Mural Arts Philadelphia

BackgroundOpioid overdoses in the USA have increased to unprecedented levels. Administration of the opioid antagonist naloxone can prevent overdoses.ObjectiveThis study was conducted to reveal the pharmacoepidemiologic patterns in naloxone prescribing to Medicaid patients from 2018 to 2021 as well as Medicare in 2019.DesignObservational pharmacoepidemiologic studySettingUS Medicare and Medicaid naloxone claimsInterventionThe Medicaid State Drug Utilisation Data File was utilised to extract information on the number of prescriptions and the amount prescribed of naloxone at a national and state level. The Medicare Provider Utilisation and Payment was also utilised to analyse prescription data from 2019.Outcome measuresStates with naloxone prescription rates that were outliers of quartile analysis were noted.ResultsThe number of generic naloxone prescriptions per 100 000 Medicaid enrollees decreased by 5.3%, whereas brand naloxone prescriptions increased by 245.1% from 2018 to 2021. There was a 33.1-fold difference in prescriptions between the highest (New Mexico=1809.5) and lowest (South Dakota=54.6) states in 2019. Medicare saw a 30.4-fold difference in prescriptions between the highest (New Mexico) and lowest states (also South Dakota) after correcting per 100 000 enrollees.ConclusionsThis pronounced increase in the number of naloxone prescriptions to Medicaid patients from 2018 to 2021 indicates a national response to this widespread public health emergency. Further research into the origins of the pronounced state-level disparities is warranted.

We study the problem of offline learning in automated decision systems under the contextual bandits model. We are given logged historical data consisting of contexts, (randomized) actions, and (nonnegative) rewards. A common goal is to evaluate what would happen if different actions were taken in the same contexts, so as to optimize the action policies accordingly. The typical approach to this problem, inverse probability weighted estimation (IPWE) [Bottou et al., 2013], requires logged action probabilities, which may be missing in practice due to engineering complications. Even when available, small action probabilities cause large uncertainty in IPWE, rendering the corresponding results insignificant. To solve both problems, we show how one can use policy improvement (PIL) objectives, regularized by policy imitation (IML). We motivate and analyze PIL as an extension to Clipped-IPWE, by showing that both are lower-bound surrogates to the vanilla IPWE. We also formally connect IML to IPWE variance estimation [Swaminathan and Joachims 2015] and natural policy gradients. Without probability logging, our PIL-IML interpretations justify and improve, by reward-weighting, the state-of-art cross-entropy (CE) loss that predicts the action items among all action candidates available in the same contexts. With probability logging, our main theoretical contribution connects IML-underfitting to the existence of either confounding variables or model misspecification. We show the value and accuracy of our insights by simulations based on Simpson's paradox, standard UCI multiclass-to-bandit conversions and on the Criteo counterfactual analysis challenge dataset.

Objectives: Normobaric oxygenation (NBO) has been shown to provide neuroprotection in vivo and in vitro . Yet, a recent Phase 2 clinical trial investigating NBO therapy in acute ischemic stroke was terminated due to questionable therapeutic benefit. NBO therapy alone may be insufficient to produce improved outcomes. In our recent study, we demonstrated a strong neuroprotective effect of ethanol at a dose of 1.5 g/kg (equivalent to the human legal driving limit). In this study, we sought to identify whether low-dose ethanol administration enhances the neuroprotection offered by NBO and whether combined administration of NBO with ethanol is associated with reduced apoptosis. Methods: Sprague-Dawley rats were subjected to right middle cerebral artery occlusion (MCAO) for 2 h, followed by reperfusion. Ischemic animals received either an intraperitoneal injection of 1.0 g/kg ethanol, 2 h of 100% NBO, or both ethanol and NBO. The Cell Death Detection ELISA Assay (Roche) was performed to determine apoptotic cell death at 24 h after reperfusion. Levels of pro-apoptotic (Caspase-3, Bcl-2-associated X-BAX, and Apoptosis-Inducing Factor-AIF) and anti-apoptotic proteins (Bcl-2 and Bcl-xL) were determined by Western blot analysis at 3 and 24 h after reperfusion. Results: As expected, untreated ischemic rats had the highest apoptotic cell death. Combined NBO/ethanol therapy decreased cell death by 48%, as compared to 29% with ethanol and 22% with NBO. Similarly, combined NBO/ethanol therapy promoted the greatest expression of anti-apoptotic factors and the lowest expression of pro-apoptotic proteins at 3 h after reperfusion. This effect was maintained at 24 h and even more pronounced for AIF and Caspase-3. Conclusions: Given singularly, NBO and ethanol improved the degree of cell death, decreased the expression of pro-apoptotic proteins, and increased the expression of anti-apoptotic proteins. Yet, when administered together, their effects largely compounded. These results suggest a synergistic neuroprotection offered by NBO with ethanol, which may be attributed at least in part to their shared role in modulating neuronal apoptosis.

The goal of this project was to increase the high-temperature strength of the H-Series of cast austenitic stainless steels by 50% and the upper use temperature by 86 to 140 degrees fahrenheit (30 to 60 degrees celsius). Meeting this goal is expected to result in energy savings of 35 trillion Btu/year by 2020 and energy cost savings of approximately $230 million/year. The higher-strength H-Series cast stainless steels (HK and HP type) have applications for the production of ethylene in the chemical industry, for radiant burner tubes and transfer rolls for secondary processing of steel in the steel industry, and for many applications in the heat treating industry, including radiant burner tubes. The project was led by Duraloy Technologies, Inc., with research participation by Oak Ridge National Laboratory (ORNL) and industrial participation by a diverse group of companies.