Prospective longitudinal study of subcortical brain volumes in individuals at high familial risk of mood disorders with or without subsequent onset of depression

Subcortical volumetric brain abnormalities have been observed in mood disorders. However, it is unknown whether these reflect adverse effects predisposing to mood disorders or emerge at illness onset. Magnetic resonance imaging was conducted at baseline and after two years in 111 initially unaffected young adults at increased risk of mood disorders because of a close family history of bipolar disorder and 93 healthy controls (HC). During the follow-up, 20 high-risk subjects developed major depressive disorder (HR-MDD), with the others remaining well (HR-well). Volumes of the lateral ventricles, caudate, putamen, pallidum, thalamus, hippocampus and amygdala were extracted for each hemisphere. Using linear mixed-effects models, differences and longitudinal changes in subcortical volumes were investigated between groups (HC, HR-MDD, HR-well). There were no significant differences for any subcortical volume between groups controlling for multiple testing. Additionally, no significant differences emerged between groups over time. Our results indicate that volumetric subcortical brain abnormalities of these regions using the current method appear not to form familial trait markers for vulnerability to mood disorders in close relatives of bipolar disorder patients over the two-year time period studied. Moreover, they do not appear to reduce in response to illness onset at least for the time period studied.

Heat stress, heat strain, and productivity in Washington State tree fruit harvesters

Thesis (Master's)--University of Washington, 2016-06

Environmental Enteropathy and Fibroblast Growth Factor 21 are associated with early childhood growth in Bangladesh

Thesis (Ph.D.)--University of Washington, 2016-06

Polysemy advantage with abstract but not concrete words

Peer reviewed

Polysemy advantage with abstract but not concrete words

Peer reviewed

Quality of life in children with advanced cancer: a report from the PediQUEST study

CONTEXT: Modifiable factors of health-related quality of life (HRQOL) are poorly described among children with advanced cancer. Symptom distress may be an important factor for intervention. OBJECTIVES: We aimed to describe patient-reported HRQOL and its relationship to symptom distress. METHODS: Prospective, longitudinal data from the multicenter Pediatric Quality of Life and Symptoms Technology study included primarily patient-reported symptom distress and HRQOL, measured at most weekly with the Memorial Symptoms Assessment Scale and Pediatric Quality of Life inventory, respectively. Associations were evaluated using linear mixed-effects models adjusting for sex, age, cancer type, intervention arm, treatment intensity, and time since disease progression. RESULTS: Of 104 enrolled patients, 49% were female, 89% were white, and median age was 12.6 years. Nine hundred and twenty surveys were completed over nine months of follow-up (84% by patients). The median total Pediatric Quality of Life score was 74 (interquartile range 63-87) and was "poor/fair" (e.g., <70) 38% of the time. "Poor/fair" categories were highest in physical (53%) and school (48%) compared to emotional (24%) and social (16%) subscores. Thirteen of 24 symptoms were independently associated with reductions in overall or domain-specific HRQOL. Patients commonly reported distress from two or more symptoms, corresponding to larger HRQOL score reductions. Neither cancer type, time since progression, treatment intensity, sex, nor age was associated with HRQOL scores in multivariable models. Among 25 children completing surveys during the last 12 weeks of life, 11 distressing symptoms were associated with reductions in HRQOL. CONCLUSION: Symptom distress is strongly associated with HRQOL. Future research should determine whether alleviating distressing symptoms improves HRQOL in children with advanced cancer.

Fluoxetine for maintenance of remission and to improve quality of life in patients with Crohn's disease: a pilot randomized placebo-controlled trial.

BACKGROUND AND AIMS: Previous studies have shown that antidepressants reduce inflammation in animal models of colitis. The present trial aimed to examine whether fluoxetine added to standard therapy for Crohn's disease [CD] maintained remission, improved quality of life [QoL] and/or mental health in people with CD as compared to placebo. METHODS: A parallel randomized double-blind placebo controlled trial was conducted. Participants with clinically established CD, with quiescent or only mild disease, were randomly assigned to receive either fluoxetine 20 mg daily or placebo, and followed for 12 months. Participants provided blood and stool samples and completed mental health and QoL questionnaires. Immune functions were assessed by stimulated cytokine secretion [CD3/CD28 stimulation] and flow cytometry for cell type. Linear mixed-effects models were used to compare groups. RESULTS: Of the 26 participants, 14 were randomized to receive fluoxetine and 12 to placebo. Overall, 14 [54%] participants were male. The mean age was 37.4 [SD=13.2] years. Fluoxetine had no effect on inflammatory bowel disease activity measured using either the Crohn's Disease Activity Index [F(3, 27.5)=0.064, p=0.978] or faecal calprotectin [F(3, 32.5)=1.08, p=0.371], but did have modest effects on immune function. There was no effect of fluoxetine on physical, psychological, social or environmental QoL, anxiety or depressive symptoms as compared to placebo [all p>0.05]. CONCLUSIONS: In this small pilot clinical trial, fluoxetine was not superior to placebo in maintaining remission or improving QoL. [ID: ACTRN12612001067864.].

Cognitive-behavioural therapy for inflammatory bowel disease: 24-month data from a randomised controlled trial

PURPOSE: There is ongoing controversy on the effectiveness of psychotherapy in inflammatory bowel disease (IBD). In the few small studies, cognitive-behavioural therapy (CBT) has been shown to alleviate symptoms of anxiety or depression. However, there is little research on the impact of CBT on physical outcomes in IBD and no studies on long-term effectiveness of CBT.

METHODS: The present two-arm pragmatic randomised controlled trial aimed to establish the impact of CBT on disease course after 24 months of observation. The study compared standard care plus CBT (+CBT) with standard care alone (SC). CBT was delivered over 10 weeks, face-to-face (F2F) or online (cCBT). The data were analysed using linear mixed-effects models.

RESULTS: CBT did not significantly influence disease activity as measured by disease activity indices at 24 months (Crohn's Disease Activity Index (CDAI), p = 0.92; Simple Clinical Colitis Activity Index (SCCAI), p = 0.88) or blood parameters (C-reactive protein (CRP), p < 0.62; haemoglobin (Hb), p = 0.77; platelet, p = 0.64; white cell count (WCC), p = 0.59) nor did CBT significantly affect mental health, coping or quality of life (all p > 0.05).

CONCLUSIONS: Therefore, we conclude that CBT does not influence the course of IBD over 24 months. Given the high rate of attrition, particularly in the CBT group, future trials should consider a personalised approach to psychotherapy, perhaps combining online and one-to-one therapist time.

Environmental controls on spatial patterns in the long-term persistence of giant kelp in central California

As marine management measures increasingly protect static areas of the oceans, it is important to make sure protected areas capture and protect persistent populations. Rocky reefs in many temperate areas worldwide serve as habitat for canopy-forming macroalgae and these structure-forming species of kelps (order Laminariales) often serve as important habitat for a great diversity of species. Macrocystis pyrifera is the most common canopy-forming kelp species found along the coast of California, but the distribution and abundance of M. pyrifera varies in space and time. The purpose of this study is to determine what environmental parameters are correlated with and their relative contribution to the spatial and temporal persistence of M. pyrifera along the central coast of California and how well those environmental parameters can be used to predict areas where this species is more likely to persist. Nine environmental variables considered in this study included depth of the seafloor, structure of the rocky reef, proportion of rocky reef, size of kelp patch, biomass of kelp within a patch, distance from the edge of a kelp patch, sea surface temperature, wave orbital velocities, and population connectivity of individual kelp patches. Using a generalized linear mixed effects model (GLMM), the persistence of M. pyrifera was significantly associated with seven of the nine variables considered: depth, complexity of the rocky reef, proportion of rock, patch biomass, distance from the edge of a patch, population connectivity, and wave orbital velocities. These seven environmental variables were then used to predict the persistence of kelp across the central coast, and these predictions were compared to a reserved dataset of M. pyrifera persistence, which was not used in the creation of the GLMM. The environmental variables were shown to accurately predict the persistence of M. pyrifera within the central coast of California (r = 0.71, P < 0.001). Because persistence of giant kelp is important to the community structure of kelp forests, understanding those factors that support persistent populations of M. pyrifera will enable more effective management of these ecosystems.

Small area estimation using linear mixed models with heterogeneous covariance structures of random effects

The problem of Small Area Estimation is about how to produce reliable estimates of domain characteristics when the sample sizes within the domain is very small ou even zero.

Robust linear mixed models with skew-normal independent distributions from a Bayesian perspective

Linear mixed models were developed to handle clustered data and have been a topic of increasing interest in statistics for the past 50 years. Generally. the normality (or symmetry) of the random effects is a common assumption in linear mixed models but it may, sometimes, be unrealistic, obscuring important features of among-subjects variation. In this article, we utilize skew-normal/independent distributions as a tool for robust modeling of linear mixed models under a Bayesian paradigm. The skew-normal/independent distributions is an attractive class of asymmetric heavy-tailed distributions that includes the skew-normal distribution, skew-t, skew-slash and the skew-contaminated normal distributions as special cases, providing an appealing robust alternative to the routine use of symmetric distributions in this type of models. The methods developed are illustrated using a real data set from Framingham cholesterol study. (C) 2009 Elsevier B.V. All rights reserved.

Likelihood ratio tests for variance components in linear mixed models

Although the asymptotic distributions of the likelihood ratio for testing hypotheses of null variance components in linear mixed models derived by Stram and Lee [1994. Variance components testing in longitudinal mixed effects model. Biometrics 50, 1171-1177] are valid, their proof is based on the work of Self and Liang [1987. Asymptotic properties of maximum likelihood estimators and likelihood tests under nonstandard conditions. J. Amer. Statist. Assoc. 82, 605-610] which requires identically distributed random variables, an assumption not always valid in longitudinal data problems. We use the less restrictive results of Vu and Zhou [1997. Generalization of likelihood ratio tests under nonstandard conditions. Ann. Statist. 25, 897-916] to prove that the proposed mixture of chi-squared distributions is the actual asymptotic distribution of such likelihood ratios used as test statistics for null variance components in models with one or two random effects. We also consider a limited simulation study to evaluate the appropriateness of the asymptotic distribution of such likelihood ratios in moderately sized samples. (C) 2008 Elsevier B.V. All rights reserved.

Feasible estimation of generalized linear mixed models (GLMM) with weak dependency between groups

This paper presents a two-step pseudo likelihood estimation technique for generalized linear mixed models with the random effects being correlated between groups. The core idea is to deal with the intractable integrals in the likelihood function by multivariate Taylor's approximation. The accuracy of the estimation technique is assessed in a Monte-Carlo study. An application of it with a binary response variable is presented using a real data set on credit defaults from two Swedish banks. Thanks to the use of two-step estimation technique, the proposed algorithm outperforms conventional pseudo likelihood algorithms in terms of computational time.

Parameter estimation of Poisson generalized linear mixed models based on three different statistical principles : a simulation study

Generalized linear mixed models are flexible tools for modeling non-normal data and are useful for accommodating overdispersion in Poisson regression models with random effects. Their main difficulty resides in the parameter estimation because there is no analytic solution for the maximization of the marginal likelihood. Many methods have been proposed for this purpose and many of them are implemented in software packages. The purpose of this study is to compare the performance of three different statistical principles - marginal likelihood, extended likelihood, Bayesian analysis-via simulation studies. Real data on contact wrestling are used for illustration.

Assessment of variance components in nonlinear mixed-effects elliptical models

The issue of assessing variance components is essential in deciding on the inclusion of random effects in the context of mixed models. In this work we discuss this problem by supposing nonlinear elliptical models for correlated data by using the score-type test proposed in Silvapulle and Silvapulle (1995). Being asymptotically equivalent to the likelihood ratio test and only requiring the estimation under the null hypothesis, this test provides a fairly easy computable alternative for assessing one-sided hypotheses in the context of the marginal model. Taking into account the possible non-normal distribution, we assume that the joint distribution of the response variable and the random effects lies in the elliptical class, which includes light-tailed and heavy-tailed distributions such as Student-t, power exponential, logistic, generalized Student-t, generalized logistic, contaminated normal, and the normal itself, among others. We compare the sensitivity of the score-type test under normal, Student-t and power exponential models for the kinetics data set discussed in Vonesh and Carter (1992) and fitted using the model presented in Russo et al. (2009). Also, a simulation study is performed to analyze the consequences of the kurtosis misspecification.