Showing papers by "Xihong Lin published in 2002"
TL;DR: Regression models were used to estimate rates of BMD change and to examine BMD as a function of age, and evidence for age at onset of bone loss at the lumbar spine was inconclusive.
Abstract: The aims of this prospective cohort study were to determine rates of premenopausal and early postmenopausal bone loss, age at onset of bone loss, and whether rates of bone loss depend on baseline bone mineral density (BMD). The cohort of 614 women aged 24-44 years at baseline from the longitudinal Michigan Bone Health Study was followed for 6 years beginning in 1992-1993. Up to five BMD measurements of the lumbar spine (L(2-4)) and the femoral neck were obtained through 1998-1999 by using dual x-ray absorptiometry and were standardized (as z scores) relative to a young adult, female BMD distribution. Regression models were used to estimate rates of BMD change and to examine BMD as a function of age. At the lumbar spine, the rate of BMD change for premenopausal women varied with time. At the femoral neck, the rate of change was -1.6% (95% confidence interval: -0.9%, -2.3%) of a z score annually (annual loss of 0.3% of baseline BMD (g/cm(2))). Evidence for age at onset of bone loss at the lumbar spine was inconclusive. Bone loss began by the midtwenties at the femoral neck. Additional annual change of -0.7% (95% confidence interval: -0.2%, -1.2%) of a z score was observed at the femoral neck for each unit increase in BMD z score at baseline.
103 citations
TL;DR: Evidence is provided suggesting that for the marginal model, marginal smoothing and penalized regression splines are not local in their behavior, and evidence suggests that when using spline methods, it is worthwhile to account for the correlation structure.
Abstract: We consider nonparametric regression in a longitudinal marginal model of generalized estimating equation (GEE) type with a time-varying covariate in the situation where the number of observations per subject is finite and the number of subjects is large. In such models, the basic shape of the regression function is affected only by the covariate values and not otherwise by the ordering of the observations. Two methods of estimating the nonparametric function can be considered: kernel methods and spline methods. Recently, surprising evidence has emerged suggesting that for kernel methods previously proposed in the literature, it is generally asymptotically preferable to ignore the correlation structure in our marginal model and instead assume that the data are independent, that is, working independence in the GEE jargon. As seen through equivalent kernel results, in univariate independent data problems splines and kernels have similar behavior; smoothing splines are equivalent to kernel regression with a s...
95 citations
TL;DR: In this article, the authors analyzed changes in treatment practices in outpatient methadone treatment units from a national panel study and found that a substantial percentage of units did not respond during the follow-up.
Abstract: This article analyzes changes in treatment practices in outpatient methadone treatment units from a national panel study. The analysis of this dataset is challenging due to several difficulties, including multiple longitudinal outcomes, nonignorable nonresponses, and missing covariates. Specifically, the data included several variables that measure the effectiveness of methadone treatment practices for each unit. A substantial percentage of units (33%%) did not respond during the follow-up. These dropout units tended to be units with less effective treatment practices; the dropout mechanism thus may be nonignorable. Finally, the time-varying covariates for the units that dropped out were missing at the time of dropout. A valid analysis hence needs to address these three issues simultaneously. Our approach assumes that the observed outcomes measure a latent variable (e.g., treatment practice effectiveness) with error. We model the relationship between this latent variable and covariates using a linear mixe...
92 citations
TL;DR: It is demonstrated that if the ascertainment scheme and data cannot be modeled properly, then the resulting ascertainment-adjusted analysis produces parameter estimates that generally do not reflect the true values in either the original population or the ascertained subpopulation.
Abstract: Ascertainment-adjusted parameter estimates from a genetic analysis are typically assumed to reflect the parameter values in the original population from which the ascertained data were collected. Burton et al. (2000) recently showed that, given unmodeled parameter heterogeneity, the standard ascertainment adjustment leads to biased parameter estimates of the population-based values. This finding has important implications in complex genetic studies, because of the potential existence of unmodeled genetic parameter heterogeneity. The authors further stated the important point that, given unmodeled heterogeneity, the ascertainment-adjusted parameter estimates reflect the true parameter values in the ascertained subpopulation. They illustrated these statements with two examples. By revisiting these examples, we demonstrate that if the ascertainment scheme and the nature of the data can be correctly modeled, then an ascertainment-adjusted analysis returns population-based parameter estimates. We further demonstrate that if the ascertainment scheme and data cannot be modeled properly, then the resulting ascertainment-adjusted analysis produces parameter estimates that generally do not reflect the true values in either the original population or the ascertained subpopulation.
43 citations
TL;DR: Test the hypothesis that IEDs are related to seizures during sleep while adjusting for log delta power (LDP), a continuous measure of sleep depth, to show whether interictal epileptiform discharges increase, decrease, or are unchanged before epileptic seizures has implications for the pathophysiology of epilepsy.
Abstract: Summary: Purpose: Whether interictal epileptiform discharges (IEDs) increase, decrease, or are unchanged before epileptic seizures has implications for the pathophysiology of epilepsy. Prior studies relating IEDs and seizures have not demonstrated a change in IEDs before seizures. However, they have not controlled for changes in the depth of sleep. Our objective was to test the hypothesis that IEDs are related to seizures during sleep while adjusting for log delta power (LDP), a continuous measure of sleep depth. Methods: Twenty-two seizures during sleep were identified in 16 subjects with epilepsy admitted for presurgical monitoring. The IEDs that occurred in the hour of sleep before each seizure were used to test the relation between IEDs and seizure occurrence. Sleep depth was measured by LDP (quantity of 1to 4-Hz activity in 30-s epochs), and records were scored visually for sleep staging and for IEDs. Multivariate logistic regression analyses were applied. Results: Adjusting for LDP, number of seizures before the current seizure, quartile of the night, and total number of IEDs that occurred during the night, IED did not increase or decrease before seizures (p > 0.1). The rate of IEDs increased directly with LDP (p 4 0.0001), as shown in prior work. Conclusions: IEDs are not activated or suppressed before seizures during sleep, suggesting that different pathophysiologic processes underlie these two phenomena. These results corroborate prior studies, while providing a more advanced analysis by adjusting for sleep depth and applying multivariate logistic regression analyses. Key Words: Epilepsy— Seizure—Interictal epileptiform discharges—Sleep—Statistics.
17 citations