous values in the primary analysis. Using data simulation, this article evaluates the performance of different adjustment strategies for continuous and binary outcomes where the covariate‐outcome relationship (via the link function) was either linear or non‐linear. Given the utility of covariate adjustment for addressing missing data, we also considered settings with complete or missing outcome data. Analysis methods included linear or logistic regression with no adjustment for the stratification variable, adjustment for randomisation categories, or adjustment for continuous values assuming a linear covariate‐outcome relationship or allowing for non‐linearity using fractional polynomials or restricted cubic splines. Unadjusted analysis performed poorly throughout. Adjustment approaches that misspecified the underlying covariate‐outcome relationship were less powerful and, alarmingly, biased in settings where the stratification variable predicted missing outcome data. Adjustment for randomisation categories tends to involve the highest degree of misspecification, and so should be avoided in practice. To guard against misspecification, we recommend use of flexible approaches such as fractional polynomials and restricted cubic splines when adjusting for continuous stratification variables in randomised trials. [ABSTRACT FROM AUTHOR]

Simulation studies are powerful tools in epidemiology and biostatistics, but they can be hard to conduct successfully. Sometimes unexpected results are obtained. We offer advice on how to check a simulation study when this occurs, and how to design and conduct the study to give results that are easier to check. Simulation studies should be designed to include some settings in which answers are already known. They should be coded in stages, with data-generating mechanisms checked before simulated data are analysed. Results should be explored carefully, with scatterplots of standard error estimates against point estimates surprisingly powerful tools. Failed estimation and outlying estimates should be identified and dealt with by changing data-generating mechanisms or coding realistic hybrid analysis procedures. Finally, we give a series of ideas that have been useful to us in the past for checking unexpected results. Following our advice may help to prevent errors and to improve the quality of published simulation studies. [ABSTRACT FROM AUTHOR]

For simulation studies that evaluate methods of handling missing data, we argue that generating partially observed data by fixing the complete data and repeatedly simulating the missingness indicators is a superficially attractive idea but only rarely appropriate to use. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Background: Patient and public involvement (PPI) in trials aims to enhance research by improving its relevance and transparency. Planning for statistical analysis begins at the design stage of a trial within the protocol and is refined and detailed in a Statistical Analysis Plan (SAP). While PPI is common in design and protocol development it is less common within SAPs. This study aimed to reach consensus on the most important and relevant statistical analysis items within an SAP to involve patients and the public. Methods: We developed a UK-based, two-round Delphi survey through an iterative consultation with public partners, statisticians, and trialists. The consultation process started with 55 items from international guidance for statistical analysis plans. We aimed to recruit at least 20 participants per key stakeholder group for inclusion in the final analysis of the Delphi survey. Participants were asked to vote on each item using a Likert scale from 1 to 9, where a rating of 1 to 3 was labelled as having 'limited importance'; 4 to 6 as 'important but not critical' and 7 to 9 as 'critical' to involve patients and the public. Results from the second round determined consensus on critical items for PPI. Results: The consultation exercise led to the inclusion of 15 statistical items in the Delphi survey. We recruited 179 participants, of whom 72% (129: 36 statisticians, 29 patients or public partners, 25 clinical researchers or methodologists, 27 trial managers, and 12 PPI coordinators) completed both rounds. Participants were on average 48 years old, 60% were female, 84% were White, 64% were based in England and 84% had at least five years' experience in trials. Four items reached consensus regarding critical importance for patient and public involvement: presentation of results to trial participants; summary and presentation of harms; interpretation and presentation of findings in an academic setting; factors impacting how well a treatment works. No consensus was reached for the remaining 11 items. In general, the results were consistent across stakeholder groups. Discussion: We identified four critical items to involve patients and the public in statistical analysis plans. The remaining 11 items did not reach consensus and need to be considered in a case-by-case basis with most responders considering patient and public involvement important (but not critical). Our research provides a platform to enable focused future efforts to improve patient and public involvement in trials and enhance the relevance of statistical analyses to patients and the public. [ABSTRACT FROM AUTHOR]

The marginality principle guides analysts to avoid omitting lower‐order terms from models in which higher‐order terms are included as covariates. Lower‐order terms are viewed as "marginal" to higher‐order terms. We consider how this principle applies to three cases: regression models that may include the ratio of two measured variables; polynomial transformations of a measured variable; and factorial arrangements of defined interventions. For each case, we show that which terms or transformations are considered to be lower‐order, and therefore marginal, depends on the scale of measurement, which is frequently arbitrary. Understanding the implications of this point leads to an intuitive understanding of the curse of dimensionality. We conclude that the marginality principle may be useful to analysts in some specific cases but caution against invoking it as a context‐free recipe. [ABSTRACT FROM AUTHOR]

Background: The population-level summary measure is a key component of the estimand for clinical trials with time-to-event outcomes. This is particularly the case for non-inferiority trials, because different summary measures imply different null hypotheses. Most trials are designed using the hazard ratio as summary measure, but recent studies suggested that the difference in restricted mean survival time might be more powerful, at least in certain situations. In a recent letter, we conjectured that differences between summary measures can be explained using the concept of the non-inferiority frontier and that for a fair simulation comparison of summary measures, the same analysis methods, making the same assumptions, should be used to estimate different summary measures. The aim of this article is to make such a comparison between three commonly used summary measures: hazard ratio, difference in restricted mean survival time and difference in survival at a fixed time point. In addition, we aim to investigate the impact of using an analysis method that assumes proportional hazards on the operating characteristics of a trial designed with any of the three summary measures. Methods: We conduct a simulation study in the proportional hazards setting. We estimate difference in restricted mean survival time and difference in survival non-parametrically, without assuming proportional hazards. We also estimate all three measures parametrically, using flexible survival regression, under the proportional hazards assumption. Results: Comparing the hazard ratio assuming proportional hazards with the other summary measures not assuming proportional hazards, relative performance varies substantially depending on the specific scenario. Fixing the summary measure, assuming proportional hazards always leads to substantial power gains compared to using non-parametric methods. Fixing the modelling approach to flexible parametric regression assuming proportional hazards, difference in restricted mean survival time is most often the most powerful summary measure among those considered. Conclusion: When the hazards are likely to be approximately proportional, reflecting this in the analysis can lead to large gains in power for difference in restricted mean survival time and difference in survival. The choice of summary measure for a non-inferiority trial with time-to-event outcomes should be made on clinical grounds; when any of the three summary measures discussed here is equally justifiable, difference in restricted mean survival time is most often associated with the most powerful test, on the condition that it is estimated under proportional hazards. [ABSTRACT FROM AUTHOR]

In network meta‐analysis, studies evaluating multiple treatment comparisons are modeled simultaneously, and estimation is informed by a combination of direct and indirect evidence. Network meta‐analysis relies on an assumption of consistency, meaning that direct and indirect evidence should agree for each treatment comparison. Here we propose new local and global tests for inconsistency and demonstrate their application to three example networks. Because inconsistency is a property of a loop of treatments in the network meta‐analysis, we locate the local test in a loop. We define a model with one inconsistency parameter that can be interpreted as loop inconsistency. The model builds on the existing ideas of node‐splitting and side‐splitting in network meta‐analysis. To provide a global test for inconsistency, we extend the model across multiple independent loops with one degree of freedom per loop. We develop a new algorithm for identifying independent loops within a network meta‐analysis. Our proposed models handle treatments symmetrically, locate inconsistency in loops rather than in nodes or treatment comparisons, and are invariant to choice of reference treatment, making the results less dependent on model parameterization. For testing global inconsistency in network meta‐analysis, our global model uses fewer degrees of freedom than the existing design‐by‐treatment interaction approach and has the potential to increase power. To illustrate our methods, we fit the models to three network meta‐analyses varying in size and complexity. Local and global tests for inconsistency are performed and we demonstrate that the global model is invariant to choice of independent loops. [ABSTRACT FROM AUTHOR]

Many trials use stratified randomisation, where participants are randomised within strata defined by one or more baseline covariates. While it is important to adjust for stratification variables in the analysis, the appropriate method of adjustment is unclear when stratification variables are affected by misclassification and hence some participants are randomised in the incorrect stratum. We conducted a simulation study to compare methods of adjusting for stratification variables affected by misclassification in the analysis of continuous outcomes when all or only some stratification errors are discovered, and when the treatment effect or treatment‐by‐covariate interaction effect is of interest. The data were analysed using linear regression with no adjustment, adjustment for the strata used to perform the randomisation (randomisation strata), adjustment for the strata if all errors are corrected (true strata), and adjustment for the strata after some errors are discovered and corrected (updated strata). The unadjusted model performed poorly in all settings. Adjusting for the true strata was optimal, while the relative performance of adjusting for the randomisation strata or the updated strata varied depending on the setting. As the true strata are unlikely to be known with certainty in practice, we recommend using the updated strata for adjustment and performing subgroup analyses, provided the discovery of errors is unlikely to depend on treatment group, as expected in blinded trials. Greater transparency is needed in the reporting of stratification errors and how they were addressed in the analysis. [ABSTRACT FROM AUTHOR]