ACCESSIBLE SUMMARY This paper reports on the issue of resuscitation in mental health inpatient environments. It reviews the literature on national standards and best practice when emergency situations arise in mental health settings. The discussion on the best practice literature takes place alongside the reporting of a national evaluation of how National Patient Safety Agency improvement guidelines for the provision for life support, and resuscitation for mental health service users was effectively implemented across health-care providers in England. Methods used to establish the effective use of the guidelines include feedback from clinical staff and staff responsible for the implementation of the new national standards for resuscitation. Serious incident data were also compared prior to the release of the national guidelines and after the guideline release dates. This included looking at events around choking and cardiac/respiratory arrest in inpatient areas. There were five deaths post-implementation of the guidelines that were considered to have serious enough error associated with the resuscitation process. This was down from 18 prior to the release of the guidelines. However, our survey showed that despite organisations reporting 100% compliance with the implementation of the guidelines, around half of frontline clinical staff were not aware of them. Although our survey responses show a contradiction between organisational and clinical staff awareness, our analysis suggests a reduction in moderate and severe harm cases and of deaths. There is evidence of a reduction in the worst types of error resulting in death, albeit with small numbers. In 2008, the National Patient Safety Agency issued a Rapid Response Report (RRR) aimed at health-care organisations providing inpatient care for mental health and learning disability patients, requiring organisations to make proper provision for life support and resuscitation for these patients. This paper examines whether effective implementation of the RRR recommendations had occurred across health providers in England. The methods used in this paper are the following: (1) Questionnaires were distributed nationally to clinical staff and implementation leads; (2) A national comparison of the number and severity of pre- and post-RRR release-related incidents involving choking/cardiac/respiratory arrest in Mental Health and Learning Disabilities settings was conducted; (3) Organisational compliance with the patient safety alert for all National Health Service Organisations in England was measured. There were five deaths post-implementation of the RRR that were considered to have serious enough error associated with the resuscitation. This was down from 18 deaths pre the RRR release. Although our survey responses show a contradiction between organisational implementation and clinical staff awareness, our analysis suggests a reduction in moderate and severe harm cases and of deaths. There is evidence of a reduction in the worst types of error resulting in death, albeit with small numbers.