D. Andrew Edwards, Jr. Formal Opinion University Counsel No. 2003-F1 The State University of New York The Capitol Albany, New York 12224-0341
Dear Mr. Edwards:
You have asked whether a physician who is treating a legally incapacitated patient and who concludes that administration of cardiopulmonary resuscitation would be "medically futile"1 has the authority to enter a do-not-resuscitate order over the objection of the patient's surrogate or health care agent. We conclude that in these circumstances entry of a do-not-resuscitate order would violate Public Health Law § 2965 and is therefore not authorized.
Your question arises from an apparent conflict between the governing statutes and a pamphlet issued jointly in 1992 by the New York State Department of Health, the New York State Task Force on Life and the Law, the Medical Society of the State of New York, and the Hospital Association of New York State. See New York State Department of Health, et al., Do-Not-Resuscitate Orders: Questions and Answers for Health CareProfessionals (2d ed. 1992) [hereinafter Questions and Answers]. The pamphlet says that where cardiopulmonary resuscitation would be "medically futile," the attending physician may enter a do-not-resuscitate order without the consent of the patient's surrogate or health care agent where the judgment of futility is confirmed by a second physician.2
To the extent this advice indicates that a physician may enter a do-not-resuscitate order without obtaining the consent of a reasonably-available health care agent or surrogate, in our view, it is inconsistent with Public Health Law § 2965 and with regulations promulgated by the Department of Health. Though the views reflected in this aspect of the Questions and Answers publication have support within the medical community, they have been explicitly rejected by the Legislature.
Statutory Framework
In New York, do-not-resuscitate orders are governed by article 29-B of the Public Health Law, entitled "Orders Not to Resuscitate." As defined by that article, an "order not to resuscitate" is an order instructing medical personnel "not to attempt cardiopulmonary resuscitation in the event a patient suffers cardiac or respiratory arrest." Pub. Health Law § 2961(17). The order must be included in writing in the patient's chart, id. § 2962(2), and is subject to periodic review, id. § 2970. As the statutory definition suggests, an order not to resuscitate ordinarily is entered in anticipation of a future cardiac or respiratory arrest.3
The Legislature enacted article 29-B "to clarify and establish the rights and obligations of patients, their families, and health care providers regarding cardiopulmonary resuscitation and the issuance of orders not to resuscitate." Id. § 2960 (legislative findings and purpose). To this end, the article is exhaustive; that is, it identifies all the circumstances in which a physician is authorized to enter an order not to resuscitate a patient.4 See id. § 2962(2) ("It shall be lawful for the attending physician to issue an order not to resuscitate a patient, provided that the order has been issued pursuant to therequirements of this article.") (emphasis supplied).
Article 29-B makes the consent of the patient or the patient's agent or surrogate the principal source of the physician's power to enter an order not to resuscitate. Indeed, in its statement of "Legislative findings and purpose," the Legislature summarized its intent as follows: "The Legislature finds that . . . it is appropriate for an attending physician, in certain circumstances, to issue an order not to attempt cardiopulmonary resuscitation of a patient where appropriate consent hasbeen obtained." Id. § 2960 (emphasis supplied); see also id. § 2962(1) ("Every person admitted to a hospital shall be presumed to consent to the administration of cardiopulmonary resuscitation . . .unless there is consent to the issuance of an order not to resuscitate asprovided in this article.") (emphasis supplied).
Where a patient has the capacity to consent to entry of a do-not-resuscitate order, it is, of course, the patient whose consent is required. Id. § 2964. The patient's consent is effective even if the patient later becomes incapacitated. Id. § 2965(1)(b) (consent of surrogate or agent is not required for incapacitated patient where patient had consented to order not to resuscitate prior to losing capacity).
If the patient has not consented, a do-not-resuscitate order can be entered based upon the consent of a "health care agent." See id. § 2961(8).5 A decision by a health care agent, duly appointed by the patient, takes "priority over decisions by any other person, except the patient or as otherwise provided in the health care proxy." Id. § 2965(1)(c). Where the patient is incapacitated but has previously appointed a health care agent to make medical decisions on his or her behalf, the health care agent stands in the patient's shoes. See id. §§ 2962(5), 2982(1). Accordingly, in this setting, the consent of the health care agent, if one is available, "must be obtained prior to issuing an order not to resuscitate the patient." Id. § 2965(1)(a).
Where a patient is incapacitated and has not appointed a health care agent, the decision whether to consent to the entry of a do-not-resuscitate order falls next to a "surrogate." Id. §§ 2961(21), 2965(2). The categories of persons permitted to act as a surrogate are identified in section 2965(2) of the Public Health Law. A surrogate can be a committee or guardian appointed pursuant to article 17-A of the Surrogate's Court Procedure Act (concerning individuals with developmental disabilities); otherwise, it must be a near relative or close friend of the patient. Pub. Health Law § 2965(2). The person chosen to serve as the surrogate must be reasonably available, willing to make a decision about the issuance of an order not to resuscitate, and competent to make the decision. Id.
In keeping with the fact that the patient has not granted the surrogate the power to make decisions on his or her behalf, the surrogate plays a somewhat different role in the process than does the patient or the patient's health care agent. The patient or the patient's health care agent may consent to the entry of a do-not-resuscitate order without any particular medical finding by the physician. See id. §§ 2964, 2965(1). In contrast, a surrogate may only consent to the entry of a do-not-resuscitate order if there has been a determination by an attending physician, with the concurrence of another physician, that either: (1) the patient has a terminal condition; (2) the patient is permanently unconscious; (3) administration of cardiopulmonary resuscitation would be medically futile; or (4) resuscitation would impose an extraordinary burden on the patient given the patient's condition. Id. § 2965(3).
Although the statute makes the existence of one of these four circumstances a prerequisite to the entry of a do-not-resuscitate order where the order is based on the consent of a surrogate, none of these circumstances provides an independent basis for the entry of a do-not-resuscitate order over the objection of or without consent of the surrogate. If a surrogate is reasonably available and is willing and able to make a decision, the physician cannot dispense with the surrogate's consent, any more than the physician can dispense with the consent of a competent patient or health care agent; the surrogate's consent "must be obtained." Id. § 2965(1)(a).
Section 2966 does provide limited authority for the issuance of a do-not-resuscitate order for an incapacitated adult who did not consent to a do-not-resuscitate order prior to losing capacity, and for whom no health care agent or surrogate is reasonably available. In this, and only this, circumstance, section 2966 permits the attending physician to enter a do-not-resuscitate order without consent, either on the basis of a determination by two physicians that resuscitative efforts would be medically futile, or as directed by court order. Pub. Health Law §2966(1). But if a health care agent has been appointed, or a surrogate is reasonably available, then the consent of the health care agent or surrogate "must be obtained prior to issuing an order not to resuscitate." Id. § 2965(1)(a).
It is clear from these statutes that article 29-B does not permit physicians who conclude that resuscitative efforts would be "medically futile" to independently override the refusal of the surrogate or the health care agent to give consent or to enter an order without consulting (and obtaining the consent of) a reasonably-available health care agent or surrogate. Public Health Law § 2965 specifically requires the consent of the agent or surrogate and makes "medical futility" one of four medical determinations that will justify giving effect to the consent of a surrogate; it does not make medical futility an independent basis for the entry of a do-not-resuscitate order. Public Health Law § 2966 makes medical futility the basis for entry of a do-not-resuscitate order only where no health care agent or surrogate is reasonably available. Regulations promulgated by the Department of Health are fully consistent with this statutory scheme.See 10 NYCRR 405.43(f).
Moreover, the statutory scheme contemplates frank discussion between the physician and health care agent or surrogate about the patient's diagnosis and prognosis and the foreseeable risks and benefits of cardiopulmonary resuscitation. See Pub. Health Law § 2962(3) (requiring the attending physician to provide such information to the person giving consent). In most cases, such open discussion should result in agreement regarding issuance of an order not to resuscitate. Where it does not, a physician seeking to enter a do-not-resuscitate order over the objection of a health care agent or surrogate has two options. The physician must first bring the dispute before the hospital's dispute mediation system, pursuant to Public Health Law § 2972. However, persons appointed to participate in the dispute mediation system do not have the authority to determine whether a do-not-resuscitate order shall be issued. Pub. Health Law § 2972(5). If mediation does not resolve the dispute and the physician wishes to persist in his or her efforts to enter a do-not-resuscitate order over the objection of a health care agent or surrogate, the physician must commence a court action. Id. § 2973.
Legislative History
With respect to the question you pose, the plain meaning of article 29-B is consistent with the policies and purposes underlying its enactment.Cf. Insurance Co. of N. Am. v. ABB Power Generation, Inc., 91 N.Y.2d 180,186 (1997) ("the literal meaning of the [statute's] text should not be followed where it is patently inconsistent with the policies or purpose of the statute or where the result would be absurd"). Responding to "a need to clarify the rights and obligations of patients, their families, and health care providers," Pub. Health Law § 2960 (legislative findings and purpose), the Legislature, in enacting article 29-B, sought to balance the risk of medically inappropriate resuscitation with the need to safeguard the patient's interest in continued treatment.
There are divergent views in the medically community as to how best to accommodate these two interests. The view that physicians should be permitted to override the surrogate or health care agent in cases of medical futility has substantial support. See, e.g., American Medical Association, Guidelines for the Appropriate Use of Do-Not-ResuscitateOrders(1990) ("if, in the judgment of the treating physician, CPR would be futile, the treating physician may enter a do-not-resuscitate order into the patient's record"). Indeed, some commentators have argued that it is absurd to deny physicians the power to decide unilaterally whether to enter do-not-resuscitate orders in cases involving medical futility.See, e.g., Rita T. Layson Terrance McConnell, Must Consent AlwaysBe Obtained for a Do-Not-Resuscitate Order?, 156 Archives Internal Med. 2617, 2619-20 (Dec. 9/23, 1996). However, other commentators have argued cogently that decisions about resuscitation should be left to the patient or her surrogate. See, e.g., Stuart J. Youngner, Who Defines Futility?, 260 JAMA 2094 (Oct. 14, 1988); Paul C. Sorum, Limiting CardiopulmonaryResuscitation, 57 Alb. L. Rev. 617, 622-23 (1994).
The Legislature was cognizant of the concerns underlying these competing views when it adopted Public Health Law § 2965. In a report issued in April 1986 containing the proposed legislation that eventually became article 29-B of the Public Health Law, the New York State Task Force on Life and the Law6 identified "medically inappropriate resuscitation" and the entry of do-not-resuscitate orders without consent as the two principal problems driving the need for legislation. New York State Task Force on Life and the Law,Do-Not-Resuscitate Orders: The Proposed Legislation and Report of the NewYork State Task Force on Life and the Law 6-7 (1st ed. 1986). The report explained that where medical personnel attempt to resuscitate a patient who is certain to suffer repeated arrests in a short period before death occurs, the "outcome may be a more traumatic death rather than a prolongation of life." Id. at 7-8.
In spite of the Task Force's concerns about medically inappropriate resuscitation, the legislation proposed by the Task Force made the surrogate's consent a prerequisite to the entry of a do-not-resuscitate order. See id. at 37, 80-81. The Task Force's report explicitly stated: "If the attending physician believes that CPR is not medically appropriate for the patient, he must identify the proper surrogate to make a decision on the patient's behalf." Id. at 37. The report emphasized, "[w]hile the physician's advice and guidance to the surrogate are critical, the surrogate must act as an independent decision maker," and explained that "[t]he independence of the surrogate and physician provides greater protection for the patient." Id. Although the Task Force recognized that the proposed legislation "does not resolve the dilemma of resuscitation which yields greater pain or discomfort than benefit," it concluded that its proposed compromise appropriately reflected a presumption in favor of resuscitation "where the decision making process cannot adequately safeguard against the risk of a decision which does not serve the patient's interests in continued treatment." Id. at 45.
The Legislature subsequently adopted a statutory scheme that closely followed that recommended by the Task Force in its report. As the Task Force proposed, the surrogate's consent was made a prerequisite to the entry of a do-not-resuscitate order, even in cases of medical futility. Pub. Health Law § 2965(1)(a). Because the Legislature plainly considered the concerns that militate in favor of permitting physicians to override the surrogate in cases of medical futility, these concerns provide no grounds for overriding the careful compromise it reached in enacting article 29-B.
Conclusion
We conclude that where a patient is incapacitated and did not consent to the entry of a do-not-resuscitate order prior to becoming incapacitated, the Public Health Law requires the physician to obtain the consent of the patient's health care agent or surrogate before entering a do-not-resuscitate order, even if the physician concludes that administration of cardiopulmonary resuscitation would be "medically futile." Only where no health care agent was appointed by the patient and no competent surrogate is reasonably available and willing to make a decision may the physician enter a do-not-resuscitate order based upon medical futility without obtaining an agent's or surrogate's consent, and then only upon the concurrence of another physician that resuscitative efforts would be medically futile or after obtaining a court order. To dispute the decision of the health care agent or surrogate, the physician must proceed to mediation and, if the dispute remains unresolved, commence a court action.
Very truly yours,
ELIOT SPITZER, Attorney General
1 For purposes of the statutory scheme governing orders not to resuscitate, "cardiopulmonary resuscitation" is defined as "measures . . . to restore cardiac function or to support ventilation in the event of cardiac or respiratory arrest" and "medically futile" means either situations in which such measures will not be successful in restoring cardiac and respiratory function or situations in which "the patient will experience repeated arrest in a short period before death occurs." Pub. Health Law § 2961(4),(12).
2 Q: When can the attending physician enter a DNR order based on medical futility:
If the physician determines that CPR would be medically futile, the physician may enter a DNR order on that basis provided that he or she takes the following steps:
The physician must discuss the DNR order with the patient, agent, or surrogate, if possible;
The judgment of futility must be confirmed by a second physician authorized by the hospital to render concurring opinions on DNR matters; and
The physician must enter the order in the patient's chart and inform the patient, agent, or surrogate. The order will not require the consent of the agent or surrogate. . . .
Q: What if the health care agent or surrogate refuses to consent to a DNR order and the physician believes that CPR would be futile for the patient?
The attending physician must seek a second opinion. If the second physician concurs that CPR will be futile, as futility is defined by the law, and the concurrence is written in the chart, the attending physician may enter the order on grounds of futility, but must inform the agent or surrogate.
Questions and Answers, supra, at 28.
3 This opinion does not address whether, in the absence of a do-not-resuscitate order, a decision to forego or terminate resuscitative efforts made after an arrest occurs could ever be considered the legal equivalent of an order not to resuscitate under Public Health Law article 29-B. The Questions and Answers publication advises physicians that a decision made after an arrest occurs to forego resuscitative efforts is governed not by the statutes in article 29-B, but "by evolving standards of care, professional guidelines, and, when applicable, Health Department regulations." Questions and Answers, supra, at 1-2. The publication also instructs physicians that, in this post-arrest setting, a finding of "futility" will justify a decision to forego resuscitation. Id. at 2. You have advised that you do not seek an opinion on the accuracy of this aspect of the publication. Your exclusive concern is the situation where a physician enters a do-not-resuscitate order in anticipation of a future cardiac or respiratory arrest.
4 Article 29-B establishes enforceable standards of conduct for issuance of do-not-resuscitate orders, not merely recommended guidelines or a "safe harbor." See Pub. Health Law §§ 12, 12-b, 2973(3).
5 Appointment of and decisionmaking by a health care agent are governed by article 29-C of the Public Health Law.
6 The New York State Task Force on Life and the Law was established by Executive Order in 1984 with a mandate to study and recommend public policy on a number of issues arising from medical advances, including the decision-making process involved in the issuance of do-not-resuscitate orders. See Executive Order No. 56, 9 NYCRR 4.56 (Dec. 20, 1984). The Task Force was established with the Commissioner of Health as its chair and members from the medical, ethical, legal and religious communities, as well as interested laypersons. See id.