Diagnosing Melancholia

Letter to the Editor

Sir: In their recent article (1999, Supplement 20) reporting results from a double-blind randomized treatment of late-life depression, Mulsant et al. found a lack of statistically significant difference between the rates of response to nortriptyline and paroxetine. The authors report that over 6 weeks of treatment, the tolerability of and response to therapeutic plasma drug levels were similar in their patients, irrespective of inpatient versus outpatient status or the presence or absence of melancholia. As the authors noted, these results are in discord with several reports including 2 multicenter, double-blind studies from the Danish University Antidepressant Group and 1 from the New York State Psychiatric Institute (NYSPI).