To the Editor: We read with interest the article by Fardet et al. in the May issue of the Journal (1) and wish to raise concerns with the authors' interpretation that glucocorticoid treatment is associated with higher risk for depression and severe neuropsychiatric disorders.

First, some of the definitions used in the study were vague, and symptoms such as “low mood” were used to identify depression. Psychiatric referral, a more precise marker of neuropsychiatric problems, was similar between the exposed and unexposed groups.

Second, the authors reported a “high incidence of neuropsychiatric adverse events in the first 3 months of glucocorticoid treatment.” Most patients received glucocorticoid treatment for acute exacerbation of obstructive lung disease (asthma or chronic obstructive pulmonary disease) or for lower respiratory tract infection. For these conditions, the duration of glucocorticoid treatment is usually less than 14 days. While neuropsychiatric adverse events may occur during treatment, the time to event extended to 3 months. The lack of a clear temporal association is a major limitation of this study since the event could occur from 2 to 10 weeks after the termination of treatment.

Lastly, depression was the most frequently reported neuropsychiatric adverse event reported in this study. The authors failed to address the known association between the clinical conditions requiring glucocorticoid treatment (2–4) and depression, as well as the association between systemic inflammation (5) and depression. Glucocorticoid treatment (as well as initial daily dose) is a marker of disease severity associated with increased systemic inflammation. This cause-and-effect relationship was not accounted for in the study, and unexposed patients were not matched by disease severity.

For these reasons, we believe that an alternative interpretation should have been presented that takes into account the association of depression and glucocorticoid treatment with increased severity of underlying clinical condition. The alternative interpretation should clarify that for clinical conditions associated with depression, an acute deterioration—associated with increased systemic inflammation and/or glucocorticoid resistance (6)—may worsen symptoms of depression.