Bayesian estimation of a random effects heteroscedastic probit model

Summary Bayesian analysis is given of a random effects binary probit model that allows for heteroscedasticity. Real and simulated examples illustrate the approach and show that ignoring heteroscedasticity when it exists may lead to biased estimates and poor prediction. The computation is carried out by an efficient Markov chain Monte Carlo sampling scheme that generates the parameters in blocks. We use the Bayes factor, cross‐validation of the predictive density, the deviance information criterion and Receiver Operating Characteristic (ROC) curves for model comparison.     

Noninvasive tests for arterial structure, function, and compliance: do they identify risk or diagnose disease?

BACKGROUND: Brachial artery reactivity (BAR), carotid intima-media thickness (IMT), and applanation tonometry for evaluation of total arterial compliance may provide information about preclinical vascular disease. We sought to determine whether these tests could be used to identify patients with coronary artery disease (CAD) without being influenced by their ability to identify those at risk for CAD developing. METHODS: We studied 100 patients and compared 3 groups: 35 patients with known CAD; 34 patients with symptoms and risk factors but no CAD identified by stress echocardiography (risk group); and 31 control subjects. BAR and IMT were measured using standard methods, and total arterial compliance was calculated by the pulse-pressure method from simultaneous radial applanation tonometry and pulsed wave Doppler of the left ventricular outflow. Ischemia was identified as a new or worsening wall-motion abnormality induced by stress. RESULTS: In a comparison between the control subjects and patients either at risk for developing CAD or with CAD, the predictors of risk for CAD were: age (P =.01); smoking history (P =.002); hypercholesterolemia (P =.002); and hypertension (P =.004) (model R = 0.82; P =.0001). The independent predictors of CAD were: IMT (P =.001); BAR (P =.04); sex (P =.005); and hypertension (P =.005) (model R = 0.80; P =.0001). CONCLUSION: IMT, BAR, and traditional cardiovascular risk factors appear to identify patients at risk for CAD developing. However, only IMT was significantly different between patients at risk for developing CAD and those with overt CAD.     

Influence on binding of third-order torque to second-order angulation

Using an earlier model, which described the critical contact angle for binding from second-order angulation alone, a more generalized model is derived that combines the effects of angulation and torque. From this vantage point, the onset of binding is evaluated for 3 scenarios: second-order angulation alone, third-order torque only, and a combination of second-order angulation and third-order torque. These scenarios are detailed by plotting the critical contact angle for binding against the torque angle as a function of 10 wire dimensions (16 x 16, 16 x 22, 17 x 17, 17 x 22, 17 x 25, 18 x 18, 18 x 22, 18 x 25, 19 x 25, and 21 x 25 mil), 4 bracket widths (70, 100, 130, and 160 mil), and 4 bracket slots (18, 20.5, 22, and 24.5 mil). From these plots, we learn that each wire base dimension (eg, an 18-mil base as found in 18 x 18-mil, 18 x 22-mil and 18 x 25-mil archwires) has a common maximum critical contact angle for binding. Moreover, each wire-slot combination has a common maximum torque angle, which is independent of bracket width. Finally, we learn that archwire-bracket combinations that use a metric 0.5-mm slot might have some advantages with regard to torquing--given the current philosophy that light, continuous forces are more favorable.     

Adenosine and Sleep

Over the last several decades the idea that adenosine (Ado) plays a role in sleep control was postulated due in large part to pharmacological studies that showed the ability of Ado agonists to induce sleep and Ado antagonists to decrease sleep. A second wave of research involving in vitro cellular analytic approaches and subsequently, the use of neurochemical tools such as microdialysis, identified a population of cells within the brainstem and basal forebrain arousal centers, with activity that is both tightly coupled to thalamocortical activation and under tonic inhibitory control by Ado. Most recently, genetic tools have been used to show that Ado receptors regulate a key aspect of sleep, the slow wave activity expressed during slow wave sleep. This review will briefly introduce some of the phenomenology of sleep and then summarize the effect of Ado levels on sleep, the effect of sleep on Ado levels, and recent experiments using mutant mouse models to characterize the role for Ado in sleep control and end with a discussion of which Ado receptors are involved in such control. When taken together, these various experiments suggest that while Ado does play a role in sleep control, it is a specific role with specific functional implications and it is one of many neurotransmitters and neuromodulators affecting the complex behavior of sleep. Finally, since the majority of adenosine-related experiments in the sleep field have focused on SWS, this review will focus largely on SWS; however, the role of adenosine in REM sleep behavior will be addressed.     

Americans' Awareness, Knowledge, and Behaviors Regarding Fats: 2006-2007

In recent years, epidemiologic and clinical studies, public and regulatory policy activity, and media coverage have focused on issues related to trans fats. To help increase awareness and understanding of trans fats and other fats, the American Heart Association (AHA) launched the “Face the Fats” national consumer education campaign in April 2007. The AHA commissioned a quantitative tracking survey between 2006 and 2007 to measure changes in consumer awareness, knowledge, and behaviors related fats and oils and their perceived impact on heart disease. The survey was conducted by Cogent Research. Data were collected during March 2006 and May 2007. At both time points, the survey included a representative sample of the American population age 18 to 65 years (n=1,000). The sampling plan for the survey was designed based on the 2000 and 2003 US Census. The margin of error was ±3.10 percentage points. Awareness of trans fats increased during the 1-year study period. In 2007, 92% of respondents were aware of trans fats, an increase from 84% in 2006 (P<0.05). The 2007 level was similar to the awareness of saturated fats (93%). Perceptions that certain fats and oils heighten the risk of heart disease increased for trans fats (73% in 2007 vs 63% in 2006; P<0.05), saturated fats (77% in 2007 vs 73% in 2006; P<0.05), and partially hydrogenated oils (56% in 2007 vs 49% in 2006; P<0.05). Knowledge about food sources of different fats remained low. On an unaided basis, 21% could name three food sources of trans fats in 2007, up from 17% in 2006 (P<0.05). Knowledge of food sources of saturated fat remained unchanged at 30% in 2007. Significantly more respondents in 2007 reported behavioral changes related to trans fat information, such as buying food products because they show “zero trans fat” on labels or packages (37% in 2007 vs. 32% in 2006; P<0.05). Between 2006 and 2007, consumer awareness about trans fats increased and attained awareness levels similar to saturated fats. The increased awareness is associated with improved self-reported behaviors in grocery shopping. Nonetheless, overall knowledge, especially regarding food sources of saturated and trans fats, remains relatively low, underscoring the need for heightened consumer education activities. The positive change in consumer awareness about trans fats is likely attributable to the wide range of messages available to them, including the AHA “Face the Fats” national consumer education campaign.