Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation
This study aims to develop and validate the use of machine learning-based prediction models to select individualized pharmacological treatment for patients with depressive disorder. This study used data from Taiwan’s National Health Insurance Research Database. Patients with incident depressive diso...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Journal of Personalized Medicine |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4426/11/12/1316 |
| _version_ | 1797503093509718016 |
|---|---|
| author | Chi-Shin Wu Albert C. Yang Shu-Sen Chang Chia-Ming Chang Yi-Hung Liu Shih-Cheng Liao Hui-Ju Tsai |
| author_facet | Chi-Shin Wu Albert C. Yang Shu-Sen Chang Chia-Ming Chang Yi-Hung Liu Shih-Cheng Liao Hui-Ju Tsai |
| author_sort | Chi-Shin Wu |
| collection | DOAJ |
| description | This study aims to develop and validate the use of machine learning-based prediction models to select individualized pharmacological treatment for patients with depressive disorder. This study used data from Taiwan’s National Health Insurance Research Database. Patients with incident depressive disorders were included in this study. The study outcome was treatment failure, which was defined as psychiatric hospitalization, self-harm hospitalization, emergency visits, or treatment change. Prediction models based on the Super Learner ensemble were trained separately for the initial and the next-step treatments if the previous treatments failed. An individualized treatment strategy was developed for selecting the drug with the lowest probability of treatment failure for each patient as the model-selected regimen. We emulated clinical trials to estimate the effectiveness of individualized treatments. The area under the curve of the prediction model using Super Learner was 0.627 and 0.751 for the initial treatment and the next-step treatment, respectively. Model-selected regimens were associated with reduced treatment failure rates, with a 0.84-fold (95% confidence interval (CI) 0.82–0.86) decrease for the initial treatment and a 0.82-fold (95% CI 0.80–0.83) decrease for the next-step. In emulation of clinical trials, the model-selected regimen was associated with a reduced treatment failure rate. |
| first_indexed | 2024-03-10T03:45:35Z |
| format | Article |
| id | doaj.art-18fb6058ebcf4a67bf70842d3881da43 |
| institution | Directory Open Access Journal |
| issn | 2075-4426 |
| language | English |
| last_indexed | 2024-03-10T03:45:35Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Personalized Medicine |
| spelling | doaj.art-18fb6058ebcf4a67bf70842d3881da432023-11-23T09:07:50ZengMDPI AGJournal of Personalized Medicine2075-44262021-12-011112131610.3390/jpm11121316Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial EmulationChi-Shin Wu0Albert C. Yang1Shu-Sen Chang2Chia-Ming Chang3Yi-Hung Liu4Shih-Cheng Liao5Hui-Ju Tsai6National Centre for Geriatrics and Welfare Research, National Health Research Institutes, Zhunan 350, TaiwanDigital Medicine Center, Institute of Brain Science, National Yang-Ming Chiao-Tung University, Taipei 112, TaiwanInstitute of Health Behaviours and Community Sciences, College of Public Health, National Taiwan University, Taipei 112, TaiwanDepartment of Psychiatry, Chang Gung Memorial Hospital, Linkou and Chang Gung University, Taoyuan 333, TaiwanDepartment of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei 106, TaiwanDepartment of Psychiatry, College of Medicine, National Taiwan University Hospital, National Taiwan University, Taipei 100, TaiwanInstitute of Population Health Sciences, National Health Research Institutes, Zhunan 350, TaiwanThis study aims to develop and validate the use of machine learning-based prediction models to select individualized pharmacological treatment for patients with depressive disorder. This study used data from Taiwan’s National Health Insurance Research Database. Patients with incident depressive disorders were included in this study. The study outcome was treatment failure, which was defined as psychiatric hospitalization, self-harm hospitalization, emergency visits, or treatment change. Prediction models based on the Super Learner ensemble were trained separately for the initial and the next-step treatments if the previous treatments failed. An individualized treatment strategy was developed for selecting the drug with the lowest probability of treatment failure for each patient as the model-selected regimen. We emulated clinical trials to estimate the effectiveness of individualized treatments. The area under the curve of the prediction model using Super Learner was 0.627 and 0.751 for the initial treatment and the next-step treatment, respectively. Model-selected regimens were associated with reduced treatment failure rates, with a 0.84-fold (95% confidence interval (CI) 0.82–0.86) decrease for the initial treatment and a 0.82-fold (95% CI 0.80–0.83) decrease for the next-step. In emulation of clinical trials, the model-selected regimen was associated with a reduced treatment failure rate.https://www.mdpi.com/2075-4426/11/12/1316anti-depressive agentsmachine learningprecision medicine |
| spellingShingle | Chi-Shin Wu Albert C. Yang Shu-Sen Chang Chia-Ming Chang Yi-Hung Liu Shih-Cheng Liao Hui-Ju Tsai Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation Journal of Personalized Medicine anti-depressive agents machine learning precision medicine |
| title | Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation |
| title_full | Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation |
| title_fullStr | Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation |
| title_full_unstemmed | Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation |
| title_short | Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation |
| title_sort | validation of machine learning based individualized treatment for depressive disorder using target trial emulation |
| topic | anti-depressive agents machine learning precision medicine |
| url | https://www.mdpi.com/2075-4426/11/12/1316 |
| work_keys_str_mv | AT chishinwu validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT albertcyang validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT shusenchang validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT chiamingchang validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT yihungliu validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT shihchengliao validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation AT huijutsai validationofmachinelearningbasedindividualizedtreatmentfordepressivedisorderusingtargettrialemulation |