The experiences of adult patients, families, and healthcare professionals of CPR decision-making conversations in the United Kingdom: A qualitative systematic review

Cardiopulmonary resuscitation (CPR) may be attempted after cardiac arrest; however, when arrest is expected as part of dying and CPR is unlikely to benefit the patient, UK guidance recommends anticipatory discussions and documentation of recommendations not to attempt CPR. Recent UK legal and professional guidance emphasizes involving patients and families in CPR decision-making, moving away from paternalistic models. Reports during the COVID-19 pandemic highlighted deficits in involvement. CPR discussions are clinically, emotionally, and ethically challenging and can significantly affect patients, families, and HCPs. Multiple UK processes guide these discussions, and perceptions are shaped by societal, cultural, and economic factors. This review aimed to explore how CPR decision-making conversations in the UK are experienced by adult patients, families, and HCPs.

The review situates findings within broader literature on difficult conversations. Prior work in advance care planning (ACP) and end-of-life (EOL) discussions shows variability in patients’ desire for involvement and inconsistent HCP attitudes, with families often feeling HCPs disengage from them. An international review of DNACPR decision-making noted ethical challenges for clinicians. A previous integrative UK review (2000–2016) found patients preferred family involvement and GP-based settings for CPR discussions, but substantial policy and practice changes have occurred since (e.g., the Tracey judgment and the ReSPECT process). The present review adds contemporary, in-depth qualitative synthesis, highlighting barriers to patient/family involvement, ethical tensions, and differences across care settings.

Design: Qualitative systematic review guided by the Cochrane Handbook chapter for qualitative evidence and reported according to PRISMA. Protocol registered on PROSPERO (CRD42022304689). Information sources and search: A specialist health librarian supported strategy development. Databases searched: PubMed, Embase (OVID), Emcare (OVID), CINAHL (EBSCO), and PsycInfo (ProQuest). Searches limited to studies published from 1 January 2012 to reflect current UK practice. Search terms combined concepts for resuscitation (including DNACPR), conversations/communication, experiences/attitudes, and UK location. Detailed strategies are provided in Appendix A. Eligibility criteria: Adults (patients about whom CPR decisions were made), their families, and HCPs (doctors, nurses, AHPs) in UK settings; exposure was recommendations about CPR; outcome was experiences of CPR decision-making conversations; qualitative empirical designs in peer-reviewed journals. Exclusions: conference abstracts, opinions, reviews, theses, purely quantitative studies, pediatric/neonatal studies, and non-UK settings. Study selection: After deduplication, titles/abstracts and then full texts were independently double-screened by two reviewers with adjudication by discussion and, if required, a third reviewer. A 10% sample was independently checked for reliability. Quality appraisal: Two reviewers independently assessed included studies using the CASP qualitative checklist. No studies were excluded based on quality. Data extraction and analysis: Two reviewers independently extracted data using a piloted pro forma. Thematic synthesis (line-by-line coding → descriptive themes → higher-level analytical themes) was undertaken. Reflexivity: The lead author (a Resuscitation HCP) reflected on potential bias, supported by supervision and focus on review aims.

- Study yield: 684 records identified; 443 screened; 106 full texts reviewed; 10 qualitative studies included. - Participant focus: 9 studies focused on HCPs; 1 on patients; none on families. HCP studies all included doctors; 3 included nurses; 1 included AHPs. Settings: 4 hospital-based, 4 general practice/community, 2 mixed. - Theme 1: Initiation of conversations - Who initiates: Perspectives varied. Many patients, GPs, and hospital clinicians thought GPs were well placed due to rapport; some GPs felt specialists were better due to condition-specific knowledge. Patients wanted a known and trusted HCP. - Prompts: Clinical deterioration, poor prognosis, transition to EOL/palliative care; occasionally improvement prompted review of prior DNACPR. Patients (and sometimes relatives) also initiated queries. Nurses sometimes prompted doctors to start discussions; organizational prompts (policies/EPR flags) existed. - Timing/location: Doctors triaged limited time toward patients at higher risk of imminent deterioration. Community discussions occurred in GP surgeries, home, or care homes; hospital discussions after ward rounds/MDTs or during visiting hours. GPs’ settings were viewed as more private and conducive; hospitals were seen as busy/overwhelming by patients. - Theme 2: Involvement of patients and families - Patient involvement: Some doctors avoided involving patients perceived as likely to be distressed or unlikely to benefit/understand, framing non-involvement as kindness. Patients in the single patient-focused study reported not being involved; several wanted involvement despite difficulty, while others deferred to clinicians or feared ‘tempting fate.’ - Family involvement: With patients lacking capacity, families supplied valuable contextual information. When patients had capacity, patients wanted family support, but clinicians often viewed families as potential sources of conflict or undue influence, leading to disagreements. - Managing reactions: Reactions were unpredictable; misunderstandings were common (e.g., equating DNACPR with cessation of all treatment). Serial conversations improved understanding but required time that HCPs felt they lacked. - Theme 3: Influences on content of conversations - Setting: Hospital clinicians emphasized acute interventions (ICU, ventilation); community clinicians emphasized chronic care plans, potential hospitalization, hospice, and place of death. - HCP views/emotions/context: Some HCPs questioned value in discussing CPR with relatively well patients; others limited scope to CPR to avoid overload. Some considered pre-operative discussions demotivating and inappropriate before potentially curative care. Patients felt too unwell during acute admissions to meaningfully engage or recall discussions. - Emotions/uncertainty: HCPs feared upsetting patients/families and losing trust; past negative experiences reduced confidence. Prognostic uncertainty complicated hypothetical discussions; clinicians preferred open approaches when uncertain and closed/persuasive approaches when trajectories were clear. - Theme 4: Conversation outcomes - Emotional impact: Distress was common for patients and families; sometimes relief and acceptance followed having a plan. HCPs often felt anxious or drained; good discussions felt rewarding. - Disagreement: Occurred when doctors recommended against CPR and patients/families disagreed, sometimes leading to second opinions or deference to patient autonomy against clinical judgment. - Incomplete conversations: Some discussions remained unfinished due to distress or ongoing disagreement, with plans to revisit later. - Cross-cutting ethical tension: HCPs consistently grappled with balancing respect for autonomy and non-maleficence (avoiding harm), influencing who to involve, how to communicate, and how to manage conflict. - Overall: Marked inconsistency exists in how CPR conversations occur, in patients’ desire for involvement, and in views on family roles. HCPs require training and pastoral support; patients and families benefit from CPR education and post-discussion support.

The review addresses how adult patients, families, and HCPs experience CPR decision-making conversations in UK settings. Findings show variability in initiation, timing, and setting; mixed preferences for involvement; and significant influence of clinician attitudes, emotions, and prognostic uncertainty on discussion content. These factors contribute to emotional burden, disagreements, and occasionally incomplete outcomes. The synthesis highlights a pervasive ethical tension between respecting autonomy and protecting from harm, mirroring broader literature on ACP/EOL discussions. Given legal and professional mandates for involvement, practice gaps remain, likely contributing to moral distress, burnout, and suboptimal patient/family experiences. The results underscore the need for system-level and educational interventions, better cross-setting collaboration, and clearer communication strategies to align practice with guidance and to reduce moral distress among HCPs.

Experiences of CPR decision-making conversations in the UK are complex and influenced by context, people, attitudes, and emotions. There is inconsistency in how these conversations occur. CPR discussions raise ethical challenges for HCPs. Further research is required to understand patient and family perspectives of these conversations. HCPs need training and pastoral support in conducting CPR discussions, and patients and families need education on CPR recommendations and emotional support after discussions.

- Limited participant perspectives: Only one included study captured patient experiences; none focused on family members. Most evidence reflects doctors’ views, with limited inclusion of nurses and AHPs. - Mixed quality reporting: Studies varied in reporting quality; some lacked explicit consideration of researcher–participant relationships and design justification. - Search scope constraints: UK-focused terms may have missed relevant studies not explicitly labeling UK locations. Studies where CPR was embedded within broader planning conversations might not have been captured if CPR was not specifically addressed. - Context specificity: Findings are specific to UK policy/practice and may not generalize internationally.