Please use this identifier to cite or link to this item: https://www.arca.fiocruz.br/handle/icict/67754

Type
Article
Copyright
Open access
Collections
Share

Statistics
Metadata
Show full item record

COMPARING CAUSAL RANDOM FOREST AND LINEAR REGRESSION TO ESTIMATE THE INDEPENDENT ASSOCIATION OF ORGANISATIONAL FACTORS WITH ICU EFFICIENCY
Bastos, Leonardo S. L. et al. | Date Issued: 2024
Author
Bastos, Leonardo S. L.
Wortel, Safira A.
Bakhshi-Raiez, Ferishta
Abu-Hanna, Ameen
Dongelmans, Dave A.
Salluh, Jorge I. F.
Zampieri, Fernando G.
Burghi, Gastón
Hamacher, Silvio
Bozza, Fernando Augusto
Keizer, Nicolette F. de
Soares, Marcio
Affilliation
Pontifical Catholic University of Rio de Janeiro. Department of Industrial Engineering. Rio de Janeiro, RJ, Brazil.
Amsterdam UMC Location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / National Intensive Care Evaluation Foundation. Amsterdam UMC location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / Amsterdam Public Health. Digital Health and Methodology. Amsterdam, The Netherlands.
Amsterdam UMC Location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / National Intensive Care Evaluation Foundation. Amsterdam UMC location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / Amsterdam Public Health. Digital Health and Methodology. Amsterdam, The Netherlands.
Amsterdam UMC Location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / Amsterdam Public Health. Digital Health and Methodology. Amsterdam, The Netherlands.
National Intensive Care Evaluation Foundation. Amsterdam UMC Location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / Amsterdam Public Health. Digital Health and Methodology. Amsterdam, The Netherlands / Department of Intensive Care. Amsterdam UMC. Amsterdam, The Netherlands.
D'Or Institute for Research and Education. Rio de Janeiro, RJ, Brazil / Brazilian Research in Intensive Care Network. São Paulo, SP, Brazil / Program Federal University of Rio de Janeiro. Internal Medicine. PostGraduate. Rio de Janeiro, RJ, Brazil.
D'Or Institute for Research and Education. Rio de Janeiro, RJ, Brazil / Brazilian Research in Intensive Care Network. São Paulo, SP, Brazil.
Hospital Maciel. Intensive Care Unit. Montevideo, Uruguay.
Pontifical Catholic University of Rio de Janeiro. Department of Industrial Engineerin. Rio de Janeiro, RJ, Brazil.
D'Or Institute for Research and Education. Rio de Janeiro, RJ, Brazil / Brazilian Research in Intensive Care Network. São Paulo, SP, Brazil / Fundação Oswaldo Cruz. Instituto Nacional de Infectologia Evandro Chagas. Laboratório de Pesquisa Clínica em Medicina Intensiva. Rio de Janeiro, RJ, Brasil.
Amsterdam UMC Location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / National Intensive Care Evaluation Foundation. Amsterdam UMC location University of Amsterdam. Department of Medical Informatics. Amsterdam, The Netherlands / Amsterdam Public Health. Digital Health and Methodology. Amsterdam, The Netherlands.
D'Or Institute for Research and Education. Rio de Janeiro, RJ, Brazil / Brazilian Research in Intensive Care Network. São Paulo, SP, Brazil.
Abstract
Purpose: Parametric regression models have been the main statistical method for identifying average treatment effects. Causal machine learning models showed promising results in estimating heterogeneous treatment effects in causal inference. Here we aimed to compare the application of causal random forest (CRF) and linear regression modelling (LRM) to estimate the effects of organisational factors on ICU efficiency. Methods: A retrospective analysis of 277,459 patients admitted to 128 Brazilian and Uruguayan ICUs over three years. ICU efficiency was assessed using the average standardised efficiency ratio (ASER), measured as the average of the standardised mortality ratio (SMR) and the standardised resource use (SRU) according to the SAPS-3 score. Using a causal inference framework, we estimated and compared the conditional average treatment effect (CATE) of seven common structural and organisational factors on ICU efficiency using LRM with interaction terms and CRF. Results: The hospital mortality was 14 %; median ICU and hospital lengths of stay were 2 and 7 days, respectively. Overall median SMR was 0.97 [IQR: 0.76,1.21], median SRU was 1.06 [IQR: 0.79,1.30] and median ASER was 0.99 [IQR: 0.82,1.21]. Both CRF and LRM showed that the average number of nurses per ten beds was independently associated with ICU efficiency (CATE [95 %CI]: -0.13 [-0.24, -0.01] and -0.09 [-0.17,-0.01], respectively). Finally, CRF identified some specific ICUs with a significant CATE in exposures that did not present a significant average effect. Conclusion: In general, both methods were comparable to identify organisational factors significantly associated with CATE on ICU efficiency. CRF however identified specific ICUs with significant effects, even when the average effect was nonsignificant. This can assist healthcare managers in further in-dept evaluation of process interventions to improve ICU efficiency.
Keywords
Intensive care unit
Organisational characteristics
Causal random forest
Efficiency
Causal inference
 
Publisher
Elsevier
Citation
BASTOS, Leonardo S. L. et al. Comparing causal random forest and linear regression to estimate the independent association of organisational factors with ICU efficiency. International Journal of Medical Informatics, v. 191, p. 1-7, Nov. 2024.
DOI
10.1016/j.ijmedinf.2024.105568
ISSN
1386-5056