CPR decision making in the hospital setting

In an attempt to improve completion of cardiopulmonary resuscitation (CPR) forms in acute hospitals, many district health boards (DHBs) have improved the clarity of documentation. Current forms within our DHB were introduced in 2015, and when adequately completed, allow documentation of various levels of care/intervention to be easily communicated between health professionals.

We conducted a small audit on completion of CPR forms within our older adult rehabilitation wards (mean age 88 years) before introduction of the new CPR forms, and found that 28 of 41 (68%) patients had documented CPR decisions (unpublished audit). Twenty-three of these (82%) were completed as ‘not for resuscitation’, however, only seven (30%) had been discussed with patients or family members.

We recently conducted a repeat audit of CPR documentation to assess whether improved forms has led to improved documentation on one rehabilitation ward. Within this audit we also had information regarding patient frailty status. Fifty-six notes were reviewed (mean age 87 years, 26 (46%) of which were frail). CPR forms were completed in 35 (63%). We found no significant differences in age, sex, frailty level (Reported Edmonton Frailty Scale1), medication number or Charlson Comorbidity Index2 (mean 1.7) between those having completed documents and those not (Chi-squared test for categorical data, T-test for comparing means between groups): ie, CPR completion was not preferentially documented in those older, frailer or with greater comorbidities. Further, of those that did have CPR documentation, 12 (34%) were for ‘full CPR’ resuscitation, and 23 (66%) for active resuscitation but with variable limits documented. Of those who were for full CPR, only one patient had this discussed with them. A limitation of this study was the very small sample size.

We wonder whether it is easier to assume the default position of documenting a patient for full CPR rather than engaging in potentially difficult conversations. This however, may not be in the patients’ best interests or wishes. Of those for active but limited resuscitation, a greater number of discussions occurred with patients at 11 (48%), but even this was less than ideal. When specifically looking at the frail subgroup (n=26), 18 (69%) had documents completed and six of these were for full resuscitation. None of these decisions were discussed with the patient. We do not believe these decisions were necessarily incorrect. We do believe that good medical practice should include discussions about these issues with patients, particularly so in frail, older adults.

These results are unsurprising, and are consistent with overseas studies. In the 2012 UK National Confidential Enquiry into Patient Outcome and Death (NCEPOD) report,3 only 10% of patients had CPR documentation at admission, with CPR status considered in 13% at the time of first consultant review. Brown et al reviewed clinical notes of 99 older adults in South Australia, finding only one third had current documentation, which was frequently ambiguous. Decisions regarding CPR have been particularly prominent in the UK, where doctors now have a legal duty to consult with and inform patients if placing a do not resuscitate (DNR) order. This landmark decision by England’s Court of Appeal was made after the family of Janet Tracey, who died in 2011, went to judicial review after neither she nor the family were consulted when a DNR order was placed.4 In the UK at least, the issue that such conversations may cause distress is no longer a sufficient (legal) reason not to engage with the patient or appropriate family members. Aside from the potential legal ramifications, there is overseas evidence that older adults wish to be involved with, and informed of resuscitation decisions.5,6

Completion of CPR and other related documents rely on much more than an adequately constructed document, although this is obviously the first step. Decisions are frequently complex and emotionally challenging. This makes it time consuming for already busy doctors. What are the barriers to adequate completion in New Zealand? Do we equip doctors to perform this task? Sulmasy et al7 suggests not: investigators found physician confidence about discussing DNR orders was lower than for other medical discussions. Addressing these issues will go some way to improving this important aspect of health care. Meanwhile, Barbara Hayes8 proposes a clinical model on which to base these ethical difficult discussions and decisions. She recommends three steps: 1) judging patient’s illness and expected response to CPR; 2) considering the ethical implications of CPR/ no CPR; 3) discussion with patient and/or family. Further details of her discussion goals can be found in her publication.

This problem is not going away: lifespan is increasing, and numbers of the very old are increasing faster than other age groups.9 In patients over 85 years, 29% of deaths occur in acute public hospitals in New Zealand.10 Part of our role as doctors is to ensure patients die with dignity and a full grasp of the information relevant to the patient’s/family’s wishes will help in this regard. We believe improving CPR documentation will assist in achieving this goal.

‍