Table 3.
Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of parent-reports for predicting weight classification with parent-reported BMI z-score.
| Parent-report overall (n = 55) | Parent-report ≤ 7d TG (n = 37) | Parent-report ≤ 7d TG + Direct (n = 24) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| NM (n%) | OW (n%) | OB (n%) | NM (n%) | OW (n%) | OB (n%) | NM (n%) | OW (n%) | OB (n%) | |
| Sensitivitya | 66 | 38 | 60 | 60 | 71 | 50 | 53 | 100 | 50 |
| Specificityb | 55 | 67 | 98 | 83 | 60 | 97 | 86 | 55 | 100 |
| PPVc | 72 | 26 | 75 | 88 | 29 | 67 | 90 | 31 | 100 |
| NPVd | 48 | 78 | 96 | 50 | 90 | 94 | 43 | 100 | 96 |
| Accuracye | 62 | 60 | 95 | 68 | 62 | 92 | 63 | 63 | 96 |
NM, normal weight; OW, overweight; OB, obese.
aSensitivity = True Positives / (True Positives + False Negatives).
bSpecificity = True Negatives / (True Negatives + False Positives).
cPositive Predictive Value = True Positives / (True Positives + False Positives).
dNegative Predictive Value = True Negatives / (True Negatives + False Negatives).
eAccuracy = (True Positives + True Negatives) / Total “n.”