1088

Journal of Pain and Symptom Management

Vol. 49 No. 6 June 2015

Original Article

Advance Care Planning Does Not Adversely Affect Hope or Anxiety Among Patients With Advanced Cancer Michael J. Green, MD, MS, FACP, Jane R. Schubart, PhD, MS, MBA, Megan M. Whitehead, MSW, Elana Farace, PhD, Erik Lehman, MS, and Benjamin H. Levi, MD, PhD Penn State College of Medicine, Hershey, Pennsylvania, USA

Abstract Context. Many physicians avoid advance care planning (ACP) discussions because they worry such conversations will lead to psychological distress. Objectives. To investigate whether engaging in ACP using online planning tools adversely affects hope, hopelessness, or anxiety among patients with advanced cancer. Methods. Patients with advanced cancer and an estimated survival of two years or less (Intervention group) and a Control group were recruited at a tertiary care academic medical center (2007e2012) to engage in ACP using an online decision aid (‘‘Making Your Wishes Known’’). Pre/post and between-group comparisons were made, including hope (Herth Hope Index), hopelessness (Beck Hopelessness Scale), and anxiety (State Trait Anxiety Inventory). Secondary outcomes included ACP knowledge, self-determination, and satisfaction. Results. A total of 200 individuals completed the study. After engaging in ACP, there was no decline in hope or increase in hopelessness in either the Control or Intervention group. Anxiety was likewise unchanged in the Control group but decreased slightly in the Intervention group. Knowledge of ACP (% correct answers) increased in both the groups, but more so in the Intervention group (13% increase vs. 4%; P < 0.01). Self-determination increased slightly in both groups, and satisfaction with the ACP process was greater (P < 0.01) in the Intervention than Control group. Conclusion. Engaging in ACP with online planning tools increases knowledge without diminishing hope, increasing hopelessness, or inducing anxiety in patients with advanced cancer. Physicians need not avoid ACP out of concern for adversely affecting patients’ psychological well-being. J Pain Symptom Manage 2015;49:1088e1096. Ó 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved. Key Words Advance care planning, hope, anxiety, decision aids

Introduction In the U.S., it is well established that competent adults have the ethical and legal right to decide whether to receive life-sustaining medical treatment,1 and that advance care planning (ACP) can help ensure that patients receive treatment consistent with their values, goals, and preferences.2 Effective planning not only decreases the burden on family

Preliminary data from this research were presented at the Society of General Internal Medicine 34th Annual Meeting; May 7, 2011, Phoenix, AZ and at the International Conference on Communication in Healthcare, October 18, 2011, Chicago, IL.

Ó 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

decision makers3 but also increases the quality of care at the end of life.4 Furthermore, studies have shown that when individuals fail to plan for their medical future and then lose decision-making capacity, they often receive unwanted medical treatment,5e7 incur unnecessary costs,8,9 endure needless suffering,10,11 and leave loved ones feeling unduly burdened.12

Address correspondence to: Michael J. Green, MD, MS, FACP, Departments of Humanities and Medicine, Penn State College of Medicine, Room C1743, 500 University Drive, Hershey, PA 17033, USA. E-mail: [email protected] Accepted for publication: November 22, 2014.

0885-3924/$ - see front matter http://dx.doi.org/10.1016/j.jpainsymman.2014.11.293

Vol. 49 No. 6 June 2015

Hope, Anxiety, and Advance Care Planning

As such, physicians, ethicists, lawyers, and others advocate that individuals plan in advance by considering and documenting their wishes for medical treatment, so that they will be prepared if and when they become incapacitated. These documents, known as advance directives, typically describe what treatments people want and do not want, and identify who they would want to make medical decisions if unable to speak for themselves. Although advance directives have been widely promoted as a way to communicate a person’s wishes, they are not without criticism.13e15 Most individuals don’t complete these documents,16e19 even when they have life-threatening illnesses.20,21 Clinicians often worry that patients lack sufficient medical knowledge to ensure informed decision making or that their treatment preferences change over time,22 thus leading to advance directive documents that do not accurately represent the individuals’ wishes.15 Even more vexing, some physicians express concern that the process of ACP will diminish patients’ hope and raise their anxiety,23,24 leading these clinicians to avoid such conversations altogether. To overcome some of these barriers, we developed a multifaceted, interactive, online decision aid for ACP called ‘‘Making Your Wishes Known: Planning Your Medical Future’’ (MYWK, available online at www. makingyourwishesknown.com).25 This program was developed over several years by a multidisciplinary team with expertise in medicine, nursing, geriatrics, decision analysis, law, graphic design, and instructional design, and has been pilot tested among diverse groups of users. The MYWK tool was created to be user-friendly, accessible, and educational, including audio and video to assist users through the process of ACP. In doing so, it provides tailored education about common medical conditions that can result in decisional incapacity, as well as medical treatments often introduced in life or death situations. The MYWK aid guides an individual to choose a spokesperson, prioritize values and goals, match treatment options with priorities, and generate a printable advance directive document articulating one’s wishes. To help users through the complex process of making decisions under conditions of uncertainty, the program uses a decision aid based on multi-attribute utility theory, a framework that has individuals rank and rate the importance of various factors that influence choices.26,27 Multiple studies with diverse patient populations have demonstrated that MYWK is easy to use,28e30 effective for improving knowledge about end-of-life medical decisions,31,32 accurate in representing patients’ wishes,33 and can help health care providers make decisions on behalf of incapacitated patients.33,34

1089

Even with such data, some physicians remain wary of ACP, out of concern that the process of preparing for the end of one’s life might itself be harmful by diminishing hope and/or raising anxiety. Because this concern has not been systematically studied, we sought to determine whether hope is diminished or anxiety raised when patients with advanced cancer engage in ACP using either our multifaceted decision aid (MYWK) or a standard advance directive form.

Methods Study Design In the context of a broader investigation into the use of MYWK among patients with advanced cancer, we conducted a single-center, institutional, review boardeapproved, single-blind, randomized controlled trial of MYWK versus standard ACP. We hypothesized that 1) hope and anxiety would not be adversely affected by engaging in the process of ACP and 2) compared with standard ACP, using a computerbased decision aid (MYWK) would result in greater knowledge and self-determination, and higher levels of satisfaction.

Population and Screening Between 2007 and 2012, we recruited 200 patients with advanced cancer from the Penn State Milton S. Hershey Medical Center. Inclusion criteria included patients who had Stage IV or other poor prognosis cancer (acute myeloid leukemia; multiple myeloma; primary malignant brain tumors; and cancer of the esophagus, stomach, liver, lung, pancreas, or any metastases); had an anticipated life expectancy of two years or less; were English speaking; were not severely depressed; had sufficient cognitive capacity to complete the study; and could read at or above an eighth-grade level. Exclusion criteria included inability to read through Word 26 of the Wide-Range Achievement Test-3;35 a score 20 or greater on the Beck Depression Inventory-II;36 or a score of 23 or less on the Folstein Mini-Mental State Examination.37 Prior completion of an advance directive was not an exclusion criteria, although we did control for this in our analysis.

Recruitment Patients were recruited from clinics that treat adult patients with cancer, including hematology/oncology, radiation oncology, surgical oncology, otolaryngology, neurosurgery, and the breast cancer clinic. Clinic patients were prescreened for eligibility based on diagnosis and prognosis. The study coordinator reviewed the list of prescreened patients with their health care providers and eliminated anyone who failed to meet

1090

Green et al.

the eligibility criteria. Eligible patients were mailed a letter of introduction along with an overview of the study and a stamped, preaddressed opt-out form. Those who did not opt out were telephoned about two weeks later and invited to schedule an in-person study visit at the institutional General Clinical Research Center (GCRC). At the visit, informed consent was obtained and further eligibility screening was conducted. Individuals who screen-failed were given the opportunity to engage in ACP using either MYWK or standard ACP materials. All participants (including those who screen-failed) were offered a $25 gift card.

Randomization Consenting individuals were randomized into either an Intervention group or a Control group using a stratified permuted-block randomization approach.38 To maintain balance between groups and strata, the participants were stratified by gender and age (#65 or >65 years), with every four contiguous patients grouped together to form a block. Within each block, the order of the group assignments was randomly permutated using a random number generator.

Intervention Group and Control Group Descriptions At the GCRC, all participants were provided a laptop computer, headphones, and study instruments. Participants in the Intervention group engaged in ACP using MYWK, completing the intervention in one sitting. As described previously, MYWK is an online decision aid that provides education about conditions that commonly lead to loss of decisional capacity, and treatments typically used to sustain life in those circumstances. It then helps users think about and articulate their values, goals, and preferences for medical care and generates an advance directive document based on user responses. Detailed descriptions of this decision aid are available in previous publications.25,39,40 Participants in the Control group were provided educational materials developed by the American Hospital Association,41 along with a modified, online version of a state-approved advance directive form previously identified as being one of the best in the U.S.42 As with MYWK, participants completed the tool in one sitting. This tool asks users to indicate their treatment preferences if terminally ill or permanently unconscious, and to name a surrogate decision maker. Unlike MYWK, this advance directive did not include values clarification exercises, education about medical conditions/treatments, or a decision support component for assisting in decision making.

Patient Characteristics Patient characteristics were collected at baseline including age, gender, race, ethnicity, education, religious

Vol. 49 No. 6 June 2015

affiliation, marital status, employment status, recent hospitalization, number of comorbid illnesses, functional status (using the Eastern Cooperative Oncology Group [ECOG] Performance Status),43 prior computer use, and prior completion of an advance directive.

Primary Outcome Measures Patients’ hope, hopelessness, and anxiety were the primary outcomes, and were measured at baseline and immediately post-intervention. Although hope and hopelessness are related, they are conceptually distinct, and the construct of hopelessness44 has been particularly important in identifying those individuals with terminal illness who are at high risk for suicide. For this study, we use McClement and Chochinov’s definition of hope in patients with advanced cancer, which identifies hope as an active coping mechanism, protecting individuals from the experience of distress and suffering.45 For hopelessness, we follow Sullivan’s description of hopelessness as not simply the absence of hope, but rather a maladaptive attachment to a form of hope.46 Hopefulness was assessed using the Herth Hope Index,47,48 a 12-item instrument that has been validated for use in clinical and research settings.49 It has high criterion validity, internal consistency (Cronbach’s alpha ¼ 0.97),48 and test-retest reliability (0.91), and has been shown to be accurate across diverse populations, including individuals with life-threatening illnesses47 and cancer.50 Hopelessness was measured using a validated sevenitem version of the Beck Hopelessness Scale,51 adapted from the original52 to address items problematic for patients with terminal conditions, such as asking individuals to imagine their life 10 years from now. Among patients with terminal illness, the seven-item version has reliability and concurrent validity for measuring hopelessness, and is more strongly correlated with other measures of end-of-life despair than the original version.51 Anxiety was assessed using the state anxiety subscale of the State Trait Anxiety Inventory (STAI), a widely used, valid, and reliable instrument53 that reflects an individual’s temporal state anxiety. The subscale contains 20 statements rated on a four-point scale to describe feelings at a particular moment in time. The STAI has been used in varied clinical and research settings, including among patients with chronic and life-threatening medical conditions.54

Secondary Outcome Measures Knowledge of ACP was assessed at baseline and immediately post-intervention using an internally validated, 19-item multiple-choice instrument that measures understanding of issues related to ACP content and

Vol. 49 No. 6 June 2015

Hope, Anxiety, and Advance Care Planning

process. Self-determination was assessed at baseline and immediately post-intervention using a nine-item instrument adapted for this project from an empowerment survey created by Pellino et al.55 Satisfaction with the ACP process was assessed post-intervention using an instrument developed for this project that consists of 12 face-valid questions about the process of ACP across four domains: 1) quality of information provided, 2) helpfulness in values clarification, 3) helpfulness in decision making, and 4) helpfulness in communicating wishes.

Statistical Analysis Basic summary statistics were computed for all variables and were compared at baseline by study group, using a Chi-squared test for categorical variables including exact tests when cell counts were small, and a two-sample t-test or a Wilcoxon rank-sum test for continuous variables, depending on the distribution of the variable. Composite summary scores were calculated for the following measures, namely Anxiety, Herth Hope Index, Beck Hopelessness, Self-Determination, and Satisfaction with ACP. The calculation involved taking the sum of the nonmissing items and multiplying it by the total number of possible items divided by the number of available and non-missing items for the participant (sum  possible/available) to situate results on a common scale. If a participant responded to all items on an instrument, the composite score was simply the

1091

sum of the items because the weighting factor was 1. This method was applied for participants with at least 66% of the items answered to allow for a small number of missing items. A linear mixed effects model that included a factor for the study visit, a factor for the study group, and the interaction between the two factors was used to compare the mean change in the outcome variables from baseline to the post-intervention visit within and between study groups for all outcome variables measured at both time points. In compliance with an intent-to-treat philosophy, this model incorporated all baseline and post-intervention responses regardless of having complete data for both time points. For outcome variables measured only at the postintervention visit, a linear fixed effects model was used to compare the means between study groups. Baseline characteristics that differed between groups (i.e., having a prior advance directive or living will, and hours spent using a computer) were included as covariates in all models to adjust for variation. All analyses were performed using SAS software v. 9.2 (SAS Institute Inc., Cary, NC).

Results Patient Characteristics About 30 clinicians identified 1988 patients, of whom 1441 met eligibility criteria, and 200 were enrolled (Fig. 1). A total of 541 (27%) of identified

Referrals: 1,988

Returned opt-out form: 545 Declined participation: 530 Unable to reach: 217 Interested, but did not follow-up: 100 Poor health/in nursing facility: 180 Died before contact: 107 Not eligible: 37 No show for appointment: 72 Randomized: 200

Intervention (MYWK): 99

Control: 101

2 Lost to Follow-up Completed: 97

Completed: 101

1 Lost to Follow-up* 3 Revoked Consent

1 Lost to Follow-up* 1 Revoked Consent

Deceased 53

Alive 42

Deceased 50

Alive 47

Fig. 1. Flow of patients through study phases. *Of the four patients who revoked consent, two were too sick to continue and one enrolled in a clinical trial. No further information is available about the other individual or about the two lost to follow-up.

1092

Green et al.

Table 1 Reasons for Nonparticipation (N ¼ 1988 Referrals) Reason

n (%)

Not eligible for study Unable to reach Too ill to participate Died before contact Not English speaking Screen fail No show or did not follow-up Declined participation Returned opt-out form Unknowndjust not interested Study too time consuming Transportation issues Already have living will Feeling too ill Not interested in research Not ready to address end-of-life issues Discomfort with computers Declined consent

217 180 107 23 14 172

(11) (9) (5) (1) (

Advance Care Planning Does Not Adversely Affect Hope or Anxiety Among Patients With Advanced Cancer.

Many physicians avoid advance care planning (ACP) discussions because they worry such conversations will lead to psychological distress...
208KB Sizes 1 Downloads 10 Views