Search

Clinicians iteratively adjust treatment approaches to improve outcomes, but to date, automatable approaches for continuous learning of risk factors as these adjustments are made are lacking. We combined a large-scale, comprehensive real-world Learning Health System infrastructure (LHSI), with automated statistical profiling, visualization, and artificial intelligence (AI) approach to test evidence-based discovery of clinical factors for three endpoints: dysphagia, xerostomia, and 3-year survival for head and neck cancer patients. Records for 964 patients treated for head and neck cancers with conventional fractionation between 2017 and 2022 were used. Combined information on demographics, diagnosis and staging, social determinants of health measures, chemotherapy, radiation therapy dose volume histogram curves, treatment details, laboratory values, and outcomes from the LHSI to winnow evidence for 485 candidate features. Univariate statistical profiling was performed using bootstrap resampling to detail confidence intervals for the following thresholds and metrics: area under the curve (AUC), sensitivity (SN), specificity (SP), F1, diagnostic odds ratio (DOR), P values for Wilcoxon Rank Sum (WRS), Kolmogorov-Smirnov (KS), and logistic fits of distributions detailed predictive evidence of individual features. Parsimonious XGBoost models were constructed with 10-fold cross validation using training (70%), validation (10%), and test (20%) sets. Probabilistic models utilizing statistical profiling logistic fits of distributions were used to benchmark XGBoost models. Incidence of dysphagia ≥ grade 3 within 1 year of treatment was low (11%). Xerostomia ≥ grade 2 (39% to 16%) and survival ≤ 3 years decreased (25% to 15%) over the time range. The strongest grade 2 xerostomia predictor was Glnd_Submand_Low: D15% [Gy] ≥ 45.2 with a logistic model quantifying a gradual rather than an abrupt increase in probability (13.5 + 0.18 (x-41.0 Gy)). Strongest predictive factors for lower likelihood of death by 3 years were GTV_High: Volume [cc] ≤ 21.1, GTV_Low: Volume [cc] ≤ 57.5, Baseline Neutrophil-Lymphocyte Ratio (NLR) ≤ 5.6, Monocyte-Lymphocyte Ratio (MLR) ≤0.56, Platelet-Lymphocyte ratio (PLR) ≤ 202.5. All predictors had WRS and KS P values < 0.02. Statistical profiling enabled detailing gains of XGBoost models with respect to individual features. Time period reductions in distribution of GTV volumes correlated with reductions in death by 3 years. Combined use of LHSI, Statistical Profiling and Artificial Intelligence provided a basis for automating evidence-based discovery. Benchmarking AI models with simple probabilistic models provided a means of understanding when results are driven by general areas of overall risk vs. more complex interactions. The method can form a new approach to continuous learning and evidence-based development of clinical trial testable hypothesis and stratifications. [ABSTRACT FROM AUTHOR]

Higher education faculty members in educational leadership are members of an intellectual community that stress scholarly activity, yet they serve preservice and in-service school leaders who value applications of knowledge to problems of practice and policy. The professional periodicals that professors rank highest in terms of quality and greatest usage reflect this balancing of interests. This recent exploratory study addressed four questions. First, which professional periodicals do educational leadership professors in member institutions of the University Council for Educational Administration (UCEA) rank as the "best"? Second, which criteria do these educational leadership professors use as the basis for these rankings? Third, which professional periodicals do these educational leadership professors read most often? Finally, which periodicals outside of education do they read regularly? (Contains 4 tables.)

This study examined superintendents' perceptions of current performance evaluation practices and their preferences for those practices. Specifically, 1,125 superintendents across the nation received a questionnaire exploring evaluation effectiveness and results-based procedures. Findings indicate practices are neither uniformly effective nor results-based by the standards set for this study. Though more than 90 percent of superintendents were evaluated at least annually, some responses raised serious doubts about the effectiveness of the practices used. One limitation of effective evaluations was the lack of training and knowledge among board members. Also, superintendents preferred results-based evaluations, yet personality traits and process skills were emphasized over results-based criteria. Because findings show little evidence that evaluation processes have changed to accommodate the accountability movement, the authors contend superintendents must educate their boards in establishing effective and results-based practices. (Contains 15 tables.)

Examines the rankings of professional journals by superintendents hired from inside the school district compared with those hired from outside. Finds little or no differences between two groups in the rankings of four top journals: "Educational Leadership", "Phi Delta Kappan", "The School Administrator," and "The American School Board Journal." (Contains 32 references.)(PKP)

Discusses three sequential questions boards should reach consensus on before beginning a superintendent search. Questions involve district's position related to education issues deemed important to the community, a new superintendent's personality traits and leadership skills required to address these issues, and whether the board should seek to fill the position from inside or outside the district. (PKP)

Reports results of national survey of superintendents' top reading choices from a list of 34 periodicals in educational leadership. Superintendents' top-ranked periodicals are "Educational Leadership,""Phi Delta Kappan,""School Administrator," and "American School Board Journal." (Contains 24 references.) (PKP)