Little is known about the clinical relevance of the Hamilton Rating Scale for Depression (HAMD-17) total scores. It is unclear how total scores translate into clinical severity, or what commonly used measures for response (reduction from baseline of>= 50% in the total score) and remission (total HAMD-17 score<= 7) mean from a clinical perspective. We therefore compared: (a) the percentage and absolute change in the HAMD-17 total scores with Clinical Global Impression-Improvement (CGI-I); (b) the absolute and percentage change in the HAMD-17 total scores with Clinical Global Impression-Severity (CGI-S) absolute change; and (c) the percentage and absolute change in the HAMD-17 total scores with CGI-I in the subgroups of patients with<= median and> median HAMD-17 total scores at baseline. The method used was equipercentile linking of HAMD-17 and CGI ratings from 43 drug trials in patients with Major Depressive Disorder (MDD) (n = 7131). Our results confirm the validity of the commonly used measures for remission and response in MDD trials: a CGI-I score of 2 ('much improved') corresponded to a reduction from baseline of> 50% and< 60%, and a CGI-I score of 1 ('very much improved') to a reduction of> 75% and< 85%. The CGI-S score of 1 ('normal., not at all ill') corresponded to the HAMD-17 total score of< 5 and the CGI-S score of 2 ('borderline mentally ill') to the score between 6 and 8. An effect of baseline illness severity was observed.