Advance directives in an intensive care unit: Experiences and recommendations of critical care nurses and physicians.
To understand the benefits and limitations of advance directives to direct decisions about the care of critically ill patients in the Intensive Care Unit (ICU).
Qualitative ethnographic study describing the perspectives of a purposive sample of critical care nurses and physicians.
Twenty-two bed open medical-surgical ICU in a large community hospital in the Midwestern United States where staff care for patients at high risk of death (defined as diagnosis of sepsis, late stages of cancer, persistent vegetative state, organ failure in two or more systems and with a mortality rate 50% or more).
Purposive sample of registered nurses (n=14), attending physicians (n=7) and fellow physicians (n=3) with advanced training in critical care medicine, consenting to provide information about the use of advance directives in the care of their patients at high risk for death in this ICU.
Informal interviews while registered nurse (n=14) and physician (n=10) participants were working in ICU. Data were recorded in the researcher's journal using verbatim quotes. A formal semi-structured follow-up interview was conducted at a location away from the hospital with a purposive sample of registered nurses (n=5) and attending physicians (n=7) who had participated in an earlier informal interview. Formal interviews were audio-recorded, transcribed verbatim and analyzed into themes describing content with regard to the participants' clinical experiences with use of advanced directives in ICU and recommendations for improving advanced care planning for ICU patients.
From the perspective of all of the participants, advance directives had an extremely limited ability to direct decisions about the care of critically ill patients in ICU. Participants felt that the negative aspects of such directives included 1) difficulty communicating their contents across settings and providers therefore could not prevent unwanted life support, 2) lacking the specificity required to address clinical situations faced by patients in real life, 3) employing vague and confusing terminology, which could not be interpreted for decision-making, 4) physicians had to identify a terminal prognosis for the directive to be valid, which was present for exceedingly few patients during the study period, 5) impressions that litigious, difficult families overturned such patient instructions, 6) patients/surrogate decision makers did not choose to share directives in some cases, and 7) patients and families who did not want to discuss death and dying would not elect to complete directives. Although participants felt that advance directives were a flawed approach to communicating patient wishes, they agreed that these previous discussions between patients and their families had several benefits, which included prompting discussions about end of life care or quality of life as well as providing participants with the opportunity to use the advance directive to decrease moral burden by shifting responsibility for choosing end-of-life options away from family members.
The author reports that while participants have identified many known issues with advance directives, one new finding was that a patient hesitated to share her advance directive for fear of physicians prematurely stopping life support. The author suggested that advance directives may help settle disagreements about the appropriate use of life support treatments before legal solutions are sought, although restricted the benefit of such clarity to a small proportion of patients who have either persistent vegetative state or diagnosed terminal illness. Participants in the study suggested that preparation of advance directives could either benefit or hinder family discussions. They offered examples of families who had beneficial family discussions in the process of writing advance directives but also offered examples of cases where advance directives were written in order to avoid such conversations. Advance directives were used by participants to reframe the decision to stop using life support (which would result in death) as respecting the patient's expressed wishes and thereby alleviate the (perceived) moral burden of grieving families. Finally, the author recognized that characteristics of the sample might limit the transferability of findings: a relatively homogeneous sample of white nurses and physicians, a community ICU with mostly medical patients, and attending physicians with pulmonary medicine certification.
This manuscript is one of four recently published reports of an ethnographic study undertaken during Dr. Gutierrez's doctoral dissertation in order to examine how prognostic information was communicated to inform end-of-life decisions in ICU (Gutierrez, 2010). The research expands on her previous research examining moral distress among critical care nurses. As a critical care nurse herself, Gutierrez was both an "insider" to the ICU culture as well as an "outsider" to the ICU where the study was conducted. She used both informal and formal interviewing to accomplish data collection, which allowed her to benefit from expertise of professionals during their clinical practice and yet permitted more considered reflection on their practice during the later formal interview.
In this study, as in many others that investigate communication around end-of-life decision-making, the focus on decision-making may be a misnomer. One must have alternative options to be truly faced with a decision, therefore what is generally referred to as the "end-of-life decision" is actually a discrete choice within a larger decision problem about whether to use life support to delay death, or to provide uniquely comfort care to manage the dying patient. Both the researcher's analysis and the participant quotes reflect a more focused aim: to communicate sufficiently well to gain (patient and) family support for a particular course of action (comfort care leading to end of life) that is also already preferred by the healthcare team. Indeed, when physicians refrain from communicating uncertain prognostic information, families are not invited to consider the larger decision problem. There are several themes that support the observation that participants in Gutierrez's study may have been seeking family agreement with comfort care through the advance directives. First, advanced directives were not celebrated by participants for their ability to secure aggressive life support treatment for patients who wanted to live despite their critical illness, but were criticized for their inability to "prevent unwanted aggressive treatments that prolong dying". Next, family members were observed to struggle with "letting go," which led to demands to continue life support treatment and threats of litigation; such desperate measures might not be employed by families if there was truly a choice with multiple viable options being presented. Participants had already made a choice (to focus on comfort at end of life) and were critical of families' inability to come to the same conclusion despite not being introduced to the decision problem earlier. Therefore, in the research under review, failure to consider the end-of-life decision more broadly may have in turn influenced attention to the data collected, the analysis, and the outcomes of the study.
The research clearly supports the impression that physicians hold the power when it comes to end-of-life decision-making. It would be interesting to investigate further whether there was justified concern that advanced directives might be used to frame choice when the physician was ready to "throw in the towel too early". In a related study, Schenker and colleagues (2012) reported that the strength of the physician's belief that life support should be withdrawn or withheld was the single significant predictor of offering comfort care as an option, where the latter option was offered to less than half of families despite a mortality rate in their study of 72%. Therefore, at least half of the families in that study were not invited to consider alternate options to life support for the patient. Such omissions do not seem to meet current professional and legal standards for informed consent. Schenker and colleagues go on to argue that a failure to discuss available treatment options is problematic because families cannot then prepare emotionally for potential withholding or withdrawal of the treatment later, and may experience death as a sudden event, which potentially contributes to complicated grief among family members. Future investigations could investigate how power could be equalized among healthcare teams and families to promote improved communication and decision-making.
Advance directives may be a helpful adjunct to advance care planning as a process of deliberation between patients and families in advance of critical illness. The concept "advance directive" is advanced here as a relatively problematic vehicle for "end-of-life decision-making". Unfortunately, this is not a new perspective. Halpern (2012) also advances some convincing critique of the structure of advance directives, but suggests in detail how they can be engineered in both content and delivery to overcome well-described cognitive biases, which impede decision-making at the end of life. More work is needed to understand how members of the healthcare team in ICU should integrate the patient's message from an advance directive in ongoing dialogue about whether the use of life support technologies are wanted in care at the end of life. The current challenge seems to be in identifying how physicians, nurses, families and patients in critical care settings are best supported to engage in explicit compassionate deliberation for this high stakes decision in the context of uncertain outcomes and complex health challenges.
Gutierrez, K. M. (2010). Communication of prognostic information in an ICU at end of life: Practices among and between nurses, physicians and family members (Doctoral Dissertation). University of Minnesota, MN. Retrieved from ProQuest Dissertations & Theses (3411843).
Halpern, S.D. (2012). Shaping end-of-life care: Behavioural economics and advance directives. Seminars in Respiratory and Critical Care Medicine, 33, 393-400.
Schenker, Y., Tiver, G.A., Hong, S.Y., & White, D.B. (2012). Association between physicians' beliefs and the option of comfort care for critically ill patients. Intensive Care Medicine, 38,1607-1615.
Jennifer Kryworuchko, PhD, RN, CNCC(C)
Assistant Professor, College of Nursing, University of Saskatchewan
|Printer friendly Cite/link Email Feedback|
|Date:||Dec 22, 2012|
|Previous Article:||Critical thinking.|
|Next Article:||Seasonal flu shot.|