Perceptions of the Term Palliative Care. ANN MORSTAD BOLDT, M.S., 1 FOUZA YUSUF, M.S., M.P.H., 1 and BRUCE P. HIMELSTEIN, M.D.

Transcription

1 JOURNAL OF PALLIATIVE MEDICINE Volume 9, Number 5, 2006 Mary Ann Liebert, Inc. Perceptions of the Term Palliative Care ANN MORSTAD BOLDT, M.S., 1 FOUZA YUSUF, M.S., M.P.H., 1 and BRUCE P. HIMELSTEIN, M.D. 2 ABSTRACT Objectives: To assess parents and health care providers perceptions of the name and description of a pediatric palliative care (PPC) program. Methods: Survey conducted at three pediatric health care sites; asked respondents (parents and staff) about their likelihood to use a program identified either as palliative care or supportive care, as well as their understanding and feelings about the program before and after reading a program description. Results: Response rate was 89% (195/220); 184 were considered evaluable. Parent respondents in the supportive care group scored significantly higher (Mann-Whitney test, p 0.003) on likelihood to use program (mean score, 4.22, n 60) than those in the palliative care group (mean score, 3.58, n 45) before each read the program description. However, this group difference disappeared (p 0.582) after reading the description (mean scores 4.50, 4.38, respectively; n 48, n 40, respectively). The name palliative care evoked more negative emotions compared to the supportive care name in parents, and reading the program description led to more positive feelings. In staff, reading the program description significantly increased likelihood to use the program for those in the Palliative Care group only (4.22 to 4.44; p 0.05; n 41). Staff also had more positive feelings about the program called supportive care, and rated this name best most frequently. Conclusion: Better definition of and explanation to families and health care providers about what palliative care programs offer may improve perceptions about palliative care and increase program utilization. INTRODUCTION PALLIATIVE CARE FOR CHILDREN is the art and science of family-centered care aimed at enhancing quality of life and minimizing suffering. 1 The American Academy of Pediatrics (AAP) espoused an integrated model of palliative care, including relief of symptoms and provision of psychological, social, and spiritual care for children living with life-threatening or terminal conditions. 2 Palliative care as a philosophy, as a model of health care delivery derived from hospice care, and as an evolving medical subspecialty are often confused. Palliative care is not hospice care, although these services may overlap for patients nearing death. According to an analysis conducted by Meghani, 3 the evolving concepts of hospice and palliative care are lagging in clinical practice compared to contemporary literature, and this may threaten the validity and timing of helping patients and families. Many barriers to pediatric palliative care ex- 1 Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin, and Palliative Care Program at Children s Hospital of Wisconsin, Milwaukee, Wisconsin. 2 Shafi Medical Center, Milwaukee, Wisconsin. 1128

2 PERCEPTIONS OF THE TERM PALLIATIVE CARE 1129 ist. 2,4,5 Communication surrounding a child s uncertain prognosis or potential death is complicated and difficult. 6,7 The issues and tasks involved in caring for a child with a life-threatening illness are emotionally charged, and parents may have a difficult time making decisions and searching for help. Aspects of possible death, including its planning, acceptance, and meaning may create a negative perception of palliative care programs. Leaders in pediatric palliative care encourage introduction of the service and philosophy earlier in the disease trajectory and in patient populations beyond cancer, such as those with complex chronic conditions. 2,8,9 However, health care consumers and providers may not be aware of this movement because of limited exposure, in part because of the relative newness of this specialty. Two international surveys found that the majority of respondents either never heard of palliative care until they were referred or they reported much confusion about the intent and type of services available. 10,11 Pediatric palliative care providers debate changing the name of programs, changing people s perception of what palliative care is and when it is beneficial. Chris Hubble comments that while staff may tend to avoid the term palliative care, there is no better word to replace it. 12 Many providers creating new palliative care services often ask, What should we call our program? How parents and providers perceive a palliative care program may affect utilization of services. We therefore conducted a pilot study with aims to understand perceptions of and feelings associated with the words palliative care. Our hypotheses were (1) the words palliative care may be a barrier to program use, (2) the use of the alternative descriptor supportive care will increase likelihood of program use and (3) providing a nonthreatening program description (death-related words removed) will increase likelihood of program use for all study participants from before to after reading the description. METHODS Study design This was a randomized, parallel-group survey study with one between-group manipulation (two program name groups) and one withingroup manipulation (providing a program description for subjects to read between assessments). In addition, we surveyed two independent subject populations (parents and staff). Study site The Children s Hospital of Wisconsin (CHW) is a 220-bed freestanding children s hospital with a long-standing pediatric palliative care program. Two inpatient units were surveyed. The outpatient offsite location was a busy seven-physician general pediatric clinic also in Milwaukee, Wisconsin. Subjects Human subject protection. After receiving support from the heads of the study sites, we obtained approval to conduct the study by the Institutional Review Board at CHW. No protected health information (PHI) was accessed or collected during this study; a waiver of authorization for HIPAA purposes was granted. The survey was anonymous and confidential. Subject selection. We surveyed two subject populations: parents and health care providers (staff). We selected subjects by convenience sample of site staff or parents whose children were patients on the 24-bed Hematology/Oncology/Transplant unit (HOT), the 24-bed Pediatric Intensive Care Unit (PICU), or an offsite general pediatric clinic. No formal inclusion/exclusion criteria were used. The research nurse asked potential subjects if they would like to complete a brief, informal survey about a pediatric health program. If so, the subject read the survey consent. At that time, the research nurse pulled a code letter from an envelope; the code referred to one of the two program names thereby assigning the subject to a name group. Sample size. Our primary outcome variable of Likelihood of Program Use was scored on a fivepoint Likert scale. We proposed a clinically significant ( 0.05) Name group difference as 1 point, with a conservative standard deviation of 2.0; accepting power of 0.80, we required a minimum sample of 50 subjects per name group for each population of parents and of staff. Anticipating a 10% rejection/attrition rate based on prior survey administration in these patient populations, we sought to survey 55 parents per name group (total of 110 parents) and 55 staff

3 1130 members per name group (total of 110 staff members). The study was not powered to detect differences between our three study sites. Because of the actual distribution of data, however, the power analysis above was retested for nonparametric testing. The Mann-Whitney U test with 50 subjects per name group can detect with 80% power, that for any pair of subjects (one from each name group), the subject from one group will score higher than the subject from the other group 2 of 3 times (2:1). Survey administration The research nurse gave each assigned subject a packet of materials identified throughout as either palliative care or supportive care. Subjects completed the written survey within the next 24- hour period; the research nurse ensured that each participant completed only one survey, and that the subject answered the first page of questions prior to reading the program description. Items asked respondents about the program name in terms of their knowledge of the words palliative care/supportive care (yes/no); likelihood to use the program in the hypothetical situation of caring for a child with life-threatening illness (from definitely no to definitely yes); opinions/attitudes about the program, and ratings on eight feelings (each rated from not at all to extremely). Subjects were then instructed to read a nonthreatening program description with the assigned program name, from which the words dying, death, hospice, and bereavement were omitted. This page was followed by a repeated set of questions asking about the program. We also asked participants to rate (from poor to excellent) six possible program names (care, total care, comfort care, advanced care, palliative care, supportive care), and included background items on age, education, and gender of respondent. Survey content and program description are available upon request. Surveys were designed and produced on SurveyPlus 2000 (Autodata Systems, Minneapolis, MN) software. Subjects filled in their responses by circle-filling or writing in comments. Data analysis Scanned data from the three study sites were pooled within the assigned Name groups. Analyses were conducted in SPSS (v. 10, SPSS Inc., Chicago, IL) separately for the parent population MORSTAD BOLDT ET AL. and the staff population. Demographic data were analyzed to assess comparability between assigned Name groups. A p value 0.05 was considered significant. To evaluate the effect of the two program names on hypothetical program use before participants read the program description (hypotheses 1 and 2), we conducted the Mann-Whitney U test at the first time point (T1), between both name groups, in parent and staff populations separately. We also examined this effect after subjects read the description (T2). To evaluate the effect of reading the program description on hypothetical program use (hypothesis 3), we conducted the Wilcoxon signed rank test at T1 and T2 within each name group. Additional nonparametric analyses were conducted to examine eight feelings evoked by the two program names before and after reading the program description. Descriptive analyses were done to evaluate responses to other suggested program names within each assigned Name group. Mann-Whitney U tests were done to compare ratings for each potential program name between assigned name groups. We also asked respondents if they knew what the words palliative care/supportive care meant (yes/no) and analyzed results by Pearson 2 test and McNemar test. Finally, write-in comments on subject definition and opinion of a palliative care/supportive care program were examined thematically only and results presented for additional insight into the subjects perceptions of the program. RESULTS Subject population Survey rates. Of 220 subjects who were approached to take the survey, 195 surveys were administered (a response rate of 89%). Of the 195 surveys administered, 184 (94%) were considered evaluable, defined as: the subject completed at minimum Part 1, Item 3 (our primary endpoint at T1). We pooled data from the HOT unit, PICU, and the offsite general pediatric clinic. See Table 1A for site response rates. Of the evaluable surveys, 105 were filled out by parents and 79 by staff respondents. The staff sample fell below our projected number due to limits of available providers at the off-site clinic.

4 PERCEPTIONS OF THE TERM PALLIATIVE CARE 1131 TABLE 1A. Demographics. The palliative care group and supportive care group were comparable on the background items age, education, and gender in both parents and staff groups (Table 1B). Likelihood of program use SITE RESPONSE RATES Parent group Staff group Site N N HOT PICU General Pediatric Total HOT, hematology/oncology transplant unit; PICU, pediatric intensive care unit. Results at T1 and T2 for the two program name groups, produced separately by parent and staff populations, are presented in Tables 2 and 3, respectively. Parents. At T1, parents with palliative care materials stated they were significantly (p 0.05) less likely to use the program than those with supportive care materials (Table 2A). At T2, parents with palliative care materials were as likely to use the program as those with supportive care materials. Within each name group, parent likelihood to use the program increased significantly after reading the program description (Table 2B). Staff. Staff likelihood to use the program did not differ between assigned Name groups at either timepoint, T1 or T2 (Table 3A). Staff likelihood to use the program increased significantly after reading the program description within the palliative care group only (Table 3B). Feelings evoked by the program name Parents. At both T1 and T2, significant differences (p 0.05) between the two name groups were determined for feelings evaluated, with the supportive care group having more positive feelings across the spectrum compared to the palliative care group (Fig. 1A). In the parent population, significant improvements towards the positive from T1 to T2 were seen within each group (Fig. 1B). Staff. At T1 and T2, significant differences between the name groups were determined for many of the feelings evaluated, with the supportive care group of staff having generally more positive response to the name compared to the palliative care group (Fig. 2A). Significant improvements in the feelings measured before and after reading the description were seen within each group (Fig. 2B). Responses to other program names Figures 3A and 3B display ratings on six potential program names. Parents rated the name supportive care program with the highest scores and palliative care program with the lowest scores, regardless of their name group assignment (descriptive only). Staff also rated the name supportive care program with the highest scores but care program with the lowest scores, regardless of their name group assignment (descriptive only). For each potential program name, the palliative care group expressed ratings similar to the supportive care group, in both parent and staff populations. Thus the name group data were pooled for display. The only exception occurred in the staff population, where the potential name care program was rated significantly higher in the palliative care group compared to the supportive care group (Mann-Whitney U 479.0, p 0.022). Self-perceived knowledge Before reading the program description (T1), significantly more parents in the supportive care group answered yes they knew what these TABLE 1B. BACKGROUND ITEMS* Palliative Care (PC) Supportive Care (SC) Parent group Age (mean years) Education (mean code ) Gender (n) M 7 M 7 F 37 F 48 Staff group Age (mean years) Education (mean code ) Gender (n) M 6 M 5 F 34 F 31 *No significant differences (P 0.05) in background items were found between PC versus SC groups Education code: 2 completed high school, 3 some college, 4 Bachelor degree, 5 graduate degree

5 1132 MORSTAD BOLDT ET AL. TABLE 2. LIKELIHOOD OF PROGRAM USE* BEFORE AND AFTER READING THE PROGRAM DESCRIPTION Parent Population a. Effect of Program Name Between Groups (at each time-point) Likelihood to p-value use time Name group N Mean score (2-tailed) Before (T1) Palliative care Supportive care After (T2) Palliative care Supportive care b. Effect of Program Description Within Groups (over time) p-value Name group Time N (2-tailed) Palliative care T1, T Supportive care T1, T *Rating scale: 1 definitely no, 2 probably no, 3 not sure, 4 probably yes, 5 definitely yes Pooled data for 3 study sites (HOT, PICU, Off-site) N: evaluable population those who completed at least Part 1 Item 3 (likelihood item), numbers evaluable decreased over time due to respondents not completing the item at T2 Statistically significant at P 0.05 words (their assigned program name) meant than in the palliative care group (57.6% versus 36.4, , p 0.033). This difference held at T2 with 100% of the supportive care group answering yes compared to 89.7% of the palliative care group ( , p 0.020). Reading the description significantly increased the percentage of parents responding yes within each name group (McNemar test; PC: p 0.000; SC: p 0.000). In the staff population, no differences between the name groups were observed at either time TABLE 3. LIKELIHOOD OF PROGRAM USE* BEFORE AND AFTER READING THE PROGRAM DESCRIPTION Staff Population a. Effect of Program Name Between Groups (at each time-point) Likelihood to p-value use time Name group N Mean score (2-tailed) Before (T1) Palliative care Supportive care After (T2) Palliative care Supportive care b. Effect of Program Description Within Groups (over time) p-value Name group Time N (2-tailed) Palliative care T1, T Supportive care T1, T *Rating scale: 1 definitely no, 2 probably no, 3 not sure, 4 probably yes, 5 definitely yes Pooled data for 3 study sites (HOT, PICU, Off-site) N: evaluable population those who completed at least Part 1 Item 3 (likelihood item), numbers evaluable decreased over time due to respondents not completing the item at T2 Statistically significant at P 0.05

6 1133 PERCEPTIONS OF THE TERM PALLIATIVE CARE FIG. 1. Feelings evoked by program name in parents. Significant (Mann-Whitney, p 0.05) differences between name groups were seen: (a) at T1 before reading the description for all feelings; (b) at T2 before reading the description for all feelings except mad. Also significant (Wilcoxon, p 0.05) changes over time (T1, T2) were seen: within the palliative care group for hopeful, relieved, and good; within the supportive care group for hopeful, relieved, and good. point for self-perceived knowledge. Reading the description significantly increased the percentage of staff responding yes within the supportive care group only (McNemar test; SC: p 0.016). DISCUSSION This pilot study was designed to learn what parents and health care providers think and feel FIG. 2. Feelings evoked by program name in staff. Significant (Mann-Whitney, p 0.05) differences between name groups were seen: (a) at T1 before reading the description for sad, fearful, mad, good, and nervous; (b) at T2 before reading the description for uncertain, sad, fearful, mad, and nervous. Also significant (Wilcoxon, p 0.05) changes over time (T1, T2) were seen within the palliative care group for sad, hopeful, relieved, good, and nervous; within the supportive care group for uncertain, relieved, and good.

7 1134 MORSTAD BOLDT ET AL. FIG. 3. Ratings of various program names. about the term palliative care. We found that initially, parents were more likely to use a program if it was called supportive care versus palliative care, a difference that disappeared after parents read a nonthreatening program description. Parents also expressed more positive feelings about a program named supportive care versus palliative care. This difference remained after they read about the program services, but significant improvements in feelings were seen within each name group between time points. When rating six potential program names, parents rated supportive care highest. Most of these effects were also seen in the staff population, although overall, they were less influenced by the assigned group name and by reading the program description. Limitations to this study include a moderately unequal randomization of patients into the two name groups, a slight attrition factor, and fewer healthcare providers than desired. Furthermore, because a portion of the participants were parents of patients without life-threatening illness, interpretations must be extrapolated to a palliative population. Our results suggest two phenomena. First, families and staff may not understand what palliative care is; second, the term palliative care is associated with negative feelings. Despite reading the program description and reporting an increased rating for likelihood to use the program, the palliative care group still had more negative feelings compared to the supportive care group, and the supportive care program was rated highest of the potential names listed. Comfort care was rated second highest in both populations, which suggests it also is less threatening and foreign. When respondents were asked to describe the program, comments included: the name supportive care was vague and the concept is nebulous, while another wrote Supportive care sounds like false advertising. Another respondent suggested that we use a more direct word than palliative. When asked to define palliative care many respondents said, Can t even guess, and, I have no idea. Should we change the name of palliative care programs? Providing accurate information seems to be a must-do mission.13 After simply reading a short program description, respondents knew more, felt better, and were more likely to use the program. This was especially true for the group that received program materials labeled palliative care, who started with poorer understanding and perhaps as a result, more negative perception. In editorials, Chamberlain14 and Cacchione15 discuss the impact of language on communication about and utilization of end-of-life care programs, such as hospice. Chamberlain proposes that the words themselves may be at the root of the problem and using new terms may overcome the communication barrier that keeps programs from joining the mainstream of accepted and acceptable care options. However, he suggests that

8 PERCEPTIONS OF THE TERM PALLIATIVE CARE 1135 any new term used would end up providing the same barrier (i.e., relaying a meaning perceived by people as close-to-death, dying, no more treatment, and last resort). Chamberlain also examines the alternate hypothesis that this public perception problem may not be caused by the name itself but perhaps by the deeper nature and meaning of the thing that the term represents. Parties that regulate such care can contribute to these perception problems by writing troubling eligibility criteria that perpetuate the image of no treatment and close-to-death. 14 Cacchione and the Coalition for Supportive Care of the Dying found that health care workers were reluctant to refer patients to related studies because of the association between the words and the concepts of dying. He too suggests that physicians do not like to refer patients to hospice programs because they do not want to take away hope. 15 This belief seems heightened in pediatric situations. 4 Progress may occur when we educate patients, families, and the public on what pediatric palliative care programs actually provide. For health care providers, practice guidelines 16 and curricula in pediatric palliative care are now available, 17,18 and learning can be enhanced by using the many teachable moments that present themselves during the provision of care. 20 Results from the Canadian study 10 suggest that information on palliative care should be disseminated through the media, community, churches, and health care providers. How should this dissemination and education take place, both internally and externally? There is an inherent tension in hospital marketing between promoting both cure rates and palliative care, which can be especially confusing for those who don t know or acknowledge that they are not mutually exclusive endeavors. Broaching the topic of pediatric palliative care brought interesting and honest reaction from the participants of this project. As this subspecialty grows, soliciting impressions from parents and staff is invaluable to effective program development. This pilot study demonstrates that providing accurate information to families and health care providers about what palliative care programs offer improves perceptions about palliative care and increased likelihood of program utilization. And perhaps, although vague, supportive care may be a more comfortable idea and name for the program. ACKNOWLEDGMENTS The authors wish to thank Laura Nigh, R.N., B.S.N., for assistance with data collection and Bruce Camitta, M.D., for guidance and thoughtful review of the manuscript. This project was funded in part by the Small Grants Program, Department of Research, Children s Hospital of Wisconsin (Ann Morstad Boldt, M.S.) and by the Midwest Athletes against Childhood Cancer (MACC) Fund, Medical College of Wisconsin (Bruce P. Himelstein, M.D.). Dr. Himelstein is a Faculty Scholar of the Project on Death in America. Preliminary results from this project were presented at the Annual Assembly of the American Academy of Hospice and Palliative Medicine, Phoenix, January 2004, and the European Association of Palliative Care Research Forum, Stresa, Italy, June REFERENCES 1. Kane J, Himelstein B: Palliative care for children. In: Berger AM, Portenoy RK, Weissman D (eds): Palliative Care & Supportive Oncology, 2nd ed. Lippincott Williams & Wilkins: Philadelphia, 2002, pp American Academy of Pediatrics. Committee on Bioethics and Committee on Hospital Care: Palliative care for children. Pediatrics 2000;106(2 Pt 1): Meghani SH: A concept analysis of palliative care in the United States. J Adv Nurs 2004;46: Himelstein BH, Hilden JM, Boldt AM, Weissman D: Medical progress: Pediatric palliative care. N Engl J Med 2004;350: Field MJ, Behrman RE (eds): When Children Die: Improving Palliative and End-of-Life Care for Children and Their Families. Washington, D.C.: Institute of Medicine Report. The National Academies Press, Kreicbergs U, Valdimarsdottir U, Onelov E, Henter JI, Steineck G: Talking about death with children who have severe malignant disease. N Engl J Med 2004;351: Browning D: To show our humanness-relational and communicative competence in pediatric palliative care. Bioethics Forum 2002;18: Feudtner C, Hays RM, Haynes G, Geyer JR, Neff JM, Koepsell TD: Deaths attributed to pediatric complex chronic conditions: National trends and implications for supportive care services. Pediatrics 2001;107:E Freyer DR: Care of the dying adolescent: special considerations. Pediatrics 2004;113: Claxton-Oldfield S, Claxton-Oldfield J, Rishchynski G: Understanding the term palliative care : A

9 1136 Canadian survey. Am J Hosp Palliat Med 2004;21: Voltz R, Akabayashi A, Reese C, Ohi G, Sass HM: Organization and patients perception of palliative care: A cross-cultural comparison. Palliative Med 1997;11: Hubble C: Word choices in pediatric palliative care medicine. National Alliance for Children with Life- Threatening Conditions (NACWLTC) Physician Commentary _November.asp (Last accessed June 10, 2004). 13. McCracken S, Colburn K, Pastorik L: Hospice care: Mapping the message. Am J Hosp Palliat Care 1994;11: Chamberlain BH: What s in a name? Am J Hosp Palliat Care 2001;18: Cacchione P: What s another word for euphemism? Am J Hosp Palliat Care 2000;17: National Consensus Project for Quality Palliative Care. Clinical Practice Guidelines. consensusproject.org/guidelines.html (Last accessed September 20, 2004). 17. Initiative for Pediatric Palliative Care (IPPC). IPPC curriculum. (Last accessed October 2, 2004). MORSTAD BOLDT ET AL. 18. American Association of Colleges of Nursing. End-of- Life Nursing Education Consortium (ELNEC) Project. (Last accessed October 4, 2004). 19. Medical College of Wisconsin and American Academy of Hospice and Palliative Medicine. End-of-Life Palliative Education Resource Center (EPERC). (Last accessed October 4, 2004). 20. Sahler OJ, Frager G, Levetown M, Cohn FG, Lipson MA. Medical education about end-of-life care in the pediatric setting: Principles, challenges, and opportunities. Pediatrics 2000;105(3 Pt 1): Address reprint requests to: Ann Morstad Boldt, M.S. Department of Pediatrics Medical College of Wisconsin 999 N. 92nd Street, CCC C350 Milwaukee, WI aboldt@chw.org