Table 6.
Subgroup statistics for the effects of benzodiazepines on the non-cued baseline startle response in rats
| Articles | Experiments | Animals | SMD 95% CI | I2 (%) | |
|---|---|---|---|---|---|
| All experiments | 17 | 27 | 684 | − 1.13 [− 1.47, − 0.80] | 75 |
| Drugs | |||||
| Alprazolam | 1 | 1 | 24 | Not pooled | |
| Chlordiazepoxide | 7 | 9 | 300 | − 1.56 [− 2.16, − 0.96] | 75 |
| Diazepam | 11 | 14 | 288 | − 0.81 [− 1.30, − 0.32] | 79 |
| Midazolam | 1 | 1 | 24 | Not pooled | |
| Oxazepam | 1 | 2 | 48 | Not pooled | |
| Strain | |||||
| F344 | 1 | 1 | 24 | Not pooled | |
| HL | 1 | 1 | 20 | Not pooled | |
| LE | 0 | 0 | 0 | NA | |
| RORO | 1 | 1 | 24 | Not pooled | |
| SDR | 6 | 10 | 188 | − 0.72 [− 1.22, − 0.22] | 79 |
| Wi | 8 | 14 | 428 | − 1.22 [− 1.63, − 0.81] | 56 |
| Time of testing | 2 | 3 | 176 | Not pooled | |
| Active period | 11 | 17 | 368 | − 1.22 [− 1.65, − 0.78] | 73 |
| Passive period | 4 | 7 | 140 | − 1.18 [− 1.87, − 0.49] | 83 |
| NR | |||||
| Shock intensity (mA) | |||||
| High (≥ 0.8) | 1 | 2 | 32 | Not pooled | |
| Low (< 0.3) | 3 | 3 | 60 | Not pooled | |
| Moderate (0.3 ≤ intensity < 0.8) | 14 | 22 | 592 | − 1.14 [− 1.53, − 0.76] | 79 |
| Total number of cue-shock pairings | |||||
| High (> 10 pairings) | 13 | 20 | 412 | − 1.35 [− 1.73, − 0.97] | 71 |
| Moderate (3–10 pairings) | 4 | 7 | 272 | − 0.56 [− 1.17, − 0.04] | 75 |
| Low (< 3 pairings) | 0 | 0 | 0 | NA | |
| Highest startle probe intensity (dB; calculated relative to background noise) | |||||
| High (> 35) | 11 | 18 | 536 | − 1.46 [− 1.85, − 1.06] | 74 |
| Moderate (20 ≤ intensity < 35) | 3 | 4 | 68 | Not pooled | |
| Low (< 20) | 0 | 0 | 0 | NA | |
| NA (intensity not reported) | 3 | 5 | 80 | − 0.57 [− 1.35, 0.21] | 83 |
| Study design | |||||
| Between-subjects | 13 | 18 | 368 | − 0.88 [− 1.30, − 0.46] | 67 |
| Within-subjects (Latin square) | 4 | 9 | 316 | − 1.64 [− 2.23, − 1.05] | 84 |
| Quality indicators | |||||
| Blinding reported | |||||
| No | 15 | 24 | 480 | − 1.20 [− 1.58, − 0.82] | 76 |
| Yes | 1 | 1 | 48 | Not pooled | |
| Yes, not blinded | 1 | 2 | 156 | Not pooled | |
| Sample size calculation | |||||
| No | 16 | 25 | 528 | − 1.19 [− 1.54, − 0.83] | 75 |
| Yes | 1 | 2 | 156 | Not pooled | |
| Allocation bias | |||||
| No measures reported | 5 | 6 | 232 | − 0.83 [− 1.53, − 0.13] | 26 |
| Random allocation | 4 | 5 | 124 | − 1.56 [− 2.38, − 0.73] | 87 |
| Matching (any kind) | 8 | 16 | 328 | − 1.14 [− 1.59, − 0.69] | 76 |
Subgroups were omitted from the analysis if these contained fewer than five experiments, from fewer than three different articles
SMD standardized mean difference (Hedge’s g), CI confidence interval, F344 Fischer 344, LE Lewis, LH Lister Hooded, NA not applicable, RORO Fuellinsdorf Albino, SDR Sprague–Dawley, Wi Wistar