Using text mining to glean insights from COVID-19 literature.

The purpose of this study is to develop a text clustering-based analysis of COVID-19 research articles. Owing to the proliferation of published COVID-19 research articles, researchers need a method for reducing the number of articles they have to search through to find material relevant to their expertise. The study analyzes 83,264 abstracts from research articles related to COVID-19. The textual data are analysed using singular value decomposition (SVD) and the expectation-maximisation (EM) algorithm. Results suggest that text clustering can both reveal hidden research themes in the published literature related to COVID-19, and reduce the number of articles that researchers need to search through to find material relevant to their field of interest.

Recovery and Youth: An Integrative Review.

Although rates of alcohol and other substance use disorders in adolescents have been estimated for decades, little is known about the prevalence, pathways, and predictors of remission and long-term recovery among adolescents. This article provides an integrative review of the literature on youth recovery. A final selection of 39 relevant articles was grouped into five sections: treatment outcomes, special emphasis populations, recovery-oriented systems of care, families, and non-abstinence-based approaches. The review recommends more adolescent research in three basic areas: more research about medication-assisted treatment and recovery as well as harm reduction approaches for adolescents; expansion of research on recovery practices for youth who do not receive treatment due to personal choice or societal disparities; and more life course research, which may begin with adolescent participants and extend across the life span. Additionally, the authors suggest the recovery capital model for adolescents and the neuroscience of addiction may provide additional precision and direction for the study of youth recovery.

Vaccination Perceptions of College Students: With and without Vaccination Waiver.

INTRODUCTION: The resurgence of vaccine preventable diseases occurs more often among intentionally unvaccinated individuals, placing at direct risk young adults not caught up on vaccinations. The objectives of this study were to characterize the sociodemographic characteristics of young adults with and without vaccination waivers and identify their perceived benefits, barriers, and influencers of vaccination. METHODS: Young adults (n = 964) from a Midwestern rural university responded to a survey (fall 2015-spring 2016) designed to identify their perception toward vaccination. Instrument consistency was measured using the Cronbach α-scores. The Chi-square test was used to test any sociodemographic differences and Mann-Whitney U-tests results for differences between exempt and non-exempt students. Analysis occurred in spring 2017. RESULTS: A little over one-third of young adults with a vaccination waiver were not up to date on their vaccinations, and think that vaccinations can cause autism. The biggest identifiable benefit was effective control against disease. The surveyed young adults ranked the out of pocket cost associated with vaccination as the most important barrier and safe and easy to use vaccines as the most important influencer of vaccination. CONCLUSION: Young adults who have had a vaccination waiver appear to not be up to date on their vaccinations. Vaccine administration programs, such as university campus clinics, would benefit from addressing perceptions unique to young adults with and without a vaccine waiver. This would subsequently better provide young adults a second shot for getting appropriately caught up on vaccinations.

Leadership for Public Health 3.0: A Preliminary Assessment of Competencies for Local Health Department Leaders.

BACKGROUND: The foundational public health services model V1.0, developed in response to the Institute of Medicine report For the Public's Health: Investing in a Healthier Future identified important capabilities for leading local health departments (LHDs). The recommended capabilities include the organizational competencies of leadership and governance, which are described as consensus building among internal and external stakeholders. Leadership through consensus building is the main characteristic of Democratic Leadership. This style of leadership works best within the context of a competent team. Not much is known about the competency structure of LHD leadership teams. The objectives of this study characterize the competency structure of leadership teams in LHDs and identify the relevance of existing competencies for the practice of leadership in public health. MATERIALS AND METHODS: The study used a cross-sectional study design. Utilizing the workforce taxonomy six management and leadership occupation titles were used as job categories. The competencies were selected from the leadership and management domain of public health competencies for the Tier -3, leadership level. Study participants were asked to rank on a Likert scale of 1-10 the relevance of each competency to their current job category, with a rank of 1 being least important and a rank of 10 being most important. The instrument was administered in person. DATA: Data were collected in 2016 from 50 public health professionals serving in leadership and management positions in a convenience sample of three LHDS. RESULTS: The competency of most relevance to the highest executive function category was that of "interaction with interrelated systems." For sub-agency level officers the competency of most relevance was "advocating for the role of public health." The competency of most relevance to Program Directors/Managers or Administrators was "ensuring continuous quality improvement." The variation between competencies by job category suggests there are distinct underlying relationships between the competencies by job category.

Efficacy and Safety of Proton Pump Inhibitors in the Long-Term Aspirin Users: A Meta-Analysis of Randomized Controlled Trials.

BACKGROUND: Long-term aspirin use in cardiovascular disease prevention may result in gastrointestinal bleeding. Although proton pump inhibitors (PPI) have been shown to reduce the risks of peptic ulcers and dyspeptic symptoms in long-term aspirin users in the randomized controlled trials, there are safety concerns about the long-term use of PPI. STUDY QUESTION: What is the safety and efficacy of PPI in patients using aspirin in long term for prevention of cardiovascular diseases and stroke? METHODS: We searched MEDLINE, EMBASE, CENTRAL, CINAHL, ProQuest, and relevant references from inception through February 2015, and used random-effects model for meta-analysis. RESULTS: A total of 10 publications from 9 studies (n = 6382) were included in the meta-analysis. Compared with control, PPI reduced the risks of peptic ulcers [risk ratio (RR): 0.19; 95% confidence interval: 0.13-0.26; P < 0.00001], gastric ulcers [0.24 (0.16-0.35); P < 0.00001], duodenal ulcers [0.12 (0.05-0.29); P < 0.00001], bleeding ulcers [0.22 (0.10-0.51); P = 0.0004], and erosive esophagitis [0.14 (0.07-0.28); P < 0.00001]. PPI increased the resolution of epigastric pain [1.13 (1.03-1.25); P = 0.01], heartburn [1.24 (1.18-1.31); P < 0.00001], and regurgitation [1.26 (1.13-1.40); P < 0.0001], but did not increase the risks of all-cause mortality [1.72 (0.61-4.87); P = 0.31], cardiovascular mortality [1.80 (0.59-5.44); P = 0.30], nonfatal myocardial infarction/ischemia [0.56 (0.22-1.41); P = 0.22], ischemic stroke/transient ischemic attack [1.09 (0.34-3.53); P = 0.89] and other adverse events. CONCLUSIONS: The PPI seems to be effective in preventing peptic ulcers and erosive esophagitis and in resolution of dyspeptic symptoms without increasing adverse events, cardiac risks or mortality in long-term aspirin users.

Comparison of the predictive qualities of three prognostic models of colorectal cancer.

Most discoveries of cancer biomarkers involve construction of a single model to determine predictions of survival.. 'Data-mining' techniques, such as artificial neural networks (ANNs), perform better than traditional methods, such as logistic regression. In this study, the quality of multiple predictive models built on a molecular data set for colorectal cancer (CRC) was evaluated. Predictive models (logistic regressions, ANNs, and decision trees) were compared, and the effect of techniques for variable selection on the predictive quality of these models was investigated. The Kolmogorov-Smirnoff (KS) statistic was used to compare the models. Overall, the logistic regression and ANN methods outperformed use of a decision tree. In some instances (e.g., for a model that included 'all variables without tumor stage' and use of a decision tree for variable selection), the ANN marginally outperformed logistic regression, although the difference between the accuracy of the KS statistic was minimal (0.80 versus 0.82). Regardless of the variable(s) and the methods for variable selection, all three predictive models identified survivors and non-survivors with the same level of statistical accuracy.

Using an empirical binomial hierarchical Bayesian model as an alternative to analyzing data from multisite studies.

This article explores the statistical methodologies used in demonstration and effectiveness studies when the treatments are applied across multiple settings. The importance of evaluating and how to evaluate these types of studies are discussed. As an alternative to standard methodology, the authors of this article offer an empirical binomial hierarchical Bayesian model as a way to effectively evaluate multisite studies. An application of using the Bayesian model in a real-world multisite study is given.

Comparison of predicted probabilities of proportional hazards regression and linear discriminant analysis methods using a colorectal cancer molecular biomarker database.

BACKGROUND: Although a majority of studies in cancer biomarker discovery claim to use proportional hazards regression (PHREG) to the study the ability of a biomarker to predict survival, few studies use the predicted probabilities obtained from the model to test the quality of the model. In this paper, we compared the quality of predictions by a PHREG model to that of a linear discriminant analysis (LDA) in both training and test set settings. METHODS: The PHREG and LDA models were built on a 491 colorectal cancer (CRC) patient dataset comprised of demographic and clinicopathologic variables, and phenotypic expression of p53 and Bcl-2. Two variable selection methods, stepwise discriminant analysis and the backward selection, were used to identify the final models. The endpoint of prediction in these models was five-year post-surgery survival. We also used linear regression model to examine the effect of bin size in the training set on the accuracy of prediction in the test set. RESULTS: The two variable selection techniques resulted in different models when stage was included in the list of variables available for selection. However, the proportion of survivors and non-survivors correctly identified was identical in both of these models. When stage was excluded from the variable list, the error rate for the LDA model was 42% as compared to an error rate of 34% for the PHREG model. CONCLUSIONS: This study suggests that a PHREG model can perform as well or better than a traditional classifier such as LDA to classify patients into prognostic classes. Also, this study suggests that in the absence of the tumor stage as a variable, Bcl-2 expression is a strong prognostic molecular marker of CRC.

Predicting 5-year survival of colorectal carcinoma patients using data mining methods.

We compared the accuracy of 3 data-mining models, neural-network, decision-tree, and logistic-regression, in predicting the 5-year survival of patients with colorectal cancer. The database consisted of patient demographics, pathologic features, and levels of expression of 2 biomarkers (p53 and Bcl-2). All 3 methods demonstrated acceptable accuracy, from 64% to 70%. The neural-network model had the best specificity (80%) and accuracy (70%) but lowest sensitivity (59%). Both logistic-regression and decision-models demonstrated comparable sensitivity (72%).