Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction
Purpose: To develop and internally validate a machine-learning algorithm to reliably predict cost after anterior cruciate ligament reconstruction (ACLR). Methods: A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent el...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-12-01
|
| Series: | Arthroscopy, Sports Medicine, and Rehabilitation |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666061X21002029 |
| _version_ | 1819101859640508416 |
|---|---|
| author | Yining Lu, M.D. Kyle Kunze, M.D. Matthew R. Cohn, M.D. Ophelie Lavoie-Gagne, M.D. Evan Polce, B.S. Benedict U. Nwachukwu, M.D., M.B.A. Brian Forsythe, M.D. |
| author_facet | Yining Lu, M.D. Kyle Kunze, M.D. Matthew R. Cohn, M.D. Ophelie Lavoie-Gagne, M.D. Evan Polce, B.S. Benedict U. Nwachukwu, M.D., M.B.A. Brian Forsythe, M.D. |
| author_sort | Yining Lu, M.D. |
| collection | DOAJ |
| description | Purpose: To develop and internally validate a machine-learning algorithm to reliably predict cost after anterior cruciate ligament reconstruction (ACLR). Methods: A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective ACLR from 2015 to 2016. Features included in initial models consisted of patient characteristics (age, sex, insurance status, income, medical comorbidities as classified by the Clinical Classifications Software diagnosis code) as well as intraoperative variables (type of anesthesia and procedure-specific factors). Models were generated to predict total charges using 4 algorithms: random forest, extreme gradient boost, elastic net penalized regression, and support vector machines with radial kernels. Training was performed with 10-fold cross-validation followed by internal validation via 0.632 bootstrapping. Model discriminative performance was assessed by area under the receiver operating characteristic curve, calibration, and the Brier score. Decision curve analysis was performed to demonstrate the net benefit of using the final model in practice. Results: In total, 7,311 patients undergoing ambulatory ACLR were included. The random forest model demonstrated the best performance assessed via internal validation (area under the curve = 0.85), calibration, and the Brier score (0.208). Cost incurred was influenced by anesthesia type, operating room time, and number of chronic comorbidities. Decision curve analysis revealed a net benefit for use of the random forest model and the model was integrated into a web-based open-access application. Conclusions: The random forest model predicted cost after ambulatory ACLR using a large, statewide database with good performance. The top variables found to predict increased charges were general anesthesia, operating room time, meniscal repair, self-pay insurance, patient neighborhood characteristics, and number of chronic conditions. Level of Evidence: III, retrospective cohort study. |
| first_indexed | 2024-12-22T01:25:22Z |
| format | Article |
| id | doaj.art-0cc9dfe9de1945c3b46c07d49e4868a0 |
| institution | Directory Open Access Journal |
| issn | 2666-061X |
| language | English |
| last_indexed | 2024-12-22T01:25:22Z |
| publishDate | 2021-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Arthroscopy, Sports Medicine, and Rehabilitation |
| spelling | doaj.art-0cc9dfe9de1945c3b46c07d49e4868a02022-12-21T18:43:37ZengElsevierArthroscopy, Sports Medicine, and Rehabilitation2666-061X2021-12-0136e2033e2045Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament ReconstructionYining Lu, M.D.0Kyle Kunze, M.D.1Matthew R. Cohn, M.D.2Ophelie Lavoie-Gagne, M.D.3Evan Polce, B.S.4Benedict U. Nwachukwu, M.D., M.B.A.5Brian Forsythe, M.D.6Department of Orthopaedic Surgery, Mayo Clinic, Rochester, Minnesota, U.S.A.Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, New York, U.S.A.Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois, U.S.A.Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois, U.S.A.Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois, U.S.A.Department of Orthopaedic Surgery, Mayo Clinic, Rochester, Minnesota, U.S.A.Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois, U.S.A.; Address correspondence to Brian Forsythe, M.D., Rush University Medical Center, 1611 W Harrison St., Ste 400, Chicago, IL 60612.Purpose: To develop and internally validate a machine-learning algorithm to reliably predict cost after anterior cruciate ligament reconstruction (ACLR). Methods: A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective ACLR from 2015 to 2016. Features included in initial models consisted of patient characteristics (age, sex, insurance status, income, medical comorbidities as classified by the Clinical Classifications Software diagnosis code) as well as intraoperative variables (type of anesthesia and procedure-specific factors). Models were generated to predict total charges using 4 algorithms: random forest, extreme gradient boost, elastic net penalized regression, and support vector machines with radial kernels. Training was performed with 10-fold cross-validation followed by internal validation via 0.632 bootstrapping. Model discriminative performance was assessed by area under the receiver operating characteristic curve, calibration, and the Brier score. Decision curve analysis was performed to demonstrate the net benefit of using the final model in practice. Results: In total, 7,311 patients undergoing ambulatory ACLR were included. The random forest model demonstrated the best performance assessed via internal validation (area under the curve = 0.85), calibration, and the Brier score (0.208). Cost incurred was influenced by anesthesia type, operating room time, and number of chronic comorbidities. Decision curve analysis revealed a net benefit for use of the random forest model and the model was integrated into a web-based open-access application. Conclusions: The random forest model predicted cost after ambulatory ACLR using a large, statewide database with good performance. The top variables found to predict increased charges were general anesthesia, operating room time, meniscal repair, self-pay insurance, patient neighborhood characteristics, and number of chronic conditions. Level of Evidence: III, retrospective cohort study.http://www.sciencedirect.com/science/article/pii/S2666061X21002029 |
| spellingShingle | Yining Lu, M.D. Kyle Kunze, M.D. Matthew R. Cohn, M.D. Ophelie Lavoie-Gagne, M.D. Evan Polce, B.S. Benedict U. Nwachukwu, M.D., M.B.A. Brian Forsythe, M.D. Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction Arthroscopy, Sports Medicine, and Rehabilitation |
| title | Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction |
| title_full | Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction |
| title_fullStr | Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction |
| title_full_unstemmed | Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction |
| title_short | Artificial Intelligence Predicts Cost After Ambulatory Anterior Cruciate Ligament Reconstruction |
| title_sort | artificial intelligence predicts cost after ambulatory anterior cruciate ligament reconstruction |
| url | http://www.sciencedirect.com/science/article/pii/S2666061X21002029 |
| work_keys_str_mv | AT yininglumd artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT kylekunzemd artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT matthewrcohnmd artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT ophelielavoiegagnemd artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT evanpolcebs artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT benedictunwachukwumdmba artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction AT brianforsythemd artificialintelligencepredictscostafterambulatoryanteriorcruciateligamentreconstruction |