Home About us Contact
|
Home About us Contact | ||
|
|
||
Longitudinal Outcomes (longitudinal + outcome)
Selected AbstractsScore Test for Conditional Independence Between Longitudinal Outcome and Time to Event Given the Classes in the Joint Latent Class ModelBIOMETRICS, Issue 1 2010Hélène Jacqmin-Gadda Summary Latent class models have been recently developed for the joint analysis of a longitudinal quantitative outcome and a time to event. These models assume that the population is divided in,G,latent classes characterized by different risk functions for the event, and different profiles of evolution for the markers that are described by a mixed model for each class. However, the key assumption of conditional independence between the marker and the event given the latent classes is difficult to evaluate because the latent classes are not observed. Using a joint model with latent classes and shared random effects, we propose a score test for the null hypothesis of independence between the marker and the outcome given the latent classes versus the alternative hypothesis that the risk of event depends on one or several random effects from the mixed model in addition to the latent classes. A simulation study was performed to compare the behavior of the score test to other previously proposed tests, including situations where the alternative hypothesis or the baseline risk function are misspecified. In all the investigated situations, the score test was the most powerful. The methodology was applied to develop a prognostic model for recurrence of prostate cancer given the evolution of prostate-specific antigen in a cohort of patients treated by radiation therapy. [source] A Sensitivity Analysis for Shared-Parameter Models for Incomplete Longitudinal OutcomesBIOMETRICAL JOURNAL, Issue 1 2010An Creemers Abstract All models for incomplete data either explicitly make assumptions about aspects of the distribution of the unobserved outcomes, given the observed ones, or at least implicitly imply such. One consequence is that there routinely exist a whole class of models, coinciding in their description of the observed portion of the data but differing with respect to their "predictions" of what is unobserved. Within such a class, there always is a single model corresponding to so-called random missingness, in the sense that the mechanism governing missingness depends on covariates and observed outcomes, but given these not further on unobserved outcomes. We employ these results in the context of so-called shared-parameter models where outcome and missingness models are connected by means of common latent variables or random effects, to devise a sensitivity analysis framework. Precisely, the impact of varying unverifiable assumptions about unobserved measurements on parameters of interest is studied. Apart from analytic considerations, the proposed methodology is applied to assess treatment effect in data from a clinical trial in toenail dermatophyte onychomycosis. While our focus is on longitudinal outcomes with incomplete outcome data, the ideas developed in this paper are of use whenever a shared-parameter model could be considered. [source] Longitudinal outcome in patients with bipolar disorder assessed by life-charting is influenced by DSM-IV personality disorder symptomsBIPOLAR DISORDERS, Issue 1 2003Peter J Bieling Objectives:, Few studies have examined the question of how personality features impact outcome in bipolar disorder (BD), though results from extant work and studies in major depressive disorder suggest that personality features are important in predicting outcome. The primary purpose of this paper was to examine the impact of DSM-IV personality disorder symptoms on long-term clinical outcome in BD. Methods:, The study used a ,life-charting' approach in which 87 BD patients were followed regularly and treated according to published guidelines. Outcome was determined by examining symptoms over the most recent year of follow-up and personality symptoms were assessed with the Structured Clinical Interview for DSM-IV (SCID-II) instrument at entry into the life-charting study. Results:, Patients with better outcomes had fewer personality disorder symptoms in seven out of 10 disorder categories and Cluster A personality disorder symptoms best distinguished euthymic and symptomatic patients. Conclusions:, These results raise important questions about the mechanisms linking personality pathology and outcome in BD, and argue that conceptual models concerning personality pathology and BD need to be further developed. Treatment implications of our results, such as need for psychosocial interventions and treatment algorithms, are also described. [source] Amyloid imaging in mild cognitive impairment subtypes,ANNALS OF NEUROLOGY, Issue 5 2009David A. Wolk MD Objective We utilized the amyloid imaging ligand Pittsburgh Compound B (PiB) to determine the presence of Alzheimer's disease (AD) pathology in different mild cognitive impairment (MCI) subtypes and to relate increased PiB binding to other markers of early AD and longitudinal outcome. Methods Twenty-six patients with MCI (13 single-domain amnestic-MCI [a-MCI], 6 multidomain a-MCI, and 7 nonamnestic MCI) underwent PiB imaging. Twenty-three had clinical follow-up (21.2 ± 16.0 [standard deviation] months) subsequent to their PiB scan. Results Using cutoffs established from a control cohort, we found that 14 (54%) patients had increased levels of PiB retention and were considered "amyloid-positive." All subtypes were associated with a significant proportion of amyloid-positive patients (6/13 single-domain a-MCI, 5/6 multidomain a-MCI, 3/7 nonamnestic MCI). There were no obvious differences in the distribution of PiB retention in the nonamnestic MCI group. Predictors of conversion to clinical AD in a-MCI, including poorer episodic memory, and medial temporal atrophy, were found in the amyloid-positive relative to amyloid-negative a-MCI patients. Longitudinal follow-up demonstrated 5 of 13 amyloid-positive patients, but 0 of 10 amyloid-negative patients, converted to clinical AD. Further, 3 of 10 amyloid-negative patients "reverted to normal." Interpretation These data support the notion that amyloid-positive patients are likely to have early AD, and that the use of amyloid imaging may have an important role in determining which patients are likely to benefit from disease-specific therapies. In addition, our data are consistent with longitudinal studies that suggest a significant percentage of all MCI subtypes will develop AD. Ann Neurol 2009;65:557,568 [source] Robust Joint Modeling of Longitudinal Measurements and Competing Risks Failure Time DataBIOMETRICAL JOURNAL, Issue 1 2009Ning Li Abstract Existing methods for joint modeling of longitudinal measurements and survival data can be highly influenced by outliers in the longitudinal outcome. We propose a joint model for analysis of longitudinal measurements and competing risks failure time data which is robust in the presence of outlying longitudinal observations during follow-up. Our model consists of a linear mixed effects sub-model for the longitudinal outcome and a proportional cause-specific hazards frailty sub-model for the competing risks data, linked together by latent random effects. Instead of the usual normality assumption for measurement errors in the linear mixed effects sub-model, we adopt a t -distribution which has a longer tail and thus is more robust to outliers. We derive an EM algorithm for the maximum likelihood estimates of the parameters and estimate their standard errors using a profile likelihood method. The proposed method is evaluated by simulation studies and is applied to a scleroderma lung study (© 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [source] A Bayesian Semiparametric Survival Model with Longitudinal MarkersBIOMETRICS, Issue 2 2010Song Zhang Summary We consider inference for data from a clinical trial of treatments for metastatic prostate cancer. Patients joined the trial with diverse prior treatment histories. The resulting heterogeneous patient population gives rise to challenging statistical inference problems when trying to predict time to progression on different treatment arms. Inference is further complicated by the need to include a longitudinal marker as a covariate. To address these challenges, we develop a semiparametric model for joint inference of longitudinal data and an event time. The proposed approach includes the possibility of cure for some patients. The event time distribution is based on a nonparametric Pólya tree prior. For the longitudinal data we assume a mixed effects model. Incorporating a regression on covariates in a nonparametric event time model in general, and for a Pólya tree model in particular, is a challenging problem. We exploit the fact that the covariate itself is a random variable. We achieve an implementation of the desired regression by factoring the joint model for the event time and the longitudinal outcome into a marginal model for the event time and a regression of the longitudinal outcomes on the event time, i.e., we implicitly model the desired regression by modeling the reverse conditional distribution. [source] Multiple-Imputation-Based Residuals and Diagnostic Plots for Joint Models of Longitudinal and Survival OutcomesBIOMETRICS, Issue 1 2010Dimitris Rizopoulos Summary The majority of the statistical literature for the joint modeling of longitudinal and time-to-event data has focused on the development of models that aim at capturing specific aspects of the motivating case studies. However, little attention has been given to the development of diagnostic and model-assessment tools. The main difficulty in using standard model diagnostics in joint models is the nonrandom dropout in the longitudinal outcome caused by the occurrence of events. In particular, the reference distribution of statistics, such as the residuals, in missing data settings is not directly available and complex calculations are required to derive it. In this article, we propose a multiple-imputation-based approach for creating multiple versions of the completed data set under the assumed joint model. Residuals and diagnostic plots for the complete data model can then be calculated based on these imputed data sets. Our proposals are exemplified using two real data sets. [source] A prepubertal and early adolescent bipolar disorder-I phenotype: review of phenomenology and longitudinal courseBIPOLAR DISORDERS, Issue 4 2003James L Craney Objective: Phenomenology, assessment, longitudinal, and psychosocial findings from an ongoing, controlled, prospective study of 93 subjects with a prepubertal and early adolescent bipolar disorder phenotype (PEA-BP) will be reviewed. Methods: Unlike adult-onset bipolar disorder, for which there were over 50 years of systematic investigations, there were a paucity of rigorous data and much controversy and skepticism about the existence and characteristics of prepubertal-onset mania. With this background, issues to address for investigation of child-onset mania included the following: (i) What to do about the differentiation of mania from attention-deficit hyperactivity disorder (ADHD). (ii) How to deal with the ubiquity of irritability as a presenting symptom in multiple child psychiatry disorders. (iii) Development of a research instrument to assess prepubertal manifestations of adult mania (i.e. children do not ,max out' credit cards or have four marriages). (iv) How to distinguish normal childhood happiness and expansiveness from pathologically impairing elated mood and grandiosity. Results: To address these issues, a PEA-BP phenotype was defined as DSM-IV mania with elated mood and/or grandiosity as one inclusion criterion. This criterion ensured that the diagnosis of mania was not made using only criteria that overlapped with those for ADHD, and that subjects had at least one of the two cardinal symptoms of mania (i.e. elated mood and grandiose behaviors). Subjects were aged 10.9 years (SD = 2.6) and age of onset of the current episode at baseline was 7.3 years (SD = 3.5). Validation of PEA-BP was shown by reliable assessment, 6-month stability, and 1- and 2-year diagnostic longitudinal outcome. PEA-BP resembled the severest form of adult-onset mania by presenting with a chronic, mixed mania, psychotic, continuously (ultradian) cycling picture. Conclusion: Counterintuitively, typical 7-year-old children with PEA-BP were more severely ill than typical 27 year olds with adult-onset mania. Moreover, longitudinal data strongly supported differentiation of PEA-BP from ADHD. [source] Are there empirically supported and clinically useful subtypes of alcohol dependence?ADDICTION, Issue 2006Victor M. Hesselbrock ABSTRACT Aims This paper provides an overview of several multidimensional empirically derived typologies of alcohol use disorders that have been derived primarily for research purposes in relation to their clinical utility. Methods Studies using multivariate statistical methods for identifying homogeneous groups of subjects were selected for inclusion. Theoretically based typologies were not included in this review. Results While formal diagnostic criteria typically identify separate categories of alcohol abuse and dependence, several studies using different statistical methods consistently suggest as many as four homogeneous types of alcoholism: a chronic/severe type, a depressed/anxious type, a mildly affected type and an antisocial type. Conclusions Even though the longitudinal outcomes of few empirically derived subtypes have been examined, alcoholism typologies remain a viable and potentially valuable tool for investigating etiological pathways, the effectiveness of treatments and the long-term course of alcohol use disorders. [source] Early Therapeutic Alliance as a Predictor of Treatment Outcome for Adolescent Cannabis Users in Outpatient TreatmentTHE AMERICAN JOURNAL ON ADDICTIONS, Issue 2006Guy S. Diamond PhD The association of early alliance to treatment attendance and longitudinal outcomes were examined in 356 adolescents participating in a randomized clinical trial targeting cannabis use. Both patient and therapist views of alliance were examined, and outcomes were evaluated over 12 months after numerous other sources of variance were controlled. Patient-rated alliance predicted a reduction in cannabis use at three and six months and a reduction in substance-related problem behaviors at six months. Therapist-rated alliance did not predict outcomes. Neither patient nor therapist alliance ratings were associated with attendance. The findings support the important and often overlooked role that alliance can play in treating substance abusing, often delinquent, adolescents. [source] A Sensitivity Analysis for Shared-Parameter Models for Incomplete Longitudinal OutcomesBIOMETRICAL JOURNAL, Issue 1 2010An Creemers Abstract All models for incomplete data either explicitly make assumptions about aspects of the distribution of the unobserved outcomes, given the observed ones, or at least implicitly imply such. One consequence is that there routinely exist a whole class of models, coinciding in their description of the observed portion of the data but differing with respect to their "predictions" of what is unobserved. Within such a class, there always is a single model corresponding to so-called random missingness, in the sense that the mechanism governing missingness depends on covariates and observed outcomes, but given these not further on unobserved outcomes. We employ these results in the context of so-called shared-parameter models where outcome and missingness models are connected by means of common latent variables or random effects, to devise a sensitivity analysis framework. Precisely, the impact of varying unverifiable assumptions about unobserved measurements on parameters of interest is studied. Apart from analytic considerations, the proposed methodology is applied to assess treatment effect in data from a clinical trial in toenail dermatophyte onychomycosis. While our focus is on longitudinal outcomes with incomplete outcome data, the ideas developed in this paper are of use whenever a shared-parameter model could be considered. [source] A Bayesian Semiparametric Survival Model with Longitudinal MarkersBIOMETRICS, Issue 2 2010Song Zhang Summary We consider inference for data from a clinical trial of treatments for metastatic prostate cancer. Patients joined the trial with diverse prior treatment histories. The resulting heterogeneous patient population gives rise to challenging statistical inference problems when trying to predict time to progression on different treatment arms. Inference is further complicated by the need to include a longitudinal marker as a covariate. To address these challenges, we develop a semiparametric model for joint inference of longitudinal data and an event time. The proposed approach includes the possibility of cure for some patients. The event time distribution is based on a nonparametric Pólya tree prior. For the longitudinal data we assume a mixed effects model. Incorporating a regression on covariates in a nonparametric event time model in general, and for a Pólya tree model in particular, is a challenging problem. We exploit the fact that the covariate itself is a random variable. We achieve an implementation of the desired regression by factoring the joint model for the event time and the longitudinal outcome into a marginal model for the event time and a regression of the longitudinal outcomes on the event time, i.e., we implicitly model the desired regression by modeling the reverse conditional distribution. [source] Latent Pattern Mixture Models for Informative Intermittent Missing Data in Longitudinal StudiesBIOMETRICS, Issue 2 2004Haiqun Lin Summary. A frequently encountered problem in longitudinal studies is data that are missing due to missed visits or dropouts. In the statistical literature, interest has primarily focused on monotone missing data (dropout) with much less work on intermittent missing data in which a subject may return after one or more missed visits. Intermittent missing data have broader applicability that can include the frequent situation in which subjects do not have common sets of visit times or they visit at nonprescheduled times. In this article, we propose a latent pattern mixture model (LPMM), where the mixture patterns are formed from latent classes that link the longitudinal response and the missingness process. This allows us to handle arbitrary patterns of missing data embodied by subjects' visit process, and avoids the need to specify the mixture patterns a priori. One assumption of our model is that the missingness process is assumed to be conditionally independent of the longitudinal outcomes given the latent classes. We propose a noniterative approach to assess this key assumption. The LPMM is illustrated with a data set from a health service research study in which homeless people with mental illness were randomized to three different service packages and measures of homelessness were recorded at multiple time points. Our model suggests the presence of four latent classes linking subject visit patterns to homeless outcomes. [source] Latent Transition Regression for Mixed OutcomesBIOMETRICS, Issue 3 2003Diana L. Miglioretti Summary. Health status is a complex outcome, often characterized by multiple measures. When assessing changes in health status over time, multiple measures are typically collected longitudinally. Analytic challenges posed by these multivariate longitudinal data are further complicated when the outcomes are combinations of continuous, categorical, and count data. To address these challenges, we propose a fully Bayesian latent transition regression approach for jointly analyzing a mixture of longitudinal outcomes from any distribution. Health status is assumed to be a categorical latent variable, and the multiple outcomes are treated as surrogate measures of the latent health state, observed with error. Using this approach, both baseline latent health state prevalences and the probabilities of transitioning between the health states over time are modeled as functions of covariates. The observed outcomes are related to the latent health states through regression models that include subject-specific effects to account for residual correlation among repeated measures over time, and covariate effects to account for differential measurement of the latent health states. We illustrate our approach with data from a longitudinal study of back pain. [source] | |||||||||||||