Algorithmic identification of atypical diabetes in electronic health record (EHR) systems.
<h4>Aims</h4>Understanding atypical forms of diabetes (AD) may advance precision medicine, but methods to identify such patients are needed. We propose an electronic health record (EHR)-based algorithmic approach to identify patients who may have AD, specifically those with insulin-suffi...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0278759 |
| _version_ | 1797856367530213376 |
|---|---|
| author | Sara J Cromer Victoria Chen Christopher Han William Marshall Shekina Emongo Evelyn Greaux Tim Majarian Jose C Florez Josep Mercader Miriam S Udler |
| author_facet | Sara J Cromer Victoria Chen Christopher Han William Marshall Shekina Emongo Evelyn Greaux Tim Majarian Jose C Florez Josep Mercader Miriam S Udler |
| author_sort | Sara J Cromer |
| collection | DOAJ |
| description | <h4>Aims</h4>Understanding atypical forms of diabetes (AD) may advance precision medicine, but methods to identify such patients are needed. We propose an electronic health record (EHR)-based algorithmic approach to identify patients who may have AD, specifically those with insulin-sufficient, non-metabolic diabetes, in order to improve feasibility of identifying these patients through detailed chart review.<h4>Methods</h4>Patients with likely T2D were selected using a validated machine-learning (ML) algorithm applied to EHR data. "Typical" T2D cases were removed by excluding individuals with obesity, evidence of dyslipidemia, antibody-positive diabetes, or cystic fibrosis. To filter out likely type 1 diabetes (T1D) cases, we applied six additional "branch algorithms," relying on various clinical characteristics, which resulted in six overlapping cohorts. Diabetes type was classified by manual chart review as atypical, not atypical, or indeterminate due to missing information.<h4>Results</h4>Of 114,975 biobank participants, the algorithms collectively identified 119 (0.1%) potential AD cases, of which 16 (0.014%) were confirmed after expert review. The branch algorithm that excluded T1D based on outpatient insulin use had the highest percentage yield of AD (13 of 27; 48.2% yield). Together, the 16 AD cases had significantly lower BMI and higher HDL than either unselected T1D or T2D cases identified by ML algorithms (P<0.05). Compared to the ML T1D group, the AD group had a significantly higher T2D polygenic score (P<0.01) and lower hemoglobin A1c (P<0.01).<h4>Conclusion</h4>Our EHR-based algorithms followed by manual chart review identified collectively 16 individuals with AD, representing 0.22% of biobank enrollees with T2D. With a maximum yield of 48% cases after manual chart review, our algorithms have the potential to drastically improve efficiency of AD identification. Recognizing patients with AD may inform on the heterogeneity of T2D and facilitate enrollment in studies like the Rare and Atypical Diabetes Network (RADIANT). |
| first_indexed | 2024-04-09T20:39:16Z |
| format | Article |
| id | doaj.art-96e0a38aa0204d458c0b848812c671d4 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-09T20:39:16Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-96e0a38aa0204d458c0b848812c671d42023-03-30T05:31:30ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-011712e027875910.1371/journal.pone.0278759Algorithmic identification of atypical diabetes in electronic health record (EHR) systems.Sara J CromerVictoria ChenChristopher HanWilliam MarshallShekina EmongoEvelyn GreauxTim MajarianJose C FlorezJosep MercaderMiriam S Udler<h4>Aims</h4>Understanding atypical forms of diabetes (AD) may advance precision medicine, but methods to identify such patients are needed. We propose an electronic health record (EHR)-based algorithmic approach to identify patients who may have AD, specifically those with insulin-sufficient, non-metabolic diabetes, in order to improve feasibility of identifying these patients through detailed chart review.<h4>Methods</h4>Patients with likely T2D were selected using a validated machine-learning (ML) algorithm applied to EHR data. "Typical" T2D cases were removed by excluding individuals with obesity, evidence of dyslipidemia, antibody-positive diabetes, or cystic fibrosis. To filter out likely type 1 diabetes (T1D) cases, we applied six additional "branch algorithms," relying on various clinical characteristics, which resulted in six overlapping cohorts. Diabetes type was classified by manual chart review as atypical, not atypical, or indeterminate due to missing information.<h4>Results</h4>Of 114,975 biobank participants, the algorithms collectively identified 119 (0.1%) potential AD cases, of which 16 (0.014%) were confirmed after expert review. The branch algorithm that excluded T1D based on outpatient insulin use had the highest percentage yield of AD (13 of 27; 48.2% yield). Together, the 16 AD cases had significantly lower BMI and higher HDL than either unselected T1D or T2D cases identified by ML algorithms (P<0.05). Compared to the ML T1D group, the AD group had a significantly higher T2D polygenic score (P<0.01) and lower hemoglobin A1c (P<0.01).<h4>Conclusion</h4>Our EHR-based algorithms followed by manual chart review identified collectively 16 individuals with AD, representing 0.22% of biobank enrollees with T2D. With a maximum yield of 48% cases after manual chart review, our algorithms have the potential to drastically improve efficiency of AD identification. Recognizing patients with AD may inform on the heterogeneity of T2D and facilitate enrollment in studies like the Rare and Atypical Diabetes Network (RADIANT).https://doi.org/10.1371/journal.pone.0278759 |
| spellingShingle | Sara J Cromer Victoria Chen Christopher Han William Marshall Shekina Emongo Evelyn Greaux Tim Majarian Jose C Florez Josep Mercader Miriam S Udler Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. PLoS ONE |
| title | Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. |
| title_full | Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. |
| title_fullStr | Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. |
| title_full_unstemmed | Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. |
| title_short | Algorithmic identification of atypical diabetes in electronic health record (EHR) systems. |
| title_sort | algorithmic identification of atypical diabetes in electronic health record ehr systems |
| url | https://doi.org/10.1371/journal.pone.0278759 |
| work_keys_str_mv | AT sarajcromer algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT victoriachen algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT christopherhan algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT williammarshall algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT shekinaemongo algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT evelyngreaux algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT timmajarian algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT josecflorez algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT josepmercader algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems AT miriamsudler algorithmicidentificationofatypicaldiabetesinelectronichealthrecordehrsystems |