Extending the DeLong algorithm for comparing areas under correlated receiver operating characteristic curves with missing data.

A nonparametric method proposed by DeLong et al in 1988 for comparing areas under correlated receiver operating characteristic curves is used widely in practice. However, the DeLong method as implemented in popular software quietly deletes individuals with any missing values, yielding potentially invalid and/or inefficient results. We simplify the DeLong algorithm using ranks and extend it to accommodate missing data by using a mixed model approach for multivariate data. Simulation results demonstrate the validity and efficiency of our procedure for data missing at random. We illustrate our proposed procedure in SAS, Stata, and R using the original DeLong data.

A Comparison of Treatment Effect Sizes in Matched Phase 2 and Phase 3 Trials of Advanced Therapeutics in Inflammatory Bowel Disease: Systematic Review and Meta-Analysis.

INTRODUCTION: Phase 2 trials are fundamental to the rational and efficient design of phase 3 trials. We aimed to determine the relationship of treatment effect size estimates from phase 2 and phase 3 clinical trials on advanced therapeutics in inflammatory bowel disease. METHODS: MEDLINE, EMBASE, CENTRAL, and the Cochrane library were searched from inception to December 19, 2022, to identify paired phase 2 and 3 placebo-controlled induction studies of advanced therapeutics for Crohn's disease (CD) and ulcerative colitis (UC). Treatment effect sizes were expressed as a risk ratio (RR) between the active arm and placebo arm. For the same therapeutics, RRs from phase 2 trials were divided by the RR from phase 3 trial to quantify the relationship of effect sizes between phases. RESULTS: Twenty-two studies (9 phase 2 trials, 13 phase 3 trials) were included for CD and 30 studies (12 phase 2 trials, 18 phase 3 trials) for UC. In UC (pooled RR 0.72; 95% confidence interval: 0.58-0.86; RR <1 indicates smaller treatment effect sizes in phase 2 trials), but not CD (pooled RR 1.01; 95% confidence interval: 0.84-1.18), phase 2 trials systematically underestimated treatment effect sizes for the primary endpoint compared with phase 3 trials. The underestimation was observed for clinical, but not endoscopic, endpoints in UC. DISCUSSION: Treatment effect sizes for the primary and clinical endpoints were similar across clinical trial phases in CD, but not UC, where only endoscopic endpoints were comparable. This will help inform clinical development plans and future trial design.

HLA-DQA1∗05 Genotype and Immunogenicity to Tumor Necrosis Factor-α Antagonists: A Systematic Review and Meta-analysis.

BACKGROUND & AIMS: Identifying patients at high risk of immunogenicity is important when selecting tumor necrosis factor (TNF)-α antagonists in patients with immune-mediated inflammatory diseases (IMIDs). We evaluated the association HLA-DQA1∗05 genotype and risk of immunogenicity with TNF-α antagonists. METHODS: Through a systematic review through July 14, 2022, we identified studies in patients with IMIDs treated with TNF-α antagonists, which reported the risk of immunogenicity and/or secondary loss of response in patients with HLA-DQA1∗05 variants. Primary outcome was risk of immunogenicity. We performed random effects meta-analysis and used GRADE to appraise certainty of evidence. RESULTS: On meta-analysis of 13 studies (3756 patients; median follow-up, 12 months; 41% with variants), HLA-DQA1∗05 variants were associated with 75% higher risk of immunogenicity compared with non-carriers (relative risk, 1.75; 95% confidence interval, 1.37-2.25) with considerable heterogeneity (I2 = 62%) (low certainty evidence). Positive and negative predictive values of HLA-DQA1∗05 variants for predicting immunogenicity were 30% and 80%, respectively. Proactive therapeutic drug monitoring, but not concomitant use of IMMs, IMIDs, and TNF-α antagonist-type, modified this association. Patients with HLA-DQA1∗05 variants experienced 2.2-fold higher risk of secondary loss of response (6 cohorts; relative risk, 2.24; 95% confidence interval, 1.67-3.00; I2 = 0%) (moderate certainty evidence). CONCLUSION: Variants in HLA-DQA1∗05 are associated with an increased risk in immunogenicity and secondary loss of response in patients with IMIDs treated with TNF-α antagonists. However, the positive and negative predictive value is moderate, and decisions on concomitant use of IMMs to prevent immunogenicity should be individualized based on all factors that influence drug clearance.

A rank-based approach to design and analysis of pretest-posttest randomized trials, with application to COVID-19 ordinal scale data.

Randomized controlled trials with a pretest-posttest design frequently yield ordered categorical outcome data. Focusing on the estimation of the win probability that a treated participant would have a better score than (or win over) a control participant, we developed methods for analysis and sample size planning for such trials. We exploited the analysis of covariance framework with the dependent variable being individual participants' win fractions at posttest and the covariate being the win fractions at pretest. The win fractions were obtained using the mid-ranks of the ordinal data. Simulation evaluation based on a recent randomized trial on COVID-19 suggests that the methods perform very well. A sample SAS code for data analysis is presented.

Parametric and nonparametric methods for confidence intervals and sample size planning for win probability in parallel-group randomized trials with Likert item and Likert scale data.

Data on the Likert scale are ubiquitous in medical research, including randomized trials. Statistical analysis of such data may be conducted using the means of raw scores or the rank information of the scores. In the context of parallel-group randomized trials, we quantify treatment effects by the probability that a subject in the treatment group has a better score than (or a win over) a subject in the control group. Asymptotic parametric and nonparametric confidence intervals for this win probability and associated sample size formulas are derived for studies with only follow-up scores, and those with both baseline and follow-up measurements. We assessed the performance of both the parametric and nonparametric approaches using simulation studies based on real studies with Likert item and Likert scale data. The simulation results demonstrate that even without baseline adjustment, the parametric methods did not perform well, in terms of bias, interval coverage percentage, balance of tail error, and assurance of achieving a pre-specified precision. In contrast, the nonparametric approach performed very well for both the unadjusted and adjusted win probability. We illustrate the methods with two examples: one using Likert item data and the other using Like scale data. We conclude that non-parametric methods are preferable for two-group randomization trials with Likert data. Illustrative SAS code for the nonparametric approach using existing procedures is provided.

Most Placebo-Controlled Trials in Inflammatory Bowel Disease were Underpowered Because of Overestimated Drug Efficacy Rates: Results from a Systematic Review of Induction Studies.

BACKGROUND AND AIMS: Most pharmaceutical clinical trials for inflammatory bowel disease [IBD] are placebo-controlled and require effect size estimation for a drug relative to placebo. We compared expected effect sizes in sample size calculations [SSCs] to actual effect sizes in IBD clinical trials. METHODS: MEDLINE, EMBASE, CENTRAL and the Cochrane library were searched from inception to March 26, 2021, to identify placebo-controlled induction studies for luminal Crohn's disease [CD] and ulcerative colitis [UC] that reported an SSC and a primary endpoint of clinical remission/response. Expected effects were subtracted from actual effects, and interquartile ranges [IQRs] for each corresponding median difference were calculated. Linear regression was used to assess whether placebo or drug event rate misspecifications were responsible for these differences. RESULTS: Of eligible studies, 36.9% [55/149] were excluded because of incomplete SSC reporting, yielding 94 studies [46 CD, 48 UC]. Treatment effects were overestimated in CD for remission (-12.6% [IQR: -16.3 to -1.6%]), in UC for remission (-10.2% [IQR: -16.5 to -5.6%]) and in CD for response (-15.3% [IQR: -27.1 to -5.8%]). Differences observed were due to overestimated drug event rates, whereas expected and actual placebo event rates were similar. A meta-regression demonstrated associations between overestimated treatment effect sizes and several trial characteristics: isolated ileal disease, longer CD duration, extensive colitis [UC], single-centre, phase 2 and no endoscopic endpoint component [UC]. CONCLUSION: Overestimation of IBD therapy efficacy rates resulted in smaller-than-expected treatment effects. These results should be used to inform SSCs and trial design for IBD drug development.

Distribution-Free Approach to the Design and Analysis of Randomized Stroke Trials With the Modified Rankin Scale.

BACKGROUND: Many methods have been suggested for analyzing the modified Rankin Scale (mRS). However, there lacks a unified approach to analysis and sample size determination that properly uses the ordinal nature of the data. We propose a simple method for CI estimation and corresponding sample size determination. METHODS: We quantify treatment effect by the win probability (WinP) that a randomly selected patient in the treatment group has an equal or a better mRS score than a patient in the control group. Thus, a win probability of 0.5 means no effect, likened to a draw in competitive sports. We estimate the win probability and its SE based on the ranks of mRS scores, where tied scores are handled by average ranks. Corresponding methods for hypothesis testing, CI estimation, and sample size determination are derived. The methods are evaluated with a simulation study based on real data from 10 randomized stroke trials that used mRS as the outcome measure. RESULTS: Simulation results demonstrated that the methods performed very well in terms of CI coverage, tail errors, and assurance to achieving the prespecified precision. Because the methods are very simple, we implemented them in an Excel spreadsheet, requiring only user inputs on frequencies of mRS scores in 2 comparison groups. CONCLUSIONS: Sound statistical methods are important for the success of randomized stroke trials. The proposed methods and associated spreadsheet should prove useful for stroke researchers in the planning and analysis of randomized trials. Meta-analysis has also been made easy for trials with ordinal scores.

The Atrial Fibrillation Therapies after ER visit: Outpatient Care for Patients with Acute AF - The AFTER3 Study.

BACKGROUND: Visits to the emergency room (ER) for atrial fibrillation/flutter (AF) are common, but follow-up care is rarely systematically organized and is often delayed. PURPOSE: We conducted a pilot program to develop a systematic, protocol-based system of care for patients presenting to the ER with a primary diagnosis of AF. METHODS: Consecutive patients presenting to the ER with ECG-documented AF at an urban teaching hospital were treated according to a guideline-based care protocol, including a patient toolkit at ER discharge, and systematic referral to a rapid access AF clinic. Consenting patients received questionnaires on AF knowledge, patient satisfaction, and the AFEQT questionnaire at first visit and three-month follow-up. RESULTS: Of the 321 patients with AF, 244 (76%) were discharged from the ER and 166 (68%) were referred to the AF clinic for urgent follow-up. Among 166 referred, 144 (87%) were seen, within a median 10.5 days (IQR 6-16.5 days); 128 (89%) patients agreed to participate in the study and 81% received a toolkit in the ER. The mean age of patients seen in AF clinic was 63.6±13.2 years and 59% were male. Eighty-seven percent were aware of their diagnosis, stroke risk (82%), possible complications (90%), treatment options (86%) and benefits of adherence (86%). Severity of Atrial Fibrillation class was > 2 in 51% at baseline; AFEQT scores increased from baseline (56.4±25.5) to three months post-ER visit (76.4±20.0), a moderately large improvement in QOL (p<0.0001). Seventy eight percent of patients with CHA2DS2-VASc score > 1 were treated with an oral anticoagulant. CONCLUSIONS: A systematic program to improve patient transition of care from the ER to community clinic was associated with prompt, guideline-based care, and high levels of patient disease awareness. Quality of life scores improved substantially between the index ER visit and 3 months post-visit.

Dronedarone and Captisol-enabled amiodarone in an experimental cardiac arrest.

OBJECTIVE: To compare the energy required for defibrillation and postshock outcomes after the administration of dronedarone, amiodarone, and placebo in a porcine model of cardiac arrest. METHODS: Forty-two pigs were randomized to amiodarone, dronedarone, or control treatments. After induction of ventricular fibrillation, compressions and ventilations were performed for 3 minutes and treatment was administered over 30 seconds. If defibrillation was unsuccessful, cardiopulmonary resuscitation continued and repeated shocks were administered every 2 minutes with continual hemodynamic monitoring for a total duration of 30 minutes. RESULTS: The cumulative energy required for defibrillation was 570 ± 422 J for dronedarone, 441 ± 365 J for amiodarone, and 347 ± 281 J for control (P = not significant). Survival at 30 minutes was 1 (7.1%) for dronedarone compared with 11 (78.6%) for control (P = 0.001). Mortality in the dronedarone group was because of refibrillation in 3 (21.4%) cases, atrioventricular block in 1 (7.1%) case, and hypotension not because of bradycardia in 9 (64.3%) cases. Two minutes after successful defibrillation, systolic aortic pressure was lower in dronedarone versus control (86.6 ± 26.9 vs. 110 ± 15.1 mm Hg; P = 0.035). CONCLUSIONS: The administration of dronedarone resulted in a significant reduction in survival and both systolic aortic and coronary perfusion pressure compared with control.

Idiopathic granulomatous mastitis masquerading as a postbiopsy abscess.

Idiopathic granulomatous mastitis (IGM) is a rare condition of unknown etiology with nonspecific findings. We present an unusual case of IGM manifesting after breast biopsy in a 42-year-old Turkish woman. IGM should be considered in the differential diagnosis when mastitis, carcinoma, and systemic diseases have been excluded and especially in the setting of a postbiopsy infection that is not responsive to antibiotic therapy.

Life-style and genital human papillomavirus in a cross-sectional survey in Shanxi Province, China.

BACKGROUND: To determine the age-stratified HPV prevalence rate and the risk factors of life-style associated with HPV infection among women in rural China. METHODS: An age-stratified, cross-sectional survey of 941 women between 16-59 years old was conducted in rural China. Carcinogenic HPV infection was determined using Digene's Hybrid Capture II HPV DNA test and interviews of life-style were conducted. Odds ratios (OR) and 95% confidence interval (CI) from the logistic regression models were used to determine the risk factors associated with HPV. RESULTS: Among 941 women, 745 who had sexual intercourse underwent a cervical examination. The prevalence rate of 13 carcinogenic HPV infections among women 20-59 years old was 15.97%. the rate of HPV prevalence in the 25-34 age group was statistically lower than that in 20-24 and 35-59 age groups (X²=13.3, P=0.0013). The OR of bathing every 7-19 days, 20-180 days, less than once every 180 days vs. bathing at least once a week were 1.19, 1.83 and 2.29 respectively and they had a dose-response relationship (Trend Test: P=0.003). The OR of women aged 25-34 age group vs. 20-24 age group was 0.40 (0.16 - 0.97) and the OR of bathing once every 180 days or less vs. at least once weekly was 2.22 (1.14 - 4.33) adjusted for the other confounding factors,. CONCLUSIONS: The HPV prevalence rate was lowest among child bearing women aged 25-34 year. Also, personal hygiene is significantly associated with the HPV infection in this area, regardless of age.

Medical students' preferences in radiology education a comparison between the Socratic and didactic methods utilizing powerpoint features in radiology education.

RATIONALE AND OBJECTIVES: The Socratic method has long been a traditional teaching method in medicine and law. It is currently accepted as the standard of teaching in clinical wards, while the didactic teaching method is widely used during the first 2 years of medical school. There are arguments in support of both styles of teaching. MATERIALS AND METHODS: After attending a radiology conference demonstrating different teaching methods, third-year and fourth-year medical students were invited to participate in an online anonymous survey. RESULTS: Of the 74 students who responded, 72% preferred to learn radiology in an active context. They preferred being given adequate time to find abnormalities on images, with feedback afterward from instructors, and they thought the best approach was a volunteer-based system of answering questions using the Socratic method in the small group. They desired to be asked questions in a way that was constructive and not belittling, to realize their knowledge deficits and to have daily pressure to come prepared. The respondents thought that pimping was an effective teaching tool, supporting previous studies. CONCLUSIONS: When teaching radiology, instructors should use the Socratic method to a greater extent. Combining Socratic teaching with gentle questioning by an instructor through the use of PowerPoint is a preferred method among medical students. This information is useful to improve medical education in the future, especially in radiology education.

Evaluation of medical student experience using medical student created StudentPACS flash based PACS simulator tutorials for learning radiological topics.

RATIONALE AND OBJECTIVES: With studies regularly containing hundreds of images, the authors believe that the ability to efficiently review numerous images and identify findings is an important skill to teach medical students. Using the StudentPACS Adobe Flash extension, created within their department, the authors created StudentPACS modules that provide users with a virtual picture archiving and communication system environment, in which findings can be selected by mouse, triggering questions with referenced answers. The aim was to assess medical students' impressions of how learning from these modules compared to their personal experiences learning radiology from textbooks or static images. MATERIALS AND METHODS: StudentPACS modules were created by medical students on elective under the supervision of resident and attending radiologists. MS I to IV students were then asked to complete StudentPACS modules that tied in with their current coursework, followed by an anonymous survey. Approximately 293 students participated. RESULTS: The majority of students reported that StudentPACS modules were either equivalent to or better than learning from static images or textbooks (90 +/- 3% [257 of 285], P < .00002), were not difficult to use (85 +/- 4% [248 of 293], P < .00002), presented them with clinical content that tied in well with the depicted imaging (90 +/- 3% [263 of 293], P < .00002), and taught them new information (69 +/- 5% [202 of 293], P < .00002). Most respondents felt the StudentPACS modules presented information they would find useful in clinical practice (91 +/- 3% [266 of 293], P < .00002), reported satisfactory experiences using StudentPACS modules as a source of self-directed learning material (79 +/- 5% [232 of 293], P < .00002), and stated that they would use StudentPACS modules for learning different topics in the future (85.6 +/- 4% [244 of 285], P < .00002). CONCLUSION: Medical students found using StudentPACS modules at least equivalent to, if not better than, using static books or annotated images.