A Reproducible Model Based on Clinical Text for Predicting Suicidal Behavior.

Suicide risk models are critical for prioritizing patients for intervention. We demonstrate a reproducible approach for training text classifiers to identify patients at risk. The models were effective in phenotyping suicidal behavior (F1=.94) and moderately effective in predicting future events (F1=.63).

Effects of preoperative pulmonary function on perioperative outcomes after robotic-assisted pulmonary lobectomy.

INTRODUCTION: Effects of pulmonary function test (PFT) results on perioperative outcomes were investigated after robotic-assisted video-thoracoscopic (RAVT) pulmonary lobectomy. METHODS: We retrospectively analyzed 706 consecutive patients who underwent RAVT lobectomy by one surgeon over 10.8 years. Preoperative (preop) forced expiratory volume in 1 s as a percent of predicted (FEV1%) was used to group patients as having normal FEV1% (≥80%) versus reduced FEV1% (<80%). Demographics, preop comorbidities, intraoperative (intraop) and postoperative (postop) complications, perioperative outcomes, and median survival time (MST) were compared across patients with normal vs. reduced FEV1% using Chi-Square (X2), Fisher's Exact test, Student's t-test, Kruskal-Wallis test, or Kaplan-Meier analysis respectively, with significance at p ≤ 0.05. Multivariable analysis was performed for perioperative outcomes to investigate the differences across patients in the FEV1% groups. RESULTS: There were 470 patients with normal FEV1% and 236 patients with reduced FEV1%. The two FEV1% groups did not differ in intraop or postop complication rates, except for higher postop other arrhythmia requiring intervention (p = 0.004), prolonged air leak >5 days (p = 0.002), mucous plug formation (p = 0.009), hypoxia (p < 0.001), and pneumonia (p = 0.002), and total postop complications (p < 0.001) in reduced-FEV1% patients. Reduced FEV1% correlated with increased intraop estimated blood loss (p < 0.0001) and skin-to-skin operative time (p < 0.0001). Median overall survival in patients with normal FEV1% was 93.20 months (95% CI: 76.5-126.0) versus 58.9 months (95% CI: 50.4-68.4) in patients with reduced FEV1% (p = 0.0004). CONCLUSION: Patients should have PFTs conducted before surgery to determine at-risk patients. However, RAVT pulmonary lobectomy is feasible and safe even in patients with reduced FEV1%.