Thus far, no algorithms have been developed to automatically extract patients who meet Asthma Predictive Index (API) criteria from the Electronic health records (EHR) yet. Our objective is to develop and validate a natural language processing (NLP) algorithm to identify patients that meet API criteria. This is a cross-sectional study nested in a birth cohort study in Olmsted County, MN. Asthma status ascertained by manual chart review based on API criteria served as gold standard. NLP-API was developed on a training cohort (n = 87) and validated on a test cohort (n = 427). Criterion validity was measured by sensitivity, specificity, positive predictive value and negative predictive value of the NLP algorithm against manual chart review for asthma status. Construct validity was determined by associations of asthma status defined by NLP-API with known risk factors for asthma. Among the eligible 427 subjects of the test cohort, 48% were males and 74% were White. Median age was 5.3 years (interquartile range 3.6–6.8). 35 (8%) had a history of asthma by NLP-API vs. 36 (8%) by abstractor with 31 by both approaches. NLP-API predicted asthma status with sensitivity 86%, specificity 98%, positive predictive value 88%, negative predictive value 98%. Asthma status by both NLP and manual chart review were significantly associated with the known asthma risk factors, such as history of allergic rhinitis, eczema, family history of asthma, and maternal history of smoking during pregnancy (p value < 0.05). Maternal smoking [odds ratio: 4.4, 95% confidence interval 1.8–10.7] was associated with asthma status determined by NLP-API and abstractor, and the effect sizes were similar between the reviews with 4.4 vs 4.2 respectively. NLP-API was able to ascertain asthma status in children mining from EHR and has a potential to enhance asthma care and research through population management and large-scale studies when identifying children who meet API criteria.

Background: Subclinical hypothyroidism (SCH) has been associated with increased risk of adverse pregnancy outcomes in some, but not all, studies. Uncertainty remains regarding the impact of levothyroxine (LT4) therapy on improving health outcomes in pregnant women with SCH. The objective of this study was to assess the potential benefits of LT4 therapy in pregnant women with SCH. Methods: The medical records were reviewed of pregnant women with SCH, defined as an elevated serum thyrotropin (TSH) of >2.5 mIU/L for the 1st trimester or >3 mIU/L for the 2nd and 3rd trimesters, but ≤10 mIU/L. Pregnant women were divided into two groups depending on whether they received LT4 (group A) or not (group B). Pregnancy loss and other pre-specified adverse outcomes were evaluated during follow-up. Results: There were 82 women in group A and 284 in group B. Group A had a higher body mass index ( p = 0.04) and a higher serum TSH level ( p < 0.0001) compared with group B. Group A had fewer pregnancies lost ( n = 5 [6.1%] vs. n = 25 [8.8%]; p = 0.12), low birth weight (LBW) offspring (1.3% vs. 10%; p < 0.001), and no neonates with a five-minute Apgar score ≤7 (0% vs. 7%; p < 0.001) compared with group B. Other pregnancy-related adverse outcomes were similar between the two groups. Inferences remained unchanged after considering different models to adjust for potential predictors of outcome. Conclusions: LT4 therapy is associated with a decreased risk of LBW and a low Apgar score among women with SCH. This association awaits confirmation in randomized trials before the widespread use of LT4 therapy in pregnant women with SCH.

Background: There are significant variabilities in guideline-concordant documentation in asthma care. However, assessing clinician’s documentation is not feasible using only structured data but requires labor-intensive chart review of electronic health records (EHRs). A certain guideline element in asthma control factors, such as review inhaler techniques, requires context understanding to correctly capture from EHR free text.Methods: The study data consist of two sets: (1) manual chart reviewed data—1039 clinical notes of 300 patients with asthma diagnosis, and (2) weakly labeled data (distant supervision)—27,363 clinical notes from 800 patients with asthma diagnosis. A context-aware language model, Bidirectional Encoder Representations from Transformers (BERT) was developed to identify inhaler techniques in EHR free text. Both original BERT and clinical BioBERT (cBERT) were applied with a cost-sensitivity to deal with imbalanced data. The distant supervision using weak labels by rules was also incorporated to augment the training set and alleviate a costly manual labeling process in the development of a deep learning algorithm. A hybrid approach using post-hoc rules was also explored to fix BERT model errors. The performance of BERT with/without distant supervision, hybrid, and rule-based models were compared in precision, recall, F-score, and accuracy.Results: The BERT models on the original data performed similar to a rule-based model in F1-score (0.837, 0.845, and 0.838 for rules, BERT, and cBERT, respectively). The BERT models with distant supervision produced higher performance (0.853 and 0.880 for BERT and cBERT, respectively) than without distant supervision and a rule-based model. The hybrid models performed best in F1-score of 0.877 and 0.904 over the distant supervision on BERT and cBERT.Conclusions: The proposed BERT models with distant supervision demonstrated its capability to identify inhaler techniques in EHR free text, and outperformed both the rule-based model and BERT models trained on the original data. With a distant supervision approach, we may alleviate costly manual chart review to generate the large training data required in most deep learning-based models. A hybrid model was able to fix BERT model errors and further improve the performance.

Little is known about the impact of socioeconomic status (SES) as a key element of social determinants of health on intensive care unit (ICU) outcomes for adults. We assessed whether a validated individual SES index termed HOUSES (HOUsing-based SocioEconomic status index) derived from housing features was associated with short-term outcomes of critical illness including ICU mortality, ICU-free days, hospital-free days, and ICU readmission. We performed a population-based cohort study of adult patients living in Olmsted County, Minnesota, admitted to 7 intensive care units at Mayo Clinic from 2011 to 2014. We compared outcomes between the lowest SES group (HOUSES quartile 1 [Q1]) and the higher SES group (HOUSES Q2-4). We stratified the cohort based on age (50 years), those with higher SES (HOUSES Q2-4) compared to those with lower SES (HOUSES Q1) had lower mortality rates (hazard ratio = 0.72; 95% CI: 0.56-0.93; adjusted P = .01), increased ICU-free days (mean 1.08 days; 95% CI: 0.34-1.84; adjusted P = .004), and increased hospital-free days (mean 1.20 days; 95% CI: 0.45-1.96; adjusted P = .002). There were no differences in ICU readmission rates (OR = 0.74; 95% CI: 0.55-1.00; P = .051). Individual-level SES may be an important determinant or predictor of critical care outcomes in older adults. Housing-based socioeconomic status may be a useful tool for enhancing critical care research and practice.