Prevalence of Ethical Issues in Patients with Advanced Cancer: Secondary Analysis of Pooled Data from the Development and Validation Cohorts of the PALCOM Scale for Complexity of Palliative Care Needs

2. Materials and Methods

We conducted a secondary analysis of the pooled data from the development and validation cohorts of the PALCOM scale, which measures the complexity of palliative needs in patients with advanced cancer. Both cohorts were prospective and multicentre, with a maximum follow-up of 6 months. They applied the same inclusion criteria and used the same methodology for both the assessment and the recording of variables.

2.1. Setting and Study Periods

The development cohort, aimed at constructing a model for complexity of palliative care needs (PALCOM scale), included patients from November 2012 to January 2013. The validation cohort, aimed at confirming the consistency and predictive value of the PALCOM scale, included patients from December 2020 to April 2021. Both studies involved multiple public healthcare centres across all levels of care in the Autonomous Community of Catalonia (Spain): primary care, hospitals, hospital and home palliative care (PC), and medium-long stay units. The development cohort included 24 healthcare centres (16 primary care, 3 hospitals, 3 home palliative care teams, and 2 medium-long stay centres). The validation cohort included 7 healthcare centres (1 primary care, 2 hospitals, 3 home PC teams, and 1 medium-long stay centre).

The results from the development and validation cohorts, along with a secondary analysis of the prevalence of ethical issues based solely on the development cohort, were published in 2018, 2023, and 2021, respectively [9,10,11,13]. Current study presents a secondary analysis of the prevalence of ethical issues using pooled data from both cohorts. The objectives are to increase sample size and consistency, compare results between the two cohorts, and analyze their evolution during follow-up (Figure 1).

2.2. PALCOM Scale of Palliative Complexity

The PALCOM scale is a predictive tool for assessing complexity of palliative care needs, consisting of five multidimensional domains: (a) symptomatic burden; (b) refractory pain; (c) functional decline; (d) socio-family risk factors; and (e) ethical issues (Table 1). Each domain is scored dichotomously, with 0 indicating absence and 1 indicating the presence of the corresponding variables [9,10,11]. The final score is the sum of the values across all domains. Patients scoring 4-5 are classified as high complexity and require systematic, intensive intervention by multidisciplinary palliative care (PC) teams, whether shared with referring teams or not. Those scoring 2-3 are classified as medium complexity and need systematic intervention by multidisciplinary PC teams, at an intensity adjusted to their needs, whether shared with referring teams or not. Patients scoring 0-1 are classified as low complexity and generally do not require systematic intervention by multidisciplinary PC teams, except for occasional consultations.

2.3. Definition of Ethical Issues

The identification of ethical issues can depend on the individual interpretation of observers. Therefore, the research team agreed on the following operational definition for both cohorts: a discrepancy in shared decision-making in a specific situation where it is necessary to choose between opposing options, morally or existentially acceptable from the individual perspective of the different participants in the deliberation. The observed ethical issues were classified according to the sub-categories identified in the analysis of the fifth domain of the PALCOM scale in the development cohort [9]: a) Related to information; b) Related to the proportionality of therapeutic and supportive measures (discrepancies in proportionality among family members, between the patient/family and the care team, and among care team members); c) Related to research and clinical trials; d) Related to palliative sedation; e) Related to the desire to hasten death; f) Other unspecified issues.

2.4. Objectives

The main objective of this study was to determine the overall prevalence of ethical issues in patients with advanced cancer. The secondary objectives were: a) to identify the prevalence of different sub-categories of ethical issues; b) to compare the prevalence between the two cohorts and assess their consistency; c) to describe the evolution of these issues over the 6-month follow-up period of the study.

2.5. Inclusion Criteria

We proposed to consecutively include in the study all patients who met the following inclusion criteria: age ≥ 18 years, diagnosis of advanced cancer, clinically estimated life expectancy ≤ 6 months, and signed informed consent.

2.6. Main Outcome Variable

The main variable in this study was the prevalence of ethical issues in patients with end-of-life cancer at the baseline visit. The prevalence of ethical issues was evaluated both globally and broken down into the aforementioned sub-categories (information, proportionality, research, palliative sedation, miscellaneous) at the baseline visit and monthly during follow-up.

2.7. Descriptive Variables

The following descriptive and clinical variables were recorded and analyzed: age, gender, primary origin of the cancer, survival, place of death, level of palliative care complexity, and domains of the PALCOM scale: a) high symptom burden (≥5 symptoms of at least moderate intensity, in a systematic record of 10 symptoms based on the Edmonton Symptom Assessment Scale model [24]); b) potentially refractory pain, according to the Edmonton Classification System for Cancer Pain [25,26,27]; c) Karnofsky Performance Status; d) socio-familial risk factors.

2.8. Data Collection, Sample Size and Statistical Analysis

All study variables were recorded directly by the treating professional in an electronic data collection notebook, coded to ensure the confidentiality of participants’ personal data. No individuals outside the research project had access to the coded database.

In both cohorts, the sample size estimation was based on previous studies and the maximum recruitment capacity of the participating centres, considering that the variables had to be recorded under the conditions of their daily clinical practice. The prevalence of patients presenting ethical issues and the frequencies of sub-categories were evaluated both generally and stratified by the level of complexity of the PALCOM scale and by month of follow-up. The prevalence observed in the development and validation cohorts were compared, considering that they were conducted at different times and by different care teams. Categorical or dichotomous variables were analyzed using absolute and relative frequencies. Continuous variables were described by calculating the mean value and standard deviation (SD) with a 95% confidence interval. For the comparison of variables according to complexity levels, Fisher’s exact test and the non-parametric Mann-Whitney-Wilcoxon test were used.

3. Results

A total of 607 patients were included in the pooled analysis, with 324 (53.4%) from the development cohort and 283 (46.6%) from the validation cohort. Monthly follow-up data were available only for the validation cohort, as this variable was not recorded in the development cohort.

In the pooled data, the mean age of the patients was 70 years (SD ±11), and 350 were men (57.7%). Three hundred fifty-five patients were included in the hospital (58.5%), and 255 (41.5%) in community health centres. The most common primary cancer origins were lung (23.1%), followed by colon (15.5%), prostate (7.6%), and breast (6.8%). Five hundred patients had metastatic disease (82.4%), and 107 had advanced loco-regional disease (17.6%). At the time of inclusion, 462 (76.1%) patients were undergoing cancer treatment. The most frequent symptoms were asthenia (93.6%), pain (80.7%), anorexia (78.9%), anxiety (69.5%), sadness (69.4%), and insomnia (60.6%). The frequencies of the five domains of the PALCOM scale were: a) high symptom burden 44.5%; b) potentially refractory pain 56.2%; c) Karnofsky Performance Status ≤60% 43.5%; d) socio-familial risk factors 66.7%; e) ethical issues 23.4%. Three hundred seventy-nine patients (62.4%) died before the end of the maximum follow-up period of 6 months. No significant differences were observed between the two cohorts in terms of symptom frequencies or the domains of the PALCOM scale (Table 2 and Table 3).

3.1. Prevalence of Ethical Issues

In the pooled data, 126 patients (20.7%) presented at least one ethical issue (59 from the development cohort and 67 from the validation cohort) (Table 4). A total of 204 ethical issues were recorded (117 in the development cohort and 87 in the validation cohort). Therefore, the average number of ethical issues per patient was 1.6 (1.9 in the development cohort and 1.3 in the validation cohort). No significant differences were observed between the two cohorts.

In the pooled data, 79 patients (13.0%) presented issues related to information. This was slightly more frequent in the development cohort (15.7%) than in the validation cohort (10.0%), but the difference was not statistically significant. In the pooled data, 95 patients (15.6%) presented issues related to the proportionality of therapeutic or supportive measures, with no significant differences between the development cohort (16.7%) and the validation cohort (14.5%). The discrepancies in proportionality were: a) among different family members in 47 patients (7.7%); b) between the patient or their family and the healthcare team in 27 patients (4.4%); c) among different members of the healthcare team in 25 patients (4.1%). No significant differences were observed between the frequencies in the development and validation cohorts. In the pooled data, 18 patients (2.9%) presented issues related to research, with no significant differences between the development cohort (2.5%) and the validation cohort (3.5%). In the pooled data, 11 patients (1.8%) explicitly, consistently, and repeatedly expressed the desire to hasten death, with no significant differences between the development cohort (1.2%) and the validation cohort (2.4%).

3.2. Evolution During Follow-Up

The monthly probability of presenting ethical issues during the 6-month follow-up was evaluable only in the validation cohort. A total of 67 patients (23.7%) presented ethical issues at the baseline visit. Among the patients who were alive in the first month of follow-up, 44 (17.1%) presented ethical issues, 32 (15.9%) in the second month, 25 (15.9%) in the third month, 18 (14.1%) in the fourth month, 18 (15.8%) in the fifth month, and 14 (14.4%) in the sixth month (Figure 2). The probability of presenting ethical issues was significantly higher at the baseline visit (p<0.001), with no significant differences in this monthly probability from the first month of follow-up onwards.

3.3. Prevalence of Ethical Issues According to Palliative Complexity Level on the PALCOM Scale

The level of palliative complexity according to the PALCOM scale depends on the interaction of five assessment domains: symptom burden, potentially refractory pain, functional impairment, socio-familial risk, and ethical issues. The presence of 4-5 affected domains corresponds to high complexity cases, 2-3 to medium complexity, and 0-1 to low complexity.

In the aggregated data, 118 patients (19.5%) were classified as having low palliative complexity, 306 (50.5%) as medium, and 182 (30.0%) as high. Higher complexity levels are significantly associated with increased six-month mortality (low 43.2%, medium 62.7%, high 75.8%) (p<0.001) and a higher probability of in-hospital death (low 7.6%, medium 16.0%, high 28.6%) (p<0.001). The likelihood of presenting ethical issues, both in the aggregated data and when broken down between the two cohorts, is significantly higher at greater levels of palliative complexity according to the PALCOM scale (p<0.001): low 4.2%; medium 19.5%; and high 30.8% (Table 3).

4.1. Overall Prevalence of Ethical Issues

In the polled data, 21% of patients presented at least one ethical issue according to the treating professionals. It is noteworthy that many of these patients simultaneously presented multiple ethical issues (an average of 1.6 ethical issues per patient). It is clear that ethical issues can cause significant discomfort for all parties involved in their discussion. The first step in addressing such a problem is its identification, and the second is to manage the conflict explicitly and respectfully. Conflict management skills can be essential for health professionals in resolving these issues. In fact, it is noteworthy that the prevalence of ethical issues in this study was significantly higher at the baseline visit compared to the prevalence observed during the subsequent six months of follow-up. These data suggest that early identification and intervention can reduce the frequency of ethical issues and, therefore, improve the outcomes of patient support measures. Overall, we consider these findings to be novel and relevant due to the high observed prevalence and the lack of previously published data in this area.

4.1.1. Ethical Issues Related to Information

Communication between doctor and patient is a process of information exchange aimed at providing the patient with the necessary knowledge to make reflective and autonomous decisions in a specific clinical context. It is an asymmetric and bidirectional process. The healthcare professional, who has knowledge of the clinical data, must explain and advise the patient clearly, truthfully, comprehensibly, and personally about everything concerning their health, and must recognize the value and meaning the patient attributes to the received information [28,29,30]. Available literature confirms that adequate communication can improve adherence to proposed treatments, quality of life, and patient satisfaction, as well as reduce uncertainty and emotional impact [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45].

In the complex process of informing about the diagnosis, extent, treatment options, and prognosis of cancer, ethical issues may arise. These issues are related to both the quality and quantity of the information provided, which is often modulated according to the professional’s criteria in response to the patient’s or family’s reaction, as well as the timing and context in which the information is offered. The healthcare professional must manage multiple and unpredictable coping styles to the life-threatening nature of the disease, where the patient’s expectations and hopes are not always realistic, and where there may be tendencies towards compassionate limitation of information by the family.

In this study, 13% of patients presented one of the following decision-making problems related to information: a) information exceeding the patient’s desired limits; b) adaptive denial; c) conspiracy of silence. Information exceeding the patient’s desired limits was described as that which caused a high emotional impact and hindered their participation in decision-making because it was provided rapidly in all its aspects, without respecting the patient’s opinion on the amount, timing, or context in which it was given. Denial was described as an adaptive process in which the patient, implicitly or explicitly, preferred not to know all aspects of their disease, especially the prognosis, or adopted alternative arguments to avoid recognizing a reality they felt was unmanageable. The conspiracy of silence was defined as the explicit agreement of the family to avoid providing information, especially the prognosis, with a compassionate intention based on the belief that the patient does not have sufficient emotional resources to face the life-threatening situation.

Various studies indicate that between 20% and 69% of patients with advanced and incurable cancer are unaware of both the prognosis and the palliative intent of chemotherapy or radiotherapy treatment [40,46,47,48,49,50]. Communication with the patient is an ethical and clinical imperative for all healthcare professionals. If a patient has insufficient information, it is the healthcare professional’s duty to expand it truthfully, comprehensibly, and personally to ensure autonomy in decision-making. In this study, the prevalence of ethical issues related to information was limited to situations where the previously provided information or the adaptive reaction to it generated an ethical issue that hindered shared decision-making with the patient. This is likely the reason for the observed differences between the general record on the lack of knowledge about the disease prognosis mentioned earlier (20-69%) and the data provided by this study (13%).

In 2018, the healthcare ethics consultation service at Memorial Sloan Kettering Cancer Center in New York published a retrospective study on the prevalence of ethical issues consulted. According to this study, 8% of these consultations were related to communication problems and 13% to cases of denial by the patient or their family [51].

4.2.2. Ethical Issues Related to Proportionality

Deliberation on the proportionality of healthcare interventions caused an ethical issue in 16% of the cases included in this study. The observed issues focused on two distinct situations: futility and treatment refusal. Futility is defined as a medical act considered useless or disproportionate due to its unlikely efficacy and lack of reasonable benefit in an individual clinical condition, according to available scientific evidence [52,53]. Refusal of treatment that healthcare professionals consider appropriate requires careful deliberation but is not ethically disputable if it is based on the patient’s reflective reasoning and the conditions of their autonomy of will are met (cognitive competence, complete information, and absence of external interferences) [54,55,56].

In this study, most discrepancies regarding proportionality were observed among the patient’s family members (8%). Less frequently, proportionality conflicts were recorded between the patient/family and the healthcare team (4%) or among the different healthcare professionals involved in the patient’s care (4%). Discrepancies regarding the withdrawal of life support were observed in very few cases (0.5%).

In the previously mentioned study by the Memorial Sloan Kettering Cancer Center in New York, 22% of the consultations to the healthcare ethics service were related to the futility of healthcare interventions, 5% to treatment refusal, and 2% to the withdrawal of life support. Additionally, they observed that in 11% of these consultations, there were discrepancies among the patient’s family members, in 3% between the patient/family and the healthcare team, and in 5% among the members of the treating team [51].

4.2.3. Ethical Issues Related to Research

Research in patients with advanced cancer who have not responded to conventional treatments undeniably represents an ethical challenge. The vulnerability of these patients and the desperate need to find therapies that delay the progression of their disease can lead to misinterpretation of information, affect the informed consent process, and encourage acceptance of inclusion in an early-phase clinical trial [57,58,59,60,61].

In this study, ethical issues related to research were detected in 3% of the included patients. As expected, all patients participating in clinical trials received verbal and written information and signed an informed consent form. However, researchers identified that some of these patients’ expectations were not fully aligned with the actual expected outcomes, or they accepted inclusion in the trial out of fear of being abandoned by the healthcare system. It is noteworthy that conflicts related to research represented 2% of the consultations to the healthcare ethics service at Memorial Sloan Kettering Cancer Center in New York [51].

4.2.4. Ethical Issues Related to Palliative Sedation

Palliative sedation is defined as the deliberate administration of sedatives in the necessary doses and combinations to reduce the level of consciousness of a terminal patient for the time required to adequately relieve one or more refractory symptoms [62,63,64]. Approximately 25% (10-50%) of patients with advanced cancer require palliative sedation in the final days of life [64].

Both the ethical aspects and procedures for the application of terminal sedation, as well as the differential criteria with euthanasia, are widely accepted. However, in daily clinical practice, dilemmas can occasionally arise related to the interpretation of the ethical requirements of palliative sedation: a) definition of refractory symptoms; b) adequacy of sedative doses proportional to the goal of reducing the level of consciousness, not shortening life span; c) involvement of the patient or their family in the sedation decision (explicit, implicit, or delegated consent).

Currently, ethical issues related to palliative sedation are infrequent. In fact, in this study, only one case was detected where the decision for palliative sedation was ethically complex (0.15%). In 2016, a study based on a survey of professionals was published to determine the frequency and difficulty of resolving 20 previously agreed-upon ethical dilemmas in the end-of-life process. The analysis of the responses concluded that the frequency and difficulty of ethical issues related to palliative sedation were low, ranking 18th among the 20 ethical dilemmas examined [16]. In the study by the healthcare ethics service at Memorial Sloan Kettering Cancer Center in New York, no consultations related to palliative sedation were recorded [51].

4.2.5. Ethical Issues Related to the Desire to Hasten Death

In the context of incurable cancer, some patients may wish to end their lives. The desire to hasten death is a complex phenomenon influenced by various dimensions of human suffering (physical, emotional, social, spiritual) [65,66,67,68,69]. In some cases, intolerable and refractory physical or psychological suffering may be the determining factor. In others, it may be physical dependency, unwanted loneliness, lack of social support, or the feeling of being a burden to others. It can also appear in patients who have serenely accepted death and consider the remaining time to live as useless and painful.

In this study, the desire to hasten death was recorded as the idea of not wanting to continue living in conditions that the patient considered intolerable and that violated their concept of personal dignity, argued reflectively and repeatedly. Therefore, patients who occasionally expressed the desire to die in the context of a specific crisis of physical or psychological suffering, or those who expressed it as a symptom of major depression, were not included. Identification was carried out in the context of the study’s systematic multidimensional evaluation, according to the clinical judgment of the treating professional, without using any instrument specifically designed to detect the desire to hasten death.

The prevalence of the desire to hasten death in this study was 1.8%. The available literature on the prevalence of the desire to hasten death is very heterogeneous, ranging from 1.5% to 35%, depending on the detection instrument used, the agreed cutoff point, and the care setting [67,68,69].

In recent years, euthanasia has been a topic of social debate in many developed countries. The core of the discussion is whether a person suffering from an incurable and progressive disease that causes intolerable and refractory physical or psychological suffering can decide when, how, and where to die. It is a discussion that seeks to harmonize the right to life and the right to freedom and dignity from a pluralistic and possibilistic perspective. Seven countries worldwide have approved laws regulating euthanasia [70]. In Spain, where this study was conducted, euthanasia was legalized in March 2021 [71]. It is noteworthy that the study presented was carried out before this law came into effect (June 2021). We do not know if the implementation of this law will influence the free and spontaneous expression of the desire to hasten death.