Longitudinal Analysis of Pulmonary Function in Survivors of Congenital Diaphragmatic Hernia.

OBJECTIVE: To analyze longitudinal trends of pulmonary function testing in patients with congenital diaphragmatic hernia (CDH) followed in our multidisciplinary clinic. STUDY DESIGN: This was a retrospective cohort study of CDH patients born between 1991 and 2013. A linear mixed effects model was fitted to estimate the trends of percent predicted forced expiratory volume in 1 second (FEV1pp), percent predicted forced vital capacity (FVCpp), and FEV1/FVC over time. RESULTS: Of 268 patients with CDH who survived to discharge, 119 had at least 1 pulmonary function test study. The FEV1pp (P < .001), FVCpp (P = .017), and FEV1/FVC (P = .001) decreased with age. Compared with defect size A/B, those with defect size C/D had lower FEV1pp by an average of 11.5% (95% CI, 2.9%-20.1%; P = .010). A history of oxygen use at initial hospital discharge also correlated with decreased FEV1pp by an average of 8.0% (95% CI, 1.2%-15.0%; P = .023). CONCLUSIONS: In a select cohort of CDH survivors, average pulmonary function declines with age relative to expected population normative values. Those with severe CDH represent a population at risk for worsening pulmonary function test measurements who may benefit from recognition and monitoring for complications.

The geometry of the lactation curve based on Wood's equation: a two-step prediction

Lactation records from cows of the southwestern Paraná state, Brazil, form the dataset of this study. We applied the information-theoretic approach to evaluate the ability of the nonlinear Wood, Brody, Dijkstra, and Gamma functions to fit to these data by employing a two-step technique based on nonlinear mixed-effects models and generalized linear mixed-effects models. Wood's equation was fitted with the combination of a first-order autoregressive correlation structure and a variance function to account for heteroscedasticity. This version was the best choice to mimic lactation records. Some geometric attributes of Wood's model were deduced, mainly the ascending specific rate from parturition to peak milk yield and the descending specific rate as a measure of the lactation persistence of the milk yield at peak production. Breed and parity order of the cows were assumed as fixed effects to obtain a reliable model fitting process. Regardless of breed, first-order parity cows had greater persistency than their older counterparts, and the greater the ascending rate of milk yield from the parturition to the peak, the sharper the decrease in milk yield post-peak; therefore, the rates (absolute values) of ascending and descending phases correlated positively. Nonetheless, the actual estimated values of the descending phase rates are negative. Wood's equation was flexible enough to mimic either concave- and convex-shaped lactation profiles. The correlations between both peak milk yield and random estimates for ß with total milk yield per lactation were positive. However, peak milk yield might not be the only variable used for ranking cows; the total milk yield integrates all information of the lactation profile through the estimated parameters of Wood's equation.(AU)

Influence assessment in censored mixed-effects models using the multivariate Student's-t distribution.

In biomedical studies on HIV RNA dynamics, viral loads generate repeated measures that are often subjected to upper and lower detection limits, and hence these responses are either left- or right-censored. Linear and non-linear mixed-effects censored (LMEC/NLMEC) models are routinely used to analyse these longitudinal data, with normality assumptions for the random effects and residual errors. However, the derived inference may not be robust when these underlying normality assumptions are questionable, especially the presence of outliers and thick-tails. Motivated by this, Matos et al. (2013b) recently proposed an exact EM-type algorithm for LMEC/NLMEC models using a multivariate Student's-t distribution, with closed-form expressions at the E-step. In this paper, we develop influence diagnostics for LMEC/NLMEC models using the multivariate Student's-t density, based on the conditional expectation of the complete data log-likelihood. This partially eliminates the complexity associated with the approach of Cook (1977, 1986) for censored mixed-effects models. The new methodology is illustrated via an application to a longitudinal HIV dataset. In addition, a simulation study explores the accuracy of the proposed measures in detecting possible influential observations for heavy-tailed censored data under different perturbation and censoring schemes.