Novel Pediatric Height Outlier Detection Methodology for Electronic Health Records via Machine Learning With Monotonic Bayesian Additive Regression Trees.

OBJECTIVE: To create a new methodology that has a single simple rule to identify height outliers in the electronic health records (EHR) of children. METHODS: We constructed 2 independent cohorts of children 2 to 8 years old to train and validate a model predicting heights from age, gender, race and weight with monotonic Bayesian additive regression trees. The training cohort consisted of 1376 children where outliers were unknown. The testing cohort consisted of 318 patients that were manually reviewed retrospectively to identify height outliers. RESULTS: The amount of variation explained in height values by our model, R2 , was 82.2% and 75.3% in the training and testing cohorts, respectively. The discriminatory ability to assess height outliers in the testing cohort as assessed by the area under the receiver operating characteristic curve was excellent, 0.841. Based on a relatively aggressive cutoff of 0.075, the outlier sensitivity is 0.713, the specificity 0.793; the positive predictive value 0.615 and the negative predictive value is 0.856. CONCLUSIONS: We have developed a new reliable, largely automated, outlier detection method which is applicable to the identification of height outliers in the pediatric EHR. This methodology can be applied to assess the veracity of height measurements ensuring reliable indices of body proportionality such as body mass index.

A Systematic Review of Micronutrient Deficiencies in Pediatric Inflammatory Bowel Disease.

BACKGROUND: This systematic review critically analyzes the current research on micronutrient deficiency in children with inflammatory bowel disease (IBD) and synthesizes these data to provide evidence-based guidelines for nutritional surveillance in this population. METHODS: We searched 5 databases (Ovid Medline, PubMed, Scopus, CINAHL, and Cochrane Library) for studies evaluating micronutrients in patients with IBD using the following inclusion criteria: 1) original research, 2) published 1996 or later; 3) published in English; 4) human subjects; and 5) containing pediatric data. Studies were reviewed and included based on the strength of research methods. Data on the prevalence of micronutrient deficiencies in pediatric patients with IBD and risk factors for micronutrient deficiency in these patients were extracted from included studies and compared and discussed in preparation of the proposed guidelines and manuscript. RESULTS: A total of 39 studies were included in the final review. The data presented in these studies show that iron deficiency and vitamin D deficiency are common in pediatric patients with IBD. Vitamin B12 and folate deficiency are rare. Zinc deficiency, while not common, occurs at a higher rate in patients with Crohn's disease than in healthy controls. There was limited data on vitamins A, E, and C, and selenium, but deficiency of these micronutrients seems rare. CONCLUSIONS: We recommend annual surveillance of iron and vitamin D in pediatric patients with IBD regardless of disease activity or phenotype. Zinc should be monitored annually in patients with Crohn's disease. There is insufficient evidence to support routine screening for other micronutrient deficiencies.