Vanderbilt University Medical Center

Developmental Language Disorder

APT-DLD
Version 1.0, July 2020

Automated Phenotyping Tool for identifying DLD cases in health-systems data (APT-DLD) is an algorithm for classifying/identifying developmental language disorder cases in electronic health records system data. APT-DLD can be used to:
1. Identify pediatric DLD cases from electronic health record systems using ICD9 and ICD10 codes
2. Study epidemiology and population-level charateristics of DLD from EHRs

The How-To guide for using APT-DLD is provided in the files listed below.

Owner Phenotyping Groups: 
Final

Diverticular Disease Severity, Left Colonic

This algorithm builds off prior phenotyping work from Pacheco & Thompson available in the PheKB phenotype "Diverticulosis and Diverticulitis" as well as the manuscripts from Joo et al (2023)(1) and De Roo et al (2023) (2) . The objective is to approximate diverticular disease severity from the electronic medical record into groups of Diverticulosis, Mild Diverticulitis, and Operative or Recurrent Inpatient Diverticulitis.

Final

Opioid-exposed infant clinical indicators

Objective 

We leveraged existing data from a single electronic health care system in the southeastern United States to demonstrate the feasibility of measuring quality indicators for the hospital-based care of opioid-exposed newborns using existing data infrastructure. Additionally, we identified other key variables related to the care of opioid-exposed maternal-infant dyads.   

Patients and Methods 

Final

Peanut Allergy

Food allergy is defined as an immune response that occurs reproducibly to a given food, typically an immunoglobulin E (IgE)-mediated clinical reaction to specific protein epitopes.  Over the last 20-30 years, food allergy has grown into a major public health problem.  Peanut allergy is a common type of food allergy that accounts for a disproportionate number of fatal and near-fatal anaphylactic events amongst all the common food allergens.

Final

Systemic lupus erythematosus (SLE)

We used Vanderbilt’s Synthetic Derivative (SD), a de-identified version of the EHR, with 2.5 million subjects. We selected all individuals with at least one SLE ICD-9 code (710.0) yielding 5959 individuals. To create a training set, 200 were randomly selected for chart review. A subject was defined as a case if diagnosed with SLE by a rheumatologist, nephrologist, or dermatologist.

Final