Psychometric properties of sideline assessment tests*
Author | Type of athletes | Athletes (n) | Concussed | Controls | Test and/or criterion | Sensitivity (%) | Specificity (%) | Test–retest reliability | AUC |
Symptoms | |||||||||
McCrea et al 19 | College football | 1631 | 94 | 56 | 89 | 100 | |||
Putukian et al 22 | College athletes | 263 | 32 | 23 | SCAT2 | 84 | 100 | ||
Chin et al 23 | High school and college athletes | 2018 | 166 | 164 | 0.88 | ||||
Resch et al 120 | College athletes | 40 | 40 | Revised Head Injury Scale | 98 | 100 | |||
Garcia et al 40 | College athletes | 733 | SCAT3 | 93 | 97 | 0.98 | |||
Broglio et al 33 | College athletes | 4360 | 0.40† | ||||||
Total | 3192 | 1065 | 283 | ||||||
Standardized Assessment of Concussion | |||||||||
Barr and McCrea15 | High school and college football | 1313 | 50 | 68 | 3-point decline | 72 | 94 | 0.55‡ | |
McCrea et al 19 | High school and college football | 1325 | 63 | 55 | 3-point decline | 78 | 95 | 0.48§ | |
McCrea et al 17 | High school and college football | 2385 | 91 | <10th percentile of normative | 79 | ||||
McCrea et al 19 | College football | 1631 | 94 | 56 | ? | 80 | 91 | ||
Echlin et al 121 | Ice hockey (age 16–21) | 67 | 21 | – | 1-point decline | 54 | |||
Barr et al 16 | High school and college football | 823 | 59 | 31 | ? | 46 | 87 | ||
Marinides et al 20 | College athletes | 217 | 30 | 2-point decline | 52 | 82 | |||
Galetta et al 21 | Hockey/lacrosse youth/college | 332 | 12 | 14 | 2-point decline | 20 | 21 | 0.68 | |
Putukian et al 22 | College athletes | 263 | 32 | 23 | <10th percentile of normative | 41 | 91 | ||
Chin et al 23 | High school and college athletes | 2018 | 166 | 164 | 0.39† | 0.56 | |||
Broglio et al 33 | College athletes | 4874 | 0.39† | ||||||
Total | 15 284 | 618 | 411 | ||||||
BESS | |||||||||
McCrea et al 19 | College football | 1631 | 94 | 56 | Modified BESS | 36 | 95 | ||
Broglio et al 122 | Young adults | 48 | BESS | 0.60¶ | |||||
Barr et al 16 | High school and college football | 823 | 59 | 31 | Modified BESS | 31 | 71 | ||
Putukian et al 22 | College athletes | 263 | 32 | 23 | Modified BESS | 25 | 100 | ||
Chin et al 23 | High school and college athletes | 2018 | 166 | 164 | Modified BESS | 0.54† | 0.56 | ||
Broglio et al 33 | College athletes | 2894 | BESS | 0.41† | |||||
Total | 4735 | 351 | 274 | ||||||
Oculomotor (KD) | |||||||||
Galetta et al 27 | Football, men’s/women’s basketball | 219 | 10 | Worsening of KD time | 100 | ||||
Leong et al 123 | Boxing | Worsening of KD >5 s | 100 | 100 | 0.9† | ||||
Galetta et al 21 | Hockey/lacrosse youth/college | 332 | 12 | 14 | Worsening of KD time | 75 | 93 | 0.92 | |
Leong et al 28 | College football, men’s/women’s basketball | 127 | 11 | Worsening of KD time | 89 | 0.95† | |||
King et al 124 | Amateur rugby | 94 | 100 | 0.92† | |||||
Marinides et al 20 | Football, women’s lacrosse, soccer | 217 | 30 | Worsening of KD time | 79 | ||||
Seidman et al 24 | High school football | 343 | 9 | Worsening of KD time | 100 | 100 | |||
Dhawan et al 29 | Youth hockey | 141 | 20 | Worsening of KD >5 s | 100 | 91 | |||
Fuller et al 125 | Elite English rugby | 145 | Worsening of KD time | 60 | 39 | 0.51 | |||
Hecimovich et al 126 | Australian football | 22 | 22 | Worsening of KD time | 98 | 96 | 0.91† | ||
Professional football | 1223 | 84 | 63 | Worsening of KD | 84 | 62 | 0.88† | ||
Broglio et al 33 | College athletes | 755 | 0.74† | ||||||
Eddy et al 127 | Recreational college athletes | 63 | 0.90† | ||||||
Total | 2041 | 310 | 99 | ||||||
Clinical reaction time (dropped weighted stick) | |||||||||
Eckner128 | College football, wrestling, women’s soccer | 102 | 0.65† | ||||||
Eckner et al 47 | High school and college athletes | 28 | 28 | 90% CI | 50 | 86 | |||
Broglio et al 33 | College athletes | 261 | 0.32† | ||||||
Total |
Test–retest reliability:
*Study selection criteria: athletes competing at any level of sport using any sideline screening assessment or studies with test–retest reliability of included assessments. All studies were of high risk of bias as assessed using Quality Assessment of Diagnostic Accuracy Studies - 2, except for Fuller et al,125 which was of low risk of bias.
†Intraclass correlation coefficient.
‡Reliable change index.
§Pearson’s correlation coefficient.
¶Generalisability coefficient.
AUC, area under the curve; KD, King-Devick; SCAT, Sports Concussion Assessment Tool.