Background
Prognostic beliefs of patients with advanced cancer have been examined extensively.1–5 Existing research consistently notes the high prevalence of inaccurate prognostic beliefs,4 revealing a misalignment between patients’ expressed desires to understand their prognosis and the prognostic beliefs they hold.6–9 Previous work has also shown that patients reporting inaccurate prognostic beliefs receive more aggressive treatments and incur greater healthcare costs at the end of life.10–15 Resultantly, these patients may not receive goal-concordant care at the end of life.
Gaps in patient–physician communication, particularly physicians’ reluctance to discuss prognosis, have been cited frequently as a cause of inaccurate beliefs.8,16 However, patients’ beliefs can also be shaped by their own illness experience; if it does not align with the prognostic information they receive, the resultant dissonance may affect how they process and internalize prognostic information. This dissonance, alongside its related coping mechanisms and cognitive biases,16,17 may cause patients to confound their health status with their prognosis. A previous study showed that patients with poorer health were more likely to accept their prognosis and vice versa.18 As patients’ life expectancy increases with advanced therapies and their health status fluctuates over this longer period, it is possible that their prognostic beliefs also change over time. However, most research on prognostic beliefs has been cross-sectional. Although some longitudinal explorations of changes in prognostic beliefs have emerged recently,19–24 they assessed beliefs during the end-of-life-period; changes over a longer duration have not been studied.21,25
We aimed to assess the extent to which prognostic beliefs change and the direction of these changes in patients with metastatic cancer over a period of up to 54 months. We focused on patients’ perceptions about disease curability, henceforth referred to as “prognostic belief.” We hypothesized that patients’ prognostic belief will be unstable over time and will change not only toward accurate belief but also toward inaccurate and uncertain belief. We also hypothesized that patients with better health (lower symptom burden, greater physical and functional well-being) will be more likely to have an inaccurate belief, whereas those with recent experience of a sudden deterioration in health or unplanned hospital admission (UHA) will be more likely to have an accurate prognostic belief. Last, we posited that patients’ prognostic belief will be more likely to change toward accuracy as their health deteriorates.
Methods
Study Setting, Design, and Participants
We analyzed data from the Cost of Medical Care of Patients with Advanced Serious Illness in Singapore (COMPASS) study; study details are reported in the previously published study protocol.26 COMPASS is a prospective cohort of patients with advanced cancer in Singapore, a developed and rapidly aging Southeast Asian country. The study recruited 600 adult patients with stage IV solid metastatic cancer from July 2016 through March 2018, who were then surveyed every 3 months until death or dropout. Additional eligibility criteria are presented in supplemental eFigure 1 (available with this article at JNCCN.org). Surveys were conducted by trained interviewers in person, over the phone, or online in the patients’ preferred language (English, Mandarin, or Malay). Patients were asked about their prognostic beliefs every 9 months.
We merged survey data with administrative data from the 2 recruiting hospitals, which treat approximately 70% of all public sector oncology cases,26 and 6 community hospices out of 9 that provide home or inpatient hospice care in the country.27
All patients provided signed consent and the study was approved by the SingHealth Centralised Institutional Review Board (2015/2781).
Outcome Variables
Prognostic Belief/Change in Prognostic Belief
At each time point, patients were asked whether they were currently undergoing treatment for their illness. Patients who responded affirmatively were then asked about their beliefs about disease curability. Based on their response to the question, “Do you think your current treatment will cure your disease?” we classified patients’ prognostic beliefs as accurate (“No”), inaccurate (“Yes”), or uncertain (“Unsure”). This question, adapted from the Cancer Care Outcomes Research and Surveillance study,28 is commonly used to assess patients’ understanding of prognosis.1 Because metastatic cancers remain largely incurable despite advancements in therapies and variability in prognosis based on the cancer type, we considered acknowledgment of disease incurability by patients undergoing treatment as reporting an accurate prognostic belief.
Change in prognostic belief was assessed between 2 consecutive responses and classified into 1 of 4 categories, based on the category to which patients changed: no change, change to accurate belief, change to inaccurate belief, and change to uncertain belief. In a sensitivity analysis, we further disaggregated the “no change” category (supplemental eTable 1).
Independent Variables
We used 4 variables to assess health status: symptom burden, physical and functional well-being scores, and recent UHA.
Symptom Burden Score
We calculated the symptom burden score based on patients’ rating of the severity of 10 symptoms commonly experienced toward the end of life, adapted from the Functional Assessment of Chronic Illness Therapy – Palliative Care (FACIT-Pal) measure.29 Patients were asked to rate their experience of symptoms such as pain, shortness of breath, and weight loss in the past week on a Likert scale of 0 (not at all) to 4 (very much). The score is a sum of the ratings for patients who rated at least 5 symptoms, adjusted by the number of responses. It ranges from 0 to 40, with a higher score indicating greater symptom burden.
Physical/Functional Well-Being Score
We used the physical and functional well-being subscales of the Functional Assessment of Cancer Therapy – General (FACT-G V4)30 tool, which is commonly used to assess quality of life for patients with cancer. Both subscales range from 0 to 28, with a higher score indicating greater well-being.
Change in Health Status
We assessed change in health status in 3 ways: change in symptom burden, change in physical well-being, and change in functional well-being scores. For each variable, the change was calculated as the difference in scores between 2 consecutive responses, adjusted for the number of months in the intervening period.
Recent UHA
We defined a recent UHA as an accident and emergency visit that progressed to an inpatient admission, with the admission date occurring in the 3 months preceding each survey. We identified these using administrative inpatient and emergency department data from participating hospitals.
Covariates
We controlled for age, education (primary and below; secondary; and tertiary and above), type of cancer (breast, gastrointestinal, genitourinary, respiratory, and other), survey response number, and palliative care visits. Palliative care visit is a binary variable defined as the use of palliative care services (hospital outpatient palliative care or inpatient/home/day care at a community hospice) in the 3 months preceding each survey, identified using administrative records. Such visits may not only have mitigated patients’ symptom burden but also may have exposed them to new information, explicit or implicit, about their prognosis.31,32 Similarly, because we expected that patients would both get sicker and learn more about their disease over time, we also controlled for the survey response number.
Statistical Analysis
We limited the analytical sample to patients who reported their prognostic beliefs at least once during the survey period (n=587). We first described the distribution of prognostic beliefs and changes in prognostic beliefs at each time point. Then, we estimated the association between the 2 outcome variables, prognostic belief and change in prognostic belief, and health status. To accommodate these nominal categorical outcome variables, we used a set of multinomial logistic regression models for each outcome. In the second set of models examining change in prognostic belief, we limited the sample to patients who reported their prognostic beliefs at least twice (n=338).
The main independent variables in the first set of models (outcome: prognostic belief) were symptom burden, physical well-being, and functional well-being scores—which were included in separate models due to high correlation—and recent UHA. We controlled for patients’ survey response number, palliative care visits, age, education, and type of cancer in all models.
The main independent variables in the second set of models (outcome: change in prognostic belief) included changes in symptom burden, physical well-being, and functional well-being scores, which were tested in separate models. We controlled for all aforementioned covariates except UHA and palliative care visits because they only accounted for the 3 months preceding the survey, which did not correspond to the 9 months or longer between consecutive surveys.
Hausman tests confirmed the validity of using random effects to model individual heterogeneity (P<.05) for all but 2 models; we used random effects models in the main results and tested the sensitivity of coefficients to fixed effects (supplemental eTable 2A, B).
Missing responses comprised <5% of the sample (supplemental eFigure 2), and nonresponse was not significantly associated with any indicator of health status. Nonetheless, we conducted complete case analysis to check the sensitivity of our results to missing responses (supplemental eTable 3A [n=482]; eTable 3B [n=233]). Stata, version 17.0 (StataCorp LLC) was used to conduct all analysis.
Results
Demographic details of the sample are described in Table 1. The mean age of patients at baseline was 61 years. The sample was composed largely of patients with respiratory (27.9%) and gastrointestinal (25.7%) cancers. In the 3 months preceding the baseline survey, 13.5% of patients experienced UHA, and 4.3% had a palliative care visit. At baseline, 28.9% of patients reported accurate belief, an equivalent proportion reported inaccurate belief, and the remaining 42.6% reported uncertain belief. Over time, the proportion of patients with an accurate belief increased: nearly 80% reported an accurate belief in month 54 (Figure 1). That said, patients’ prognostic beliefs changed at each time point, both toward and away from an accurate belief (supplemental eFigure 3), as summarized in Figure 2. On average, at each time point, 21% of patients changed to an uncertain or inaccurate belief, and 20% changed to an accurate belief (Figure 2). Of the patients who answered the survey at least twice during the study period, 24% changed from having an accurate belief to an uncertain or inaccurate belief at some point during the follow-up. Notably, patients’ prognostic beliefs changed up to 6 times, and on average, 28% of each patients’ total responses constituted a change in belief from the previous time point.
Sample Characteristics at Baseline
Prognostic beliefs at each time point.
Citation: Journal of the National Comprehensive Cancer Network 21, 10; 10.6004/jnccn.2023.7052
Changes in prognostic belief from previous time point.
Citation: Journal of the National Comprehensive Cancer Network 21, 10; 10.6004/jnccn.2023.7052
Consistent with our hypotheses, patients who experienced greater symptom burden were less likely to report an inaccurate (relative risk ratio [RRR], 0.87; 95% CI, 0.84–0.90) or uncertain belief (RRR, 0.90; 95% CI, 0.87–0.92) compared with an accurate prognostic belief (Table 2). Analogously, patients with greater physical and functional well-being were more likely to report an inaccurate (RRR, 1.10; 95% CI, 1.06–1.15 [model 2] and RRR, 1.10; 95% CI, 1.06–1.14 [model 3], respectively) or an uncertain belief (RRR, 1.08; 95% CI, 1.05–1.12 [model 2]) compared with an accurate prognostic belief. Contrary to our hypothesis, patients who experienced a recent UHA were also more likely to have an inaccurate (RRR, 2.71 [model 1]; 2.40 [model 2]; 2.24 [model 3]) or uncertain belief (RRR, 2.34 [model 1]; 2.09 [model 2]; 1.81 [model 3]) than an accurate belief.
Association of Prognostic Belief With Health Status
Table 3 shows that, in line with our hypothesis, an increase in symptom burden was associated with a greater risk of patients changing toward accurate belief (RRR, 1.75; 95% CI, 1.33–2.31) compared with reporting no change. Improvements in physical and functional well-being were also negatively associated with change to an accurate prognostic belief, albeit at a 10% significance level. Patients were also significantly less likely to change to an inaccurate belief the longer they remained in the study (ie, the greater their response number [RRR, 0.74; 95% CI, 0.59–0.92]).
Association of Change in Prognostic Belief With Changes in Health Status
Results from the sensitivity analyses were consistent with the main results. An increase in symptom burden was associated with a change to accurate belief even after disaggregating the “no change” category (supplemental eTable 1). In the fixed effects models, symptom burden and physical well-being continued to be significantly associated with prognostic beliefs, although coefficients were expectedly less precise (eTable 2A, B). The complete case analysis yielded qualitatively similar results as well, other than functional well-being score, because the model failed to converge (eTable 3A, B).
Discussion
We used data from a prospective cohort of patients with stage IV solid metastatic cancer to examine changes in patients’ prognostic beliefs, up to 54 months. Given that many patients are seeing improvements in life expectancy despite the incurable nature of their illness while also receiving treatments to effectively palliate their symptoms, our study provides a fuller picture of the evolution of patients’ prognostic beliefs. We found that beliefs about disease curability changed often, both toward and away from an accurate belief, and were consistently associated with patients’ changing health status as captured by their symptom burden, physical well-being, and recent UHA. We found that 24% of patients who reported their prognostic beliefs at least twice changed from having an accurate belief to an inaccurate or uncertain belief at some point during the follow-up. This proportion of change from accurate to less accurate beliefs is similar to that reported in recent longitudinal examinations of prognostic beliefs.25,33,34 We found that on average, 7.1% of patients changed to inaccurate belief and 14.3% changed to uncertain belief at each time point. Previous studies have reported that between 8% and 20% of patients change from an accurate to an inaccurate belief between surveys.25,33,34 The slightly lower proportion of change to inaccurate beliefs in our sample may be explained by the longer time period under analysis, as the sample evolves to a majority of accurate beliefs over time. Alternately, it may be that these changes are captured as “change to uncertain” observations instead, allowing us to discern that even when patients change to less accurate beliefs, they appear to do so hesitantly. Such changes further demonstrate the limitation of understanding prognostic belief as a static or stable construct, or one that will only change toward accuracy.
We showed that patients’ prognostic belief evolves dynamically with their health status. We found that patients were consistently less likely to report an accurate belief when they experienced better health (lower symptom burden and greater physical well-being). Correspondingly, patients were more likely to report a change to accurate belief if they experienced an increase in symptom burden or decrease in physical well-being. This suggests the possibility of cognitive dissonance when the patient’s illness experience contradicted the prognostic information provided to them.35 To resolve the tension between their positive self-image or experience of health and poor prognosis, they may have avoided “cognitively integrating” their prognosis, and thereby reported inaccurate beliefs. Existing literature also suggests that emotional numbness36 and cognitive biases such as hope and denial may influence how patients metabolize prognostic information.37–39 Contrary to our hypothesis, we found that patients were less likely to report an accurate belief if they had experienced a recent UHA. This is somewhat surprising because patients may or may not feel physically better after an UHA, but the experience of having survived and overcome an unexpected deterioration in health possibly created a self-deception or misattribution bias that made them more optimistic about disease curability.40 In this case, the prognostic information provided to patients may be at odds with their improved health status, prompting them to resolve this dissonance by rationalizing that their treatment may cure them.41
Our results have clinical implications. Physicians’ reluctance to discuss prognosis is recognized as a significant barrier to effective patient–physician communication, but the flip side—patients’ reluctance to accept prognosis—arguably deserves closer attention.42,43 Although improvements in patient–physician communication must focus on meeting patients’ information needs, clear delivery of prognostic information, and supporting patients in internalizing this information, physicians must balance these considerations with the psychological benefits offered by coping strategies that may manifest as inaccurate beliefs.17 Given the cognitive dissonance patients appear to be experiencing, the insistence that patients report correct understanding of their prognosis should be revisited. Physicians must also consider that patients’ prognostic beliefs can change as their illness evolves and health status changes, even years after they have been living with an incurable illness.17,28,41 Keeping in mind that patients may experience “backsliding” to less accurate beliefs even after reporting an accurate belief, conversations about goals of care must occur regularly to address and account for these changes.
On the methodological front, our results underscore the need to understand prognostic belief as a dynamically evolving construct, and call into question the framing of accurate belief about disease curability as prognostic awareness.44 In addition to its lack of conceptually consistent definition,3,45 the term “prognostic awareness” implies that patients’ responses are a function of the information provided to them. Our results show otherwise.
This study, however, is not without limitations. First, the study design prevents us from drawing causal inferences. Because patients’ prognostic beliefs possibly influenced their experience and reporting of health status, further research into the determinants of prognostic belief is warranted. Second, we witnessed sizeable numbers of deaths along with some dropouts and nonresponses over the 54 months. This sample evolution is unsurprising given patients’ diagnosis of metastatic cancer. We were unable to control for mortality as we studied prognostic belief prospectively from baseline survey. Upon examining the sensitivity of results to missingness, we found that the main results remained broadly unchanged. Third, our findings were based on responses to one question that focused on patients’ belief about disease curability; the word “cure” has been previously noted as potentially limiting.44 However, previous studies with composite prognostic awareness measures report low correspondence between different questions, indicating that using multiple measures of a poorly understood construct might produce noisy results.18 Hence, we chose not to include subjective life expectancy or knowledge of cancer stage in our analysis and framed responses as “beliefs” instead of “awareness.” Fourth, our study would have benefitted from more data on prognostic information received by patients apart from the palliative care consults and visits they experienced. Last, we did not assess additional possible confounding factors, including use of advanced anticancer therapies, alternative/complementary treatments, or enrollment in clinical trials.
Conclusions
We studied long-term changes in patients’ prognostic beliefs and offered insights into previously unstudied dimensions of prognostic belief, namely the extent to which it evolves over a long period of time, its association with changing health status, and the prevalence of changes from accurate to less accurate beliefs. We found that prognostic beliefs are associated with changing health status: when patients feel better, they are less likely to hold an accurate prognostic belief, which suggests the presence of cognitive dissonance. Our findings have implications for treating physicians who must note that even after patients have reported an accurate understanding of prognosis, their beliefs may change as their health status evolves. Thus, conversations about goals of care must occur regularly to accommodate these changes.
References
- 1.↑
Applebaum AJ, Kolva EA, Kulikowski JR, et al. Conceptualizing prognostic awareness in advanced cancer: a systematic review. J Health Psychol 2014;19:1103–1119.
- 2.↑
Vlckova K, Tuckova A, Polakova K, et al. Factors associated with prognostic awareness in patients with cancer: a systematic review. Psychooncology 2020;29:990–1003.
- 3.↑
Kühne F, Hermann M, Preisler M, et al. Prognostic awareness in advanced disease: a review update and concept analysis. Front Psychol 2021;12:629050.
- 4.↑
Chen CH, Kuo SC, Tang ST. Current status of accurate prognostic awareness in advanced/terminally ill cancer patients: systematic review and meta-regression analysis. Palliat Med 2017;31:406–418.
- 5.↑
Ozdemir S, Ng S, Wong WHM, et al. Advanced cancer patients’ prognostic awareness and its association with anxiety, depression and spiritual well-being: a multi-country study in Asia. Clin Oncol (R Coll Radiol) 2022;34:368–375.
- 6.↑
Innes S, Payne S. Advanced cancer patients’ prognostic information preferences: a review. Palliat Med 2009;23:29–39.
- 7.↑
Hagerty RG, Butow PN, Ellis PM, et al. Communicating with realism and hope: incurable cancer patients’ views on the disclosure of prognosis. J Clin Oncol 2005;23:1278–1288.
- 8.↑
Chen CH, Tang ST. Prognostic disclosure and its influence on cancer patients. J Cancer Res Pract 2014;1:103–112.
- 9.↑
Saracino RM, Polacek LC, Applebaum AJ, et al. Health information preferences and curability beliefs among patients with advanced cancer. J Pain Symptom Manage 2021;61:121–127.
- 10.↑
Wen FH, Chen JS, Chang WC, et al. Accurate prognostic awareness and preference states influence the concordance between terminally ill cancer patients’ states of preferred and received life-sustaining treatments in the last 6 months of life. Palliat Med 2019;33:1069–1079.
- 11.↑
Tang ST, Wen FH, Hsieh CH, et al. Preferences for life-sustaining treatments and associations with accurate prognostic awareness and depressive symptoms in terminally ill cancer patients’ last year of life. J Pain Symptom Manage 2016;51:41–51.e1.
- 12.↑
Weeks JC, Cook EF, O’Day SJ, et al. Relationship between cancer patients’ predictions of prognosis and their treatment preferences. JAMA 1998;279:1709–1714.
- 13.↑
De Haes H, Koedoot N. Patient centered decision making in palliative cancer treatment: a world of paradoxes. Patient Educ Couns 2003;50:43–49.
- 14.↑
Balasubramanian I, Finkelstein E, Malhotra R, et al. Healthcare cost trajectories in the last 2 years of life among patients with a solid metastatic cancer: a prospective cohort study. J Natl Compr Canc Netw 2022;20:997–1004.e3.
- 15.↑
Prigerson HG, Bao Y, Shah MA, et al. Chemotherapy use, performance status, and quality of life at the end of life. JAMA Oncol 2015;1:778–784.
- 16.↑
Butow PN, Clayton JM, Epstein RM. Prognostic awareness in adult oncology and palliative care. J Clin Oncol 2020;38:877–884.
- 17.↑
Greer JA, Applebaum AJ, Jacobsen JC, et al. Understanding and addressing the role of coping in palliative care for patients with advanced cancer. J Clin Oncol 2020;38:915–925.
- 18.↑
Kurita K, Siegler EL, Reid MC, et al. It is not what you think: associations between perceived cognitive and physical status and prognostic understanding in patients with advanced cancer. J Pain Symptom Manage 2018;56:259–263.
- 19.↑
Chen CH, Wen FH, Chou WC, et al. Associations of prognostic-awareness-transition patterns with end-of-life care in cancer patients’ last month. Support Care Cancer 2022;30:5975–5989.
- 20.↑
Chen JS, Wen FH, Chou WC, et al. Terminally ill cancer patients’ distinct symptom-functional patterns/states are differentially associated with their accurate prognostic awareness in the last six months of life. J Palliat Med 2019;22:782–789.
- 21.↑
Chen CH, Wen FH, Hou MM, et al. Transitions in prognostic awareness among terminally ill cancer patients in their last 6 months of life examined by multi-state Markov modeling. Oncologist 2017;22:1135–1142.
- 22.↑
Ozdemir S, Chaudhry I, Ng S, et al. Prognostic awareness and its association with health outcomes in the last year of life. Cancer Med 2023;12:4801–4808.
- 23.↑
Temel JS, Greer JA, Admane S, et al. Longitudinal perceptions of prognosis and goals of therapy in patients with metastatic non-small-cell lung cancer: results of a randomized study of early palliative care. J Clin Oncol 2011;29:2319–2326.
- 24.↑
Loucka M, Vlckova K, Tuckova A, et al. Prognostic awareness in advanced cancer patients and their caregivers: a longitudinal cohort study. Psychooncology 2021;30:1449–1456.
- 25.↑
Hasegawa T, Okuyama T, Uemura T, et al. Prognostic awareness and discussions of incurability in patients with pretreated non-small cell lung cancer and caregivers: a prospective cohort study. Oncologist 2022;27:982–990.
- 26.↑
Teo I, Singh R, Malhotra C, et al. Cost of medical care of patients with advanced serious illness in Singapore (COMPASS): prospective cohort study protocol. BMC Cancer 2018;18:459.
- 27.↑
Zhuang Q, Chong PH, Ong WS, et al. Longitudinal patterns and predictors of healthcare utilization among cancer patients on home-based palliative care in Singapore: a group-based multi-trajectory analysis. BMC Med 2022;20:313.
- 28.↑
Weeks JC, Catalano PJ, Cronin A, et al. Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med 2012;367:1616–1625.
- 29.↑
Lyons KD, Bakitas M, Hegel MT, et al. Reliability and validity of the Functional Assessment of Chronic Illness Therapy-Palliative care (FACIT-Pal) scale. J Pain Symptom Manage 2009;37:23–32.
- 30.↑
Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. J Clin Oncol 1993;11:570–579.
- 31.↑
Emiloju OE, Djibo DAM, Ford JG. Association between the timing of goals-of-care discussion and hospitalization outcomes in patients with metastatic cancer. Am J Hosp Palliat Care 2020;37:433–438.
- 32.↑
Liu PH, Landrum MB, Weeks JC, et al. Physicians’ propensity to discuss prognosis is associated with patients’ awareness of prognosis for metastatic cancers. J Palliat Med 2014;17:673–682.
- 33.↑
Gray TF, Plotke R, Heuer L, et al. Perceptions of prognosis and end-of-life care outcomes in patients with advanced lung and gastrointestinal cancer. Palliat Med 2023;37:740–748.
- 34.↑
Agarwal R, Shin P, Knezevic A, et al. Accuracy of curability expectations in patients with gastrointestinal cancers. Cancer Med 2023;12:20–29.
- 36.↑
Maciejewski PK, Prigerson HG. Emotional numbness modifies the effect of end-of-life discussions on end-of-life care. J Pain Symptom Manage 2013;45:841–847.
- 37.↑
Chochinov HM, Tataryn DJ, Wilson KG, et al. Prognostic awareness and the terminally ill. Psychosomatics 2000;41:500–504.
- 38.↑
Loh KP, Mohile SG, Epstein RM, et al. Willingness to bear adversity and beliefs about the curability of advanced cancer in older adults. Cancer 2019;125:2506–2513.
- 39.↑
Finkelstein EA, Baid D, Cheung YB, et al. Hope, bias and survival expectations of advanced cancer patients: a cross-sectional study. Psychooncology 2021;30:780–788.
- 40.↑
Echarte LE, Bernacer J, Larrivee D, et al. Self-deception in terminal patients: belief system at stake. Front Psychol 2016;7:117.
- 41.↑
Jackson VA, Jacobsen J, Greer JA, et al. The cultivation of prognostic awareness through the provision of early palliative care in the ambulatory setting: a communication guide. J Palliat Med 2013;16:894–900.
- 42.↑
Mone S, Kerr H. Prognostic awareness in advanced cancer: an integrative literature review. BMJ Support Palliat Care 2021;11:53–58.
- 43.↑
Gattellari M, Butow PN, Tattersall MHN, et al. Misunderstanding in cancer patients: why shoot the messenger? Ann Oncol 1999;10:39–46.
- 45.↑
Brenner K, Greer JA, Jackson V, et al. Development of a prognostic awareness impact scale for patients with advanced cancer. J Palliat Med 2022;25:445–454.
