The Clinical Global Impressions scale: errors in understanding and use.

Area of Science:

• Psychopharmacology
• Clinical Trial Methodology
• Biostatistics

Background:

• The Clinical Global Impressions Severity (CGI-S) and Improvement (CGI-I) scales are critical for assessing drug efficacy in new psychopharmacology applications.
• Ensuring the integrity of these widely used measures is essential for accurate drug development and regulatory review.

Purpose of the Study:

• To investigate whether clinical trial investigators incorporate information unrelated to treatment efficacy into their CGI-S and CGI-I ratings.
• To determine the impact of indication-unrelated adverse events on the validity of CGI ratings in psychopharmacology studies.

Main Methods:

• Forty-five principal investigators rated patient narratives for major depressive disorder or generalized anxiety disorder using CGI-S and CGI-I scales.
• Investigators were randomized to receive narratives with or without unrelated medical or psychiatric adverse events.
• A survey assessed investigators' perceptions of how unrelated adverse events influence their CGI ratings.

Main Results:

• CGI-S and CGI-I ratings were significantly more severe and indicated less improvement when narratives included unrelated adverse events (P < .04).
• Despite self-reported objectivity, investigators' ratings were demonstrably influenced by unrelated medical and psychiatric adverse events.
• A significant difference in CGI-S ratings was observed when unrelated medical adverse events were present (P < .03).

Conclusions:

• The inclusion of indication-irrelevant adverse events in CGI ratings compromises the validity of the CGI as a measure of treatment efficacy.
• This bias may lead to a failure in detecting true efficacy signals in psychopharmacology clinical trials.
• Standardizing rating procedures and investigator training is crucial to mitigate bias and ensure reliable efficacy assessments.