Analysis of the documents shows that all the dimensions of end-of-life care found in the literature and included in the framework (see Table 1 : Thematic grid) are echoed in the statements of the most representative organizations committed to the definition of policies and guidelines for palliative and end-of-life care. It is worth noting that all the national organizations found according to our research strategy belong to English speaking countries. This might be due to the fact that it was in these countries that the palliative care movement first developed and flourished in the 60s and 70s.
In general, the "sub-areas" of symptom control (i.e. A1, A2 and A3) as well as those referring to relational and social issues (i.e. B1, B2, B3 and B4) are more widely covered by the documents than the "sub-areas" related to "preparation" (i.e. C1, C2 and C3) and to "existential condition" (i.e. D1 and D2). This result is consistent with what is stated by several studies showing that the control of symptoms and of the psychosocial dimension of dying [15, 17, 25, 26, 35–41], is given a higher relevance than the control of the dying process by the patient himself [15, 19, 25, 26, 34, 51, 55].
With regard to symptoms, the control of pain and of psychological distress (i.e. A1 and A2) is acknowledged as fundamental, while being assisted by a staff member in order to make the process of dying more comfortable (i.e. A3) is considered as less relevant. This result seems to be counteracted by the evidence from the literature, which shows that being comfortable is seen as important both by patients and by health care professionals .
As to the relational and social dimension, a large number of documents state that individual preferences as well as personal values and beliefs (i.e. B1) should be respected and honoured. This issue has been extensively discussed in the literature [4, 12, 23, 25, 51, 56, 60] and is particularly relevant for patients sharing cultural values which are different from those dominant in society .
Most documents combine the respect for personal beliefs and values with the importance of addressing one's spiritual needs and of facilitating religious practices (i.e. D2), thus showing consideration for individual preferences both from the relational and from the existential perspective. However, the importance attributed to respect for individual preferences seems to be in contradiction with the minor weight lent to the choice of the place of dying (i.e. C2).
Further discrepancies can be found between issues related to preparation and issues related to the relational and social dimension of dying. Indeed, many documents recognise the importance of good communication between the patient and the caring staff (i.e. B3), and state that communication should include information about diagnosis and prognosis, as well as the discussion of issues related to death and dying. Yet, this result jars with the fact that only a few documents refer to the awareness of diagnosis and of impending death (i.e. C1), an omission which is even more striking since how often Western surveys address this issue [4, 12, 23, 25, 51, 53, 57, 60].
It might be argued that, due to the discrepancies between the element of preparation and the relational and social area, it is not possible to derive from the documents an integrated model of best palliative care practice. In particular, it might be suggested that the documents do not offer a coherent model for policies directed to the actual empowerment of patients in the decision-making process. This is especially evident with regard to end-of-life decisions. While there is a general agreement between the documents that patients should take part in end-of-life choices, withdrawing and withholding of life-sustaining treatments are viewed as the result of an evaluation that is mainly up to the doctor. Indeed, the consistency of end-of-life choices with the patient's wishes is given only a minor relevance.
Another result that is worth discussing is the small relevance given to being in control of oneself (i.e. C3). Only one third of the documents refers to this item, whereas it is one of the most important elements of end-of-life care present in the literature . This might be due to the fact that the maintenance of control is a multifaceted and a patient centred issue, which can only with difficulty be addressed by policy statements on palliative and end-of-life care. The same conclusion may be suggested for completion and meaningfulness at the end of life (i.e. D1), which is given a minor relevance in the documents, while several qualitative studies on patients, families and caregivers account for this item [12, 17, 23, 56, 60, 62].
On the other hand, there seems to be no persuasive explanation to the fact that minimising the burden of caring on the family is hardly covered in the statements. While, according to the literature, freedom from financial and physical burden is considered by patients and caregivers as one of the most important component of a end-of-life care [17, 56, 58, 59, 61] only a few documents address this issue. It might be suggested that, in general, the statements tend to consider "financial and physical" support to the family as less relevant than "emotional" support. In fact, a great number of documents consider emotional support to the family (i.e. B2) as reasonably important. Nevertheless, this consideration does not prevent from observing that minimizing the burden of care is generally underestimated in the documents.
Finally, it is worth noting that a definition of quality of life at the end of life is lacking, whereas this issue is widely covered by the documents. Of course, the concept of quality of life is highly individual and fluid, and it might be difficult to give a precise definition of this notion. However, several studies have already proved the possibility of developing instruments to assess both the quality of life for dying patients and the quality of care at the end of life [52, 57, 59, 63, 64]. Therefore, it might be observed that the documents generally fail to address the issue of quality of life in a consistent and precise manner.