PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “longitudinal data”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources
PubMed2006

Structural inference in transition measurement error models for longitudinal data.

We propose a new class of models, transition measurement error models, to study the effects of covariates and the past responses on the current response in longitudinal studies when one of the covariates is measured with error. We show that the response variable conditional on the error-prone covariate follows a complex transition mixed effects model. The naive model obtained by ignoring the measurement error correctly specifies the transition part of the model, but misspecifies the covariate effect structure and ignores the random effects. We next study the asymptotic bias in naive estimator obtained by ignoring the measurement error for both continuous and discrete outcomes. We show that the naive estimator of the regression coefficient of the error-prone covariate is attenuated, while the naive estimators of the regression coefficients of the past responses are generally inflated. We then develop a structural modeling approach for parameter estimation using the maximum likelihood estimation method. In view of the multidimensional integration required by full maximum likelihood estimation, an EM algorithm is developed to calculate maximum likelihood estimators, in which Monte Carlo simulations are used to evaluate the conditional expectations in the E-step. We evaluate the performance of the proposed method through a simulation study and apply it to a longitudinal social support study for elderly women with heart disease. An additional simulation study shows that the Bayesian information criterion (BIC) performs well in choosing the correct transition orders of the models.

Aged↗
PubMed2001

An alternative parameterization of the general linear mixture model for longitudinal data with non-ignorable drop-outs.

This paper considers the mixture model methodology for handling non-ignorable drop-outs in longitudinal studies with continuous outcomes. Recently, Hogan and Laird have developed a mixture model for non-ignorable drop-outs which is a standard linear mixed effects model except that the parameters which characterize change over time depend also upon time of drop-out. That is, the mean response is linear in time, other covariates and drop-out time, and their interactions. One of the key attractions of the mixture modelling approach to drop-outs is that it is relatively easy to explore the sensitivity of results to model specification. However, the main drawback of mixture models is that the parameters that are ordinarily of interest are not immediately available, but require marginalization of the distribution of outcome over drop-out times. Furthermore, although a linear model is assumed for the conditional mean of the outcome vector given time of drop out, after marginalization, the unconditional mean of the outcome vector is not, in general, linear in the regression parameters. As a result, it is not possible to parsimoniously describe the effects of covariates on the marginal distribution of the outcome in terms of regression coefficients. The need to explicitly average over the distribution of the drop-out times and the absence of regression coefficients that describe the effects of covariates on the outcome are two unappealing features of the mixture modelling approach. In this paper we describe a particular parameterization of the general linear mixture model that circumvents both of these problems.

Anti-Asthmatic Agents↗
PubMed2001

Genotypic correlates of resistance to HIV-1 protease inhibitors on longitudinal data: the role of secondary mutations.

Direct sequencing of the pol gene was assessed retrospectively with protease inhibitor susceptibility in a longitudinal study. A total of 134 samples from 26 patients were analysed at regular intervals up to 2 years. Patients were included in virological failure despite indinavir, ritonavir or saquinavir based triple-drug therapy. Both the type and number of certain secondary protease mutations modulated the effect of primary mutations on phenotypic resistance. This was notably applicable to L101/V, and to lesser extents to A711V/T. However, combinations of primary mutations, including 154V could predict resistance to the drug used and nelfinavir in more than 80%. In contrast, in vitro cross-resistance to amprenavir was rarely encountered. In addition, there was a relationship between a higher number of key mutations and poorer virological and clinical outcomes, respectively, from 6 and 3 months on. The key mutations were the protease mutations independently conferring phenotypic resistance and/or the reverse transcriptase mutations predicting treatment outcome. This relationship was independent from drug history, viral load and CD4 cell count measurements. In summary, even on a small sample size, sequence-based genotyping seems to be a good prognostic marker when performed longitudinally. In the context of primary resistance mutations, including additional secondary mutations, it may be useful in the prediction of phenotypic and clinical resistance. This should be assessed to optimize treatment monitoring before emergence of broadly cross-resistant virus.

Acquired Immunodeficiency Syndrome↗
PubMed2001

Comparing personal trajectories and drawing causal inferences from longitudinal data.

This review considers statistical analysis of data from studies that obtain repeated measures on each of many participants. Such studies aim to describe the average change in populations and to illuminate individual differences in trajectories of change. A person-specific model for the trajectory of each participant is viewed as the foundation of any analysis having these aims. A second, between-person model describes how persons very in their trajectories. This two-stage modeling framework is common to a variety of popular analytic approaches variously labeled hierarchical models, multilevel models, latent growth models, and random coefficient models. Selected published examples reveal how the approach can be flexibly adapted to represent development in domains as diverse as vocabulary growth in early childhood, academic learning, and antisocial propensity during adolescence. The review then considers the problem of drawing causal inferences from repeated measures data.

Adolescent↗
PubMed1999

Glycemic control in diabetic American Indians. Longitudinal data from the Strong Heart Study.

OBJECTIVE: To describe glycemic control and identify correlates of elevated HbA1c levels in diabetic American Indians participating in the Strong Heart Study, which is a longitudinal study of cardiovascular disease in American Indians in Arizona, Oklahoma, South Dakota, and North Dakota. RESEARCH DESIGN AND METHODS: This analysis is based on data from the baseline (1989-1992) and first follow-up (1994-1995) examinations of the Strong Heart Study. The 1,581 diabetic participants included in this analysis were aged 45-74 years at baseline, were diagnosed with diabetes before and at baseline, and had their HbA1c levels measured at follow-up. HbA1c was used as the index of glycemic control. Characteristics that may affect glycemic control were evaluated for cross-sectional and longitudinal relationships by analysis of covariance and multiple regression. RESULTS: There was no significant difference between median HbA1c at baseline (8.4%) and at follow-up (8.5%). Sex, age (inversely), and insulin and oral hypoglycemic agent therapy were significantly related to HbA1c levels in both the cross-sectional and longitudinal analyses. Current smoking, prior use of alcohol, and duration of diabetes were significant only for the cross-sectional data. Baseline HbA1c significantly and positively predicted HbA1c levels at follow-up. Comparison of HbA1c by therapy type shows that insulin therapy produced a significant decrease in HbA1c between the baseline and follow-up examinations. CONCLUSIONS: Glycemic control was poor among diabetic American Indians participating in the Strong Heart Study. Women, patients taking insulin or oral hypoglycemic agents, and younger individuals had the worst control of all the participants. Baseline HbA1c, and weight loss predicted worsening of control, whereas insulin therapy predicted improvement in control. Additional therapies and/or approaches are needed to improve glycemic control in this population.

Administration, Oral↗
PubMed1978

Mixed longitudinal data on skeletal age from a group of Dutch children living in Utrecht and surroundings.

The height and weight of 1132 children, aged 8.0--17.0 years studied in 1970--72 in a semi-longitudinal survey in the Dental Institute of Utrecht, are compared with similar data from the national Dutch survey of 1965. Children measured in 1970--72 are somewhat taller than, but have the same weight as, those of the same ages in 1965. The increase in height since 1965 appears to be primarily due to the sub-group enrolled in vocational (as opposed to general) education. The maximal yearly increments in height and weight of the girls occurred between 11.0 and 12.0 years, and between 12.0 and 13.0 years, respectively. For boys the maximal increments in height and weight occurred between 14.0 and 15.0 years. Using the Tanner-Whitehouse 2 method, the skeletal age of this group of children was determined and compared with similar data from British standards. The results of the twenty-bone skeletal age indicated that Dutch boys, and to a lesser extent, girls, mature slightly later than English children at the approximate age range of 10.0--13.0 years and 8.0--10.0 years, respectively. After this age they follow roughly the growth curve of the British standards. The annual increment in skeletal age, plotted with chronological age as a time base, shows a peak for boys as well as girls that coincides with peak height velocity. The Carpal skeletal ages of boys and girls are almost identical in all age-groups with those of the British children, while the RUS skeletal age shows a much greater variability in the different age-groups. The variation in mean velocity (maturity points) between the two populations appears to be more marked in the RUS bones than in the round bones. The TW 1 skeletal age of each subject was plotted against the total TW 2, RUS or Carpal skeletal ages of the same individual. Equations are given for converting TW 1 skeletal ages into total TW 2 or RUS skeletal ages.

Adolescent↗
PubMed1993

Clustering on the basis of longitudinal data.

A menu-drive PC program, ZDIST, for computing the distances between the estimated polynomial growth curves of subjects who have been followed longitudinally is described, illustrated, and made available to interested readers. These distances can be computed on the basis of the individual growth curves themselves and/or from estimates of individuals' growth velocity and acceleration curves. The resulting distance matrices can be saved in ASCII format and subsequently imported into any clustering program which accepts this type of input, e.g. SYSTAT.

Achondroplasia↗
PubMed2006

Relationship status and testosterone in North American heterosexual and non-heterosexual men and women: cross-sectional and longitudinal data.

Previous research has found that single heterosexual (Het) men have higher salivary testosterone (T) concentrations than partnered Het men. Here, we used both longitudinal and cross-sectional analyses to examine a more diverse population (n = 258) that included Het and non-heterosexual (Non-Het) women and men. Results showed that, for Het men (but not Het women) and Non-Het women (but not Non-Het men), baseline T was significantly lower in partnered than unpartnered individuals. Longitudinal analyses indicated that changes in partnered status were not associated with changes in testosterone concentrations; instead, women and men with lower T at baseline were significantly more likely to be partnered at follow-up. These findings thus suggest that partnered status is associated with stable, trait-level T values, rather than current state. Furthermore, the observed effect is limited to individuals (male or female) who are oriented toward female partners. The results are discussed in terms of evolutionary trade-offs between single and multiple partners, and the possibility of female choice and/or disinterest.

Adult↗
PubMed1980

[Longitudinal data of physical growth of healthy children. II. Height, weight, skinfold thickness of children aged 1.5--16 years (author's transl)].

From 1968--1978 a longitudinal study was performed concerning development of height, weight and skinfold thickness in 709 boys and 711 girls 1.5--16 years old. Increase in height during time studied amounted to 92 cm in boys and 82 cm in girls. Mean increases in weight amounted to 49.3 kg and 41.2 kg respectively. Boys had highest increase in growth from 13 to 15 years, girls from 11 to 13 years. Skinfold has been thicker in girls than in boys. Triceps Skinfolds had been ped wave-like in both sexes. Following an unchanged decreased phase, the skinfolds remaining developed its thickness constantly. The phase of stagnation paralleled time of highest increase in growth.

Adolescent↗
PubMed2006

A nonlinear model with latent process for cognitive evolution using multivariate longitudinal data.

Cognition is not directly measurable. It is assessed using psychometric tests, which can be viewed as quantitative measures of cognition with error. The aim of this article is to propose a model to describe the evolution in continuous time of unobserved cognition in the elderly and assess the impact of covariates directly on it. The latent cognitive process is defined using a linear mixed model including a Brownian motion and time-dependent covariates. The observed psychometric tests are considered as the results of parameterized nonlinear transformations of the latent cognitive process at discrete occasions. Estimation of the parameters contained both in the transformations and in the linear mixed model is achieved by maximizing the observed likelihood and graphical methods are performed to assess the goodness of fit of the model. The method is applied to data from PAQUID, a French prospective cohort study of ageing.

Aged↗
PubMed2005

Stage of change transitions and processes of change, decisional balance, and self-efficacy in smokers: a transtheoretical model validation using longitudinal data.

Interactions were examined between stage of change transitions and intraindividual increases or decreases in the processes of change, pros and cons of smoking, and situational temptations longitudinally. A total of 786 ever smokers was assessed 2 times, 6 months apart, with respect to the transtheoretical model (TTM) constructs. Two significant discriminant functions within initial precontemplators and 1 significant function within initial contemplators were found. Ten out of 15 TTM variables contributed to at least 1 function. The functions mainly distinguished between preabstinence (precontemplation, contemplation, or preparation) and abstinence (action or maintenance) stages of change, that is, between current and former smokers. This is one of the few studies providing a longitudinal validation of the postulates of the TTM.

Adolescent↗
PubMed2000

Analysis of left-censored longitudinal data with application to viral load in HIV infection.

The classical model for the analysis of progression of markers in HIV-infected patients is the mixed effects linear model. However, longitudinal studies of viral load are complicated by left censoring of the measures due to a lower quantification limit. We propose a full likelihood approach to estimate parameters from the linear mixed effects model for left-censored Gaussian data. For each subject, the contribution to the likelihood is the product of the density for the vector of the completely observed outcome and of the conditional distribution function of the vector of the censored outcome, given the observed outcomes. Values of the distribution function were computed by numerical integration. The maximization is performed by a combination of the Simplex algorithm and the Marquardt algorithm. Subject-specific deviations and random effects are estimated by modified empirical Bayes replacing censored measures by their conditional expectations given the data. A simulation study showed that the proposed estimators are less biased than those obtained by imputing the quantification limit to censored data. Moreover, for models with complex covariance structures, they are less biased than Monte Carlo expectation maximization (MCEM) estimators developed by Hughes (1999) Mixed effects models with censored data with application to HIV RNA Levels. Biometrics 55, 625-629. The method was then applied to the data of the ALBI-ANRS 070 clinical trial for which HIV-1 RNA levels were measured with an ultrasensitive assay (quantification limit 50 copies/ml). Using the proposed method, estimates obtained with data artificially censored at 500 copies/ml were close to those obtained with the real data set.

Journal Article↗