Modeling mortality risk effects of cigarettes and smokeless tobacco: results from the National Health Interview Survey Linked Mortality File Data.

BACKGROUND: Cigarettes and smokeless tobacco (SLT) products are among a wide range of tobacco products that are addictive and pose a significant health risk. In this study, we estimated smoking- and SLT use-related mortality hazard ratios (HRs) among U.S. adults by sex, age group, and cause of death, for nine mutually exclusive categories of smoking and/or SLT use. METHODS: We used data from the public-use National Health Interview Survey Linked Mortality with mortality follow-up through 2015. We used Cox proportional hazard models to estimate mortality HRs, adjusted by race/ethnicity, education, poverty level, body mass index, and tobacco-use status. RESULTS: With never users as reference group, HRs for smoking-related diseases for male exclusive current smokers aged 35-64 and 65+ were 2.18 (95% confidence interval [CI]: 1.79-2.65), and 2.45 (95% CI: 2.14-2.79), respectively. Similar significant HR estimates were found for females and for all-cause mortality (ACM) and other-cause mortality (OCM) outcomes. HRs for exclusive current SLT users were only significant for males aged 35-64 for ACM (HR: 2.04, 95% CI: 1.27-3.27) and OCM (HR: 2.80, 95% CI: 1.50-5.25). HRs for users who switched from cigarettes to SLT products were significant for males aged 65+ for smoking-related diseases (HR: 2.06, 95% CI: 1.47-2.88), SLT-related diseases (HR: 1.99, 95% CI: 1.36-2.89), and ACM (HR: 1.63, 95% CI: 1.21-2.19). CONCLUSIONS: Male exclusive current SLT users aged 35-64 had a significant HR for ACM and OCM outcomes, suggesting that deaths not attributed to SLT use could be contributing to the ACM elevated HR for exclusive current SLT users.

Pseudo maximum likelihood approach for the analysis of multivariate left-censored longitudinal data.

The linear mixed effects model based on a full likelihood is one of the few methods available to model longitudinal data subject to left censoring. However, a full likelihood approach is complicated algebraically because of the large dimension of the numeric computations, and maximum likelihood estimation can be computationally prohibitive when the data are heavily censored. Moreover, for mixed models, the complexity of the computation increases as the dimension of the random effects in the model increases. We propose a method based on pseudo likelihood that simplifies the computational complexities, allows a wide class of multivariate models, and that can be used for many different data structures including settings where the level of censoring is high. The motivation for this work comes from the need for a joint model to assess the joint effect of pro-inflammatory and anti-inflammatory biomarker data on 30-day mortality status while simultaneously accounting for longitudinal left censoring and correlation between markers in the analysis of Genetic and Inflammatory Markers for Sepsis study conducted at the University of Pittsburgh. Two markers, interleukin-6 and interleukin-10, which naturally are correlated because of a shared similar biological pathways and are left-censored because of the limited sensitivity of the assays, are considered to determine if higher levels of these markers is associated with an increased risk of death after accounting for the left censoring and their assumed correlation. Copyright © 2016 John Wiley & Sons, Ltd.

CBER's Experience With Adaptive Design Clinical Trials.

There is considerable interest among pharmaceutical and other medical product developers in adaptive clinical trials, in which knowledge learned during the course of a trial affects ongoing conduct or analysis of the trial. When the FDA released a draft Guidance document on adaptive design clinical trials in early 2010, expectations were high that it would lead to an increase in regulatory submissions involving adaptive design features, particularly for confirmatory trials. A 6-year (2008-2013) retrospective survey was performed within the Center for Biologics Evaluation and Research (CBER) at the FDA to gather information regarding the submission and evaluation of adaptive design trial proposals. We present an up-to-date summary of adaptive design proposals seen in CBER and provide an overview of our experiences. We share our concerns regarding the statistical issues and operational challenges raised during the review process for adaptive design trials. We also provide general recommendations for developing proposals for such trials. Our motivation in writing this paper was to encourage the best study design proposals to be submitted to CBER. Sometimes these can be adaptive, and sometimes a simpler design is most efficient.