Bias correction in random effects models with sparse binary responses.

Sparse correlated binary data are frequently encountered in many applications involving either rare event cases or small sample sizes. In this study, we consider correlated binary data and a logit random effects model framework. We discuss h-likelihood estimates and how the computational procedure is affected by sparseness. We propose an adjustment to the fitting process that involves the adaption of the regression calibration method to the estimation of random effects. Using this adjustment, we correct for the bias in the random effects estimates resulting in better properties for the fixed effects estimates of the model. This is supported by the results of the simulation study that was conducted under different sparseness levels. The proposed adjusted h-likelihood estimation approach is also used for the analysis of two real meta-analyses data sets.

Use of the Lagrange Multiplier Test for Assessing Measurement Invariance Under Model Misspecification.

This article studies the Type I error, false positive rates, and power of four versions of the Lagrange multiplier test to detect measurement noninvariance in item response theory (IRT) models for binary data under model misspecification. The tests considered are the Lagrange multiplier test computed with the Hessian and cross-product approach, the generalized Lagrange multiplier test and the generalized jackknife score test. The two model misspecifications are those of local dependence among items and nonnormal distribution of the latent variable. The power of the tests is computed in two ways, empirically through Monte Carlo simulation methods and asymptotically, using the asymptotic distribution of each test under the alternative hypothesis. The performance of these tests is evaluated by means of a simulation study. The results highlight that, under mild model misspecification, all tests have good performance while, under strong model misspecification, the tests performance deteriorates, especially for false positive rates under local dependence and power for small sample size under misspecification of the latent variable distribution. In general, the Lagrange multiplier test computed with the Hessian approach and the generalized Lagrange multiplier test have better performance in terms of false positive rates while the Lagrange multiplier test computed with the cross-product approach has the highest power for small sample sizes. The asymptotic power turns out to be a good alternative to the classic empirical power because it is less time consuming. The Lagrange tests studied here have been also applied to a real data set.

Beyond the Words: Comparing Interpersonal Engagement Between Maternal and Paternal Infant-Directed Speech Acts.

The present study investigates the way infants express their emotions in relation to parental feelings between maternal and paternal questions and direct requests. We therefore compared interpersonal engagement accompanying parental questions and direct requests between infant-mother and infant-father interactions. We video-recorded spontaneous communication between 11 infant-mother and 11 infant-father dyads-from the 2nd to the 6th month-in their home. The main results of this study are summarized as follows: (a) there are similarities in the way preverbal infants use their affections in spontaneous interactions with their mothers and fathers to express signs of sensitivity in sharing knowledge through questions and direct requests; and (b) the developmental trajectories of face-to-face emotional coordination in the course of parental questions descend in a similar way for both parents across the age range of this study. Regarding the developmental trajectories of emotional non-coordination, there is evidence of a linear trend in terms of age difference between the parents' gender with fathers showing the steeper slope. The results are discussed in relation to the theory of intersubjectivity.

Weighted cumulative sum tests for random effect models with binary responses.

Correlated binary responses are very commonly encountered in many disciplines like, for example, medical studies. The development of goodness-of-fit tests is essential for examining the adequacy of the fitted models. The objective of this article is to provide weighted modifications of cumulative sums or moving cumulative sums of residuals for testing goodness-of-fit of random effects logistic regression models. The proposed weights can be interpreted as the residuals of a weighted linear regression of an omitted covariate on the covariates already included in the fixed part of the model. These processes lead to supremum statistics whose null distribution is derived using simulation. Results from a simulation study suggest better performance of the weighted when compared to the unweighted supremum statistics. The proposed tests are illustrated using a real data example.

Goodness of fit tests for random effect models with binary responses.

Correlated binary responses are very common in longitudinal and repeated measures studies. Random effects models are often used to analyze such data and the h-likelihood estimating procedure provides an inferential tool. The objective of this study is to introduce goodness-of-fit score statistics for the fixed part of these models. The proposed statistics are based on processes that partition the observations into mutually exclusive groups. Weighted versions of these statistics are also introduced and are based on the correlation between an appropriately adjusted candidate covariate for entrance into the model and the model residuals. A simulation study indicates that the weighted statistics perform better than their unweighted counterparts, whereas the statistics that are based on the partitioning of the covariate space seem to perform slightly better compared with those based on other grouping procedures. The use of the proposed statistics is illustrated using a real data example.

Alteration of α-N-acetylgalactosaminidase (nagalase) concentration in alcohol-dependent individuals without liver disease, during the detoxification therapy.

BACKGROUND: The present study aimed to investigate for the first time, the alteration of α-N-acetylgalactosaminidase (nagalase) concentration in alcohol-dependent individuals without liver disease, before, during and at the end of the detoxification therapy. METHODS: Forty-eight alcohol-dependent individuals without liver disease who were admitted for alcohol detoxification, and eighty-four healthy controls participated in this study. Patients' blood was obtained upon admission, two weeks later and after the completion of the detoxification period (4-5 weeks). Nagalase concentration in serum was assessed by enzyme-linked immunosorbent assay. RESULTS: Nagalase concentration was significantly elevated in the patient samples in all serum collections as compared to the normal controls, with a progressive fall from admission to discharge (p-value<0.001). Values differed significantly among the three time points, with a net shift to decrease, but remained still high, above normal control level at the end of the therapy. No significant correlations were detected among the nagalase levels and the liver enzymes values. Moreover, no significant correlation was found between the alterations of nagalase concentrations and the amount of consumed alcohol. CONCLUSIONS: The high nagalase concentrations in alcohol abuse might be associated with macrophage impairment through decreasing the endogenous macrophage-activating factor (MAF) production by Gc-protein. The possible pathogenetic association between nagalase activity and alcohol overconsumption remains a matter of further investigation. Nagalase could also serve as a marker of alcohol overconsumption for the evaluation of alcohol-dependent individuals before, as well as during the detoxification therapy.

Characteristics of the Human Upper Gastrointestinal Contents in the Fasted State Under Hypo- and A-chlorhydric Gastric Conditions Under Conditions of Typical Drug - Drug Interaction Studies.

OBJECTIVE: Evaluate the impact of reduced gastric acid secretion after administration of two acid-reducing agents on the physicochemical characteristics of contents of upper gastrointestinal lumen of fasted adults. MATERIALS AND METHODS: Eight healthy male adults, fasted from food for 12 h, participated in a three-phase crossover study. Phase 1: No drug treatment prior to aspirations. Phase 2: Oral administration of 40 mg pantoprazole at ~9 am the last 3 days prior to aspirations and at ~7 am on aspiration day. Phase 3: Oral administration of 20 mg famotidine at ~7 pm prior to aspirations and at ~7 am on aspiration day. Samples from the contents of upper gastrointestinal lumen were aspirated for 50 min, after administration of 240 ml table water at ~9 am. RESULTS: Reduction of gastric acid secretion was accompanied by reduced buffer capacity, chloride ion concentration, osmolality and surface tension in stomach and by increased pH (up to ~0.7 units) in upper small intestine during the first 50 min post-water administration. The mechanism of reduction of acid secretion seems to be important for the buffer capacity in stomach and for the surface tension in upper gastrointestinal lumen. CONCLUSIONS: Apart from gastric pH, reduced acid secretion affects physicochemical characteristics of contents of upper gastrointestinal lumen which may be important for the performance of certain drugs/products in the fasted state.

Weighted pairwise likelihood estimation for a general class of random effects models.

Models with random effects/latent variables are widely used for capturing unobserved heterogeneity in multilevel/hierarchical data and account for associations in multivariate data. The estimation of those models becomes cumbersome as the number of latent variables increases due to high-dimensional integrations involved. Composite likelihood is a pseudo-likelihood that combines lower-order marginal or conditional densities such as univariate and/or bivariate; it has been proposed in the literature as an alternative to full maximum likelihood estimation. We propose a weighted pairwise likelihood estimator based on estimates obtained from separate maximizations of marginal pairwise likelihoods. The derived weights minimize the total variance of the estimated parameters. The proposed weighted estimator is found to be more efficient than the one that assumes all weights to be equal. The methodology is applied to a multivariate growth model for binary outcomes in the analysis of four indicators of schistosomiasis before and after drug administration.

A Composite Likelihood Inference in Latent Variable Models for Ordinal Longitudinal Responses.

The paper proposes a composite likelihood estimation approach that uses bivariate instead of multivariate marginal probabilities for ordinal longitudinal responses using a latent variable model. The model considers time-dependent latent variables and item-specific random effects to be accountable for the interdependencies of the multivariate ordinal items. Time-dependent latent variables are linked with an autoregressive model. Simulation results have shown composite likelihood estimators to have a small amount of bias and mean square error and as such they are feasible alternatives to full maximum likelihood. Model selection criteria developed for composite likelihood estimation are used in the applications. Furthermore, lower-order residuals are used as measures-of-fit for the selected models.

The impact of comorbidity and other clinical and sociodemographic factors on health-related quality of life in Greek patients with Parkinson's disease.

OBJECTIVES: This study was designed to evaluate the impact of other common self-reported comorbid disorders (hypertension, dyslipidemia, ischemic heart disease, diabetes mellitus, minor stroke, arthritis, low back pain or osteoporosis and depression) on health-related Quality of Life (HRQoL) of Parkinson's disease (PD) patients and to explore the association of their HRQoL with various sociodemographic and clinical factors. METHODS: Data about age, gender, education, occupation, income, marital and residential status, social relations, disease duration, functional status, treatment and concomitant diseases were collected of 139 Greek patients (68 men and 71 women) with PD. Patients were consecutively recruited from the outpatient clinic of the first Neurology Department of Athens National University at Aeginition Hospital. Disease severity was assessed using the unified Parkinson's disease rating scale including Hoehn and Yahr and Schwab and England (S&E) scales. HRQoL was measured by the specific Parkinson's disease questionnaire (PDQ-39). A multivariate multiple regression model with normal errors was used for the statistical analysis. RESULTS: The main determinants of HRQoL were low degree of independence measured by the S&E scale (F = 35.942, p < 0.001), social isolation (F = 20.508, p < 0.001), disease duration (F = 14.983, p < 0.001), sleep (F = 6.507, p = 0.013) and gastrointestinal disturbances (F = 4.643, p = 0.035) and the presence of depression (F = 6.022, p = 0.017). CONCLUSION: Among the other chronic comorbidities only depression was associated with a poor HRQoL in PD patients. Functional dependence and social isolation contributed most to worse HRQoL. Our findings suggest that adequate social support and management of depression, sleep and gastrointestinal disturbances could reduce the distress and improve HRQoL in patients with PD.

The effect of diode laser (980 nm) treatment on aggressive periodontitis: evaluation of microbial and clinical parameters.

OBJECTIVE: The aim was to compare the effect of scaling and root planing (SRP) alone, diode laser treatment (LAS) alone, and SRP combined with LAS (SRP + LAS) on clinical and microbial parameters in patients with aggressive periodontitis. MATERIALS AND METHODS: Thirty patients with aggressive periodontitis were assessed for plaque, bleeding on probing (BOP), probing pocket depth (PPD), and clinical attachment level (CAL). Four plaque samples were randomly obtained, one from each quadrant that was randomly assigned to SRP alone, SRP + LAS, LAS alone, and control (CRL). A 980-nm diode laser was used in continuous mode at 2 W power. Plaque samples were collected 2 wk, 12 wk, and 6 mo post-treatment. The levels of Porphyromonas gingivalis, Tannerella forsythia, Aggregatibacter actinomycetemcomitans, Treponema denticola, and total bacterial load (TBL) were evaluated using ssrRNA probes. RESULTS: Bacterial counts were decreased with all three treatment modalities and they did not reach baseline levels at 6 mo post-treatment. The SRP + LAS group showed statistically significantly lower TBL and bacterial levels of P. gingivalis and T. denticola at 6 mo post-treatment compared to SRP or LAS treatments alone. At the end of the observation period significant differences were observed for PPD and CAL between the SRP + LAS group and both the SRP alone and LAS alone groups. No differences were detected for percentage of plaque and percentage of BOP between any of the treatment groups at 6 mo post-treatment. CONCLUSIONS: Within the limits of this study, diode laser-assisted treatment with SRP showed a superior effect over SRP or LAS alone for certain microbial and clinical parameters in patients with aggressive periodontitis over the 6-mo monitoring period.

Latent variable models for multivariate longitudinal ordinal responses.

The paper proposes a full information maximum likelihood estimation method for modelling multivariate longitudinal ordinal variables. Two latent variable models are proposed that account for dependencies among items within time and between time. One model fits item-specific random effects which account for the between time points correlations and the second model uses a common factor. The relationships between the time-dependent latent variables are modelled with a non-stationary autoregressive model. The proposed models are fitted to a real data set.

Detection of gingival crevicular fluid cytokines in children and adolescents with and without fixed orthodontic appliances.

OBJECTIVE: To study the expression of IL-1beta, IL-4, and IL-8 in the gingival crevicular fluid (GCF) of children, adolescents, and young adults with and without fixed orthodontic appliances. MATERIAL AND METHODS: Eighty systemically healthy children and adolescents participated in the study: 56 aged between 8 and 16 years without any orthodontic appliance (Group A) and 24 aged between 10 and 20 years having worn fixed orthodontic appliances for at least 12 months (Group B). Clinical examination included presence or absence of plaque, probing depth, bleeding on probing, and gingival overgrowth. GCF was collected by means of Durapore strips from four randomly selected sites per subject. The contents of interleukin-1 beta (IL-1beta), interleukin-4 (IL-4), and interleukin-8 (IL-8) were detected by ELISA, measured as total amounts (pg/30s) and expressed in log scale. RESULTS: Statistically significant differences were noted for the mean log IL-1beta, IL-4, and IL-8 between the two groups: Group B showed significantly higher mean levels in log IL-1beta and log IL-8 compared to Group A. Mean levels of log IL-4 were lower in Group B, although they did not reach statistical significance. Furthermore, mean levels of log IL-1beta and log IL-8 were associated with bleeding sites (p<0.001) and gingival overgrowth, while mean level of log IL-4 was associated with non-bleeding sites and no gingival overgrowth (p<0.001). CONCLUSION: Our findings suggest that fixed orthodontic appliances result in an increase in the expression of IL-1beta and IL-8. This may reflect biologic activity in the periodontium during orthodontic tooth movement.

Cytokine profile in gingival crevicular fluid of aggressive periodontitis: influence of smoking and stress.

BACKGROUND: Cigarette smoking and stress are considered risk factors that have been associated with periodontal disease progression. Conflicting results have been reported concerning the direct influence of smoking on the subgingival microbiota of periodontitis patients. Cytokine production may also be influenced by smoking and stress leading to an imbalance that disturbs the host-parasite relationship. AIM: The objective of the present study was to evaluate the influence of cigarette smoking on the gingival crevicular fluid (GCF) levels of interleukin (IL)-1beta, IL-4, IL-6 and IL-8 in aggressive or early onset periodontitis (EOP) patients and in healthy controls (H), psychosocial stress being considered as modifying factor. MATERIAL AND METHODS: Sixty-five EOP and 35 periodontally healthy individuals participated in this cross-sectional study. All the participants were interviewed about their smoking habits and their stressful social events. Clinical examination included the assessment of plaque index (PI), bleeding on probing (BOP), clinical attachment level (CAL) and probing pocket depth (PPD). GCF was collected using durapore strips, from four sites per patient, randomly selected in each quadrant. The total amounts of IL-1beta, IL-4, IL-6 and IL-8 were measured in a total of 400 samples using commercially available enzyme-linked immunosorbent assays. RESULTS: All clinical parameters were significantly higher in the EOP group compared to the H group. There were no significant differences between EOP smokers and EOP non-smokers with regard to plaque accumulation, CAL and PPD of the sampling sites, whereas mean CAL and PPD of the diseased sites were greater in EOP smokers than in EOP non-smokers. In addition, EOP smokers seemed to have significantly less BOP and greater bone loss compared to EOP non-smokers. Significant interactions between "EOP" and "smoking" were present for total amounts of IL-1beta and IL-4. IL-1beta, IL-6 and IL-8 showed significant main effects with healthy smokers and healthy non-smokers, respectively. For IL-8, stress presented a statistically significant interaction with smoking status and EOP (F=4.742, p=0.030). More specifically EOP smokers were statistically affected by stress. CONCLUSIONS: Smoking influences host-related factors including cytokine network. The relative importance of smoking and stress-related alterations and their precise mode of action in increasing the risk of aggressive periodontitis remains to be elucidated.