Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre
Objective To develop models to predict opportunities for improvement in trauma care and compare the performance of these models to the currently used audit filters.Design Retrospective registry-based study.Setting Single-centre, Scandinavian level one equivalent trauma centre.Participants 8220 adult...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMJ Publishing Group
2025-06-01
|
| Series: | BMJ Open |
| Online Access: | https://bmjopen.bmj.com/content/15/6/e099624.full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850224570741555200 |
|---|---|
| author | Martin Gerdin Wärnberg Lovisa Strömmer Jonatan Attergrim Kelvin Szolnoky Olof Brattström Martin Jacobsson Gunilla Wihlke |
| author_facet | Martin Gerdin Wärnberg Lovisa Strömmer Jonatan Attergrim Kelvin Szolnoky Olof Brattström Martin Jacobsson Gunilla Wihlke |
| author_sort | Martin Gerdin Wärnberg |
| collection | DOAJ |
| description | Objective To develop models to predict opportunities for improvement in trauma care and compare the performance of these models to the currently used audit filters.Design Retrospective registry-based study.Setting Single-centre, Scandinavian level one equivalent trauma centre.Participants 8220 adult trauma patients screened for opportunities for improvement between 2013 and 2022.Primary and secondary outcome measures Two machine learning models (logistic regression and XGBoost) and the currently used audit filters were compared. Internal validation by an expanding window approach with annual updates was used for model evaluation. Performance measured by discrimination, calibration, sensitivity and false positive rate of opportunities for improvement prediction.Results A total of 8220 patients, with a mean age of 45 years, were analysed; 69% were men with a mean injury severity score of 12. Opportunities for improvement were identified in 496 (6%) patients. Both the logistic regression and XGBoost models were well-calibrated, with intercalibration indices of 0.02 and 0.02, respectively. The models demonstrated higher areas under the receiver operating characteristic curve (AUCs) (logistic regression: 0.71; XGBoost: 0.74). The XGBoost model had a lower false positive rate at a similar sensitivity (false positive rate: 0.63). The audit filters had an AUC of 0.62 and a false positive rate of 0.67.Conclusions The logistic regression and XGBoost models outperformed audit filters in predicting opportunities for improvement among adult trauma patients and can potentially be used to improve systems for selecting patients for trauma peer review. |
| format | Article |
| id | doaj-art-e747208b014d43328de7231501641ddd |
| institution | OA Journals |
| issn | 2044-6055 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | BMJ Publishing Group |
| record_format | Article |
| series | BMJ Open |
| spelling | doaj-art-e747208b014d43328de7231501641ddd2025-08-20T02:05:35ZengBMJ Publishing GroupBMJ Open2044-60552025-06-0115610.1136/bmjopen-2025-099624Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centreMartin Gerdin Wärnberg0Lovisa Strömmer1Jonatan Attergrim2Kelvin Szolnoky3Olof Brattström4Martin Jacobsson5Gunilla Wihlke6Department of Global Public Health, Karolinska Institutet, Stockholm, SwedenDepartment of Surgery, Capio S:t Görans Hospital, Stockholm, SwedenDepartment of Global Public Health, Karolinska Institutet, Stockholm, SwedenDepartment of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, SwedenDepartment of Anesthesiology, Mora Hospital, Mora, SwedenDepartment of Biomedical Engineering and Health Systems, KTH Royal Institute of Technology, Stockholm, SwedenPerioperative Medicine and Intensive Care, Karolinska University Hospital, Stockholm, SwedenObjective To develop models to predict opportunities for improvement in trauma care and compare the performance of these models to the currently used audit filters.Design Retrospective registry-based study.Setting Single-centre, Scandinavian level one equivalent trauma centre.Participants 8220 adult trauma patients screened for opportunities for improvement between 2013 and 2022.Primary and secondary outcome measures Two machine learning models (logistic regression and XGBoost) and the currently used audit filters were compared. Internal validation by an expanding window approach with annual updates was used for model evaluation. Performance measured by discrimination, calibration, sensitivity and false positive rate of opportunities for improvement prediction.Results A total of 8220 patients, with a mean age of 45 years, were analysed; 69% were men with a mean injury severity score of 12. Opportunities for improvement were identified in 496 (6%) patients. Both the logistic regression and XGBoost models were well-calibrated, with intercalibration indices of 0.02 and 0.02, respectively. The models demonstrated higher areas under the receiver operating characteristic curve (AUCs) (logistic regression: 0.71; XGBoost: 0.74). The XGBoost model had a lower false positive rate at a similar sensitivity (false positive rate: 0.63). The audit filters had an AUC of 0.62 and a false positive rate of 0.67.Conclusions The logistic regression and XGBoost models outperformed audit filters in predicting opportunities for improvement among adult trauma patients and can potentially be used to improve systems for selecting patients for trauma peer review.https://bmjopen.bmj.com/content/15/6/e099624.full |
| spellingShingle | Martin Gerdin Wärnberg Lovisa Strömmer Jonatan Attergrim Kelvin Szolnoky Olof Brattström Martin Jacobsson Gunilla Wihlke Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre BMJ Open |
| title | Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre |
| title_full | Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre |
| title_fullStr | Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre |
| title_full_unstemmed | Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre |
| title_short | Predicting opportunities for improvement in trauma care using machine learning: a retrospective registry-based study at a major trauma centre |
| title_sort | predicting opportunities for improvement in trauma care using machine learning a retrospective registry based study at a major trauma centre |
| url | https://bmjopen.bmj.com/content/15/6/e099624.full |
| work_keys_str_mv | AT martingerdinwarnberg predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT lovisastrommer predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT jonatanattergrim predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT kelvinszolnoky predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT olofbrattstrom predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT martinjacobsson predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre AT gunillawihlke predictingopportunitiesforimprovementintraumacareusingmachinelearningaretrospectiveregistrybasedstudyatamajortraumacentre |