Skip to main content

Table 3 Confusion matrix of the interpreted risk by patients versus the gold standard (nephrologists judgement) and performance (i.e. precision and recall) for clinical scenarios where an action was needed (i.e. at least medium risk) and those where no action was needed (i.e. low risk)

From: Presentation of laboratory test results in patient portals: influence of interface design on risk interpretation and visual search behaviour

Presentation

Interpreted risk by participants

Clinical scenarios (gold standard)

Performance

Action needed

No action needed

Precision

Recall

Baseline

Action needed

28

6

0.82

0.70

No action needed

12

14

0.54

0.70

Contextualised

Action needed

29

7

0.81

0.72

No action needed

11

13

0.54

0.65

Grouped

Action needed

28

9

0.76

0.70

No action needed

12

11

0.48

0.55