Bayesian model comparison in cosmology with Population Monte Carlo

We use Bayesian model selection techniques to test extensions of the standard flat LambdaCDM paradigm. Dark-energy and curvature scenarios, and primordial perturbation models are considered. To that end, we calculate the Bayesian evidence in favour of each model using Population Monte Carlo (PMC), a new adaptive sampling technique which was recently applied in a cosmological context. The Bayesian evidence is immediately available from the PMC sample used for parameter estimation without further computational effort, and it comes with an associated error evaluation. Besides, it provides an unbiased estimator of the evidence after any fixed number of iterations and it is naturally parallelizable, in contrast with MCMC and nested sampling methods. By comparison with analytical predictions for simulated data, we show that our results obtained with PMC are reliable and robust. The variability in the evidence evaluation and the stability for various cases are estimated both from simulations and from data. For the cases we consider, the log-evidence is calculated with a precision of better than 0.08. Using a combined set of recent CMB, SNIa and BAO data, we find inconclusive evidence between flat LambdaCDM and simple dark-energy models. A curved Universe is moderately to strongly disfavoured with respect to a flat cosmology. Using physically well-motivated priors within the slow-roll approximation of inflation, we find a weak preference for a running spectral index. A Harrison-Zel'dovich spectrum is weakly disfavoured. With the current data, tensor modes are not detected; the large prior volume on the tensor-to-scalar ratio r results in moderate evidence in favour of r=0.

Short and long term mortality rates after a lower limb amputation

Objective To determine mortality rates after a first lower limb amputation and explore the rates for different subpopulations. Methods Retrospective cohort study of all people who underwent a first amputation at or proximal to transtibial level, in an area of 1.7 million people. Analysis with Kaplan-Meier curves and Log Rank tests for univariate associations of psycho-social and health variables. Logistic regression for odds of death at 30-days, 1-year and 5-years. Results 299 people were included. Median time to death was 20.3 months (95%CI: 13.1; 27.5). 30-day mortality = 22%; odds of death 2.3 times higher in people with history of cerebrovascular disease (95%CI: 1.2; 4.7, P = 0.016). 1 year mortality = 44%; odds of death 3.5 times higher for people with renal disease (95%CI: 1.8; 7.0, P < 0.001). 5-years mortality = 77%; odds of death 5.4 times higher for people with renal disease (95%CI: 1.8; 16.0,P = 0.003). Variation in mortality rates was most apparent in different age groups; people 75–84 years having better short term outcomes than those younger and older. Conclusions Mortality rates demonstrated the frailty of this population, with almost one quarter of people dying within 30-days, and almost half at 1 year. People with cerebrovascular had higher odds of death at 30 days, and those with renal disease and 1 and 5 years, respectively.

Tools for compositional data with a total

Compositional data analysis usually deals with relative information between parts where the total (abundances, mass, amount, etc.) is unknown or uninformative. This article addresses the question of what to do when the total is known and is of interest. Tools used in this case are reviewed and analysed, in particular the relationship between the positive orthant of D-dimensional real space, the product space of the real line times the D-part simplex, and their Euclidean space structures. The first alternative corresponds to data analysis taking logarithms on each component, and the second one to treat a log-transformed total jointly with a composition describing the distribution of component amounts. Real data about total abundances of phytoplankton in an Australian river motivated the present study and are used for illustration.

A mobile app offering distractions and tips to cope with cigarette craving: A qualitative study

Background Despite considerable effort, most smokers relapse within a few months after quitting due to cigarette craving. The widespread adoption of mobile phones presents new opportunities to provide support during attempts to quit. Objective To design and pilot a mobile app "DistractMe" to enable quitters to access and share distractions and tips to cope with cigarette cravings. Methods A qualitative study with 14 smokers who used DistractMe on their mobiles during the first weeks of their quit attempt. Based on interviews, diaries, and log data, we examined how the app supported quitting strategies. Results Three distinct techniques of coping when using DistractMe were identified: diversion, avoidance, and displacement. We further identified three forms of engagement with tips for coping: preparation, fortification, and confrontation. Overall, strategies to prevent cravings and their effects (avoidance, displacement, preparation, and fortification) were more common than immediate coping strategies (diversion and confrontation). Tips for coping were more commonly used than distractions to cope with cravings, because they helped to fortify the quit attempt and provided opportunities to connect with other users of the application. However, distractions were important to attract new users and to facilitate content sharing. Conclusions Based on the qualitative results, we recommend that mobile phone-based interventions focus on tips shared by peers and frequent content updates. Apps also require testing with larger groups of users to assess whether they can be self-sustaining.

Dacomitinib versus erlotinib in patients with EGFR-mutated advanced nonsmall-cell lung cancer (NSCLC): Pooled subset analyses from two randomized trials

Background: The irreversible epidermal growth factor receptor (EGFR) inhibitors have demonstrated efficacy in NSCLC patients with activating EGFR mutations, but it is unknown if they are superior to the reversible inhibitors. Dacomitinib is an oral, small-molecule irreversible inhibitor of all enzymatically active HER family tyrosine kinases. Methods: The ARCHER 1009 (NCT01360554) and A7471028 (NCT00769067) studies randomized patients with locally advanced/metastatic NSCLC following progression with one or two prior chemotherapy regimens to dacomitinib or erlotinib. EGFR mutation testing was performed centrally on archived tumor samples. We pooled patients with exon 19 deletion and L858R EGFR mutations from both studies to compare the efficacy of dacomitinib to erlotinib. Results: One hundred twenty-one patients with any EGFR mutation were enrolled; 101 had activating mutations in exon 19 or 21. For patients with exon19/21 mutations, the median progression-free survival was 14.6 months [95% confidence interval (CI) 9.0–18.2] with dacomitinib and 9.6 months (95% CI 7.4–12.7) with erlotinib [unstratified hazard ratio (HR) 0.717 (95% CI 0.458–1.124), two-sided log-rank, P = 0.146]. The median survival was 26.6 months (95% CI 21.6–41.5) with dacomitinib versus 23.2 months (95% CI 16.0–31.8) with erlotinib [unstratified HR 0.737 (95% CI 0.431–1.259), two-sided log-rank, P = 0.265]. Dacomitinib was associated with a higher incidence of diarrhea and mucositis in both studies compared with erlotinib. Conclusions: Dacomitinib is an active agent with comparable efficacy to erlotinib in the EGFR mutated patients. The subgroup with exon 19 deletion had favorable outcomes with dacomitinib. An ongoing phase III study will compare dacomitinib to gefitinib in first-line therapy of patients with NSCLC harboring common activating EGFR mutations (ARCHER 1050; NCT01774721). Clinical trials number: ARCHER 1009 (NCT01360554) and A7471028 (NCT00769067).