Laboratory (phase I) and on-field (phase II) validity and player (phase III) and practitioner (phase IV) feasibility results
| Biocore-FRI | HitIQ | ORB | Prevent | |
| Phase I | ||||
| Linear CCC (95% CI) | 0.981 (0.967 to 0.989) | 0.944 (0.906 to 0.967) | 0.445 (0.294 to 0.575) | 0.980 (0.967 to 0.989) |
| Angular CCC (95% CI) | 0.990 (0.983 to 0.994) | 0.983 (0.970 to 0.990) | 0.533 (0.379 to 0.658) | 0.982 (0.969 to 0.990) |
|
Total CCC
( 95% CI ) |
0.986
(0.980 to 0.990) |
0.965
(0.949 to 0.976) |
0.525
(0.420 to 0.617) |
0.984
(0.977 to 0.989) |
| Linear mean bias (LOA) | −1.2% (−17.1% to 14.8%) | −4.0% (−33.7% to 25.8%) | −36.4% (−78.3% to 5.6%) | −1.0% (−16.9% to 14.8%) |
| Angular mean bias (LOA) | −2.3% (−17.0% to 12.4%) | −4.2% (−24.5% to 16.2%) | −51.0% (−96.1% to −6.0%) | −6.8% (−21.7% to 8.0%) |
| Phase II (false-positive; guided analysis ) | ||||
| Player matches (n) | 17 | 28 | 9 | 27 |
| iMG-triggered events (n) n per player, per match (median (IQR)) | 201 9 (6, 12) | 405 13 (8, 20) | 179 9 (6, 19) | 745 23 (17, 34) |
| True-positives (n) n per player, per match (median (IQR)) | 162 9 (6, 11) | 245 8 (5, 11) | 66 1 (0, 7) | 666 22 (15, 32) |
| False-positives (n) n per player, per match (median (IQR)) | 4 0 (0, 0) | 28 1 (0, 2) | 59 3 (2, 6) | 45 1 (0, 2) |
| True-positives and false-positives PPV (95% CI) | 0.98 (0.94 to 0.99) | 0.90 (0.86 to 0.93) | 0.53 (0.44 to 0.62) | 0.94 (0.92 to 0.95) |
| Inactive assumed false-positives (n) n per player, per match (median (IQR)) | 30 0 (0, 0) | 21 0 (0, 1) | 7 0 (0, 0) | 0 0 (0, 0) |
| True-positives, false-positives and inactive assumed false-positives PPV (95% CI) |
0.83
(0.77 to 0.88) |
0.83
(0.79 to 0.87) |
0.50
(0.41 to 0.59) | 0.94 (0.92 to 0.95) |
| All assumed false-positives (n) n per player, per match (median (IQR)) | 35 0 (0, 1) | 132 3.5 (1, 5.5) | 54 4 (1, 13) | 34 0 (0, 1) |
|
True-positives, false-positives and all assumed false-positives PPV
(95% CI) |
0.81
(0.74 to 0.86) |
0.60
(0.56 to 0.65) |
0.37
(0.30 to 0.44) |
0.89
(0.87 to 0.92) |
| Phase II (false-negative; unguided analysis) | ||||
| True-positive (n) | 47 | 49 | 20 | 82 |
| False-negative (n) | 46 | 73 | 8 | 27 |
|
Sensitivity (%)
(95% CI) |
0.51
(0.41 to 0.61) |
0.40
(0.31 to 0.49) |
0.71
(0.54 to 0.88) |
0.75
(0.67 to 0.83) |
| Phase III | ||||
| Fit (mean %±SD) (range) | 77±31 (16, 100) | 88±12 (63, 100) | 65±22 (22, 94) | 85±11 (67, 100) |
| Comfort (median (IQR)) | 6 (5, 7) | 8 (6, 8) | 5 (3, 7) | 8 (7, 8) |
| Function (mean %±SD) (range) | 55±35 (16, 100) | 61±31 (21, 95) | 43±31 (11, 83) | 67±22 (44, 94) |
| Phase IV | ||||
| Preparation (SUS out of 100; mean±SD) (range) | 51.3±21.7 (25, 78) | 71.3±20.7 (55, 100) | – | 83.8±21.3 (53, 95) |
| Management (SUS out of 100; mean±SD) (range) | 50.6±28.5 (20, 93) | 78.8±11.1 (68, 100) | – | 80.0±20.7 (50, 98) |
All assumed false-positives is the sum of off-camera and inactive assumed false-positives; off-camera assumed false-positives are shown in online supplemental table 1. Preparation and management are reported as mean and SD.
CCC, concordance correlation coefficient; FRI, Football Research Inc; LOA, limits of agreement; PPV, positive predictive value; SUS, System Usability Scale.