Artificial intelligence to identify harmful alcohol use after early liver transplant for alcohol-associated hepatitis.

Early liver transplantation (LT) for alcohol-associated hepatitis (AH) is the fastest growing indication for LT, but prediction of harmful alcohol use post-LT remains limited. Among 10 ACCELERATE-AH centers, we examined psychosocial evaluations from consecutive LT recipients for AH from 2006 to 2017. A multidisciplinary panel used content analysis to develop a maximal list of psychosocial variables. We developed an artificial intelligence model to predict post-LT harmful alcohol use. The cohort included training (N = 91 among 8 centers) and external validation (N = 25 among 2 centers) sets, with median follow-up of 4.4 (IQR 3.0-6.0) years post-LT. In the training set, AUC was 0.930 (95%CI 0.862-0.998) with positive predictive value of 0.891 (95%CI 0.620-1.000), internally validated through fivefold cross-validation. In the external validation set, AUC was 0.692 (95%CI 0.666-0.718) with positive predictive value of 0.82 (95%CI 0.625-1.000). The model identified specific variables related to social support and substance use as highly important to predict post-LT harmful alcohol use. We retrospectively developed and validated a model that identified psychosocial profiles at LT predicting harmful alcohol use post-LT for AH. This preliminary model may inform selection and post-LT management for AH and warrants prospective evaluation in larger studies among all alcohol-associated liver disease being considered for early LT.

Use of standardized letters of recommendation for otolaryngology head and neck surgery residency and the impact of gender.

OBJECTIVES: To evaluate gender-based differences in narrative letters of recommendation (NLORs) and standardized letters of recommendation (SLORs) submitted for applicants to an otolaryngology head and neck surgery (OHNS) residency program. STUDY DESIGN: Retrospective review. METHODS: Nine hundred fifty-eight letters of recommendation (LORs) submitted to one OHNS program in 2013 to 2014 were analyzed. RESULTS: NLORs (n = 590) and SLORs (n = 368) were reviewed. Male writers composed over 85% of all LORs. Compared with male writers, female writers of LORs were more likely to compose a letter of minimal assurance (P < 0.025). Female writers of SLORs were more likely to rank applicants higher in communication skills (P < 0.035) and match potential (P < 0.045). Analysis of LORs by applicant gender revealed that female applicants were more likely than male applicants to be described as team players (P < 0.025) and less likely to receive a letter of minimal assurance (P < 0.001). Compared with SLORs, NLORs written for male applicants were more likely to reference their leadership potential (P < 0.001). Female applicants were less likely to be described as "bright" (P < 0.001) and more likely to have their appearance mentioned (P < 0.03) in NLORs when compared to SLORs. CONCLUSION: Although we found that SLORs reduce gender biases compared to NLORs, there remain significant differences in how female and male applicants are described. Greater awareness of historical biases likely has contributed to more equitable letter writing, although impactful gender biases remain when reviewing applicants. LEVEL OF EVIDENCE: NA. Laryngoscope, 127:2738-2745, 2017.