Reporting on covariate adjustment in randomised controlled trials before and after revision of the 2001 CONSORT statement: a literature review

Analysing cluster randomised controlled trials using GLMM, GEE1, GEE2, and QIF: results from four case studies

BACKGROUND

Using four case studies, we aim to provide practical guidance and recommendations for the analysis of cluster randomised controlled trials.

METHODS

Four modelling approaches (Generalized Linear Mixed Models with parameters estimated by maximum likelihood/restricted maximum likelihood; Generalized Linear Models with parameters estimated by Generalized Estimating Equations (1st order or second order) and Quadratic Inference Function, for analysing correlated individual participant level outcomes in cluster randomised controlled trials were identified after we reviewed the literature. We systematically searched the online bibliography databases of MEDLINE, EMBASE, PsycINFO (via OVID), CINAHL (via EBSCO), and SCOPUS. We identified the above-mentioned four statistical analytical approaches and applied them to four case studies of cluster randomised controlled trials with the number of clusters ranging from 10 to 100, and individual participants ranging from 748 to 9,207. Results were obtained for both continuous and binary outcomes using R and SAS statistical packages.

RESULTS

The intracluster correlation coefficient (ICC) estimates for the case studies were less than 0.05 and are consistent with the observed ICC values commonly reported in primary care and community-based cluster randomised controlled trials. In most cases, the four methods produced similar results. However, in a few analyses, quadratic inference function produced different results compared to the generalized linear mixed model, first-order generalized estimating equations, and second-order generalized estimating equations, especially in trials with small to moderate numbers of clusters.

CONCLUSION

This paper demonstrates the analysis of cluster randomised controlled trials with four modelling approaches. The results obtained were similar in most cases, however, for trials with few clusters we do recommend that the quadratic inference function should be used with caution, and where possible a small sample correction should be used. The generalisability of our results is limited to studies with similar features to our case studies, for example, studies with a similar-sized ICC. It is important to conduct simulation studies to comprehensively evaluate the performance of the four modelling approaches.

Chronic mechanical trauma/irritation and oral carcinoma: A systematic review showing low evidence to support an association

BACKGROUND

Chronic trauma of oral mucosa, resulting from repeated and persistent mechanical irritative action of an intraoral injury agent, has repeatedly been reported to be possibly implicated in the development of oral squamous cell carcinoma (OSCC).

OBJECTIVES

The present systematic review aimed to assess whether chronic mechanical trauma can be considered a risk factor for OSCC.

DATA SOURCES

PubMed, CENTRAL (Cochrane Central Register of Controlled Trials), Scopus; EMBASE, Web of Science.

STUDY ELIGIBILITY CRITERIA

Cohort studies comparing OSCC incidence among subjects with/without chronic mechanical trauma or case-control or cross-sectional studies comparing chronic mechanical trauma among subjects with/without OSCC.

RESULTS

Only one prospective case-control study fulfilled the inclusion criteria, but the quality of the evidence provided is not enough to define trauma as a risk factor for OSCC. The main limitation is the presence of only one case-control study at high risk of bias. In the absence of strong evidence supporting the role of trauma in OSCC, a thorough discussion on trauma and carcinogenesis has been performed.

CONCLUSIONS

Available evidence does not support an active role for chronic trauma in oral carcinogenesis, neither as promoter nor as progressor factor. Prospective cohort studies able to better assess trauma in OSCC are needed.

Are we leaving money on the table in infertility RCTs? Trialists should statistically adjust for prespecified, prognostic covariates to increase power

Infertility randomized controlled trials (RCTs) are often too small to detect realistic treatment effects. Large observational studies have been proposed as a solution. However, this strategy threatens to weaken the evidence base further, because non-random assignment to treatments makes it impossible to distinguish effects of treatment from confounding factors. Alternative solutions are required. Power in an RCT can be increased by adjusting for prespecified, prognostic covariates when performing statistical analysis, and if stratified randomization or minimization has been used, it is essential to adjust in order to get the correct answer. We present data showing that this simple, free and frequently necessary strategy for increasing power is seldom employed, even in trials appearing in leading journals. We use this article to motivate a pedagogical discussion and provide a worked example. While covariate adjustment cannot solve the problem of underpowered trials outright, there is an imperative to use sound methodology to maximize the information each trial yields.

Reducing the time to successful intravenous cannulation in anaesthetised children with poor vein visibility using a near-infrared device: A randomised multicentre trial

BACKGROUND

During inhalational induction of anaesthesia for children, severe respiratory events can occur but can be rapidly treated once intravenous access is in place. Reducing the time to successful cannulation during inhalational induction for children with poor vein visibility would improve safety.

OBJECTIVE

To study the effectiveness of a near-infrared (NIR) vascular imaging device (Veinviewer) to facilitate intravenous cannulation.

DESIGN

A prospective, multicentre, randomised, open clinical trial.

SETTING

The operating rooms of three paediatric hospitals in Paris, France, from 1 October 2012 to 31 March 2016.

PATIENTS

Children up to the age of 7 years, with poor vein visibility requiring general anaesthesia.

INTERVENTION

Inhalational anaesthesia was initiated and intravenous cannulation was performed with the standard approach or with the Veinviewer Vision.

MAIN OUTCOME MEASURES

The primary outcome was the time to successful intravenous cannulation. A secondary outcome was the proportion of successful first attempts.

RESULTS

The mean time to successful intravenous cannulation was 200 (95% CI, 143 to 295) seconds in the Veinviewer and 252 (95% CI, 194 to 328) seconds for the control group: hazard ratio 1.28 (1.02 to 1.60) (P = 0.03). The adjusted hazard ratio for known predictive factors was 1.25 (0.99 to 1.56) (P = 0.06). Success at the first attempt was 64.6% (102/158) in the 'Veinviewer' group vs. 55.6% (85/153) in the 'control' group (P = 0.10).

CONCLUSION

The Veinviewer has limited value in reducing the time to successful intravenous cannulation during inhalational anaesthesia for young children with poor vein visibility. However, there is a strong trend to reducing the delay in some cases and, given its absence of side effects, it could be part of a rescue option for a difficult venous-access strategy.

CLINICAL TRIAL REGISTRATION

NCT01685866 (http://www.clinicaltrials.gov).

Principled selection of baseline covariates to account for censoring in randomized trials with a survival endpoint

The analysis of randomized trials with time-to-event endpoints is nearly always plagued by the problem of censoring. In practice, such analyses typically invoke the assumption of noninformative censoring. While this assumption usually becomes more plausible as more baseline covariates are being adjusted for, such adjustment also raises concerns. Prespecification of which covariates will be adjusted for (and how) is difficult, thus prompting the use of data-driven variable selection procedures, which may impede valid inferences to be drawn. The adjustment for covariates moreover adds concerns about model misspecification, and the fact that each change in adjustment set also changes the censoring assumption and the treatment effect estimand. In this article, we discuss these concerns and propose a simple variable selection strategy designed to produce a valid test of the null in large samples. The proposal can be implemented using off-the-shelf software for (penalized) Cox regression, and is empirically found to work well in simulation studies and real data analyses.

Trends in the crossover of patients in phase III oncology clinical trials in the USA

Background

The incorporation of crossover in randomised controlled trials is accepted as an ethical obligation, especially in cancer clinical trials. The more common type of crossover is crossover allowance, which allows patients assigned to one arm to switch to another arm if there is an established benefit in the crossover arm. In contrast, crossover-designed studies involve switching patients from all arms to a different arm as part of the study design. Crossover allowance may have advantages in patient recruitment and incorporating crossover after initial positive results fulfil ethical requirements. However, crossover can also contribute to confounding major endpoints of studies, such as overall survival or the second progression-free survival interval. For this reason, it is important to investigate and identify potential trends of crossover in clinical trials testing novel therapies.

Methods

Data about cancer clinical trials were extracted from clinicaltrials.gov. The search query was limited to completed phase III studies in adult populations. Location was limited to the USA. Date range extended from 1990 to 2019. Search query included the terms: cancer; completed- recruitment status; age: 18–65+ years; sex: all; location: USA; and study phase: phase 3. Studies were then excluded if they were not randomised controlled trials (RCTs) with the primary purpose of treatment and if they did not test cancer-related interventions.

Results

A total of 744 clinical trials were identified. There were 459 RCTs aimed at treatment, and of those, 35 utilised crossover. The start dates of these crossover trials ranged from 1997 to 2012. Thirty studies utilised crossover allowance. Prostate, breast and gastrointestinal stromal tumour cancers were the most represented cancer types in crossover studies. Among the 30 studies, the median proportion of patients who crossed over relative to the original arm assignment ranged from 2% to 88%, with a median of 57.5%.

Conclusions

The proportion of identified clinical trials with crossover compared to those without is extremely small. Crossover in clinical trials studying cancer treatment does not appear to be a widespread practice. Even though statistical approaches to mitigate confounding exist, crossover can still skew accurate reporting of the impact of experimental therapies on overall survival.

Nocebo and lessebo effects

The power of placebos is commonly associated with the placebo effect. In contrast, detrimental effects related to the use of a placebo are little studied and less well recognized. This chapter covers the nocebo and lessebo effects defined, respectively, as expectation of harm in the form of adverse events in a placebo arm and reduction of therapeutic benefit due to the uncertainty of being allocated to placebo. The lessebo effect is a more recent concept and has been described only in depression, schizophrenia and Parkinson's disease. The nocebo response was evaluated in many neurological diseases, including epilepsy, multiple sclerosis, Parkinson's disease, Alzheimer's disease, restless leg syndrome, among others. Meta-analyses of randomized controlled trials in these conditions reveal a significant variability of the magnitude of the nocebo response and that factors related to study design, study participants or neurological disease can be associated with a nocebo response, although with the opposing findings across conditions. The knowledge about neurobiological mechanisms of the nocebo effect is poor for neurological diseases, and most of the information has been generated in pain. Functional neuroimaging suggests the existence of a distinct network for the anticipation and the experience of a hyperalgesia nocebo response. Different types of neurotransmitters have been involved, including cholecystokinin, dopamine and opioids. Recognizing the potential impact of nocebo and lessebo effects, mitigating strategies are in development with application to clinical research and clinical practice, such as a contextualized informed consent process, alternative study designs and enhancement of patient-physician communication.

Primary outcome reporting in adolescent depression clinical trials needs standardization

BACKGROUND

Evidence-based health care is informed by results of randomized clinical trials (RCTs) and their syntheses in meta-analyses. When the trial outcomes measured are not clearly described in trial publications, knowledge synthesis, translation, and decision-making may be impeded. While heterogeneity in outcomes measured in adolescent major depressive disorder (MDD) RCTs has been described, the comprehensiveness of outcome reporting is unknown. This study aimed to assess the reporting of primary outcomes in RCTs evaluating treatments for adolescent MDD.

METHODS

RCTs evaluating treatment interventions in adolescents with a diagnosis of MDD published between 2008 and 2017 specifying a single primary outcome were eligible for outcome reporting assessment. Outcome reporting assessment was done independently in duplicate using a comprehensive checklist of 58 reporting items. Primary outcome information provided in each RCT publication was scored as "fully reported", "partially reported", or "not reported" for each checklist item, as applicable.

RESULTS

Eighteen of 42 identified articles were found to have a discernable single primary outcome and were included for outcome reporting assessment. Most trials (72%) did not fully report on over half of the 58 checklist items. Items describing masking of outcome assessors, timing and frequency of outcome assessment, and outcome analyses were fully reported in over 70% of trials. Items less frequently reported included outcome measurement instrument properties (ranging from 6 to 17%), justification of timing and frequency of outcome assessment (6%), and justification of criteria used for clinically significant differences (17%). The overall comprehensiveness of reporting appeared stable over time.

CONCLUSIONS

Heterogeneous reporting exists in published adolescent MDD RCTs, with frequent omissions of key details about their primary outcomes. These omissions may impair interpretability, replicability, and synthesis of RCTs that inform clinical guidelines and decision-making in this field. Consensus on the minimal criteria for outcome reporting in adolescent MDD RCTs is needed.

A Randomized Noninferiority Trial of Wearing Adjustable Glasses versus Standard and Ready-made Spectacles among Chinese Schoolchildren: Wearability and Evaluation of Adjustable Refraction III

PURPOSE

To compare wear of standard, adjustable, and ready-made glasses among children.

DESIGN

Randomized, controlled, open-label, noninferiority trial.

PARTICIPANTS

Students aged 11 to 16 years with presenting visual acuity (VA) ≤6/12 in both eyes, correctable to ≥6/7.5, subjective spherical equivalent refractive error (SER) ≤-1.0 diopters (D), astigmatism and anisometropia both <2.00 D, and no other ocular abnormalities.

METHODS

Participants were randomly allocated (1:1:1) to standard glasses, ready-made glasses, or adjustable glasses based on self-refraction. We recorded glasses wear on twice-weekly covert evaluation by head teachers (primary outcome), self-reported and investigator-observed wear, best-corrected visual acuity (BCVA) (not prespecified), children's satisfaction, and value attributed to glasses.

MAIN OUTCOME MEASURE

Proportion of glasses wear on twice-weekly covert evaluation by head teachers over 2 months.

RESULTS

Among 379 eligible participants, 127 were allocated to standard glasses (mean age, 13.7 years; standard deviation [SD], 1.0 years; 54.3% were male), 125 to ready-made (mean age, 13.6; SD, 0.83; 45.6%), and 127 to adjustable (mean age, 13.4 years; SD, 0.85; 54.3%). Mean wear proportion of adjustable glasses was significantly lower than for standard glasses (45% vs. 58%; P = 0.01), although the adjusted difference (90% confidence interval [CI], -19.0% to -3.0%) did not meet the prespecified inferiority threshold of 20%. Self-reported (90.2% vs. 84.8%, P = 0.64) and investigator-observed (44.1% vs. 33.9%, P = 0.89) wear did not differ between standard and adjustable glasses, nor did satisfaction with (P = 0.97) or value attributed to study glasses (P = 0.55) or increase in quality of life (5.53 [SD, 4.47] vs. 5.68 [SD, 4.34] on a 100-point scale, P > 0.30). Best-corrected visual acuity with adjustable glasses was better (P < 0.001) than with standard glasses. Change in power of study lenses at the end of the study (adjustable: 0.65 D, 95% CI, 0.52-0.79; standard, 0.01 D; 95% CI, -0.006 to 0.03, P < 0.001) was greater for adjustable glasses, although interobserver variation in power measurements may explain this. Lens scratches and frame damage were more common with adjustable glasses, whereas lens breakage was less common than for standard glasses.

CONCLUSIONS

Proportion of wear was lower with adjustable glasses, although VA was better and measures of satisfaction and quality of life were not inferior to standard glasses.

Scoping review on interventions to improve adherence to reporting guidelines in health research

OBJECTIVES

The goal of this study is to identify, analyse and classify interventions to improve adherence to reporting guidelines in order to obtain a wide picture of how the problem of enhancing the completeness of reporting of biomedical literature has been tackled so far.

DESIGN

Scoping review.

SEARCH STRATEGY

We searched the MEDLINE, EMBASE and Cochrane Library databases and conducted a grey literature search for (1) studies evaluating interventions to improve adherence to reporting guidelines in health research and (2) other types of references describing interventions that have been performed or suggested but never evaluated. The characteristics and effect of the evaluated interventions were analysed. Moreover, we explored the rationale of the interventions identified and determined the existing gaps in research on the evaluation of interventions to improve adherence to reporting guidelines.

RESULTS

109 references containing 31 interventions (11 evaluated) were included. These were grouped into five categories: (1) training on the use of reporting guidelines, (2) improving understanding, (3) encouraging adherence, (4) checking adherence and providing feedback, and (5) involvement of experts. Additionally, we identified lack of evaluated interventions (1) on training on the use of reporting guidelines and improving their understanding, (2) at early stages of research and (3) after the final acceptance of the manuscript.

CONCLUSIONS

This scoping review identified a wide range of strategies to improve adherence to reporting guidelines that can be taken by different stakeholders. Additional research is needed to assess the effectiveness of many of these interventions.

Intra-operative lidocaine in the prevention of vomiting after elective tonsillectomy in children: A randomised controlled trial

BACKGROUND

Postoperative vomiting (POV) is a frequent complication of tonsillectomy in children. In adult patients undergoing abdominal surgeries, the use of intravenous lidocaine infusion can prevent POV.

OBJECTIVE

To evaluate the anti-emetic effect of an intravenous lidocaine infusion used as an adjuvant to general anaesthesia, in children undergoing elective ear, nose and throat surgery.

DESIGN

Double-blind, randomised, controlled study.

SETTING

Hospital-based, single-centre study in Chile.

PATIENTS

ASA I-II children, aged 2 to 12 years, scheduled for elective tonsillectomy.

INTERVENTION

We standardised the induction and maintenance of anaesthesia. Patients were randomly allocated to lidocaine (1.5 mg kg intravenous lidocaine over 5 min followed by 2 mg kg h) or 0.9% saline (at the same rate and volume). Infusions were continued until the end of the surgery.

MAIN OUTCOME MEASURES

Presence of at least one episode of vomiting, retching or both in the first 24 h postoperatively (POV).

SECONDARY OUTCOMES

Plasma concentrations of lidocaine and postoperative pain.

RESULTS

Ninety-two children were enrolled. Primary outcome data were available for 91. In the Lidocaine group, 28 of 46 patients (60.8%) experienced POV, compared with 37 of 45 patients (82.2%) in the Saline group [difference in proportions 21.3% (95% confidence interval (CI) 2.8 to 38.8), P = 0.024]. The intention-to-treat analysis showed that when we assumed that the patient in the Saline group lost to follow-up did not have POV, the difference in proportions decreased to 19.6% (95% CI, 0.9 to 37.2), with an unadjusted odds ratio of 0.38 (95% CI, 0.15 to 0.97, P = 0.044). The odds of having POV were 62% less likely in those patients receiving lidocaine compared with patients in the Saline group. The mean lidocaine plasma concentration was 3.91 μg ml (range: 0.87 to 4.88).

CONCLUSION

Using an intravenous lidocaine infusion as an adjuvant to general anaesthesia decreased POV in children undergoing elective tonsillectomy.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT01986309.

TRAPEZE: a randomised controlled trial of the clinical effectiveness and cost-effectiveness of chemotherapy with zoledronic acid, strontium-89, or both, in men with bony metastatic castration-refractory prostate cancer

BACKGROUND

Bony metastatic castration-refractory prostate cancer is associated with a poor prognosis and high morbidity. TRAPEZE was a two-by-two factorial randomised controlled trial of zoledronic acid (ZA) and strontium-89 (Sr-89), each combined with docetaxel. All have palliative benefits, are used to control bone symptoms and are used with docetaxel to prolong survival. ZA, approved on the basis of reducing skeletal-related events (SREs), is commonly combined with docetaxel in practice, although evidence of efficacy and cost-effectiveness is lacking. Sr-89, approved for controlling metastatic pain and reducing need for subsequent bone treatments, is generally palliatively used in patients unfit for chemotherapy. Phase II analysis confirmed the safety and feasibility of combining these agents. TRAPEZE aimed to determine the clinical effectiveness and cost-effectiveness of each agent.

METHODS

Patients were randomised to receive six cycles of docetaxel plus prednisolone: alone, with ZA, with a single Sr-89 dose after cycle 6, or with both. Primary outcomes were clinical progression-free survival (CPFS: time to pain progression, SRE or death) and cost-effectiveness. Secondary outcomes were SRE-free interval (SREFI), total SREs, overall survival (OS) and quality of life (QoL). Log-rank test and Cox regression modelling were used to determine clinical effectiveness. Cost-effectiveness was assessed from the NHS perspective and expressed as cost per additional quality-adjusted life-year (QALY). An additional analysis was carried out for ZA to reflect the availability of generic ZA.

RESULTS

PATIENTS

757 randomised (median age 68.7 years; Eastern Cooperative Oncology Group scale score 0, 40%; 1, 52%; 2, 8%; prior radiotherapy, 45%); median prostate-specific antigen 143.78 ng/ml (interquartile range 50.8-353.9 ng/ml). Stratified log-rank analysis of CPFS was statistically non-significant for either agent (Sr-89, p = 0.11; ZA, p = 0.45). Cox regression analysis adjusted for stratification variables showed CPFS benefit for Sr-89 [hazard ratio (HR) 0.845, 95% confidence interval (CI) 0.72 to 0.99; p = 0.036] and confirmed no effect of ZA (p = 0.46). ZA showed a significant SREFI effect (HR 0.76; 95% CI 0.63 to 0.93; p = 0.008). Neither agent affected OS (Sr-89, p = 0.74; ZA, p = 0.91), but both increased total cost (vs. no ZA and no Sr-89, respectively); decreased post-trial therapies partly offset costs [net difference: Sr-89 £1341; proprietary ZA (Zometa(®), East Hanover, NJ, USA) £1319; generic ZA £251]. QoL was maintained in all trial arms; Sr-89 (0.08 additional QALYs) and ZA (0.03 additional QALYs) showed slight improvements. The resulting incremental cost-effectiveness ratio (ICER) for Sr-89 was £16,590, with £42,047 per QALY for Zometa and £8005 per QALY for generic ZA.

CONCLUSION

Strontium-89 improved CPFS, but not OS. ZA did not improve CPFS or OS but significantly improved SREFI, mostly post progression, suggesting a role as post-chemotherapy maintenance therapy. QoL was well maintained in all treatment arms, with differing patterns of care resulting from the effects of Sr-89 on time to progression and ZA on SREFI and total SREs. The addition of Sr-89 resulted in additional cost and a small positive increase in QALYs, with an ICER below the £20,000 ceiling per QALY. The additional costs and small positive QALY changes in favour of ZA resulted in ICERs of £42,047 (Zometa) and £8005 for the generic alternative; thus, generic ZA represents a cost-effective option. Additional analyses on the basis of data from the Hospital Episode Statistics data set would allow corroborating the findings of this study. Further research into the use of ZA (and other bone-targeting therapies) with newer prostate cancer therapies would be desirable.

STUDY REGISTRATION

Current Controlled Trials ISRCTN12808747.

FUNDING

This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 20, No. 53. See the NIHR Journals Library website for further project information.

Cognitive behavioral therapy positively affects fatigue in patients with multiple sclerosis: Results of a randomized controlled trial

Background:

Fatigue is a common symptom in multiple sclerosis (MS) and often restricts societal participation. Cognitive behavioral therapy (CBT) may alleviate MS-related fatigue, but evidence in literature is inconclusive.

Objective:

To evaluate the effectiveness of CBT to improve MS-related fatigue and participation.

Methods:

In a multi-center, assessor-masked, randomized controlled trial, participants with severe MS-related fatigue were assigned to CBT or control treatment. CBT consisted of 12 individual sessions with a psychologist trained in CBT, the control treatment consisted of three consultations with a MS nurse, both delivered over 16 weeks. Assessments were at baseline, 8, 16 (i.e. post-intervention), 26, and 52 weeks post-baseline. Primary outcomes were the Checklist Individual Strength-fatigue subscale (CIS20r fatigue) and the Impact on Participation and Autonomy questionnaire (IPA). Data were analyzed according to the intention-to-treat principle, using mixed-model analysis.

Results:

Between 2011 and 2014, 91 patients were randomized (CBT: n = 44; control: n = 47). Between-group analysis showed a positive post-intervention effect for CBT on CIS20r fatigue (T16: −6.7 (95% confidence interval (CI) = −10.7; −2.7) points) that diminished during follow-up (T52: 0.5 (95% CI = −3.6; 4.4)). No clinically relevant effects were found on societal participation.

Conclusion:

Severe MS-related fatigue can be reduced effectively with CBT in the short term. More research is needed on how to maintain this effect over the long term.

Confidence interval of difference of proportions in logistic regression in presence of covariates

Comparison of treatment differences in incidence rates is an important objective of many clinical trials. However, often the proportion is affected by covariates, and the adjustment of the predicted proportion is made using logistic regression. It is desirable to estimate the treatment differences in proportions adjusting for the covariates, similarly to the comparison of adjusted means in analysis of variance. Because of the correlation between the point estimates in the different treatment groups, the standard methods for constructing confidence intervals are inadequate. The problem is more difficult in the binary case, as the comparison is not uniquely defined, and the sampling distribution more difficult to analyze. Four procedures for analyzing the data are presented, which expand upon existing methods and generalize the link function. It is shown that, among the four methods studied, the resampling method based on the exact distribution function yields a coverage rate closest to the nominal.

Covariate adjustments in randomized controlled trials increased study power and reduced biasedness of effect size estimation

OBJECTIVES

This study aims to show that under several assumptions, in randomized controlled trials (RCTs), unadjusted, crude analysis will underestimate the Cohen's d effect size of the treatment, and an unbiased estimate of effect size can be obtained only by adjusting for all predictors of the outcome.

STUDY DESIGN AND SETTING

Four simulations were performed to examine the effects of adjustment on the estimated effect size of the treatment and power of the analysis. In addition, we analyzed data from the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study (older adults aged 65-94), an RCT with three treatment arms and one control arm.

RESULTS

We showed that (1) the number of unadjusted covariates was associated with the effect size of the treatment; (2) the biasedness of effect size estimation was minimized if all covariates were adjusted for; (3) the power of the statistical analysis slightly decreased with the number of adjusted noise variables; and (4) exhaustively searching the covariates and noise variables adjusted for can lead to exaggeration of the true effect size. Analysis of the ACTIVE study data showed that the effect sizes adjusting for covariates of all three treatments were 7.39-24.70% larger than their unadjusted counterparts, whereas the effect size would be elevated by at most 57.92% by exhaustively searching the variables adjusted for.

CONCLUSION

All covariates of the outcome in RCTs should be adjusted for, and if the effect of a particular variable on the outcome is unknown, adjustment will do more good than harm.

A review of the use of covariates in cluster randomized trials uncovers marked discrepancies between guidance and practice

OBJECTIVES

Reviews of the handling of covariates in trials have explicitly excluded cluster randomized trials (CRTs). In this study, we review the use of covariates in randomization, the reporting of covariates, and adjusted analyses in CRTs.

STUDY DESIGN AND SETTING

We reviewed a random sample of 300 CRTs published between 2000 and 2008 across 150 English language journals.

RESULTS

Fifty-eight percent of trials used covariates in randomization. Only 69 (23%) included tables of cluster- and individual-level covariates. Fifty-eight percent reported significance tests of baseline balance. Of 207 trials that reported baseline measures of the primary outcome, 155 (75%) subsequently adjusted for these in analyses. Of 174 trials that used covariates in randomization, 30 (17%) included an analysis adjusting for all those covariates. Of 219 trial reports that included an adjusted analysis of the primary outcome, only 71 (32%) reported that covariates were chosen a priori.

CONCLUSION

There are some marked discrepancies between practice and guidance on the use of covariates in the design, analysis, and reporting of CRTs. It is essential that researchers follow guidelines on the use and reporting of covariates in CRTs, promoting the validity of trial conclusions and quality of trial reports.

Analysis of time to event outcomes in randomized controlled trials by generalized additive models

BACKGROUND

Randomized Controlled Trials almost invariably utilize the hazard ratio calculated with a Cox proportional hazard model as a treatment efficacy measure. Despite the widespread adoption of HRs, these provide a limited understanding of the treatment effect and may even provide a biased estimate when the assumption of proportional hazards in the Cox model is not verified by the trial data. Additional treatment effect measures on the survival probability or the time scale may be used to supplement HRs but a framework for the simultaneous generation of these measures is lacking.

METHODS

By splitting follow-up time at the nodes of a Gauss Lobatto numerical quadrature rule, techniques for Poisson Generalized Additive Models (PGAM) can be adopted for flexible hazard modeling. Straightforward simulation post-estimation transforms PGAM estimates for the log hazard into estimates of the survival function. These in turn were used to calculate relative and absolute risks or even differences in restricted mean survival time between treatment arms. We illustrate our approach with extensive simulations and in two trials: IPASS (in which the proportionality of hazards was violated) and HEMO a long duration study conducted under evolving standards of care on a heterogeneous patient population.

FINDINGS

PGAM can generate estimates of the survival function and the hazard ratio that are essentially identical to those obtained by Kaplan Meier curve analysis and the Cox model. PGAMs can simultaneously provide multiple measures of treatment efficacy after a single data pass. Furthermore, supported unadjusted (overall treatment effect) but also subgroup and adjusted analyses, while incorporating multiple time scales and accounting for non-proportional hazards in survival data.

CONCLUSIONS

By augmenting the HR conventionally reported, PGAMs have the potential to support the inferential goals of multiple stakeholders involved in the evaluation and appraisal of clinical trial results under proportional and non-proportional hazards.

Yoga in Sedentary Adults with Arthritis: Effects of a Randomized Controlled Pragmatic Trial

OBJECTIVE

To evaluate the effect of Integral-based hatha yoga in sedentary people with arthritis.

METHODS

There were 75 sedentary adults aged 18+ years with rheumatoid arthritis (RA) or knee osteoarthritis randomly assigned to 8 weeks of yoga (two 60-min classes and 1 home practice/wk) or waitlist. Poses were modified for individual needs. The primary endpoint was physical health [Medical Outcomes Study Short Form-36 (SF-36) physical component summary (PCS)] adjusted for baseline; exploratory adjusted outcomes included fitness, mood, stress, self-efficacy, SF-36 health-related quality of life (HRQOL), and RA disease activity. In everyone completing yoga, we explored longterm effects at 9 months.

RESULTS

Participants were mostly female (96%), white (55%), and college-educated (51%), with a mean (SD) age of 52 years (12 yrs). Average disease duration was 9 years and 49% had RA. At 8 weeks, yoga was associated with significantly higher PCS (6.5, 95% CI 2.0-10.7), walking capacity (125 m, 95% CI 15-235), positive affect (5.2, 95% CI 1.4-8.9), and lower Center for Epidemiologic Studies Depression Scale (-3.0, 95% CI -4.8 - -1.3). Significant improvements (p < 0.05) were evident in SF-36 role physical, pain, general health, vitality, and mental health scales. Balance, grip strength, and flexibility were similar between groups. Twenty-two out of 28 in the waitlist group completed yoga. Among all yoga participants, significant (p < 0.05) improvements were observed in mean PCS, flexibility, 6-min walk, and all psychological and most HRQOL domains at 8 weeks with most still evident 9 months later. Of 7 adverse events, none were associated with yoga.

CONCLUSION

Preliminary evidence suggests yoga may help sedentary individuals with arthritis safely increase physical activity, and improve physical and psychological health and HRQOL. Clinical Trials NCT00349869.

Covariate adjustment had similar benefits in small and large randomized controlled trials

OBJECTIVES

Covariate adjustment is a standard statistical approach in the analysis of randomized controlled trials. We aimed to explore whether the benefit of covariate adjustment on statistical significance and power differed between small and large trials, where chance imbalance in prognostic factors necessarily differs.

STUDY DESIGN AND SETTING

We studied two large trial data sets [Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO-I), N = 30,510 and International Stroke Trial (IST), N = 18,372] repeatedly drawing random samples (500,000 times) of sizes 300 and 5,000 per arm and simulated each primary outcome using the control arms. We empirically determined the treatment effects required to fix power at 80% for all unadjusted analyses and calculated the joint probabilities in the discordant cells when cross-classifying adjusted and unadjusted results from logistic regression models (ie, P < 0.05 vs. P ≥ 0.05).

RESULTS

The power gained from an adjusted analysis for small and large samples was between 5% and 6%. Similar proportions of discordance were noted irrespective of the sample size in both the GUSTO-I and the IST data sets.

CONCLUSION

The proportions of change in statistical significance from covariate adjustment of strongly prognostic characteristics were the same for small and large trials with similar gains in statistical power. Covariate adjustment is equally recommendable in small and large trials.

Evidence for the selective reporting of analyses and discrepancies in clinical trials: a systematic review of cohort studies of clinical trials

BACKGROUND

Most publications about selective reporting in clinical trials have focussed on outcomes. However, selective reporting of analyses for a given outcome may also affect the validity of findings. If analyses are selected on the basis of the results, reporting bias may occur. The aims of this study were to review and summarise the evidence from empirical cohort studies that assessed discrepant or selective reporting of analyses in randomised controlled trials (RCTs).

METHODS AND FINDINGS

A systematic review was conducted and included cohort studies that assessed any aspect of the reporting of analyses of RCTs by comparing different trial documents, e.g., protocol compared to trial report, or different sections within a trial publication. The Cochrane Methodology Register, Medline (Ovid), PsycInfo (Ovid), and PubMed were searched on 5 February 2014. Two authors independently selected studies, performed data extraction, and assessed the methodological quality of the eligible studies. Twenty-two studies (containing 3,140 RCTs) published between 2000 and 2013 were included. Twenty-two studies reported on discrepancies between information given in different sources. Discrepancies were found in statistical analyses (eight studies), composite outcomes (one study), the handling of missing data (three studies), unadjusted versus adjusted analyses (three studies), handling of continuous data (three studies), and subgroup analyses (12 studies). Discrepancy rates varied, ranging from 7% (3/42) to 88% (7/8) in statistical analyses, 46% (36/79) to 82% (23/28) in adjusted versus unadjusted analyses, and 61% (11/18) to 100% (25/25) in subgroup analyses. This review is limited in that none of the included studies investigated the evidence for bias resulting from selective reporting of analyses. It was not possible to combine studies to provide overall summary estimates, and so the results of studies are discussed narratively.

CONCLUSIONS

Discrepancies in analyses between publications and other study documentation were common, but reasons for these discrepancies were not discussed in the trial reports. To ensure transparency, protocols and statistical analysis plans need to be published, and investigators should adhere to these or explain discrepancies.

The risks and rewards of covariate adjustment in randomized trials: an assessment of 12 outcomes from 8 studies

Background

Adjustment for prognostic covariates can lead to increased power in the analysis of randomized trials. However, adjusted analyses are not often performed in practice.

Methods

We used simulation to examine the impact of covariate adjustment on 12 outcomes from 8 studies across a range of therapeutic areas. We assessed (1) how large an increase in power can be expected in practice; and (2) the impact of adjustment for covariates that are not prognostic.

Results

Adjustment for known prognostic covariates led to large increases in power for most outcomes. When power was set to 80% based on an unadjusted analysis, covariate adjustment led to a median increase in power to 92.6% across the 12 outcomes (range 80.6 to 99.4%). Power was increased to over 85% for 8 of 12 outcomes, and to over 95% for 5 of 12 outcomes. Conversely, the largest decrease in power from adjustment for covariates that were not prognostic was from 80% to 78.5%.

Conclusions

Adjustment for known prognostic covariates can lead to substantial increases in power, and should be routinely incorporated into the analysis of randomized trials. The potential benefits of adjusting for a small number of possibly prognostic covariates in trials with moderate or large sample sizes far outweigh the risks of doing so, and so should also be considered.

Collaborative care for patients with depression and diabetes mellitus: a systematic review and meta-analysis

BACKGROUND

Diabetic patients with depression are often inadequately treated within primary care. These comorbid conditions are associated with poor outcomes. The aim of this systematic review was to examine whether collaborative care can improve depression and diabetes outcomes in patients with both depression and diabetes.

METHODS

Medline, Embase, Cochrane library and PsyINFO were systematically searched to identify relevant publications. All randomized controlled trials of collaborative care for diabetic patients with depression of all ages who were reported by depression treatment response, depression remission, hemoglobin A1c (HbA1c) values, adherence to antidepressant medication and/or oral hypoglycemic agent were included. Two authors independently screened search results and extracted data from eligible studies. Dichotomous and continuous measures of outcomes were combined using risk ratios (RRs) and mean differences (MDs) with 95% confidence intervals (CIs) either by fixed or random-effects models.

RESULTS

Eight studies containing 2,238 patients met the inclusion criteria. Collaborative care showed a significant improvement in depression treatment response (RR = 1.33, 95% CI = 1.05-1.68), depression remission (adjusted RR = 1.53, 95% CI =1.11-2.12), higher rates of adherence to antidepressant medication (RR = 1.79, 95% CI = 1.19-2.69) and oral hypoglycemic agent (RR = 2.18, 95% CI = 1.61-2.96), but indicated a non-significant reduction in HbA1c values (MD = -0.13, 95% CI = -0.46-0.19).

CONCLUSIONS

Improving depression care in diabetic patients is very necessary and important. Comparing with usual care, collaborative care was associated with significantly better depressive outcomes and adherence in patients with depression and diabetes. These findings emphasize the implications for collaborative care of diabetic patients with depression in the future.

Efficacy of composite versus ceramic inlays and onlays: study protocol for the CECOIA randomized controlled trial

Background

Dental caries is a common disease and affects many adults worldwide. Inlay or onlay restoration is widely used to treat the resulting tooth substance loss. Two esthetic materials can be used to manufacture an inlay/onlay restoration of the tooth: ceramic or composite. Here, we present the protocol of a multicenter randomized controlled trial (RCT) comparing the clinical efficacy of both materials for tooth restoration. Other objectives are analysis of overall quality, wear, restoration survival and prognosis.

Methods

The CEramic and COmposite Inlays Assessment (CECOIA) trial is an open-label, parallel-group, multicenter RCT involving two hospitals and five private practices. In all, 400 patients will be included. Inclusion criteria are adults who need an inlay/onlay restoration for one tooth (that can be isolated with use of a dental dam and has at least one intact cusp), can tolerate restorative procedures and do not have severe bruxism, periodontal or carious disease or poor oral hygiene. The decayed tissue will be evicted, the cavity will be prepared for receiving an inlay/onlay and the patient will be randomized by use of a centralized web-based interface to receive: 1) a ceramic or 2) composite inlay or onlay. Treatment allocation will be balanced (1:1). The inlay/onlay will be adhesively luted. Follow-up will be for 2 years and may be extended; two independent examiners will perform the evaluations. The primary outcome measure will be the score obtained with use of the consensus instrument of the Fédération Dentaire Internationale (FDI) World Dental Federation. Secondary outcomes include this instrument’s items, inlay/onlay wear, overall quality and survival of the inlay/onlay. Data will be analyzed by a statistician blinded to treatments and an adjusted ordinal logistic regression model will be used to compare the efficacy of both materials.

Discussion

For clinicians, the CECOIA trial results may help with evidence-based recommendations concerning the choice of materials for inlay/onlay restoration. For patients, the results may lead to improvement in long-term restoration. For researchers, the results may provide ideas for further research concerning inlay/onlay materials and prognosis.

This trial is funded by a grant from the French Ministry of Health.

Trial registration

ClinicalTrials.gov Identifier: NCT01724827

Adjusting for multiple prognostic factors in the analysis of randomised trials

Background

When multiple prognostic factors are adjusted for in the analysis of a randomised trial, it is unclear (1) whether it is necessary to account for each of the strata, formed by all combinations of the prognostic factors (stratified analysis), when randomisation has been balanced within each stratum (stratified randomisation), or whether adjusting for the main effects alone will suffice, and (2) the best method of adjustment in terms of type I error rate and power, irrespective of the randomisation method.

Methods

We used simulation to (1) determine if a stratified analysis is necessary after stratified randomisation, and (2) to compare different methods of adjustment in terms of power and type I error rate. We considered the following methods of analysis: adjusting for covariates in a regression model, adjusting for each stratum using either fixed or random effects, and Mantel-Haenszel or a stratified Cox model depending on outcome.

Results

Stratified analysis is required after stratified randomisation to maintain correct type I error rates when (a) there are strong interactions between prognostic factors, and (b) there are approximately equal number of patients in each stratum. However, simulations based on real trial data found that type I error rates were unaffected by the method of analysis (stratified vs unstratified), indicating these conditions were not met in real datasets. Comparison of different analysis methods found that with small sample sizes and a binary or time-to-event outcome, most analysis methods lead to either inflated type I error rates or a reduction in power; the lone exception was a stratified analysis using random effects for strata, which gave nominal type I error rates and adequate power.

Conclusions

It is unlikely that a stratified analysis is necessary after stratified randomisation except in extreme scenarios. Therefore, the method of analysis (accounting for the strata, or adjusting only for the covariates) will not generally need to depend on the method of randomisation used. Most methods of analysis work well with large sample sizes, however treating strata as random effects should be the analysis method of choice with binary or time-to-event outcomes and a small sample size.

Practices and impact of primary outcome adjustment in randomized controlled trials: meta-epidemiologic study

OBJECTIVE

To assess adjustment practices for primary outcomes of randomized controlled trials and their impact on the results.

DESIGN

Meta-epidemiologic study.

DATA SOURCES

25 biomedical journals with the highest impact factor according to Journal Citation Reports 2009.

STUDY SELECTION

Randomized controlled trials published in print in 2009 that reported primary outcomes. The search yielded 684 eligible papers of randomized controlled trials, of which 200 were randomly selected.

DATA EXTRACTION

Two researchers independently extracted data on study population, intervention, primary outcome, and the adjustment plan for primary outcomes. They also recorded the magnitude and statistical significance of the intervention effect with and without adjustments, and estimated whether adjustment made a difference in the level of nominal significance. They also compared the analysis plan for model adjustment in the published trial versus the trial protocol with information on the protocol collected from registries, design papers, and communication with all corresponding authors.

RESULTS

54% of the trials used stratified randomization, 96% presented baseline characteristics in the compared arms, and 46% also evaluated differences in baseline factors with statistical testing. Half of the trials performed adjusted analyses for the main outcome, as the sole analysis (29%) or along with unadjusted analyses (21%). Adjustment for stratification variables and for baseline variables was performed in 39% (42/108) and 42% (84/199) of the trials, respectively. Among 40 comparisons with both adjusted and unadjusted analyses, 43% had statistically significant effects, 40% had non-significant effects, and 18% had significant effects with only one of the two analyses, but not with the other. Information on analysis plan regarding model adjustment was available in 6% (9/162) of trial registry entries, 78% (21/27) of design papers, and 74% (40/54) of protocols obtained from authors. The analysis plan disagreed between the published trial and the registry, protocol, or design paper in 47% (28/60) of the studies.

CONCLUSIONS

There is large diversity on whether and how analyses of primary outcomes are adjusted in randomized controlled trials and these choices can sometimes change the nominal significance of the results. Registered protocols should explicitly specify adjustments plans for main outcomes and analysis should follow these plans.

SPIRIT 2013 explanation and elaboration: guidance for protocols of clinical trials

High quality protocols facilitate proper conduct, reporting, and external review of clinical trials. However, the completeness of trial protocols is often inadequate. To help improve the content and quality of protocols, an international group of stakeholders developed the SPIRIT 2013 Statement (Standard Protocol Items: Recommendations for Interventional Trials). The SPIRIT Statement provides guidance in the form of a checklist of recommended items to include in a clinical trial protocol. This SPIRIT 2013 Explanation and Elaboration paper provides important information to promote full understanding of the checklist recommendations. For each checklist item, we provide a rationale and detailed description; a model example from an actual protocol; and relevant references supporting its importance. We strongly recommend that this explanatory paper be used in conjunction with the SPIRIT Statement. A website of resources is also available (www.spirit-statement.org). The SPIRIT 2013 Explanation and Elaboration paper, together with the Statement, should help with the drafting of trial protocols. Complete documentation of key trial elements can facilitate transparency and protocol review for the benefit of all stakeholders.

Consolidated standards of reporting trials (CONSORT) and the completeness of reporting of randomised controlled trials (RCTs) published in medical journals

BACKGROUND

An overwhelming body of evidence stating that the completeness of reporting of randomised controlled trials (RCTs) is not optimal has accrued over time. In the mid-1990s, in response to these concerns, an international group of clinical trialists, statisticians, epidemiologists, and biomedical journal editors developed the CONsolidated Standards Of Reporting Trials (CONSORT) Statement. The CONSORT Statement, most recently updated in March 2010, is an evidence-based minimum set of recommendations including a checklist and flow diagram for reporting RCTs and is intended to facilitate the complete and transparent reporting of trials and aid their critical appraisal and interpretation. In 2006, a systematic review of eight studies evaluating the "effectiveness of CONSORT in improving reporting quality in journals" was published.

OBJECTIVES

To update the earlier systematic review assessing whether journal endorsement of the 1996 and 2001 CONSORT checklists influences the completeness of reporting of RCTs published in medical journals.

SEARCH METHODS

We conducted electronic searches, known item searching, and reference list scans to identify reports of evaluations assessing the completeness of reporting of RCTs. The electronic search strategy was developed in MEDLINE and tailored to EMBASE. We searched the Cochrane Methodology Register and the Cochrane Database of Systematic Reviews using the Wiley interface. We searched the Science Citation Index, Social Science Citation Index, and Arts and Humanities Citation Index through the ISI Web of Knowledge interface. We conducted all searches to identify reports published between January 2005 and March 2010, inclusive.

SELECTION CRITERIA

In addition to studies identified in the original systematic review on this topic, comparative studies evaluating the completeness of reporting of RCTs in any of the following comparison groups were eligible for inclusion in this review: 1) Completeness of reporting of RCTs published in journals that have and have not endorsed the CONSORT Statement; 2) Completeness of reporting of RCTs published in CONSORT-endorsing journals before and after endorsement; or 3) Completeness of reporting of RCTs before and after the publication of the CONSORT Statement (1996 or 2001). We used a broad definition of CONSORT endorsement that includes any of the following: (a) requirement or recommendation in journal's 'Instructions to Authors' to follow CONSORT guidelines; (b) journal editorial statement endorsing the CONSORT Statement; or (c) editorial requirement for authors to submit a CONSORT checklist and/or flow diagram with their manuscript. We contacted authors of evaluations reporting data that could be included in any comparison group(s), but not presented as such in the published report and asked them to provide additional data in order to determine eligibility of their evaluation. Evaluations were not excluded due to language of publication or validity assessment.

DATA COLLECTION AND ANALYSIS

We completed screening and data extraction using standardised electronic forms, where conflicts, reasons for exclusion, and level of agreement were all automatically and centrally managed in web-based management software, DistillerSR(®). One of two authors extracted general characteristics of included evaluations and all data were verified by a second author. Data describing completeness of reporting were extracted by one author using a pre-specified form; a 10% random sample of evaluations was verified by a second author. Any discrepancies were discussed by both authors; we made no modifications to the extracted data. Validity assessments of included evaluations were conducted by one author and independently verified by one of three authors. We resolved all conflicts by consensus.For each comparison we collected data on 27 outcomes: 22 items of the CONSORT 2001 checklist, plus four items relating to the reporting of blinding, and one item of aggregate CONSORT scores. Where reported, we extracted and qualitatively synthesised data on the methodological quality of RCTs, by scale or score.

MAIN RESULTS

Fifty-three publications reporting 50 evaluations were included. The total number of RCTs assessed within evaluations was 16,604 (median per evaluation 123 (interquartile range (IQR) 77 to 226) published in a median of six (IQR 3 to 26) journals. Characteristics of the included RCT populations were variable, resulting in heterogeneity between included evaluations. Validity assessments of included studies resulted in largely unclear judgements. The included evaluations are not RCTs and less than 8% (4/53) of the evaluations reported adjusting for potential confounding factors.   Twenty-five of 27 outcomes assessing completeness of reporting in RCTs appeared to favour CONSORT-endorsing journals over non-endorsers, of which five were statistically significant. 'Allocation concealment' resulted in the largest effect, with risk ratio (RR) 1.81 (99% confidence interval (CI) 1.25 to 2.61), suggesting that 81% more RCTs published in CONSORT-endorsing journals adequately describe allocation concealment compared to those published in non-endorsing journals. Allocation concealment was reported adequately in 45% (393/876) of RCTs in CONSORT-endorsing journals and in 22% (329/1520) of RCTs in non-endorsing journals. Other outcomes with results that were significant include: scientific rationale and background in the 'Introduction' (RR 1.07, 99% CI 1.01 to 1.14); 'sample size' (RR 1.61, 99% CI 1.13 to 2.29); method used for 'sequence generation' (RR 1.59, 99% CI 1.38 to 1.84); and an aggregate score over reported CONSORT items, 'total sum score' (standardised mean difference (SMD) 0.68 (99% CI 0.38 to 0.98)).

AUTHORS' CONCLUSIONS

Evidence has accumulated to suggest that the reporting of RCTs remains sub-optimal. This review updates a previous systematic review of eight evaluations. The findings of this review are similar to those from the original review and demonstrate that, despite the general inadequacies of reporting of RCTs, journal endorsement of the CONSORT Statement may beneficially influence the completeness of reporting of trials published in medical journals. Future prospective studies are needed to explore the influence of the CONSORT Statement dependent on the extent of editorial policies to ensure adherence to CONSORT guidance.

Assessment and implication of prognostic imbalance in randomized controlled trials with a binary outcome--a simulation study

BACKGROUND

Chance imbalance in baseline prognosis of a randomized controlled trial can lead to over or underestimation of treatment effects, particularly in trials with small sample sizes. Our study aimed to (1) evaluate the probability of imbalance in a binary prognostic factor (PF) between two treatment arms, (2) investigate the impact of prognostic imbalance on the estimation of a treatment effect, and (3) examine the effect of sample size (n) in relation to the first two objectives.

METHODS

We simulated data from parallel-group trials evaluating a binary outcome by varying the risk of the outcome, effect of the treatment, power and prevalence of the PF, and n. Logistic regression models with and without adjustment for the PF were compared in terms of bias, standard error, coverage of confidence interval and statistical power.

RESULTS

For a PF with a prevalence of 0.5, the probability of a difference in the frequency of the PF≥5% reaches 0.42 with 125/arm. Ignoring a strong PF (relative risk = 5) leads to underestimating the strength of a moderate treatment effect, and the underestimate is independent of n when n is >50/arm. Adjusting for such PF increases statistical power. If the PF is weak (RR = 2), adjustment makes little difference in statistical inference. Conditional on a 5% imbalance of a powerful PF, adjustment reduces the likelihood of large bias. If an absolute measure of imbalance ≥5% is deemed important, including 1000 patients/arm provides sufficient protection against such an imbalance. Two thousand patients/arm may provide an adequate control against large random deviations in treatment effect estimation in the presence of a powerful PF.

CONCLUSIONS

The probability of prognostic imbalance in small trials can be substantial. Covariate adjustment improves estimation accuracy and statistical power, and hence should be performed when strong PFs are observed.

Impact of single centre status on estimates of intervention effects in trials with continuous outcomes: meta-epidemiological study

OBJECTIVE

To compare estimates of intervention effects between single centre and multicentre randomised controlled trials with continuous outcomes.

DESIGN

Meta-epidemiological study.

DATA SOURCES

26 meta-analyses totalling 292 randomised controlled trials (177 single centre, 115 multicentre) with continuous outcomes published between January 2007 and January 2010 in the Cochrane database of systematic reviews.

DATA EXTRACTION

Data were extracted on characteristics of trials, single or multicentre status, risk of bias using the risk of bias tool of the Cochrane Collaboration, and results.

DATA SYNTHESIS

The intervention effects were estimated with standardised mean differences. For each meta-analysis, random effects meta-regression was used to estimate the difference in standardised mean differences between single centre and multicentre trials. Differences in standardised mean differences were then pooled across meta-analyses by a random-effects meta-analysis model. A combined difference in standardised mean differences of less than 0 indicated that single centre trials showed larger treatment effects, on average, than did multicentre trials. Because single centre trials may be more prone to publication bias and may have lower methodological quality than multicentre trials, sensitivity analyses were done with adjustment for sample size and domains of the risk of bias tool.

RESULTS

Single centre trials showed larger intervention effects than did multicentre trials (combined difference in standardised mean differences -0.09, 95% confidence interval -0.17 to -0.01, P=0.04), with low heterogeneity across individual meta-analyses (I(2)=0%, between meta-analyses variance τ(2)=0.00). Adjustment for sample size slightly attenuated the difference (-0.08, -0.17 to 0.01). Adjustment for risk of bias yielded similar estimates with wider confidence intervals, some of them crossing 0 (-0.09, -0.17 to 0.00 for overall risk of bias).

CONCLUSIONS

On average, single centre clinical trials with continuous outcomes showed slightly larger intervention effects than did multicentre trials. Further research is needed to investigate potential causes of these differences.

Early rehabilitation in critical care (eRiCC): functional electrical stimulation with cycling protocol for a randomised controlled trial

INTRODUCTION

Intensive care-acquired weakness is a common problem, leads to significant impairment in physical functioning and muscle strength, and is prevalent in individuals with sepsis. Early rehabilitation has been shown to be safe and feasible; however, commencement is often delayed due to a patient's inability to co-operate. An intervention that begins early in an intensive care unit (ICU) admission without the need for patient volition may be beneficial in attenuating muscle wasting. The eRiCC (early rehabilitation in critical care) trial will investigate the effectiveness of functional electrical stimulation-assisted cycling and cycling alone, compared to standard care, in individuals with sepsis.

METHODS AND ANALYSIS

This is a single centre randomised controlled trial. Participants (n=80) aged ≥18 years, with a diagnosis of sepsis or severe sepsis, who are expected to be mechanically ventilated for ≥48 h and remain in the intensive care ≥4 days will be randomised within 72 h of admission to (1) standard care or (2) intervention where participants will receive functional electrical muscle stimulation-assisted supine cycling on one leg while the other leg undergoes cycling alone. Primary outcome measures include: muscle mass (quadriceps ultrasonography; bioelectrical impedance spectroscopy); muscle strength (Medical Research Council Scale; hand-held dynamometry) and physical function (Physical Function in Intensive Care Test; Functional Status Score in intensive care; 6 min walk test). Blinded outcome assessors will assess measures at baseline, weekly, at ICU discharge and acute hospital discharge. Secondary measures will be evaluated in a nested subgroup (n=20) and will consist of biochemical/histological analyses of collected muscle, urine and blood samples at baseline and at ICU discharge.

ETHICS AND DISSEMINATION

Ethics approval has been obtained from the relevant institution, and results will be published to inform clinical practice in the care of patients with sepsis to optimise rehabilitation and physical function outcomes.

TRIAL REGISTRATION

Australian and New Zealand Clinical Trials Registry ACTRN12612000528853.

Human papillomaviruses in oral carcinoma and oral potentially malignant disorders: a systematic review

OBJECTIVES

Human papillomavirus (HPV) in oral carcinoma (OSCC) and potentially malignant disorders (OPMD) is controversial. The primary aim was to calculate pooled risk estimates for the association of HPV with OSCC and OPMD when compared with healthy oral mucosa as controls. We also examined the effects of sampling techniques on HPV detection rates.

METHODS

Systematic review was performed using PubMed (January 1966-September 2010) and EMBASE (January 1990-September 2010). Eligible studies included randomized controlled, cohort and cross-sectional studies. Pooled data were analysed by calculating odds ratios, using a random effects model. Risk of bias was based on characteristics of study group, appropriateness of the control group and prospective design.

RESULTS

Of the 1121 publications identified, 39 cross-sectional studies met the inclusion criteria. Collectively, 1885 cases and 2248 controls of OSCC and 956 cases and 675 controls of OPMD were available for analysis. Significant association was found between pooled HPV-DNA detection and OSCC (OR = 3.98; 95% CI: 2.62-6.02) and even for HPV16 only (OR = 3.86; 95% CI: 2.16-6.86). HPV was also associated with OPMD (OR = 3.87; 95% CI: 2.87-5.21). In a subgroup analysis of OPMD, HPV was also associated with oral leukoplakia (OR = 4.03; 95% CI: 2.34-6.92), oral lichen planus (OR = 5.12; 95% CI: 2.40-10.93), and epithelial dysplasia (OR = 5.10; 95% CI: 2.03-12.80).

CONCLUSIONS

The results suggest a potentially important causal association between HPV and OSCC and OPMD.