Tags

Type your tag names separated by a space and hit enter

Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records.
Lancet Digit Health. 2019 06; 1(2):e78-e89.LD

Abstract

BACKGROUND

Intensive-care units (ICUs) treat the most critically ill patients, which is complicated by the heterogeneity of the diseases that they encounter. Severity scores based mainly on acute physiology measures collected at ICU admission are used to predict mortality, but are non-specific, and predictions for individual patients can be inaccurate. We investigated whether inclusion of long-term disease history before ICU admission improves mortality predictions.

METHODS

Registry data for long-term disease histories for more than 230 000 Danish ICU patients were used in a neural network to develop an ICU mortality prediction model. Long-term disease histories and acute physiology measures were aggregated to predict mortality risk for patients for whom both registry and ICU electronic patient record data were available. We compared mortality predictions with admission scores on the Simplified Acute Physiology Score (SAPS) II, the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II, and the best available multimorbidity score, the Multimorbidity Index. An external validation set from an additional hospital was acquired after model construction to confirm the validity of our model. During initial model development data were split into a training set (85%) and an independent test set (15%), and a five-fold cross-validation was done during training to avoid overfitting. Neural networks were trained for datasets with disease history of 1 month, 3 months, 6 months, 1 year, 2·5 years, 5 years, 7·5 years, 10 years, and 23 years before ICU admission.

FINDINGS

Mortality predictions with a model based solely on disease history outperformed the Multimorbidity Index (Matthews correlation coefficient 0·265 vs 0·065), and performed similarly to SAPS II and APACHE II (Matthews correlation coefficient with disease history, age, and sex 0·326 vs 0·347 and 0·300 for SAPS II and APACHE II, respectively). Diagnoses up to 10 years before ICU admission affected current mortality prediction. Aggregation of previous disease history and acute physiology measures in a neural network yielded the most precise predictions of in-hospital mortality (Matthews correlation coefficient 0·391 for in-hospital mortality compared with 0·347 with SAPS II and 0·300 with APACHE II). These results for the aggregated model were validated in an external independent dataset of 1528 patients (Matthews correlation coefficient for prediction of in-hospital mortality 0·341).

INTERPRETATION

Longitudinal disease-spectrum-wide data available before ICU admission are useful for mortality prediction. Disease history can be used to differentiate mortality risk between patients with similar vital signs with more precision than SAPS II and APACHE II scores. Machine learning models can be deconvoluted to generate novel understandings of how ICU patient features from long-term and short-term events interact with each other. Explainable machine learning models are key in clinical settings, and our results emphasise how to progress towards the transformation of advanced models into actionable, transparent, and trustworthy clinical tools.

FUNDING

Novo Nordisk Foundation and Innovation Fund Denmark.

Authors+Show Affiliations

Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; Department of Intensive Care, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.Centre for IT, Medical Technology and Telephony Services, Capital Region of Denmark, Copenhagen, Denmark.Daintel, Lyngby, Denmark.Daintel, Lyngby, Denmark.Daintel, Lyngby, Denmark.Department of Intensive Care, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark.Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. Electronic address: soren.brunak@cpr.ku.dk.

Pub Type(s)

Journal Article
Research Support, Non-U.S. Gov't

Language

eng

PubMed ID

33323232

Citation

Nielsen, Annelaura B., et al. "Survival Prediction in Intensive-care Units Based On Aggregation of Long-term Disease History and Acute Physiology: a Retrospective Study of the Danish National Patient Registry and Electronic Patient Records." The Lancet. Digital Health, vol. 1, no. 2, 2019, pp. e78-e89.
Nielsen AB, Thorsen-Meyer HC, Belling K, et al. Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records. Lancet Digit Health. 2019;1(2):e78-e89.
Nielsen, A. B., Thorsen-Meyer, H. C., Belling, K., Nielsen, A. P., Thomas, C. E., Chmura, P. J., Lademann, M., Moseley, P. L., Heimann, M., Dybdahl, L., Spangsege, L., Hulsen, P., Perner, A., & Brunak, S. (2019). Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records. The Lancet. Digital Health, 1(2), e78-e89. https://doi.org/10.1016/S2589-7500(19)30024-X
Nielsen AB, et al. Survival Prediction in Intensive-care Units Based On Aggregation of Long-term Disease History and Acute Physiology: a Retrospective Study of the Danish National Patient Registry and Electronic Patient Records. Lancet Digit Health. 2019;1(2):e78-e89. PubMed PMID: 33323232.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records. AU - Nielsen,Annelaura B, AU - Thorsen-Meyer,Hans-Christian, AU - Belling,Kirstine, AU - Nielsen,Anna P, AU - Thomas,Cecilia E, AU - Chmura,Piotr J, AU - Lademann,Mette, AU - Moseley,Pope L, AU - Heimann,Marc, AU - Dybdahl,Lars, AU - Spangsege,Lasse, AU - Hulsen,Patrick, AU - Perner,Anders, AU - Brunak,Søren, Y1 - 2019/05/23/ PY - 2019/02/26/received PY - 2019/04/12/revised PY - 2019/04/15/accepted PY - 2020/12/16/entrez PY - 2019/6/1/pubmed PY - 2019/6/1/medline SP - e78 EP - e89 JF - The Lancet. Digital health JO - Lancet Digit Health VL - 1 IS - 2 N2 - BACKGROUND: Intensive-care units (ICUs) treat the most critically ill patients, which is complicated by the heterogeneity of the diseases that they encounter. Severity scores based mainly on acute physiology measures collected at ICU admission are used to predict mortality, but are non-specific, and predictions for individual patients can be inaccurate. We investigated whether inclusion of long-term disease history before ICU admission improves mortality predictions. METHODS: Registry data for long-term disease histories for more than 230 000 Danish ICU patients were used in a neural network to develop an ICU mortality prediction model. Long-term disease histories and acute physiology measures were aggregated to predict mortality risk for patients for whom both registry and ICU electronic patient record data were available. We compared mortality predictions with admission scores on the Simplified Acute Physiology Score (SAPS) II, the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II, and the best available multimorbidity score, the Multimorbidity Index. An external validation set from an additional hospital was acquired after model construction to confirm the validity of our model. During initial model development data were split into a training set (85%) and an independent test set (15%), and a five-fold cross-validation was done during training to avoid overfitting. Neural networks were trained for datasets with disease history of 1 month, 3 months, 6 months, 1 year, 2·5 years, 5 years, 7·5 years, 10 years, and 23 years before ICU admission. FINDINGS: Mortality predictions with a model based solely on disease history outperformed the Multimorbidity Index (Matthews correlation coefficient 0·265 vs 0·065), and performed similarly to SAPS II and APACHE II (Matthews correlation coefficient with disease history, age, and sex 0·326 vs 0·347 and 0·300 for SAPS II and APACHE II, respectively). Diagnoses up to 10 years before ICU admission affected current mortality prediction. Aggregation of previous disease history and acute physiology measures in a neural network yielded the most precise predictions of in-hospital mortality (Matthews correlation coefficient 0·391 for in-hospital mortality compared with 0·347 with SAPS II and 0·300 with APACHE II). These results for the aggregated model were validated in an external independent dataset of 1528 patients (Matthews correlation coefficient for prediction of in-hospital mortality 0·341). INTERPRETATION: Longitudinal disease-spectrum-wide data available before ICU admission are useful for mortality prediction. Disease history can be used to differentiate mortality risk between patients with similar vital signs with more precision than SAPS II and APACHE II scores. Machine learning models can be deconvoluted to generate novel understandings of how ICU patient features from long-term and short-term events interact with each other. Explainable machine learning models are key in clinical settings, and our results emphasise how to progress towards the transformation of advanced models into actionable, transparent, and trustworthy clinical tools. FUNDING: Novo Nordisk Foundation and Innovation Fund Denmark. SN - 2589-7500 UR - https://www.unboundmedicine.com/medline/citation/33323232/Survival_prediction_in_intensive_care_units_based_on_aggregation_of_long_term_disease_history_and_acute_physiology:_a_retrospective_study_of_the_Danish_National_Patient_Registry_and_electronic_patient_records_ L2 - https://linkinghub.elsevier.com/retrieve/pii/S2589-7500(19)30024-X DB - PRIME DP - Unbound Medicine ER -