Comparison of content
When comparing the contents of the booklets, two versions appeared: booklets that were based on the 2005 Canadian Comfort Care Booklet (the 2021 Canadian booklet, the Czech booklet, the Italian booklet, the Dutch booklet and the UK booklet) and booklets that were based on the 2017 UK booklet (the Irish booklet and sections of the 2021 Canadian booklet), see Table 1 (Additional file 2). The booklets that were based on the UK booklet thus had not used the original Canadian booklet as the starting point, but were based on the UK booklet -retaining the adaptions that were made in the UK booklet. Both healthcare professionals and family caregivers were involved in evaluating the contents of the booklets and the adaptations. This involvement ranged from participation in a study leading up to the development of the booklet  (indirect involvement), to “collaboration and co-production” which entailed team membership and contributing to key decisions . Healthcare professionals largely influenced the content revisions. Revisions often concerned the local legal frameworks and shared decision making practice.
The key topics that were present in all the booklets were the progression of dementia and possible complications such as eating difficulties and infections, the decision-making process about treatment options at the end of life, palliative care and managing symptoms such as pain, breathing problems and anxiety, the dying process and common emotions and procedures after death. These key topics were retained from the original Canadian Comfort Care Booklet.
We arranged the textual revisions of the booklets compared with the original texts in six categories: 1.Typology of treatments and symptoms at the end of life, 2. Patient and family rights and wishes, 3. Typology of decisions at the end of life, 4. Indirect or explicit messages, 5. More or less positive about prognosis, and 6. Relationship between healthcare professionals and family caregivers.
Typology of treatments and symptoms at the end of life
The booklets differed in the treatment options that they described, and the level of medical detail that was provided about symptoms and treatments. Three topics related to treatment differed the greatest between the booklets: artificial nutrition, life-terminating or life-limiting treatments and spiritual care. Whereas all booklets discussed eating difficulties in advanced dementia, the UK booklet did not contain information about tube feeding or any objections to it (Table 2 (Additional file 3), quote A). Instead, extensive information about oral hygiene was provided and this was also included in the Irish and updated Canadian booklet. The Czech booklet provided detailed information about alternative feeding and food options to address eating difficulties. The Czech and Italian booklet provided more medical information about the complications of artificial nutrition during the dying process than the other booklets. This was included because it was considered difficult to convince family caregivers not to start artificial feeding at end of life. Regarding information about euthanasia, this was not included in the Irish and UK booklets, while more elaborate information was provided in the Dutch booklet and updated information in the Canadian booklet. The Czech and Italian booklets mentioned euthanasia only to state that it is not a viable option. The Dutch and Canadian booklets had included additional information about palliative sedation. The UK, Irish and updated Canadian booklet offered information on spiritual care, which was not included in the Czech, Dutch and Italian booklets.
The extent to which the booklets offered detail about medical issues varied, and this may relate to difference in whether healthcare professionals or family caregivers were the most influential in the revision process (see also Table 1 (Additional file 2) Stakeholder Involvement). The Czech, Dutch and Irish booklets contained the most information regarding medications and physical health and the UK booklet the least. For example, the Czech, Dutch and Irish booklet provided detailed information about the breathing pattern during dying or extra information about pain management options. In contrast, the UK booklet did not speak about the medical complications that could arise after hospitalization when addressing why transfer to hospital may not be appropriate, while the other booklets did.
Patient and family rights and wishes
The varying legal systems of the jurisdictions were apparent in diverging emphasis on patient rights and wishes between the booklets. The Czech and Irish booklets stood out the most in this respect. The Czech booklet included an entire section about living wills and legally binding wishes to refuse care, driven by the Health Services Act, No. 372/2011 Coll . Emphasis was placed on acting in accordance with living wills throughout the booklet. The Irish booklet contained several sections that asked family caregivers to think about previous wishes of the person with dementia, to ensure that any decision making is aligned to the person’s previously stated will and preference. Reference was made to Ireland’s Assisted Decision-Making (Capacity) Act 2015 . The UK booklet and Dutch booklet only referred to patient wishes for specific treatment decisions, such as CPR and the use of antibiotics, respectively.
The Italian booklet clearly stated that family members provide only information for the decision-making process. This mirrors the updated Canadian booklet, that had adapted the terminology to current legislation regarding shared decision making (Table 2 (Additional file 3), quote B). That is: the family caregiver was referred to as “the patient’s legal representative” instead of “the mandatory” (a term formerly used to indicate a representative by law) and a statement was added about variations in provincial laws across the country regarding the role and rights of this person. The updated Canadian booklet, and the Dutch booklet, introduced family wishes only in the section on providing the last care after death.
Further, no mentioning of settling disagreements in court was made in the Czech, Italian and Dutch booklets. The Czech booklet also did not refer to the assistance of ethics committees. A final difference between the booklets was that only the Dutch and Canadian booklets contained information about the legal requirements for life terminating treatments.
Typology of decisions at the end of life
A similar category of differences between the booklets related to the description of end-of-life decision making. The Irish and Czech booklets emphasized best interest decision making, involving the family caregiver. The UK booklet underscored the responsibility of the clinician or medical team to facilitate this.
When describing considerations for decisions or treatment, the booklets varied in whether they stated the underlying dilemma. The UK and Italian booklets often did not include the dilemma. For example, they did not refer to pneumonia as “the older person’s best friend” (an argument against curative treatment of pneumonia, Table 2 (Additional file 3), quote C). Dilemmas were sometimes emphasized in the Czech booklet, for instance by adding the statement “even at the cost of reduced comfort” which suggests curative treatment can be incongruent with comfort care. The moral acceptability of treatment decisions was, at some places, omitted from the Dutch and Irish booklet. The decision to increase doses of morphine at the end of life to reduce suffering was therefore more a medical than a moral decision, for instance.
Indirect or explicit messages
An evident difference between the booklets was their layout. While the Dutch and original Canadian booklets contained images of moments of caring, the UK booklet contained images of nature. The Italian, Czech, Irish and updated Canadian booklets were in the middle of this spectrum and showed images of their local nursing home contexts. In addition, the Irish booklet contained images of nature.
Differences between explicit messages or more softened, indirect messages were also found in the text in terms of style. The UK booklet used more softened language, for example comparing breathing problems to asthma. This booklet also spoke about nausea or discomfort, similar to the Italian and Irish booklet. The Czech, Dutch and Canadian booklets instead mentioned vomiting and pain. The Czech booklet typified useless or harmful treatment in the last days or hours of terminal illness as “dysthanasia”, detained death, and mentioned more confrontational treatment details.
All booklets considered dementia as a terminal condition, but some booklets were more explicit about this. The Czech booklet further contained explicit statements about the non-curable and terminal nature of the disease causing dementia syndrome. Also the Irish booklet explicitly mentioned the dying phase several times. The Dutch booklet clearly related not eating and drinking to the dying phase (Table 2 (Additional file 3), quote D).
All booklets recommended a palliative care approach based on physical and psychological comfort; the Canadian and Italian booklets concluded with the statement that “That’s because the majority of people perceive that advanced and prolonged dementia is worse than death”. This statement was not incorporated in the other booklets.
More or less positive about prognosis
There was some variation within and between the booklets regarding the description of the prognosis. The Czech booklet started with the limited life expectancy and cause of death in the introduction (Table 2 (Additional file 3), quote E) and therefore appeared less positive about the prognosis compared with the other booklets. The symptoms that were described in this booklet were mainly possible causes of death, as was the case for the Canadian, Dutch and Irish booklets. The UK and Irish booklets had additionally included symptoms related to activities of daily living, describing less severe stages of dementia. This encompassed a more holistic tone and upstream approach regarding prognosis than referring only to symptoms around the end of life.
The Czech booklet was less positive about prognosis throughout the booklet, for example stating how certain treatments may not be tolerated by the person with dementia. The more positive tone about prognosis of the UK booklet was also present throughout, for example by not stating some negative consequences of treatments. The Dutch booklet was more positive about prognosis in some sections: a maximum estimate of survival was provided for people who do not eat (instead of a time window that included a shorter time estimate). However, in other sections, the Dutch booklet was less positive about prognosis: it included the statement that the “final stage can be long and exhausting”.
Relationship between healthcare professionals and family caregivers
Two booklets stood out regarding the relationship between healthcare professionals and family caregivers: the updated Canadian and Irish booklet. Both had included information about family involvement in care and this was particularly present throughout the Irish booklet (Table 2 (Additional file 3), quote F). The other booklets did not include this information, apart from sitting in at the end of life. The Irish booklet additionally referred to several healthcare disciplines throughout the booklet, which supports the multidisciplinary nature of palliative care. The other booklets mainly referred to physicians and nursing staff.
According to the final consensus mapping, all EAPC first nine domains defining optimal palliative dementia care were addressed in all the booklets, as depicted in Table 3 (Additional file 4). However, not all specific recommendations within the domains were addressed by all booklets. Recommendations with regards to `setting care goals and advance care planning’ were addressed the least, especially in the Canadian and Italian booklet, while the Irish booklet addressed some of the specific recommendations. Supporting people with mild dementia in advance care planning (recommendation 3.4) was not mentioned in any of the booklets, as all booklets described the advanced stages of dementia since the booklets are positioned at the end of life, where decision making capacity may be limited. Recommendations that were also not addressed by any of the booklets related to `Continuity of care’ (having a central care coordinator and appropriate information transfer between healthcare professionals) and to `Optimal treatment’ (interdisciplinary consultation between dementia and palliative care specialists).
The Czech booklet was the only booklet that addressed recommendation 2.5 about previously expressed preferences regarding place of care (domain 2: Person-centered care). An explicit statement about avoiding the use of restraints (recommendation 6.3, domain: Avoiding burdensome treatment) was found only in the Irish Booklet.
Based on our overall findings, we present guidance statements regarding the contents of informational booklets for family caregivers about dementia and palliative care  in Box 2 (Additional file 5). This may inform future updates or wider adoption of the booklets and support the development of other educational materials for family caregivers in this area.