Skip to main content

Table 4 Proportion missed, workload savings, and estimated time savings for the dual independent screening simulations

From: The semi-automation of title and abstract screening: a retrospective exploration of ways to leverage Abstrackr’s relevance predictions in systematic and rapid reviews

Systematic review name

Proportion missed, single reviewer, n (%) a

Dual independent screening, fully automated simulation

Dual independent screening, semi-automated simulation

Proportion missed, n (%)

Workload savings, n (%)

Time savings, hours (days)

Proportion missed, n (%)

Workload savings, n (%)

Time savings, hours (days)

Biomarkers

1 (2)

0 (0)

1612 (47)

13 (2)

6 (13)

1109 (32)

9 (1)

Brain injury

2 (5)

0 (0)

6062 (48)

51 (6)

0 (0)

3936 (31)

33 (4)

Activity and pregnancy

11 (11)

1 (1)

2728 (47)

23 (3)

1 (1)

2409 (41)

20 (3)

Concussion

0 (0)

0 (0)

1239 (43)

10 (1)

0 (0)

601 (21)

5 (< 1)

Antipsychotics

4 (3)

2 (2)

11,956 (49)

100 (12)

3 (2)

9839 (40)

82 (10)

Digital technologies for pain

0 (0)

0 (0)

2462 (46)

21 (3)

0 (0)

2141 (40)

18 (2)

Treatments for bronchiolitis

10 (7)

1 (1)

5661 (48)

47 (6)

1 (1)

5005 (43)

42 (5)

VBAC

5 (24)

3 (14)

4892 (48)

41 (5)

3 (14)

3402 (33)

28 (4)

Visual acuity

0 (0)

0 (0)

11,029 (49)

92 (11)

0 (0)

7390 (33)

62 (8)

Experience of bronchiolitis

12 (43)

1 (1)

451 (35)

4 (< 1)

1 (1)

340 (26)

3 (< 1)

Experiences of UTIs

0 (0)

0 (0)

1293 (43)

11 (1)

0 (0)

447 (15)

4 (< 1)

  1. UTI urinary tract infection; VBAC vaginal birth after caesarean section
  2. aProportion missed (retrospective) had the screening been completed by the second reviewer in isolation