The effect of correlation structure on treatment contrasts estimated from incomplete clinical trial data with likelihood-based repeated measures compared with last observation carried forward ANOVA

Three-year outcomes of a randomized clinical trial of perioperative vaginal estrogen as adjunct to native tissue vaginal apical prolapse repair

BACKGROUND

As the muscular and connective tissue components of the vagina are estrogen responsive, clinicians may recommend vaginal estrogen to optimize tissues preoperatively and as a possible means to reduce prolapse recurrence, but long-term effects of perioperative intravaginal estrogen on surgical prolapse management are uncertain.

OBJECTIVE

This study aimed to compare the efficacy of perioperative vaginal estrogen vs placebo cream in reducing composite surgical treatment failure 36 months after native tissue transvaginal prolapse repair.

STUDY DESIGN

This was an extended follow-up of a randomized superiority trial conducted at 3 tertiary US sites. Postmenopausal patients with bothersome anterior or apical vaginal prolapse were randomized 1:1 to 1-g conjugated estrogen cream (0.625 mg/g) or placebo, inserted vaginally twice weekly for ≥5 weeks preoperatively and continued twice weekly for 12 months postoperatively. All participants underwent vaginal hysterectomy (if the uterus was present) and standardized uterosacral or sacrospinous ligament suspension at the surgeon's discretion. The primary report's outcome was time to failure by 12 months postoperatively, defined by a composite outcome of objective prolapse of the anterior or posterior walls beyond the hymen or the vaginal apex descending below one-third the total vaginal length, subjective bulge symptoms, and/or retreatment. After 12 months, participants could choose to use-or not use-vaginal estrogen for atrophy symptom bother. The secondary outcomes included Pelvic Organ Prolapse Quantification points, subjective prolapse symptom severity using the Patient Global Impression of Severity and the Patient Global Impression of Improvement, and prolapse-specific subscales of the 20-Item Pelvic Floor Distress Inventory and the Pelvic Floor Impact Questionnaire-Short Form 7. Data were analyzed as intent to treat and "per protocol" (ie, ≥50% of expected cream use per medication diary).

RESULTS

Of 206 postmenopausal patients, 199 were randomized, and 186 underwent surgery. Moreover, 164 postmenopausal patients (88.2%) provided 36-month data. The mean age was 65.0 years (standard deviation, 6.7). The characteristics were similar at baseline between the groups. Composite surgical failure rates were not significantly different between the estrogen group and the placebo group through 36 months, with model-estimated failure rates of 32.6% (95% confidence interval, 21.6%-42.0%) and 26.8% (95% confidence interval, 15.8%-36.3%), respectively (adjusted hazard ratio, 1.55; 95% confidence interval, 0.90-2.66; P=.11). The results were similar for the per-protocol analysis. Objective failures were more common than subjective failures, combined objective and subjective failures, or retreatment. Using the Patient Global Impression of Improvement, 75 of 80 estrogen participants (94%) and 72 of 76 placebo participants (95%) providing 36-month data reported that they were much or very much better 36 months after surgery (P>.99). These data included reports from 51 of 55 "surgical failures." Pelvic Organ Prolapse Quantification measurements, Patient Global Impression of Severity scores, and prolapse subscale scores of the 20-Item Pelvic Floor Distress Inventory and Pelvic Floor Impact Questionnaire-Short Form 7 all significantly improved for both the estrogen and placebo groups from baseline to 36 months postoperatively without differences between the groups. Of the 160 participants providing data on vaginal estrogen usage at 36 months postoperatively, 40 of 82 participants (49%) originally assigned to the estrogen group were using prescribed vaginal estrogen, and 47 of 78 participants (60%) assigned to the placebo group were using vaginal estrogen (P=.15).

CONCLUSION

Adjunctive perioperative vaginal estrogen applied ≥5 weeks preoperatively and 12 months postoperatively did not improve surgical success rates 36 months after uterosacral or sacrospinous ligament suspension prolapse repair. Patient perception of improvement remained very high at 36 months.

Role of Common Antihypertensives in the Growth of Abdominal Aortic Aneurysm at the Presurgical Stage

Background: Whether drug therapy slows the growth of abdominal aortic aneurysms (AAAs) in the Japanese population remains unknown. Methods and Results: In a multicenter prospective open-label study, patients with AAA at the presurgical stage (mean [±SD] AAA diameter 3.27±0.58 cm) were randomly assigned to treatment with candesartan (CAN; n=67) or amlodipine (AML; n=64) considering confounding factors (statin use, smoking, age, sex, renal function), with effects of blood pressure control minimized setting a target control level. The primary endpoint was percentage change in AAA diameter over 24 months. Secondary endpoints were changes in circulating biomarkers (high-sensitivity C-reactive protein [hs-CRP], malondialdehyde-low-density lipoprotein, tissue-specific inhibitor of metalloproteinase-1, matrix metalloproteinase [MMP] 2, MMP9, transforming growth factor-β1, plasma renin activity [PRA], angiotensin II, aldosterone). At 24 months, percentage changes in AAA diameter were comparable between the CAN and AML groups (8.4% [95% CI 6.23-10.59%] and 6.5% [95% CI 3.65-9.43%], respectively; P=0.23]. In subanalyses, AML attenuated AAA growth in patients with comorbid chronic kidney disease (CKD; P=0.04) or systolic blood pressure (SBP) <130 mmHg (P=0.003). AML exhibited a definite trend for slowing AAA growth exclusively in never-smokers (P=0.06). Among circulating surrogate candidates for AAA growth, PRA (P=0.02) and hs-CRP (P=0.001) were lower in the AML group. Conclusions: AML may prevent AAA growth in patients with CKD or lower SBP, associated with a decline in PRA and circulating hs-CRP.

Perioperative Vaginal Estrogen as Adjunct to Native Tissue Vaginal Apical Prolapse Repair: A Randomized Clinical Trial

Importance

Surgical repairs of apical/uterovaginal prolapse are commonly performed using native tissue pelvic ligaments as the point of attachment for the vaginal cuff after a hysterectomy. Clinicians may recommend vaginal estrogen in an effort to reduce prolapse recurrence, but the effects of intravaginal estrogen on surgical prolapse management are uncertain.

Objective

To compare the efficacy of perioperative vaginal estrogen vs placebo cream on prolapse recurrence following native tissue surgical prolapse repair.

Design, Setting, and Participants

This randomized superiority clinical trial was conducted at 3 tertiary US clinical sites (Texas, Alabama, Rhode Island). Postmenopausal women (N = 206) with bothersome anterior and apical vaginal prolapse interested in surgical repair were enrolled in urogynecology clinics between December 2016 and February 2020.

Interventions

The intervention was 1 g of conjugated estrogen cream (0.625 mg/g) or placebo, inserted vaginally nightly for 2 weeks and then twice weekly to complete at least 5 weeks of application preoperatively; this continued twice weekly for 12 months postoperatively. Participants underwent a vaginal hysterectomy (if uterus present) and standardized apical fixation (either uterosacral or sacrospinous ligament fixation).

Main Outcomes and Measures

The primary outcome was time to failure of prolapse repair by 12 months after surgery defined by at least 1 of the following 3 outcomes: anatomical/objective prolapse of the anterior or posterior walls beyond the hymen or the apex descending more than one-third of the vaginal length, subjective vaginal bulge symptoms, or repeated prolapse treatment. Secondary outcomes included measures of urinary and sexual function, symptoms and signs of urogenital atrophy, and adverse events.

Results

Of 206 postmenopausal women, 199 were randomized and 186 underwent surgery. The mean (SD) age of participants was 65 (6.7) years. The primary outcome was not significantly different for women receiving vaginal estrogen vs placebo through 12 months: 12-month failure incidence of 19% (n = 20) for vaginal estrogen vs 9% (n = 10) for placebo (adjusted hazard ratio, 1.97 [95% CI, 0.92-4.22]), with the anatomic recurrence component being most common, rather than vaginal bulge symptoms or prolapse repeated treatment. Masked surgeon assessment of vaginal tissue quality and estrogenization was significantly better in the vaginal estrogen group at the time of the operation. In the subset of participants with at least moderately bothersome vaginal atrophy symptoms at baseline (n = 109), the vaginal atrophy score for most bothersome symptom was significantly better at 12 months with vaginal estrogen.

Conclusions and Relevance

Adjunctive perioperative vaginal estrogen application did not improve surgical success rates after native tissue transvaginal prolapse repair.

Trial Registration

ClinicalTrials.gov Identifier: NCT02431897.

Repeated assessments of depressive symptoms in randomized psychosocial intervention trials: best practice for analyzing symptom change over time

OBJECTIVE

Psychotherapy randomized trials rarely have tested for the best fitting model for time effects. We examined the fit of different statistical models for examining time when repeated assessments of depressive symptoms are the primary outcome.

METHOD

We used data from three studies comparing psychotherapy treatments for major depressive disorder. Outcome measures were self-report ratings for Study 1 (N = 237) and Study 2 (N = 100) and clinician ratings for Study 3 (N = 120) of depressive symptoms measured at every session (Studies 1 and 2) or monthly (Study 3). We examined the fit of the following time patterns: linear, quadratic, cubic, log transformation of time, piece-wise linear, and unstructured.

RESULTS

In Study 1, a log-linear model had the best fit (Δ Akaike information criterion [AICc] = 7.5). In Study 2, all models had essentially no support (Δ AICcs > 10) in comparison to the best fitting model, which was the unstructured model. In Study 3, the cubic model had the best fit, but it was not significantly better than a log-linear (Δ AICc = 3.5) or unstructured model (Δ AICc = 2.5).

CONCLUSIONS

Trials should routinely compare different time models, including an unstructured model, when repeated measures of depressive symptoms are the primary outcome.

Mindful Self-Care for Caregivers: A Proof of Concept Study Investigating a Model for Embedded Caregiver Support in a Pediatric Setting

We conducted an open-trial proof of concept study to determine the safety, acceptability, and feasibility of Mindful Self-Care for Caregivers (MSCC) for parents of children with Autism Spectrum Disorders (ASD) (N = 13). The intervention was offered as a co-located care model in a pediatric specialty center where the participants' children received care. Results demonstrated that the intervention was: highly acceptable to all stakeholders (i.e., participants, the group facilitator, and center administration) and could be conducted safely by a masters-level practitioner with minimal resources. Further, secondary measures support a hypothesized interventional model of MSCC, demonstrating gains in mindfulness skills and sense of competency in the parenting role reduced perceived stress and depression in parents of children with ASD.

Bayesian Method of Borrowing Study-Level Historical Longitudinal Control Data for Mixed-Effects Models with Repeated Measures

Bringing historical control information into a new trial appropriately holds the promise of more efficient trial design with more accurate estimates, increased power, and fewer patients allocated to inefficacious control group, provided the historical control data are sufficiently similar to the concurrent control. Interest has been growing over the past few decades in leveraging historical clinical trial on the control arm. However, most of the current historical borrowing methods focus on incorporating patient-level historical control information at only one time point. In this work, we propose a Bayesian hierarchical Mixed effect Models for Repeated Measures to incorporate aggregated study-level longitudinal historical control estimates into the concurrent trial that collected repeated longitudinal data. The simulation study demonstrates that, as compared to one time point data analysis approach, leveraging longitudinal historical control data produces greater power enhancement and mitigates the power loss when the missing data under missing at random mechanism is present. Our work also helps fill the gap of lack of methods borrowing historical longitudinal control data from the published summarized estimates when patient-level control data are not available.

Feasibility of brief online gatekeeper training for Japanese university students: A randomized controlled trial

Gatekeeper training for university students is a promising suicide prevention approach. However, there is a lack of comparative studies with control groups and the effectiveness of online gatekeeper training programs is unclear. We investigated the effectiveness of brief online gatekeeper training for Japanese university students. Participants were divided into two groups (training or control) and answered surveys at three time points (pretest, posttest, follow-up). The gatekeeper training improved students' knowledge. Skills were also improved post-training but returned to baseline level at follow-up. A brief online program may improve knowledge, but the effect is limited.

Clinical determinants of cerebrovascular reactivity in very preterm infants during the transitional period

BACKGROUND

Preterm infants are at enhanced risk of brain injury due to altered cerebral haemodynamics during postnatal transition. This observational study aimed to assess the clinical determinants of transitional cerebrovascular reactivity and its association with intraventricular haemorrhage (IVH).

METHODS

Preterm infants <32 weeks underwent continuous monitoring of cerebral oxygenation and heart rate over the first 72 h after birth. Serial cranial and cardiac ultrasound assessments were performed to evaluate the ductal status and to diagnose IVH onset. The moving correlation coefficient between cerebral oxygenation and heart rate (TOHRx) was calculated. Linear mixed-effect models were used to analyse the impact of relevant clinical variables on TOHRx. The association between TOHRx and IVH development was also assessed.

RESULTS

Seventy-seven infants were included. A haemodynamically significant patent ductus arteriosus (hsPDA) (β = 0.044, 95% CI: 0.007-0.081) and ongoing dopamine treatment (β = 0.096, 95% CI: 0.032-0.159) were associated with increasing TOHRx, indicating impaired cerebrovascular reactivity. A significant association between TOHRx, mean arterial blood pressure (β = -0.004, 95% CI: -0.007, -0.001) and CRIB-II score (β = 0.007, 95% CI: 0.001-0.015) was also observed. TOHRx was significantly higher in infants developing high-grade IVH compared to those without IVH.

CONCLUSIONS

Dopamine treatment, low blood pressure, hsPDA and high CRIB-II are associated with impaired cerebrovascular reactivity during postnatal transition, with potential implications on IVH development.

IMPACT

The correlation coefficient between cerebral oxygenation and heart rate (TOHRx) provides a non-invasive estimation of cerebrovascular reactivity, whose failure has a potential pathogenic role in the development of IVH in preterm infants. This study shows that cerebrovascular reactivity during the transitional period improves over time and is affected by specific clinical and therapeutic factors, whose knowledge could support the development of individualized neuroprotective strategies in at-risk preterm infants. The evidence of increased TOHRx in infants developing high-grade compared to low-grade or no IVH during the transitional period further supports the role of impaired cerebrovascular reactivity in IVH pathophysiology.

Consideration of the adaptive randomization allocation ratio in the presence of treatment group heteroscedasticity in clinical trials

For randomized clinical trials, subjects' variance structures may vary over time among treatment groups, resulting in the heteroscedasticity of residuals in a regression analysis. Commonly used methods that assume equal variance among all treatment groups may not be able to control for a type I error. When the variances are indeed the same across treatment groups, an equal randomization allocation ratio will yield the greatest study power. However, out of ethical concern or urgent need for rare disease clinical trials, more patients may have to be allocated to the study drug arm. In these situations, an unequal randomization ratio should be considered. We propose a group variance-covariance and structures-based method to adapt the randomization ratio after interim analysis. We use simulations to compare commonly used statistical methods for continuous endpoints in assessing the impact of heteroscedasticity in equal and unequal randomization ratios and examine the extent to which the findings are affected by missing data.

The emotion-elicited late positive potential is stable across five testing sessions

Many studies have examined associations between neural and behavioral markers of attention to emotion and individual differences in affective functioning. However, the majority of these studies are cross-sectional, and examine associations between brain, behavior, and individual differences at one or two time-points, limiting our understanding of the extent to which these neural responses reflect trait-like patterns of attention. The present study used the Emotional Interrupt paradigm, and examined the stability and trajectory of behavioral (i.e., reaction time to targets following task-irrelevant appetitive, neutral, and aversive images), and neural responses to images (i.e., the late positive potential or LPP), across five sessions separated by one week in 86 individuals. Additionally, we examined the extent to which the LPP and behavioral measures were sensitive to naturally occurring daily fluctuations in positive and negative affect. Results indicate that, though the magnitude of the conditional LPP waveforms decreased over time, the degree of emotional modulation (i.e., differentiation of emotional from neutral) did not; in fact, differentiation of appetitive from neutral increased over time. Behavioral responses were similarly stable across sessions. Additionally, we largely did not observe significant effects of state positive and negative affect on the LPP or behavior over time. Finally, the LPP elicited by appetitive images significantly predicted reaction time to targets following these images. These data suggest that neural and behavioral markers of attention to motivationally salient cues may be trait-like in nature, and may be helpful in future studies seeking to identify markers of vulnerability for diverse forms of psychopathology.

Improved Decision-Making Confidence Using Item-Based Pharmacometric Model: Illustration with a Phase II Placebo-Controlled Trial

This study aimed to illustrate how a new methodology to assess clinical trial outcome measures using a longitudinal item response theory–based model (IRM) could serve as an alternative to mixed model repeated measures (MMRM). Data from the EXACT (Exacerbation of chronic pulmonary disease tool) which is used to capture frequency, severity, and duration of exacerbations in COPD were analyzed using an IRM. The IRM included a graded response model characterizing item parameters and functions describing symptom-time course. Total scores were simulated (month 12) using uncertainty in parameter estimates. The 50th (2.5th, 97.5th) percentiles of the resulting simulated differences in average total score (drug minus placebo) represented the estimated drug effect (95%CI), which was compared with published MMRM results. Furthermore, differences in sample size, sensitivity, specificity, and type I and II errors between approaches were explored. Patients received either oral danirixin 75 mg twice daily (n = 45) or placebo (n = 48) on top of standard of care over 52 weeks. A step function best described the COPD symptoms-time course in both trial arms. The IRM improved precision of the estimated drug effect compared to MMRM, resulting in a sample size of 2.5 times larger for the MMRM analysis to achieve the IRM precision. The IRM showed a higher probability of a positive predictive value (34%) than MMRM (22%). An item model–based analysis data gave more precise estimates of drug effect than MMRM analysis for the same endpoint in this one case study.

Design of a Randomized Clinical Trial of Perioperative Vaginal Estrogen Versus Placebo With Transvaginal Native Tissue Apical Prolapse Repair (Investigation to Minimize Prolapse Recurrence of the Vagina using Estrogen: IMPROVE)

OBJECTIVES

To provide the rationale and design for a randomized, double-blind clinical trial of conjugated estrogen vaginal cream (applied for at least 5 weeks preoperatively and continued twice-weekly through 12 months postoperatively) compared with placebo in postmenopausal women with symptomatic pelvic organ prolapse undergoing a standardized transvaginal native tissue apical repair.

METHODS

Study population, randomization process, study cream intervention, masking of participants and evaluators, placebo cream manufacture, standardized surgical intervention, and collection of adverse events are described. The primary outcome of surgical success is a composite of objectively no prolapse beyond the hymen and the vaginal cuff descending no more than one third the vaginal length; subjectively, no sense of vaginal pressure or bulging; and no retreatment for prolapse at 12 months. Time-to-failure postoperatively will be compared in the 2 groups with continued surveillance to 36 months. Secondary outcomes assessed at baseline, preoperatively (ie, after at least 5 weeks of study cream), and postoperatively at 6 month intervals include validated condition-specific and general quality-of-life metrics, overall impression of improvement, sexual function, vaginal atrophy symptoms, and body image. Challenges unique to this study include design and manufacture of placebo and defining and measuring study drug adherence.

RESULTS

Recruitment of 204 women is complete with 197 randomized. There have been 174 surgeries completed with 15 more pending; 111 have completed their 12 month postoperative visit.

CONCLUSIONS

This trial will contribute evidence-based information regarding the effect of perioperative vaginal estrogen as an adjunct therapy to standardized transvaginal native tissue prolapse surgical repair.

Comparison of Precision and Accuracy of Five Methods to Analyse Total Score Data

Total score (TS) data is generated from composite scales consisting of several questions/items, such as the Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS). The analysis method that most fully uses the information gathered is item response theory (IRT) models, but these are complex and require item-level data which may not be available. Therefore, the TS is commonly analysed with standard continuous variable (CV) models, which do not respect the bounded nature of data. Bounded integer (BI) models do respect the data nature but are not as extensively researched. Mixed models for repeated measures (MMRM) are an alternative that requires few assumptions and handles dropout without bias. If an IRT model exists, the expected mean and standard deviation of TS can be computed through IRT-informed functions-which allows CV and BI models to estimate parameters on the IRT scale. The fit, performance on external data and parameter precision (when applicable) of CV, BI and MMRM to analyse simulated TS data from the MDS-UPDRS motor subscale are investigated in this work. All models provided accurate predictions and residuals without trends, but the fit of CV and BI models was improved by IRT-informed functions. The IRT-informed BI model had more precise parameter estimates than the IRT-informed CV model. The IRT-informed models also had the best performance on external data, while the MMRM model was worst. In conclusion, (1) IRT-informed functions improve TS analyses and (2) IRT-informed BI models had more precise IRT parameter estimates than IRT-informed CV models.

Study protocol of the multi-centre, randomised controlled trial of the Frankfurt Early Intervention Programme A-FFIP versus early intervention as usual for toddlers and preschool children with Autism Spectrum Disorder (A-FFIP study)

BACKGROUND

Naturalistic developmental behavioural interventions (NDBI) have been shown to improve autism-specific symptoms in young children with Autism Spectrum Disorder (ASD). NDBI approaches, such as the ASD-specific Frankfurt Early Intervention Programme for ASD (A-FFIP), are based on ASD-specific developmental and learning aspects. A-FFIP is a low-intensity intervention which can easily be implemented in the local health care/social welfare system. The aim of the present study is to establish 1-year efficacy of the manualised early intervention programme A-FFIP in toddlers and preschool children with ASD. It is hypothesised that A-FFIP will result in improved ASD-specific symptoms compared to early intervention as usual (EIAU). Child- and family-specific secondary outcomes, as well as moderators and mediators of outcome, will be explored.

METHODS/DESIGN

A prospective, multi-centre, parallel-group, randomised controlled, phase-III trial comparing A-FFIP versus EIAU. A total of 134 children (A-FFIP: 67, EIAU: 67) aged 24-66 months at baseline assessment meeting the criteria for ASD (DSM-5) will be included. The primary outcome is the absolute change of the total score of the Brief Observation of Social Communication Change (BOSCC-AT) between baseline (T2) and 1-year follow-up (T6). The treatment effect will be tested, adjusted for relevant covariates applying a mixed model for repeated measures. Secondary outcomes are BOSCC social communication and repetitive-behaviour scores, single ASD symptoms, language, cognition, psychopathology, parental well-being and family quality of life. Predictors, moderators and mediating mechanisms will be explored.

DISCUSSION

If efficacy of the manualised A-FFIP early intervention is established, the current study has the potential to change clinical practice strongly towards the implementation of a low-intensity, evidence-based, natural early intervention in ASD. Early intervention in ASD requires specialist training, which subsequently needs to be developed or included into current training curricula.

TRIAL REGISTRATION

German Registry for Clinical Trials (Deutscher Register Klinischer Studien, DRKS); ID: 00016330. Retrospectively registered on 4 January 2019. URL: https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00016330.

Effectiveness of the Stepping Stones Triple P group parenting program in reducing comorbid behavioral problems in children with autism

Children with autism spectrum disorders often exhibit comorbid behavioral problems. These problems have an impact on the severity of the core symptoms, the progression of the disorder as well as on the families’ quality of life. We evaluated the effectiveness of the Stepping Stones Triple P group parent training program as a supplementary intervention in the treatment of children with autism spectrum disorder. Therefore, we employed a single group repeated measures design and assessed child variables via parents’ and teachers’ judgments at four successive time points. The participants were parents of 24 children with autism spectrum disorder aged between 3.6 and 12 years. We found a significant reduction of comorbid behavioral problems in the children, primarily in the parents’ judgment at follow-up. Furthermore, a reduction of the autism spectrum disorder core symptoms emerged. The teachers’ judgment particularly revealed an improvement in children’s social relationships. Effect sizes were large ( ƞ2 ranging from 0.14 to 0.23). The findings demonstrate the effectiveness of the Stepping Stones Triple P as a supplementary intervention for reducing comorbid behavioral problems in the treatment of children with autism spectrum disorder. Higher parental self-efficacy and parental attributions, including parents’ ability to influence child problem behaviors, are discussed as important factors for the effectiveness of Stepping Stones Triple P.

Computerized cognitive training in young adults with depressive symptoms: Effects on mood, cognition, and everyday functioning

BACKGROUND

Computerized cognitive training (CCT) has previously improved cognition and mood in people with depression. Existing research has not determined if the benefits following CCT are specific to the content of CCT or are a function of participation in an engaging activity. In this double-blind randomized controlled trial, we tested whether executive functioning and processing speed (EF/PS)-focused CCT could outperform verbal ability-focused CCT.

METHODS

46 young adults with at least mild depressive symptoms (HDRS ≥ 10) were recruited from the community and randomized to either EF/PS CCT or verbal ability CCT. Participants trained on their mobile device 5 days per week for 8 weeks. Depressive severity, everyday functioning, and cognition were evaluating pre and post-training.

RESULTS

The EF/PS group had greater gains in executive functioning and processing speed than the verbal group. There were no differences between groups in mood or everyday functioning improvement, though the EF/PS obtained equivalent improvement with half the training time. Both groups saw significant improvements in self and clinician-rated depressive severity, everyday functioning, and cognition.

LIMITATIONS

There was no waitlist control condition and the sample consisted of individuals with mild depressive symptoms and not diagnosed major depressive disorder.

CONCLUSIONS

CCT is associated with improved mood, cognition, and everyday functioning, though the type of CCT content does not differentially impact depressive symptom change. EF/PS focused CCT has greater impact on processing speed and executive functioning and leads to equivalent mood/everyday functioning gains as verbal-focused CCT more efficiently. Common factors remain plausible drivers of CCT's therapeutic effects.

Efficient two-step multivariate random effects meta-analysis of individual participant data for longitudinal clinical trials using mixed effects models

BACKGROUND

Mixed effects models have been widely applied in clinical trials that involve longitudinal repeated measurements, which possibly contain missing outcome data. In meta-analysis of individual participant data (IPD) based on these longitudinal studies, joint synthesis of the regression coefficient parameters can improve efficiency, especially for explorations of effect modifiers that are useful to predict the response or lack of response to particular treatments.

METHODS

In this article, we provide a valid and efficient two-step method for IPD meta-analyses using the mixed effects models that adequately addresses the between-studies heterogeneity using random effects models. The two-step method overcomes the practical difficulties of computations and modellings of the heterogeneity in the one-step method, and enables valid inference without loss of efficiency. We also show the two-step method can effectively circumvent the modellings of the between-studies heterogeneity of the variance-covariance parameters and provide valid and efficient estimators for the regression coefficient parameters, which are the primary objects of interests in the longitudinal studies. In addition, these methods can be easily implemented using standard statistical packages, and enable synthesis of IPD from different sources (e.g., from different platforms of clinical trial data sharing systems).

RESULTS

To assess the proposed method, we conducted simulation studies and also applied the method to an IPD meta-analysis of clinical trials for new generation antidepressants. Through the numerical studies, the validity and efficiency of the proposed method were demonstrated.

CONCLUSIONS

The two-step approach is an effective method for IPD meta-analyses of longitudinal clinical trials using mixed effects models. It can also effectively circumvent the modellings of the between-studies heterogeneity of the variance-covariance parameters, and enable efficient inferences for the regression coefficient parameters.

Allowing for uncertainty due to missing and LOCF imputed outcomes in meta-analysis

The use of the last observation carried forward (LOCF) method for imputing missing outcome data in randomized clinical trials has been much criticized and its shortcomings are well understood. However, only recently have published studies widely started using more appropriate imputation methods. Consequently, meta‐analyses often include several studies reporting their results according to LOCF. The results from such meta‐analyses are potentially biased and overprecise. We develop methods for estimating summary treatment effects for continuous outcomes in the presence of both missing and LOCF‐imputed outcome data. Our target is the treatment effect if complete follow‐up was obtained even if some participants drop out from the protocol treatment. We extend a previously developed meta‐analysis model, which accounts for the uncertainty due to missing outcome data via an informative missingness parameter. The extended model includes an extra parameter that reflects the level of prior confidence in the appropriateness of the LOCF imputation scheme. Neither parameter can be informed by the data and we resort to expert opinion and sensitivity analysis. We illustrate the methodology using two meta‐analyses of pharmacological interventions for depression.

Variation in analytic transparency in recent efficacy studies of antidepressant medication

The validity of clinical trial results is influenced by researchers' decisions regarding the management of missing data. Inadequate management of missing data has been identified as a significant source of bias that can result in an overestimation of drug efficacy. Transparency related to the management of missing data is essential to assess the strength of evidence reported in publications. In a subset of 17 randomised clinical trials for two new antidepressant medications, we present a case study in which we examined investigators' decisions regarding how to handle missing data and if their chosen method took into account, possible violations of analytic requirements that could affect results. The majority of trials (76%) concluded that there was a benefit of antidepressant treatment and in 94% the methodology for handling missing data was identifiable. Of these, 50% imputed data using the last observation carried forward and half used a mixed-effects model repeated measure approach. Most reports did not provide a rationale for the method used, and no trials described analyses regarding differences between completers and dropouts. Sensitivity analysis was inconsistently reported and correction for multiple comparisons was not uniformly applied. Lack of transparency for analytic choices related to handling of missing data testing was common in this subset of RCTs. Because management of missing data can directly influence the quality of study results, it is critical that journal editors develop and enforce standards for methodological transparency.

Laparoscopic anti-reflux surgery for the treatment of idiopathic pulmonary fibrosis (WRAP-IPF): a multicentre, randomised, controlled phase 2 trial

BACKGROUND

Abnormal acid gastro-oesophageal reflux (GER) is hypothesised to play a role in progression of idiopathic pulmonary fibrosis (IPF). We aimed to determine whether treatment of abnormal acid GER with laparoscopic anti-reflux surgery reduces the rate of disease progression.

METHODS

The WRAP-IPF trial was a randomised controlled trial of laparoscopic anti-reflux surgery in patients with IPF and abnormal acid GER recruited from six academic centres in the USA. We enrolled patients with IPF, abnormal acid GER (DeMeester score of ≥14·7; measured by 24-h pH monitoring) and preserved forced vital capacity (FVC). We excluded patients with a FVC below 50% predicted, a FEV₁/FVC ratio of less than 0·65, a history of acute respiratory illness in the past 12 weeks, a body-mass index greater than 35, and known severe pulmonary hypertension. Concomitant therapy with nintedanib and pirfenidone was allowed. The primary endpoint was change in FVC from randomisation to week 48, in the intention-to-treat population with mixed-effects models for repeated measures. This trial is registered with ClinicalTrials.gov, number NCT01982968.

FINDINGS

Between June 1, 2014, and Sept 30, 2016, we screened 72 patients and randomly assigned 58 patients to receive surgery (n=29) or no surgery (n=29). 27 patients in the surgery group and 20 patients in the no surgery group had an FVC measurement at 48 weeks (p=0·041). Intention-to-treat analysis adjusted for baseline anti-fibrotic use demonstrated the adjusted rate of change in FVC over 48 weeks was -0·05 L (95% CI -0·15 to 0·05) in the surgery group and -0·13 L (-0·23 to -0·02) in the non-surgery group (p=0·28). Acute exacerbation, respiratory-related hospitalisation, and death was less common in the surgery group without statistical significance. Dysphagia (eight [29%] of 28) and abdominal distention (four [14%] of 28) were the most common adverse events after surgery. There was one death in the surgery group and four deaths in the non-surgery group.

INTERPRETATION

Laparoscopic anti-reflux surgery in patients with IPF and abnormal acid GER is safe and well tolerated. A larger, well powered, randomised controlled study of anti-reflux surgery is needed in this population.

FUNDING

US National Institutes of Health National Heart, Lung and Blood Institute.

A simple correction to completer analyses and improvement on baseline observation carried forward

Replacing missing data using the baseline observation carried forward (BOCF) technique is known to be fraught with problems. Despite recommendations to the contrary, BOCF and the related last observation carried forward (LOCF) continue to be used in some fields of research. We first show the use of BOCF in testing for a change in a single sample is essentially equivalent to what results from a completer analysis. Next, we derive a simple method based only on summary statistics for adjusting inference from a completer analysis in situations where the estimand of the completer analysis is expected to be close to that of the full analysis set. For those with missing follow-up data, the method assumes a mean change of zero from baseline, with variance equivalent to that estimated from the patients who complete; in other words, it is essentially BOCF analysis with plausible variance inflation. The extension to two samples is considered and a similar adjustment of the completer analysis is presented. The method presented here can be used to reanalyze results from publications that treat missing data inadequately, to adjust for missing data in power calculations and as a quick way of checking the plausibility of an appropriate treatment of missing data. These practical tools are meant to complement, rather than compete with, established methods for dealing with missing data. Numerical simulations are used to elucidate some properties of the proposed method and, finally, we apply the method to data from a recent trial which evaluated weight loss programs.

Escitalopram versus paroxetine controlled release in major depressive disorder: a randomized trial

OBJECTIVE

There are no direct comparisons between escitalopram and paroxetine controlled release in patients with major depressive disorder (MDD).

METHODS

We conducted a 24-week, rater-masked, randomized trial of escitalopram (5-20 mg/day) versus paroxetine controlled release (12.5-50 mg/day) in patients with MDD (UMIN000011191). Patients with the diagnosis of moderate-to-severe MDD (a 17-item Hamilton Rating Scale for Depression [HAMD-17], with total score at baseline being ≥20) were recruited to participate in a parallel, randomized, controlled trial. The primary outcome for efficacy was an improvement in the 21-item HAMD (HAMD-21) total score at 24 weeks. The secondary outcomes were the response, remission, and discontinuation rates and the incidence of individual adverse events.

RESULTS

A total of 88 patients with MDD (males, 61.4%; mean age, 40.8±13.4 years) were recruited. The discontinuation rate was 58.0% (escitalopram, 55.8%; paroxetine controlled release, 60.0%). Both escitalopram and paroxetine controlled-release treatment groups exhibited significant reduction in the HAMD-21 total score at 2, 4, 8, 12, and 24 weeks from the baseline. However, there were no significant differences in the HAMD-21 total score, response rate, remission rate, and discontinuation rate at any time point between the groups. In addition, there were no significant differences in the incidence of any individual adverse events (eg, nausea, vomiting, and somnolence) between the treatment groups.

CONCLUSION

Our results suggest that escitalopram and paroxetine controlled release had similar efficacy and safety profiles in patients with MDD. One of the primary limitations of this study is the small sample size.

Alemtuzumab improves quality-of-life outcomes compared with subcutaneous interferon beta-1a in patients with active relapsing-remitting multiple sclerosis

Background:

Alemtuzumab was superior on clinical and magnetic resonance imaging (MRI) outcomes versus subcutaneous interferon beta-1a in phase 3 trials in patients with relapsing-remitting multiple sclerosis.

Objective:

To examine quality-of-life (QoL) outcomes in the alemtuzumab phase 3 trials.

Methods:

Patients who were treatment naive (Comparison of Alemtuzumab and Rebif® Efficacy in Multiple Sclerosis I [CARE-MS I]) or had an inadequate response to prior therapy (CARE-MS II) received annual courses of alemtuzumab 12 mg/day at baseline (5 days) and Month 12 (3 days) or subcutaneous interferon beta-1a 44 µg three times/week. QoL was measured every 6 or 12 months using Functional Assessment of Multiple Sclerosis (FAMS), European Quality of Life-5 Dimensions (EQ-5D) and its visual analog scale (EQ-VAS), and 36-Item Short-Form Survey (SF-36).

Results:

Statistically significant improvements from baseline with alemtuzumab were observed on all three QoL instruments at the earliest post-baseline assessment and sustained through Year 2. Statistically significant greater QoL improvements over subcutaneous interferon beta-1a were seen at all time points in CARE-MS II with FAMS, EQ-VAS and SF-36 physical component summary, and in CARE-MS I with FAMS.

Conclusion:

Patients treated with alemtuzumab had improvements in physical, mental, and emotional QoL regardless of treatment history. Improvements were significantly greater with alemtuzumab versus subcutaneous interferon beta-1a on both disease-specific and general measures of QoL.

Analysis of an incomplete longitudinal composite variable using a marginalized random effects model and multiple imputation

The total score, which is calculated as the sum of scores in multiple items or questions, is repeatedly measured in longitudinal clinical studies. A mixed effects model for repeated measures method is often used to analyze these data; however, if one or more individual items are not measured, the method cannot be directly applied to the total score. We develop two simple and interpretable procedures that infer fixed effects for a longitudinal continuous composite variable. These procedures consider that the items that compose the total score are multivariate longitudinal continuous data and, simultaneously, handle subject-level and item-level missing data. One procedure is based on a multivariate marginalized random effects model with a multiple of Kronecker product covariance matrices for serial time dependence and correlation among items. The other procedure is based on a multiple imputation approach with a multivariate normal model. In terms of the type-1 error rate and the bias of treatment effect in total score, the marginalized random effects model and multiple imputation procedures performed better than the standard mixed effects model for repeated measures analysis with listwise deletion and single imputations for handling item-level missing data. In particular, the mixed effects model for repeated measures with listwise deletion resulted in substantial inflation of the type-1 error rate. The marginalized random effects model and multiple imputation methods provide for a more efficient analysis by fully utilizing the partially available data, compared to the mixed effects model for repeated measures method with listwise deletion.

Online attention modification for social anxiety disorder: replication of a randomized controlled trial

Social anxiety disorder (SAD) models posit vigilance for external social threat cues and exacerbated self-focused attention as key in disorder development and maintenance. Evidence indicates a modified dot-probe protocol may reduce symptoms of SAD; however, the efficacy when compared to a standard protocol and long-term maintenance of treatment gains remains unclear. Furthermore, the efficacy of such protocols on SAD-related constructs remains relatively unknown. The current investigation clarified these associations using a randomized control trial replicating and extending previous research. Participants with SAD (n = 113; 71% women) were randomized to complete a standard (i.e. control) or modified (i.e. active) dot-probe protocol consisting of 15-min sessions twice weekly for four weeks. Self-reported symptoms were measured at baseline, post-treatment, and 4-month and 8-month follow-ups. Hierarchical linear modeling indicated significant self-reported reductions in symptoms of social anxiety, fear of negative evaluation, trait anxiety, and depression, but no such reductions in fear of positive evaluation. Symptom changes did not differ based on condition and were maintained at 8-month follow-up. Attentional biases during the dot-probe task were not related to symptom change. Overall, our results replicate support for the efficacy of both protocols in reducing symptoms of SAD and specific related constructs, and suggest a role of exposure, expectancy, or practice effects, rather than attention modification, in effecting such reductions. The current results also support distinct relationships between fears of negative and positive evaluation and social anxiety. Further research focused on identifying the mechanisms of change in attention modification protocols appears warranted.

Impact of a community-based lifestyle intervention program on health-related quality of life

BACKGROUND

The presence of risk factors for type 2 diabetes and cardiovascular disease, or the conditions themselves, contributes to lower health-related quality of life (HRQoL) among adults. Although community-based lifestyle intervention programs have been shown to be effective for improving risk factors for these diseases, the impact of these interventions on HRQoL has rarely been described.

PURPOSE

To examine changes in HRQoL following participation in the Group Lifestyle Balance program, a community translation of the Diabetes Prevention Program lifestyle intervention for adults with prediabetes and/or the metabolic syndrome.

METHODS

Participants enrolled in the 12-month, 22-session intervention program (N = 223) completed the EuroQol Health Questionnaire (EQ-5D-3L) at baseline, 6, and 12 months. Linear mixed-effects regression models determined change in EQ-5D-visual analog scale (VAS) and Index scores post-intervention.

RESULTS

Mean EQ-5D-VAS was improved by +7.38 (SE = 1.03) at 6 months and by +6.73 (SE = 1.06) at 12 months post-intervention (both; p < 0.0001). Mean changes in EQ-5D index values were +0.00 (SE = 0.01; NS) and +0.01 (SE = 0.01; p < 0.05), respectively. Adjusted for age, baseline score, and achieving intervention goals, mean change in EQ-5D-VAS was +11.83 (SE = 1.61) at 6 months and +11.23 (SE = 1.54) at 12 months (both; p < 0.0001). Adjusted mean change in EQ-5D index value was +0.04 (SE = 0.01) at 6 months and +0.05 (SE = 0.01) at 12 months (both; p < 0.01).

CONCLUSION

Participation in a community lifestyle intervention program resulted in improved HRQoL among adults with prediabetes and/or the metabolic syndrome. These benefits to HRQoL, together with improved clinical and behavioral outcomes, should increase the appeal of such programs for improving health.

Imputation strategies for blood pressure data nonignorably missing due to medication use

Background Underlying or untreated blood pressure (BP) is often an outcome of interest, but is unobservable when study participants are on anti-hypertensive medications. Untreated levels are not missing at random but would be higher among those on such medication. In such cases, standard methods of analysis may lead to bias.

Purpose BPs obtained at the private physician's office (out-of-study BPs) at the time of prescription of anti-hypertensive medications were available from Phase II of the Trials of Hypertension Prevention (TOHP) and were used to adjust for the potential bias.

Methods Observed out-of-study BPs were used to estimate the conditional expectation and variance of the unobserved unmedicated study BPs. For those with no physician data, imputation from bootstrap samples of out-of-study BPs was used. An iterative method based on the EM algorithm was used to estimate the unknown study parameters in a random-effects model using multiple imputations. This was compared to an alternative model for the out-of-study BPs based on a theoretical truncated normal distribution, and to standard analyses, including both multivariate repeated measures and last-observation-carried-forward (LOCF) analyses, using data from Phase II of TOHP.

Results Differences between methods were seen in the decline in BP over time in the reference group, where the changes from baseline to 36 months were 3.0 in univariate analyses, 2.4 using LOCF, and 2.6 in the multivariate analysis, compared to 2.0 or 1.7 in the imputation analyses, depending on the number of physician visits. Estimated intervention effects tended to be slightly larger using the imputation methods. Limitations out-of-study measures may not be available for other studies.

Conclusions Because the proposed strategy was based on an empirically observed distribution for out-of-study BP, fewer assumptions about the missing data were made. These data may be useful in suggesting imputation strategies for other studies.

Eliciting and using expert opinions about dropout bias in randomized controlled trials

Background The analysis of clinical trials with dropout usually assumes the missing data are `missing at random', i.e. given an individual's past observed data, their probability of dropout does not depend on their present outcome. However, in many settings this assumption is implausible, so it is sensible to assess the robustness of conclusions to departures from missing at random.

Purpose To develop a practical, accessible, approach that allows expert opinions about the degree of departure from missing at random in the analysis of a clinical trial to be meaningfully and accurately elicited and incorporated in sensitivity analysis.

Methods We elicit experts' prior beliefs about the mean difference between missing and observed outcomes in each trial arm. Then we perform a Bayesian synthesis of the information in the trial data with that in the experts' prior, using (i) a full Bayesian analysis for which we give WinBUGS code, and (ii) a simple approximate formula for the estimated treatment effect and its standard error. We illustrate our approach by re-analysing a recent trial of interventions to improve the quality of peer review.

Results In the peer review trial, the approximate formula agreed well with the full Bayesian analysis, and both showed substantially larger standard errors than an analysis assuming missing at random.

Limitations Strictly, the method is only applicable if the outcome is normally distributed. We did not elicit the full bivariate prior distribution, and instead used a sensitivity analysis. Our approach is not designed to incorporate prior beliefs about the intervention effect itself.

Conclusions Our proposed approach allows for the greater uncertainty introduced by missing data that are potentially informatively missing. It can therefore claim to be a truly conservative method, unlike methods such as `last observation carried forward'. It is practical and accessible to non-statisticians. It should be considered as part of the design and analysis of future clinical trials. Clinical Trials 2007; 4: 125—139. http:/ctj.sagepub.com

A randomized trial of aripiprazole vs blonanserin for the treatment of acute schizophrenia and related disorders

OBJECTIVE

There has been no direct comparison of aripiprazole and blonanserin for schizophrenia treatment. We conducted a 24-week, rater-masked, randomized trial of aripiprazole (6-30 mg/d) vs blonanserin (4-24 mg/d) in schizophrenia patients who were not taking any antipsychotic medication for more than 2 weeks before enrollment (UMIN000011194).

METHODS

The primary outcome measure for efficacy was improvement of Positive and Negative Syndrome Scale (PANSS) total score at week 24. Secondary outcomes were PANSS subscale scores, 21-item Hamilton Rating Scale for Depression (HAMD-21) score, response rate, discontinuation rate, and individual adverse events.

RESULTS

Forty-four patients were recruited. The discontinuation rate was 86.4% in the aripiprazole group and 68.2% in the blonanserin treatment group. There was no significant difference in mean time to discontinuation between the groups. Although both treatment groups showed significant reductions in the PANSS total score, PANSS subscale scores, and HAMD-21 scores at week 24, the magnitudes of the changes did not differ between the groups. There were no significant differences in the incidences of adverse events including somnolence, extrapyramidal symptoms, prolactin-related adverse events, and weight change between the groups.

CONCLUSION

Our results suggest similar efficacy and safety profiles of aripiprazole and blonanserin in the patients with schizophrenia. Double-blind controlled studies are needed to further explore the efficacy and safety of aripiprazole and blonanserin in schizophrenia.

'Physical Activity 4 Everyone' school-based intervention to prevent decline in adolescent physical activity levels: 12 month (mid-intervention) report on a cluster randomised trial

BACKGROUND

Adolescence is a recognised period of physical activity decline, particularly among low-income communities. We report the 12-month (midpoint) effects of a 2-year multicomponent physical activity intervention implemented in disadvantaged secondary schools.

METHODS

A cluster randomised trial was undertaken in 10 secondary schools located in disadvantaged areas in New South Wales, Australia. Students in Grade 7 were recruited, with follow-up in Grade 8. The intervention was guided by socioecological theory and included seven physical activity strategies, and six implementation adoption strategies. The primary outcome was mean minutes of moderate-to-vigorous physical activity (MVPA) per day assessed using Actigraph GT3X accelerometers. Outcome data were analysed using repeated measures linear mixed models.

RESULTS

At baseline, 1150 (93%) students participated in the data collection (mean age 12 years, 48% boys) and 1050 (79%) students participated at 12-month follow-up. By the 12-month follow-up, the six implementation adoption strategies had been used to support schools to deliver four of the seven physical activity elements. There was a significant group-by-time interaction for mean minutes of MVPA per day in favour of the intervention group (adjusted difference between groups at follow-up=3.85 min, 95% CI (0.79 to 6.91), p ≤ 0.01), including significantly more vigorous physical activity (2.45 min, p ≤ 0.01), equating to 27 min more MVPA per week.

SUMMARY

At 12-month follow-up, the intervention had reduced the decline in physical activity among adolescents from disadvantaged schools. The intervention may assist students to meet physical activity guidelines.

Comparison of bias-corrected covariance estimators for MMRM analysis in longitudinal data with dropouts

In longitudinal clinical trials, some subjects will drop out before completing the trial, so their measurements towards the end of the trial are not obtained. Mixed-effects models for repeated measures (MMRM) analysis with “unstructured” (UN) covariance structure are increasingly common as a primary analysis for group comparisons in these trials. Furthermore, model-based covariance estimators have been routinely used for testing the group difference and estimating confidence intervals of the difference in the MMRM analysis using the UN covariance. However, using the MMRM analysis with the UN covariance could lead to convergence problems for numerical optimization, especially in trials with a small-sample size. Although the so-called sandwich covariance estimator is robust to misspecification of the covariance structure, its performance deteriorates in settings with small-sample size. We investigated the performance of the sandwich covariance estimator and covariance estimators adjusted for small-sample bias proposed by Kauermann and Carroll ( J Am Stat Assoc 2001; 96: 1387–1396) and Mancl and DeRouen ( Biometrics 2001; 57: 126–134) fitting simpler covariance structures through a simulation study. In terms of the type 1 error rate and coverage probability of confidence intervals, Mancl and DeRouen’s covariance estimator with compound symmetry, first-order autoregressive (AR(1)), heterogeneous AR(1), and antedependence structures performed better than the original sandwich estimator and Kauermann and Carroll’s estimator with these structures in the scenarios where the variance increased across visits. The performance based on Mancl and DeRouen’s estimator with these structures was nearly equivalent to that based on the Kenward–Roger method for adjusting the standard errors and degrees of freedom with the UN structure. The model-based covariance estimator with the UN structure under unadjustment of the degrees of freedom, which is frequently used in applications, resulted in substantial inflation of the type 1 error rate. We recommend the use of Mancl and DeRouen’s estimator in MMRM analysis if the number of subjects completing is ( n + 5) or less, where n is the number of planned visits. Otherwise, the use of Kenward and Roger’s method with UN structure should be the best way.

Superiority of duloxetine to placebo in improving diabetic neuropathic pain: Results of a randomized controlled trial in Japan

Aims/Introduction:  Duloxetine has been suggested to exert analgesic effects by activating the descending inhibitory system through inhibition of serotonin (5‐HT) and noradrenaline (NA) reuptake. This randomized controlled trial investigated the efficacy and safety of duloxetine in Japanese patients with diabetic neuropathic pain (DNP).

Materials and Methods:  Duloxetine 40 or 60 mg/day or placebo was given orally once daily for 12 weeks. The primary efficacy measure was weekly mean 24‐h average pain severity score on the 11‐point Numerical Rating Scale.

Results:  At 12 weeks vs baseline, the 24‐h average pain score (adjusted mean ± SE) was significantly improved in the combined duloxetine (−2.47 ± 0.18) and duloxetine 40 mg (−2.41 ± 0.21) and 60 mg groups (−2.53 ± 0.21) as compared with the placebo group (−1.61 ± 0.18). Duloxetine also exerted significant improvements over the placebo in nearly all secondary outcome measures including 24‐h worst pain, night pain, Brief Pain Inventory (BPI) pain scores, Patient’s Global Impression of Improvement (PGI‐I) and health outcome measures, namely, various BPI interference scores. The incidence of adverse events (AE) was higher in the duloxetine groups than in the placebo group (duloxetine overall, 84.8%; duloxetine 40 mg, 84.7%; duloxetine 60 mg, 84.9%; placebo, 73.7%). Most AE were mild or moderate in severity, and resolved or relieved. There were no clinically significant safety concerns.

Conclusions:  Duloxetine 40 or 60 mg/day showed superiority over the placebo at reducing pain scores in patients with DNP. Duloxetine is safe, efficacious and clinically useful in the management of DNP. This trial was registered with ClinicalTrials.gov (no. NCT‐00552175). (J Diabetes Invest, doi: 10.1111/j.2040‐1124.2010.00073.x, 2010)

Early prediction of blonanserin response in Japanese patients with schizophrenia

BACKGROUND

Blonanserin is a second-generation antipsychotic used for the treatment of schizophrenia in Japan and Korea. The present study aimed to examine early prediction of blonanserin in patients with schizophrenia.

METHODS

An 8-week, prospective, single-arm, flexible-dose clinical trial of blonanserin in patients with schizophrenia was conducted under real-world conditions. The inclusion criteria were antipsychotic naïve, and first-episode schizophrenia patients or schizophrenia patients with no consumption of any antipsychotic medication for more than 4 weeks before enrollment in this study. The positive predictive value, negative predictive value, sensitivity, specificity, and predictive power were calculated for the response status at week 4 to predict the subsequent response at week 8.

RESULTS

Thirty-seven patients were recruited (56.8% of them had first-episode schizophrenia), and 28 (75.7%) completed the trial. At week 8, blonanserin was associated with a significant improvement in the Positive and Negative Syndrome Scale (PANSS) total score (P<0.0001) and in positive (P<0.0001), negative (P<0.0001), and general subscale scores (P<0.0001). In terms of percentage improvement of PANSS total scores from baseline to week 8, 64.9% of patients showed a ≥20% reduction in the PANSS total score and 48.6% showed a ≥30% reduction. However, 8.1% of patients experienced at least one adverse event. Using the 20% reduction in the PANSS total score at week 4 as a definition of an early response, the negative predictive values for later responses (ie, reductions of ≥30 and ≥40 in the PANSS total scores) were 88.9% and 94.1%, respectively. The specificities were 80.0% and 51.6%, respectively.

CONCLUSION

Our results suggest that the blonanserin response at week 4 could predict the later response at week 8.

Azithromycin and cough-specific health status in patients with chronic obstructive pulmonary disease and chronic cough: a randomised controlled trial

BACKGROUND

Macrolides reduce exacerbations in patients with COPD. Their effects on health status has not been assessed as primary outcome and is less clear. This study assessed the effects of prophylactic azithromycin on cough-specific health status in COPD-patients with chronic productive cough.

METHODS

In this randomised controlled trial 84 patients met the eligibility criteria: age of ≥40 years, COPD GOLD stage ≥2 and chronic productive cough. The intervention-group (n = 42) received azithromycin 250 mg 3 times a week and the control-group (n = 42) received a placebo. Primary outcome was cough-specific health status at 12 weeks, measured with the Leicester Cough Questionnaire (LCQ). Secondary outcomes included generic and COPD-specific health status and exacerbations. Changes in adverse events and microbiology were monitored.

RESULTS

Mean age of participants was 68 ± 10 years and mean FEV1 was 1.36 ± 0.47 L. The improvement in LCQ total score at 12 weeks was significantly greater with azithromycin (difference 1.3 ± 0.5, 95% CI 0.3;2.3, p = 0.01) and met the minimal clinically important difference. Similar results were found for the domain scores, and COPD-specific and generic health status questionnaires. Other secondary endpoints were non-significant. No imbalances in adverse events were found.

CONCLUSIONS

Prophylactic azithromycin improved cough-specific health status in COPD-patients with chronic productive cough to a clinically relevant degree.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01071161.

Comparing HAMD(17) and HAMD subscales on their ability to differentiate active treatment from placebo in randomized controlled trials

BACKGROUND

The 17-item Hamilton depression rating scale (HAMD(17)) is the standard efficacy outcome in antidepressant clinical trials. It is criticized for multidimensionality and poorly discriminating treatment from placebo. HAMD subscales may overcome these limitations and reduce the sample size of clinical trials. This study compared the discriminative performance of the HAMD(17) and three established HAMD subscales (Bech, Maier-Philipp, Gibbons) across a range of antidepressants with different mechanisms of action.

METHODS

We analyzed data from 24 clinical trials including 3692 patients randomized to tricyclic or tetracyclic antidepressants (TCAs or TeCAs), selective serotonin reuptake inhibitors (SSRIs) or placebo. Data were analyzed using a mixed model for repeated measurements (MMRM). Standardized effect sizes for the HAMD(17) and subscales were derived for every time-point, and their effect on sample size was evaluated.

RESULTS

For TCAs and TeCAs vs. placebo, the HAMD(17) consistently provided the highest standardized effects. The sample size to establish efficacy at week six was >25 percent smaller than for any of the subscales. However, for SSRIs vs. placebo, the HAMD(17) provided slightly smaller standardized effects and was the least efficient outcome. There were no relevant differences between the subscales.

LIMITATIONS

Data were derived exclusively from mirtazapine trials. Conclusions are restricted to clinical trial settings.

CONCLUSIONS

Comparative performance of the HAMD(17) and various subscales strongly depends on type of antidepressant. Results support using HAMD(17) as primary endpoint in clinical trials, but it will be beneficial to pro-actively include subscales as additional endpoints to successfully establish treatment effects of new antidepressants.

Comparisons of Analysis Methods for Proof-of-Concept Trials

Drug development struggles with high costs and time consuming processes. Hence, a need for new strategies has been accentuated by many stakeholders in drug development. This study proposes the use of pharmacometric models to rationalize drug development. Two simulated examples, within the therapeutic areas of acute stroke and type 2 diabetes, are utilized to compare a pharmacometric model-based analysis to a t-test with respect to study power of proof-of-concept (POC) trials. In all investigated examples and scenarios, the conventional statistical analysis resulted in several fold larger study sizes to achieve 80% power. For a scenario with a parallel design of one placebo group and one active dose arm, the difference between the conventional and pharmacometric approach was 4.3- and 8.4-fold, for the stroke and diabetes example, respectively. Although the model-based power depend on the model assumptions, in these scenarios, the pharmacometric model-based approach was demonstrated to permit drastic streamlining of POC trials.CPT: Pharmacometrics & Systems Pharmacology (2013) 2, e23; doi:10.1038/psp.2012.24; advance online publication 16 January 2013.

On the practical application of mixed effects models for repeated measures to clinical trial data

The use of mixed effects models for repeated measures (MMRM) for clinical trial analyses has recently gained broad support as a primary analysis methodology. Some questions of practical implementation detail remain, however. For example, whether and how to incorporate clinical trial data that is collected at nonprotocol-specified timepoints or clinic visits has not been systematically studied. In this paper, we compare different methods for applying MMRM to trials wherein data is available at protocol-specified timepoints, as well as nonprotocol-specified timepoints due to patient early discontinuation. The methods under consideration included observed case MMRM, per protocol visits MMRM, interval last observation carried forward (LOCF) MMRM, and a hybrid of the per protocol visits and interval LOCF MMRM approaches. Simulation results reveal that the method that best controls the type I error rate is the per protocol visits method. This method is also associated with the least precision among the competing methods. Thus, in confirmatory clinical trials wherein control of type I error rates is critical, per protocol visits MMRM is recommended. However, in exploratory trials where strict type I error control is not as critical, one may prefer interval LOCF MMRM due to its increased precision. Points to consider with respect to both study design (e.g., assigning schedule of events) and subsequent analysis are offered.

The Illness Density Index (IDI): A longitudinal measure of treatment efficacy

BACKGROUND

A reliable and meaningful quantitative index of success is paramount in the trial of any new treatment. However, existing methods for defining response and remission for treatments tested for psychiatric disorders are limited in that they often minimize the variance in change over time among individual patients and generally use arbitrarily chosen levels of functioning at specified times during treatment.

PURPOSE

To suggest and determine the properties of an alternative measure of treatment success, the Illness Density Index (IDI), that may be more sensitive to fluctuations in symptoms over the course of treatment compared to existing measures.

METHODS

We examined data from 64 depressed patients with multiple assessments of the Hamilton Depression Rating Scale (HDRS) over 12 weeks of randomized treatment in order to compare and contrast varying numerical definitions of response and remission, including percent change and linear slope over time.

RESULTS

Examination of the indices comparing the within-sample rank of individual patients revealed that these indices agree in cases where patients have little or no response as well as clear and sustained response, while they differ in patients who have a slow (or late) response as well as relapse during the treatment course.

LIMITATIONS

The measure may not be useful for all types of studies, especially short-term treatment trials.

CONCLUSIONS

The IDI is highly correlated with both categorical (e.g., remission) and continuous (e.g., percent change) definitions of treatment success. Furthermore, it differentiates certain trajectories of change that current definitions do not. Thus, the proposed index may be a valuable addition to current measures of efficacy, especially when trying to identify biological substrates of illness or predictors of long-term outcome.

Prednisone, azathioprine, and N-acetylcysteine for pulmonary fibrosis

BACKGROUND

A combination of prednisone, azathioprine, and N-acetylcysteine (NAC) has been widely used as a treatment for idiopathic pulmonary fibrosis. The safety and efficacy of this three-drug regimen is unknown.

METHODS

In this randomized, double-blind, placebo-controlled trial, we assigned patients with idiopathic pulmonary fibrosis who had mild-to-moderate lung-function impairment to one of three groups -- receiving a combination of prednisone, azathioprine, and NAC (combination therapy), NAC alone, or placebo -- in a 1:1:1 ratio. The primary outcome was the change in longitudinal measurements of forced vital capacity during a 60-week treatment period.

RESULTS

When approximately 50% of data had been collected (with 77 patients in the combination-therapy group and 78 in the placebo group), a planned interim analysis revealed that patients in the combination-therapy group, as compared with the placebo group, had an increased rate of death (8 vs. 1, P=0.01) and hospitalization (23 vs. 7, P<0.001). These observations, coupled with no evidence of physiological or clinical benefit for combination therapy, prompted the independent data and safety monitoring board to recommend termination of the combination-therapy group at a mean follow-up of 32 weeks. Data from the ongoing comparison of the NAC-only group and the placebo group are not reported here.

CONCLUSIONS

Increased risks of death and hospitalization were observed in patients with idiopathic pulmonary fibrosis who were treated with a combination of prednisone, azathioprine, and NAC, as compared with placebo. These findings provide evidence against the use of this combination in such patients. (Funded by the National Heart, Lung, and Blood Institute and the Cowlin Family Fund; ClinicalTrials.gov number, NCT00650091.).

Efficacy of duloxetine for the treatment of generalized anxiety disorder: implications for primary care physicians

OBJECTIVE

This study examined the efficacy and tolerability of duloxetine, a dual reuptake inhibitor of serotonin and norepinephrine, for the treatment of patients with generalized anxiety disorder (GAD).

METHOD

Patients were ≥ 18 years old and recruited from 5 European countries, the United States, and South Africa. The study had a 9-week, multicenter, randomized, double-blind, fixed-dose, placebo-controlled, parallel-group design. A total of 513 patients (mean age = 43.8 years; 67.8% female) with a DSM-IV-defined GAD diagnosis received treatment with duloxetine 60 mg/day (N = 168), duloxetine 120 mg/day (N = 170), or placebo (N = 175). The primary efficacy measure was the Hamilton Rating Scale for Anxiety (HAM-A) total score. Secondary measures included the Sheehan Disability Scale, HAM-A psychic and somatic anxiety factor scores, and HAM-A response, remission, and sustained improvement rates. The study was conducted from July 2004 to September 2005.

RESULTS

Both groups of duloxetine-treated patients demonstrated significantly greater improvements in anxiety symptom severity compared with placebo-treated patients as measured by HAM-A total score and HAM-A psychic and somatic anxiety factor scores (p values ranged from ≤ .01 to ≤ .001). Duloxetine-treated patients had greater functional improvements in Sheehan Disability Scale global and specific domain scores (p ≤ .001) than placebo-treated patients. Both duloxetine doses also resulted in significantly greater HAM-A response, remission, and sustained improvement rates compared with placebo (p values ranged from ≤ .01 to ≤ .001). The rate of study discontinuation due to adverse events was 11.3% for duloxetine 60 mg and 15.3% for duloxetine 120 mg versus 2.3% for placebo (p ≤ .001).

CONCLUSION

The results of this study demonstrate that duloxetine 60 mg/day and 120 mg/day were efficacious and well tolerated and thus may provide primary care physicians with a useful pharmacologic intervention for GAD.

CLINICAL TRIALS REGISTRATION

ClinicalTrials.gov identifier NCT00122824.

Structural models describing placebo treatment effects in schizophrenia and other neuropsychiatric disorders

Large variation in placebo response within and among clinical trials can substantially affect conclusions about the efficacy of new medications in psychiatry. Developing a robust placebo model to describe the placebo response is important to facilitate quantification of drug effects, and eventually to guide the design of clinical trials for psychiatric treatment via a model-based simulation approach. In addition, high dropout rates are very common in the placebo arm of psychiatric clinical trials. While developing models to evaluate the effect of placebo response, the data from patients who drop out of the trial should be considered for accurate interpretation of the results. The objective of this paper is to review the various empirical and semi-mechanistic models that have been used to quantify the placebo response in schizophrenia trials. Pros and cons of each placebo model are discussed. Additionally, placebo models used in other neuropsychiatric disorders like depression, Alzheimer's disease and Parkinson's disease are also reviewed with the objective of finding those placebo models that could be useful for clinical studies of both acute and chronic schizophrenic disease conditions. Better understanding of the patterns of dropout and the factors leading to dropouts are crucial in identifying the true placebo response. We therefore also review dropout models that are used in the development of models for treatment effects and in the optimization of clinical trials by simulation approaches. The use of an appropriate modelling strategy that is capable of identifying the potential sources of variable placebo responses and dropout rates is recommended for improving the sensitivity in discriminating between the effects of active treatment and placebo.

Handling baselines in repeated measures analyses with missing data at random

In longitudinal clinical studies, after randomization at baseline, subjects are followed for a period of time for development of symptoms. A mixed model for repeated measures (MMRM) can be used to analyze data from such studies. Fitzmaurice et al. (2004) outlined five approaches for handling baseline responses in an MMRM analysis. They are: (1) Retain the baselines as part of the outcome vector and make no assumptions about group differences in the mean response at baseline. (2) Retain the baselines as part of the outcome vector and assume the group means are equal at baseline. (3) Subtract the baselines from all of the remaining post-baseline responses, and analyze the differences from baseline. (4) Use the baselines as a covariate in the analysis of the post-baseline responses, assuming homogeneous regression slopes. (5) Use the baselines as a covariate in the analysis of the post-baseline responses, allowing different regression slopes. In this paper, we evaluate these five approaches in the presence of data missing at random. We evaluate the approaches based on the bias of the estimate and the coverage accuracy of the confidence interval. The results suggest that strategies 2 and 5 are recommended.

Should we treat depression with drugs or psychological interventions? A reply to Ioannidis

We reply to the Ioannidis's paper "Effectiveness of antidepressants; an evidence based myth constructed from a thousand controlled trials." We disagree that antidepressants have no greater efficacy than placebo. We present the efficacy from hundreds of trials in terms of the percentage of patients with a substantial clinical response (a 50% improvement or more symptomatic reduction). This meta-analysis finds that 42-70% of depressed patients improve with drug and 21%-39% improve with placebo. The response benefit of antidepressant treatment is 33%-11% greater than placebo. Ioannidis argues that it would be vanishingly smaller because systematic biasing in these clinical trials would reduce the drug-placebo difference to zero. Ioannidis' argument that antidepressants have no benefit is eroded by his failures of logic because he does not present any evidence that there are a large number of studies where placebo is substantially more effective than drug. (To reduce to zero, one would also have to show that some of the unpublished studies find placebo better than drug and have substantial systematic or methodological bias). We also present the empirical evidence showing that these methodological concerns generally have the opposite effect of what Ioannidis argues, supporting our contention that the measured efficacy of antidepressants likely underestimates true efficacy. Our most important criticism is Ioannidis' basic underlying argument about antidepressants that if the existing evidence is imperfect and methods can be criticized, then this proves that antidepressant are not efficacious. He presents no credible evidence that antidepressants have zero effect size. Valid arguments can point out difficulties with the data but do not prove that a given drug had no efficacy. Indeed better evidence might prove it was more efficacious that originally found. We find no empirical or ethical reason why psychiatrists should not try to help depressed patients with drugs and/or with psychotherapeutic/behavioral treatments given evidence of efficacy even though our treatment knowledge has limitations. The immense suffering of patients with major depression leads to ethical, moral, professional and legal obligations to treat patients with the best available tools at our disposal, while diligently and actively monitoring for adverse effects and actively revising treatment components as necessary.