A Comparative Meta-Analysis of Clinical Global Impressions Change in Antidepressant Trials

Abstract

Two scales of the Clinical Global Impressions (CGI) Scale are frequently used in antidepressant trials. No research has systematically addressed how CGI change compares to change on established measures such as the Hamilton Depression Rating Scale (HAM-D), Montgomery-Asberg Depression Rating Scale, or Beck Depression Inventory. The current meta-analysis examined 75 antidepressant trials in which the CGI was used along with at least one other popular depression measure. The CGI-Severity scale was significantly more conservative than the HAM-D in rating change in double-blind trials, but not in open trials. The Beck Depression Inventory was significantly more conservative than the CGI-Severity. The CGI-Improvement scale was significantly more liberal than the HAM-D or Montgomery-Asberg Depression Rating Scale. Rater bias or scale content may explain differences between measures. Given the often substantial differences between instruments, researchers should use a variety of measures rather than relying on any single tool in assessing treatment response.

Assessment of mood: Guides for clinicians

“…The original study reported that the MADRS was more sensitive in distinguishing between responder and nonresponders than the HAMD [8]. However, a systematic review of antidepressant trials where both HAMD and MADRS were used revealed no statistically significant difference in effect sizes between the two rating scales [42].…”

“…A post hoc analysis of seven antidepressant clinical trials of major depression (total N=1927), based on the anchor-based approach which correlates the HAMD scores with the Clinical Global Impression-Severity scores, has established the following interpretative guideline [41] Measuring change Validity for change measure. A systematic review of antidepressant trials where both HAMD and MADRS were used revealed no statistically significant difference in effect sizes between the two rating scales [42]. In other words, the HAMD may be as sensitive to change as the MADRS, which was originally developed to be more sensitive to treatment changes than its predecessors [8].…”

An Open-Label Pilot Study to Evaluate the Efficacy of Sildenafil Citrate in Middle-Aged Men With Late-Onset Dysthymia

“…Sexual functioning was evaluated with the IIEF, a 15-item selfadministered questionnaire that consists of the following 5 domains: erectile function, orgasmic function, sexual desire, intercourse satisfaction, and overall satisfaction (Cappelleri et al, 1999). Overall illness severity and global change were assessed with the clinical global impression (CGI) illness severity, and change scales (CGIc) (Spielmans and McFall, 2006).…”

Hamilton depression rating subscales to predict antidepressant treatment outcome in the early course of treatment

“…The predictive values for treatment outcome of the full HAMD 17 were already replicated in a much larger dataset bySzegedi et al (2009).4 The caveat of the CGI scales is that they are a global measure of mental wellbeing and are also not an ideal treatment outcome criterion: compared to the HAMD 17 , they have proven to be rather conservative regarding the remission classification (CGI-S shows less change than the HAMD 17 ) and more liberal regarding the response classification (CGI-I is more likely to show treatment response than the HAMD 17 )(Spielmans and McFall, 2006).…”