Anne-Marie Slowther

Background Do not attempt cardiopulmonary resuscitation decisions have been widely criticised. The Recommended Summary Plan for Emergency Care and Treatment (ReSPECT) process was developed to facilitate shared decisions between patients and clinicians in relation to emergency treatments, including cardiopulmonary resuscitation. Objective To explore how, when and why ReSPECT plans are made and what effects the plans have on patient outcomes. Design A mixed-methods evaluation, comprising (1) a qualitative study of ReSPECT decision-making processes, (2) an interrupted time series examining process and survival outcomes following in-hospital cardiac arrest and (3) a retrospective observational study examining factors associated with ReSPECT recommendations and patient outcomes. Setting NHS acute hospitals and primary care and community services in England (2017–2020). Participants Hospital doctors, general practitioners, nurses, patients and families. Data sources The following sources were used: (1) observations of ReSPECT conversations at six hospitals and conversations with clinicians, patient, families and general practitioners, (2) survey and freedom of information data from hospitals participating in the National Cardiac Arrest Audit and (3) a review of inpatient medical records, ReSPECT forms and NHS Safety Thermometer data. Results By December 2019, the ReSPECT process was being used in 40 of 186 (22%) acute hospitals. In total, 792 of 3439 (23%) inpatients, usually those identified at risk of deterioration, had a ReSPECT form. Involvement of the patient and/or family was recorded on 513 of 706 (73%) ReSPECT forms reviewed. Clinicians said that lack of time prevented more conversations. Observed conversations focused on resuscitation, but also included other treatments and the patient’s values and preferences. Conversation types included open-ended conversations, with clinicians actively eliciting the patients’ wishes and preferences, a persuasive approach, swaying the conversation towards a decision aligned with medical opinion, and simply informing the patient/relative about a medical decision that had already been made. The frequency of harms reported on the NHS Safety Thermometer was similar among patients with or without a ReSPECT form. Hospital doctors and general practitioners gave different views on the purpose of the ReSPECT process and the type of recommendations they would record. Limitations The research was undertaken within the first 2 years following the implementation of ReSPECT. Local policies meant that doctors led these conversations. Most patients were seriously ill, which limited opportunities for interviews. Incomplete adoption of the ReSPECT process and problems associated with the NHS Safety Thermometer tool affected the evaluation on clinical outcomes.