Nonparametric association analysis of bivariate left‐truncated competing risks data

We develop time‐varying association analyses for onset ages of two lung infections to address the statistical challenges in utilizing registry data where onset ages are left‐truncated by ages of entry and competing‐risk censored by deaths. Two types of association estimators are proposed based on conditional cause‐specific hazard function and cumulative incidence function that are adapted from unconditional quantities to handle left truncation. Asymptotic properties of the estimators are established by using the empirical process techniques. Our simulation study shows that the estimators perform well with moderate sample sizes. We apply our methods to the Cystic Fibrosis Foundation Registry data to study the relationship between onset ages of Pseudomonas aeruginosa and Staphylococcus aureus infections.     

Competing Risks and Time‐Dependent Covariates

Time‐dependent covariates are frequently encountered in regression analysis for event history data and competing risks. They are often essential predictors, which cannot be substituted by time‐fixed covariates. This study briefly recalls the different types of time‐dependent covariates, as classified by Kalbfleisch and Prentice [The Statistical Analysis of Failure Time Data, Wiley, New York, 2002] with the intent of clarifying their role and emphasizing the limitations in standard survival models and in the competing risks setting. If random (internal) time‐dependent covariates are to be included in the modeling process, then it is still possible to estimate cause‐specific hazards but prediction of the cumulative incidences and survival probabilities based on these is no longer feasible. This article aims at providing some possible strategies for dealing with these prediction problems. In a multi‐state framework, a first approach uses internal covariates to define additional (intermediate) transient states in the competing risks model. Another approach is to apply the landmark analysis as described by van Houwelingen [Scandinavian Journal of Statistics 2007, 34 , 70–85] in order to study cumulative incidences at different subintervals of the entire study period. The final strategy is to extend the competing risks model by considering all the possible combinations between internal covariate levels and cause‐specific events as final states. In all of those proposals, it is possible to estimate the changes/differences of the cumulative risks associated with simple internal covariates. An illustrative example based on bone marrow transplant data is presented in order to compare the different methods.     

Identification of a correlation between Helicobacter pylori infection and Graves' disease

Background: viral and bacterial antigens have been suspected to be able to mimic the antigenic profile of the thyroid cell membrane and to play an important role in the onset of the autoimmune diseases, such as Graves’ disease and Hashimoto thyroiditis. The Helicobacter pylori infection is worldwide diffused and is present in the developed countries up to 50% of the population. The presence of the cytotoxin‐associated gene A antigens identifies the most virulent strains of the bacterium. Previous studies have demonstrated the possible correlation between the Helicobacter pylori and Hashimoto’s thyroiditis but these results are controversial. Aims: We studied the prevalence rate of this bacterium in the Graves’ disease and two selected subgroups such as the hyperthyroid patients, at the first time of diagnosis, and the euthyroid methimazole‐treated patients. Materials and Methods: We analyzed Helicobacter pylori in fresh stool samples with an enzyme immunoassay method and the presence of cytotoxin‐associated gene A antigens with a serological test. Results: Our results show that a significative increased rate of prevalence is present in Graves' patients, when the disease is ongoing, with an overall prevalence of the strains expressing the cytotoxin‐associated gene A antigens compared to the control group. Conclusions: The association between the Helicobacter pylori and Graves’ disease suggests a possible role of this bacterium in the onset and/or the maintenance of the disease.     

Joint Modeling of Multivariate Longitudinal Data and Competing Risks Using Multiphase Sub-models

In many clinical studies that involve follow-up, it is common to observe one or more sequences of longitudinal measurements, as well as one or more time to event outcomes. A competing risks situation arises when the probability of occurrence of one event is altered/hindered by another time to event. Recently, there has been much attention paid to the joint analysis of a single longitudinal response and a single time to event outcome, when the missing data mechanism in the longitudinal process is non-ignorable. We, in this paper, propose an extension where multiple longitudinal responses are jointly modeled with competing risks (multiple time to events). Our shared parameter joint model consists of a system of multiphase non-linear mixed effects sub-models for the multiple longitudinal responses, and a system of cause-specific non-proportional hazards frailty sub-models for competing risks, with associations among multiple longitudinal responses and competing risks modeled using latent parameters. The joint model is applied to a data set of patients who are on mechanical circulatory support and are awaiting heart transplant, using readily available software. While on the mechanical circulatory support, patient liver and renal functions may worsen and these in turn may influence one of the two possible competing outcomes: (i) death before transplant; (ii) transplant. In one application, we propose a system of multiphase cause-specific non-proportional hazard sub-model where frailty can be time varying. Performance under different scenarios was assessed using simulation studies. By using the proposed joint modeling of the multiphase sub-models, one can identify: (i) non-linear trends in multiple longitudinal outcomes; (ii) time-varying hazards and cumulative incidence functions of the competing risks; (iii) identify risk factors for the both types of outcomes, where the effect may or may not change with time; and (iv) assess the association between multiple longitudinal and competing risks outcomes, where the association may or may not change with time.     

Time‐Dependent Predictive Accuracy in the Presence of Competing Risks

Summary Competing risks arise naturally in time‐to‐event studies. In this article, we propose time‐dependent accuracy measures for a marker when we have censored survival times and competing risks. Time‐dependent versions of sensitivity or true positive (TP) fraction naturally correspond to consideration of either cumulative (or prevalent) cases that accrue over a fixed time period, or alternatively to incident cases that are observed among event‐free subjects at any select time. Time‐dependent (dynamic) specificity (1–false positive (FP)) can be based on the marker distribution among event‐free subjects. We extend these definitions to incorporate cause of failure for competing risks outcomes. The proposed estimation for cause‐specific cumulative TP/dynamic FP is based on the nearest neighbor estimation of bivariate distribution function of the marker and the event time. On the other hand, incident TP/dynamic FP can be estimated using a possibly nonproportional hazards Cox model for the cause‐specific hazards and riskset reweighting of the marker distribution. The proposed methods extend the time‐dependent predictive accuracy measures of Heagerty, Lumley, and Pepe (2000, Biometrics 56, 337–344) and Heagerty and Zheng (2005, Biometrics 61, 92–105).     

Disentangling facilitation and seed dispersal from environmental heterogeneity as mechanisms generating associations between savanna plants

Question: How can we disentangle facilitation and seed dispersal from environmental heterogeneity as mechanisms causing spatial associations of plant species? Location: Semi‐arid savanna in the Kimberley Thorn Bushveld, South Africa. Methods: We developed a two‐step protocol for the statistical differentiation of association‐promoting mechanisms in plants based on the Acacia erioloba–Grewia flava association. Individuals of the savanna shrub G. flava and the tree A. erioloba were mapped on four study plots. Disentangling the mechanism causing the association of G. flava and A. erioloba involved tests of three spatial and one non‐spatial null model. The spatial null models include homogeneous and heterogeneous Poisson processes for spatial randomness based on the bivariate spatial point patterns of the four plots. With the non‐spatial analysis, we determined the relationship between the canopy diameter of A. erioloba trees and presence or absence of G. flava shrubs in the tree understorey to find whether shrub presence requires a minimum tree canopy diameter. Results: We first showed a significant positive spatial association of the two species. Thereafter, the non‐spatial analysis supported an exclusion of environmental heterogeneity as the sole cause of this positive association. We found a minimum tree size under which no G. flava shrubs occurred. Conclusions: Our two‐step analysis showed that it is unlikely that heterogeneous environmental conditions caused the spatial association of A. erioloba and G. flava. Instead, this association may have been caused by seed dispersal and/or facilitation (e.g. caused by hydraulic lift and/or nitrogen fixation by the host tree).     

Estimating Subject‐Specific Dependent Competing Risk Profile with Censored Event Time Observations

Summary In a longitudinal study, suppose that the primary endpoint is the time to a specific event. This response variable, however, may be censored by an independent censoring variable or by the occurrence of one of several dependent competing events. For each study subject, a set of baseline covariates is collected. The question is how to construct a reliable prediction rule for the future subject's profile of all competing risks of interest at a specific time point for risk‐benefit decision making. In this article, we propose a two‐stage procedure to make inferences about such subject‐specific profiles. For the first step, we use a parametric model to obtain a univariate risk index score system. We then estimate consistently the average competing risks for subjects who have the same parametric index score via a nonparametric function estimation procedure. We illustrate this new proposal with the data from a randomized clinical trial for evaluating the efficacy of a treatment for prostate cancer. The primary endpoint for this study was the time to prostate cancer death, but had two types of dependent competing events, one from cardiovascular death and the other from death of other causes.     

Fine‐scale spatial and temporal population genetics of Aedes japonicus,a new US mosquito,reveal multiple introductions

The newly introduced mosquito Aedes japonicus has expanded from its original range in Northeastern Asia to 29 US states (including Hawaii) plus Canada and northern Europe. Our objectives were to test an earlier hypothesis of multiple introductions of this species to the Northeastern US and evaluate putative temporal changes in genetic makeup. Using a panel of seven microsatellite loci, we confirmed the existence of two abundant genetic forms in specimens originally collected in 1999–2000 (F_ST value based on microsatellite data = 0.26) that matches the disjunctive distribution of mitochondrial haplotypes. To examine the distribution of the two genetic ‘types’ across Pennsylvania we created a fine‐scale genetic map of Ae. japonicus using 439 specimens collected from 54 Pennsylvania counties in 2002–2003. We also made direct comparisons between collections in 1999–2000 and new collections made in 2004–2005 obtained from the same areas in the northeastern US. We observed that the strong association between mtDNA haplotype and microsatellite signature seen in 1999–2000 had weakened significantly by 2002 across Pennsylvania, a trend continued to some extent in 2004–2005 in PA, NJ, and NY, indicating that once easily distinguishable separate introductions are merging. The two expanding genetic forms create a complex correlation between spatial and genetic distances. The existence of multiple introductions would be obscured without sampling early and across time with highly polymorphic molecular markers. Our results provide a high‐resolution analysis of the spatial and temporal dynamics of a newly introduced disease vector and argue that successive introductions may be a common pattern for invasive mosquitoes.     

Prevalence of the fibrinogen β‐chain,angiotensin‐converting enzyme and plasminogen activator inhibitor‐1 polymorphisms in Costa Rican young adults with thrombotic disease

Thrombotic disease is a multifactorial condition that involves both classical and genetic risk factors. We studied the association between the classical risk factors of hypertension and smoking, and polymorphisms on the genes of the angiotensin‐converting enzyme (ACE), the β‐chain of fibrinogen (FG), and the plasminogen activator inhibitor‐1 (PAI‐1) in patients with venous and arterial thrombosis. The present investigation is a retrospective case–control study. A total of 340 participants were analyzed, including 162 patients and 178 healthy controls. Hypertension and smoking showed a significant association with thrombotic disease (p < 0.05) but FG level was found significant risk factor only for the venous thrombosis (VT) group (p < 0.04). Significant differences between thrombotic groups were found for the studied polymorphisms of PAI‐1 (p < 0.0014), but for both FG β‐chain gene polymorphisms, none of the molecular analyses showed a positive sample for any mutating allele (p > 0.05). For the ACE polymorphism, the I allele present a protective effect in the general thrombotic group. This is one of the first reports in a Latin‐American population dealing with these molecular markers and thrombotic diseases. Copyright © 2010 John Wiley & Sons, Ltd.     

A bivariate cure-mixture approach for modeling familial association in diseases

For modeling correlation in familial diseases with variable ages at onset, we propose a bivariate model that incorporates two types of pairwise association, one between the lifetime risk or the overall susceptibility of two individuals and one between the ages at onset between two susceptible individuals. For estimation, we consider a two-stage estimation procedure similar to that of Shih (1998, Biometrics 54, 1115-1128). We evaluate the properties of the estimators through simulations and compare the performance with that from a bivariate survival model that allows correlation between ages at onset only. We apply the methodology to breast cancer using the kinship data from the Washington Ashkenazi Study. We also discuss potential applications of the proposed method in the area of cure modeling.     

Reverse causality and confounding and the associations of overweight and obesity with mortality

Objective: To examine the effect of reverse causality and confounding on the association of BMI with all‐cause and cause‐specific mortality. Research Methods and Procedures: Data from two large prospective studies were used. One (a community‐based cohort) included 8327 women and 7017 men who resided in two Scottish towns at the time of the baseline assessment in 1972–1976; the other (an occupational cohort) included 4016 men working in the central belt of Scotland at the time of the baseline assessment in 1970–1973. Participants in both cohorts were ages 45 to 64 years at baseline; the follow‐up period was 28 to 34 years. Results: In age‐adjusted analyses that did not take account of reverse causality or smoking, there was no association between being overweight (BMI 25 to <30 kg/m²) and mortality, and weak to modest associations between obesity (BMI ≥30 kg/m²) and mortality. There was a strong association between smoking and lower BMI in women and men in both cohorts (all p < 0.0001). Among never‐smokers and with the first 5 years of deaths removed, overweight was associated with an increase in all‐cause mortality (relative risk ranging from 1.12 to 1.38), and obesity was associated with a doubling of risk in men in both cohorts (relative risk, 2.10 and 1.96, respectively) and a 60% increase in women (relative risk, 1.56). In both never‐smokers and current smokers, being overweight or obese was associated with important increases in the risk of cardiovascular disease. Discussion: These findings demonstrate that with appropriate control for smoking and reverse causality, both overweight and obesity are associated with important increases in all‐cause and cause‐specific mortality, and in particular with cardiovascular disease mortality.     

Age and Growth Patterns of Brazil Nut Trees (Bertholletia excelsa Bonpl.) in Amazonia,Brazil

Various techniques have been used to estimate the age of Brazil nut trees (Bertholletia excelsa Bonpl.), but these techniques produce large discrepancies. Here, we first verified that two individuals of known ages from a plantation in central Amazonia, Brazil, have a congruent number of growth rings. The indexed average tree‐ring curve was significantly correlated with total precipitation during the rainy season (November–June) over a 50‐yr period, confirming the annual nature of the tree rings. Second, we analyzed Brazil nut trees from two populations in the Trombetas (eastern Amazon) and Purus (central Amazon) regions, performing tree‐ring analysis to estimate tree age and diameter increment rates. We compared age–diameter relationships, mean passage time through 10‐cm diameter size classes, and growth trajectories of individual trees. The maximum age of Brazil nut trees analyzed was 361 yr in the Purus and 401 yr in the Trombetas. Trees at the Purus site had higher mean diameter increment rates and showed more variation compared to trees at the Trombetas site. Individual growth trajectories show that the majority of trees attained the canopy by direct growth, while a smaller number passed through one release or one suppression event before becoming established in the canopy. None of the trees passed through multiple release and suppression events. The age estimations presented here are comparable to previous tree‐ring analyses for the species, and the observed growth patterns support earlier work indicating B. excelsa as a gap‐dependent tree species.     

Circadian variation in clinical features and outcome of intracerebral hemorrhage: The INTERACT studies

Previous studies consistently reported a diurnal variation in the occurrence of intracerebral hemorrhage (ICH), with a morning peak. However, limited knowledge exists on the circadian pattern of ICH severity and outcome. This study aimed to determine possible associations between ICH onset time and admission severity and 90-day outcomes using the combined data set of the pilot and main-phase Intensive blood pressure (BP) reduction in an acute cerebral hemorrhage trial (INTERACT). The ICH onset time was categorized into three groups (1: 00:00–07:59; 2: 08:00–15:59; and 3: 16:00–23:59). We found an association between onset time and low Glasgow Coma Scale score: aOR (time 1: 1.72, 95% CI 1.12–2.66; time 3: 1.95, 95% CI 1.31–2.89, p = 0.003; in comparison to time 2). There was no association between onset time and volume of ICH (adjusted p = 0.354) or 90-day outcomes of death or major disability, and death and major disability separately (all adjusted p > 0.4). The results showed that more severe cases of ICH patients, defined by a reduced level of consciousness, had late afternoon to early morning stroke onset, but this was unrelated to baseline hematoma volume or location. There was no circadian influence on ICH clinical outcome.     

Family‐based genome scan for age at onset of late‐onset Alzheimer's disease in whole exome sequencing data

Alzheimer's disease (AD) is a common and complex neurodegenerative disease. Age at onset (AAO) of AD is an important component phenotype with a genetic basis, and identification of genes in which variation affects AAO would contribute to identification of factors that affect timing of onset. Increase in AAO through prevention or therapeutic measures would have enormous benefits by delaying AD and its associated morbidities. In this paper, we performed a family‐based genome‐wide association study for AAO of late‐onset AD in whole exome sequence data generated in multigenerational families with multiple AD cases. We conducted single marker and gene‐based burden tests for common and rare variants, respectively. We combined association analyses with variance component linkage analysis, and with reference to prior studies, in order to enhance evidence of the identified genes. For variants and genes implicated by the association study, we performed a gene‐set enrichment analysis to identify potential novel pathways associated with AAO of AD. We found statistically significant association with AAO for three genes (WRN, NTN4 and LAMC3) with common associated variants, and for four genes (SLC8A3, SLC19A3, MADD and LRRK2) with multiple rare‐associated variants that have a plausible biological function related to AD. The genes we have identified are in pathways that are strong candidates for involvement in the development of AD pathology and may lead to a better understanding of AD pathogenesis.     

Carnivoran resource and habitat use in the context of a Late Miocene faunal turnover episode

We investigate resource and habitat use by apex predators through stable isotope analysis at two Spanish Late Miocene localities: Los Valles de Fuentidueña (~9.6 Ma, LVF) and Cerro de los Batallones (~9.1 Ma, BAT). The temporal window represented by LVF and BAT was crucial in the shaping of the current Iberian mammalian structure because it corresponds to the initial stages of a faunal turnover episode and regional environmental change at ~9.5–8.5 Ma (Vallesian–Turolian transition), associated with an increase in the seasonality of precipitation. Herbivore and carnivore δ¹³C and δ¹⁸O values do not point to significant changes in either the vegetation cover (a woodland to mesic C₃ grassland) or the hydrological regime during the time lapse represented between LVF and BAT. This suggests that the environmental shift recorded around the Vallesian–Turolian boundary may have occurred later in time, since LVF and BAT ages are synchronic with the onset of the turnover event. From the standpoint of predator–prey evaluation by means of stable isotope analysis, statistical post hoc tests, mixing model output, and the assessment of niche occupation by LVF and BAT carnivores point to high levels of interspecific competition among large active carnivores, albeit some genera, such as the amphicyonid Magericyon and specially the hyaenid Lycyaena, seemed to avoid competition by taking prey from a more open habitat. Despite the drop in diversity and change in faunal components observed between the LVF and BAT assemblages, a high degree of resource and habitat competition is evident from stable isotope data.     

Bromeliads provide shelter against fire to mutualistic spiders in a fire‐prone landscape

1. A key challenge in the study of mutualistic interactions is understanding sources of variation that strengthen or weaken these interactions. In spider–plant mutualisms, spiders benefit plants by improving plant nutrition and protecting plants from herbivory. Although the benefits of plants to spider growth and survival are often claimed, they are rarely demonstrated. 2. In this study, empirical evidence is provided that bromeliads (Bromelia balansae, Bromeliaceae) are essential for the resilience of the mutualistic bromeliad‐living jumping spider populations (Psecas chapoda, Salticidae) after a fire event, sheltering spiders from the heat of the flames. 3. Spider populations were compared before and after a natural fire event and it was shown that spiders of different ages survived the fire. The survival of such individuals allowed the population of P. chapoda spiders to recover rapidly, returning to pre‐fire levels in 5 months. 4. Bromeliads reduced the susceptibility of P. chapoda spiders to burning, and this mutualistic relationship contributed to the resilience of the spider population after a fire event. It is suggested that frequent fires in fire‐prone landscapes may have strengthened this spider–plant relationship, contributing to the maintenance and evolution of this association.     

Saitohin Q7R polymorphism is associated with late‐onset Alzheimer's disease susceptibility among caucasian populations: a meta‐analysis

Saitohin (STH) Q7R polymorphism has been reported to influence the individual's susceptibility to Alzheimer's disease (AD); however, conclusions remain controversial. Therefore, we performed this meta‐analysis to explore the association between STH Q7R polymorphism and AD risk. Systematic literature searches were performed in the PubMed, Embase, Cochrane Library and Web of Science for studies published before 31 August 2016. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to assess the strength of the association using a fixed‐ or random‐effects model. Subgroup analyses, Galbraith plot and sensitivity analyses were also performed. All statistical analyses were performed with STATA Version 12.0. A total of 19 case–control studies from 17 publications with 4387 cases and 3972 controls were included in our meta‐analysis. The results showed that the Q7R polymorphism was significantly associated with an increased risk of AD in a recessive model (RR versus QQ+QR, OR = 1.27, 95% CI = 1.01–1.60, P = 0.040). After excluding the four studies not carried out in caucasians, the overall association was unchanged in all comparison models. Further subgroup analyses stratified by the time of AD onset, and the quality of included studies provided statistical evidence of significant increased risk of AD in RR versus QQ+QR model only in late‐onset subjects (OR = 1.56, 95% CI = 1.07–2.26, P = 0.021) and in studies with high quality (OR = 1.37, 95% CI = 1.01–1.86, P = 0.043). This meta‐analysis suggests that the RR genotype in saitohin Q7R polymorphism may be a human‐specific risk factor for AD, especially among late‐onset AD subjects and caucasian populations.     

Quantitative trait loci for resistance to fish pasteurellosis in gilthead sea bream (Sparus aurata)

Fish pasteurellosis is a bacterial disease causing important losses in farmed fish, including gilthead sea bream, a teleost fish of great relevance in marine aquaculture. We report in this study a QTL analysis for resistance to fish pasteurellosis in this species. An experimental population of 500 offspring originating from eight sires and six dams in a single mass‐spawning event was subjected to a disease challenge with Photobacterium damselae subsp. piscicida (Phdp), the causative agent of fish pasteurellosis. A total of 151 microsatellite loci were genotyped in the experimental population, and half‐sib regression QTL analysis was carried out on two continuous traits, body length at time of death and survival, and for two binary traits, survival at day 7 and survival at day 15, when the highest peaks of mortality were observed. Two significant QTLs were detected for disease resistance. The first one was located on linkage group LG3 affecting late survival (survival at day 15). The second one, for overall survival, was located on LG21, which allowed us to highlight a potential marker (Id13) linked to disease resistance. A significant QTL was also found for body length at death on LG6 explaining 5–8% of the phenotypic variation.     

Preliminary evidence of an effect of cerebellar volume on postural sway in FMR1 premutation males

Recent evidence suggests that early changes in postural control may be discernible among females with premutation expansions (55–200 CGG repeats) of the fragile X mental retardation 1 (FMR1) gene at risk of developing fragile X‐associated tremor ataxia syndrome (FXTAS). Cerebellar dysfunction is well described in males and females with FXTAS, yet the interrelationships between cerebellar volume, CGG repeat length, FMR1 messenger RNA (mRNA) levels and changes in postural control remain unknown. This study examined postural sway during standing in a cohort of 22 males with the FMR1 premutation (ages 26–80) and 24 matched controls (ages 26–77). The influence of cerebellar volume, CGG repeat length and FMR1 mRNA levels on postural sway was explored using multiple linear regression. The results provide preliminary evidence that increasing CGG repeat length and decreasing cerebellar volume were associated with greater postural sway among premutation males. The relationship between CGG repeat length and postural sway was mediated by a negative association between CGG repeat size and cerebellar volume. While FMR1 mRNA levels were significantly elevated in the premutation group and correlated with CGG repeat length, FMR1 mRNA levels were not significantly associated with postural sway scores. These findings show for the first time that greater postural sway among males with the FMR1 premutation may reflect CGG repeat‐mediated disruption in vulnerable cerebellar circuits implicated in postural control. However, longitudinal studies in larger samples are required to confirm whether the relationships between cerebellar volume, CGG repeat length and postural sway indicate greater risk for neurological decline.     

Improving the time-machine: estimating date of birth of grade II gliomas

Objectives: Here we present a model aiming to provide an estimate of time from tumour genesis, for grade II gliomas. The model is based on a differential equation describing the diffusion–proliferation process. We have applied our model to situations where tumour diameter was shown to increase linearly with time, with characteristic diametric velocity. Materials and methods: We have performed numerical simulations to analyse data, on patients with grade II gliomas and to extract information concerning time of tumour biological onset, as well as radiology and distribution of model parameters. Results and conclusions: We show that the estimate of tumour onset obtained from extrapolation using a constant velocity assumption, always underestimates biological tumour age, and that the correction one should add to this estimate is given roughly by 20/v (year), where v is the diametric velocity of expansion of the tumour (expressed in mm/year). Within the assumptions of the model, we have identified two types of tumour: the first corresponds to very slowly growing tumours that appear during adolescence, and the second type corresponds to slowly growing tumours that appear later, during early adulthood. That all these tumours become detectable around a mean patient age of 30 years could be interesting for formulation of strategies for early detection of tumours.