Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study
Abstract BackgroundDiabetes-related lower extremity complications, such as foot ulceration and amputation, are on the rise, currently affecting nearly 131 million people worldwide. Methods for early detection of individuals at high risk remain elusive. While data-driven diabet...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2025-05-01
|
| Series: | JMIR Diabetes |
| Online Access: | https://diabetes.jmir.org/2025/1/e60141 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849472232172027904 |
|---|---|
| author | Alyce S Adams Catherine Lee Gabriel Escobar Elizabeth A Bayliss Brian Callaghan Michael Horberg Julie A Schmittdiel Connie Trinacty Lisa K Gilliam Eileen Kim Nima S Hejazi Lin Ma Romain Neugebauer |
| author_facet | Alyce S Adams Catherine Lee Gabriel Escobar Elizabeth A Bayliss Brian Callaghan Michael Horberg Julie A Schmittdiel Connie Trinacty Lisa K Gilliam Eileen Kim Nima S Hejazi Lin Ma Romain Neugebauer |
| author_sort | Alyce S Adams |
| collection | DOAJ |
| description |
Abstract
BackgroundDiabetes-related lower extremity complications, such as foot ulceration and amputation, are on the rise, currently affecting nearly 131 million people worldwide. Methods for early detection of individuals at high risk remain elusive. While data-driven diabetic polyneuropathy algorithms exist, high-performing, clinically useful tools to assess risk are needed to improve clinical care.
ObjectiveThis study aimed to develop an electronic medical record–based machine learning algorithm that would predict lower extremity complications.
MethodsWe conducted a retrospective longitudinal cohort study to predict the risk of lower extremity complications within 24 months of an initial diagnosis of diabetic polyneuropathy. From an initial cohort of 468,162 individuals with at least 1 diagnosis of diabetic polyneuropathy at one of 2 multispecialty health care systems (based in northern California and Colorado) between April 2012 and December 2016, we created an analytic cohort of 48,209 adults with continuous enrollment, who were newly diagnosed with no evidence of end-of-life care. The outcome was any lower extremity complication, including foot ulceration, osteomyelitis, gangrene, or lower extremity amputation. We randomly split the data into training (38,569/48209; 80%) and testing (9,640/48209; 20%) datasets. In the training dataset, we used super Learner (SL), an ensemble learning method that employs cross-validation and combines multiple candidate risk predictors, into a single risk predictor. We evaluated the performance of the SL risk predictor in the testing dataset using the receiver operating characteristic curve and a calibration plot.
ResultsOf the 48,209 individuals in the cohort, 2327 developed a lower extremity complication during follow-up. The SL risk estimator exhibited good discrimination (AUC=0.845, 95% CI 0.826-0.863) and calibration. A modified version of our SL algorithm, simplified to facilitate real-world adoption, had only slightly reduced discrimination (AUC=0.817, 95%CI 0.797-0.837). The modified version slightly outperformed the naïve logistic regression model (AUC=0.804, 95% CI 0.783-0.825) in terms of precision gained relative to the frequency of alerts and number of patients that needed to be evaluated.
ConclusionsWe have built a machine learning–based risk estimator with the potential to improve clinical detection of diabetic patients at high risk for lower extremity complications at the time of an initial diabetic polyneuropathy diagnosis. The algorithm exhibited good discriminant validity and calibration using only data from the electronic medical record. Additional research will be needed to identify optimal contexts and strategies for maximizing algorithmic fairness in both interpretation and deployment. |
| format | Article |
| id | doaj-art-006730fef7934201ab8e49e8d59a85f9 |
| institution | Kabale University |
| issn | 2371-4379 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Diabetes |
| spelling | doaj-art-006730fef7934201ab8e49e8d59a85f92025-08-20T03:24:36ZengJMIR PublicationsJMIR Diabetes2371-43792025-05-0110e60141e6014110.2196/60141Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort StudyAlyce S Adamshttp://orcid.org/0000-0002-0146-4996Catherine Leehttp://orcid.org/0000-0003-0008-9052Gabriel Escobarhttp://orcid.org/0000-0003-2540-3327Elizabeth A Baylisshttp://orcid.org/0000-0002-5264-0670Brian Callaghanhttp://orcid.org/0000-0002-8885-6748Michael Horberghttp://orcid.org/0000-0002-6971-7646Julie A Schmittdielhttp://orcid.org/0000-0001-8995-5345Connie Trinactyhttp://orcid.org/0000-0002-4799-1994Lisa K Gilliamhttp://orcid.org/0000-0001-7054-1409Eileen Kimhttp://orcid.org/0009-0007-5179-505XNima S Hejazihttp://orcid.org/0000-0002-7127-2789Lin Mahttp://orcid.org/0000-0002-7009-8213Romain Neugebauerhttp://orcid.org/0000-0002-6085-4564 Abstract BackgroundDiabetes-related lower extremity complications, such as foot ulceration and amputation, are on the rise, currently affecting nearly 131 million people worldwide. Methods for early detection of individuals at high risk remain elusive. While data-driven diabetic polyneuropathy algorithms exist, high-performing, clinically useful tools to assess risk are needed to improve clinical care. ObjectiveThis study aimed to develop an electronic medical record–based machine learning algorithm that would predict lower extremity complications. MethodsWe conducted a retrospective longitudinal cohort study to predict the risk of lower extremity complications within 24 months of an initial diagnosis of diabetic polyneuropathy. From an initial cohort of 468,162 individuals with at least 1 diagnosis of diabetic polyneuropathy at one of 2 multispecialty health care systems (based in northern California and Colorado) between April 2012 and December 2016, we created an analytic cohort of 48,209 adults with continuous enrollment, who were newly diagnosed with no evidence of end-of-life care. The outcome was any lower extremity complication, including foot ulceration, osteomyelitis, gangrene, or lower extremity amputation. We randomly split the data into training (38,569/48209; 80%) and testing (9,640/48209; 20%) datasets. In the training dataset, we used super Learner (SL), an ensemble learning method that employs cross-validation and combines multiple candidate risk predictors, into a single risk predictor. We evaluated the performance of the SL risk predictor in the testing dataset using the receiver operating characteristic curve and a calibration plot. ResultsOf the 48,209 individuals in the cohort, 2327 developed a lower extremity complication during follow-up. The SL risk estimator exhibited good discrimination (AUC=0.845, 95% CI 0.826-0.863) and calibration. A modified version of our SL algorithm, simplified to facilitate real-world adoption, had only slightly reduced discrimination (AUC=0.817, 95%CI 0.797-0.837). The modified version slightly outperformed the naïve logistic regression model (AUC=0.804, 95% CI 0.783-0.825) in terms of precision gained relative to the frequency of alerts and number of patients that needed to be evaluated. ConclusionsWe have built a machine learning–based risk estimator with the potential to improve clinical detection of diabetic patients at high risk for lower extremity complications at the time of an initial diabetic polyneuropathy diagnosis. The algorithm exhibited good discriminant validity and calibration using only data from the electronic medical record. Additional research will be needed to identify optimal contexts and strategies for maximizing algorithmic fairness in both interpretation and deployment.https://diabetes.jmir.org/2025/1/e60141 |
| spellingShingle | Alyce S Adams Catherine Lee Gabriel Escobar Elizabeth A Bayliss Brian Callaghan Michael Horberg Julie A Schmittdiel Connie Trinacty Lisa K Gilliam Eileen Kim Nima S Hejazi Lin Ma Romain Neugebauer Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study JMIR Diabetes |
| title | Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study |
| title_full | Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study |
| title_fullStr | Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study |
| title_full_unstemmed | Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study |
| title_short | Estimating the Risk of Lower Extremity Complications in Adults Newly Diagnosed With Diabetic Polyneuropathy: Retrospective Cohort Study |
| title_sort | estimating the risk of lower extremity complications in adults newly diagnosed with diabetic polyneuropathy retrospective cohort study |
| url | https://diabetes.jmir.org/2025/1/e60141 |
| work_keys_str_mv | AT alycesadams estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT catherinelee estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT gabrielescobar estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT elizabethabayliss estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT briancallaghan estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT michaelhorberg estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT julieaschmittdiel estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT connietrinacty estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT lisakgilliam estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT eileenkim estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT nimashejazi estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT linma estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy AT romainneugebauer estimatingtheriskoflowerextremitycomplicationsinadultsnewlydiagnosedwithdiabeticpolyneuropathyretrospectivecohortstudy |