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 775 records · Page 43Linked to original sources
PubMed2003

A longitudinal approach for constructing beta-expectation tolerance intervals.

A new method is presented for constructing beta-expectation tolerance limits (TLs) for longitudinal data with error components regression structure. The TLs are mainly intended for small samples where the variation between and within subjects is large. In such cases the new method is superior to existing methods, which are based on a cross-sectional approach and which do not utilize the longitudinal structure of the data. Simulation studies show that the mean length of the TLs can be much reduced by using the new approach, while at the same time the beta-expectation property is maintained. The gain from using the longitudinal approach furthermore increases with the beta-expectation level. The results are demonstrated on data consisting of measurements of glutamate concentrations in brains from rats. Here, the cross-sectional approach due to Wilks give 90-99% TLs, which are up to 101-230% wider than are those obtained with the new approach.

Animals↗
PubMed2006

Maternal perceptions of her child's body weight in infancy and early childhood and their relation to body weight status at age 7.

INTRODUCTION: Maternal perception of her child's weight status has been hypothesised to affect a child's weight development. OBJECTIVE: The aim of this analysis was to determine in how far the maternal weight perception of her child's weight at different ages is related to its future body weight status. MATERIALS AND METHODS: Longitudinal data on body weight, height and skinfolds from the Dortmund Nutritional and Anthropometric Longitudinally Designed Study were used to determine the risk of being overweight at age 7. Complete data on anthropometry, maternal weight perception and confounding variables were available for 253 children. RESULTS AND DISCUSSION: Maternal weight perception assessments at age 6 months, 12 months, 2 years and 4 years of age were related to body weight status as well as changes in body mass index standard deviation score (BMI-SDS) and skinfold SDS between age at assessment and age 7. With respect to the risk of being overweight at age 7, no independent effect of maternal perception was found. When changes in anthropometric measures were considered, it could be observed that children whose weight was considered too low gained more weight until age 7, and those who were considered to be too heavy lost more weight as compared to the children whose weight was considered to be just right. Among infants aged 6 months who were above the 85th percentile, maternal misperception appeared to promote an unfavourable weight development. These latter results support the hypothesis that maternal weight perception might affect a child's weight development. CONCLUSION: As our data suggest, this effect might begin to operate already in infancy.

Adolescent↗
PubMed2003

The effects of religious commitment on the attitudes and behavior of teens regarding premarital childbirth.

Using the NELS data set from the 1988-1992 period, this study used a longitudinal data set to address whether very religious teens maintained different attitudes and behaviors regarding premarital childbirth than their less religious counterparts. The results of this study support the notion that a high level of religious commitment among adolescents makes it less likely that teens will have permissive attitudes about premarital childbirth and also makes it less likely that they will become never married single parents during their high school years. The variables for "religiosity" were taken from the NELS surveys for 1988 and 1990. The variables measuring attitudes and behaviors regarding premarital childbirth were obtained from the 1992 data set. This procedure helped address any issues regarding the direction of causality. The significance of these results is discussed.

Adolescent↗
PubMed1994

Cox regression analysis of multivariate failure time data: the marginal approach.

Multivariate failure time data are commonly encountered in scientific investigations because each study subject may experience multiple events or because there exists clustering of subjects such that failure times within the same cluster are correlated. In this paper, I present a general methodology for analysing such data, which is analogous to that of Liang and Zeger for longitudinal data analysis. This approach formulates the marginal distributions of multivariate failure times with the familiar Cox proportional hazards models while leaving the nature of dependence among related failure times completely unspecified. The baseline hazard functions for the marginal models may be identical or different. Simple estimating equations for the regression parameters are developed which yield consistent and asymptotically normal estimators, and robust variance-covariance estimators are constructed to account for the intra-class correlation. Simulation results demonstrate that the large-sample approximations are adequate for practical use and that ignoring the intra-class correlation could yield rather misleading variance estimators. The proposed methodology has been fully implemented in a simple computer program which also incorporates several alternative approaches. Detailed illustrations with data from four clinical or epidemiologic studies are provided.

Adolescent↗
PubMed1990

The power to detect differences in average rates of change in longitudinal studies.

With considerable current interest in longitudinal epidemiologic studies, little is available regarding sample size requirements. This paper considers a method for analysis of longitudinal data, where one compares the mean rates of change for two or more groups, and proposes a statistic for use in determining sample size requirements. One calculates individual rates of change with least squares estimates of slopes of individuals' responses regressed over time. The assumption of linear change over time, while clearly not applicable for some data, applies to many biological measurements, either as recorded or with some transformation. The variances of these estimated slopes have two components: within-individual variability based on measurement error and length and frequency of follow-up, and true between-individual slope variability. It is assumed that measurement error is the same for all subjects, so that the total variances differ due to differences in follow-up. The question addressed is: when can one use the usual ANOVA F statistic to compare group means of estimated slopes? Expected mean squares demonstrate that this F is appropriate when either group has the same number of subjects, or when each subject has the same length and frequency of follow-up. A procedure for computing power and sample size is presented, where one can specify the maximum detectable difference in any two average slopes. Moment estimation and maximum likelihood estimation of variance components from prior data are discussed.

Humans↗
PubMed2004

A short history of data banking in the United States from 1974 to 2003.

There have been 4 major longitudinal data banking efforts within the United States: ARAMIS, the Western Consortium, and the individual data banks of Drs. Ted Pincus and Fred Wolfe. ARAMIS began in the 1970s, and helped to develop the language and methodology of rheumatology data banks using biannual surveys. The National Data Bank for Rheumatic Diseases used the ARAMIS model beginning in the late 1990s to form a very large contemporary rheumatology data bank. Hybrid models using both survey data and clinical data were put into practice by Pincus and Wolfe, and by Paulus at the Western Consortium.

Arthritis, Rheumatoid↗
PubMed1992

Parametric variance function estimation for nonnormal repeated measurement data.

Zeger and Liang (1986, Biometrics 42, 121-130) proposed a procedure for analyzing nonnormal longitudinal data in the context of the generalized linear model. This procedure is extended to model variance heterogeneity, allowing the observations to come from distributions with different scale parameters. Loss of efficiency is evaluated when heterogeneity of scale factor is ignored.

Analysis of Variance↗
PubMed1994

Time series models based on generalized linear models: some further results.

This paper considers the problem of extending the classical moving average models to time series with conditional distributions given by generalized linear models. These models have the advantage of easy construction and estimation. Statistical modelling techniques are also proposed. Some simulation results and an illustrative example are reported to illustrate the methodology. The models will have potential applications in longitudinal data analysis.

Biometry↗
PubMed1994

An application of maximum likelihood and generalized estimating equations to the analysis of ordinal data from a longitudinal study with cases missing at random.

Data are analysed from a longitudinal psychiatric study in which there are no dropouts that do not occur completely at random. A marginal proportional odds model is fitted that relates the response (severity of side effects) to various covariates. Two methods of estimation are used: generalized estimating equations (GEE) and maximum likelihood (ML). Both the complete set of data and the data from only those subjects completing the study are analysed. For the completers-only data, the GEE and ML analyses produce very similar results. These results differ considerably from those obtained from the analyses of the full data set. There are also marked differences between the results obtained from the GEE and ML analysis of the full data set. The occurrence of such differences is consistent with the presence of a non-completely-random dropout process and it can be concluded in this example that both the analyses of the completers only and the GEE analysis of the full data set produce misleading conclusions about the relationships between the response and covariates.

Age Factors↗
PubMed2003

Alcohol intake and glycemia in American Indians: the strong heart study.

The goal of this study was to explore the relationship between alcohol intake and glycemia and type 2 diabetes in American Indians aged 45 to 74 years. Data were obtained from participants in the Strong Heart Study, a longitudinal study of 13 American Indian communities in 3 geographic areas in the United States. Alcohol consumption was determined by self-reported alcohol intake history. Participants previously diagnosed with diabetes were excluded from the analysis. Analysis of covariance (ANCOVA) was used to estimate the adjusted means of blood glucose for alcohol intake categories. Logistic regression was used to calculate the odds ratios (ORs) and 95% confidence intervals (CIs) to estimate the association between alcohol intake and type 2 diabetes in the cross-sectional analysis and between alcohol intake and glucose intolerance using longitudinal data. Fasting and 2-hour plasma glucose concentrations showed an inverse J-shaped curve across categories of alcohol intake. Using never drinkers as the referent group in cross-sectional analysis, light drinkers had a significantly lower risk of having diabetes (OR, 0.66; 95% CI, 0.44 to 0.99); among drinkers, heavy drinkers had a higher, although not significant, prevalence of diabetes. Longitudinal analysis showed no significant worsening of glucose tolerance across levels of alcohol intake. Subanalyses stratified by body mass index (BMI) did not show differences between obese and non-obese participants in the relationship between alcohol intake and glucose tolerance. Although plasma glucose concentration showed a shallow, inverse J-shaped association across levels of increasing alcohol intake in American Indians aged 45 to 74 years, alcohol intake did not appear to significantly increase the risk for worsening glucose tolerance. Thus, alcohol intake does not appear to be a determinant of diabetes risk in this population.

Aged↗
PubMed2003

Adult age differences in multiple cognitive functions: differentiation, dedifferentiation, or process-specific change?

Adult age differences in covariance structures of latent variables of vocabulary, list recall, speed, working memory, and text recall, were analyzed to test hypotheses of structural changes with age. There were baseline data from 613 men and women aged 30-97, data from a second wave of testing from 322 people, and complete longitudinal data from 289 people. There were age differences in the size but not configuration of factor loadings cross-sectionally but not longitudinally. There were no changes in factor standard deviations or covariances. Findings did not support models of dedifferentiation with age.

Adult↗
PubMed2005

[Determinants of postpartum weight retention according to skin color among women in Rio de Janeiro, Brazil].

OBJECTIVE: To investigate factors associated with postpartum weight retention according to skin color in the city of Rio de Janeiro, Brazil, as part of the effort to help control overweight and obesity. METHODS: We carried out a cohort study with data collection at four points after delivery: 15 days, 2 months, 6 months, and 9 months. The 450 participants (15 to 45 years of age) were recruited from a maternity center and a city health center in Rio de Janeiro. Postpartum weight retention was defined as the dependent variable. The following covariates were investigated: maternal age, marital status, ability to read a letter, smoking, monthly family income, number of children, age at first delivery, a score for physical activity at work, and a score for physical activity during leisure time. The data were analyzed using multivariate linear mixed-effects models for longitudinal data. RESULTS: The factors that remained associated with postpartum weight retention for white women were time after delivery (beta = -0.0061, P = 0.0002), marital status (married/single beta = -0.9279, P = 0.0457), and age at first delivery (beta = -0.1553, P = 0.0364). The factors for brown women were time after delivery (beta = -0.0062, P < 0.0001), marital status (married/single beta = -0.6000, P = 0.0145), and work physical activity score (beta = -0.1068, P = 0.0003). The factors for black women were smoking (beta = -1.4635, P = 0.0535) and score for physical activity during leisure time (beta = -0.2422, P = 0.0145). CONCLUSIONS: The factors associated with postpartum weight retention were different for the three skin-color categories. Nevertheless, the data do not indicate the need for different interventions for the three skin-color groups. That is particularly true because the physical activity scores were among the variables associated with weight retention, and any initiative to control obesity should take physical activity into consideration. On the other hand, given the socioeconomic differences among the three skin-color categories, new studies should be developed to investigate in more detail the pattern of postpartum weight retention according to skin color.

Adolescent↗
PubMed1996

MIXREG: a computer program for mixed-effects regression analysis with autocorrelated errors.

MIXREG is a program that provides estimates for a mixed-effects regression model (MRM) for normally-distributed response data including autocorrelated errors. This model can be used for analysis of unbalanced longitudinal data, where individuals may be measured at a different number of timepoints, or even at different timepoints. Autocorrelated errors of a general form or following an AR(1), MA(1), or ARMA(1,1) form are allowable. This model can also be used for analysis of clustered data, where the mixed-effects model assumes data within clusters are dependent. The degree of dependency is estimated jointly with estimates of the usual model parameters, thus adjusting for clustering. MIXREG uses maximum marginal likelihood estimation, utilizing both the EM algorithm and a Fisher-scoring solution. For the scoring solution, the covariance matrix of the random effects is expressed in its Gaussian decomposition, and the diagonal matrix reparameterized using the exponential transformation. Estimation of the individual random effects is accomplished using an empirical Bayes approach. Examples illustrating usage and features of MIXREG are provided.

Adolescent↗
PubMed2005

The role of depressive symptoms in predicting sex with multiple and high-risk partners.

Existing research examining the role of psychopathology on HIV risk behaviors has relied heavily upon cross-sectional data, making it difficult to assess causal relationships. This paper uses longitudinal data to examine the relationship between depressive symptoms and sexual risk behaviors in a community sample of 332 inner-city drug users. Results from repeated measures analysis using generalized estimating equations suggest that high depressive symptoms may be causally related to having sex with multiple partners (odds ratio = 1.55, CI: 1.12, 2.14) and having sex with injection drug users (odds ratio = 1.57, CI: 1.11, 2.22) and crack users (odds ratio = 1.37, CI: 1.02, 1.86). Findings support greater attention to depressive symptoms in HIV prevention interventions for illicit drug users to reduce sexual risk behaviors.

Adult↗
PubMed2005

Longitudinal models of growth and survival applied to the early detection of Alzheimer's disease.

This article explores new statistical methodologies for using longitudinal data in the early prediction of Alzheimer's disease (AD). Specifically, the authors examine some new techniques that allow the joint or "shared" estimation of longitudinal components based on both duration (survival) and quantitative changes (growth curves). These new shared growth-survival parameter models may be used to characterize the declining functions that anticipate the onset of AD. The authors apply these models to data from the Kungsholmen Project, a longitudinal study of aging in Stockholm, Sweden. They examine age-based survival-frailty models for the onset of AD, latent growth-decline curve models for changes in cognition over age, and 3 alternative forms of models for the shared relationships of survival and early cognitive decline. The accuracy and reliability of this approach is considered for a better understanding of the developmental course of AD in these data, including the potential removal of biases due to subject selection.

Adult↗
PubMed1998

Time to nursing home admission for persons with Alzheimer's disease: the effect of health care system characteristics.

OBJECTIVE: To study the influence of state health care system characteristics on time to nursing home admission (NHA) for persons with Alzheimer's disease (AD). METHOD: Up to nine years of Consortium to Establish a Registry for Alzheimer's Disease (CERAD) data on 639 non-Latino White individuals were merged with longitudinal data from the 28 states in which the CERAD participants resided. The state variables reflected characteristics of each state's long-term care (LTC) system, including Medicaid LTC spending practices and the supply of LTC providers. Cox Proportional Hazards Models with time-varying covariates were used to evaluate the risk factors associated with time to NHA. RESULTS: There was differential influence of state variables by marital status. For unmarried non-Latino White persons with AD, a higher percentage of Medicaid LTC spending on home and community-based services (HCBS) was significantly associated with a longer time to NHA. For married persons, a greater number of home health agencies was associated with a longer time to NHA. Other associations also varied by marital status. CONCLUSION: Study findings support the utility of targeted continued expanded provision of HCBS by states and provide a basis for future research regarding the impact of changing state health care systems on LTC utilization for persons with AD.

Aged↗
PubMed2000

Mass change, environmental variability and female fertility in wild Propithecus verreauxi.

Accurate estimates of mass and size are important in a wide range of research questions in population and evolutionary biology, and yet such data are still rare for wild primates. This study presents detailed longitudinal data from a large population of wild indriids, and demonstrates links between fluctuations in body mass, environmental cycles, and reproduction. Understanding these links is a necessary step toward explaining the function and evolution of distinctive features of lemur biology and behavior. During the first 12 years of an ongoing study of the sifaka, Propithecus verreauxi verreauxi, at Beza Mahafaly in southwest Madagascar, 320 animals were captured and weighed throughout the year. Adult males and females exhibit seasonal cycles of mass loss, with females losing significantly more mass than males. In 2 drought years this pattern was especially pronounced. Compared to lighter females, females who were heavier at the time of the mating season were more likely to give birth in the following birth season. By showing (1) seasonally greater mass loss in reproductive females compared to males, particularly in drought years, (2) a close link between female mass and fertility, and (3) an uncoupling of the periods of highest body mass and of gestation and lactation, these results suggest that energy acquisition and storage are critically important in the life history strategies of female sifaka, and that "capital breeding" may be a feature of sifaka reproductive strategies.

Animals↗