Direct verbatim quotes from the interviews are used to illustrate the findings and facilitate the discussion. Bold text was used to indicate what participants emphasised. Words added by the researcher for clarifications are enclosed in [] and […] represents omitted text. Three narratives were generated: (i) Shifting expectations; (ii) Relational interdependency and (iii) Trigger points.

This narrative focuses on insights concerning how PwD’s decision-making and expectations toward their future changed over time as the disease progressed. The shifting expectations are essential to understand since they tend to influence PwD’s ACP process throughout. Under this narrative, two types of plans regarding PwD’s future expectations were identified: predefined discussion and agreement (what PwD had previously expressed and wanted) and interim decisions made by carers (what carers deemed best or thought that the PwD would have wanted).

During the early stage of dementia, daily decisions on routines activities and expectations about their potential future life were typically similar to the pre-diagnosis period and were made by PwD themselves since they were able to express and maintain their physical independence and needs (Basic Activities of Daily Living (ADLs)) as well as complex activities to live independently in the community (Instrumental Activities of Daily Living (IADLs) [28]. Such independence was indicated from Mike who had had the Alzheimer’s disease for 6 months:

‘I get up in the morning, have breakfast, go out and buy a paper at the paper shop’s just around the corner. Then we might be going for a walk or going up a hill and various other things. But nothing much has changed really, so far, to be honest’.- Mike (69), living with young onset Alzheimer’s for 6 months.

Mike positioned himself as an unchanged person. However, he hinted at some lingering uncertainty that his daily life had not changed ‘so far’, implying his awareness that this could change. Nevertheless, he preferred to avoid discussing and planning for his future self:

‘Mike: …I try not to think very much about the far…part of the [my dementia] [pause] [abruptly]. Yeah, I just don’t like thinking about that. Butfor now, you know, life’s good, and we can have a good time. Interviewer: And just focus on the present. Things you can do and so on? Mike: Yeah.

His uncertainty avoidance was paralleled with the acknowledgement of his decline associated with dementia when he recounted his direct experiences of caring for his parents, both of whom had dementia. This was narrated along with his professional experience to dementia and how he was aware that ‘things will get worse’ in the future:

Mike: I know that this [dementia] is in my family so…[pause]…it’s quite scary.

Interviewer: is it scary because you see…what they had become? At the later stage?

Mike: [nodded] Yeah […] And as a Church of Scotland minister, I have seen it a lot. I…took …[searching for words]…funeral services for many, many people who have been having this sort of (diagnosis of dementia) so yeah… (I know what I will become in the future) […] [I’m aware that] things [my dementia] will get worse. But for the moment, I try and live in the moment these days and just not to…get ahead of myself.

Nevertheless, despite Mike’s personal and professional experiences with PwD, he appeared to avoid planning for his imagined future self and preferred ‘not to get ahead’ of himself, indicating an ongoing avoidance of the future and focusing on the present.

Mike’s coping mechanism is similar to Jodie who preferred to ‘take every day as it comes’:

I do what I want, when I want and it’s great. I don’t think about it [future plans] too much. I just take every day as it come. .- Jodie (56), living with young onset Alzheimer’s for three years.

Nevertheless, she expressed a declining ability to keep up with her friends due to her dementia symptoms:

[…] we have our small group [our book club members], sometimes, I’m finding it hard to keep up [with conversations] Or…did I miss something, sort of feeling. It’s not their fault, it’s just them talking as normal (but I still cannot keep up with them).

Conversations with both Mike and Jodie revealed the story of individuals that continued to take control of their routines despite the lingering uncertainty towards the future or fading independence. At this early stage of dementia, participants’ expectations for the future tended to focus on daily decisions or activities. They preferred not to address future decisions as it required them to envisage a different, perhaps baffling future self, which they preferred to avoid.

As the disease progressed, PwD’s daily decisions would often be made along with and by carers due to PwD’s declining ADLs and IADLs. Carers could act as PwD’s advocate and safeguard the PwD’s decision, similar to Fiona who supported her mother to thoroughly discuss her ACP with the rest of the family:

She [mum] wants to stay at home for as long as she can. There’s a care home just at the end of the road and she thinks that’s where she should go. She also wanted no resuscitation thing in place. So, I’ve just dealt with that. It caused some upset in the family but [the family] do acknowledge it is not their choice – Fiona (59), carer of her mother who is living with Alzheimer’s for 20 years.

Alternatively, carers might feel that they needed to safeguard PwD, co-create plans with them, or even take over their planning entirely, as indicated by Graeme who decided everything for Beth, his wife who is living with Alzheimer’s disease. Graeme recounted that they did not discuss ACP in-depth at all since Beth felt that she was ‘not ready’ and ‘not there yet’. Consequently, all decisions about her daily life as well as decisions about ACP were made and adjusted by Graeme. This stance was grounded in their commitment to each other as spouses which they believed that they were well positioned to make the best decisions for each other:

The other commitment I made to her [as a spouse] is that “I will not do anything to you or for you that I wouldn’t expect you to do for me if it was the other way around”. So, we have that commitment and she knows I wouldn’t (do anything else except that) […] My plan is to keep her home as long as possible, with as much help and support as necessary. Until such time, if she got to stage that if she was dangerous to herself, to me or the neighbours, then she may have to go into a more secure environment. But I will keep her at home for as long as possible- Graeme (70), carer of his wife for seven years.

Graeme’s stance to take over Beth’s decisions contrasted with David who tended to discuss his ACP as a joint changing needs with his wife who is his carer. He recounted how he thoroughly planned his ACP ‘to the last detail’ with his wife as follows:

David: We [my wife and I] have planned our lives now down to the last detail. Planning is important because you don’t know the route of your journey through dementia whether it’s going to be quick or slow […] My wife says I need to keep a memory box now and this should be useful to her as well as me later on when she wants to remind me of things and I’m keen to do that.[…] We’ve planned. No doubt we’ve made some mistakes but we’re quite proud of the planning that we’ve done. […] We’ve also planned how we’re going to treat each other as life goes on. And she also is not to treat me like a child. If there are any decisions to be made, I’d really like to be a part of them.[…] [we planned everything together because] My wife’s opinions are very important, they are vital, and so is her welfare… –David (66), living with Alzheimer’s for two years.

David was fully aware of the unpredictable disease trajectory and how his ‘journey through dementia’ could be ‘quick or slow’. Consequently, David used this knowledge as leverage to thoroughly discuss expectations for the future and plan decisions with his wife. Their discussions comprised two types of planning on David’s part: pre-defined discussion (‘the planning that we’ve done’) and potential interim and iterative planning for the future (‘how we’re going to treat each other as life goes on’). These different types of planning emphasised the iterative nature of ACPs; that is, ACPs need to be reviewed and, where necessary, revised over time by the persons that would be affected by such decisions. By discussing their plans together, David and his wife could strategise on how to safeguard their decisions about their future lives and, over time, revisit and adjust their plans in ways that are practical for them both whilst accommodating expectations for the future.

Although, their pre-defined discussion may change in the future, these co-created discussions indicate a relational, iterative approach of ACP between David and his wife thus might prove more realistic for both. For example, the extract shows the co-constructed planning process that occurred between David and his wife when she suggested he create a ‘memory box’, which he was ‘keen to do’. This suggestion highlights the relational nature of the couple; that is, her input also influenced David’s decisions on what he wanted to be enshrined in his ACP as well as potentially affecting both their future lives when ACP decisions may be enacted. As such, David’s future plans did not totally originate from his individual rationalisations, rather, the plans were relationally founded and were originated from his co-existence with his wife (self-in-relation-with others) based perhaps on intimate knowledge of a past lived together.

David’s extract suggests the shifting expectations of his decisions and future plans from individual needs to a shared-decision process with the person who was involved in his care: his wife. This transition juxtaposed and challenged the public narrative of individualistic ACP and emphasised the relational collective shared decision-making between PwD and their key persons. Consequently, this emphasises the importance of PwD’s ACP that will gradually transform over time and need to be co-constructed with others.

This narrative explored the notion of relationality within interrelationships between PwD, their family and HCPs that influence the ACP process. Dementia symptomatology can variably affect how or if a PwD can conceptualise and express their wishes for the future within the context of planning and writing their ACP [8, 29]. Participants suggested several symptoms that can variously influence their ACP over time which are PwD’s declining reading and writing capabilities as well as fluctuating memory. As such, for most PwD, the support from their key person (s) can help to overcome the difficulty surround the ACP process. In this study, the key persons often were PwD’s carers or professionals who have an established, trusting relationship with PwD. Specially, their tacit understanding and appropriate support within the relationship seemed to allow PwD to initiate and revise an ACP accordingly.

One example came from Mary who had difficulty reading writing any documents due to her dementia symptoms but found her dementia link worker useful to support her ACP process:

Mary: I did them [ACP] with my…dementia worker. Over long periods. And I actually just put it in yesterday.

Interviewer: So you and your dementia worker create that together?

Mary: Yes we did it together. Well, I talked and she typed (laugh) […] ‘cause I struggle (with words and forms) um…that’s another thing. Forms, paperwork, they’re all of a struggle and so she just asked questions [think]. Yes, she asked. We’ve been over questions and then she would…[think] type it.- Mary (51), living with Alzheimer’s for three years.

From the extract, it appears that Mary’s dementia link worker discussed ACP with her over an extended period, indicating their established relationship. This would have allowed the link worker to understand Mary’s preferences, her ways of communicating and concerns and provide Mary with appropriate support to complete her ACP. This is suggested when the dementia link worker transcribed Mary’s verbal preferences of her ACP (‘I talked and she typed’) thus mitigating Mary’s ‘struggle’ to understand words and forms and helping with the ACP documentation process. Therefore, Mary’s ACP was perceived as ‘completed’ due to Mary’s trusting relationship with the link worker and the link worker’s skillsets to support the process.

The interrelationships amongst carers and other family members also influenced the ACP process and how it was likely to be adhered to, as Clara highlighted the ‘excellent support’ she received from her extended family who adhered to her mother’s ACP which possibly stemmed from her mother’s early discussion of ACP and clear communication to all relevant persons regarding her ACP:

She [mum] was very clear and communicating to everybody around her, how she wanted things to be. As a result, they [my aunt and uncles] have been hugely supportive of mum and of me taking care of mum. They have been an excellent support to me. There hasn’t been any conflict or any disagreements about how to proceed with any of her treatment. – Clara (49), carer of her mother for six years.

Conversely, Donna, who was also caring for her mother. She ‘created’ her mother’s ACP ‘years ago’ with her sister yet her male siblings were not involved nor engaged with the discussion. Consequently, the lack of engagement from her male siblings resulted in a ‘fighting’ in her family over their mother’s care due to ‘different understanding’ of their mother’s conditions and changes that needed to be made:

I think what was lacking is [us siblings] having some day sitting down and being a little devil’s advocate and saying “Well, what happens this and what happens that”. It was just decided that ABC will happen, and this is what we’ll do [yet things did not happen that way][…] There was a breakdown within our family unit [after I revised mum’s ACP without telling my brothers]. A lot of fighting, which probably could have been avoided if things were just different and we thought things through…-Donna (50), carer of her mother for 15 years.

These extracts highlighted the complex interrelationships between PwD, carers and extended families that influenced the ACP process. Primary carers might feel the need to initiate an ACP with PwD, but their intentions might not be in line with the extended family. Consequently, the ACP might not be initiated due to this divergence.

Similarly, the extended family might perceive the initial ACP discussion as final. Yet PwD and carers may wish to revise their ACP to make it more reflective of changed circumstances. This intention could potentially be misunderstood by the extended family, thus creating discrepancies in care or disagreement over the ACP.

The perceived variable degrees of support from HCPs can function as either facilitators or inhibitors of the ACP process. Several participants framed their understanding of this phenomena as ‘postcode lottery’ effect. This conceptual understanding differs from the literature which defines the phrase as “variations in health care between different geographical areas that appear arbitrary and unlinked to health need” [30] (p.1). In this study, the effects of ‘postcode lottery’ goes beyond the geographical areas and included the relationships between participants and HCPs as well as the perceived variable quality of service that participants felt.

The ‘postcode lottery’ effect was suggested by Mary when she recounted the contrasting experiences of positive support from her dementia link worker and the lack of support when caring for her mother-in-law, who also had Alzheimer’s disease:

Mary: The support here is very good whereas where I used to stay, I wouldn’t have.

had what I have here and that’s only…less than 20 miles [away] and the support is just.

totally different! I count myself lucky. I’ve got my dementia worker, I’ve got a CPN nurse. I have.

my consultant […] I don’t know about all these things when I cared for my mother-in-law. I didn’t know about carer support, [dementia support] groups and things. I think it’s a postcode lottery, to be honest.

As a PwD, Mary had already completed her own ACP. Yet her experience as a family carer of a PwD was different. Mary did not find the support from HCPs for her and her mother-in-law helpful; Mary was not aware of the support to which she and her mother-in-law were entitled (‘carer support groups and things’). As such, this resulted in the lack of ACP discussion between Mary, her mother-in-law and HCPs. She pointed out that the difference in support was based on the area that she lived and how that had a direct impact on her ACP process. This created a surprising contrast in support from HCPs between two areas which were ‘less than 20 miles away’ from each other. Hence, Mary framed her situation as ‘lucky’ as it was unclear why she and her mother-in-law received different levels of care and support that PwD are entitled to.

This narrative encapsulates the timing and reasoning behind the initiation and revision of ACP and examines if an optimal time for the process can be deduced. The majority of participants recounted pivotal moments that triggered the initiation and revision of ACP— the ACP trigger points — throughout the dementia journey ranging from before dementia diagnosis, immediately after the diagnosis or years after. These ACP trigger points encompassed broader discussions beyond medical aspects, emphasising the totality of ACP.

Initially, the contents of ACP that participants discussed tended to be based on broad discussion towards future scenarios and lacked details. Donald who was caring for his wife (Jackie, living with Alzheimer’s disease) created their ACPs together prior to Jackie’s diagnosis after his retirement, making this their first trigger point:

Once I retired, we tried to get some of these [ACP] sorted out so it would be about 2000 and…14-15-ish (2014 or 2015) when we did that? Then we updated it. We transferred it to the local solicitor when we moved [to where we live right now], so that’s been in place about five years. We were prompted by the fact that my parents had left it rather late and it was all a bit difficult towards the end, so we determined to be proactive, so probably predated Jackie’s dementia’. – Donald (70, carer of his wife) and Jackie (68, living with Alzheimer’s for five years).

During 2014–2015, Jackie had not been diagnosed with dementia yet. As such, their ACP discussion was focused on non-medical aspects of their situation which was adequately reflective of Jackie’s needs. Later, they revised Jackie’s ACP with her GP to include her future care and end-of-life care preferences one year after her diagnosis of dementia (e.g., their second trigger point):

Jackie: It was the GP [that started the ACP conversation]. She said had I had the plans [ACP], or did I have any thoughts about the future. And we talked about it and she wrote down this (my DNACPR). [speaking to Donald] She wrote down this plan, wasn’t she?

Donald: Yes, she did this [pointed to Jackie’s ACP document] (reading out Jackie’s ACP documents) […] There is a note on the bottom which says, “please involve the patient and husband in all health care decisions”. It does include, I think, a sense of discussing with relatives, so that we can…not overrule but, but perhaps interpret sensibly, what Jackie’s [wishes are] about?

The couple included Jackie’s future care preferences and explicitly stated in the ACP document that Donald would be involved with her future decisions. As such, Jackie’s revised ACP had changed from the original one in 2014 and was more reflective of their ‘current’ wishes between Donald and Jackie’s, as seen from the note on the bottom of her ACP. After this revision, Donald pointed that he should be able to safeguard Jackie’s preferences as to what she would have wanted (‘not overrule but, but perhaps interpret sensibly’).

Participants often considered the ‘official diagnosis’ of dementia as another trigger point for ACP discussion. This moment led to either the initiation or revision of ACP. PwD who had already had any prior discussion on their ACP, the discussion afterwards tended be expanded to include place of care and end-of-life care, which, again, emphasises the iterative nature of ACP:

Clara: At her diagnosis [when mum created her ACP with me]. She was very determined to write her will and deal with power of attorney. […] She was pretty vocal about how she wanted me to support her.

Apart from the official moment of diagnosis, several PwD interpreted other key poignant situations as their ACP trigger point. Such situations could be years after the diagnosis, as told by Christopher:

Christopher: [when I realised that I needed my ACP] At the conference. [I met the support group for] some of the…some people with dementia and…few of them [PwD] are dead and put everything in place. And that’s when. it made me realize it [death] could…err…happen at any time. You never know your minutes. So…get it [ACP] in place now because it’s going to save a lot of heartaches…and confusion…and stuff […].

Interviewer: was it your wife that you discussed your plan with?.

Christopher: Me wife and…[think] meself, me son and daughter. We sat down and discussed… [searching for words] it with them.[…] It’s already been in place [ACP]. I’ve got one for me health, and I’ve got the errr do not resuscitate.[DNACPR] in place. We did that about after a few years [after my diagnosis]. –Christopher (68), living with Alzheimer’s for 11 years.

Christopher retold the key moment of realisation happened to him at a conference years after his diagnosis when the death of other PwD were the main trigger that enabled him to become aware of the benefits of ACP (‘put everything in place’). As such, to ‘prevent the heartaches and confusion’ within his family, he decided to discuss ACP together with others.

These extracts suggest various events that acted as ACP trigger points that triggered PwD to conceptualise, initiate or revise their ACP across the disease trajectory. As such, it is highly unlikely to locate a generic, single ideal ‘window of opportunity’ period for PwD to discuss their ACP. The ACP trigger points from this study are summarised in Figs. 2 and 3, illustrating when participants had discussed their ACP throughout dementia trajectory and the situations around their trigger points (see Figs. 2 and 3).

The ACP trigger points throughout the dementia trajectory

Reasons and situations around the ACP trigger points