Using publicly available data to predict recreational cannabis legalization at the county-level: A machine learning approach

BACKGROUND

There is substantial geographic variability in local cannabis policies within states that have legalized recreational cannabis. This study develops an interpretable machine learning model that uses county-level population demographics, sociopolitical factors, and estimates of substance use and mental illness prevalences to predict the legality of recreational cannabis sales within each U.S. county.

METHODS

We merged data and selected 14 model inputs from the 2010 Census, 2012 County Presidential Data from the MIT Elections Lab, and Small Area Estimates from the National Surveys on Drug Use and Health (NSDUH) from 2010 to 2012 at the county level. County policies were labeled as having recreational cannabis legal (RCL) if the sale of recreational cannabis was allowed anywhere in the county in 2014, resulting in 92 RCL and 3002 non-RCL counties. We used synthetic data augmentation and minority oversampling techniques to build an ensemble of 1000 logistic regressions on random sub-samples of the data, withholding one state at a time and building models from all remaining states. Performance was evaluated by comparing the predicted policy conditions with the actual outcomes in 2014.

RESULTS

When compared to the actual RCL policies in 2014, the ensemble estimated predictions of counties transitioning to RCL had a macro f1 average score of 0.61. The main factors associated with legalizing county-level recreational cannabis sales were the prevalences of past-month cannabis use and past-year cocaine use.

CONCLUSION

By leveraging publicly available data from 2010 to 2012, our model was able to achieve appreciable discrimination in predicting counties with legal recreational cannabis sales in 2014, however, there is room for improvement. Having demonstrated model performance in the first handful of states to legalize cannabis, additional testing with more recent data using time to event models is warranted.

Lower Emotional Exhaustion among Employees Is Associated with Intentional Incorporation of Animals into Residential Care Settings

Secondary effects of animal-integrated programming on residential care center (RCC) staff and organizational culture are not well understood. We explored emotional exhaustion among RCC employees both in facilities that incorporated animals and those that did not incorporate animals into the therapeutic environment. We conducted a survey throughout a large midwestern RCC system in the United States to determine relationships between organizational culture, emotional exhaustion, and the intentionality by which animals were incorporated into programming. Data were analyzed by examining associations between variables of interest using chi-square or t-tests, and linear mixed-effects modeling was used to identify potential confounding effects due to differences in children served within RCCs. Staff from RCCs that used animals intentionally reported lower emotional exhaustion (p = 0.006), and higher average workplace safety (p = 0.024) and psychological safety (p < 0.001). Integrating animals into RCC programming is associated with elements of a strong organizational culture. It is possible that animal-integrated programming has a positive impact on the facility culture and workforce, and/or that RCCs with strong pre-existing cultures are more likely to use animal-integrated programming.

Hypercholesterolemia Accelerates Both the Initiation and Progression of Angiotensin II-induced Abdominal Aortic Aneurysms

OBJECTIVE

This study determined whether hypercholesterolemia would contribute to both the initiation and progression of angiotensin (Ang)II-induced abdominal aortic aneurysms (AAAs) in mice.

METHODS AND RESULTS

To determine whether hypercholesterolemia accelerates the initiation of AAAs, male low-density lipoprotein (LDL) receptor -/- mice were either fed one week of Western diet prior to starting AngII infusion or initiated Western diet one week after starting AngII infusion. During the first week of AngII infusion, mice fed normal diet had less luminal expansion of the suprarenal aorta compared to those initiated Western diet after the first week of AngII infusion. The two groups achieved comparable luminal dilation on week 2 through week 6 of AngII infusion as monitored by ultrasound. To determine whether hypercholesterolemia contributed to the progression of established AAAs, male LDL receptor -/- mice were fed Western diet and infused with AngII for 4 weeks. Mice with established AAAs were then stratified into two groups based on luminal diameters measured by ultrasound. While AngII infusion was continued for another 8 weeks in both groups, mice in one group were continuously fed Western diet, but diet in the other group was switched to normal laboratory diet. In the latter group, plasma cholesterol concentrations were reduced rapidly to approximately 500 mg/dl within one week after the diet was switched from Western diet to normal laboratory diet. Luminal expansion progressed constantly in mice continuously fed Western diet, whereas no continuous expansion was detected in mice that were switched to normal laboratory diet.

CONCLUSION

Hypercholesterolemia accelerates both the initiation of AAAs and progression of established AAAs in AngII-infused male LDL receptor -/- mice.

CLINICAL RELEVANCE

Hypercholesterolemia is modestly associated with AAAs in observational or retrospective clinical studies. It is not feasible to study whether hypercholesterolemia contributes to the initiation of AAAs or progression of established AAAs in human. This study using AngII-induced AAA mouse model provides solid evidence that hypercholesterolemia contributes to both the initiation and progression of AAAs, supporting that statin therapy at any stage of AAA development may be beneficial to hypercholesterolemic patients with AAAs.

Immune System Activation in Perioperative Thrombectomy Patients: Preliminary Retrospective Study

BACKGROUND

Ischemic stroke is a devastating condition resulting in significant morbidity. Mechanical thrombectomy is now standard for large vessel occlusive stroke. Neuroinflammation is known to play important roles in ischemic stroke. Our aims were to examine our thrombectomy procedures and preliminarily examine systemic immune response in relation to thrombectomy changes.

METHODS

A retrospective review was conducted on mechanical thrombectomy cases from July 2011 through December 2015. Primary outcomes were time to recanalization, final Thrombolysis in Cerebral Infarction score, procedural complications, National Institutes of Health Stroke Scale improvement, and changes in white blood cell (WBC) count.

RESULTS

One-hundred and twenty-nine procedures were performed. We found a significant difference between WBC count on admission and WBC count post thrombectomy day 1 for patients with >90 minutes to recanalization (P = 0.006107). There was a positive association between WBC change and absolute National Institutes of Health Stroke Scale change among females (P = 0.0273) but not among males.

CONCLUSIONS

Overall, we found that systemic immune response has close relationships with speed of recanalization and preliminarily may shift differently on the basis of sex.

Early signatures of bleeding and mortality in patients on left ventricular assist device support: novel methods for personalized risk-stratification

Background: Left ventricular assist devices (LVADs) provide support for patients with end-stage heart failure. The aims of this study were to determine whether baseline analysis and early trends in routine laboratory data, platelet activity, and thromboinflammatory biomarkers following LVAD implantation reveal trends that predict personalized risks of one-year gastrointestinal (GI) bleeding, stroke, pump thrombosis, drive-line infections and mortality in patients on LVAD support. Methods: We performed an observational study at the University of Kentucky with 61 participants who underwent first-time LVAD implantation. Blood was collected at baseline and post-op days 0, 1, 3 and 6 as well as clinical follow-up. Demographics, clinical characteristics, one-year adverse events and routine laboratory data were collected from electronic medical records. Platelet function and plasma biomarkers were profiled. Results: Evaluation of routine laboratory results revealed that sustained thrombocytopenia and increased mean platelet volume (MPV) were associated with development of GI bleeding and mortality. Platelet function at follow-up visit predicted one-year bleeding events. Thrombotic biomarker sCD40L strongly predicted one-year GI bleeding at baseline before implantation and within the first week following LVAD implant. Conclusions: Early trends in routine bloodwork and platelet function may serve as novel signatures of patients at risk to experience adverse events.

Dioxin-like PCB 126 increases intestinal inflammation and disrupts gut microbiota and metabolic homeostasis

The gut microbiome is sensitive to diet and environmental exposures and is involved in the regulation of host metabolism. Additionally, gut inflammation is an independent risk factor for the development of metabolic diseases, specifically atherosclerosis and diabetes. Exposures to dioxin-like pollutants occur primarily via ingestion of contaminated foods and are linked to increased risk of developing cardiometabolic diseases. We aimed to elucidate the detrimental impacts of dioxin-like pollutant exposure on gut microbiota and host gut health and metabolism in a mouse model of cardiometabolic disease. We utilized 16S rRNA sequencing, metabolomics, and regression modeling to examine the impact of PCB 126 on the microbiome and host metabolism and gut health. 16S rRNA sequencing showed that gut microbiota populations shifted at the phylum and genus levels in ways that mimic observations seen in chronic inflammatory diseases. PCB 126 reduced cecum alpha diversity (0.60 fold change; p = 0.001) and significantly increased the Firmicutes to Bacteroidetes ratio (1.63 fold change; p = 0.044). Toxicant exposed mice exhibited quantifiable concentrations of PCB 126 in the colon, upregulation of Cyp1a1 gene expression, and increased markers of intestinal inflammation. Also, a significant correlation between circulating Glucagon-like peptide-1 (GLP-1) and Bifidobacterium was evident and dependent on toxicant exposure. PCB 126 exposure disrupted the gut microbiota and host metabolism and increased intestinal and systemic inflammation. These data imply that the deleterious effects of dioxin-like pollutants may be initiated in the gut, and the modulation of gut microbiota may be a sensitive marker of pollutant exposures.

A weighted U-statistic for genetic association analyses of sequencing data

With advancements in next-generation sequencing technology, a massive amount of sequencing data is generated, which offers a great opportunity to comprehensively investigate the role of rare variants in the genetic etiology of complex diseases. Nevertheless, the high-dimensional sequencing data poses a great challenge for statistical analysis. The association analyses based on traditional statistical methods suffer substantial power loss because of the low frequency of genetic variants and the extremely high dimensionality of the data. We developed a Weighted U Sequencing test, referred to as WU-SEQ, for the high-dimensional association analysis of sequencing data. Based on a nonparametric U-statistic, WU-SEQ makes no assumption of the underlying disease model and phenotype distribution, and can be applied to a variety of phenotypes. Through simulation studies and an empirical study, we showed that WU-SEQ outperformed a commonly used sequence kernel association test (SKAT) method when the underlying assumptions were violated (e.g., the phenotype followed a heavy-tailed distribution). Even when the assumptions were satisfied, WU-SEQ still attained comparable performance to SKAT. Finally, we applied WU-SEQ to sequencing data from the Dallas Heart Study (DHS), and detected an association between ANGPTL 4 and very low density lipoprotein cholesterol.