Background and objectives. Clinical care protocols, such as CODITOX for severe acute poisonings, attempt to codify and coordinate care processes to prevent delays in situations where timing is a key factor. This study aimed to measure the degree of adherence to CODITOX and compare patterns of activation of the protocol over time. Methods. Noninterventional, retrospective, analytical study of CODITOX activations in 2012. We compared poisonings attended according to the CODITOX protocol to those attended without CODITOX activation in the emergency department's resuscitation unit. Results. Out of 90 poisoning cases, 12 were attended following the CODITOX (group A). In 36 cases (group B), patients were not attended according to protocol even though they presented with criteria that should have activated it; in the remaining 42 cases, criteria for activating the protocol were not present and it was not used. Complications developed in 75% of group-A cases and 28% of group-B cases (P=.006). In group A, 41.7% of the patients were admitted to the intensive care unit (25% in group B); mortality was 4.4% and there were no significant between-group differences. The times between arrival and administration of charcoal and between arrival and administration of an antidote tended to be shorter in group A, but again the between-group differences were not statistically significant and did not affect duration of hospital stay. Conclusions. The CODITOX protocol is not used in all cases that meet the activation criteria. Although the cases treated according to protocol were more serious, no clinical, therapeutic, or outcome differences could be identified between patients treated or not treated according to the CODITOX. The protocol-treated patients developed more complications, however.