Step 1: Panel formation

Participants were recruited by sending advertising flyers to Australian and international dementia organisations, support groups and research centres.

Step 2: Literature search and survey development

In order to inform the content of the initial survey, a systematic search of the ‘grey’ and academic literature was conducted in February to March, and October to December 2013 to gather statements about how to help an older person who is developing cognitive impairment or has dementia, or delirium. The search was conducted using Google Australia, Google UK, Google USA, Google Books and Google Scholar. The key search terms were (elderly OR aged AND dementia AND help OR communicate OR care* OR respond OR support OR tips), (elderly OR aged AND delirium AND help OR communicate OR care* OR respond OR support OR tips), (elderly OR aged AND confusion OR disoriented OR attention OR memory OR mood OR alertness OR hallucination AND help OR communicate OR care* OR respond OR support OR tips), and (older person OR elderly OR aged AND Alzheimer*AND help OR communicate OR care* OR respond OR support OR tips), (seeing a doctor AND confused OR dementia OR Alzheimer AND when doctor), and (confused OR dementia OR Alzheimer AND discussing OR talking about AND diagnosis OR loss OR dying).

As per other similar Delphi studies (e.g. [27, 33, 36]), the first 50 websites, books and peer-reviewed journal articles for each search term were retrieved and reviewed for relevant information, after duplicates were excluded. The decision to only examine the first 50 websites, books and journal articles for each search term is based on previous Delphi studies that found that the quality of the resources declined rapidly after the first 50 [39]. Any links appearing on the websites were also reviewed. Websites, articles and books were excluded if they did not contain information about how a non-professional can support an older person who is developing cognitive impairment, or has dementia or delirium. The content of a total of 500 websites, 29 books and 7 journal articles was analysed to develop the Round 1 survey. Figure 1 summarises the results of the literature search.

A working group, consisting of staff from Mental Health First Aid Australia and the University of Melbourne, translated the relevant information from the literature search into helping statements that were clear, actionable, and contained only one idea. The statements were used to form the Round 1 survey that was administered to the expert panels via SurveyMonkey. The panel members were asked to rate each of the helping statements, using a 5-point scale (‘essential’, ‘important’, ‘don’t know/depends’, ‘unimportant’ or ‘should not be included’), according to whether or not they thought the statement should be included in the guidelines. See Additional file 1 for a copy of the surveys.

Step 3: Data collection and analysis

Data were collected in three survey rounds administered between October 2013 and July 2015. In Round 1, panel members completed the survey developed using the literature search, and also had the opportunity to provide qualitative data in the form of comments or suggestions for new helping statements. The qualitative data was analysed using qualitative analysis software (TAMS Analyzer).

The participants’ comments were thematically analysed and the working group created new items to cover helping ideas that were not included in the first survey. This new content was translated into clear and actionable statements for the Round 2 survey.

Panel members were given a summary report of Round 1 that included a list of the items that were endorsed and rejected, as well as the items that were to be re-rated in the next round. The report included the panel percentages of each rating, as well as the panel member’s individual scores for each item to be re-rated. This allowed the participants to compare their ratings with each expert panel’s consensus rating and consider whether to maintain or change their answer when re-rating an item.

The procedures for Rounds 2 and 3 were the same as described above with several exceptions. Round 2 included new items from the Round 1 comments, as well as items to be re-rated, whereas Round 3 only included the re-rated items from Round 2. There was no opportunity for comments in Round 2 or Round 3, and if a re-rated item did not receive an ‘essential’ or ‘important’ rating by 80 % or more of each panel, it was rejected. Round 3 consisted of any new items in Round 2 that needed to be re-rated, according to the above criteria.

Although not originally planned, Round 3 also included one new item that was derived from a newly published book about challenging sexual behaviours. The original literature search did not reveal many items about this topic and the working group thought that, if endorsed by the panels, it would make for a more comprehensive set of guidelines. This item received endorsement in Round 3, and therefore did not require re-rating in another survey round.

Step 4: Guidelines development

All of the endorsed statements were written, by the first author, into the guidelines document. This was done by grouping similar items and re-writing them into continuous prose for ease of reading. Where possible, statements were combined and repetition was deleted. Original wording of the items was retained as much as possible. Some items were given examples and explanatory notes to clarify the advice. The working group reviewed this draft to ensure that the structure and the language were appropriate for the target audience of the guidelines. The draft guidelines were then given to panel members for final comment, feedback and endorsement.

Participants

Item ratings

A total of 656 items were rated over the three rounds to yield a total of 389 endorsed items and 267 rejected items (see Additional file 2 for a list of the endorsed and rejected items). Figure 2 presents the information about the total number of items rated, endorsed and rejected over the three rounds. The endorsed items formed the basis for the guidelines.

Differences between expert panels

There was a strong positive correlation between the two panels in the percentage endorsement for whether items should be included in the guidelines (r = 0.81). However, there were also some areas of disagreement. Items that were rejected by one panel but endorsed by the other, and that received notably higher or lower rating (±10 %), are noted below. A 10 % difference was chosen, as this was used in previous studies [36, 40, 41].

Items rejected by the professional panel by ±10 %

Items rejected by the carer advocate panel by ±10 %

Guidelines development

Approaching and talking to the person

These guidelines include advice on how to approach and talk to a person about a decline in cognitive functioning, including what to do if the person is reluctant to admit to cognitive problems or seek professional help. Often it will be family, friends or neighbours who first notice a person’s decline in cognitive function [17]. Mental Health First aid training based on guidelines developed in a similar way has been shown to increase course participants’ ability to recognise mental health problems in a vignette and to reduce their stigmatising attitudes [24, 41]. It is hoped that training based on these guidelines will help course participants more readily recognise the signs of, and reduce the stigma associated with dementia, promoting earlier help seeking.

Challenging behaviours

This research developed a list of strategies that a helper can use when they encounter challenging behaviours in the person with dementia. They include strategies for managing resistance, arguments, agitation, anger, aggression, inappropriate sexual behaviour, delusions and hallucinations. Challenging behaviours have a pervasive negative effect on carers, as compared to other challenges experienced, e.g. assisting with daily needs [43]. Research suggests that interventions that target carer knowledge and skills, such as these guidelines, are one important way to reduce carer burden and the resultant negative health effects on both the carer and the person with dementia [14, 44].

Decision making

Making decisions on behalf of a loved one has been identified as an area of significant distress for carers of people with dementia [4] and these guidelines include advice on discussing and making decisions about sensitive topics, such as moving to a higher level of care, planning for the future and driving cessation. The decision to change to a higher level of care can be emotionally difficult for carers [45], and decision aids, similar to these guidelines, may be helpful. Although the research into the usefulness of decision aids is limited, one study reported that carers found them helpful and that they reduced subjective carer burden [46].

The results of this study indicate that while helpers are an important supportive factor in driving safety and cessation, it is the medical professional’s responsibility to make the decision for when the person should stop driving. One important way that a helper may be involved in decisions around driving is to report unsafe driving to the health professional. This means that it is important that helpers have some guidance on discussing and making decisions about driving. Furthermore, this research recommends and gives guidance on how helpers can take a proactive role in planning for driving cessation, which may help make the transition to not driving easier for the person with dementia [47].

Crisis situations

Two crisis situations that are covered by these guidelines are wandering and delirium, both of which can lead to significant harm and even death if not addressed appropriately [48]. The majority of the literature on wandering is focused on assessing and managing wandering behaviour from the perspective of a care facility or a family carer (e.g. [49]). Our guidelines diverge from the literature in that they provide ‘first aid’ advice on what family and non-professional helpers or a member of the public can do if they encounter someone who is wandering. The advice includes how to recognise and approach a person who is wandering, ways to ascertain the person’s identity, and how and when to get emergency services involved.

Delirium commonly occurs in people with dementia, with a prevalence rate of between 22 % and 89 % in community and hospital populations [50]. The long-term outcomes of delirium in a person with dementia include increased cognitive and functional decline, nursing home placement and death [51, 52]. Given these poor outcomes, it is important that delirium is recognised and treated quickly. It is hoped that these guidelines will improve the ability of helpers to recognise, seek treatment for and assist in managing delirium, reducing the negative long-term consequences.

Limitations

A limitation of this study was the size of the carer advocate panel. A panel size of 23 has been found to yield stable results in a simulation study (Akins et al. 2005). While there were 24 experts in the carer advocate panel in Round 1, this reduced to 22 experts in subsequent rounds.

People with dementia were not invited to participate in this research—a notable limitation, but it was impossible to screen cognitive ability of participants to determine if they were able to meaningfully complete the surveys. Another limitation is the possibility that some panel members were asked to advise on statements that were beyond their expertise, possibly leading to a lack of inclusion of useful items. Furthermore, while participants were able to provide comments in Round 1 of the survey, they were not able to discuss their comments and opinions with others. Panel members may have held biases or made incorrect assumptions that were unchallenged because there was no opportunity for discussion. It is possible that key actions were omitted from the guidelines because of this.

Another possible limitation is the exclusion of items that would be helpful at certain stages of dementia, but were rejected because they received a ‘depends’ rating from the majority of participants. Furthermore, the majority of the panel members were female and mostly from Australia and New Zealand, which may have introduced bias. Finally, these guidelines were developed for English-speaking Western countries and further research is needed to adapt them for other cultures.

Implementation

Guidelines in themselves may not ensure change in supportive behaviours [37]. Therefore, these guidelines will be used to inform the contents of the Mental Health First Aid for the Older Person course. This course, due to be released later in 2016, will teach participants how to assist an older person who is developing cognitive impairment or has dementia, or delirium, as well as other mental health problems and crises. Although the course will initially be released for Australia, the training materials will also be made available to the organizations in over 20 countries that have adopted and adapted the Mental Health First Aid program from Australia.