Skip to main content

Table 4 Relevance of the extracted solutions

From: Creating efficiencies in the extraction of data from randomized trials: a prospective evaluation of a machine learning and text mining tool

Report Section

Data Element

Fully Relevant Solutions, n (%)a

Partially Relevant Solutions, n (%)a

Fully Irrelevant Solutions, n (%)a

Publication information

First author name

63 (84.0)

0 (0.0)

12 (16.0)

Date of publication

64 (85.3)

0 (0.0)

11 (14.7)

Digital object identifier

62 (82.7)

0 (0.0)

13 (17.3)

Meta information

Funding source

16 (21.3)

38 (50.7)

21 (28.0)

Funding number

52 (69.3)

13 (17.3)

10 (13.3)

Registration number

62 (82.7)

5 (6.7)

8 (10.7)

Enrollment

Eligibility criteria

0 (0.0)

66 (88.0)

9 (12.0)

Sample size

31 (41.3)

38 (50.7)

6 (8.0)

Enrollment start date

34 (45.3)

15 (20.0)

26 (34.7)

Enrollment end date

37 (49.3)

13 (17.3)

25 (33.3)

Early stopping

70 (93.3)

3 (4.0)

2 (2.7)

Intervention

Experimental arm(s)

10 (13.3)

56 (74.7)

9 (12.0)

Control arm(s)

16 (21.3)

49 (65.3)

10 (13.3)

Route of administration

53 (70.7)

12 (16.0)

10 (13.3)

Dose

36 (48.0)

22 (29.3)

17 (22.7)

Frequency of administration

39 (52.0)

14 (18.7)

22 (29.3)

Duration of treatment

22 (29.3)

25 (33.3)

28 (37.3)

Outcome

Primary outcome(s)

38 (50.7)

24 (32.0)

13 (17.3)

Primary outcome time point

7 (9.3)

42 (56.0)

26 (34.7)

Secondary outcome(s)

23 (30.7)

31 (41.3)

21 (28.0)

Secondary outcome time point

15 (20.0)

27 (36.0)

33 (44.0)

Summary measure

Median (IQR), n

36 (16 to 53)

22 (12 to 38)

13 (10 to 22)

Median (IQR), %

48.0 (21.3 to 70.7)

29.3 (16.0 to 50.7)

17.3 (13.3 to 29.3)

  1. Values in italics typeface fall at or below the limit of the lowest quartile
  2. aOut of a total 75 solutions per data element (i.e., one solution per data element, per trial). Partially correct solutions were those that included relevant information but either also included erroneous information or fell short of including all essential details