Communicating With Realism and Hope: Incurable Cancer Patients' Views on the Disclosure of Prognosis
http://www.100md.com
《临床肿瘤学》
the Medical Psychology Research Unit, University of Sydney, Sydney, New South Wales, Australia
ABSTRACT
PATIENTS AND METHODS: One hundred twenty-six (58%) of 218 patients invited onto the study participated. Eligible patients were the consecutive metastatic cancer patients of 30 oncologists, who were diagnosed within 6 weeks to 6 months before recruitment, over 18 years of age, and without known mental illness. Patients completed a postal survey measuring patient preferences for the manner of delivery of prognostic information, including how doctors might instill hope.
RESULTS: Ninety-eight percent of patients wanted their doctor to be realistic, provide an opportunity to ask questions, and acknowledge them as an individual when discussing prognosis. Doctor behaviors rated the most hope giving included offering the most up to date treatment (90%), appearing to know all there is to know about the patient's cancer (87%), and saying that pain will be controlled (87%). The majority of patients indicated that the doctor appearing to be nervous or uncomfortable (91%), giving the prognosis to the family first (87%), or using euphemisms (82%) would not facilitate hope. Factor analysis revealed six general styles and three hope factors; the most strongly endorsed styles were realism and individualized care and the expert/positive/collaborative approach. A range of demographic, psychological, and disease factors were associated with preferred general and hope-giving styles, including anxiety, information-seeking behavior, expected survival, and age.
CONCLUSION: The majority of patients preferred a realistic and individualized approach from the cancer specialist and detailed information when discussing prognosis.
INTRODUCTION
Prior research demonstrates that a clear majority of cancer patients in the Western world reports a preference for detailed information about their disease and expected outcome,9-13 although information needs can vary across different phases of the illness.14-16 Legal rulings have emphasized the responsibility of doctors to provide all necessary information in some jurisdictions.17 Nevertheless, patients often misunderstand the status of their disease and the aim of treatment13 and commonly overestimate their life expectancy.18 This may impact on decision making, particularly concerning anticancer treatment that may have side effects and reduce quality of life.19,20
Issues that concern clinicians about communicating life expectancy to patients include how much information to give, difficulties with prognostication, and disclosing prognostic uncertainty without increasing anxiety, reducing trust, and destroying hope.5,14,21-25 Others, however, believe expressing this prognostic ambiguity carefully to patients is one means of engendering hope4,19 and that providing information decreases anxiety associated with ignorance and uncertainty.14,26-28
The need for optimism and hope to be sustained in the process of honestly delivering bad news and a limited life expectancy is an ideal expressed by both doctors and patients.4,5,12,15,29 However, there is a delicate balance between fostering realistic hope and unethically creating unrealistic expectations of longevity.5,19,21,29 Furthermore, hope is a broad concept that can hold different meanings for each individual. Similarly, prognosis, a term often thought of as synonymous with life expectancy, encompasses broader issues surrounding the future course of the disease and how this may impact on the quality of life of the patient.19,27,30,31 Despite the complex issues surrounding this topic, apart from one study on communicating hope in the diagnosis of early-stage cancer,31 research is lacking in determining how patients define hope and how health professionals communicate hope to patients.14,32
Much of the communication literature has focused on how to break bad news.6,33,34 Few studies or guidelines have targeted the communication of prognosis specifically, and these few studies focus on either early-stage disease10,35 or palliative and end-of-life issues.12,13,29,36,37 There is a lack of evidence-based information on discussing a poor prognosis, a context that arguably demands more resources from both doctors and patients.5,38 Furthermore, although patient preferences for general information have been associated with demographic, psychological, and disease variables, for example, age,11,26,28,39 sex,11,16,39 religiosity,16 education level,26 involvement preference scores,16,40 anxiety and depression levels,35,41 and disease status,16,28 it has not yet been clarified whether such variables specifically influence metastatic cancer patient preferences for prognostic information.
In the current study, we aimed to identify the context and the way in which patients with incurable metastatic cancer wanted to be informed about their prognosis and to explore what features in the delivery of prognostic information they would experience as more or less hopeful. We hypothesized that the majority of patients would want detailed prognostic information as well as communication that facilitated hope. We also expected some variation in preferences according to patient demographic, disease, and psychological characteristics.
PATIENTS AND METHODS
Measures
Survey development. A written survey was used to elicit patient preferences for the content and format of prognostic discussion. To develop survey items, key themes were abstracted from the published literature, including an earlier qualitative study of breast cancer patients5 and analysis of audiotapes of initial oncology consultations.42,43 These themes included patient desire for information and involvement,9,16,26,40 honesty, and the provision of hope when prognosis is communicated.5,12,16,36,44 The instrument was reviewed by oncologists, health professionals, and members of a consumer advocacy group and piloted among 10 patients with metastatic cancer. Minor revisions were made in response to reviewers' and pilot participants' feedback.
Demographics and disease details. Participants were asked about the following: age, relationship status, occupation, highest educational level achieved, medical or allied health training, language spoken at home, parents' country of birth, whether or not they have children, and whether or not they have a religious denomination or spiritual belief. Oncologists provided patient disease information, including type of cancer, dates of diagnosis of primary and metastatic cancer, mode of treatment, and estimated survival.
Doctor style. Thirty-five doctor informational and supportive behaviors when discussing prognosis, which were derived from earlier studies of prognostic information preferences,10,31 were listed. Patients rated their response to each item on a 5-point Likert scale (agree completely, agree, neutral, disagree, and disagree completely). Examples of items included, "I would like my cancer specialist to: (a) check my understanding of what he/she has told me, (b) be realistic, (c) be optimistic, and (d) tell me about cancer support groups."
Definition of hope. Participants were provided with four exemplar hope definitions, using the words of patients from a previous study.45 The definitions were as follows: a feeling or expectation (1) that things can go well; (2) that because one thing has gone wrong, it doesn't mean that other things will not go well; (3) that you have just as good chances (if not better) as the next person of having the best outcome; (4) that you can still enjoy a good quality of life, even if life expectancy is uncertain; or (5) other. Participants were asked to indicate what hope means to them in their current situation; they could select more than one item and/or write their own definition under other. Participants were also asked a separate open question about what else (apart from doctor communication) helps them to be hopeful.
What patients find hopeful. Thirty doctor behaviors that might convey or discourage hope, such as "gave me survival statistics," "appeared nervous or uncomfortable," and "was occasionally humorous," were listed. The items were derived from Sardell and Trierweiler31 and our earlier qualitative study exploring ideal prognostic communication in the metastatic setting.5 Participants were asked to indicate on a 5-point Likert scale how hopeful each of the behaviors would make them feel (ie, very hopeful, hopeful, neutral, not hopeful, and hopeless).
Preference for prognostic information, timing, and manner of presentation. Patients were asked to indicate their preference for specific prognostic facts, mode of presentation of statistics, and the timing of prognostic information and who should initiate the discussion. These results have been presented elsewhere.46
Information and involvement preferences. Participants' preferences for information and involvement were elicited using the seven-item binary Information subscale of the Krantz Health Opinion Survey ({alpha} = .74 in this sample).47
Depression and anxiety. Levels of depression and anxiety were measured by the 14-item Hospital Anxiety and Depression Scale self-assessment tool devised by Zigmond and Snaith48 ({alpha} = .84 and {alpha} = .81 in this sample).49,50
Statistical Analyses
Open questions and patient preferences for the doctor's style and hope-giving items were analyzed using descriptive statistics. Doctor style and hope items were then entered separately into two factor analyses using varimax rotation with Kaiser normalization. Total scores for the identified factors were calculated by summing items with weights more than 0.3 on each factor. If an item loaded onto more than one factor, it was included in the factor on which it had the highest weight and excluded from the other factor(s). Associations between demographic, psychological, and disease variables (described earlier) and each factor (the dependent variable) were explored in univariate analyses (using linear regression, t tests, and ANOVA). Associated variables that were found to be significant at the 0.25 level in univariate analyses51 were entered into binary logistic regressions. Because anxiety and depression scores were highly correlated (Pearson's correlation, r = 0.57; P < .01), only one of these scores (the more significant on univariate analysis) was entered into multivariate analyses.
RESULTS
Of the 218 patients approached to participate, 10 were ineligible, and 22 refused. Of the remainder, 126 patients completed the survey. The most common reason cited for not completing the survey was ill health. No significant differences were found between those who completed the survey and those who did not complete the survey on the variables of age (F2,194 = 0.180; P = .672), sex ({chi}2 = 0.294; df = 1; P = .588), clinic where recruited (metropolitan or nonmetropolitan; {chi}2 = 2.540; df = 1; P = .111), type of cancer ({chi}2 = 4.341; df = 4; P = .362), and time since diagnosis of metastatic disease (F2,192 = 0.008; P = .930).
Fifty-six percent of participants were male, whereas 25% had breast cancer, 18% had colorectal cancer, 16% had prostate cancer, 10% had lung cancer, and 31% had other cancer types. The mean time since diagnosis of metastases was 13 weeks (range, 1 to 39 weeks; standard deviation [SD], 8.7 months). The majority of patients had an expected survival estimated by their oncologist of months (42%), or years (42%) and were receiving either systemic anticancer therapy or radiotherapy (92%; Table 1).
Information and Involvement Preferences
The average score on the Krantz Health Opinion Survey: Information Preferences Subscale was 3.9 (SD, 2.15), which is in the medium score range, indicating that, overall, participants did not have either a particularly high or low desire to ask questions or to be informed about medical decisions.46
Anxiety and Depression
Twenty-three percent and 19% of patients fell into the possible case range (scores, 8 to 10) for anxiety and depression respectively. Ten percent and 7% of patients fell into the probable case range (scores, 11 to 21) for anxiety or depression, respectively.48
Preference for Doctor Style
Of the 35 doctor behaviors listed, the most preferred included for the doctor to be realistic, provide an opportunity to ask questions, and acknowledge the patient as an individual when discussing prognosis (all 98%; Table 2). Less-preferred items were for the oncologist to discuss the patient's financial situation (37% of patients), give the patient his or her prognosis with another medical person present (28% of patients), give the patient an audiotape of their discussion (19% of patients), and give the patient his or her prognosis over the telephone (14% of patients; Table 2).
Hope-Giving Behavior
Of the 30 listed doctor behaviors, those rated as being most hopeful included being offered the most up to date treatment (90%), the oncologist appearing to know all there is to know about the patient's cancer (87%), the occasional use of humor (80%), being told that the pain will be controlled (87%), and being told all treatment options (83%; Table 3).
Behaviors that were rated as causing the patient to feel not hopeful or hopeless were the doctor appearing to be nervous or uncomfortable (91%), giving the prognosis to the family first (87%), the use of euphemisms (82%), avoiding talking about cancer and only discussing treatment (75%), and giving the good news first and then the bad news (72%); however, almost half of the patients rated this last item as neutral (46%).
Thirty percent of participants rated giving statistics about how long they will live as likely to make them feel hopeful; however, similar percentages of patients rated this item as not hopeful (32%) and neutral (38%). Similarly, participants rated expressing uncertainty about the course of the cancer (ie, said that the course of the cancer cannot be predicted) evenly across the options, with 35%, 30%, and 35% rating this as hopeful, not hopeful, and neutral, respectively (Table 3).
Patient Definition of Hope
Some participants ticked one or more of the four exemplars of hope; however, the majority (62%) wrote their own definition with or without ticking an exemplar. The most commonly endorsed definition was "that you can still enjoy a good quality of life even if life expectancy is uncertain" (19%) and "a feeling or expectation that things can go well" (14%). Fewer endorsed the items of "that you have just as much chance as the next person of having a good outcome" (3%) and "that because one thing has gone wrong, it doesn't mean that other things will go wrong" (3%).
Of the alternatives proposed, the most common themes were quality of life and fulfillment of goals (23%; for example, "To get on with life, make sure you make the most of it for as long as you can; set a distant goal and work like hell to get there"), receiving the best treatment that would provide the best cancer and symptom control (8%; for example, "Hope for me is a general expectation that everyone involved in my treatment is doing their best"), and the hope for cure or remission (8%; for example, "... hope that they may find a cure or extend my life").
General Factors Influencing Hopefulness
One hundred four (83%) of the 126 respondents completed the open question on general factors influencing hopefulness. The majority (87%) stated that family helped them to be hopeful. Other factors cited were religious beliefs or spirituality (28%), friends (24%), a positive attitude (13%), his or her children or grandchildren (12%) or partner (10%), and scientific advances in cancer care (9%).
Factor Analysis
Factor analysis of the 35 doctor style items revealed six factors that accounted for approximately 57% of the total variance (Table 4). These styles were as follows: factor 1, realism and individualized care (providing realistic and direct information tailored to the individual); factor 2, emotional support (providing information on support services and an openness to discuss patients' fears and concerns); factor 3, facilitation of coping with dying (displaying openness to discuss concerns about dying and providing information on palliative care services); factor 4, provision of information (ensuring patient understanding and providing information materials such as publications and audiotape of consultation); factor 5, emphasizing all options (providing information on complementary therapies, discussing optimistic future scenarios, and suggesting a second opinion); and factor 6, a personal approach (sitting next to the patient and sharing personal information; Table 4).
Three factors were identified among the 30 hope items, which accounted for approximately 54% of the variance (Table 5). These were as follows: factor 1, expert/positive/collaborative (expertise, humor, and inclusion of patient as part of the team); factor 2, avoidant (avoiding or appearing uncomfortable about discussing the cancer, using euphemisms, and giving the prognosis to others first); and factor 3, empathic (expressing one's own feelings or asking the patient about his or her own reaction to the prognosis; Table 5). One item ("asked me if I would like a second opinion") did not weigh on any of the factors.
The strength of preference for these styles was determined by calculating the total scores for each factor. Because different factors included varying numbers of items, the totals were normalized to allow direct comparison. Thus, scores could range from 20 to 100. The most strongly endorsed doctor style was realism and individualized care (mean score, 88.9; SD, 9.6), and the least endorsed doctor style was personal approach (mean score, 59.9; SD, 14.1). The most strongly endorsed hope-giving style was the expert/positive/collaborative approach (mean score, 78.4; SD, 14.4).
Variables Significantly Associated With Preferences
Doctor style. Table 6 lists the means scores on the general doctor style factors by those variables found to be significant in multivariate analyses. Higher anxiety scores (t102 = 2.463; P = .02), having a partner (t102 = 2.00; P = .048), English speaking at home (t102 = –2.363; P = .02), expected survival of years as opposed to weeks or months (t102 = 2.23; P = .03), and longer time since diagnosis of metastatic disease (t102 = 2.09; P = .04) were found to significantly and independently predict preference for the oncologist to have a realistic and individualized approach (factor 1). Higher Krantz Information Subscale scores (t102 = 2.88; P = .005) were also significantly associated with this factor.
Age was the only variable significantly associated with preference for factor 2, with younger patients more likely to prefer an emotionally supportive approach (t103 = –4.53; P < .001). Higher anxiety scores (t115 = 2.80; P = .006) and female sex (t115 = –2.68; P = .008) predicted preference for the facilitation of the coping with dying approach (factor 3). Patients who scored higher on information seeking (t108 = 3.10; P = .002) were more likely to prefer the informative approach (factor 4). Younger patients (t114 = –3.65; P < .001) were more likely to prefer the emphasizing all options approach (factor 5). Patients who stated they had a religious belief (t109 = 3.40; P = .001) were more likely to prefer the personal approach (factor 6; Table 6).
Hope-giving style. Table 6 lists the mean scores on the hope-giving factors by those variables found to be significant in multivariate analyses. Older patients rated the expert/positive/collaborative (factor 1) and empathic (factor 3) approaches as significantly more hope giving than younger patients (factor 1: t97 = 2.08; P = .04; factor 3: t102 = 3.18; P = .002). Anxiety was also significantly associated with these factors (factor 1: t97 = 2.20; P = .03; factor 3: t102 = 2.43; P = .02; Table 6). None of the variables explored were found to be significantly associated with the avoidant approach (factor 2).
DISCUSSION
There was some variability in preferences according to patient characteristics, which allows some insight into who may prefer which style. Overall, younger, anxious patients placed a stronger focus on emotional support for themselves and their families. This is understandable given that these patients are more likely to have a dependent family whose support needs during their illness and after their death may be a primary concern for them. These patients seemed to have high needs and high expectations for medical care. This is in accordance with other studies that report that younger patients have a preference for more information and have high unmet needs.26,52
Patients who had been diagnosed with metastatic disease for a longer period and those with longer expected survival were more likely to want realism (factor 1). Perhaps patients are better able to cope with realism over time. In earlier work, we found that patients who are in routine follow-up, without relapse, develop stronger preferences for involvement in decision making over time.16 A longer expected survival would be understandably easier to discuss frankly. Interestingly, because these data were derived from oncologists' prognostic estimates, it would seem that these patients had some awareness of their own prognosis.
English speakers were also more likely to prefer realism. Those who speak another language at home are possibly influenced by a culture where avoidance and paternalism is more common. It is well documented that, in some cultures, communication of prognosis is viewed as harmful and brutal.53-57 Overall, however, the majority of patients in all groups preferred a realistic approach. Generalizations concerning a group linked by cultural factors can lead to stereotyping. Thus, information about beliefs and practices within particular cultures and groups never obviates the need for exploring individual preferences and needs. Notably, patients who did have a religious belief wanted this to be acknowledged by the cancer specialist in the context of discussing prognosis.
Only two variables were associated with preferences for hope-giving behaviors. Older patients were more likely to rate both the expert and empathic approaches as more hope giving. However, older patients were less likely to prefer the emotionally supportive approach as a general doctor style. It seems that older patients want direct emotional support from their doctor as opposed to more formal organized support. This supports previous findings in the literature.58,59 Anxiety was also associated with these factors, although, on examination of the means, it seems that the relationship may be curvilinear, with patients scoring high or low showing less interest in these approaches. This is difficult to interpret, and further research may clarify these relationships.
Some of the results are not consistent with current practice guidelines.37,60 For example having another medical person present during prognostic discussion and offering an audiotape of the discussion were not endorsed by participants in this study. A personal approach that involved sitting next to the patient or sharing some personal information was also less valued. These results are similar to those of Schofield et al61 who reported that patient preferences do not always reflect published consensus guidelines, which may need to be modified in light of this feedback.
There were several limitations to the study. Cultural factors have not been well considered because non-English speakers were excluded from the sample. It is known that cancer patients from other cultures have different views and information needs.53-57 It is also not certain, although probable, that these results would be relevant to the cancer populations of Britain and North America. Furthermore, some subgroups in the sample (eg, particular cancers) are quite small, and therefore, this study did not allow detection of differences between these groups. A larger sample may have produced more significant and reliable results.
Only 30 of 106 oncologists invited to participate in this study took part. Although no demographic or practice differences were observed between those who accepted and those who actively refused to participate, more subtle differences (such as interest in communication) that we did not measure may have differentiated these groups and, therefore, influenced the results. Furthermore, no data were available on oncologists who did not respond at all.
Not all oncologists may have complied with the instruction to recruit consecutive eligible patients. They may have invited patients who felt more comfortable with their diagnosis or who were less symptomatic; few patients were receiving symptomatic care only. Although this is characteristic of outpatient oncology patients with a recent diagnosis of metastases,42,62-65 the potential for bias in the study sample remains significant.
It is possible that participants were more comfortable with addressing prognostic issues than nonparticipants. There was a low prevalence of probable cases of anxiety and depression in the sample; however, the levels were similar to levels of anxiety and depression reported in a sample of 159 patients with advanced disease recruited in a large English study,49 which suggests that we did not have a particularly psychologically resilient sample.
The exemplar hope items and predictor variables included in this study were chosen on primarily empirical grounds (for example, those variables that had been found to be associated with general information preferences previously) rather than on theoretical grounds. Relevant theory (apart from the work on monitoring and blunting of Miller66) is lacking. There is a need for further development of appropriate theory to underpin future work in this field.
Nevertheless, this study provides the first data of its kind in this area, and it is hoped that the data will prove useful for clinicians struggling to communicate effectively with their patients with incurable cancer. Future research could use a more qualitative, in-depth approach to better explore some of the issues raised by these data.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We are grateful for the contribution of M. Boyer, F. Boyle, S. Clarke, J. Clayton, S. Crossing, D. Goldstein, F. Kirsten, M. Stockler, A. Sullivan, S. Crossing, and S.K. Lo, and we thank all participating oncologists and their patients.
NOTES
Supported by New South Wales Cancer Council, Australia.
Presented at the 15th International Symposium of the Multinational Association of Supportive Care in Cancer, Berlin, Germany, June 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Buckman R: Breaking bad news: Why is it still so difficult BMJ 288:1597-1599, 1984
2. Ptacek JT, Eberhardt TL: Breaking bad news: A review of the literature. JAMA 276:496-502, 1996
3. Ptacek JT, Ptacek JJ: Patients' perceptions of receiving bad news about cancer. J Clin Oncol 19:4160-4164, 2001
4. Christakis NA: Death Foretold: Prophecy and Prognosis in Medical Care. Chicago, IL, University of Chicago Press, 1999
5. Butow PN, Dowsett S, Hagerty RG, et al: Communicating prognosis to patients with metastatic disease: What do they really want to know Support Care Cancer 10:161-168, 2002
6. Baile WF, Glober GA, Lenzi R, et al: Discussing disease progression and end-of-life decisions. Oncology 13:1021-1031, 1999
7. Goldberg R, Guadagnoli E, Silliman RA, et al: Cancer patients' concerns: Congruence between patients and primary care physicians. J Cancer Educ 5:193-199, 1990
8. Ong LML, de Haes JCJM, Hoos AM, et al: Doctor-patient communication: A review of the literature. Soc Sci Med 40:903-918, 1995
9. Jenkins V, Fallowfield L, Saul J: Information needs of patients with cancer: Results from a large study in UK cancer centres. Br J Cancer 84:48-51, 2001
10. Lobb EA, Kenny DT, Butow PN, et al: Women's preferences for discussion of prognosis in early breast cancer. Health Expect 4:48-57, 2001
11. Derdiarian AK: Informational needs of recently diagnosed cancer patients. Nurs Res 35:276-281, 1986
12. Greisinger AJ, Lorimor RJ, Aday LA, et al: Terminally ill cancer patients: Their most important concerns. Cancer Pract 5:147-154, 1997
13. Sapir R, Catane R, Kaufman B, et al: Cancer patient expectations of and communication with oncologists and oncology nurses: The experience of an integrated oncology and palliative care service. Support Care Cancer 8:458-463, 2000
14. Northouse PG, Northouse LL: Communication and cancer: Issues confronting patients, health professionals, and family members. J Psychosocial Oncol 5:17-46, 1987
15. Leydon GM, Boulton M, Moynihan C, et al: Cancer patients' information needs and information seeking behaviour: In depth interview study. BMJ 320:909-912, 2000
16. Butow PN, McLean M, Dunn S, et al: The dynamics of change: Cancer patients' preferences for information, involvement and support. Ann Oncol 8:857-863, 1997
17. Goldberg RJ: Disclosure of information to adult cancer patients: Issues and update. J Clin Oncol 2:948-955, 1984
18. Chan A, Woodruff R: Communicating with patients with advanced cancer. J Palliat Care 13:29-33, 1997
19. Nuland SB: Hope and the Cancer Patient, in Nuland SB: How We Die. London, United Kingdom, Vintage, 1997, pp 222-242
20. Weeks JC, Cook EF, O'Day SJ, et al: Relationship between cancer patients' predictions of prognosis and their treatment preferences. JAMA 279:1709-1714, 1998
21. Kodish E, Post SG: Oncology and hope. J Clin Oncol 13:1817, 1995
22. Delvecchio Good MJ, Good BJ, Schaffer C, et al: American oncology and the discourse on hope. Cult Med Psychiatry 14:59-79, 1990
23. Christakis NA, Lamont EB, Smith JL, et al: Extent and determinants of error in doctors' prognoses in terminally ill patients: Prospective cohort study. BMJ 320:469-473, 2000
24. Ogden J, Fuks K, Gardner M, et al: Doctors expression of uncertainty and patient confidence. Patient Educ Couns 48:171-176, 2002
25. McIntosh J: Process of communication, information seeking and control associated with cancer: A selective review of the literature. Soc Sci Med 8:167-182, 1974
26. Cassileth BR, Zupkis RV, Sutton-Smith K, et al: Information and participation preferences among cancer patients. Ann Intern Med 92:832-836, 1980
27. Slevin ML: Talking about cancer: How much is too much Br J Hosp Med 38: 56:58-59, 1987
28. Fallowfield L, Ford S, Lewis S: No news is not good news: Information preferences of patients with cancer. Psychooncology 4:197-202, 1995
29. Kutner JS, Steiner JF, Corbett KK, et al: Information needs in terminal illness. Soc Sci Med 48:1341-1352, 1999
30. Nowotny ML: Every tomorrow, a vision of hope. J Psychosocial Oncol 9:117-126, 1991
31. Sardell AN, Trierweiler SJ: Disclosing the cancer diagnosis: Procedures that influence patient hopefulness. Cancer 72:3355-3365, 1993
32. Miller VD, Knapp ML: The Post Nuntio Dilemma: Approaches to Communicating With the Dying. Beverly Hills, CA, Sage, 1986
33. Girgis A, Sanson-Fisher RW: Breaking bad news: Consensus guidelines for medical practitioners. J Clinl Oncol 13:2449-2456, 1995
34. Girgis A, Sanson-Fisher RW: Breaking bad news. 1: Current best advice for clinicians. Behav Med 24:53-59, 1998
35. Kaplowitz SA, Campo S, Chui WT: Cancer patients' desire for communication of prognosis information. Health Commun 14:221-241, 2002
36. Wenrich MD, Curtis JR, Shannon SE, et al: Communicating with dying patients within the spectrum of medical care from terminal diagnosis to death. Arch Intern Med 161:868-874, 2001
37. National Health and Medical Research Council: Psychosocial Clinical Practice Guidelines: Information, Support and Counseling for Women With Breast Cancer. Canberra, Australia, National Health and Medical Research Council, 2000
38. Kaplowitz SA, Osuch JR, Safron D, et al: Physician communication with seriously ill cancer patients: Results of a survey of physicians, in de Vries B (ed): End of Life Issues: Interdisciplinary and Multidimensional Perspectives. New York, NY, Springer Publishing Company, 1999, pp 205-227
39. Parker PA, Baile WF, de Moor C, et al: Breaking bad news about cancer: Patients' preferences for communication. J Clin Oncol 19:2049-2056, 2001
40. Degner LF, Kristjanson LJ, Bowman D, et al: Information needs and decisional preferences in women with breast cancer. JAMA 277:1485-1492, 1997
41. Marwit SJ, Datson SL: Disclosure preferences about terminal illness: An examination of decision-related factors. Death Studies 26:1-20, 2002
42. Gattellari M, Voigt KJ, Butow PN, et al: When the treatment goal is not cure: Are cancer patients equipped to make informed decisions J Clin Oncol 20:503-513, 2002
43. Leighl N, Gattellari M, Butow P, et al: Discussing adjuvant cancer therapy. J Clin Oncol 19:1768-1778, 2001
44. Peteet JR, Abrams HE, Ross DM, et al: Presenting a diagnosis of cancer: Patients' views. J Fam Prac 32:577-581, 1991
45. MacCormack T, Simonian J, Lim J, et al: "Someone who cares": A qualitative investigation of cancer patients' experiences of psychotherapy. Psychooncology 10:52-65, 2001
46. Hagerty RG, Butow PN, Ellis PA, et al: Cancer patient preferences for communication of prognosis in the metastatic setting. J Clin Oncol 22:1721-1730, 2004
47. Krantz DS, Baum A, Wideman M: Assessment of preferences for self-treatment and information in health care. J Pers Soc Psychol 39:977-990, 1980
48. Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr Scand 67:361-370, 1983
49. Ibbotson T, Maguire P, Selby P, et al: Screening for anxiety and depression in cancer patients: The effects of disease and treatment. Eur J Cancer Care 30A:37-40, 1994
50. Bejelland I, Dahl AA, Haug TT, et al: The validity of the Hospital Anxiety and Depression Scale: An updated literature review. J Psychosom Res 52:69-77, 2002
51. Hosmer D, Lemeshow S: Applied Logistic Regression (ed 2). New York, NY, John Wiley and Sons, 2000
52. Girgis A, Sanson-Fisher RW, Schofield MJ: Is there consensus between breast cancer patients and providers on guidelines for breaking bad news Behav Med 25:69-77, 1999
53. Butow PN, Tattersall MHN, Goldstein D: Communication with cancer patients in culturally diverse societies. Ann N Y Acad Sci 809:317-329, 1997
54. Goldstein D, Thewes B, Butow PN: Communicating in a multicultural society II: Greek community attitudes towards cancer in Australia. Intern Med J 32:289-296, 2002
55. Blackhall LJ, Murphy ST, Frank G, et al: Ethnicity and attitudes toward patient autonomy. JAMA 274:820-825, 1995
56. Grassi L, Giraldi T, Messina EG, et al: Physicians' attitudes to and problems with truth-telling to cancer patients. Support Care Cancer 8:40-45, 2000
57. Huang X, Meiser B, Butow PN, et al: Attitudes and information needs of Chinese migrant cancer patients and their relatives. Aust N Z J Med 29:207-213, 1999
58. Bauman LJ, Gervey R, Siegel K: Factors associated with cancer patients' participation in support groups. J Psychosocial Oncol 10:1-20, 1992
59. Sanson-Fisher R, Girgis A, Boyes A, et al: The unmet supportive care needs of patients with cancer: Supportive Care Review Group. Cancer 88:226-237, 2000
60. National Breast Cancer Centre, Management of Advanced Breast Cancer Working Group: Clinical Practice Guidelines for the Management of Breast Cancer. Canberra, Australia, National Health and Medical Research Council, 2001, pg 217
61. Schofield PE, Beeney LJ, Thompson JF, et al: Hearing the bad news of a cancer diagnosis: The Australian melanoma patient's perspective. Ann Oncol 12:365-371, 2001
62. Mackillop WJ, Quirt CF: Measuring the accuracy of prognostic judgments in oncology. J Clin Epidemiol 50:21-29, 1997
63. Milsted RAV, Tattersall MHN, Fox RM, et al: Cancer chemotherapy: What have we achieved Lancet 1:1343-1346, 1980
64. Lamont EB, Christakis NA: Prognostic disclosure to patients with cancer near the end of life. Ann Intern Med 134:1096-1105, 2001
65. Chow E, Fung K, Panzarella T, et al: A predictive model for survival in metastatic cancer patients attending an outpatient palliative radiotherapy clinic. Int J Radiat Oncol Biol Phys 53:1291-1302, 2002
66. Miller SM: Monitoring versus blunting styles of coping with cancer influence the information patients want and need about their disease. Cancer 76:167-177, 1995(Rebecca G. Hagerty, Phyll)
ABSTRACT
PATIENTS AND METHODS: One hundred twenty-six (58%) of 218 patients invited onto the study participated. Eligible patients were the consecutive metastatic cancer patients of 30 oncologists, who were diagnosed within 6 weeks to 6 months before recruitment, over 18 years of age, and without known mental illness. Patients completed a postal survey measuring patient preferences for the manner of delivery of prognostic information, including how doctors might instill hope.
RESULTS: Ninety-eight percent of patients wanted their doctor to be realistic, provide an opportunity to ask questions, and acknowledge them as an individual when discussing prognosis. Doctor behaviors rated the most hope giving included offering the most up to date treatment (90%), appearing to know all there is to know about the patient's cancer (87%), and saying that pain will be controlled (87%). The majority of patients indicated that the doctor appearing to be nervous or uncomfortable (91%), giving the prognosis to the family first (87%), or using euphemisms (82%) would not facilitate hope. Factor analysis revealed six general styles and three hope factors; the most strongly endorsed styles were realism and individualized care and the expert/positive/collaborative approach. A range of demographic, psychological, and disease factors were associated with preferred general and hope-giving styles, including anxiety, information-seeking behavior, expected survival, and age.
CONCLUSION: The majority of patients preferred a realistic and individualized approach from the cancer specialist and detailed information when discussing prognosis.
INTRODUCTION
Prior research demonstrates that a clear majority of cancer patients in the Western world reports a preference for detailed information about their disease and expected outcome,9-13 although information needs can vary across different phases of the illness.14-16 Legal rulings have emphasized the responsibility of doctors to provide all necessary information in some jurisdictions.17 Nevertheless, patients often misunderstand the status of their disease and the aim of treatment13 and commonly overestimate their life expectancy.18 This may impact on decision making, particularly concerning anticancer treatment that may have side effects and reduce quality of life.19,20
Issues that concern clinicians about communicating life expectancy to patients include how much information to give, difficulties with prognostication, and disclosing prognostic uncertainty without increasing anxiety, reducing trust, and destroying hope.5,14,21-25 Others, however, believe expressing this prognostic ambiguity carefully to patients is one means of engendering hope4,19 and that providing information decreases anxiety associated with ignorance and uncertainty.14,26-28
The need for optimism and hope to be sustained in the process of honestly delivering bad news and a limited life expectancy is an ideal expressed by both doctors and patients.4,5,12,15,29 However, there is a delicate balance between fostering realistic hope and unethically creating unrealistic expectations of longevity.5,19,21,29 Furthermore, hope is a broad concept that can hold different meanings for each individual. Similarly, prognosis, a term often thought of as synonymous with life expectancy, encompasses broader issues surrounding the future course of the disease and how this may impact on the quality of life of the patient.19,27,30,31 Despite the complex issues surrounding this topic, apart from one study on communicating hope in the diagnosis of early-stage cancer,31 research is lacking in determining how patients define hope and how health professionals communicate hope to patients.14,32
Much of the communication literature has focused on how to break bad news.6,33,34 Few studies or guidelines have targeted the communication of prognosis specifically, and these few studies focus on either early-stage disease10,35 or palliative and end-of-life issues.12,13,29,36,37 There is a lack of evidence-based information on discussing a poor prognosis, a context that arguably demands more resources from both doctors and patients.5,38 Furthermore, although patient preferences for general information have been associated with demographic, psychological, and disease variables, for example, age,11,26,28,39 sex,11,16,39 religiosity,16 education level,26 involvement preference scores,16,40 anxiety and depression levels,35,41 and disease status,16,28 it has not yet been clarified whether such variables specifically influence metastatic cancer patient preferences for prognostic information.
In the current study, we aimed to identify the context and the way in which patients with incurable metastatic cancer wanted to be informed about their prognosis and to explore what features in the delivery of prognostic information they would experience as more or less hopeful. We hypothesized that the majority of patients would want detailed prognostic information as well as communication that facilitated hope. We also expected some variation in preferences according to patient demographic, disease, and psychological characteristics.
PATIENTS AND METHODS
Measures
Survey development. A written survey was used to elicit patient preferences for the content and format of prognostic discussion. To develop survey items, key themes were abstracted from the published literature, including an earlier qualitative study of breast cancer patients5 and analysis of audiotapes of initial oncology consultations.42,43 These themes included patient desire for information and involvement,9,16,26,40 honesty, and the provision of hope when prognosis is communicated.5,12,16,36,44 The instrument was reviewed by oncologists, health professionals, and members of a consumer advocacy group and piloted among 10 patients with metastatic cancer. Minor revisions were made in response to reviewers' and pilot participants' feedback.
Demographics and disease details. Participants were asked about the following: age, relationship status, occupation, highest educational level achieved, medical or allied health training, language spoken at home, parents' country of birth, whether or not they have children, and whether or not they have a religious denomination or spiritual belief. Oncologists provided patient disease information, including type of cancer, dates of diagnosis of primary and metastatic cancer, mode of treatment, and estimated survival.
Doctor style. Thirty-five doctor informational and supportive behaviors when discussing prognosis, which were derived from earlier studies of prognostic information preferences,10,31 were listed. Patients rated their response to each item on a 5-point Likert scale (agree completely, agree, neutral, disagree, and disagree completely). Examples of items included, "I would like my cancer specialist to: (a) check my understanding of what he/she has told me, (b) be realistic, (c) be optimistic, and (d) tell me about cancer support groups."
Definition of hope. Participants were provided with four exemplar hope definitions, using the words of patients from a previous study.45 The definitions were as follows: a feeling or expectation (1) that things can go well; (2) that because one thing has gone wrong, it doesn't mean that other things will not go well; (3) that you have just as good chances (if not better) as the next person of having the best outcome; (4) that you can still enjoy a good quality of life, even if life expectancy is uncertain; or (5) other. Participants were asked to indicate what hope means to them in their current situation; they could select more than one item and/or write their own definition under other. Participants were also asked a separate open question about what else (apart from doctor communication) helps them to be hopeful.
What patients find hopeful. Thirty doctor behaviors that might convey or discourage hope, such as "gave me survival statistics," "appeared nervous or uncomfortable," and "was occasionally humorous," were listed. The items were derived from Sardell and Trierweiler31 and our earlier qualitative study exploring ideal prognostic communication in the metastatic setting.5 Participants were asked to indicate on a 5-point Likert scale how hopeful each of the behaviors would make them feel (ie, very hopeful, hopeful, neutral, not hopeful, and hopeless).
Preference for prognostic information, timing, and manner of presentation. Patients were asked to indicate their preference for specific prognostic facts, mode of presentation of statistics, and the timing of prognostic information and who should initiate the discussion. These results have been presented elsewhere.46
Information and involvement preferences. Participants' preferences for information and involvement were elicited using the seven-item binary Information subscale of the Krantz Health Opinion Survey ({alpha} = .74 in this sample).47
Depression and anxiety. Levels of depression and anxiety were measured by the 14-item Hospital Anxiety and Depression Scale self-assessment tool devised by Zigmond and Snaith48 ({alpha} = .84 and {alpha} = .81 in this sample).49,50
Statistical Analyses
Open questions and patient preferences for the doctor's style and hope-giving items were analyzed using descriptive statistics. Doctor style and hope items were then entered separately into two factor analyses using varimax rotation with Kaiser normalization. Total scores for the identified factors were calculated by summing items with weights more than 0.3 on each factor. If an item loaded onto more than one factor, it was included in the factor on which it had the highest weight and excluded from the other factor(s). Associations between demographic, psychological, and disease variables (described earlier) and each factor (the dependent variable) were explored in univariate analyses (using linear regression, t tests, and ANOVA). Associated variables that were found to be significant at the 0.25 level in univariate analyses51 were entered into binary logistic regressions. Because anxiety and depression scores were highly correlated (Pearson's correlation, r = 0.57; P < .01), only one of these scores (the more significant on univariate analysis) was entered into multivariate analyses.
RESULTS
Of the 218 patients approached to participate, 10 were ineligible, and 22 refused. Of the remainder, 126 patients completed the survey. The most common reason cited for not completing the survey was ill health. No significant differences were found between those who completed the survey and those who did not complete the survey on the variables of age (F2,194 = 0.180; P = .672), sex ({chi}2 = 0.294; df = 1; P = .588), clinic where recruited (metropolitan or nonmetropolitan; {chi}2 = 2.540; df = 1; P = .111), type of cancer ({chi}2 = 4.341; df = 4; P = .362), and time since diagnosis of metastatic disease (F2,192 = 0.008; P = .930).
Fifty-six percent of participants were male, whereas 25% had breast cancer, 18% had colorectal cancer, 16% had prostate cancer, 10% had lung cancer, and 31% had other cancer types. The mean time since diagnosis of metastases was 13 weeks (range, 1 to 39 weeks; standard deviation [SD], 8.7 months). The majority of patients had an expected survival estimated by their oncologist of months (42%), or years (42%) and were receiving either systemic anticancer therapy or radiotherapy (92%; Table 1).
Information and Involvement Preferences
The average score on the Krantz Health Opinion Survey: Information Preferences Subscale was 3.9 (SD, 2.15), which is in the medium score range, indicating that, overall, participants did not have either a particularly high or low desire to ask questions or to be informed about medical decisions.46
Anxiety and Depression
Twenty-three percent and 19% of patients fell into the possible case range (scores, 8 to 10) for anxiety and depression respectively. Ten percent and 7% of patients fell into the probable case range (scores, 11 to 21) for anxiety or depression, respectively.48
Preference for Doctor Style
Of the 35 doctor behaviors listed, the most preferred included for the doctor to be realistic, provide an opportunity to ask questions, and acknowledge the patient as an individual when discussing prognosis (all 98%; Table 2). Less-preferred items were for the oncologist to discuss the patient's financial situation (37% of patients), give the patient his or her prognosis with another medical person present (28% of patients), give the patient an audiotape of their discussion (19% of patients), and give the patient his or her prognosis over the telephone (14% of patients; Table 2).
Hope-Giving Behavior
Of the 30 listed doctor behaviors, those rated as being most hopeful included being offered the most up to date treatment (90%), the oncologist appearing to know all there is to know about the patient's cancer (87%), the occasional use of humor (80%), being told that the pain will be controlled (87%), and being told all treatment options (83%; Table 3).
Behaviors that were rated as causing the patient to feel not hopeful or hopeless were the doctor appearing to be nervous or uncomfortable (91%), giving the prognosis to the family first (87%), the use of euphemisms (82%), avoiding talking about cancer and only discussing treatment (75%), and giving the good news first and then the bad news (72%); however, almost half of the patients rated this last item as neutral (46%).
Thirty percent of participants rated giving statistics about how long they will live as likely to make them feel hopeful; however, similar percentages of patients rated this item as not hopeful (32%) and neutral (38%). Similarly, participants rated expressing uncertainty about the course of the cancer (ie, said that the course of the cancer cannot be predicted) evenly across the options, with 35%, 30%, and 35% rating this as hopeful, not hopeful, and neutral, respectively (Table 3).
Patient Definition of Hope
Some participants ticked one or more of the four exemplars of hope; however, the majority (62%) wrote their own definition with or without ticking an exemplar. The most commonly endorsed definition was "that you can still enjoy a good quality of life even if life expectancy is uncertain" (19%) and "a feeling or expectation that things can go well" (14%). Fewer endorsed the items of "that you have just as much chance as the next person of having a good outcome" (3%) and "that because one thing has gone wrong, it doesn't mean that other things will go wrong" (3%).
Of the alternatives proposed, the most common themes were quality of life and fulfillment of goals (23%; for example, "To get on with life, make sure you make the most of it for as long as you can; set a distant goal and work like hell to get there"), receiving the best treatment that would provide the best cancer and symptom control (8%; for example, "Hope for me is a general expectation that everyone involved in my treatment is doing their best"), and the hope for cure or remission (8%; for example, "... hope that they may find a cure or extend my life").
General Factors Influencing Hopefulness
One hundred four (83%) of the 126 respondents completed the open question on general factors influencing hopefulness. The majority (87%) stated that family helped them to be hopeful. Other factors cited were religious beliefs or spirituality (28%), friends (24%), a positive attitude (13%), his or her children or grandchildren (12%) or partner (10%), and scientific advances in cancer care (9%).
Factor Analysis
Factor analysis of the 35 doctor style items revealed six factors that accounted for approximately 57% of the total variance (Table 4). These styles were as follows: factor 1, realism and individualized care (providing realistic and direct information tailored to the individual); factor 2, emotional support (providing information on support services and an openness to discuss patients' fears and concerns); factor 3, facilitation of coping with dying (displaying openness to discuss concerns about dying and providing information on palliative care services); factor 4, provision of information (ensuring patient understanding and providing information materials such as publications and audiotape of consultation); factor 5, emphasizing all options (providing information on complementary therapies, discussing optimistic future scenarios, and suggesting a second opinion); and factor 6, a personal approach (sitting next to the patient and sharing personal information; Table 4).
Three factors were identified among the 30 hope items, which accounted for approximately 54% of the variance (Table 5). These were as follows: factor 1, expert/positive/collaborative (expertise, humor, and inclusion of patient as part of the team); factor 2, avoidant (avoiding or appearing uncomfortable about discussing the cancer, using euphemisms, and giving the prognosis to others first); and factor 3, empathic (expressing one's own feelings or asking the patient about his or her own reaction to the prognosis; Table 5). One item ("asked me if I would like a second opinion") did not weigh on any of the factors.
The strength of preference for these styles was determined by calculating the total scores for each factor. Because different factors included varying numbers of items, the totals were normalized to allow direct comparison. Thus, scores could range from 20 to 100. The most strongly endorsed doctor style was realism and individualized care (mean score, 88.9; SD, 9.6), and the least endorsed doctor style was personal approach (mean score, 59.9; SD, 14.1). The most strongly endorsed hope-giving style was the expert/positive/collaborative approach (mean score, 78.4; SD, 14.4).
Variables Significantly Associated With Preferences
Doctor style. Table 6 lists the means scores on the general doctor style factors by those variables found to be significant in multivariate analyses. Higher anxiety scores (t102 = 2.463; P = .02), having a partner (t102 = 2.00; P = .048), English speaking at home (t102 = –2.363; P = .02), expected survival of years as opposed to weeks or months (t102 = 2.23; P = .03), and longer time since diagnosis of metastatic disease (t102 = 2.09; P = .04) were found to significantly and independently predict preference for the oncologist to have a realistic and individualized approach (factor 1). Higher Krantz Information Subscale scores (t102 = 2.88; P = .005) were also significantly associated with this factor.
Age was the only variable significantly associated with preference for factor 2, with younger patients more likely to prefer an emotionally supportive approach (t103 = –4.53; P < .001). Higher anxiety scores (t115 = 2.80; P = .006) and female sex (t115 = –2.68; P = .008) predicted preference for the facilitation of the coping with dying approach (factor 3). Patients who scored higher on information seeking (t108 = 3.10; P = .002) were more likely to prefer the informative approach (factor 4). Younger patients (t114 = –3.65; P < .001) were more likely to prefer the emphasizing all options approach (factor 5). Patients who stated they had a religious belief (t109 = 3.40; P = .001) were more likely to prefer the personal approach (factor 6; Table 6).
Hope-giving style. Table 6 lists the mean scores on the hope-giving factors by those variables found to be significant in multivariate analyses. Older patients rated the expert/positive/collaborative (factor 1) and empathic (factor 3) approaches as significantly more hope giving than younger patients (factor 1: t97 = 2.08; P = .04; factor 3: t102 = 3.18; P = .002). Anxiety was also significantly associated with these factors (factor 1: t97 = 2.20; P = .03; factor 3: t102 = 2.43; P = .02; Table 6). None of the variables explored were found to be significantly associated with the avoidant approach (factor 2).
DISCUSSION
There was some variability in preferences according to patient characteristics, which allows some insight into who may prefer which style. Overall, younger, anxious patients placed a stronger focus on emotional support for themselves and their families. This is understandable given that these patients are more likely to have a dependent family whose support needs during their illness and after their death may be a primary concern for them. These patients seemed to have high needs and high expectations for medical care. This is in accordance with other studies that report that younger patients have a preference for more information and have high unmet needs.26,52
Patients who had been diagnosed with metastatic disease for a longer period and those with longer expected survival were more likely to want realism (factor 1). Perhaps patients are better able to cope with realism over time. In earlier work, we found that patients who are in routine follow-up, without relapse, develop stronger preferences for involvement in decision making over time.16 A longer expected survival would be understandably easier to discuss frankly. Interestingly, because these data were derived from oncologists' prognostic estimates, it would seem that these patients had some awareness of their own prognosis.
English speakers were also more likely to prefer realism. Those who speak another language at home are possibly influenced by a culture where avoidance and paternalism is more common. It is well documented that, in some cultures, communication of prognosis is viewed as harmful and brutal.53-57 Overall, however, the majority of patients in all groups preferred a realistic approach. Generalizations concerning a group linked by cultural factors can lead to stereotyping. Thus, information about beliefs and practices within particular cultures and groups never obviates the need for exploring individual preferences and needs. Notably, patients who did have a religious belief wanted this to be acknowledged by the cancer specialist in the context of discussing prognosis.
Only two variables were associated with preferences for hope-giving behaviors. Older patients were more likely to rate both the expert and empathic approaches as more hope giving. However, older patients were less likely to prefer the emotionally supportive approach as a general doctor style. It seems that older patients want direct emotional support from their doctor as opposed to more formal organized support. This supports previous findings in the literature.58,59 Anxiety was also associated with these factors, although, on examination of the means, it seems that the relationship may be curvilinear, with patients scoring high or low showing less interest in these approaches. This is difficult to interpret, and further research may clarify these relationships.
Some of the results are not consistent with current practice guidelines.37,60 For example having another medical person present during prognostic discussion and offering an audiotape of the discussion were not endorsed by participants in this study. A personal approach that involved sitting next to the patient or sharing some personal information was also less valued. These results are similar to those of Schofield et al61 who reported that patient preferences do not always reflect published consensus guidelines, which may need to be modified in light of this feedback.
There were several limitations to the study. Cultural factors have not been well considered because non-English speakers were excluded from the sample. It is known that cancer patients from other cultures have different views and information needs.53-57 It is also not certain, although probable, that these results would be relevant to the cancer populations of Britain and North America. Furthermore, some subgroups in the sample (eg, particular cancers) are quite small, and therefore, this study did not allow detection of differences between these groups. A larger sample may have produced more significant and reliable results.
Only 30 of 106 oncologists invited to participate in this study took part. Although no demographic or practice differences were observed between those who accepted and those who actively refused to participate, more subtle differences (such as interest in communication) that we did not measure may have differentiated these groups and, therefore, influenced the results. Furthermore, no data were available on oncologists who did not respond at all.
Not all oncologists may have complied with the instruction to recruit consecutive eligible patients. They may have invited patients who felt more comfortable with their diagnosis or who were less symptomatic; few patients were receiving symptomatic care only. Although this is characteristic of outpatient oncology patients with a recent diagnosis of metastases,42,62-65 the potential for bias in the study sample remains significant.
It is possible that participants were more comfortable with addressing prognostic issues than nonparticipants. There was a low prevalence of probable cases of anxiety and depression in the sample; however, the levels were similar to levels of anxiety and depression reported in a sample of 159 patients with advanced disease recruited in a large English study,49 which suggests that we did not have a particularly psychologically resilient sample.
The exemplar hope items and predictor variables included in this study were chosen on primarily empirical grounds (for example, those variables that had been found to be associated with general information preferences previously) rather than on theoretical grounds. Relevant theory (apart from the work on monitoring and blunting of Miller66) is lacking. There is a need for further development of appropriate theory to underpin future work in this field.
Nevertheless, this study provides the first data of its kind in this area, and it is hoped that the data will prove useful for clinicians struggling to communicate effectively with their patients with incurable cancer. Future research could use a more qualitative, in-depth approach to better explore some of the issues raised by these data.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We are grateful for the contribution of M. Boyer, F. Boyle, S. Clarke, J. Clayton, S. Crossing, D. Goldstein, F. Kirsten, M. Stockler, A. Sullivan, S. Crossing, and S.K. Lo, and we thank all participating oncologists and their patients.
NOTES
Supported by New South Wales Cancer Council, Australia.
Presented at the 15th International Symposium of the Multinational Association of Supportive Care in Cancer, Berlin, Germany, June 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Buckman R: Breaking bad news: Why is it still so difficult BMJ 288:1597-1599, 1984
2. Ptacek JT, Eberhardt TL: Breaking bad news: A review of the literature. JAMA 276:496-502, 1996
3. Ptacek JT, Ptacek JJ: Patients' perceptions of receiving bad news about cancer. J Clin Oncol 19:4160-4164, 2001
4. Christakis NA: Death Foretold: Prophecy and Prognosis in Medical Care. Chicago, IL, University of Chicago Press, 1999
5. Butow PN, Dowsett S, Hagerty RG, et al: Communicating prognosis to patients with metastatic disease: What do they really want to know Support Care Cancer 10:161-168, 2002
6. Baile WF, Glober GA, Lenzi R, et al: Discussing disease progression and end-of-life decisions. Oncology 13:1021-1031, 1999
7. Goldberg R, Guadagnoli E, Silliman RA, et al: Cancer patients' concerns: Congruence between patients and primary care physicians. J Cancer Educ 5:193-199, 1990
8. Ong LML, de Haes JCJM, Hoos AM, et al: Doctor-patient communication: A review of the literature. Soc Sci Med 40:903-918, 1995
9. Jenkins V, Fallowfield L, Saul J: Information needs of patients with cancer: Results from a large study in UK cancer centres. Br J Cancer 84:48-51, 2001
10. Lobb EA, Kenny DT, Butow PN, et al: Women's preferences for discussion of prognosis in early breast cancer. Health Expect 4:48-57, 2001
11. Derdiarian AK: Informational needs of recently diagnosed cancer patients. Nurs Res 35:276-281, 1986
12. Greisinger AJ, Lorimor RJ, Aday LA, et al: Terminally ill cancer patients: Their most important concerns. Cancer Pract 5:147-154, 1997
13. Sapir R, Catane R, Kaufman B, et al: Cancer patient expectations of and communication with oncologists and oncology nurses: The experience of an integrated oncology and palliative care service. Support Care Cancer 8:458-463, 2000
14. Northouse PG, Northouse LL: Communication and cancer: Issues confronting patients, health professionals, and family members. J Psychosocial Oncol 5:17-46, 1987
15. Leydon GM, Boulton M, Moynihan C, et al: Cancer patients' information needs and information seeking behaviour: In depth interview study. BMJ 320:909-912, 2000
16. Butow PN, McLean M, Dunn S, et al: The dynamics of change: Cancer patients' preferences for information, involvement and support. Ann Oncol 8:857-863, 1997
17. Goldberg RJ: Disclosure of information to adult cancer patients: Issues and update. J Clin Oncol 2:948-955, 1984
18. Chan A, Woodruff R: Communicating with patients with advanced cancer. J Palliat Care 13:29-33, 1997
19. Nuland SB: Hope and the Cancer Patient, in Nuland SB: How We Die. London, United Kingdom, Vintage, 1997, pp 222-242
20. Weeks JC, Cook EF, O'Day SJ, et al: Relationship between cancer patients' predictions of prognosis and their treatment preferences. JAMA 279:1709-1714, 1998
21. Kodish E, Post SG: Oncology and hope. J Clin Oncol 13:1817, 1995
22. Delvecchio Good MJ, Good BJ, Schaffer C, et al: American oncology and the discourse on hope. Cult Med Psychiatry 14:59-79, 1990
23. Christakis NA, Lamont EB, Smith JL, et al: Extent and determinants of error in doctors' prognoses in terminally ill patients: Prospective cohort study. BMJ 320:469-473, 2000
24. Ogden J, Fuks K, Gardner M, et al: Doctors expression of uncertainty and patient confidence. Patient Educ Couns 48:171-176, 2002
25. McIntosh J: Process of communication, information seeking and control associated with cancer: A selective review of the literature. Soc Sci Med 8:167-182, 1974
26. Cassileth BR, Zupkis RV, Sutton-Smith K, et al: Information and participation preferences among cancer patients. Ann Intern Med 92:832-836, 1980
27. Slevin ML: Talking about cancer: How much is too much Br J Hosp Med 38: 56:58-59, 1987
28. Fallowfield L, Ford S, Lewis S: No news is not good news: Information preferences of patients with cancer. Psychooncology 4:197-202, 1995
29. Kutner JS, Steiner JF, Corbett KK, et al: Information needs in terminal illness. Soc Sci Med 48:1341-1352, 1999
30. Nowotny ML: Every tomorrow, a vision of hope. J Psychosocial Oncol 9:117-126, 1991
31. Sardell AN, Trierweiler SJ: Disclosing the cancer diagnosis: Procedures that influence patient hopefulness. Cancer 72:3355-3365, 1993
32. Miller VD, Knapp ML: The Post Nuntio Dilemma: Approaches to Communicating With the Dying. Beverly Hills, CA, Sage, 1986
33. Girgis A, Sanson-Fisher RW: Breaking bad news: Consensus guidelines for medical practitioners. J Clinl Oncol 13:2449-2456, 1995
34. Girgis A, Sanson-Fisher RW: Breaking bad news. 1: Current best advice for clinicians. Behav Med 24:53-59, 1998
35. Kaplowitz SA, Campo S, Chui WT: Cancer patients' desire for communication of prognosis information. Health Commun 14:221-241, 2002
36. Wenrich MD, Curtis JR, Shannon SE, et al: Communicating with dying patients within the spectrum of medical care from terminal diagnosis to death. Arch Intern Med 161:868-874, 2001
37. National Health and Medical Research Council: Psychosocial Clinical Practice Guidelines: Information, Support and Counseling for Women With Breast Cancer. Canberra, Australia, National Health and Medical Research Council, 2000
38. Kaplowitz SA, Osuch JR, Safron D, et al: Physician communication with seriously ill cancer patients: Results of a survey of physicians, in de Vries B (ed): End of Life Issues: Interdisciplinary and Multidimensional Perspectives. New York, NY, Springer Publishing Company, 1999, pp 205-227
39. Parker PA, Baile WF, de Moor C, et al: Breaking bad news about cancer: Patients' preferences for communication. J Clin Oncol 19:2049-2056, 2001
40. Degner LF, Kristjanson LJ, Bowman D, et al: Information needs and decisional preferences in women with breast cancer. JAMA 277:1485-1492, 1997
41. Marwit SJ, Datson SL: Disclosure preferences about terminal illness: An examination of decision-related factors. Death Studies 26:1-20, 2002
42. Gattellari M, Voigt KJ, Butow PN, et al: When the treatment goal is not cure: Are cancer patients equipped to make informed decisions J Clin Oncol 20:503-513, 2002
43. Leighl N, Gattellari M, Butow P, et al: Discussing adjuvant cancer therapy. J Clin Oncol 19:1768-1778, 2001
44. Peteet JR, Abrams HE, Ross DM, et al: Presenting a diagnosis of cancer: Patients' views. J Fam Prac 32:577-581, 1991
45. MacCormack T, Simonian J, Lim J, et al: "Someone who cares": A qualitative investigation of cancer patients' experiences of psychotherapy. Psychooncology 10:52-65, 2001
46. Hagerty RG, Butow PN, Ellis PA, et al: Cancer patient preferences for communication of prognosis in the metastatic setting. J Clin Oncol 22:1721-1730, 2004
47. Krantz DS, Baum A, Wideman M: Assessment of preferences for self-treatment and information in health care. J Pers Soc Psychol 39:977-990, 1980
48. Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr Scand 67:361-370, 1983
49. Ibbotson T, Maguire P, Selby P, et al: Screening for anxiety and depression in cancer patients: The effects of disease and treatment. Eur J Cancer Care 30A:37-40, 1994
50. Bejelland I, Dahl AA, Haug TT, et al: The validity of the Hospital Anxiety and Depression Scale: An updated literature review. J Psychosom Res 52:69-77, 2002
51. Hosmer D, Lemeshow S: Applied Logistic Regression (ed 2). New York, NY, John Wiley and Sons, 2000
52. Girgis A, Sanson-Fisher RW, Schofield MJ: Is there consensus between breast cancer patients and providers on guidelines for breaking bad news Behav Med 25:69-77, 1999
53. Butow PN, Tattersall MHN, Goldstein D: Communication with cancer patients in culturally diverse societies. Ann N Y Acad Sci 809:317-329, 1997
54. Goldstein D, Thewes B, Butow PN: Communicating in a multicultural society II: Greek community attitudes towards cancer in Australia. Intern Med J 32:289-296, 2002
55. Blackhall LJ, Murphy ST, Frank G, et al: Ethnicity and attitudes toward patient autonomy. JAMA 274:820-825, 1995
56. Grassi L, Giraldi T, Messina EG, et al: Physicians' attitudes to and problems with truth-telling to cancer patients. Support Care Cancer 8:40-45, 2000
57. Huang X, Meiser B, Butow PN, et al: Attitudes and information needs of Chinese migrant cancer patients and their relatives. Aust N Z J Med 29:207-213, 1999
58. Bauman LJ, Gervey R, Siegel K: Factors associated with cancer patients' participation in support groups. J Psychosocial Oncol 10:1-20, 1992
59. Sanson-Fisher R, Girgis A, Boyes A, et al: The unmet supportive care needs of patients with cancer: Supportive Care Review Group. Cancer 88:226-237, 2000
60. National Breast Cancer Centre, Management of Advanced Breast Cancer Working Group: Clinical Practice Guidelines for the Management of Breast Cancer. Canberra, Australia, National Health and Medical Research Council, 2001, pg 217
61. Schofield PE, Beeney LJ, Thompson JF, et al: Hearing the bad news of a cancer diagnosis: The Australian melanoma patient's perspective. Ann Oncol 12:365-371, 2001
62. Mackillop WJ, Quirt CF: Measuring the accuracy of prognostic judgments in oncology. J Clin Epidemiol 50:21-29, 1997
63. Milsted RAV, Tattersall MHN, Fox RM, et al: Cancer chemotherapy: What have we achieved Lancet 1:1343-1346, 1980
64. Lamont EB, Christakis NA: Prognostic disclosure to patients with cancer near the end of life. Ann Intern Med 134:1096-1105, 2001
65. Chow E, Fung K, Panzarella T, et al: A predictive model for survival in metastatic cancer patients attending an outpatient palliative radiotherapy clinic. Int J Radiat Oncol Biol Phys 53:1291-1302, 2002
66. Miller SM: Monitoring versus blunting styles of coping with cancer influence the information patients want and need about their disease. Cancer 76:167-177, 1995(Rebecca G. Hagerty, Phyll)