Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice
Physician turnover places a heavy burden on the healthcare industry, patients, physicians, and their families. Having a mechanism in place to identify physicians at risk for departure could help target appropriate interventions that prevent departure. We have collected physician characteristics, ele...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9891518/?tool=EBI |
| _version_ | 1811171970615607296 |
|---|---|
| author | Kevin Lopez Huan Li Hyung Paek Brian Williams Bidisha Nath Edward R. Melnick Andrew J. Loza |
| author_facet | Kevin Lopez Huan Li Hyung Paek Brian Williams Bidisha Nath Edward R. Melnick Andrew J. Loza |
| author_sort | Kevin Lopez |
| collection | DOAJ |
| description | Physician turnover places a heavy burden on the healthcare industry, patients, physicians, and their families. Having a mechanism in place to identify physicians at risk for departure could help target appropriate interventions that prevent departure. We have collected physician characteristics, electronic health record (EHR) use patterns, and clinical productivity data from a large ambulatory based practice of non-teaching physicians to build a predictive model. We use several techniques to identify possible intervenable variables. Specifically, we used gradient boosted trees to predict the probability of a physician departing within an interval of 6 months. Several variables significantly contributed to predicting physician departure including tenure (time since hiring date), panel complexity, physician demand, physician age, inbox, and documentation time. These variables were identified by training, validating, and testing the model followed by computing SHAP (SHapley Additive exPlanation) values to investigate which variables influence the model’s prediction the most. We found these top variables to have large interactions with other variables indicating their importance. Since these variables may be predictive of physician departure, they could prove useful to identify at risk physicians such who would benefit from targeted interventions. |
| first_indexed | 2024-04-10T17:23:53Z |
| format | Article |
| id | doaj.art-8ac89ae2f7e74bcea8c5f67d0177a79d |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-10T17:23:53Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-8ac89ae2f7e74bcea8c5f67d0177a79d2023-02-05T05:30:58ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01182Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practiceKevin LopezHuan LiHyung PaekBrian WilliamsBidisha NathEdward R. MelnickAndrew J. LozaPhysician turnover places a heavy burden on the healthcare industry, patients, physicians, and their families. Having a mechanism in place to identify physicians at risk for departure could help target appropriate interventions that prevent departure. We have collected physician characteristics, electronic health record (EHR) use patterns, and clinical productivity data from a large ambulatory based practice of non-teaching physicians to build a predictive model. We use several techniques to identify possible intervenable variables. Specifically, we used gradient boosted trees to predict the probability of a physician departing within an interval of 6 months. Several variables significantly contributed to predicting physician departure including tenure (time since hiring date), panel complexity, physician demand, physician age, inbox, and documentation time. These variables were identified by training, validating, and testing the model followed by computing SHAP (SHapley Additive exPlanation) values to investigate which variables influence the model’s prediction the most. We found these top variables to have large interactions with other variables indicating their importance. Since these variables may be predictive of physician departure, they could prove useful to identify at risk physicians such who would benefit from targeted interventions.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9891518/?tool=EBI |
| spellingShingle | Kevin Lopez Huan Li Hyung Paek Brian Williams Bidisha Nath Edward R. Melnick Andrew J. Loza Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice PLoS ONE |
| title | Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice |
| title_full | Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice |
| title_fullStr | Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice |
| title_full_unstemmed | Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice |
| title_short | Predicting physician departure with machine learning on EHR use patterns: A longitudinal cohort from a large multi-specialty ambulatory practice |
| title_sort | predicting physician departure with machine learning on ehr use patterns a longitudinal cohort from a large multi specialty ambulatory practice |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9891518/?tool=EBI |
| work_keys_str_mv | AT kevinlopez predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT huanli predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT hyungpaek predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT brianwilliams predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT bidishanath predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT edwardrmelnick predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice AT andrewjloza predictingphysiciandeparturewithmachinelearningonehrusepatternsalongitudinalcohortfromalargemultispecialtyambulatorypractice |