| Methods |
Amitriptyline vs placebo
RCT
41 GPs in UK |
| Participants |
N=178 Amitriptyline n=67, placebo n=74; 18‐64 yrs
Research diagnostic criteria for major, minor or intermittent depression HAMD score >=6.
Exclusions: referral to a psychiatrist, antidepressant in previous 3 months, drug/alcohol problems, phobic, Generalised Anxiety Disorder (GAD), Obsessive Compulsive Disorder (OCD), HAMD >=27
Major depression: 67% of intervention group, 74% placebo;
Minor or intermittent 33% in intervention group, 26% in placebo group. |
| Interventions |
Amitriptyline 75mg in first week then 100mg for second week, increasing to 125‐175 mg if needed.
Cointervention with benzodiazepines was allowed.
Psychiatrist did all assessments as home visits. |
| Outcomes |
21% (37) drop outs at 4 weeks.
Intention to treat analysis of the 141 remaining.
Results at 6 weeks:
HAMD mean reduction 9.3 in Amitriptyline group and 6.1 in placebo group. Mean difference 3.2 (95% CI 1.3‐5 P<0.001); HAMD mean 5.4 (SD 3.8), n=67 in Amitriptyline group vs mean 8.7 (SD 9.6) n= 74 in placebo group at 6 weeks.
At 1 week: HAMD mean 11 (SD 3.8) n=67 vs mean 13.5 (SD 3.5) n=74 in placebo group.
At 2 weeks: mean 7.8 (SD 3.8) n= 67 in Amitriptyline group vs mean 10.4 (SD 3.5) n=74 in placebo group.
At 3 weeks: mean 6.0, (SD 3.8) n=67 in Amitriptyline group vs mean 10.5 (SD 3.5) n=74 in placebo group.
At 4 weeks: mean 5.0 (SD 3.8) n=67 vs mean 9.0 (SD 3.5) n=74 in placebo group.
Paykel et al 1988 found amitriptyline more effective for major depression (HAMD 13‐24) than minor depression (HAMD 6‐12).
Adverse effects leading to withdrawal: 18/67 in amitriptyline group dropped out due to dry mouth, increase in appetite and weight gain. 8/74 in placebo poor compliance and 3/74 withdrawal due to treatment failure (worsening depression).
Withdrawal for any reason: 23/67 in amitriptyline group and 14/74 in the placebo group. |
| Notes |
Psychiatrist did all interviews
Median dose Amitriptyline = 125mg |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Allocation concealment (selection bias) |
Low risk |
A ‐ Adequate |
