Joint modeling of a linear mixed effects model for selfesteem from mean ages 13 to 22 and a generalized linear model for anxiety disorder at mean age 33 (original) (raw)
Related papers
Longitudinal trends in self-reported anxiety. Effects of age and birth cohort during 25 years
BMC psychiatry, 2017
Anxiety has been suggested to increase among young individuals, but previous studies on longitudinal trends are inconclusive. The aim of this study was to analyze longitudinally, the changes over time of prevalence of self-reported anxiety in the Swedish population between 1980/1981 and 2004/2005, in different birth cohorts and age groups. A random sample of non-institutionalized persons aged 16-71 years was interviewed every eighth year. Self-reported anxiety was assessed using the question" Do you suffer from nervousness, uneasiness, or anxiety?" (no; yes, mild; yes, severe). Mixed models with random intercepts were used to estimate changes in rates of anxiety (mild or severe) within different age groups and birth cohorts and in males and females separately. In addition to three time-related variables - year of interview, age at the time of the interview, and year of birth -the following explanatory variables were included: education, urbanization, marital status, smokin...
The Age of Onset of Anxiety Disorders
The Canadian Journal of Psychiatry, 2016
Objective: The objective was to estimate the age of onset (AOO) for all anxiety disorders and for specific subtypes. Gender differences in the AOO of anxiety disorders were examined, as were the influence of study characteristics on reported AOOs. Methods: Seven electronic databases were searched up to October 2014, with keywords representing anxiety disorder subtypes, AOO, and study design. The inclusion criteria were studies using a general population sample that provided data on the AOO for all anxiety disorders, or specific anxiety disorders, according to DSM-III-R, DSM-IV, or ICD-10 criteria. Results: There were 1028 titles examined, which yielded 24 studies meeting the inclusion criteria. Eight studies reported the AOO and gender. Meta-analysis found a mean AOO of all anxiety disorders of 21.3 years (95% CI 17.46 to 25.07). Separation anxiety disorder, specific phobia, and social phobia had their mean onset before the age of 15 years, whereas the AOO of agoraphobia, obsessive-...
Core dimensions of anxiety and depression change independently during adolescence
Journal of abnormal psychology, 2017
The developmental trajectories of emotional disorder symptoms during adolescence remain elusive, owing in part to a shortage of intensive longitudinal data. In the present study, we charted the temporal course of the tripartite model of anxiety and depression-which posits an overarching negative affect dimension and specific anhedonia and anxious arousal dimensions-over adolescence and emerging adulthood to construct a developmental map of the core dimensions of emotional disorders. We recruited 604 high school juniors, overselecting those at high risk for emotional disorders, and assessed the tripartite symptom domains 5 times annually. Latent curve modeling revealed that negative affect and anxious arousal declined over follow up, whereas anhedonia did not. Moreover, the correlation in rate of change varied across pairs of symptom domains. Change in negative affect was moderately correlated with change in anxious arousal, but change in anhedonia was not significantly related to ch...
Comprehensive Psychiatry, 2010
The study used person-oriented analyses to identify subgroups of individuals who exhibit different patterns of depressive and anxiety disorders over the course of adolescence and young adulthood. Using latent class growth analysis, six trajectory classes were identified. Two classes were mainly characterized by depressive disorders; one was mainly characterized by anxiety disorders; two were characterized by temporally different patterns of comorbidity; and one class was characterized by the absence of psychopathology. Classes characterized largely by depressive disorders differed in persistence and degree of comorbidity with anxiety disorders. Classes that were characterized by anxiety disorders differed in persistence, age of onset, and constellation of specific anxiety disorders. Female participants were more likely to belong to classes characterized by fluctuations in the course of depressive and anxiety disorders; sex differences were not observed in classes characterized by persistent depressive and anxiety disorders. Offspring of depressed parents were more likely to have a depressive course, whereas offspring of parents with anxiety disorders tended to have a course characterized by anxiety disorder. The findings indicate that several subgroups of adolescents exist with distinct longitudinal trajectories of depressive and anxiety disorders, and these trajectory classes are associated with different risk factors.
Development and psychopathology, 2014
In this study, we prospectively examined developmental trajectories of five anxiety disorder symptom dimensions (generalized anxiety disorder, panic disorder, school anxiety, separation anxiety disorder, and social anxiety disorder) from early to late adolescence in a community sample of 239 adolescents, assessed annually over 8 years. Latent growth modeling indicated different developmental trajectories from early into late adolescence for the different anxiety disorder symptoms, with some symptoms decreasing and other symptoms increasing over time. Sex differences in developmental trajectories were found for some symptoms, but not all. Furthermore, latent class growth analysis identified a normal developmental profile (including a majority of adolescents reporting persistent low anxiety disorder symptoms over 8 years) and an at-risk developmental profile (including a minority of adolescents reporting persistent high anxiety disorder symptoms over 8 years) for all of the anxiety di...
Real longitudinal data analysis for real people: Building a good enough mixed model
Statistics in Medicine, 2000
Mixed effect models have become very popular, especially for the analysis of longitudinal data. One challenge is how to build a good enough mixed effects model. In this paper, we suggest a systematic strategy for addressing this challenge and introduce easily implemented practical advice to build mixed effect models. A general discussion of scientific strategies motivates the recommended five step procedure for model fitting. The need to model both the mean structure (the fixed effects) and the covariance structure (the random effects and residual error) creates the fundamental flexibility and complexity. Some very practical recommendations help conquer the complexity. Centering, scaling, and full-rank coding all predictor variables radically improves the chances of convergence, computing speed, and numerical accuracy. Applying computational and assumption diagnostics from univariate linear models to mixed model data greatly helps detect and solve related computational problems. Applying computational and assumption diagnostics from univariate linear models to mixed model data can radically improve the chances of convergence, computing speed, and numerical accuracy. The approach helps fit more general covariance models, a crucial step in selecting a credible covariance model needed for defensible inference. A detailed demonstration of the recommended strategy is based on data from a published study of a randomized trial of a multicomponent intervention to prevent young adolescents' alcohol use. The discussion highlights a need for additional covariance and inference tools for mixed models. The discussion also highlights the need for improving how scientists and statisticians teach and review the process of finding a good enough mixed model.
Testing a hierarchical model of anxiety and depression in adolescents: A tri-level model
Journal of Anxiety Disorders, 2010
A strong association between anxiety and depression has been demonstrated through genotypic findings (e.g., Kendler, Prescott, Myers, & Neale, 2003) and phenotypic results at both symptom (e.g., Clark & Watson, 1991) and diagnostic levels (e.g., Kessler, Chiu, Demler, & Walters, 2005). Elucidating the structural relationships among anxiety and depressive symptoms will greatly benefit research into the classification and etiology of disorders of anxiety and depression. For example, Watson (2005) cites structural evidence as a reason to update classification of the mood and anxiety disorders in the upcoming fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Also, the valid characterization of anxiety and depression phenotypes will greatly benefit study of behavioral genetics through potentially enabling such phenotypes to be matched to genotypes. It is important to understand relationships among anxiety and depressive symptoms for individuals in their late teens, a period referred to as late adolescence or emerging adulthood, as rates of unipolar mood disorders and some anxiety disorders demonstrate large increases during this period (e.g., Burke, Burke, Regier, & Rae, 1990; Hankin et al., 1998). Further, although the DSM has a separate section for disorders usually first diagnosed in infancy, childhood, or adolescence (including separation anxiety disorder), most mood and anxiety disorders are specified as presenting very similarly across childhood, adolescence, and adulthood (with minor exceptions). Therefore, it is important to understand the structural relations among mood and anxiety symptoms during adolescence and how these relations compare to those found in childhood and adulthood. Phenotypic structural models to date have mainly examined the tripartite model proposed by Clark and Watson (1991). The tripartite model represents the shared and unique features of anxiety and depressive disorders using three factors: negative affect (NA), positive affect, and physiological hyperarousal. Shared symptoms, or those prevalent in both types of disorders, are represented by NA. Symptoms of anhedonia and the absence of positive affect are specific to depression while symptoms of somatic tension and physiological hyperarousal are specific to anxiety. The tripartite model has been supported in child and adolescent clinical (Chorpita, Albano, & Barlow, 1998) and nonclinical community samples (Anthony, Lonigan, Hooe, & Phillips,
Psychological Methods, 2012
With increasing popularity, growth curve modeling is more and more often considered as the first choice for analyzing longitudinal data. While the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this paper we compare three approaches for analyzing longitudinal data: repeated measures ANOVA, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this paper we first indicate the kinds of data that are appropriately modeled by each, and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of AIC and BIC in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.
Depression and Anxiety, 2014
Background: Identification of youth at risk for anxiety and unipolar mood disorders (UMDs) can improve public health by targeting those who may warrant early or preventive intervention. This study examined whether endorsing core features of anxiety and UMDs predicted onset of later anxiety and UMDs across the next 7-9 years, and whether having subthreshold or subclinical manifestations of these disorders similarly predicted onset. Methods: Data from this study come from the Youth Emotion Project (YEP), a two-site investigation of common and specific risk factors for emotional disorders. Endorsement of core features of a disorder and subclinical or subthreshold anxiety and UMD diagnoses were determined using data from the Structured Clinical Interview for DSM-IV (SCID) at the baseline assessment. Participants completed annual SCIDs over the course of the next 7-9 years (depending on cohort). Results: Endorsement of panic attacks, obsessions and/or compulsions, and depression and/or anhedonia predicted onset of panic disorder, obsessive compulsive disorder, and major depressive disorder, respectively. When including all anxiety disorders in a model, only the presence of panic attacks uniquely predicted anxiety disorder onset. The presence of subclinical or subthreshold panic disorder, obsessive compulsive disorder, and social phobia at baseline predicted the full onset of these disorders over the follow-up period. Conclusions: Experiencing some symptoms of anxiety and UMDs in the absence of meeting diagnostic criteria is indicative of risk for later onsets of clinically significant DSM manifestations of these disorders. These individuals should be identified and targeted for prevention programs. Depression and Anxiety 31:207-213, 2014.