Background/Objective: Traumatic intracranial hemorrhage (tICH) accounts for significant trauma morbidity and mortality. Several studies have developed prognostic models for tICH outcomes, but previous models face limitations, including poor generalizability and limited accuracy. The objective was to develop a prognostic model and determine predictors of mortality using the largest trauma database in the U.S., applying rigorous analytical methodology with true hold-out-set model validation. Methods: We identified 248,536 patients in the National Trauma Data Bank (NTDB) from 2012 to 2016 with a diagnosis code associated with tICH. For each admission, we collected demographic information, systolic blood pressure, blood alcohol level (BAL), Glasgow Coma Score (GCS), Injury Severity Score (ISS), presence of epidural/subdural/subarachnoid/intraparenchymal hemorrhage, comorbidities, complications, trauma center level, and trauma center region. Our final study population was 212,666 patients following exclusion of records with missing data. The dependent variable was patient death. Linear support vector machine (SVM) classification was carried out with recursive feature selection. Model performance was assessed using holdout 10-fold cross-validation. Results: Cross-validation demonstrated a mean accuracy of 0.792 (95% CI 0.783–0.799). Accuracy, precision, recall, and AUC were 0.827, 0.309, 0.750, and 0.791, respectively. In the final model, high ISS, advanced age, subdural hemorrhage, and subarachnoid hemorrhage were associated with increased mortality, while high GCS verbal and motor subscores, current smoker, BAL beyond the legal limit, and level 1 trauma center were associated with decreased mortality. Conclusions: A linear SVM model was developed for tICH, with nine features selected as predictors of mortality. These findings are applicable to multiple hemorrhage subtypes and may benefit the triage of high risk patients upon admission. While many studies have attempted to create models to predict mortality in TBI, we sought to confirm those predictors using modern modeling approaches, machine learning, and true hold-out test sets, using the largest available TBI database in the U.S. We find that while the predictors we identify are consistent with prior reports, overall prediction accuracy is somewhat lower than prior reports when assessed more rigorously.

2.

Background Morbidity and mortality from prostate cancer (PCa) are known to vary heavily based on socioeconomic and demographic risk factors. We sought to describe prescreening PSA (prostate‐specific antigen) counseling (PPC) rates amongst male‐to‐female transgender (MtF‐TG) patients and non‐TG patients using the behavioral risk factor surveillance system (BRFSS). Methods We used the survey data from 2014, 2016, and 2018 BRFSS and included respondents aged 40–79 years who completed the “PCa screening” and “sexual orientation and gender identity” modules. We analyzed differences in age, education level, income level, marital status, and race/ethnicity using Pearson's χ2 tests. The association of PPC with MtF‐TG status and other patient characteristics was evaluated using multivariate logistic regression. Results A total of 175,383 respondents were included, of which 0.3% identified as MtF‐TG. Overall, 62.4% of respondents reported undergoing PPC. On univariate analysis, PPC rates were lower among MtF‐TG respondents when compared to the non‐TG group (58.3% vs. 62.4%, p = 0.03). MtF‐TG respondents were also more likely to report lower education level (p < 0.01), lower‐income level (p < 0.01), and were less likely to be white (p < 0.01) than non‐TG respondents. However, multivariate analysis adjusting for these respondent features demonstrated an association between higher income and higher education levels with increased odds of PPC, but no association was demonstrated between MtF‐TG status and PPC rates. PPC rates for the MtF‐TG and non‐TG populations did not change significantly over time. Conclusions Although PPC was less frequently reported among MtF‐TG respondents than in the non‐TG group on univariate analysis, this association was not demonstrated when controlling for confounders, including education and income levels. Instead, on multivariate analysis, low education and income levels were more predictive of PPC rates. Further research is needed to ensure equivalent access to prescreening counseling for patients across the socioeconomic and gender identity spectrum.