Machine Learning-Based Early Warning Systems for Acute Care Utilization During Systemic Therapy for Cancer
BACKGROUND: Emergency department visits and hospitalizations frequently occur during systemic therapy for cancer. We developed and evaluated a longitudinal warning system for acute care use.
METHODS: Using a retrospective population-based cohort of patients who started intravenous systemic therapy for nonhematologic cancers between July 1, 2014, and June 30, 2020, we randomly separated patients into cohorts for model training, hyperparameter tuning and model selection, and system testing. Predictive features included static features, such as demographics, cancer type, and treatment regimens, and dynamic features, such as patient-reported symptoms and laboratory values. The longitudinal warning system predicted the probability of acute care utilization within 30 days after each treatment session. Machine learning systems were developed in the training and tuning cohorts and evaluated in the testing cohort. Sensitivity analyses considered feature importance, other acute care endpoints, and performance within subgroups.
RESULTS: The cohort included 105,129 patients who received 1,216,385 treatment sessions. Acute care followed 182,444 (15.0%) treatments within 30 days. The ensemble model achieved an area under the receiver operating characteristic curve of 0.742 (95% CI, 0.739-0.745) and was well calibrated in the test cohort. Important predictive features included prior acute care use, treatment regimen, and laboratory tests. If the system was set to alarm approximately once every 15 treatments, 25.5% of acute care events would be preceded by an alarm, and 47.4% of patients would experience acute care after an alarm. The system underestimated risk for some treatment regimens and potentially underserved populations such as females and non-English speakers.
CONCLUSIONS: Machine learning warning systems can detect patients at risk for acute care utilization, which can aid in preventive intervention and facilitate tailored treatment. Future research should address potential biases and prospectively evaluate impact after system deployment.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2023 |
---|---|
Erschienen: |
2023 |
Enthalten in: |
Zur Gesamtaufnahme - volume:21 |
---|---|
Enthalten in: |
Journal of the National Comprehensive Cancer Network : JNCCN - 21(2023), 10 vom: 01. Okt., Seite 1029-1037.e21 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Grant, Robert C [VerfasserIn] |
---|
Links: |
---|
Themen: |
Acute care |
---|
Anmerkungen: |
Date Completed 23.10.2023 Date Revised 30.10.2023 published: Print Citation Status MEDLINE |
---|
doi: |
10.6004/jnccn.2023.7046 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM363487026 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM363487026 | ||
003 | DE-627 | ||
005 | 20231226093443.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231226s2023 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.6004/jnccn.2023.7046 |2 doi | |
028 | 5 | 2 | |a pubmed24n1211.xml |
035 | |a (DE-627)NLM363487026 | ||
035 | |a (NLM)37856226 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Grant, Robert C |e verfasserin |4 aut | |
245 | 1 | 0 | |a Machine Learning-Based Early Warning Systems for Acute Care Utilization During Systemic Therapy for Cancer |
264 | 1 | |c 2023 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 23.10.2023 | ||
500 | |a Date Revised 30.10.2023 | ||
500 | |a published: Print | ||
500 | |a Citation Status MEDLINE | ||
520 | |a BACKGROUND: Emergency department visits and hospitalizations frequently occur during systemic therapy for cancer. We developed and evaluated a longitudinal warning system for acute care use | ||
520 | |a METHODS: Using a retrospective population-based cohort of patients who started intravenous systemic therapy for nonhematologic cancers between July 1, 2014, and June 30, 2020, we randomly separated patients into cohorts for model training, hyperparameter tuning and model selection, and system testing. Predictive features included static features, such as demographics, cancer type, and treatment regimens, and dynamic features, such as patient-reported symptoms and laboratory values. The longitudinal warning system predicted the probability of acute care utilization within 30 days after each treatment session. Machine learning systems were developed in the training and tuning cohorts and evaluated in the testing cohort. Sensitivity analyses considered feature importance, other acute care endpoints, and performance within subgroups | ||
520 | |a RESULTS: The cohort included 105,129 patients who received 1,216,385 treatment sessions. Acute care followed 182,444 (15.0%) treatments within 30 days. The ensemble model achieved an area under the receiver operating characteristic curve of 0.742 (95% CI, 0.739-0.745) and was well calibrated in the test cohort. Important predictive features included prior acute care use, treatment regimen, and laboratory tests. If the system was set to alarm approximately once every 15 treatments, 25.5% of acute care events would be preceded by an alarm, and 47.4% of patients would experience acute care after an alarm. The system underestimated risk for some treatment regimens and potentially underserved populations such as females and non-English speakers | ||
520 | |a CONCLUSIONS: Machine learning warning systems can detect patients at risk for acute care utilization, which can aid in preventive intervention and facilitate tailored treatment. Future research should address potential biases and prospectively evaluate impact after system deployment | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
650 | 4 | |a Machine Learning | |
650 | 4 | |a acute care | |
650 | 4 | |a emergency department visits | |
650 | 4 | |a warning systems | |
700 | 1 | |a He, Jiang Chen |e verfasserin |4 aut | |
700 | 1 | |a Khan, Ferhana |e verfasserin |4 aut | |
700 | 1 | |a Liu, Ning |e verfasserin |4 aut | |
700 | 1 | |a Podolsky, Sho |e verfasserin |4 aut | |
700 | 1 | |a Kaliwal, Yosuf |e verfasserin |4 aut | |
700 | 1 | |a Powis, Melanie |e verfasserin |4 aut | |
700 | 1 | |a Notta, Faiyaz |e verfasserin |4 aut | |
700 | 1 | |a Chan, Kelvin K W |e verfasserin |4 aut | |
700 | 1 | |a Ghassemi, Marzyeh |e verfasserin |4 aut | |
700 | 1 | |a Gallinger, Steven |e verfasserin |4 aut | |
700 | 1 | |a Krzyzanowska, Monika K |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Journal of the National Comprehensive Cancer Network : JNCCN |d 2003 |g 21(2023), 10 vom: 01. Okt., Seite 1029-1037.e21 |w (DE-627)NLM156397676 |x 1540-1413 |7 nnns |
773 | 1 | 8 | |g volume:21 |g year:2023 |g number:10 |g day:01 |g month:10 |g pages:1029-1037.e21 |
856 | 4 | 0 | |u http://dx.doi.org/10.6004/jnccn.2023.7046 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 21 |j 2023 |e 10 |b 01 |c 10 |h 1029-1037.e21 |