Gravem 1987.
| Methods | Six‐week, double‐blind, multicentre trial | |
| Participants | In‐ and out‐patients who were referred to hospital for a depression requiring drug treatment. The patient's depression was classified as endogenous or non‐endogenous by means of the Newcastle Depression Scale I. Age range: 18‐70 years Exclusion criteria: serious physical disease, pregnancy, previous resistance to therapy with amitriptyline or citalopram in doses considered to be adequate. |
|
| Interventions | Citalopram: 27 participants Amitriptyline: 24 participants Citalopram dose range: 20‐60 mg/day Amitriptyline dose range: 75‐225 mg/day Additionally treatment was not allowed apart from low doses of diazepam or nitrazepam for severe anxiety or insomnia, if necessary. |
|
| Outcomes | Primary outcome: endpoint total score on the 10‐item Comprehensive Psychopathological Rating Scale (CPRS) sub‐scale for depression. | |
| Notes | This study was funded by Lundbeck (citalopram manufacturer). No signed informed consent was required, neither from the patient nor from his relatives. The clinician informed the patient of the object of the study and that he/she was quite free to participate. At that time there were no ethical committees in Norway to evaluate the design of study (Health Autorities approved the study). One suicide attempted in citalopram group. |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | No reliable information provided (no data about sequence generation). |
| Allocation concealment (selection bias) | Unclear risk | No information reported. |
| Blinding (performance bias and detection bias) All outcomes | Low risk | Quote: "tablets of identical appearance were prepared". Probably done. |
| Incomplete outcome data (attrition bias) All outcomes | Unclear risk | No information about incomplete data in each group. |
| Selective reporting (reporting bias) | High risk | No MADRS scores were reported (neither at baseline nor at endpoint). Response rate and remission rate are missing. |
| Other bias | Unclear risk | Sponsorship bias cannot be ruled out. |