Estimation of symptom-free days in generalized anxiety disorder

The efficacy of treatments for generalized anxiety disorder has usually been measured in terms of response or remission of symptoms. These endpoints, however, may not adequately capture the transient periods of symptom abatement and relapse characteristic of chronic psychiatric disorders. Here, we evaluate the measurement of treatment effectiveness in terms of the number of symptom-free days (SFDs).

A pooled analysis was performed of data from five manufacturer-initiated trials of venlafaxine extended-release (XR) in patients with generalized anxiety disorder without co-morbid major depressive disorder. The trials were randomized, double-blind, placebo-controlled and of 8 weeks duration (total intent-to-treat population 1295 venlafaxine XR, 544 placebo). Two of the studies had extensions up to 6 months (intent-to-treat population 514 venlafaxine XR, 253 placebo). The patients were >or= 18 years of age with a Hamilton Rating Scale for Anxiety (HAM-A) score of >or= 18.

SFDs were estimated using weekly scores on the HAM-A. Values of 7 and 0 SFDs, respectively, were assigned to each week the patient had a HAM-A score of or= 18 (the minimum threshold for anxiety). Fractional SFD values were assigned proportionately to weekly HAM-A scores between 7 and 18.

The median (inter-quartile range) SFDs were 19 (2-36) for venlafaxine XR and 10 (0-27) for placebo in the 8-week studies (p < 0.0001). In the 6-month extension studies the SFDs were 102 (27-139) for venlafaxine XR and 36 (0-94) for placebo (p < 0.0001).

SFDs differentiate between active treatment and placebo in clinical trials and may be an appropriate measure of treatment effectiveness.