Comparison of Pocock and Simon's covariate-adaptive randomization procedures in clinical trials.

When multiple influential covariates need to be balanced during a clinical trial, stratified blocked randomization and covariate-adaptive randomization procedures are frequently used in trials to prevent bias and enhance the validity of data analysis results. The latter approach is increasingly used in practice for a study with multiple covariates and limited sample sizes. Among a group of these approaches, the covariate-adaptive procedures proposed by Pocock and Simon are straightforward to be utilized in practice. We aim to investigate the optimal design parameters for the patient treatment assignment probability of their developed three methods. In addition, we seek to answer the question related to the randomization performance when additional covariates are added to the existing randomization procedure. We conducted extensive simulation studies to address these practically important questions.

Association of Win-Loss Record With Neuropsychiatric Symptoms and Brain Health Among Professional Fighters.

OBJECTIVE: Repetitive head impacts in professional fighting commonly lead to head injuries. Increased exposure to repetitive head trauma, measured by the number of professional fights and years of fighting, has been associated with slower processing speed and smaller brain volumes. The impact of win-loss outcomes has been investigated in other sports, with several studies suggesting that individuals on losing teams experience more head injuries. Here, the authors hypothesized that fighters with a worse fight record would exhibit poorer brain health outcomes. METHODS: The Professional Fighters Brain Health Study examined changes in neuropsychiatric symptoms, regional brain volume, and cognition among professional boxers and mixed martial arts fighters. These data were used to evaluate the relationship between win-loss ratios and brain health outcomes among professional fighters (N=212) by using validated neuropsychiatric symptom and cognitive measures and MRI data. RESULTS: Retired fighters with a better record demonstrated more impulsiveness (B=0.21, df=48) and slower processing speed (B=-0.42, df=31). More successful fighters did not perform better than fighters with worse records on any neuropsychiatric or cognitive test. Retired fighters with better fight records had smaller brain volumes in the subcortical gray matter, anterior corpus callosum, left and right hippocampi, left and right amygdala, and left thalamus. More successful active fighters had a smaller left amygdala volume. CONCLUSIONS: These findings suggest that among retired fighters, a better fight record was associated with greater impulsiveness, slower processing speed, and smaller brain volume in certain regions. This study shows that even successful fighters experience adverse effects on brain health.

Response adaptive randomization design for a two-stage study with binary response.

Response adaptive randomization has the potential to treat more participants in better treatments in a trial to benefit participants. We propose optimal response adaptive randomization designs for a two-stage study with binary response, having the smallest expected sample size or the fewest expected number of failures. Equal randomization is used in the first stage, and data from the first stage is used to determine the adaptive sample size ratio in the second stage. In the proposed optimal designs, the type I error rate and the statistical power are calculated from the asymptotic normal distributions. The new designs that minimize the expected number of failures have the advantage over the existing optimal randomized designs to substantially reduce the number of failures.

Two-stage response adaptive randomization designs for multi-arm trials with binary outcome.

In recent years, adaptive randomization methods have gained significant popularity in clinical research and trial design due to their ability to provide both efficiency and flexibility in adjusting the statistical procedures of ongoing clinical trials. For a study to compare multiple treatments, a multi-arm two-stage design could be utilized to select the best treatment from the first stage and further compare that treatment with control in the second stage. The traditional design used equal randomization in both stages. To better utilize the interim results from the first stage, we propose to develop response adaptive randomization two-stage designs for a multi-arm clinical trial with binary outcome. Two allocation methods are considered: (1) an optimal allocation based on a sequential design; (2) the play-the-winner rule. Optimal multi-arm two-stage designs are obtained under three criteria: minimizing the expected number of failures, minimizing the average expected sample size, and minimizing the expected sample size under the null hypothesis. Simulation studies show that the proposed adaptive design based on the play-the-winner rule has good performance. A phase II trial for patients with pancreas adenocarcinoma and a germline BRCA/PALB2 mutation was used to illustrate the application of the proposed response adaptive randomization designs.

Traumatic encephalopathy syndrome: application of new criteria to a cohort exposed to repetitive head impacts.

OBJECTIVE: To examine the characteristics of those who fulfil the recent National Institute of Neurological Disease and Stroke (NINDS) Consensus Diagnostic Criteria for Traumatic Encephalopathy Syndrome (TES) and test whether they show differences in MRI-based regional brain volumes, cognitive domains, and certain plasma biomarkers. METHODS: Professional fighters 35 years of age or older and/or retired were included. Participants were categorised as either having TES (TES+) or not (non-TES). TES+ participants were further subtyped by their cognitive profile. Multiple linear regression models were used to compare MRI-based regional brain volumes, cognitive performance, plasma tau and neurofilament light levels between TES- and TES+ groups. RESULTS: 176 participants (110 boxers and 66 MMA) were included in the analysis. 72 (41%)/176 were categorised as having TES, the likelihood of TES increasing with age. TES+ participants tended to be boxers, started fighting at a younger age, had more professional fights and knocked out more frequently. The TES+ group had lower regional brain volumes including both grey and white matter structures. TES+ also had lower scores on simple and choice reaction time, psychomotor speed and Trails A . CONCLUSION: The new TES criteria does distinguish a group of fighters with differences in regional brain volumes and reduced cognitive function. Our findings support the use of the NINDS criteria for TES in further research of the long-term effects of repetitive head impacts.

Acute Headache Management for Patients with Subarachnoid Hemorrhage: An International Survey of Health Care Providers.

BACKGROUND: Severe headaches are common after subarachnoid hemorrhage. Guidelines recommend treatment with acetaminophen and opioids, but patient data show that headaches often persist despite multimodal treatment approaches. Considering an overall slim body of data for a common complaint affecting patients with SAH during their intensive care stay, we set out to assess practice patterns in headache management among clinicians who treat patients with SAH. METHODS: We conducted an international cross-sectional study through a 37-question Web-based survey distributed to members of five professional societies relevant to intensive and neurocritical care from November 2021 to January 2022. Responses were characterized through descriptive analyses. Fisher's exact test was used to test associations. RESULTS: Of 516 respondents, 329 of 497 (66%) were from North America and 121 of 497 (24%) from Europe. Of 435 respondents, 379 (87%) reported headache as a major management concern for patients with SAH. Intensive care teams were primarily responsible for analgesia during hospitalization (249 of 435, 57%), whereas responsibility shifted to neurosurgery at discharge (233 of 501, 47%). Most used medications were acetaminophen (90%), opioids (66%), corticosteroids (28%), and antiseizure medications (28%). Opioids or medication combinations including opioids were most frequently perceived as most effective by 169 of 433 respondents (39%, predominantly intensivists), followed by corticosteroids or combinations with corticosteroids (96 of 433, 22%, predominantly neurologists). Of medications prescribed at discharge, acetaminophen was most common (303 of 381, 80%), followed by opioids (175 of 381, 46%) and antiseizure medications (173 of 381, 45%). Opioids during hospitalization were significantly more prescribed by intensivists, by providers managing higher numbers of patients with SAH, and in Europe. At discharge, opioids were more frequently prescribed in North America. Of 435 respondents, 299 (69%) indicated no change in prescription practice of opioids with the opioid crisis. Additional differences in prescription patterns between continents and providers and while inpatient versus at discharge were found. CONCLUSIONS: Post-SAH headache in the intensive care setting is a major clinical concern. Analgesia heavily relies on opioids both in use and in perception of efficacy, with no reported change in prescription patterns for opioids for most providers despite the significant drawbacks of opioids. Responsibility for analgesia shifts between hospitalization and discharge. International and provider-related differences are evident. Novel treatment strategies and alignment of prescription between providers are urgently needed.

Efficacy of Eravacycline Versus Best Previously Available Therapy for Adults With Pneumonia Due to Difficult-to-Treat Resistant (DTR) Acinetobacter baumannii.

BACKGROUND: Multidrug-resistant Acinetobacter baumannii remains challenging to treat. Although eravacycline has in vitro activity against this pathogen, there are no studies evaluating outcomes. OBJECTIVE: To assess the efficacy of eravacycline compared with best previously available therapy in adults with difficult-to-treat resistant (DTR) A. baumannii pneumonia. METHODS: This was a retrospective study of adults hospitalized for pneumonia with DTR A. baumannii. Patients receiving eravacycline were compared with those receiving best previously available therapy. The primary outcome was 30-day in-hospital mortality. Secondary outcomes included clinical cure at Day 14, hospital and intensive care unit (ICU) length of stay, microbiologic cure, and readmission within 90 days with a positive A. baumannii respiratory culture. RESULTS: Ninety-three patients were included, with 27 receiving eravacycline. Eravacycline was associated with higher 30-day mortality (33% vs 15%; P = 0.048), lower microbiologic cure (17% vs 59%; P = 0.004), and longer durations of mechanical ventilation (10.5 vs 6.5 days; P = 0.016). At baseline, eravacycline patients had more A. baumannii bacteremia and coinfection with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Among bacteremic patients, all 4 receiving eravacycline died by Day 30 and both patients receiving best previously available therapy survived. Upon exclusion of patients with bacteremia and SARS-CoV-2, there were no differences between the groups across any outcomes. CONCLUSIONS: Eravacycline-based combination therapy had similar outcomes to best previously available combination therapy for adults with DTR A. baumannii pneumonia. However, eravacycline should be used with caution in the setting of bacteremia as outcomes were poor in this population.

Randomized two-stage optimal design for interval-censored data.

Interval-censored data occur in a study where the exact event time of each participant is not observed but it is known to be within a certain time interval. Multiple tests were proposed for such data, including the logrank test by Sun, the proportional hazard test by Finkelstein, and the Wilcoxon-type test by Peto and Peto. We propose sample size calculations based on these tests for a parallel one-stage or two-stage design. When the proportional hazard assumption is met, the proportional hazard test and the logrank test need smaller sample sizes than the Wilcoxon-type test, and the sample size savings are substantial. But this trend is reversed when the proportional hazard assumption does not hold, and the sample size savings using the Wilcoxon-type test are sizable. An example from a lung cancer clinical trial is used to illustrate the application of the proposed sample size calculations.

Inpatient palliative care utilisation among patients with gallbladder cancer in the United States: A 10-year perspective.

OBJECTIVES: Gallbladder cancer (GBC) is a rare, poor-prognosis cancer with unique demographics, comorbidities and a paucity of research. This study investigated inpatient palliative care and its associations with demographics, comorbidities (e.g., obesity), length of stay and hospital charges in GBC in US hospitals (2007-2016). METHODS: Data were extracted from the National Inpatient Sample (NIS) database that contains deidentified clinical and nonclinical information for each hospitalisation. Inpatient palliative care utilisation was identified using the International Classification of Diseases (ICD-9 and ICD-10) codes (V66.7 and Z51.5). Generalised regression analysis was conducted with adjustment for variations in predictors. RESULTS: Of the 4921 reported GBC hospitalizations, only 10.3% encountered palliative care. Palliative care was associated with reduced hospital charges by $12,405 per hospitalisation (P < 0.0001) with no change in length of stay. Palliative care utilisation increased over time (P = 0.004). It was associated with age >80 years, with more severe disease, and in-hospital death (P < 0.0001). Obesity had a negative association with palliative care utilisation (P = 0.0029). DISCUSSION: Our novel study found that obese people were less likely to use palliative care services in GBC. Interventions are needed to increase palliative care consultation in GBC patients, particularly in obese patients.

Monte Carlo cross-validation for a study with binary outcome and limited sample size.

Cross-validation (CV) is a resampling approach to evaluate machine learning models when sample size is limited. The number of all possible combinations of folds for the training data, known as CV rounds, are often very small in leave-one-out CV. Alternatively, Monte Carlo cross-validation (MCCV) can be performed with a flexible number of simulations when computational resources are feasible for a study with limited sample size. We conduct extensive simulation studies to compare accuracy between MCCV and CV with the same number of simulations for a study with binary outcome (e.g., disease progression or not). Accuracy of MCCV is generally higher than CV although the gain is small. They have similar performance when sample size is large. Meanwhile, MCCV is going to provide reliable performance metrics as the number of simulations increases. Two real examples are used to illustrate the comparison between MCCV and CV.

Sex Differences in Cognitive Changes in De Novo Parkinson's Disease.

OBJECTIVE: To evaluate the sex differences in cognitive course over 4 years in Parkinson's disease (PD) patients with and without mild cognitive impairment (MCI) compared to controls. METHODS: Four-year longitudinal cognitive scores of 257 cognitively intact PD, 167 PD-MCI, and 140 controls from the Parkinson's Progression Markers Initiative were included. Longitudinal scores of men and women, and PD with and without MCI were compared. RESULTS: Women had better verbal memory, men had better visuospatial function. There was no interaction between sex, diagnostic group, and/or time (4-year follow-up period). CONCLUSIONS: Sex differences in cognitive course in de novo PD are similar to healthy aging. Cognitive decline rates in PD with and without MCI are similar for the first 4 years of PD.

Accurate confidence intervals for risk difference in meta-analysis with rare events.

BACKGROUND: Meta-analysis provides a useful statistical tool to effectively estimate treatment effect from multiple studies. When the outcome is binary and it is rare (e.g., safety data in clinical trials), the traditionally used methods may have unsatisfactory performance. METHODS: We propose using importance sampling to compute confidence intervals for risk difference in meta-analysis with rare events. The proposed intervals are not exact, but they often have the coverage probabilities close to the nominal level. We compare the proposed accurate intervals with the existing intervals from the fixed- or random-effects models and the interval by Tian et al. (2009). RESULTS: We conduct extensive simulation studies to compare them with regards to coverage probability and average length, when data are simulated under the homogeneity or heterogeneity assumption of study effects. CONCLUSIONS: The proposed accurate interval based on the random-effects model for sample space ordering generally has satisfactory performance under the heterogeneity assumption, while the traditionally used interval based on the fixed-effects model works well when the studies are homogeneous.

Two-stage optimal designs based on exact variance for a single-arm trial with survival endpoints.

Sample size calculation based on normal approximations is often associated with the loss of statistical power for a single-arm trial with a time-to-event endpoint. Recently, Wu (2015) derived the exact variance for the one-sample log-rank test under the alternative and showed that a single-arm one-stage study based on exact variance often has power above the nominal level while the type I error rate is controlled. We extend this approach to a single-arm two-stage design by using exact variances of the one-sample log-rank test for the first stage and the two stages combined. The empirical power of the proposed two-stage optimal designs is often not guaranteed under a two-stage design setting, which could be due to the asymptotic bi-variate normal distribution used to estimate the joint distribution of the test statistics. We adjust the nominal power level in the design search to guarantee the simulated power of the identified optimal design being above the nominal level. The sample size and the study time savings of the proposed two-stage designs are substantial as compared to the one-stage design.

The Relationship Between Fighting Style, Cognition, and Regional Brain Volume in Professional Combatants: A Preliminary Examination Using Brief Neurocognitive Measures.

OBJECTIVE: Repetitive head impacts (RHIs) in combat sports are associated with cognitive decline and brain volume reduction. While fighting style differences between boxers, mixed martial artists (MMAs), and martial artists (MAs) have resulted in a broader spectrum of injury, the effects of RHIs on MAs relative to other fighters have not yet been explored. This study aimed to determine a differential effect of fighting style on cognition and brain. SETTING: A large outpatient medical center specializing in neurological care. PARTICIPANTS, DESIGN, AND MAIN MEASURES: In total, 40 MAs, 188 boxers, and 279 MMAs were compared on baseline measures of subcortical regional brain volumes, after controlling for total brain volumes, and cognitive performance. RESULTS: Significant differences between MAs, MMAs, and boxers were observed in subcortical brain structure volumes and cognitive measures. MMAs and MAs consistently had larger volumes and higher scores than boxers. Fighting style significantly moderated the relationship between the number of professional fights and the volumes of various subcortical brain structures and performance on a measure of processing speed at baseline. CONCLUSIONS: Differences in RHIs across fighting styles may be of clinical significance. Exploring changes over time within the MA, boxer, and MMA cohorts may provide insight into longer-term discrepancies in subcortical regional brain volumes and cognitive functioning across fighting styles.

The effect of age of first exposure to competitive fighting on cognitive and other neuropsychiatric symptoms and brain volume.

It has long been established that fighting sports such as boxing and mixed martial arts can lead to head injury. Prior work from this group on the Professional Fighters Brain Health Study found that exposure to repetitive head impacts is associated with lower brain volumes and decreased processing speed in fighters. Current and previously licensed professional fighters were recruited, divided into active and retired cohorts, and matched with a control group that had no prior experience in sports with likely head trauma. This study examined the relationship between age of first exposure (AFE) to fighting sports and brain structure (MRI regional volume), cognitive performance (CNS Vital Signs, iComet C3), and clinical neuropsychiatric symptoms (PHQ-9, Barratt Impulsiveness Scale). Brain MRI data showed significant correlations between earlier AFE and smaller bilateral hippocampal and posterior corpus callosum volumes for both retired and active fighters. Earlier AFE in active fighters was correlated with decreased processing speed and decreased psychomotor speed. Retired fighters showed a correlation between earlier AFE and higher measures of depression and impulsivity. Overall, the results help to inform clinicians, governing bodies, parents, and athletes of the risks associated with beginning to compete in fighting sports at a young age.

Two-stage optimal designs with survival endpoint when the follow-up time is restricted.

BACKGROUND: Survival endpoint is frequently used in early phase clinical trials as the primary endpoint to assess the activity of a new treatment. Existing two-stage optimal designs with survival endpoint either over estimate the sample size or compute power outside the alternative hypothesis space. METHODS: We propose a new single-arm two-stage optimal design with survival endpoint by using the one-sample log rank test based on exact variance estimates. This proposed design with survival endpoint is analogous to Simon's two-stage design with binary endpoint, having restricted follow-up. RESULTS: We compare the proposed design with the existing two-stage designs, including the two-stage design with survival endpoint based on the nonparametric Nelson-Aalen estimate, and Simon's two-stage designs with or without interim accrual. The new design always performs better than these competitors with regards to the expected total study length, and requires a smaller expected sample size than Simon's design with interim accrual. CONCLUSIONS: The proposed two-stage minimax and optimal designs with survival endpoint are recommended for use in practice to shorten the study length of clinical trials.

Efficient statistical inference for a parallel study with missing data by using an exact method.

In a parallel group study comparing a new treatment with a standard of care, missing data often occur for various reasons. When the outcome is binary, the data from such studies can be summarized into a 2 × 3 contingency table, with the missing observations in the last column. When the missingness is neither related to the outcome of interest nor related to other outcomes from the study but it is covariate dependent with the sole covariate being treatment, this type of missing data mechanism is considered as missing at random. In 2016, Tian et al.  proposed three statistics to test the hypothesis that the response rate is equivalent for a parallel group study with missing data. The asymptotic limiting distributions of these test statistics were used for statistical inference. However, asymptotic approaches for testing proportions generally do not have satisfactory performance with regard to type I error rate control for a clinical trial with the sample size from small to medium. For this reason, we consider an exact approach based on maximization to provide valid and efficient statistical inference for a parallel group study with missing data. Exact approaches can guarantee the type I error rate and they are computationally feasible in this setting. We conduct extensive numerical studies to compare the performance of the exact approach based on the three statistics for a one-sided hypothesis testing problem. We conclude that the exact approach based on the likelihood ratio statistic is more powerful than the exact approach based on the other two statistics. Two real clinical trial data sets are used to illustrate the application of the proposed exact approach.

The Relations Among Depression, Cognition, and Brain Volume in Professional Boxers: A Preliminary Examination Using Brief Clinical Measures.

OBJECTIVE: Depression, neuropathology, and cognitive decline are commonly observed with repetitive head injuries (RHIs). We examined whether in boxers (a) clinically significant depression is associated with structural brain changes and cognition; (b) minimal symptoms of depression moderate the relations among RHI and brain volumes and cognition; and (c) baseline depression is associated with longitudinal cognitive changes. SETTING: Clinical Research Center. PARTICIPANTS: A total of 205 male professional boxers. DESIGN: Cross-sectional and longitudinal (subsample: n = 45; first visit to follow-up range = 1-6 years; mean = 2.61 years). MAIN MEASURES: Patient Health Questionnaire-9 depression; CNS Vital Signs cognitive battery; brain imaging. RESULTS: Clinically significant depression was associated with smaller regional volumes in insula, cingulate, orbitofrontal cortex, thalami, and middle corpus-callosum subregions; and with poorer verbal memory and psychomotor speed performance. Depression symptoms moderated the relations between RHI and bilateral thalami, left hippocampus, left medial orbitofrontal cortex, and bilateral insula volumes; but not cognition. Baseline depression was associated with poorer psychomotor speed and reaction time longitudinally and improved verbal memory performance longitudinally. CONCLUSION: Clinical depression is associated with volumetric and cognitive changes occasioning RHI exposure, and even minimal depressive symptoms may moderate the relations between exposure and brain volumes in key regions. Longitudinally, there is preliminary evidence that depression precedes cognitive changes.

Determining sample size for a binary diagnostic test in the presence of verification bias.

To compare a new binary diagnostic test with the gold standard, sensitivity and specificity are the two common measurements used to evaluate the new test. When not all the patients are verified by the gold standard due to time, budget, or cost considerations, several approaches have been proposed to compute sample size for such studies under the assumption of missing completely at random. However, the majority of them are based on asymptotic approaches that generally do not guarantee the type I and II error rates, and the remaining approaches use exact binomial distributions in sample size calculation but only the verified samples are used. In this article, for a study with verification bias, we propose computing exact sample sizes by using all the samples. The proposed approach is compared with the existing exact approach that compute sample size by using verified samples only, and the results show that the proposed approach requires fewer participants than the competitor.

Accurate unconditional p-values for a two-arm study with binary endpoints.

Unconditional exact tests are increasingly used in practice for categorical data to increase the power of a study and to make the data analysis approach being consistent with the study design. In a two-arm study with a binary endpoint, p-value based on the exact unconditional Barnard test is computed by maximizing the tail probability over a nuisance parameter with a range from 0 to 1. The traditional grid search method is able to find an approximate maximum with a partition of the parameter space, but it is not accurate and this approach becomes computationally intensive for a study beyond two groups. We propose using a polynomial method to rewrite the tail probability as a polynomial. The solutions from the derivative of the polynomial contain the solution for the global maximum of the tail probability. We use an example from a double-blind randomized Phase II cancer clinical trial to illustrate the application of the proposed polynomial method to achieve an accurate p-value. We also compare the performance of the proposed method and the traditional grid search method under various conditions. We would recommend using this new polynomial method in computing accurate exact unconditional p-values.