Improving quality care in sepsis through machine learning models

Machine learning models, such as Random Forest, may be used to improve risk adjustment in patients with sepsis. Random Forest outperformed logistic regression analysis in providing a more accurate prediction of 90-day mortality in sepsis patients.  

A study by Dr Elizabeth Munroe (University of Michigan, MI, USA) aimed to assess risk-adjustment mortality models in patients with sepsis. “Traditional models do not fully incorporate acute physiology,” she outlined [1]. “Perhaps new machine learning models can do this better.” The authors used data from the HMS-sepsis registry, including 5,303 cases of sepsis hospitalisation in 31 hospitals across Michigan. The 90-day mortality rate of the sample was 27.0%. Several machine learning models were compared with a more traditional stepwise logistic regression model for the primary outcome of 90-day mortality. The models included patient characteristics, comorbidities, and parameters of acute physiology.  

The stepwise logistic regression model had an area under the curve (AUC) of 0.77, whereas the best-performing machine learning model, Random Forest, had an AUC of 0.90. Dr Munroe explained that a disadvantage of the traditional model is its limited ability to assess interactions. In contrast, the more complex Random Forest model accounts for interactions. Furthermore, Radom Forest identified additional variables of importance for mortality in sepsis, such as creatinine and bilirubin levels, functional limitations, and dementia. 

“Risk adjustment is important in sepsis quality care improvement,” according to Dr Munroe. “This study showed that machine learning models may help to improve risk adjustment in this population. A next step may be to use the variables that were identified by machine learning models to improve our traditional model.” 

1. Munroe E, et al. Machine learning methods to address confounding in sepsis mortality rate. C94, ATS International Conference 2023, 19–24 May, Washington DC, USA. 

 

Copyright ©2023 Medicom Medical Publishers



Posted on 12 June, 202313 June, 2023