- © 2004 by American Society of Clinical Oncology
The Attitudes of Cancer Patients and Their Families Toward the Disclosure of Terminal Illness
- Young Ho Yun,
- Chang Geol Lee,
- Si-young Kim,
- Sang-wook Lee,
- Dae Seog Heo,
- Jun Suk Kim,
- Keun Seok Lee,
- Young Seon Hong,
- Jung Suk Lee and
- Chang Hoon You
- From the Research Institute and Hospital, National Cancer Center, Goyang; Yonsei University College of Medicine, Kyunghee University Hospital; Asan Medical Center, Seoul National University Hospital; Korea University Medical Center; Kangdong Sacred Heart Hospital, St Mary's Hospital, Seoul, Korea
- Address reprint requests to Young Ho Yun, MD, PhD, Quality of Cancer Care Branch, Research Institute and Hospital, National Cancer Center 809, Madu-dong, Ilsan-gu, Goyang-si, Gyeonggi-do 411-769, Korea; e-mail: lawyun{at}ncc.re.kr.
Abstract
Purpose To ascertain the attitude of cancer patients and their families toward disclosure of terminal illness to the patient.
Patients and Methods We constructed a questionnaire that included demographic and clinical information and delivered it to 758 consecutive individuals (433 cancer patients and 325 families that have a relative with cancer) at seven university hospitals and one national cancer center in Korea.
Results 380 cancer patients and one member from each of 281 families that have a relative with cancer completed the questionnaire. Cancer patients were more likely than family members to believe that patients should be informed of the terminal illness (96.1% v 76.9%; P < .001). Fifty percent of the family members and 78.3% of the patients thought that the doctor in charge should be the one who informs the patient. Additionally, 71.7% of the patients and 43.6% of the family members thought that patients should be informed immediately after the diagnosis. Stepwise multiple logistic regression indicated that the patient group was more likely than the family group to want the patient to be informed of the terminal illness (odds ratio [OR], 9.76; 95% CI, 4.31 to 22.14), by the doctor (OR, 4.00; 95% CI, 2.61 to 6.11), and immediately after the diagnosis (OR, 3.64; 95% CI, 2.45 to 5.41).
Conclusion Our findings indicated that most cancer patients want to be informed if their illness is terminal, and physicians should realize that the patient and the family unit may differ in their attitude toward such a disclosure. Our results also reflect the importance of how information is given to the patient.
INTRODUCTION
Most physicians [1-4] and patients [5-11] now accept as ethical the proposition that patients are entitled to know their diagnosis. In Korea, only 18% of physicians were likely to disclose the bad news to patients in 1982, but 81.8% agreed to do so in 1990 [12,13]. This reflects a change in attitude brought about by advances in therapy that reduced the cancer death rate, a decrease in pessimism about cancer, and an increase in concerns about a patient's right to participate in care decisions [2,7]. Most patients with recurrent cancers, however, die as a result of their malignancy, and there remains as a significant problem in some cultures that patients are not being informed when their disease progresses to a terminal phase and when treatment changes from curative to palliative [14,15]. But if patients are to take charge of their own care, challenge the disease, and make life-support decisions, they must be made aware of their condition [16-18]. Because patients who overestimate their survival time prefer more aggressive treatment [19], telling patients the truth about their terminal cancer may lead to more appropriate care. It is difficult, however, to decide what to tell patients, and how and when to tell them. Although recent studies have focused on the cancer patient's attitude toward disclosure of the diagnosis of cancer [8-11] and on the attitude toward disclosure of terminal cancer [12,20-30], few have focused specifically on the cancer patient's attitude toward disclosure of terminal illness [27,28], and no published studies have focused on the attitudes of the cancer patient's family.
Our study was designed to investigate attitudes of cancer patients and their families' attitudes toward the diagnosis of a terminal cancer.
PATIENTS AND METHODS
Participants and Procedures
Seven university hospitals and one national cancer center were involved in the study. None of the institutions had a stated policy about the disclosure of terminal illness. A member of the research team approached 758 consecutive individuals (433 cancer outpatients and 325 members of families who have a relative with cancer) who were sitting in the waiting room of surgical, medical, and radio-oncology clinics for a medical consultation or a routine follow-up visit. For enrollment in the study, subjects needed to be age 18 years or older, well enough to fill out a questionnaire or communicate with the interviewer, and well enough to provide informed consent. They were given information explaining the study and asked to participate. As patients and families were filling out the questionnaires, researchers were present to ensure that information was not shared. In cases where the patient was accompanied by family, the patients were interviewed in a separate room. The study was approved by the institutional review board of the National Cancer Center, Goyang, Korea.
Materials
We constructed a questionnaire examining attitudes toward disclosure of the diagnosis of a terminal illness. We based the questionnaire on previous studies of the disclosure of “bad news” and pilot-tested it on healthy volunteers and cancer patients. We aimed to learn whether patients and families differed in their attitudes. The questionnaire gathered the following: demographic information (age, sex, relationship to patient, level of education, income, and religiousness); clinical information (time since initial diagnosis, awareness of the cancer diagnosis, type of primary cancer, disease stage, and Eastern Cooperative Oncology Group performance status); preference for end-of-life care issues (place of care and death); and information on attitude toward the disclosure of terminal illness to the patient. A series of questions covered the following: pros and cons of informing patients of the terminal illness, and the reasons; the appropriate person to inform the patient, and the reasons; and the appropriate time to inform the patient (Table 1). In the questionnaire, terminal illness was defined as cancer that was progressing and no longer treatable by conventional therapy (surgery or radio-, chemo-, or hormone therapy).
Measures of Attitudes Toward the Disclosure of Terminal Illness
Statistical Analysis
We used a t test or χ2 test to determine significant differences in dependent and independent variables between the patient and family groups. We used univariate logistic regression analysis to estimate the odds ratio (OR) for each independent variable (the OR is the extent to which being a member of a specific group increased or decreased the probability of agreeing with the model of attitudes toward disclosing bad news). We used indicator variables for independent variables that were categoric. Odds ratios greater than 1 represent how much more likely it was for a subject in a category to believe that patients should be informed of their terminal illness, that the doctor is the appropriate person to inform the patient, and that the appropriate time to inform the patient is immediately after diagnosis. In addition, for factors significantly associated in univariate analysis, we performed stepwise multiple regression analyses for each dependent variable to assess which of the independent variables best predicted attitudes toward the disclosure of terminal cancer. Finally, we performed within-group stepwise multiple regression analyses to identify factors significantly related to attitudes toward the disclosure of bad news after controlling for the group. For these analyses, we set the significance level at P < .05. We used the SAS statistical package, version 8.1 (SAS Institute, Cary, NC, 1990).
RESULTS
Sample Characteristics
Six hundred sixty-one (87.2%) of the 758 consecutive individuals—380 (87.8%) of the patients and 281 (87%) of the family members—completed the questionnaire with informed consent while waiting for their appointment. The most common reasons given for nonparticipation were a lack of time and poor eyesight. The questionnaire took approximately 10 minutes to complete. Table 2 gives the subject characteristics. The patient group consisted of more men (P < .01), was older (P < .001), and had a lower education level (P < .001) than the family group. The patients to whom the family group subjects were related were more likely than the patient subjects to have terminal stage disease (P < .001) and a worse performance status (P < .001).
Characteristics of Study Subjects*
Difference in Attitude Toward Disclosure of Diagnosis of a Terminal Illness
The patient group was more likely than the family group to want the patient to be informed of the terminal illness (96.1% v 76.9%; P < .001; Fig 1). The reasons that subjects gave for answering “yes” or “no” to the question of disclosure are given in Table 3.
The percentage of subjects that answered: (A) Do you want the patient to be informed of the truth? (B) Who is the appropriate person to inform the patient? (C) When is the appropriate time to inform the patient?
Why Did You Answer “Yes” or “No” Regarding Disclosure of a Terminal Illness to the Patient?
The patient group was also more likely than the family group to think that the doctor in charge is the appropriate person to provide the disclosure (P < .001), and was more likely to think that the patient should be informed of a terminal illness immediately after diagnosis (P < .001; Fig 1). The reasons that subjects gave for answering “doctor in charge” or “family member” are given in Table 4.
Why Do You Think the Person in Charge Is the Appropriate One to Inform the Patient of a Terminal Illness?
Univariate Logistic Regression Analyses of Factors Related to Wanting Disclosure
Age, religiousness, relationship to patient, time since diagnosis, type of primary cancer, disease stage, and performance status were likely to influence attitudes, while sex, level of education, income, awareness of the cancer, preference for place of care and death, and diagnosis were not (Table 5).
Odds Ratios and 95% CIs From Logistic Regression Analysis Predicting Attitude Toward the Disclosure of a Terminal Illness
Stepwise Multiple Logistic Regression Analyses
In stepwise multiple logistic regression analyses, we excluded factors that were not associated in univariate analysis with attitude toward the disclosure of terminal illness. For all three of the measures of attitude toward disclosure, the first factor related to attitude was the difference between patient and family members. The patient group was more likely than the family group to want the patient to be informed of the diagnosis of terminal illness (OR, 9.76; 95% CI, 4.31 to 22.14), by the doctor (OR, 4.00; 95% CI, 2.61 to 6.11), and immediately after the diagnosis is made (OR, 3.64; 95% CI, 2.45 to 5.41). After controlling for the group difference, only disease stage was associated with attitude toward the appropriate time to inform patients of the truth. Subjects with early stage disease were more likely than subjects with terminal stage disease to want the patient to be informed immediately of the diagnosis of terminal illness (OR, 2.32; 95% CI, 1.32 to 4.09). Age, religiousness, time since initial diagnosis, and performance status showed no association in the multivariate analysis (Table 6).
Stepwise Multiple Logistic Regression Analysis of Factors Predictive of Attitudes Toward Disclosure of a Terminal Illness
Within-Group Analyses
For factors significantly associated in univariate analysis within each group, we performed stepwise multiple regression analyses for all three of the measures of attitude toward disclosure. In the patient group analysis, patients with a higher level of education were more likely than patients with a lower level of education (OR, 1.88; 95% CI, 1.05 to 3.36) to think that the doctor is the appropriate person to inform the patient of the terminal illness. In the within-family group analysis, spouses were more likely than other relatives (OR, 2.03; 95% CI, 1.17 to 3.54) to regard the doctor as the appropriate informer.
DISCUSSION
A most important finding was that 96.1% of the patient group wanted to be told if they had a terminal illness. This differed significantly from the corresponding proportion in the family group (76.9%). The two groups also differed in attitude about who should inform the patient about the terminal illness, and when. More patients than family members wanted the doctor to be the one who informed the patient and to do so immediately after the illness was diagnosed. These observations should provide persuasive data for health professionals who might withhold information about a terminal illness on the grounds that patients would prefer not to know about it.
From the patient's point of view, receiving a diagnosis of terminal illness is traumatic and may cause otherwise rational people to make irrational choices [7]. Family and physicians frequently protect patients from bad news to give them hope for the future [31]. In Japan, China, Greece, and Ethiopia, for example, physicians believe that causing patients to lose hope by telling them about the illness will only hasten their death, [20,21,28] and physicians in the United Kingdom and Italy are likely to withhold the information from the patient at the family's request [9,15]. Most physicians endorse the involvement of family members when disclosing terminal illness and may allow the family to make decisions on behalf of an incapacitated patient [32,33]. Sometimes family members may consent to far more aggressive treatment than patients would want for themselves [7]. Paternalistic decisions by physician or families may lead to dissatisfaction with the medical system, causing increased stress, financial strain, and prolonged and painful deaths as a result of unwanted, invasive care [7]. Even in the past, when most physicians believed that cancer connoted certain death, approximately 90% of patients indicated that they would want disclosure [5,6,9,34]. In practice, cancer patients want to know their prognosis so that they can resolve unfinished business, take appropriate care of themselves in cooperation with health care professionals, and relieve themselves and their families of the burden of useless treatments. Also, they believe they have a right to know the truth about their own condition. However, unless they are invited directly during a consultation, patients rarely raise the issue [35]. Thus, the physician should assess a patient's need for information about the prognosis.
Many studies have suggested that because cultures vary, Western values that promote the principle of patient autonomy may not be universally applicable [20,21,26,36-39]. Baile at al [23] provided preliminary data regarding cultural differences in disclosing cancer diagnosis and prognosis. Blackhall et al [26,40] found that ethnicity was the primary factor that influenced attitudes toward truth-telling and patient decision-making. Those results, however, reflected the attitudes of doctors, family members, and healthy people. In the present study, which involved Korean cancer patients and their families, we found a strong wish for disclosure of terminal illness among the patients themselves, and significant difference in attitude between patients and family members. Therefore, caution should be applied in holding to the family-centered model rather than exploring the patient's wishes. Our results suggest that differences in attitude toward disclosure of terminal illness existing within a single culture may be because of different roles and points of view between patients and their family members. International ethical codes and human rights law accept that the right to autonomy or self-determination is broadly perceived as necessary to human dignity [39]. However, without knowledge of the attitudes toward disclosure of a terminal illness among cancer patients, doctors, family, and healthy people in both Western and Asian populations, it is not clear whether differences in attitude are due to differences in culture, role, or point of view. Families who objected to telling the patient the truth were more likely than patients to believe that the disclosure would cause the patients to lose hope and discourage them from fighting the disease. In addition, patients who objected to being told the truth were more likely than the families to see the disclosure as meaningless and conveying a feeling of helplessness. Those were interesting findings that may underscore the importance of how physicians and families disclose the truth in clinical practice. In fact, the issue of how to disclose the truth is more important than whether or not to do so [41]. There are many recommendations on how a cancer diagnosis should be given to patients. All stress the importance of honesty, compassion, sensitivity, clarity, and allowing some measure of hope [6,42,43].
Who is the appropriate person to inform the patient of terminal illness? Our study showed that most patients and half the families would prefer disclosure by the doctor because the doctor understands the disease, could explain the situation fully, and is trusted. In contrast, approximately 20% of patients and half of families would prefer disclosure by family members because the family members are aware of the patient's emotional, psychological, and social conditions as well as the physical one, and they could cushion the impact and provide comfort. In the family group, spouses were more than twice as likely as other relatives to regard the doctor as the appropriate informer; they believed that the doctor understood the disease and could explain the situation fully. We need to further investigate the basis for the difference in such preferences according to the relationship to patients. But too frequently, physicians are ill equipped to handle that disclosure [44]. When discussions about prognosis are documented in medical charts along with a plan for end-of-life care, they can pave the way for further discussions of that care, and they can enhance the physician's relationship with the patient and the patient's family [45,46]. In addition, laws such as the US Patient Self-Determination Act, which requires that patients be informed of their right to make their own medical care decisions and supports advance directives, may improve communication between doctors and patients while providing greater assurance that patients will be treated according to their own values and preferences [47,48].
Our study had several limitations. First, the family members sitting in the waiting rooms might not have been representative of all family members. Second, our study did not allow the participants to complete the questionnaire and deliver it in a sealed envelope, so it may have been biased toward positive outcomes. Researchers, however, did not allow the patients and family members to share information, which minimized the bias caused by patients answering according to family expectations. Third, if the participating institutions were more progressive than nonparticipating institutions regarding disclosure, there could be a bias toward positive outcomes. However, most Korean hospitals, like those in this study, do not have an institutional policy regarding the disclosure of diagnoses. Fourth, clinical information depended on patient and family knowledge about the illness, such as disease stage and performance status, and those might not have been accurate. Finally, we did not compare attitudes between the patients and family members by matching. Such matching might be useful for patients and family education.
In summary, our findings suggest that regardless of their role, respondents agreed on the importance of disclosure of a terminal illness. This study underscores the fact that most patients wanted to be informed of the diagnosis of a terminal illness, but attitudes toward such disclosure may differ between patient and family. Our results also reflect the importance of how information should be given to patients. We suggest further, comprehensive investigations to ascertain the contributions of cultural and role differences to attitudes toward the disclosure of terminal illness among cancer patients, doctors, family members, and healthy people in both Western and Asian populations. This could be done through a cooperative international study.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgments
This work was supported by National Cancer Center Grant N14020.
Footnotes
-
Presented at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
- Received July 8, 2003.
- Accepted November 7, 2003.
