TABLE 1
FROM:
ACNP Task Force Report on SSRIs and Suicidal Behavior in Youth
J John Mann, Graham Emslie, Ross J Baldessarini, William Beardslee, Jan A Fawcett, Frederick K Goodwin, Andrew C Leon, Herbert Y Meltzer, Neal D Ryan, David Shaffer and Karen D Wagner
BACK TO ARTICLETable 1. Efficacy of Antidepressants for Treating Pediatric Major Depressive Disorder
| Medication | Open/double blind | Duration (weeks) | Participant ages | Number of study participants (drug and placebo) | Continuous response a | Number needed to treat NNT b | Categorical response c | Number needed to treat NNT b | Reference |
|---|---|---|---|---|---|---|---|---|---|
| SSRIs | |||||||||
| Fluoxetine | DB | 8 | 8–18 | 96 (48 flx, 48 pb) | 58 vs 32% (p=0.013)d | 5 | 56 vs 33% (p=0.02) | 5 | Emslie et al (1997) |
| Fluoxetine | DB | 9 | 8–17 | 419 (209 flx, 210 pb) | 65 vs 53% (p=0.09)d | 9 | 52 vs 37% (p=0.03) | 7 | Emslie et al (2002a, 2002b) |
| Paroxetine | DB | 8 | 12–18 | 275 (93 par, 95 imp, 87 pb) | 63 vs 50 vs 46% (p=0.02)e | 9 | 66 vs 52 vs 48% (p=0.02) | 6 | Keller et al (2001) |
| Paroxetine | DB | 12 | 13–18 | 286 (187 par, 99 pb) | 61 vs 58% (NS); UK: 75 vs 71% (NS)f | 44 | 69.2 vs 53.7% (NS) | 9 | Milin et al (1999); GSK, Final Clinical Report (Study 377) |
| Paroxetine | DB | 8 | 7–17 | 203 (101 par, 102 pb) | Change from baseline (p=0.68; NS)a | — | 49 vs 46% (NS) | 34 | GSK, Final Clinical Report (Study 701) |
| Citalopram | DB | 8 | 7–17 | 174 (89 cit, 85 pb) | Change from baseline (p=0.038)a | 9 | 47 vs 45% (NS) | 51 | Wagner et al (2004) |
| Citalopram | DB | 12 | 13–18 | 244 (124 cit, 120 pb) | NS | — | Data unavailable | — | UK report |
| Sertraline (two studies) | DB | 10 | 6–17 | 376 (189 sert, 187 pb) | 69 vs 59% (p=0.05)d | 10 | 63 vs 53% (p=0.05) | 10 | Wagner et al (2003) |
| Sertraline | 97 sert, 91 pb | 62 vs 56% | 18 | Data unavailable | — | A0501001 | |||
| Sertraline | 92 sert, 96 pb | 34 vs 46% | 9 | Data unavailable | — | A0501017 | |||
| Other new generation antidepressants (non-SSRI) | |||||||||
| Mirtazapine (two studies) | DB | 8 | 7–17 | 250 (165 mirt, 85 pb) | NS | — | NS | — | |
| Nefazodone | DB | 8 | 12–17 | 195 (99 nef, 96 pb) | Change from baseline (p=0.055; NS)a | — | 65 vs 46% (p=0.005) | — | Emslie et al (2002a, 2002b) |
| Nefazodone | DB | Data NA | 7–17 | Data unavailable | Data unavailable | — | Data unavailable | — | Emslie et al (2002a, 2002b) |
| Venlafaxine | DB | 8 | 8–17 | 161 | NS | — | NS | — | Emslie et al (2004) |
| Venlafaxine | DB | 8 | 8–17 | 193 | NS | — | NS | — | Emslie et al (2004) |
flx=fluoxetine; pb=placebo; par=paroxetine; imp=imipramine; mirt=mirtazapine; sert=sertraline; nef=nefazodone; cit=citalopram.
a Continuous response: difference between drug and placebo in the amount of improvement of depressive symptoms (based on Children's Depression Rating Scale—Revised (CDRS-R)).
b NNT is defined in the text.
c Categorical response was defined in all cases as a Clinical Global Improvement of depression of 'much' or 'very much improved'.
d Continuous response: rates of response based on a percent improvement on a depression rating scale (CDRS-R; 30% on fluoxetine trials; 40% on sertraline trials).
e Continuous response: rates of response based on minimal symptoms of depression (Hamilton Depression Rating Scale
8).
f Continuous response: rates of response based on a percent improvement on a depression rating scale (Montgomery Asberg Depression Rating Scale (MADRS); 
50% improvement).
