Table 4.
Diagnostic Accuracy of the BEDS Reference and BEDS-7 Classification Algorithms and the BES Severity Subgroupsa
| Required BEDS Item Response |
BEDS Screening Status | BED Diagnosis Status, nb |
||||||||||||||||
| Classification Method | BEDS Response | Item 1 | Item 2 | Item 3 | Item 4 | Item 5 | Item 6 | Item 7 | Item 8 | Item 9 | Item 10 | Item 11 | Item 12 | Item 13 | Totalc | |||
| Yes | No | |||||||||||||||||
| Reference algorithm | Yes | X | X | X | – | – | – | – | – | – | – | – | – | – | Positive | 3 (18.8)d | 8 | 11 |
| No | – | – | – | – | – | – | – | – | – | – | – | – | – | Negative | 13 | 67 (89.3)e | 80 | |
| Never/rarely | – | – | – | – | – | – | – | – | – | X | X | X | X | Total | 16 | 75 | 91 | |
| Sometimes | – | – | – | – | – | – | – | – | – | – | – | – | – | |||||
| Often | – | – | – | X | Xf | Xf | Xf | Xf | Xf | – | – | – | – | |||||
| Always | – | – | – | X | Xf | Xf | Xf | Xf | Xf | – | – | – | – | |||||
| BEDS-7 algorithm | Yes | X | – | X | – | – | – | – | – | – | – | – | – | – | Positive | 16 (100.0)d | 46 | 62 |
| No | – | – | – | – | – | – | – | – | – | – | – | – | – | Negative | 0 | 29 (38.7)e | 29 | |
| Never/rarely | – | – | – | – | – | – | – | – | – | X | – | – | – | Total | 16 | 75 | 91 | |
| Sometimes | – | – | – | X | – | – | X | X | X | X | – | – | – | |||||
| Often | – | – | – | X | – | – | X | X | X | – | – | – | – | |||||
| Always | – | – | – | X | – | – | X | X | X | – | – | – | – | |||||
| BES severity subgroupg | NA | Moderate/serious (18+) | 16 (100.0)d | 54 | 70 | |||||||||||||
| None/mild (≤ 17) | 0 | 27 (33.3)e | 27 | |||||||||||||||
| Total | 16 | 81 | 97 | |||||||||||||||
X indicates that the response is required for a positive classification by the BEDS, – indicates that the response is not required. Positive and negative predictive values were not estimated due to the oversampling of individuals with BED characteristics in the quantitative evaluation sample.
BED diagnosis was determined by a clinical interview using a modified version of the BED criteria (part of module H) from the Structured Clinical Interview for DSM-IV-TR Axis I Disorders–Research Version, Non-Patient Edition.
Of the 97 participants, 91 responded “yes” or “no” to item 1, item 2, and item 3.
The sensitivity is presented in parentheses and is defined as the probability that the BEDS algorithm will be positive when the participant was diagnosed with BED (ie, true positive rate).
The specificity is presented in parentheses and is defined as the probability that the BEDS algorithm will be negative when the participant was not diagnosed with BED (ie, true negative rate).
At least 3 of the 5 items (5, 6, 7, 8, and 9) marked as “often” or “always.”
Timmerman12 BES subgroups: none or mild (≤ 17 points), moderate (18 to 26 points), and serious (≥ 27 points).
Abbreviations: BEDS = Binge-Eating Disorder Screener, BES = Binge Eating Scale, NA = not applicable.