A 2014 systematic review [1] of studies, published between 1988 and 2012, shows that the majority of empirical research about the decision-making process when cancer treatment in children and adolescents is no longer curative is based primarily upon interviews with bereaved parents, and to a lesser extent, on interviews with attending oncologists, and on reviews of medical records. Only one study included statements from the affected young people. All these studies were carried out in High-income, English speaking countries.

The results from these studies cannot be extrapolated to Mexico because of the characteristics of its culture, especially those relating to feelings, values, beliefs, and patterns of communication in the decision-making process during medical treatment. Primary studies in the international literature [2, 3] show how different systems of values and ethnic identities influence the preferences of the parents as to informing their children, or not, about their imminent death. For example, in contrast to white Americans, the Japanese, the Chinese, and the Dutch avoid discussing death with a child with incurable cancer.

Lonergan [4] proposes that a culture is a set of skills, feelings, values, and beliefs that are shared by people in such a way that they can collaborate in the construction of what is good for the individual, for society, and in the end, for all human beings. Mexicans have a rich culture with many traditions. They call themselves “mestizos” (mixed), because their culture was shaped by their Meso-American (indigenous peoples), European, and African ancestors [5]. Mexican culture is also strongly influenced by the dominant culture of the USA, its neighbour to the north. As a result, in medicine, illness and death tend to be viewed through the lens of biomedical explanation and Mexican culture plays a peripheral role.

Although there is no single definition of the “good death”, it could be defined as one that is free from avoidable distress and suffering in patients, carers and families; and is in accord with patients’ and families’ wishes; while also being consistent with cultural, clinical and ethical standards for this stage of life [6, 7].

The Mexican healthcare system faces two principal challenges in providing high-quality medical attention to adolescents with cancer. First, for adolescents 15–19 years old, cancer is the 4th leading cause of death after accidents, assault and battery, and injuries; thus, cancer is the primary cause of death due to illness [8]. Mortality from cancer in adolescents is greater than that for those 0–14 years old, with an estimated 56% survival from the time of diagnosis [9]. According to Mexican national statistics, 21% of adolescents with cancer are attended in public hospitals belonging to the Mexican social security system and 79% in public hospitals that provide medical attention to those that do not have social security [9]. It should be noted that the Mexican healthcare system comprises public and private components, serving ~95% and ~5% of the population, respectively. In the public sector are the institutions of social security and the institutions and programs that attend the population without social security [10]. Although many hospitals report palliative care services, most of them function solely as “pain clinics” [11, 12]. There is no reliable estimation of the nature of these services. The available data are product of personal appreciations obtained by telephonic or email surveys that provide results that do not coincide among them [11, 12]. Nonetheless, it is estimated that <5% of the hospitals provide palliative care services to children and adolescents with cancer.

The second challenge derives from the wide variation in the quality of the medical attention that the population receives within the Mexican healthcare system. In a report published by The Economist on for the overall Quality of Death index, Mexico had a low score (2.7 out of 10), ranking 36th among 40 countries [13]. There is a lack of statistical data about Mexican adolescents; however, it is documented that, in general, in developing countries, like Mexico, >80% of patients with cancer suffer pain before dying [14]; pain is consistently one of the most feared consequences of cancer for patients and families [14]. In December 2014, the Mexican government, with the intent of improving this situation, issued a resolution in which the Secretariat of Health declared that the provision of palliative care is obligatory [15]. This pronouncement is aligned with the World Health Assembly resolution WHA67.19 of the World Health Organization [16]. Yet, despite this resolution and its compulsory nature, both family members and healthcare professionals, particularly oncologists, have only limited knowledge of the resolution and of how to translate it into clinical practice.

A study that included the review of 63 clinical medical records of Mexican adolescents with cancer, who had died between 2011 and 2014 in three of the main tertiary-care hospitals in Mexico City, shows that, of 40 adolescents diagnosed to be in terminal phase of their disease, 16 (40%) continued to receive treatment with curative ends. Of the 51 whose place of death was known, 45 (88%) died in hospital. Of the 41 who died within 30 days of their last hospitalization, very few, three of them (7%) received palliative treatment [17].

The study here reported is both timely and relevant. The evidence generated is urgently needed in order to understand the reality of the situation of adolescents with cancer, when treatment is no longer curative. There are no published studies that address this problem in low-to-middle income countries such as Mexico. The objective of the current study was to explore in-depth, to understand, and to explain the decision-making process from the perspective of attending paediatric oncologists, parents, and adolescents with incurable cancer, and to identify the ethical principles that guide such decision-making. The results of this study will contribute to design interventions that will help clinicians to confront the moment when further curative cancer treatment is futile in order to improve the quality of death for the adolescents with cancer and to promote the well-being of their families.

The synthesis of the interviews of the 32 participants revealed four themes: 1) the flow of information to inform decision-making; 2) the disclosure of the prognosis; 3) the decision-maker and the stakeholders involved in decision-making (their values, preferences, and beliefs); and 4) barriers and facilitators to decision-making. Some numerical data are presented only to provide the reader a better perspective on the data.

The findings of this study suggest that, in the relationship oncologist-parents/adolescent, the oncologists recognized that it was their duty to provide the parents/adolescents with the information that they (the oncologists) considered to be appropriate and relevant, thereby permitting the parent/adolescent to have “control” over the course of action that was to be followed; that is, the oncologists believed that their role is to orient the choice, making recommendation(s), and to give the parents/adolescents the opportunity to decide whether they accept, or not, the recommendation(s). On the one hand, this reflects a certain model of paternalism, because healthcare provision is tailored to the preferences of the attending oncologists. On the other hand, this reflects the oncologists’ belief that autonomous choice is a parental right that should be acknowledged and their personal choices respected. Beauchamp and Childress [22], Seedhouse [23], Lain [24], among others, have stated that autonomy is not a single right that can be ceded; rather, it is an intrinsic personal quality that can be enhanced, or diminished, depending upon what happens to or is done to people.

Our findings also indicate that the oncologists believe that the parents would have difficulty in making an appropriate, reasoned choice after the disclosure of therapeutic futility. This claim is based on unreliable personal criteria developed during a somewhat uncritical personal experience; for example, the oncologists said that they limited the amount of information presented, because they thought that "what the parents would remember will be minimal". In addition, they considered that the decision-making capacity of the parents was compromised and, also, that parents had little capacity to understand the scientific objectives and procedures of medical treatment. Therefore, they prefer to disclose the information about prognosis in negative terms (e.g., "your child no longer will benefit from curative chemotherapy") to induce a parental response that will allow them (the oncologists) to act paternalistically by protecting parents and, in the end, the adolescents, against potentially harming consequences of their own (parents and adolescents) decisions (i.e., therapeutic obstination). In this respect, Beauchamp and Childress [22] and Seedhouse [23], among others, have pointed out that the dialogue with parents/adolescent is not about communicating all the potentially relevant information, but rather guaranteeing a realistic comprehension of the important information. This implies helping the parents (in the present context) to reach a point at which they can make a reasoned and responsible choice; otherwise, it makes no sense to speak about respect for autonomy as an ethics principle when what is meant is the legal or institutionally valid authorization by the parents.

The fact that all the parents accepted the recommendations or the plan proposed by the attending oncologist (i.e., to suspend curative treatment and to start palliative management of the disease, without enquiring more into the risks and benefits) did not necessarily signify that that was what the parents really wanted. The respectful acceptance of medical recommendations is part of the value of courtesy and deference to authority that is characteristic of the Mexican culture [5, 25]. Directly contradicting a physician is considered to be a disrespectful and discourteous [5]. Klessing [5], Schuler [25], Napoles [26], Gao [27], among others, report that the extent to which one accepts unequal power relations, or ‘power distance’, denotes cultural configuration of the role expectation that govern social and interpersonal relationships. Thereby, it would be expected that patients from cultural groups characterized by ‘high power-distance’, like those in Mexico and Latin America countries, accept authoritative and “expert” recommendations from their doctors. Different from low power-distance culture, like the U.S., in which a patient from this type of cultural background would expect to share opinions, concerns, and beliefs with their doctor. Nonetheless, in the face of cases of disagreement between oncologist-parents/adolescent (e.g., when the parents insist in continuing curative treatment, or when the parents refuse to take the child from the hospital for fear that their child would die at home) the oncologists prefer to avoid possible legal repercussions by giving greater value to the obligation to respect the desires of the parents than to the medical value of beneficence. This behaviour manifests what is legally required —respect for the autonomy of parents/carers even if the oncologist disagrees with their views or actions [15]. However, ethics calls for oncologists to go beyond what is legally required and, as it has been mentioned, help the parents to rich a point at which they can make reasoned and responsible choice [22, 23].

Despite current guidelines in paediatric palliative care advocacy in the adolescent with capacity to make decisions about her/his medical treatment, including treatment during the terminal stage of cancer [15, 16, 28], the current study showed that the parents interviewed did not consider it necessary to inform the adolescents about their impending death. This appears to be congruent with the fact that parents were terrified after being informed that curative treatment for the cancer was no longer an option and then seeing their child in pain and with other unpleasant symptoms. In these circumstances the last thing that parents wanted was to make their child aware that curative treatment had failed. Also, to be considered is the fact that Mexicans have a familial orientation and the person always has to be thought of as a member of a family. This assertion is supported by other studies [29, 30] that report on the importance of family in decision-making at the end-of-life in Latin-American families and on the belief that truth-telling about prognosis is harmful to the patient; hence, Latin-American families, and Mexican families by inclusion, prefer to not discuss death openly. Understanding this cultural preference is important, because it forestalls us, and rightfully so, from the risk of assuming that respect for autonomy is an absolute moral obligation, as discussed below, and also from the risk of placing too much weight on top-down guidelines [15, 16] as the source of moral authority.

This study also revealed that the oncologists prescribed the suspension of curative treatment, and the start of palliative management of the cancer, when they concluded (based on tangible facts such as the number of lines of treatment employed, number of relapses, and radiological and laboratory evidence) that curative treatment was no longer beneficial. The consequent ethical question asks if, in face of therapeutic futility, the biological progression of the disease is the only important concern. This question raises some points that deserve discussion.

First, therapeutic futility is the expression of the combination of a scientific judgment and a value judgment (by the oncologist) in the sense that it acknowledges the uniqueness of the individual (the course of the disease, values, context and the physiological idiosyncrasies) [22]. The ethical and practical problem occurs when only the oncologist’s own values and objectives are considered when making the judgments. The parents’/adolescents’ perceptions of harm and benefit are idiosyncratic, peculiar to each patient or family, and depend on their way of life, on how they perceive themselves, and on the emotions and the perceptions awaken in them. Therefore, consideration of the needs, values, and preferences of the parents/adolescents is part of the medical principle of beneficence—a principle professed by the oncologists. Lain [24] and Campbell [31], among others, point out that, in a doctor-patient/family relationship, a co-operative relationship (co-execution) between the parties is indispensible.

Second, the foregoing leads naturally to another argument. Within the context of therapeutic futility, when beneficence is technically conceived and realized appropriately, it takes precedence over respect of autonomy of parents/adolescents. Beneficence is not just to do with physical functioning; it is at least equally to do with the mental life of a person/patient [22, 23]. The last statement is important in making decisions when treatment is futile. For instance, it would be inappropriate if a parent, in the face of therapeutic futility, were to insist that curative treatment be continued. In such case, the principle of beneficence would supersede the principle of respect for autonomy. Even though the parents are the primary decision makers and the oncologists have the obligation to respect parental choices, when parental preferences increase the risk that their adolescent children will be harmed, such preferences are inappropriate. This should not be read simply as attempting to maximize physical benefit or minimize suffering. To educate parents/carers about the effect of treatment on their child body, and ensuring that an extremely stressed parent/carer does not become a victim of carer burden by allowing her/him to take a rest, are both actions within a continuum that would qualify as a process to enable autonomous parents/carers. Education for health is not just to do with physical functioning, it is work for reaching wholeness; that is why ethics is process-based and dialogical, rather that rule-based.

Third, although all the oncologists considered it important “not to cause more harm” to the adolescent with terminal stage cancer, the fact that some oncologists preferred to prescribe further curative chemotherapy seemed to depend more on the importance they attached to either curing the cancer or prolonging a patient’s life. These findings are consistent with the report of Buiting et al. [32] that shows that the tendency of oncologists to extend curative chemotherapy until the terminal stage of cancer, and to strive to prolong the patients’ life, could be explained not only by patients’ and oncologists’ mutually reinforcing attitudes of not abandoning curative treatment; but also, by the oncologists’ belief that removing a patient’s hope by withdrawing or withholding curative treatment is harmful. It should be mentioned that these findings do not solve the controversy concerning quantity versus quality of life. Nonetheless, the fact that the judgments concerning the correct course of action, in the face of therapeutic futility, always should be specific to each individual patient does not mean that these judgments are simple matters of opinion or solely to be determined by the personal desires of the parents and/or adolescents. A judgment of futility requires that the diagnosis and prognosis be based on the best available research evidence along with a continuous and forthright dialogue between the oncologist and the parents/adolescent. The dialogue must lead the oncologist to understand the entire personal and social reality of the patient and his/her family and must also allow the parents and the adolescents to understand what is happening and what is possible and what is not possible. The preferences of the parents/adolescents are crucially important, but not necessarily decisive.

According to the participating oncologists, the palliative management of cancer should begin when adolescents become terminally ill; i.e., when due to the biological progression of the disease, they determine that the treatments have little or null probability of success and a higher probability of being more prejudicial that beneficial. Inherent assumption to the biomedical model, which is based on an objective science, in that explanations of disease focus on biological changes to the relative neglect of social and psychological factors [33].

Most oncologists identified the following as their major barriers for the determination of, and timely communication about, the terminal cancer stage: their own poor training in conducting end-of-life conversations; the emotional ties with adolescents they had known since childhood; the problem of having to confront some parents who “do not understand” and “do not accept the prognosis”; and the uncertainty in prognostication per se. These results are consistent with those reported in other studies [34, 35], which highlight several factors associated with the attitudes of the physicians that influence the determination of the moment to communicate that the adolescent is at the terminal cancer stage. Among such factors are: the difficulty in arriving at an accurate and precise prognosis, poor or lack of training in communication concerning end-of-life issues and discussions about them, discomfort about talking about poor prognosis, and about death and dying because it provokes emotions difficult to manage by all participants, and the time available for such communication.

The oncologists recognized that palliative care is of great significance, because it can improve the life conditions of adolescents with incurable cancer. That is why they identified the following as urgent needs for their hospitals: having a clinic or palliative care service for referral of patients, an appropriate place to discuss therapeutic futility, and training (for them, the oncologists) in psychology and/or palliative care. As to the idea that the optimal place for the terminal stage is at home, the current policy of the Mexican government is that patients should be supported to die at home, should that be their preference [15]. All the participating oncologists attributed immense importance to the home as final place for end-of-life care. But if such importance is given to the home, then appropriate support mechanisms are required for adolescents and their parents/primary carers, such as educating families on what to expect; connecting families to healthcare professionals and community support networks; bolstering parent coping and developing parental problem-solving skills; as well as psychosocial screenings to identify family and patient risk factors to support the needs of patients [36]. However, some of the participating parents preferred the hospital, either because they feared that at home they would be providing little control of the symptoms or because they placed their fears and anxieties concerning the control of pain and symptomology at the last moments of their child’s life ahead of anything else. Guidelines for palliative care [15] are necessary, because they establish what should be an appropriate professional conduct; however, they are not enough. Training in ethics must be provided to assist the oncologists in achieving and maintaining ethical and technical professional standards.

It is relevant to mention that there was a difference among the values of the oncologists, the parents, and the adolescents. Oncologists valued curing the disease and prolonging the patients’ life. They considered it important that the adolescents be involved in the decision-making. In contrast, the parents valued honesty in communication, clarity of the information provided to them, and an oncologist with a deep involvement in their relationships; for them, respect for the autonomy of their child was not as important. Likewise, the adolescents said that they valued honesty from their oncologists. The qualities valued and demanded by the parents and adolescents pertain to the principles of beneficence and non-maleficence and serve as a framework for understanding what the participants tended to value at end-of-life. Awareness of this diversity of values at play can help to avoid important clinical consequences in the oncologist−parents/adolescent relationship, such as: difficulties with informed consent, resentment toward a detached doctor, decreased satisfaction with care, and miscommunication.

The knowledge generated in this study is valuable, because it permits to understand how the decision-making process is understood and constructed by the principal agents in a terminal cancer, especially when the decisions have great consequences not only for the sick adolescent, but also for the family. Also, it provides evidence of the compelling and urgent need for undertaking a more effective implementation of palliative care, especially in developing countries such as Mexico. Likewise, it shows that appropriate training and continual education in the ethics of quotidian clinical practice is indispensible in order to achieve a good medical practice.