Intended for healthcare professionals


Views of elderly patients and their relatives on cardiopulmonary resuscitation

BMJ 1994; 308 doi: (Published 25 June 1994) Cite this as: BMJ 1994;308:1677
  1. R Morgan,
  2. D King,
  3. C Prajapati,
  4. J Rowe
  1. Department of Medicine for the Elderly, Broadgreen Hospital, Liverpool L14 3LB
  1. Correspondence to: Dr King.
  • Accepted 16 March 1994

Proposed guidelines for withholding cardiopulmonary resuscitation suggest that “when appropriate, consultations with patients or their relatives, or both, should be considered before decisions are made.”1 In practice when a decision is made not to resuscitate an elderly patient the patient is rarely consulted even if he or she is mentally competent. Relatives are more likely to be consulted. We assessed the views of both elderly patients and their relatives on this subject.

Methods and results

We interviewed 100 alert patients (abbreviated mental test score at least 8 out of 10: mean age 80.4 years; 62 women) and their legal next of kin. All were interviewed individually in private by a doctor unknown to them. The main diagnoses were; angina (28), chest infection (22), heart failure (8), stroke (8), cancer (6), miscellaneous (28). The median length of stay was 22 days (range 0-89). Patients were aware of their diagnoses and were interviewed when the acute illness was over. No patient was imminently expected to die.

The procedure for cardiopulmonary resuscitation was explained to everybody interviewed. Criteria associated with a good outcome were explained as well as its futility in certain cases and the fact that it might result in dependency before eventual death. Data were analysed using McNemar's X2 test.

Most patients and relatives (89 and 88) thought that doctors should discuss plans for cardiopulmonary resuscitation with them. Interestingly, 34 patients felt that discussion on their resuscitation status should be limited to themselves while 37 relatives thought that this should involve them but not the patient. Age was considered not to be important in cardiopulmonary resuscitation decisions (60 patients, 70 relatives). More patients than relatives thought that people with physical and mental disability should not be resuscitated (table). No relation existed between a patient's diagnosis, sex, and their views on resuscitation. Most people did not feel uncomfortable discussing their resuscitation status (98) or that of their relatives (95) but welcomed it: “I am grateful to be asked my opinion on such an important topic.” Thirteen patients had “do not resuscitate” in their casenotes. Eight of these wanted to undergo cardiopulmonary resuscitation whereas 35 of the remaining 87 (who would have been resuscitated in the event of an arrest) did not.

Responses given by pairs of patients and relatives to questions about

View this table:


Although cardiopulmonary resuscitation is available in all hospitals, not all patients are suitable candidates. In particular, people with pneumonia, hypotension, and cancer have a poor outcome, though age does not necessarily affect outcome.2

Recent studies have shown that physicians rarely make “do not resuscitate” decisions, although geriatricians are more likely to do so.2,3 They rarely discuss these decisions with patients, however, possibly because they feel uncomfortable doing so. Stolman et al showed that 30% of doctors felt uncomfortable discussing resuscitation plans with patients whereas only 9% felt uncomfortable doing so with relatives.4 Schade and Muslin found that discussion led to added distress for patients, or patients forgot about the discussion altogether,5 though their patients were young and dying. These findings contrast with ours, in which most patients welcomed discussion. Possibly our older patients may have come to terms with their own mortality, although none was in imminent danger of death.

There were many reasons why 37 relatives wanted their “loved ones” excluded from discussions about resuscitation, but it is unethical to discuss the issue with relatives but not with a mentally competent patient. The lack of concordance in 43 cases between the decision to resuscitate or not and the patient's wishes raises ethical dilemmas. Eight of 13 patients with do not resuscitate orders wanted cardiopulmonary resuscitation even though six had carcinomatosis and despite discussions on outcome and life expectancy. Legally if a patient requests cardiopulmonary resuscitation it should be provided and not doing so could result in litigation (personal communications, Medical Protection Society and Medical Defence Union). If patients' views had been taken into account routinely then 27 possible resuscitation procedures would be avoided, with streamlining of the service in terms both of maintaining dignity and saving resources. The elderly welcome discussion on cardiopulmonary resuscitation. Although this may produce ethical dilemmas it will more often ensure more appropriate decisions.

We thank Mr D Irvine, audit assistant, for his help and Dr D Ashby, medical statistician, for her advice.