Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “linear mixed model”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 307 records · Page 17Linked to original sources
PubMed2006

Modeling prostate specific antigen kinetics in patients on active surveillance.

PURPOSE: Prostate specific antigen doubling time was used to stratify patients into groups at low and high risk for progression. The prostate specific antigen kinetics in these 2 groups were modeled. MATERIALS AND METHODS: In this prospective, single-arm cohort study patients with favorable clinical parameters (stage T1b-T2b N0M0, Gleason score 7 or less, prostate specific antigen 15 ng/ml or less) were conservatively treated with watchful waiting. Evolution of serial prostate specific antigen measurements over time was estimated from a general linear mixed model of the natural log of prostate specific antigen. The corresponding average and individual prostate specific antigen doubling times were also calculated. RESULTS: Since November 1995 a total of 231 patients had at least 6 months of followup and at least 3 prostate specific antigen measurements. Based on prostate specific antigen doubling time and repeat biopsy, 93 patients fulfilled the criteria for high risk of disease progression and 138 were defined as low risk. Given the baseline status of these individuals, 2 reference average lines (high risk and low risk) were derived to model the evolution of prostate specific antigen levels and permit more rational decision making regarding the need for definitive intervention. The average prostate specific antigen doubling time was 2.97 years (95% CI 2.2-4.4) in patients allocated to the high risk group and 6.54 years (95% CI 4.8-12.3) in those at low risk. CONCLUSIONS: By applying the dynamic prognostic rule in combination with serial biopsy, a rational decision for definitive intervention based on the risk of disease progression could be optimally recommended about 2.3 years after initiated surveillance.

Aged↗
PubMed2006

Antibacterial use and Streptococcus pneumoniae penicillin resistance: A temporal relationship model.

The nature of the temporal relationship between antibacterial consumption and Streptococcus pneumoniae penicillin resistance is investigated using population level data across time. IMS Health Global Services provided national outpatient antibiotic prescription data for the years 1996-2003 from France, Spain, Italy, Germany, the United Kingdom, and the United States. Surveillance data consist of S. pneumoniae isolates obtained from a surveillance database in the same geographic regions from 1996 to 2003. A linear mixed model for repeated measures was used to analyze the association between resistance and several antibacterial classes through time. Changes in penicillin resistance through time in any country are better explained by the weighted cumulative antibacterial consumption with a 2-year lag. Narrow-spectrum penicillins are associated with lower resistance rates. Large reductions in consumption at the population level are needed to affect resistance. There is a peak level of penicillin resistance associated with cumulative exposure to a combination of antibiotic classes that is unique for every country.

Anti-Bacterial Agents↗
PubMed2005

Extracellular N-acetylaspartate depletion in traumatic brain injury.

N-Acetylaspartate (NAA) is almost exclusively localized in neurons in the adult brain and is present in high concentration in the CNS. It can be measured by proton magnetic resonance spectroscopy and is seen as a marker of neuronal damage and death. NMR spectroscopy and animal models have shown NAA depletion to occur in various types of chronic and acute brain injury. We investigated 19 patients with traumatic brain injury (TBI). Microdialysis was utilized to recover NAA, lactate, pyruvate, glycerol and glutamate, at 12-h intervals. These markers were correlated with survival and a 6-month Glasgow Outcome Score. Eleven patients died and eight survived. A linear mixed model analysis showed a significant effect of outcome and of the interaction between time of injury and outcome on NAA levels (p = 0.009 and p = 0.004, respectively). Overall, extracellular NAA was 34% lower in non-survivors. A significant non-recoverable fall was observed in this group from day 4 onwards, with a concomitant rise in lactate-pyruvate ratio and glycerol. These results suggest that mitochondrial dysfunction is a significant contributor to poor outcome following TBI and propose extracellular NAA as a potential marker for monitoring interventions aimed at preserving mitochondrial function.

Adult↗
PubMed2005

Home advantage in speed skating: evidence from individual data.

Home advantage is a well-documented phenomenon in many sports. Home advantage has been shown to exist for team sports (soccer, hockey, football, baseball, basketball) and for countries organizing sports tournaments like the Olympics and World Cup Soccer. There is also some evidence for home advantage in some individual sports, but there is a much more limited literature. This paper addresses the issue of home advantage in speed skating. From a methodological point of view, it is difficult to identify home advantage, because skaters vary in their abilities and the conditions of tournaments vary. There is a small but significant home advantage using a generalized linear mixed model, with random effects for skaters and fixed effects for skating rinks and seasons. Even though the home advantage effect exists, it is very small when compared to variation in skating times due to differences of rinks and individual abilities.

Competitive Behavior↗
PubMed2004

The analysis of repeated 'direct' measures of change illustrated with an application in longitudinal imaging.

The use of repeated measures of an outcome variable to improve statistical power and precision in randomized clinical trials and cohort studies is well documented. Linear mixed models have great utility in the analysis of such studies in many medical applications including imaging. However, in imaging studies and other applications the basic outcome can be a 'direct' measure of change in a variable, as opposed to a difference calculated by subtraction of one measured value from another. The correlation structure of such repeated measures of 'direct' change, in particular the non-independence of within-person consecutive measures, adds complexity to the analysis. In this paper, we present a family of hierarchical mixed models for the analysis of such data and explain how to implement them using standard statistical software. We illustrate the use of our models with data from a cohort of patients with Alzheimer's disease.

Alzheimer Disease↗
PubMed2004

Variation in sleepiness during early morning shifts: a mixed model approach to an experimental field study of train drivers.

The present study aimed to experimentally evaluate the effect of early morning shifts on sleep and sleepiness of train drivers during normal working conditions. A total of 17 experienced train drivers were studied during a 4.5 h drive in two directions with a 2.5 h break in between on three different shifts: an early shift that started at 05:49 h (train left at 06:18 h) and ended at 17:41h, a day shift (07:49-19:41 h), and an evening shift (09:49-21:41 h). Retrospective (since the last stop) ratings of mean sleepiness and peak sleepiness (Karolinska Sleepiness Scale--KSS: 1 = very alert, 9= very sleepy, fighting sleep, difficulty staying awake) were assessed at each stop during the drive. The results showed that sleep length was reduced (p <0.001) by 1 h and 2h, respectively, by the early shift compared to the day and evening shifts. The prevalence of severe sleepiness (KSS > or = 7) was high, especially during the early shift when 14 (82%) subjects reported at least one event during the drive. Application of the Generalized Linear Mixed Models (GLMM) to the sleepiness data showed that there was an increased risk for severe sleepiness during the early shift (OR = 4.9) that increased further with the length of the drive between stops (OR = 1.9, 15 min), suggesting an interaction between early morning shift and monotony. The findings have practical implications in risk assessment. Long drives without stops and other monotonous situations should have a higher risk rating for severe sleepiness in shifts with an early start before 06:00 h, compared to shifts that begin 2 h later.

Adult↗
PubMed2001

Bayesian modeling of the level and duration of fertility in the menstrual cycle.

Time to pregnancy studies that identify ovulation days and collect daily intercourse data can be used to estimate the day-specific probabilities of conception given intercourse on a single day relative to ovulation. In this article, a Bayesian semiparametric model is described for flexibly characterizing covariate effects and heterogeneity among couples in daily fecundability. The proposed model is characterized by the timing of the most fertile day of the cycle relative to ovulation, by the probability of conception due to intercourse on the most fertile day, and by the ratios of the daily conception probabilities for other days of the cycle relative to this peak probability. The ratios are assumed to be increasing in time to the peak and decreasing thereafter. Generalized linear mixed models are used to incorporate covariate and couple-specific effects on the peak probability and on the day-specific ratios. A Markov chain Monte Carlo algorithm is described for posterior estimation, and the methods are illustrated through application to caffeine data from a North Carolina pregnancy study.

Algorithms↗
PubMed1987

Organic geochemistry of sediments from the continental margin off southern New England, U.S.A.--Part I. Amino acids, carbohydrates and lignin.

Total organic carbon (TOC), lignin, amino acids, sugars and amino sugars were measured in recent sediments for the continental margin off southern New England. The various organic carbon fractions decreased in concentration with increasing distance from shore. The fraction of the TOC that was accounted for by these major components also decreased with increasing distance from shore. The concentration of lignin indicated that only about 3-5% of the organic carbon in the nearshore sediment was of terrestrial origin. The various fractions were highly correlated, which was consistent with a simple linear mixing model of shelf organic matter with material form the slope and rise and indicated a significant transport of sediment from the continental shelf to the continental slope and rise.

Amino Acids↗
PubMed2006

Estimation of genetic and environmental factors for melanoma onset using population-based family data.

Estimation of genetic and environmental contributions to cancers falls in the framework of generalized linear mixed modelling with several random effect components. Computational challenges remain, however, in dealing with binary or survival phenotypes. In this paper, we consider the analysis of melanoma onset in a population of 2.6 million nuclear families in Sweden, for which none of the current survival-based methodologies is feasible. We treat the disease outcome as a binary phenotype, so that the standard proportional hazard model leads to a generalized linear model with the complementary-log link function. For rare diseases this link is very close to the probit link, and thus allows the use of marginal likelihood for the estimation of the variance components. We correct for the survival length bias by censoring the parent generation within each family at the time they attain the same cumulative hazard as the child generation, thus improving the validity of the estimates. Our finding that childhood shared environment in addition to genetic factors had a considerable effect on the development of melanoma is consistent with epidemiological studies.

Adolescent↗
PubMed2005

Coinfection with Plasmodium falciparum and schistosoma haematobium: protective effect of schistosomiasis on malaria in senegalese children?

Studies with animal models have suggested the possibility of interactions between parasites during concurrent infections and have raised the question of a similar phenomenon in humans. The present survey was undertaken to assess the impact of urinary schistosomiasis on the susceptibility of children to malaria. It was carried out in Senegal between September 2001 and March 2002 among 523 children 3-15 years of age. We tested the association between Plasmodium falciparum densities and the load of Schistosoma haematobium egg excretion using a linear mixed model because data were not independent. After controlling for age, sex, and season, we showed that children lightly infected with S. haematobium (1-9 eggs/10 mL of urine) had lower P. falciparum densities than those not infected (beta = -0.34, 95% confidence interval = -0.85, -0.10), suggesting a negative interaction between both parasites.

Adolescent↗
PubMedDate not supplied

Oxygen consumption monitoring by oxygen saturation measurements in mechanically ventilated premature neonates.

PURPOSE: To assess oxygen consumption VO2) with arterial and venous oxygen saturation SaO2 and SvO2) from blood measurements for continuous monitoring of mechanically ventilated preterm neonates. METHOD: Twelve preterm neonates with gestational ages ranging from 27 to 34 weeks at birth, who were mechanically ventilated and had umbilical arterial and venous lines in place, were investigated. Analyses were performed with blood samples and continuous monitoring of VO2 from SaO2 and SvO2 measurements. RESULTS: Mean VO2 from blood samples was 3.3 mL/(kg min) (+/-2.49), and that from monitor readings was 8.8 (+/-4.49). Using curve fit analysis to predict the stability of monitor reading, the period from 8 to 28 hours following blood validation was most stable for SvO2 (cubic curve, R=0.5, P<.001). There is a considerable mix of arterial blood in the venous blood, which increased SvO2 and decreased VO2 in the blood samples during first 8 hours of monitoring following blood draws. A multivariate linear mixed model was established for VO2 measurements including related parameters. Ventilatory weaning was associated with decreased VO2. CONCLUSIONS: These findings warrant caution against interpreting VO2 from blood SvO2 and subsequent monitoring readings during mechanical ventilation for preterm neonates.

Arteries↗
PubMed2006

Maternal depression and child internalizing: the moderating role of child emotion regulation.

This study tests a model of children's emotion regulation (ER) as a moderator of the link between maternal depression and child internalizing problems. Participants were 78 children (ages 4 to 7), including 45 children of mothers with a history of childhood-onset depression (COD) and 33 children of mothers who had never been depressed. ER was assessed observationally during a laboratory mood induction. ER behaviors were empirically reduced into 3 categories: (a) negative focus on delay, (b) positive reward anticipation, and (c) behavioral distraction. Linear mixed models indicated that positive reward anticipation moderated the effects of maternal COD on children's internalizing problems, particularly if mothers had current depressive symptoms. Findings suggest that generating positive affect in the face of a potential frustration may be a protective ER strategy for children at risk for depression.

Adult↗
PubMed2004

A population pharmacokinetic model with time-dependent covariates measured with errors.

We propose a population pharmacokinetic (PK) model with time-dependent covariates measured with errors. This model is used to model S-oxybutynin's kinetics following an oral administration of Ditropan, and allows the distribution rate to depend on time-dependent covariates blood pressure and heart rate, which are measured with errors. We propose two two-step estimation methods: the second-order two-step method with numerical solutions of differential equations (2orderND), and the second-order two-step method with closed form approximate solutions of differential equations (2orderAD). The proposed methods are computationally easy and require fitting a linear mixed model at the first step and a nonlinear mixed model at the second step. We apply the proposed methods to the analysis of the Ditropan data, and evaluate their performance using a simulation study. Our results show that the 2orderND method performs well, while the 2orderAD method can yield PK parameter estimators that are subject to considerable biases.

Biometry↗
PubMed2004

Mycophenolate mofetil in IgA nephropathy: results of a 3-year prospective placebo-controlled randomized study.

BACKGROUND: Because humoral immunity is believed to play a pivotal role in the pathogenesis of IgA nephropathy (IgAN), a prospective placebo-controlled randomized study was started in patients with IgAN using mycophenolate mofetil (MMF). METHODS: A total of 34 patients with IgAN were treated with salt intake restriction, angiotensin-converting enzyme (ACE) inhibition and MMF 2 g per day (N= 21) or placebo (N= 13). After 36 months of follow-up clinical, biochemical, and radiologic data were analyzed using linear mixed models for longitudinal data and Kaplan-Meier survival analysis. RESULTS: Therapy had to be stopped prematurely in five patients. Two patients (MMF group) evolved to end-stage renal disease (ESRD). There was no difference between groups in the percentage of patients with a decrease of 25% or more in the inulin clearance or with a serum creatinine increase of 50% or more over 3 years. There was also no significant difference between groups in annualized rate of change of serum creatinine, computed by linear regression analysis. No significant difference was noted between groups for inulin clearance, serum creatinine, proteinuria, blood pressure, or other parameters of renal function. Hemoglobin and C-reactive protein were significantly lower in the MMF group compared with the placebo group. As a function of time, a significant decline in both groups was noted of proteinuria, parenchymal thickness of the kidneys and C3d. CONCLUSION: In patients with IgAN at risk for progressive disease, no beneficial effect of 3-year treatment with MMF 2 g per day could be demonstrated on renal function/outcome or proteinuria. However, larger randomized studies are needed to confirm or reject these results.

Adult↗
PubMed2025

Genetic Differences in Reactivity to the Environment Impact Psychotic-Like and Affective Reactivity in Daily Life.

BACKGROUND AND HYPOTHESIS: Consistent with diathesis-stress models, psychosis research has focused on genetic moderation of adverse environmental exposures. In contrast, the Differential Susceptibility (DS) model suggests that the same genetic variants that increase risk-inducing effects of adverse experiences also enhance beneficial effects from positive experiences. This study examined whether individuals with high genetic susceptibility to the environment showed differential psychotic-like and affective reactivity in response to positive and negative events in daily life. STUDY DESIGN: Experience sampling methodology assessed context (positive and stressful) and momentary levels of paranoia, psychotic-like experiences (PLE), and positive (PA) and negative affect (NA) in 217 non-clinical adults oversampled for schizotypy. Linear mixed models examined whether Polygenic Risk Scores of Environmental Sensitivity (PRS-ES) moderated the impact of current context on subsequent experiences. STUDY RESULTS: PRS-ES moderated positive, but not stressful, context on subsequent levels of momentary paranoia, NA, and PA, but not PLE. Genetic and environmental (G&#x2005;&#xd7;&#x2005;E) interactions indicated diathesis-stress at lower thresholds of PRS-ES, but a DS model at the highest threshold of the PRS-ES. Participants with elevated PRS-ES showed increased paranoia and NA and decreased PA in subsequent assessments when reporting low levels of positive situations, but also decreased paranoia and NA and increased PA when rating contexts as positive. CONCLUSIONS: Findings support the influence of genetic sensitivity to the environment on psychotic-like and affective reactivity in daily life, particularly in response to positive contexts. This highlights the transdiagnostic protective role of positive experiences and informs ecological momentary interventions.

Humans↗
PubMed2001

Evaluate multiple adverse events in crossover design bioequivalence clinical trials.

AIM: To establish a statistical model to appropriately evaluate the relationship between clinical adverse event and treatment. METHODS: By splitting the duration of each treatment period into several time intervals (use day as unit), a clinical adverse event was sampled in each time interval as presented or not presented. The number of presented cases was added for each time interval, and time of subjects spent in each time interval was cumulated as person days. The onset of a clinical adverse event and its duration were represented as repeatedly measured count data. By using the generalized linear mixed model with fixed- and random-effects, the relative rate of clinical adverse events relative to different treatments was modeled by the generalized estimate equation (GEE) technique. RESULTS: Example shows that not only the onset of adverse events, but also its duration and total person-days subjects spent in study would influence the relative rate of clinic adverse events. CONCLUSION: Our proposed approach is a good alternative and supplemental method for evaluating clinical adverse events.

Clinical Trials as Topic↗
PubMed2004

Effect of season and outdoor climate on litter size at birth in purebred landrace and yorkshire sows in Thailand.

The objective of the present study was to retrospectively investigate causes of variation on litter size at birth (total number of piglets born per litter (TB) and number of piglets born alive per litter (BA)) of Landrace (L) and Yorkshire (Y) sows in swine nucleus herds in Thailand. The data included sows farrowed during a four-year period from January 1998 to December 2001. The analyzed data set included observations on 8020 litters from 2199 L sows and 6919 litters from 1680 Y sows. Analysis of variance (ANOVA) was applied for statistical analyses using General Linear Mixed Model (MIXED) procedure of SAS. No breed difference was found for both TB and BA. Farrowing months significantly influenced TB and BA (P<0.001). Sows farrowed in August and September had a lower BA than sows farrowed from November to June (P<0.05). Effect of farrowing months on both TB and BA was more pronounced in primiparous compared with multiparous sows. Average minimum daily temperature during gestation negatively correlated with both TB and BA, average maximum daily temperature during gestation negatively correlated with BA and average daily humidity during gestation negatively correlated with both TB and BA. The correlations were stronger in L than in Y sows.

Analysis of Variance↗
PubMed2007

Improving point predictions of random effects for subjects at high risk.

The prediction of random effects corresponding to subject-specific characteristics (e.g. means or rates of change) can be very useful in medical and epidemiologic research. At times, one may be most interested in obtaining accurate and/or precise predictions for subjects whose characteristic places them in a tail of the distribution. While the typical posterior mean predictor dominates others in terms of overall mean squared error of prediction (MSEP), its tendency to 'overshrink' has motivated research into alternatives emphasizing other criteria. Here, we specifically target MSEP within a certain region (e.g. above a known cut-off for high risk or a specified percentile of the random effect distribution), and we consider minimizing this quantity with and without constraints on overall MSEP efficiency. We use the normal-theory random intercept model to derive prediction methods with potential to yield markedly better performance for subjects in the specified region, given a well-controlled and (if desired) modest concession of overall MSEP. Criteria geared toward classification as well as overall and regional prediction unbiasedness are also provided. We evaluate the proposed techniques and illustrate them using repeated measures data on fasting blood glucose from type 2 diabetes patients. A simulation study verifies that theoretical properties and relative performances of the proposed predictors are essentially maintained when calculating them in practice based on estimated mixed linear model parameters. Straightforward extensions to incorporate covariates and additional random effects are briefly outlined.

Bias↗