Skip to main content

Table 4 Distribution of erroneous outcomes, classified by type of clinical problem

From: A rule-based electronic phenotyping algorithm for detecting clinically relevant cardiovascular disease cases

Results wrong depending on clinical problem n (%)

Proportion

Of the total number of errors (n = 98)

The fn (n = 32)

The FP (n = 66)

Errors

CaV (n = 62)

CeV (n = 36)

Peripheral vascular disease

18 (18.37)

10 (31.25)

8 (12.12)

18 (30)

NA

Heart failure

11 (11.22)

4 (12.5)

7 (10.61)

11 (18.33)

Acute myocardial infarction

11 (11.22)

5 (15.63)

6 (9.09)

11 (18.33)

Cardiovascular symptomsa

22 (22.45)

4 (12.5)

18 (27.27)

22 (36.67)

Stroke/CeVD

29 (29.6)

2 (6.25)

27 (40.91)

NA

29 (81.58)

Transient ischemic attack

7 (7.14)

7 (21.87)

0 (0)

7 (18.42)

  1. NA not applicable
  2. aChest pain, angina pectoris, chest oppression