Background: Electroencephalography (EEG) is a routine clinical tool that is used to evaluate thalamocortical function in comatose patients. The presence or absence of reactivity in background EEG patterns to afferent stimuli is believed to be an important indicator of clinical outcome. At present, there are no guidelines or standardized testing protocols for the assessment of EEG reactivity in critically ill patients. Moreover, the inter-rater reliability of subjectively identifying presence or absence of reactivity is known to be poor. Methods: Here we report the implementation of a clinical protocol formalizing the use of afferent stimuli – name-calling, clapping, nasal tickle, central painful stimuli and tracheal suction – administered during the routine EEG evaluation of behaviourally unresponsive patients in the critical care units at London Health Sciences Centre. EEGs were evaluated by qualified electroencephalographers. Results: This retrospective observational study of consecutive patients describes the inter-rater reliability of detecting presence or absence of EEG reactivity since implementation of the clinical protocol. Moreover it evaluates the relationship between EEG reactivity and clinical outcome to determine its reliability as a prognostic tool. Conclusions: The implementation of clinical protocols to standardize testing parameters may improve the ability to provide a reliable neurologic prognosis for critically ill patients in a comatose and behaviourally unresponsive state.