Skip to main content

Table 2 Findings from the user study on the perceptions (expressed in terms of CoIEDs and RPs) by doctors and laypeople of illness severity levels. Significance levels are computed through the Mann-Whitney U test

From: Ordinal labels in machine learning: a user-centered approach to improve data validity in medical settings

Level

  

Doctor

Patient

Diff

 

Doctor

Patient

RP

RP

Doctor

 

CoIED

CoIED

median

median

vs

 

95% CI

95% CI

95% CI

95% CI

Patient

Absent

[4.74, 13.7]

[12.9, 14.6]

[4.5, 8.0]

[0.0, 0.0]

***

Mild

[22.2, 29.6]

[25.8, 27.3]

[20.5, 30.0]

[20.0, 20.0]

NS

Moderate

[42.6, 50.7]

[40.4, 42.2]

[45.0, 49.5]

[50.0, 50.0]

NS

Severe

[63.5, 71.5]

[56.83, 59.15]

[67.0, 75.5]

[75.0, 75.0]

***

Extreme

[83.5, 92.5]

[69.12, 72.51]

[90.5, 94.5]

[99.0, 100.0]

***