In August 2019 the Ontario Superior Court released its judgment in the case of Wawrzyniak v Livingstone, https://www.canlii.org/en/on/onsc/doc/2019/2019onsc4900/2019onsc4900.html

The events at the heart of this story occurred in 2008 and were widely reported at the time – and gave rise to a HPARB decision that was different from two previous CPSO judgments.

Here is a summary of the facts of the case:

Ms. W’s father, Douglas DeGuerre, age 88, had been transferred from long term care to acute care with multiple co-morbidities including diabetes, severe cardiac and respiratory disease, severe vascular ischemia, gangrene, multiple open necrotic lesions and end-stage renal disease. He was seen by vascular and orthopedic surgeons, neither of whom had any treatment to offer that would be of any benefit. He was admitted to internal medicine for supportive treatment of pain and multiorgan failure. His attending physician expected a slow decline toward death in spite of the treatments provided.

As usual, the Hospital [Sunnybrook] had a policy by which a standing order existed to initiate CPR in case of cardiac or respiratory arrest, unless a specific instruction was given to the contrary. There were several discussions with Ms. W, who was insistent that her father be deemed “full code” and refused palliative care. On day seven of his admission, Mr. DeGuerre underwent bilateral amputation of his lower limbs for the purpose of pain management, spent a few days in the ICU post-operatively, was stabilized and transferred back to the medical ward with daily follow-up from the rapid response team.

On September 22, 2008, both Dr. L (the most responsible physician) and Dr. C (a critical care physician on the rapid response team) separately assessed Mr. DeGuerre and determined that he was close to death. He was moribund, cachectic, confused, unable to clear his own secretions, with uncontrollable pain and signs of spreading gangrene. The physicians determined that he would not benefit from aggressive ICU care or CPR, which would only increase his suffering. A third intensitivist agreed. They therefore co-signed a DNR order in his chart. Dr. C attempted to contact Ms. W immediately thereafter by phone to inform her, but was unsuccessful in reaching her.   

Later that afternoon, Mr. DeGuerre’s condition abruptly deteriorated as Ms. W was visiting and he went into respiratory distress. At this time, it was Ms. W’s understanding that pursuant to her direction, her father would be treated with a “full code” resuscitation. Ms. W demanded that all resuscitative efforts be made. Dr. C arrived shortly thereafter and advised Ms. W that her father’s death was inevitable and CPR would only lead to greater pain and suffering. Ms. W was very upset that CPR was not administered and attempted to assist her father herself. Her efforts were unsuccessful and Mr. DeGuerre died. The circumstances surrounding Mr. DeGuerre’s death were extremely upsetting for everyone involved. 

Borden Ladner Gervais https://blg.com/en/News-And-Publications/Publication_5835

Ms. W twice complained to the CPSO Inquiries, Complaints and Reports Committee which twice ruled that the physicians concerned had acted appropriately.  Ms. W appealed to the Health professions Appeal and review Board (HPARB) which ruled as follows:

The Board found that the Respondents were required to obtain consent from the Applicant as SDM before replacing the “Full Code” order with the DNR order. In the absence of consent, the Respondents were required to follow the dispute dissolution procedure under the HCCA. 

https://www.canlii.org/en/on/onhparb/doc/2014/2014canlii49888/2014canlii49888.html

This ruling from 2014 – gave rise to a highly ambiguous and widely contested CPSO policy on end of life care in 2015.

The relevant section is as follows:

15.Physicians must not unilaterally make a decision regarding a no-CPR order.

1. Before writing a no-CPR order in the patient’s record, physicians must inform the patient and/or substitute decision-maker that the order will be written and the reasons why.¹³
2. If the patient or substitute decision-maker disagrees and insists that CPR be provided, physicians must engage in the conflict resolution process as outlined in this policy and must not write the no-CPR order while conflict resolution is underway.
3. If the patient experiences cardiac or respiratory arrest while conflict resolution is underway regarding the writing of a no-CPR order, physicians must provide all resuscitative efforts required by the standard of care, which may include CPR.¹⁴

https://www.cpso.on.ca/Physicians/Policies-Guidance/Policies/Planning-for-and-Providing-Quality-End-of-Life-Car#Policy

In September though the CPSO issued and Advice to the Profession companion resource

https://www.cpso.on.ca/Physicians/Policies-Guidance/Policies/Planning-for-and-Providing-Quality-End-of-Life-Car/Advice-to-the-Profession-Planning-for-and-Providin

Does the College require that consent be obtained before writing a no-CPR order?

No, the College does not require that consent be obtained prior to writing a no-CPR order. Rather, the policy emphasizes the importance of good and early communication that aims to avoid last minute decisions and intractable disagreements. The policy also does not require that physicians propose a no-CPR order be written. Instead, it requires that physicians inform the patient and/or substitute decision-maker that the order be written and the reasons why prior to writing it. Physicians can be fairly straightforward and directive in doing so, while recognizing that this may be particularly difficult news for the patient and/or their family to hear. Only if the patient or substitute decision-maker disagrees upon learning that the order will be written must physicians engage in a conflict resolution process to try and find consensus.

What happens if there is disagreement about writing the no-CPR order?

If once learning that a no-CPR order will be written, the patient and/or substitute decision-maker disagrees and insists that CPR be provided, the policy requires physicians to engage in conflict resolution as outlined in the policy. During this time, physicians must not write a no-CPR order. However, if the patient’s condition deteriorates and they experience a cardiac or respiratory event while conflict resolution is underway, physicians are permitted to make a bedside determination about which resuscitative efforts, including CPR, to provide and are only required to provide those that are within the standard of care.

Advice to the profession: Planning for and Providing Quality End-of-Life Care. (link above)

A straightforward reading of these two documents seems to provide contradictory advice.  The September 2019 Advice however is based directly on the August 2019 Ontario Superior Court Decision in  Wawrzyniak v Livingstone, (link above.)

Justice Cavanagh said:

200 After they made the medical decision that Mr. DeGuerre’s death was imminent and that CPR would almost certainly not benefit him and would only cause him harm, the defendants took steps to record and enforce their medical decision. These steps involved writing the DNR order (to preclude the otherwise automatic initiation of CPR according to Sunnybrook’s policy) and, in Dr. Chapman’s case, deciding not to offer CPR at Mr. DeGuerre’s bedside that evening. These steps are not “anything that is done” for a health related purpose within the meaning of “treatment” in the HCCA. They were processes followed by Dr. Chapman and Dr. Livingstone to comply with the applicable standard of care by deciding whether administration of CPR to Mr. DeGuerre on September 22, 2008 would be medically inappropriate and whether a medical decision should be made not be offer CPR to Mr. DeGuerre as a treatment option.

The key features here are:

1 A clinical determination by the physicians that the resuscitation attempt would almost certainly not benefit (and would be likely to cause harm to) the patient.

2. Writing the DNR order.

So, writing a DNR order when a resuscitation attempt is almost certain not to benefit the patient is not a `treatment` under the Health Care Consent Act.  And therefore, does not need consent.

In considering whether writing a DNR order, when CPR had previously been part of the plan, is a component of a “plan of treatment“ (under the HCCA)  the judge wrote:

225 A person’s health condition during the course of medical care for a serious illness is likely to change over time and a physician must adjust the treatment options which are offered to a person during the course of care to reflect his or her clinical judgment. This may involve proposing additional treatments, recommending the withholding or withdrawal of a treatment that was and is being offered (or that has already been administered), or deciding that a treatment that had been offered as part of a plan of treatment will almost certainly not benefit a patient and making the medical decision that this treatment will no longer be offered.

Again with the conclusion that consent was not required.

There is much more in this judgment.  Judge Cavanagh makes an important distinction between withdrawing and withholding (or not offering) treatment.

272 In my view, there is a vital distinction between a medical decision to “withdraw” a life support treatment which was already administered and which the patient has been receiving (that may involve interventions requiring physical interaction with the patient’s body (as was the case in Rasouli, Golubchuk and, in respect of removal of the ventilator, Sweiss)) and a decision not to offer (or, put another way, to withhold) a life support treatment such as CPR that would not involve any physical interaction with the patient’s body.

There are two things going on here.  The first is that a clinical judgment must follow the facts of the case – and a patient`s condition may deteriorate over time resulting in a clinical decision to no longer offer (withdraw) a particular treatment or intervention.  That decision is clinical and does not require consent.  However, acting on the decision may require consent – depending on the processes required to enact the decision.  The decision to write a DNR order because the patient would almost certainly not benefit does not require consent because no action or interference with the patient`s body is required – DNR is an order not to act — No CPR.  However, a decision to withdraw a treatment may require an action that interferes with, or touches the patient.  This would require consent.   This is why, according to Cavanagh the Rasouli decision went the way it did.  Consent was required to withdraw a life sustaining treatment because the process of withdrawal involved interaction with the patient`s body.

Conclusion

The CPSO has gone part way in following the judgment but has stopped short – in a somewhat confusing fashion.

The decision was clear that the requirements are not the same – consent is not required prior to writing a no-CPR order and physicians are only obliged to provide CPR when doing so is within the standard of care.

So, physicians are not obliged to provide CPR when it is not part of the standard of care.  The decision is actually clear that physicians are not obliged to offer a treatment that is outside of the standard of care.  Indeed, reading the judgment seems to suggest that physicians should not offer treatments that are outside of the standard of care.  (And I would certainly support that view.)

the College does not require that consent be obtained prior to writing a no-CPR order. Rather, the policy emphasizes the importance of good and early communication that aims to avoid last minute decisions and intractable disagreements. The policy also does not require that physicians propose a no-CPR order be written.

So far so good.

 Instead, it requires that physicians inform the patient and/or substitute decision-maker that the order be written and the reasons why prior to writing it. Physicians can be fairly straightforward and directive in doing so… (Emphasis added.)

Here is the tricky part.  The physician determines that a resuscitation attempt would be outside of the standard of care and therefore will not be offered.  No consent is required to writing a no CPR or DNR order.  However, the patient SDM must be informed – and unless or until the patient or SDM agrees the order cannot be written.  In practice that is exactly the same as requiring a patient or SDM’s consent to a DNR.  The only change is that while agreement is being pursued, e.g. through some sort of conflict resolution period, if the patient suffers a cardiac arrest during that period the physician at the bedside can make a clinical judgment as to the appropriateness of a resuscitation attempt.  Whereas before the CPSO instruction was to attempt to administer CPR.  The approach adopted by the CPSO is not consistent with the judgment.

I would suggest the following:

Physicians have an obligation not to provide and not to offer non-beneficial treatments or interventions.  Providing those non-beneficial treatments or proposing them is not part of the standard of care.  Not offering non-beneficial care is good practice.  When a resuscitation attempt would not be beneficial to the patient it should not be offered – instead the physician should write a DNR or No-CPR order and the patient or substitute decision-maker should be informed – with the reasons for the decision.  This is not a matter for negotiation or one that requires consent.  In the event of disagreement there should be a conflict resolution process.  If a patient has a cardiac arrest while the conflict is, as yet, unresolved the physician at the bedside should make a clinical judgment as to the appropriateness of initiating CPR.

There are clear ramifications for health care organizations.  There should be policies and practices that support physicians as they make clear clinical judgments that a resuscitation attempt is likely to benefit the patient, may benefit the patient, or is almost certainly unlikely to benefit the patient.   If the clinical judgment is that a resuscitation attempt will almost certainly not benefit the patient there should be policies requiring the physician to write a DNR or No-CPR order and to inform the patient or substitute decision-maker.  Organizations should develop and make available conflict resolution processes consistent with CPSO policy:

Managing Conflicts

25. In order to minimize and/or resolve conflict that can arise regarding treatment decisions, physicians must:
    1. communicate clearly, patiently, and in a timely manner information regarding the patient’s diagnosis and/or prognosis, treatment options and assessments of those options, and the availability of supportive services (e.g., social work, spiritual care, etc.) and palliative care resources;
    2. identify misinformation and/or misunderstandings that might be causing the conflict and take reasonable steps to ensure that these are corrected and that questions are answered;
    3. offer referral to another professional with expertise in the relevant area and facilitate obtaining a second opinion, as appropriate;
    4. offer consultation with an ethicist or ethics committee, as appropriate and available;
    5. where appropriate, seek legal advice regarding mediation, adjudication or arbitration processes that are available; and
    6. take reasonable steps to transfer the care of the patient to another facility or health care provider as a last resort and only when all appropriate and available methods of resolving conflict have been exhausted.²⁴

https://www.cpso.on.ca/Physicians/Policies-Guidance/Policies/Planning-for-and-Providing-Quality-End-of-Life-Car

And finally, organizations should take steps to educate staff and the communities they serve on this approach to end-of-life care.

Rob Butcher November 2019