Treatment-resistant depression (TRD) has been a controversial issue for more than 35 years. Despite a large number of publications, clinicians and researchers still know few about the prevalence of TRD since no staging method has been validated.

We sought to assess similarities and differences between the 4 TRD-staging methods published to date and assess TRD prevalence in a clinical sample. We also wanted to look for clinical factors associated with TRD.

We conducted a clinical study in a psychiatric unit of a university hospital. We designed a hetero-questionnaire to stage TRD according to the 4 methods. Psychiatric diagnosis, depression severity and cognitive status were assessed using standardized tools. Patients were not included in the study if they suffer from schizophrenia, bipolar disorder, current substance dependence or major cognitive impairment.

We recruited 37 inpatients. Twenty-four had received an inadequate treatment, four had not responded after one adequate antidepressant trial and 9 after two trials. Only 7 were resistant according to the 4 staging methods, which showed several differences. The Massachussetts General Hospital (MGH) method appeared as the most specific and easy-to-use. Complete TRD seemed much less frequent than pseudo-resistance and relative resistance with untreated comorbidity. Chronic depression and comorbidity were frequent in the TRD subgroup.

The concept of “difficult-to-treat” depression might be more appropriate for clinical practice than TRD. The MGH tool seems to fit best to clinical practice. Further research is needed to confirm these descriptive findings in a larger sample.