Skip to main content

Table 5 Classification results and contingency table for logistic regression model based on clinical data and fall risk assessment tests

From: Sensors vs. experts - A performance comparison of sensor-based fall risk assessment vs. conventional assessment in a sample of geriatric patients

model CONV
classification accuracy 72%     
sensitivity 68%   contingency table  
specificity 74%   fall within one year  
negative predictive value 77%   yes no Sum
positive predictive value 65% pred. yes 13 7 20
Brier score 0.20 pred. no 6 20 26
AUC 0.74 sum 19 27 46