Back to the basics: Rethinking partial correlation network methodology.

The Gaussian graphical model (GGM) is an increasingly popular technique used in psychology to characterize relationships among observed variables. These relationships are represented as elements in the precision matrix. Standardizing the precision matrix and reversing the sign yields corresponding partial correlations that imply pairwise dependencies in which the effects of all other variables have been controlled for. The graphical lasso (glasso) has emerged as the default estimation method, which uses ℓ1 -based regularization. The glasso was developed and optimized for high-dimensional settings where the number of variables (p) exceeds the number of observations (n), which is uncommon in psychological applications. Here we propose to go 'back to the basics', wherein the precision matrix is first estimated with non-regularized maximum likelihood and then Fisher Z transformed confidence intervals are used to determine non-zero relationships. We first show the exact correspondence between the confidence level and specificity, which is due to 1 minus specificity denoting the false positive rate (i.e., α). With simulations in low-dimensional settings (p â‰ª n), we then demonstrate superior performance compared to the glasso for detecting the non-zero effects. Further, our results indicate that the glasso is inconsistent for the purpose of model selection and does not control the false discovery rate, whereas the proposed method converges on the true model and directly controls error rates. We end by discussing implications for estimating GGMs in psychology.

Impact of CMS coverage decision on access to transcatheter aortic valve replacement.

OBJECTIVES: To assess the impact of the Centers for Medicare and Medicaid Services (CMS) national coverage determination (NCD) on access for patients with aortic stenosis (AS) with transcatheter aortic valve replacement (TAVR) in a tertiary care center. BACKGROUND: TAVR has given hope to patients with AS who are deemed inoperable. The effects of the NCD on access to patients with AS has not been evaluated. MATERIALS AND METHODS: A total of 94 inoperable AS patients were evaluated and treated from December 2011 through June of 2012 with TAVR. Patients who underwent transfemoral (TF) vs. non-TF access were compared. The CMS NCD was released on May 1, 2012 and on July 1, 2012, the nontransfemoral access program was put on hold due to lack of reimbursement. RESULTS: Patients in the TF (n = 33) and non-TF access (n = 61) groups were similar in age (85.2 ± 6.3 vs. 84.8 ± 6.6 P = 0.74) and STS mortality (9.38 ± 5.33 vs. 7.91 ± 3.69, P = 0.074). The iliofemoral arteries were larger diameter in the TF group (7.72 ± 1.49 vs. 6.21 ± 1.78, P < 0.001) and males (7.39 ± 1.81 vs. 6.1 ± 1.61 P < 0.001). More women underwent valve implantation via non-TF access (73 vs. 23%, P = 0.03). After the NCD, 21 patients who previously qualified for non-TF TAVR would not be reimbursed by CMS. Four died soon after. CONCLUSIONS: After the NCD, the proportion of inoperable patients with severe AS that can be treated with TAVR was greatly reduced due the lack of reimbursement for TAVR via non-TF access. This effect is particularly pronounced in women. © 2014 Wiley Periodicals, Inc.

Cerebral blood flow at one year after controlled cortical impact in rats: assessment by magnetic resonance imaging.

Progressive tissue loss and delayed cognitive deficits are seen in rats during the initial year after experimental traumatic brain injury (TBI). As much as 10% of parenchymal volume is lost even in the contralateral hemisphere by 1 year after controlled cortical impact (CCI) in rats. Progressive declines in cerebral blood flow (CBF) are also associated with advanced age and neurodegenerative diseases. Surprisingly, the long-term effects of TBI on CBF remain undefined. CBF was quantified by continuous arterial spin-labeled magnetic resonance imaging (MRI) and measurements of spin-lattice relaxation time in a slice through the plane of injury at 1 year after experimental TBI produced by CCI (n = 4) or sham surgery (n = 4) in rats. CBF was quantified in six regions of interest (ROIs) that were anatomically identified on the control images in each hemisphere and included a medial cortical segment (contusion-enriched, beneath the impact site, on the ipsilateral side) cortex, hippocampus, thalamus, amygdala/pyriform cortex, and hemisphere. At 1 year after injury, CBF was dramatically (96%) reduced in structures within the large cystic lesion that was seen in three of four rats and variably included cortex and hippocampus. Overall, there was an 80% reduction in CBF in the ipsilateral medial cortical segment comparing CCI and sham groups. Similarly, 52% and 67% reductions were seen in CBF in the cortical and hippocampal ROIs ipsilateral to impact (CCI vs. sham), respectively. These are regions both with marked CBF disturbances early after injury and that ultimately suffer considerable tissue loss over the 1-year interval. However, at 1 year after CCI, CBF was not different from sham in other ROIs, including ipsilateral thalamus, or either contralateral hippocampus or hemisphere. We conclude that, at 1 year after CCI, CBF is reduced in anatomic structures at or near the impact site, including injured cortex and hippocampus, and this translates into a reduction in hemispheric CBF. However, despite both significant occult tissue loss ipsilateral and contralateral to the injury and delayed cognitive deficits, widespread reductions in CBF are not observed. This suggests the possibility of remodeling or repackaging of the brain that preserves CBF outside of the cystic lesion.