Associations of Testosterone and Related Hormones With All-Cause and Cardiovascular Mortality and Incident Cardiovascular Disease in Men : Individual Participant Data Meta-analyses.

BACKGROUND: Whether circulating sex hormones modulate mortality and cardiovascular disease (CVD) risk in aging men is controversial. PURPOSE: To clarify associations of sex hormones with these outcomes. DATA SOURCES: Systematic literature review to July 2019, with bridge searches to March 2024. STUDY SELECTION: Prospective cohort studies of community-dwelling men with sex steroids measured using mass spectrometry and at least 5 years of follow-up. DATA EXTRACTION: Independent variables were testosterone, sex hormone-binding globulin (SHBG), luteinizing hormone (LH), dihydrotestosterone (DHT), and estradiol concentrations. Primary outcomes were all-cause mortality, CVD death, and incident CVD events. Covariates included age, body mass index, marital status, alcohol consumption, smoking, physical activity, hypertension, diabetes, creatinine concentration, ratio of total to high-density lipoprotein cholesterol, and lipid medication use. DATA SYNTHESIS: Nine studies provided individual participant data (IPD) (255 830 participant-years). Eleven studies provided summary estimates (n = 24 109). Two-stage random-effects IPD meta-analyses found that men with baseline testosterone concentrations below 7.4 nmol/L (<213 ng/dL), LH concentrations above 10 IU/L, or estradiol concentrations below 5.1 pmol/L had higher all-cause mortality, and those with testosterone concentrations below 5.3 nmol/L (<153 ng/dL) had higher CVD mortality risk. Lower SHBG concentration was associated with lower all-cause mortality (median for quintile 1 [Q1] vs. Q5, 20.6 vs. 68.3 nmol/L; adjusted hazard ratio [HR], 0.85 [95% CI, 0.77 to 0.95]) and lower CVD mortality (adjusted HR, 0.81 [CI, 0.65 to 1.00]). Men with lower baseline DHT concentrations had higher risk for all-cause mortality (median for Q1 vs. Q5, 0.69 vs. 2.45 nmol/L; adjusted HR, 1.19 [CI, 1.08 to 1.30]) and CVD mortality (adjusted HR, 1.29 [CI, 1.03 to 1.61]), and risk also increased with DHT concentrations above 2.45 nmol/L. Men with DHT concentrations below 0.59 nmol/L had increased risk for incident CVD events. LIMITATIONS: Observational study design, heterogeneity among studies, and imputation of missing data. CONCLUSION: Men with low testosterone, high LH, or very low estradiol concentrations had increased all-cause mortality. SHBG concentration was positively associated and DHT concentration was nonlinearly associated with all-cause and CVD mortality. PRIMARY FUNDING SOURCE: Medical Research Future Fund, Government of Western Australia, and Lawley Pharmaceuticals. (PROSPERO: CRD42019139668).

Prevalence of non-communicable diseases among household contacts of people with tuberculosis: A systematic review and individual participant data meta-analysis.

OBJECTIVE: To investigate the prevalence of non-communicable diseases among household contacts of people with tuberculosis. METHODS: We conducted a systematic review and individual participant data meta-analysis. We searched Medline, Embase and the Global Index Medicus from inception to 16 May 2023. We included studies that assessed for at least one non-communicable disease among household contacts of people with clinical tuberculosis. We estimated the non-communicable disease prevalence through mixed effects logistic regression for studies providing individual participant data, and compared it with estimates from aggregated data meta-analyses. Furthermore, we compared age and sex-standardised non-communicable disease prevalence with national-level estimates standardised for age and sex. RESULTS: We identified 39 eligible studies, of which 14 provided individual participant data (29,194 contacts). Of the remaining 25 studies, 18 studies reported aggregated data suitable for aggregated data meta-analysis. In individual participant data analysis, the pooled prevalence of diabetes in studies that undertook biochemical testing was 8.8% (95% confidence interval [CI], 5.1%-14.9%, four studies). Age-and sex-standardised prevalence was higher in two studies (10.4% vs. 6.9% and 11.5% vs. 8.4%) than the corresponding national estimates and similar in two studies. Prevalence of diabetes mellitus based on self-report or medical records was 3.4% (95% CI 2.6%-4.6%, 14 studies). Prevalence did not significantly differ compared to estimates from aggregated data meta-analysis. There were limited data for other non-communicable diseases. CONCLUSION: The prevalence of diabetes mellitus among household contacts was high while that of known diabetes was substantially lower, suggesting the underdiagnosis. tuberculosis household contact investigation offers opportunities to deliver multifaceted interventions to identify tuberculosis infection and disease, screen for non-communicable diseases and address shared risk factors.

Factors Associated With Circulating Sex Hormones in Men : Individual Participant Data Meta-analyses.

BACKGROUND: Various factors modulate circulating testosterone in men, affecting interpretation of testosterone measurements. PURPOSE: To clarify factors associated with variations in sex hormone concentrations. DATA SOURCES: Systematic literature searches (to July 2019). STUDY SELECTION: Prospective cohort studies of community-dwelling men with total testosterone measured using mass spectrometry. DATA EXTRACTION: Individual participant data (IPD) (9 studies; n = 21 074) and aggregate data (2 studies; n = 4075). Sociodemographic, lifestyle, and health factors and concentrations of total testosterone, sex hormone-binding globulin (SHBG), luteinizing hormone (LH), dihydrotestosterone, and estradiol were extracted. DATA SYNTHESIS: Two-stage random-effects IPD meta-analyses found a nonlinear association of testosterone with age, with negligible change among men aged 17 to 70 years (change per SD increase about the midpoint, -0.27 nmol/L [-7.8 ng/dL] [CI, -0.71 to 0.18 nmol/L {-20.5 to 5.2 ng/dL}]) and decreasing testosterone levels with age for men older than 70 years (-1.55 nmol/L [-44.7 ng/dL] [CI, -2.05 to -1.06 nmol/L {-59.1 to -30.6 ng/dL}]). Testosterone was inversely associated with body mass index (BMI) (change per SD increase, -2.42 nmol/L [-69.7 ng/dL] [CI, -2.70 to -2.13 nmol/L {-77.8 to -61.4 ng/dL}]). Testosterone concentrations were lower for men who were married (mean difference, -0.57 nmol/L [-16.4 ng/dL] [CI, -0.89 to -0.26 nmol/L {-25.6 to -7.5 ng/dL}]); undertook at most 75 minutes of vigorous physical activity per week (-0.51 nmol/L [-14.7 ng/dL] [CI, -0.90 to -0.13 nmol/L {-25.9 to -3.7 ng/dL}]); were former smokers (-0.34 nmol/L [-9.8 ng/dL] [CI, -0.55 to -0.12 nmol/L {-15.9 to -3.5 ng/dL}]); or had hypertension (-0.53 nmol/L [-15.3 ng/dL] [CI, -0.82 to -0.24 nmol/L {-23.6 to -6.9 ng/dL}]), cardiovascular disease (-0.35 nmol/L [-10.1 ng/dL] [CI, -0.55 to -0.15 nmol/L {-15.9 to -4.3 ng/dL}]), cancer (-1.39 nmol/L [-40.1 ng/dL] [CI, -1.79 to -0.99 nmol/L {-51.6 to -28.5 ng/dL}]), or diabetes (-1.43 nmol/L [-41.2 ng/dL] [CI, -1.65 to -1.22 nmol/L {-47.6 to -35.2 ng/dL}]). Sex hormone-binding globulin was directly associated with age and inversely associated with BMI. Luteinizing hormone was directly associated with age in men older than 70 years. LIMITATION: Cross-sectional analysis, heterogeneity between studies and in timing of blood sampling, and imputation for missing data. CONCLUSION: Multiple factors are associated with variation in male testosterone, SHBG, and LH concentrations. Reduced testosterone and increased LH concentrations may indicate impaired testicular function after age 70 years. Interpretation of individual testosterone measurements should account particularly for age older than 70 years, obesity, diabetes, and cancer. PRIMARY FUNDING SOURCE: Medical Research Future Fund, Government of Western Australia, and Lawley Pharmaceuticals. (PROSPERO: CRD42019139668).

Comparison of the Working Alliance in Blended Cognitive Behavioral Therapy and Treatment as Usual for Depression in Europe: Secondary Data Analysis of the E-COMPARED Randomized Controlled Trial.

BACKGROUND: Increasing interest has centered on the psychotherapeutic working alliance as a means of understanding clinical change in digital mental health interventions in recent years. However, little is understood about how and to what extent a digital mental health program can have an impact on the working alliance and clinical outcomes in a blended (therapist plus digital program) cognitive behavioral therapy (bCBT) intervention for depression. OBJECTIVE: This study aimed to test the difference in working alliance scores between bCBT and treatment as usual (TAU), examine the association between working alliance and depression severity scores in both arms, and test for an interaction between system usability and working alliance with regard to the association between working alliance and depression scores in bCBT at 3-month assessments. METHODS: We conducted a secondary data analysis of the E-COMPARED (European Comparative Effectiveness Research on Blended Depression Treatment versus Treatment-as-usual) trial, which compared bCBT with TAU across 9 European countries. Data were collected in primary care and specialized services between April 2015 and December 2017. Eligible participants aged 18 years or older and diagnosed with major depressive disorder were randomized to either bCBT (n=476) or TAU (n=467). bCBT consisted of 6-20 sessions of bCBT (involving face-to-face sessions with a therapist and an internet-based program). TAU consisted of usual care for depression. The main outcomes were scores of the working alliance (Working Alliance Inventory-Short Revised-Client [WAI-SR-C]) and depressive symptoms (Patient Health Questionnaire-9 [PHQ-9]) at 3 months after randomization. Other variables included system usability scores (System Usability Scale-Client [SUS-C]) at 3 months and baseline demographic information. Data from baseline and 3-month assessments were analyzed using linear regression models that adjusted for a set of baseline variables. RESULTS: Of the 945 included participants, 644 (68.2%) were female, and the mean age was 38.96 years (IQR 38). bCBT was associated with higher composite WAI-SR-C scores compared to TAU (B=5.67, 95% CI 4.48-6.86). There was an inverse association between WAI-SR-C and PHQ-9 in bCBT (B=-0.12, 95% CI -0.17 to -0.06) and TAU (B=-0.06, 95% CI -0.11 to -0.02), in which as WAI-SR-C scores increased, PHQ-9 scores decreased. Finally, there was a significant interaction between SUS-C and WAI-SR-C with regard to an inverse association between higher WAI-SR-C scores and lower PHQ-9 scores in bCBT (b=-0.030, 95% CI -0.05 to -0.01; P=.005). CONCLUSIONS: To our knowledge, this is the first study to show that bCBT may enhance the client working alliance when compared to evidence-based routine care for depression that services reported offering. The working alliance in bCBT was also associated with clinical improvements that appear to be enhanced by good program usability. Our findings add further weight to the view that the addition of internet-delivered CBT to face-to-face CBT may positively augment experiences of the working alliance. TRIAL REGISTRATION: ClinicalTrials.gov NCT02542891, https://clinicaltrials.gov/study/NCT02542891; German Clinical Trials Register DRKS00006866, https://drks.de/search/en/trial/DRKS00006866; Netherlands Trials Register NTR4962, https://www.onderzoekmetmensen.nl/en/trial/25452; ClinicalTrials.Gov NCT02389660, https://clinicaltrials.gov/study/NCT02389660; ClinicalTrials.gov NCT02361684, https://clinicaltrials.gov/study/NCT02361684; ClinicalTrials.gov NCT02449447, https://clinicaltrials.gov/study/NCT02449447; ClinicalTrials.gov NCT02410616, https://clinicaltrials.gov/study/NCT02410616; ISRCTN Registry ISRCTN12388725, https://www.isrctn.com/ISRCTN12388725?q=ISRCTN12388725&filters=&sort=&offset=1&totalResults=1&page=1&pageSize=10; ClinicalTrials.gov NCT02796573, https://classic.clinicaltrials.gov/ct2/show/NCT02796573. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.1186/s13063-016-1511-1.

REFINE-Lung implements a novel multi-arm randomised trial design to address possible immunotherapy overtreatment.

Increasing evidence suggests that some immunotherapy dosing regimens for patients with advanced cancer could result in overtreatment. Given the high costs of these agents, and important implications for quality of life and toxicity, new approaches are needed to identify and reduce unnecessary treatment. Conventional two-arm non-inferiority designs are inefficient in this context because they require large numbers of patients to explore a single alternative to the standard of care. Here, we discuss the potential problem of overtreatment with anti-PD-1 directed agents in general and introduce REFINE-Lung (NCT05085028), a UK multicentre phase 3 study of reduced frequency pembrolizumab in advanced non-small-cell lung cancer. REFINE-Lung uses a novel multi-arm multi-stage response over continuous interventions (MAMS-ROCI) design to determine the optimal dose frequency of pembrolizumab. Along with a similarly designed basket study of patients with renal cancer and melanoma, REFINE-Lung and the MAMS-ROCI design could contribute to practice-changing advances in patient care and form a template for future immunotherapy optimisation studies across cancer types and indications. This new trial design is applicable to many new or existing agents for which optimisation of dose, frequency, or duration of therapy is desirable.

Practical approaches to Bayesian sample size determination in non-inferiority trials with binary outcomes.

Bayesian analysis of a non-inferiority trial is advantageous in allowing direct probability statements to be made about the relative treatment difference rather than relying on an arbitrary and often poorly justified non-inferiority margin. When the primary analysis will be Bayesian, a Bayesian approach to sample size determination will often be appropriate for consistency with the analysis. We demonstrate three Bayesian approaches to choosing sample size for non-inferiority trials with binary outcomes and review their advantages and disadvantages. First, we present a predictive power approach for determining sample size using the probability that the trial will produce a convincing result in the final analysis. Next, we determine sample size by considering the expected posterior probability of non-inferiority in the trial. Finally, we demonstrate a precision-based approach. We apply these methods to a non-inferiority trial in antiretroviral therapy for treatment of HIV-infected children. A predictive power approach would be most accessible in practical settings, because it is analogous to the standard frequentist approach. Sample sizes are larger than with frequentist calculations unless an informative analysis prior is specified, because appropriate allowance is made for uncertainty in the assumed design parameters, ignored in frequentist calculations. An expected posterior probability approach will lead to a smaller sample size and is appropriate when the focus is on estimating posterior probability rather than on testing. A precision-based approach would be useful when sample size is restricted by limits on recruitment or costs, but it would be difficult to decide on sample size using this approach alone.

A comparison of different population-level summary measures for randomised trials with time-to-event outcomes, with a focus on non-inferiority trials.

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

Substantive model compatible multilevel multiple imputation: A joint modeling approach.

BACKGROUND: Substantive model compatible multiple imputation (SMC-MI) is a relatively novel imputation method that is particularly useful when the analyst's model includes interactions, non-linearities, and/or partially observed random slope variables. METHODS: Here we thoroughly investigate a SMC-MI strategy based on joint modeling of the covariates of the analysis model. We provide code to apply the proposed strategy and we perform an extensive simulation work to test it in various circumstances. We explore the impact on the results of various factors, including whether the missing data are at the individual or cluster level, whether there are non-linearities and whether the imputation model is correctly specified. Finally, we apply the imputation methods to the motivating example data. RESULTS: SMC-JM appears to be superior to standard JM imputation, particularly in presence of large variation in random slopes, non-linearities, and interactions. Results seem to be robust to slight mis-specification of the imputation model for the covariates. When imputing level 2 data, enough clusters have to be observed in order to obtain unbiased estimates of the level 2 parameters. CONCLUSIONS: SMC-JM is preferable to standard JM imputation in presence of complexities in the analysis model of interest, such as non-linearities or random slopes.

Choosing and changing the analysis scale in non-inferiority trials with a binary outcome.

BACKGROUND: The size of the margin strongly influences the required sample size in non-inferiority and equivalence trials. What is sometimes ignored, however, is that for trials with binary outcomes, the scale of the margin - risk difference, risk ratio or odds ratio - also has a large impact on power and thus on sample size requirement. When considering several scales at the design stage of a trial, these sample size consequences should be taken into account. Sometimes, changing the scale may be needed at a later stage of a trial, for example, when the event proportion in the control arm turns out different from expected. Also after completion of a trial, a switch to another scale is sometimes made, for example, when using a regression model in a secondary analysis or when combining study results in a meta-analysis that requires unifying scales. The exact consequences of such switches are currently unknown. METHODS AND RESULTS: This article first outlines sample size consequences for different choices of analysis scale at the design stage of a trial. We add a new result on sample size requirement comparing the risk difference scale with the risk ratio scale. Then, we study two different approaches to changing the analysis scale after the trial has commenced: (1) mapping the original non-inferiority margin using the event proportion in the control arm that was anticipated at the design stage or (2) mapping the original non-inferiority margin using the observed event proportion in the control arm. We use simulations to illustrate consequences on type I and type II error rates. Methods are illustrated on the INES trial, a non-inferiority trial that compared single birth rates in subfertile couples after different fertility treatments. Our results demonstrate large differences in required sample size when choosing between risk difference, risk ratio and odds ratio scales at the design stage of non-inferiority trials. In some cases, the sample size requirement is twice as large on one scale compared with another. Changing the scale after commencing the trial using anticipated proportions mainly impacts type II error rate, whereas switching using observed proportions is not advised due to not maintaining type I error rate. Differences were more pronounced with larger margins. CONCLUSIONS: Trialists should be aware that the analysis scale can have large impact on type I and type II error rates in non-inferiority trials.

Investigating the inequalities in route to diagnosis amongst patients with diffuse large B-cell or follicular lymphoma in England.

INTRODUCTION: Diagnostic delay is associated with lower chances of cancer survival. Underlying comorbidities are known to affect the timely diagnosis of cancer. Diffuse large B-cell (DLBCL) and follicular lymphomas (FL) are primarily diagnosed amongst older patients, who are more likely to have comorbidities. Characteristics of clinical commissioning groups (CCG) are also known to impact diagnostic delay. We assess the association between comorbidities and diagnostic delay amongst patients with DLBCL or FL in England during 2005-2013. METHODS: Multivariable generalised linear mixed-effect models were used to assess the main association. Empirical Bayes estimates of the random effects were used to explore between-cluster variation. The latent normal joint modelling multiple imputation approach was used to account for partially observed variables. RESULTS: We included 30,078 and 15,551 patients diagnosed with DLBCL or FL, respectively. Amongst patients from the same CCG, having multimorbidity was strongly associated with the emergency route to diagnosis (DLBCL: odds ratio 1.56, CI 1.40-1.73; FL: odds ratio 1.80, CI 1.45-2.23). Amongst DLBCL patients, the diagnostic delay was possibly correlated with CCGs that had higher population densities. CONCLUSIONS: Underlying comorbidity is associated with diagnostic delay amongst patients with DLBCL or FL. Results suggest a possible correlation between CCGs with higher population densities and diagnostic delay of aggressive lymphomas.

Major Depression and Survival in People With Cancer.

OBJECTIVE: The question of whether depression is associated with worse survival in people with cancer remains unanswered because of methodological criticism of the published research on the topic. We aimed to study the association in a large methodologically robust study. METHODS: We analyzed data on 20,582 patients with breast, colorectal, gynecological, lung, and prostate cancers who had attended cancer outpatient clinics in Scotland, United Kingdom. Patients had completed two-stage screening for major depression as part of their cancer care. These data on depression status were linked to demographic, cancer, and subsequent mortality data from national databases. We estimated the association of major depression with survival for each cancer using Cox regression. We adjusted for potential confounders and interactions between potentially time-varying confounders and the interval between cancer diagnosis and depression screening, and used multiple imputation for missing depression and confounder data. We pooled the cancer-specific results using fixed-effects meta-analysis. RESULTS: Major depression was associated with worse survival for all cancers, with similar adjusted hazard ratios (HRs): breast cancer (HR = 1.42, 95% confidence interval [CI] = 1.15-1.75), colorectal cancer (HR = 1.47, 95% CI = 1.11-1.94), gynecological cancer (HR = 1.36, 95% CI = 1.08-1.71), lung cancer (HR = 1.39, 95% CI = 1.24-1.56), and prostate cancer (HR = 1.76, 95% CI = 1.08-2.85). The pooled HR was 1.41 (95% CI = 1.29-1.54, p < .001, I2 = 0%). These findings were not materially different when we only considered the deaths (90%) that were attributed to cancer. CONCLUSIONS: Major depression is associated with worse survival in patients with common cancers. The mechanisms of this association and the clinical implications require further study.

Systematic evaluation and external validation of 22 prognostic models among hospitalised adults with COVID-19: an observational cohort study.

The number of proposed prognostic models for coronavirus disease 2019 (COVID-19) is growing rapidly, but it is unknown whether any are suitable for widespread clinical implementation.We independently externally validated the performance of candidate prognostic models, identified through a living systematic review, among consecutive adults admitted to hospital with a final diagnosis of COVID-19. We reconstructed candidate models as per original descriptions and evaluated performance for their original intended outcomes using predictors measured at the time of admission. We assessed discrimination, calibration and net benefit, compared to the default strategies of treating all and no patients, and against the most discriminating predictors in univariable analyses.We tested 22 candidate prognostic models among 411 participants with COVID-19, of whom 180 (43.8%) and 115 (28.0%) met the endpoints of clinical deterioration and mortality, respectively. Highest areas under receiver operating characteristic (AUROC) curves were achieved by the NEWS2 score for prediction of deterioration over 24 h (0.78, 95% CI 0.73-0.83), and a novel model for prediction of deterioration <14 days from admission (0.78, 95% CI 0.74-0.82). The most discriminating univariable predictors were admission oxygen saturation on room air for in-hospital deterioration (AUROC 0.76, 95% CI 0.71-0.81), and age for in-hospital mortality (AUROC 0.76, 95% CI 0.71-0.81). No prognostic model demonstrated consistently higher net benefit than these univariable predictors, across a range of threshold probabilities.Admission oxygen saturation on room air and patient age are strong predictors of deterioration and mortality among hospitalised adults with COVID-19, respectively. None of the prognostic models evaluated here offered incremental value for patient stratification to these univariable predictors.

The DURATIONS randomised trial design: Estimation targets, analysis methods and operating characteristics.

BACKGROUND: Designing trials to reduce treatment duration is important in several therapeutic areas, including tuberculosis and bacterial infections. We recently proposed a new randomised trial design to overcome some of the limitations of standard two-arm non-inferiority trials. This DURATIONS design involves randomising patients to a number of duration arms and modelling the so-called 'duration-response curve'. This article investigates the operating characteristics (type-1 and type-2 errors) of different statistical methods of drawing inference from the estimated curve. METHODS: Our first estimation target is the shortest duration non-inferior to the control (maximum) duration within a specific risk difference margin. We compare different methods of estimating this quantity, including using model confidence bands, the delta method and bootstrap. We then explore the generalisability of results to estimation targets which focus on absolute event rates, risk ratio and gradient of the curve. RESULTS: We show through simulations that, in most scenarios and for most of the estimation targets, using the bootstrap to estimate variability around the target duration leads to good results for DURATIONS design-appropriate quantities analogous to power and type-1 error. Using model confidence bands is not recommended, while the delta method leads to inflated type-1 error in some scenarios, particularly when the optimal duration is very close to one of the randomised durations. CONCLUSIONS: Using the bootstrap to estimate the optimal duration in a DURATIONS design has good operating characteristics in a wide range of scenarios and can be used with confidence by researchers wishing to design a DURATIONS trial to reduce treatment duration. Uncertainty around several different targets can be estimated with this bootstrap approach.

Multiple imputation methods for handling incomplete longitudinal and clustered data where the target analysis is a linear mixed effects model.

Multiple imputation (MI) is increasingly popular for handling multivariate missing data. Two general approaches are available in standard computer packages: MI based on the posterior distribution of incomplete variables under a multivariate (joint) model, and fully conditional specification (FCS), which imputes missing values using univariate conditional distributions for each incomplete variable given all the others, cycling iteratively through the univariate imputation models. In the context of longitudinal or clustered data, it is not clear whether these approaches result in consistent estimates of regression coefficient and variance component parameters when the analysis model of interest is a linear mixed effects model (LMM) that includes both random intercepts and slopes with either covariates or both covariates and outcome contain missing information. In the current paper, we compared the performance of seven different MI methods for handling missing values in longitudinal and clustered data in the context of fitting LMMs with both random intercepts and slopes. We study the theoretical compatibility between specific imputation models fitted under each of these approaches and the LMM, and also conduct simulation studies in both the longitudinal and clustered data settings. Simulations were motivated by analyses of the association between body mass index (BMI) and quality of life (QoL) in the Longitudinal Study of Australian Children (LSAC). Our findings showed that the relative performance of MI methods vary according to whether the incomplete covariate has fixed or random effects and whether there is missingnesss in the outcome variable. We showed that compatible imputation and analysis models resulted in consistent estimation of both regression parameters and variance components via simulation. We illustrate our findings with the analysis of LSAC data.

Longitudinal associations between perceptions of the neighbourhood environment and physical activity in adolescents: evidence from the Olympic Regeneration in East London (ORiEL) study.

BACKGROUND: Most UK adolescents do not achieve recommended levels of physical activity. Previous studies suggested that perceptions of the neighbourhood environment could contribute to explain differences in physical activity behaviours. We aimed to examine whether five measures of perceptions - perceived bus stop proximity, traffic safety, street connectivity, enjoyment of the neighbourhood for walking/cycling, and personal safety - were longitudinally associated with common forms of physical activity, namely walking to school, walking for leisure, and a composite measure of outdoor physical activity. We further aimed to investigate the moderating role of gender. METHODS: We used longitudinal data from the Olympic Regeneration in East London (ORiEL) study, a prospective cohort study. In 2012, 3106 adolescents aged 11 to 12 were recruited from 25 schools in 4 deprived boroughs of East London. Adolescents were followed-up in 2013 and 2014. The final sample includes 2260 adolescents surveyed at three occasions. We estimated logistic regression models using Generalised Estimating Equations to test the plausibility of hypotheses on the nature of the longitudinal associations (general association, cumulative effect, co-varying trajectories), adjusting for potential confounders. Item non-response was handled using multiple imputation. RESULTS: Longitudinal analyses indicate little evidence that perceptions of the neighbourhood are important predictors of younger adolescent physical activity. There was weak evidence that greater perceived proximity to bus stops is associated with a small decrease in the probability of walking for leisure. Results also indicate that poorer perception of personal safety decreases the probability of walking for leisure. There was some indication that better perception of street connectivity is associated with more outdoor physical activity. Finally, we found very little evidence that the associations between perceptions of the neighbourhood and physical activity differed by gender. CONCLUSIONS: This study suggests that younger adolescents' perceptions of their neighbourhood environment, and changes in these perceptions, did not consistently predict physical activity in a deprived and ethnically diverse urban population. Future studies should use situation-specific measures of the neighbourhood environment and physical activity to better capture the hypothesised processes and explore the relative roles of the objective environment, parental and adolescents' perceptions in examining differences in types of physical activity.

Multiple imputation for discrete data: Evaluation of the joint latent normal model.

Missing data are ubiquitous in clinical and social research, and multiple imputation (MI) is increasingly the methodology of choice for practitioners. Two principal strategies for imputation have been proposed in the literature: joint modelling multiple imputation (JM-MI) and full conditional specification multiple imputation (FCS-MI). While JM-MI is arguably a preferable approach, because it involves specification of an explicit imputation model, FCS-MI is pragmatically appealing, because of its flexibility in handling different types of variables. JM-MI has developed from the multivariate normal model, and latent normal variables have been proposed as a natural way to extend this model to handle categorical variables. In this article, we evaluate the latent normal model through an extensive simulation study and an application on data from the German Breast Cancer Study Group, comparing the results with FCS-MI. We divide our investigation in four sections, focusing on (i) binary, (ii) categorical, (iii) ordinal, and (iv) count data. Using data simulated from both the latent normal model and the general location model, we find that in all but one extreme general location model setting JM-MI works very well, and sometimes outperforms FCS-MI. We conclude the latent normal model, implemented in the R package jomo, can be used with confidence by researchers, both for single and multilevel multiple imputation.

Rethinking non-inferiority: a practical trial design for optimising treatment duration.

Background Trials to identify the minimal effective treatment duration are needed in different therapeutic areas, including bacterial infections, tuberculosis and hepatitis C. However, standard non-inferiority designs have several limitations, including arbitrariness of non-inferiority margins, choice of research arms and very large sample sizes. Methods We recast the problem of finding an appropriate non-inferior treatment duration in terms of modelling the entire duration-response curve within a pre-specified range. We propose a multi-arm randomised trial design, allocating patients to different treatment durations. We use fractional polynomials and spline-based methods to flexibly model the duration-response curve. We call this a 'Durations design'. We compare different methods in terms of a scaled version of the area between true and estimated prediction curves. We evaluate sensitivity to key design parameters, including sample size, number and position of arms. Results A total sample size of ~ 500 patients divided into a moderate number of equidistant arms (5-7) is sufficient to estimate the duration-response curve within a 5% error margin in 95% of the simulations. Fractional polynomials provide similar or better results than spline-based methods in most scenarios. Conclusion Our proposed practical randomised trial 'Durations design' shows promising performance in the estimation of the duration-response curve; subject to a pending careful investigation of its inferential properties, it provides a potential alternative to standard non-inferiority designs, avoiding many of their limitations, and yet being fairly robust to different possible duration-response curves. The trial outcome is the whole duration-response curve, which may be used by clinicians and policymakers to make informed decisions, facilitating a move away from a forced binary hypothesis testing paradigm.

How to check a simulation study.

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

Treatment preferences among people at risk of developing tuberculosis: A discrete choice experiment.

Diagnosing and treating people with bacteriologically-negative but radiologically-apparent tuberculosis (TB) may contribute to more effective TB care and reduce transmission. However, optimal treatment approaches for this group are unknown. It is important to understand peoples' preferences of treatment options for effective programmatic implementation of people-centred treatment approaches. We designed and implemented a discrete choice experiment (DCE) to solicit treatment preferences among adults (≥18 years) with TB symptoms attending a primary health clinic in Blantyre, Malawi. Treatment attributes included in the DCE were as follows: duration of treatment; number of tablets per dose; reduction in the risk of being unwell with TB disease; likelihood of infecting others; adverse effects from the treatment; frequency of follow up; and the annual travel cost to access care. Quantitative choice modelling with multinomial logit models estimated through frequentist and Bayesian approaches investigated preferences for the management of bacteriologically-negative, but radiographically-apparent TB. 128 participants were recruited (57% male, 43.8% HIV-positive, 8.6% previously treated for TB). Participants preferred to take any treatment compared to not taking treatment (odds ratio [OR] 5.78; 95% confidence interval [CI]: 2.40, 13.90). Treatments that reduced the relative risk of developing TB disease by 80% were preferred (OR: 2.97; 95% CI: 2.09, 4.21) compared to treatments that lead to a lower reduction in risk of 50%. However, there was no evidence for treatments that are 95% effective being preferred over those that are 80% effective. Participants strongly favoured the treatments that could completely stop transmission (OR: 7.87, 95% CI: 5.71, 10.84), and prioritised avoiding side effects (OR: 0.19, 95% CI: 0.12, 0.29). There was no evidence of an interaction between perceived TB disease risk and treatment preferences. In summary, participants were primarily concerned with the effectiveness of TB treatments and strongly preferred treatments that removed the risk of onward transmission. Person-centred approaches of preferences for treatment should be considered when designing new treatment strategies. Understanding treatment preferences will ensure that any recommended treatment for probable early TB disease is well accepted and utilized by the public.