Table 4.
Diagnostic properties of the validation studies
| Study | Screening test | Aphasia (n) | No aphasia (n) | Aphasia correctly classified (n) | No aphasia correctly classified (n) | Sensitivity (%) | Specificity (%) | Non-verified (n) | LR+ | LR− | DOR (95 % CI) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Al-Khawaja 1996 [13] | FAST | 45 | 5 | 39 | 4 | 87 | 80 | 0 | 4.4 | 0.16 | 27.5 (2.6–289.5) |
| Enderby 1987 [14] | FAST, full | 20 | 30 | 20 | 23 | 100 | 77 | 0 | 4.4 | 0.00 | ∞ |
| Enderby 1987 [14] | FAST, short | 20 | 30 | 20 | 27 | 100 | 90 | 0 | 10 | 0.00 | ∞ |
| O’Neill 1990 [15] | FAST | 23 | 31 | 22 | 19 | 96 | 61 | 0 | 2.5 | 0.07 | 35.7 (4.2–300.5) |
| Flamand-Roze 2011 [21] | LAST | 52 | 50 | 51 | 50 | 98 | 100 | 0 | ∞ | 0.02 | ∞ |
| Choi 2015 [18] | MAST* | 30 | 30 | 27 | 22 | 90 | 73 | 0 | 3.37 | 0.14 | 24.7 (5.9–104) |
| Kostalova 2008 [16] | MAST | 149 | 45 | 143 | 40 | 96 | 89 | 0 | 8.7 | 0.04 | 217.5 (63.1–749.7) |
| Romero 2012 [17] | MAST | 29 | 29 | 26 | 29 | 90 | 100 | 0 | ∞ | 0.10 | ∞ |
| Doesborgh 2003 [20] | ScreeLing | 14 | 49 | 12 | 47 | 86 | 96 | 0 | 21.5 | 0.15 | 143.3 (18.3–1124.3) |
| Al-Khawaja 1996 [13] | SST | 38 | 4 | 35 | 4 | 92 | 80 | 0 | ∞ | 0.08 | ∞ |
| Kim 2011 [22] | SVF, 60 s | 27 | 26 | 23 | 22 | 85 | 85 | 0 | 5.7 | 0.18 | 31.7 (7.0–142.5) |
| Kim 2011 [22] | SVF, 30 s | 27 | 26 | 23 | 23 | 85 | 88 | 0 | 7.1 | 0.17 | 41.8 (8.4–208.0) |
| Thommessen 1999 [19] | UAS | 8 | 29 | 6 | 26 | 75 | 90 | 0 | 7.5 | 0.28 | 26.8 (3.6–197.5) |
LR+ Likelihood Ratio of a Positive Test, LR− Likelihood Ratio of a Negative Test, DOR diagnostic odds ratio
a8 patients have traumatic brain injury
bData on the 42 stroke patients could separately be extracted