The use of the Gibbs sampler with fully conditionally specified models, where the distribution of each variable given the other variables is the starting point, has become a popular method to create imputations in incomplete multivariate data. The theoretical weakness of this approach is that the specified conditional densities can be incompatible, and therefore the stationary distribution to which the Gibbs sampler attempts to converge may not exist. This study investigates practical consequences of this problem by means of simulation. Missing data are created under four different missing data mechanisms. Attention is given to the statistical behavior under compatible and incompatible models. The results indicate that multiple imputation produces essentially unbiased estimates with appropriate coverage in the simple cases investigated, even for the incompatible models. Of particular interest is that these results were produced using only five Gibbs iterations starting from a simple draw from observed marginal distributions. It thus appears that, despite the theoretical weaknesses, the actual performance of conditional model specification for multivariate imputation can be quite good, and therefore deserves further study.

To provide a global analysis of gene expression in the aging heart, we monitored the expression of 9,977 genes simultaneously in 5-and 30-month-old male B6C3F1 mice by using high-density oligonucleotide microarrays and several statistical techniques. Aging was associated with transcriptional alterations consistent with a metabolic shift from fatty acid to carbohydrate metabolism, increased expression of extracellular matrix genes, and reduced protein synthesis. Caloric restriction (CR) started at 14 months of age resulted in a 19% global inhibition of age-related changes in gene expression. Interestingly, CR also resulted in alterations in gene expression consistent with preserved fatty acid metabolism, reduced endogenous DNA damage, decreased innate immune activity, apoptosis modulation, and a marked cytoskeletal reorganization. These observations provide evidence that aging of the heart is associated with specific transcriptional alterations, and that CR initiated in middle age may retard heart aging by inducing a profound transcriptional reprogramming.W hen started either early in life or at middle age, caloric restriction (CR) increases average and maximum lifespan, and reduces the incidence and delays the onset of spontaneous cancers and several other age-related diseases (1). Additionally, CR reduces the age-associated increase in reactive oxygen species (ROS)-induced molecular damage (2-5). Previously, we have used high-density oligonucleotide arrays to define aging and CR-related transcriptional alterations in mouse skeletal muscle (6) and brain (7). These reports provided evidence for an age-associated stress response characterized by the induction of heat-shock factors and other oxidative stress-induced transcripts. CR prevented these age-related alterations completely or partially. Both studies provided further support for the concept that oxidative stress may be an important, and perhaps underlying cause of the aging process of postmitotic tissues.The cardiac myocyte is the most energy demanding cell in the body, contracting constantly, 3 billion times or more in the average human lifespan (8), requiring large supplies of high-energy phosphates (9). Age-related changes in human and rodent hearts include a reduction in the number of myocytes (10, 11), myocyte hypertrophy (11, 12), cardiac fibrosis (13), lipofuscin pigment accumulation (14), a reduction in calcium transport across sarcoplasmic reticulum membrane (15), and alterations in the response to ␤-adrenergic stimulation (16). Collectively, these alterations likely contribute to age-related heart diseases being the leading cause of mortality in the U.S. (17). CR reduces the severity of spontaneous cardiomyopathy in male Sprague-Dawley rats (18) and prevents age-associated alterations in late diastolic function in B6D2F 1 mice (19). At the molecular level, CR reduces the concentration of both 8-hydroxydeoxyguanosine in DNA (20) and dityrosine cross-linking of proteins (21) in the heart of aging mice, and prevents somatic mitochondrial genomic rear...

No abstract

In healthcare cost-effectiveness analysis, probability distributions are typically skewed and missing data are frequent. Bootstrap and multiple imputation are well-established resampling methods for handling skewed and missing data. However, it is not clear how these techniques should be combined. This paper addresses combining multiple imputation and bootstrap to obtain confidence intervals of the mean difference in outcome for two independent treatment groups. We assessed statistical validity and efficiency of 10 candidate methods and applied these methods to a clinical data set. Single imputation nested in the bootstrap percentile method (with added noise to reflect the uncertainty of the imputation) emerged as the method with the best statistical properties. However, this method can require extensive computation times and the lack of standard software makes this method not accessible for a larger group of researchers. Using a standard unpaired t-test with standard multiple imputation without bootstrap appears to be a robust alternative with acceptable statistical performance for which standard multiple imputation software is available.