Accepted for/Published in: JMIR Bioinformatics and Biotechnology
Date Submitted: Mar 28, 2022
Open Peer Review Period: Mar 28, 2022 - May 23, 2022
Date Accepted: Sep 2, 2022
(closed for review but you can still tweet)
Treatment-discontinuation Prediction in Diabetic Patients Using a Ranking Model: Machine-Learning Model Development
ABSTRACT
Background:
Treatment discontinuation, one of the major prognostic issues in diabetes care, could be predicted using a combination of machine-learned techniques and electronic medical records.
Objective:
To generate a risk prediction model for treatment discontinuation in patients with diabetes based on machine-learned techniques.
Methods:
This model included patients with diagnostic codes indicative of diabetes at the University of Tokyo Hospital between September 3, 2012, and May 17, 2014. They were internally validated with patients from the same hospital from May 18, 2014, to January 29, 2016. The data used in this study included 7,551 patients who visited the hospital after January 1, 2004, and had diagnostic codes indicative of diabetes. Data recorded in the electronic medical records between September 3, 2012, and January 29, 2016, were used in this study. The main outcome was treatment discontinuation, which was defined as missing a scheduled clinical appointment and having no hospital visits within less than three times the average visit interval or 60 days. The treatment discontinuation risk score was calculated using parameters derived from the machine-learned ranking algorithm. The prediction capacity was evaluated using test data with C-index for performance of ranking patients, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve (AUPRC) for discrimination, in addition to calibration plot.
Results:
The treatment discontinuation risk score showed a C-index value of 0.749 (95% CI 0.655 to 0.823), AUROC of 0.758 (95% CI 0.649 to 0.857), and AUPRC of 0.713 (95% CI 0.554 to 0.841). The observed and predicted probabilities were correlated with the calibration plots.
Conclusions:
In this study, a treatment discontinuation risk score was developed for patients with diabetes by combining a machine-learned method with electronic medical records. The score calculation can be integrated into the medical records to identify patients at high risk, which can be useful in supporting diabetes care and preventing treatment discontinuation.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.