RESUMO
Objectives: The COVID-19 pandemic was disruptive for trainees and may have affected career decisions for some learners. This study examined the impact of the pandemic on emergency medicine (EM) resident perceptions of their mental health, perceptions of personal safety, and career choice regret. Methods: This was a cross-sectional survey study administered following the 2021 American Board of Emergency Medicine In-Training Examination (ITE). Survey measures included suicidal ideation (SI), COVID concerns in terms of infection prevention and control (IPC) training, COVID risk to self and/or COVID risk to family, and COVID-related career regret. COVID concerns were compared by gender and race/ethnicity using Pearson's chi-square tests. Multivariable logistic regression models were used to test the association between SI and COVID concerns, resident characteristics, and program characteristics. Results: A total of 6980 out of 8491 EM residents (82.2%) from 244 programs completed the survey. Only 1.1% of participants reported insufficient training in COVID IPC practices. Participants were concerned about COVID risk to themselves (40.3%) and to their families (63.3%) due to their job roles. These concerns were more common among women or nonbinary (vs. men); all other races/ethnicities (vs. non-Hispanic Whites); senior residents (vs. PGY-1, PGY-2 residents); and residents who were married or in relationships (vs. single or divorced). A total of 6.1% of participants reported that COVID made them reconsider choosing EM as their career. Career regret in this cohort was higher than that in the proportion (3.2%) expressing career regret in the 2020 ITE (p < 0.001). Career regret was more common among women or nonbinary (vs. men); all other races/ethnicities (vs. non-Hispanic Whites); and senior residents (vs. PGY-1, PGY-2 residents). The overall SI rate was 2.6%, which did not differ from that of the 2020 sample of EM residents (2.5%, p = 0.88). Conclusions: Many EM residents reported concerns about COVID risks to themselves and their families. Although the rate of SI remained unchanged, more EM residents reported career regret during the COVID pandemic.
RESUMO
BACKGROUND: This study aimed to use natural language processing to predict the presence of intra-abdominal injury using unstructured data from electronic medical records. METHODS: This was a random-sample retrospective observational cohort study leveraging unstructured data from injured patients taken to one of 9 acute care hospitals in an integrated health system between 2015 and 2021. Patients with International Classification of Diseases External Cause of Morbidity codes were identified. History and physical, consult, progress, and radiology report text from the first 8 hours of care were abstracted. Annotator dyads independently annotated encounters' text files to establish ground truth regarding whether intra-abdominal injury occurred. Features were extracted from text using natural language processing techniques, bag of words, and principal component analysis. We tested logistic regression, random forests, and gradient boosting machine to determine accuracy, recall, and precision of natural language processing to predict intra-abdominal injury. RESULTS: A random sample of 7,000 patient encounters of 177,127 was annotated. Only 2,951 had sufficient information to determine whether an intra-abdominal injury was present. Among those, 84 (2.9%) had an intra-abdominal injury. The concordance between annotators was 0.989. Logistic regression of features identified with bag of words and principal component analysis had the best predictive ability, with an area under the receiver operating characteristic curve of 0.9, recall of 0.73, and precision of 0.17. Text features with greatest importance included "abdomen," "pelvis," "spleen," and "hematoma." CONCLUSION: Natural language processing could be a screening decision support tool, which, if paired with human clinical assessment, can maximize precision of intra-abdominal injury identification.