Generalizability

---

Overview

• Background
• Results
  • ADHD
  • Appendicitis
  • Crohn's Disease
  • Hypothyroidism
  • Peanut Allergy
  • Sickle Cell Disease
  • Sleep Apnea
  • Steroid-Induced Osteonecrosis
  • Systemic Lupus Erythematosus

---

Background

Problem: Computational phenotype definitions may have limited generalizability because they are tailored to specific source vocabularies or hospital systems.

Solution: PhenKnowVec is built on the Observational Medical Outcomes Partnership (OMOP) common data model, which allows it to leverage mappings between source vocabularies (e.g. ICD9-CM, MEDRA, GPT) and standardized clinical terminologies (e.g. SNOMED, LOINC, RxNorm, CPT4).

Experiment: For each phenotype, we examined what information is gained and/or lost when deriving pediatric and adult patient cohorts using different the clinical code sets defined in Table 2. For all comparisons, the Source Vocabulary - Exact None code set was be used as the gold standard.

The results from these experiments are organized by phenotype and listed below.

---

Results

ADHD

Appendicitis

Crohn's Disease

Hypothyroidism

Peanut Allergy

Sickle Cell Disease

Sleep Apnea

Steroid-Induced Osteonecrosis

Systemic Lupus Erythematosus