A test of the concept of "underlying depressive illness" in the treatment of anxiety states.

In order to test the validity of the concept of anxiety states masking an underlying depressive illness, patients clinically diagnosed as suffering from anxiety or tension states were treated on a random double-blind basis for 4 weeks with either a pure anxiolytic, lorazepam, or an anxiolytic/antidepressant preparation, fluphenazine with nortriptyline. Patients' self-ratings were very similar to the physicians' ratings which showed that fluphenazine/nortriptyline was associated with significantly greater overall improvement(p less than 0.01), as well as significantly greater improvements in the group of symptoms specifically related to depression(p less than 0.05). These results suggest that a depressive element is present in an appreciable proportion of patients presenting with apparent anxiety states, and antidepressant as well as anxiolytic treatment is required. Patients selected on the basis that they had improved satisfactorily at the end of the 4-weeks' treatment were followed up for a further 3 months without medication, and the relapse rate was 24%, irrespective of previous treatment. More of the patients treated with lorazepam had to be excluded from the follow-up because of failure to improve, and these probably represented the proportion (19%) of this population with an appreciable depressive element to their illness.