Developing an automated algorithm for identification of children and adolescents with diabetes using electronic health records from the OneFlorida + clinical research network
Authors' conclusion
We developed a highly accurate EHR-based CP for diabetes in youth based on EHR data from UF Health. Consistent with prior studies, T2D was more difficult to identify using these methods.
Study context
- Population
- Individuals aged <18 years with presumed diabetes (HbA1c ≥6.5%, fasting glucose ≥126 mg/dL, random plasma glucose ≥200 mg/dL, diabetes diagnosis code or diabetes medication), UF Health System electronic health records (EHRs) 2012–2020
- Participants
- 536 randomly selected presumed cases
- Comparison
- Automated computable phenotype (CP) algorithm vs chart review for identifying diabetes type
- T1Dn = 159
- T2Dn = 64
- Outcome
- Accuracy of CP algorithm vs chart review
Methodological detailschart review by four independent reviewers, EHR-based computable phenotype algorithm
- Chart review by four independent reviewers
- EHR-based computable phenotype algorithm
Performance
As reported in the abstract
| Strategy | Sensitivity | Specificity | PPV |
|---|---|---|---|
| CP algorithm (T1D) | 94% | 98% | 96% |
| CP algorithm (T2D) | 95% | 95% | 73% |
Sensitivity
CP algorithm (T1D)94%
CP algorithm (T2D)95%
Specificity
CP algorithm (T1D)98%
CP algorithm (T2D)95%
PPV
CP algorithm (T1D)96%
CP algorithm (T2D)73%
Further findings
Diabetes type on chart review
- T1D among 536 reviewed cases
- 159
- T2D among 536 reviewed cases
- 64
Original publication
Developing an automated algorithm for identification of children and adolescents with diabetes using electronic health records from the OneFlorida + clinical research network. Diabetes Obes Metab. Published 30 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →