Application of a Natural Language Processing Algorithm to Asthma Ascertainment. An Automated Chart Review.

RATIONALE: Difficulty of asthma ascertainment and its associated methodologic heterogeneity have created significant barriers to asthma care and research. OBJECTIVES: We evaluated the validity of an existing natural language processing (NLP) algorithm for asthma criteria to enable an automated chart review using electronic medical records (EMRs). METHODS: The study was designed as a retrospective birth cohort study using a random sample of 500 subjects from the 1997-2007 Mayo Birth Cohort who were born at Mayo Clinic and enrolled in primary pediatric care at Mayo Clinic Rochester. Performance of NLP-based asthma ascertainment using predetermined asthma criteria was assessed by determining both criterion validity (chart review of EMRs by abstractor as a gold standard) and construct validity (association with known risk factors for asthma, such as allergic rhinitis). MEASUREMENTS AND MAIN RESULTS: After excluding three subjects whose respiratory symptoms could be attributed to other conditions (e.g., tracheomalacia), among the remaining eligible 497 subjects, 51% were male, 77% white persons, and the median age at last follow-up date was 11.5 years. The asthma prevalence was 31% in the study cohort. Sensitivity, specificity, positive predictive value, and negative predictive value for NLP algorithm in predicting asthma status were 97%, 95%, 90%, and 98%, respectively. The risk factors for asthma (e.g., allergic rhinitis) that were identified either by NLP or the abstractor were the same. CONCLUSIONS: Asthma ascertainment through NLP should be considered in the era of EMRs because it can enable large-scale clinical studies in a more time-efficient manner and improve the recognition and care of childhood asthma in practice.

Automated chart review utilizing natural language processing algorithm for asthma predictive index.

BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSION: 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.

What accounts for the association between late preterm births and risk of asthma?

BACKGROUND: Although results of many studies have indicated an increased risk of asthma in former late preterm (LPT) infants, most of these studies did not fully address covariate imbalance. OBJECTIVE: To compare the cumulative frequency of asthma in a population-based cohort of former LPT infants to that of matched term infants in their early childhood, when accounting for covariate imbalance. METHODS: From a population-based birth cohort of children born 2002-2006 in Olmsted County, Minnesota, we assessed a random sample of LPT (34 to 36 6/7 weeks) and frequency-matched term (37 to 40 6/7 weeks) infants. The subjects were followed-up through 2010 or censored based on the last date of contact, with the asthma status based on predetermined criteria. The Kaplan-Meier method was used to estimate the cumulative incidence of asthma during the study period. Cox models were used to estimate the hazard ratio and 95% confidence interval for the risk of asthma, when adjusting for potential confounders. RESULTS: LPT infants (n = 282) had a higher cumulative frequency of asthma than did term infants (n = 297), 29.9 versus 19.5%, respectively; p = 0.01. After adjusting for covariates associated with the risk of asthma, an LPT birth was not associated with a risk of asthma, whereas maternal smoking during pregnancy was associated with a risk of asthma. CONCLUSION: LPT birth was not independently associated with a risk of asthma and other atopic conditions. Clinicians should make an effort to reduce exposure to smoking during pregnancy as a modifiable risk factor for asthma.

Risk of Asthma in Late Preterm Infants: A Propensity Score Approach.

BACKGROUND: The risk of asthma, specifically in former late preterm infants, has not been well defined. Covariate imbalance and lack of controlling for this has led to inconsistent results in prior studies. OBJECTIVE: The objective of this study was to determine the risk of asthma in former late preterm infants using a propensity score approach. METHODS: The study was a population-based birth cohort study. Study subjects were all children born in Rochester, Minn, between 1976 and 1982. Asthma status during the first 7 years of life was assessed by applying predetermined criteria. The propensity score was formulated using 15 covariates by fitting a logistic regression model for late preterm birth versus term birth. We applied the propensity score method to match late preterm infants (34 0/7 to 36 6/7 weeks of gestation) to term infants (37 0/7 to 40 6/7 weeks of gestation) within a caliper of 0.2 standard deviation of logit of propensity score. RESULTS: Of the eligible 7040 infants, 5915 children had complete data. Before propensity score matching, late preterm infants had a higher risk of asthma (20 of 262, 7.6%) compared with full-term infants (272 of 5653, 4.8%) (P = .039). There was significant covariate imbalance between comparison groups. After matching with propensity scores, we found that former late preterm infants had a similar risk of asthma to the matched full-term infants (6.6% vs 7.7%, respectively, P = .61), and the result was consistent with covariate-adjustment Cox regression models controlling for significant covariates (P = .57). CONCLUSION: A late preterm birth history is not independently associated with childhood asthma, as the reported risk of asthma among former late preterm infants appears to be due to covariate imbalance.

ADHD and learning disabilities in former late preterm infants: a population-based birth cohort.

BACKGROUND AND OBJECTIVE: Previous studies suggest that former late preterm infants are at increased risk for learning and behavioral problems compared with term infants. These studies have primarily used referred clinical samples of children followed only until early school age. Our objective was to determine the cumulative incidence of attention deficit/hyperactivity disorder (ADHD) and learning disabilities (LD) in former late preterm versus term infants in a population-based birth cohort. METHODS: Subjects included all children born 1976 to 1982 in Rochester, MN who remained in the community after 5 years. This study focused on the comparison of subjects in 2 subgroups, late preterm (34 to <37 weeks) and term (37 to <42 weeks). School and medical records were available to identify individuals who met research criteria for ADHD and LD in reading, written language, and math. The Kaplan-Meier method was used to estimate the cumulative incidence of each condition by 19 years of age. Cox models were fit to evaluate the association between gestational age group and condition, after adjusting for maternal education and perinatal complications. RESULTS: We found no statistically significant differences in the cumulative incidence of ADHD or LD between the late preterm (N = 256) versus term (N = 4419) groups: ADHD (cumulative incidence by age 19 years, 7.7% vs 7.2%; P = .84); reading LD (14.2% vs 13.1%; P = .57); written language LD (13.5% vs 15.7%; P = .36), and math LD (16.1% vs 15.5%; P = .89). CONCLUSIONS: These data from a population-based birth cohort indicate that former late preterm infants have similar rates of LD and ADHD as term infants.