Abstract
Aim
This review considers the theoretical literature concerning communication between a person with dementia and a person without dementia.
Methods
Theoretical accounts of communication between a person with dementia and a person without dementia were identified through database searches, hand searching of reference lists and contacting experts in the topic area.
Results
Twenty-one articles were included in the final review, yielding 18 different theoretical accounts of communication in the context of dementia. Thematic synthesis showed that four categories could account for the majority of the material described in these theories, even though they were derived from very different perspectives. These categories were mutuality, personhood, anxiety reduction and environmental considerations.
Discussion
The four categories identified in the thematic synthesis were reminiscent of Kitwood’s theory of person-centred dementia care. Different interpretations of this finding are considered. Most theoretical accounts focused on the relational aspects of communication. This meant that societal and individual accounts of communication in the context of dementia were relatively impoverished in terms of theory.
Conclusions
There is widespread agreement regarding the important components of good quality communication in the context of dementia. This suggests exciting opportunities for collaboration between apparently disparate bodies of literature. Individual and societal accounts of communication in the context of dementia are relatively neglected and would benefit from more research.
Introduction
Communication difficulties are common in people living with dementia (please see Schrauf & Muller, 2014; Watson, Aizawa, Savundranayagam, & Orange, 2012 for overviews of the topic). These arise from a combination of internal and external factors including the storage and retrieval of information and the ability to deal with the dynamic and complex nature of social interaction (Schrauf & Muller, 2014; Watson et al., 2012). Communication difficulties like these may limit opportunities for autonomy and for interactions for the person with dementia (Davis, 2005; Ekman, Wahlin, Viitanen, Norberg, & Winblad, 1994). This is reflected in the experience of people with dementia who report that their greatest unmet needs are often psychosocial in nature, such as the need for company and daytime activities (Hancock, Woods, Challis, & Orrell, 2006; van der Roest et al., 2009). Qualitative work has shown that communication difficulties are also amongst the most difficult aspects of caring for a family member with dementia (Murray, Schneider, Banerjee, & Mann, 1999), and quantitative work has demonstrated an association between communication and strain on the person caring for someone who has dementia (Papastavrou, Kalokerinou, Papacostas, Tsangari, & Sourtzi, 2007). Interventions which target communication therefore have the potential to improve the quality of life of both the person with dementia and the person(s) who care for them (see e.g. Eggenberger, Heimerl, & Bennett, 2013 for a review of interventions).
Communication is a complex phenomenon which can be understood on multiple levels (Luhmann, 1992). It happens on the individual level in the sense that a person gives off particular signals, either verbal or non-verbal. Another individual must then interpret these signals, and so it may be said that communication involves individuals and relationships. The individuals and the relationship also exist within a context, and so communication also constitutes and is shaped by the environmental, societal, political and historical. Each one of these aspects of communication may (rightly or wrongly) be partitioned, and each represents a way in which communication can be understood. The lens through which communication is viewed in this fundamental sense (i.e. as a neurobiological, linguistic, psychological, societal or political event, etc.) influences how communication is thought about and theoretically framed in a contested paradigmatic arena (Suppes, 1967). Theories arising from different perspectives may be said to be commensurate to the extent that they can establish a common language with which to share ideas (Kuhn, 1962). On the other hand, theories may be incommensurable if they are unable to do so due to irreconcilable ontological and epistemological assumptions.
The current paper outlines the findings from a scoping review (Arksey & O’Malley, 2005; Levac, Colquhoun, & O’Brien, 2010) that examined the theoretical literature on communication between people with dementia and the people without dementia who are around them. Scoping reviews allow an overview of the different theories which have been used to explore a phenomenon and identify areas that warrant further investigation (Arksey & O’Malley, 2005; Booth & Carroll, 2015; Levac et al., 2010). Reviews of this kind also allow researchers to identify points of agreement between separate yet commensurate bodies of knowledge, thus presenting opportunities for multidisciplinary collaboration.
Methods
Research question
For this scoping review, the research question was ‘What theories are available in understanding communication between people with dementia and people without dementia?’ There are four components to this question: ‘theory’, ‘communication’, ‘people with dementia’ and ‘people without dementia’. Following the recommendations of Levac et al. (2010) we sought to define the sub-components of the research question as clearly as we could, given that this had ramifications for the search strategy. Concepts that were deemed to be tangential to the central question such as mathematical theories and non-verbal forms of communication such as written correspondence were excluded from the scoping review. This led to the development of the following definitions:
Theory: models, frameworks or arrangements of concepts identified by the authors in explaining communication. Communication: the moment-by-moment exchange of thoughts, feelings, needs or other information between two or more people through speaking or non-verbal methods such as sign language, sounds, gestures, movements or expressions. People with dementia: people who have been diagnosed with probable dementia.
1
This can include any of the many types of dementia, for example Alzheimer’s disease, frontotemporal dementia, Lewy body dementia, and so on. People without dementia: any person communicating with a person with dementia who does not themselves have a diagnosis of dementia. This could be in a personal or professional capacity. The communication may be for social or practical purposes.
Database identification
Databases were chosen for their breadth and variety of coverage. Medline was chosen for coverage of medical literature. Psychinfo was chosen for its coverage of psychological literature. ERIC was chosen for its coverage of education literature and CINAHL for its coverage of nursing and allied health professional literature such as speech and language therapy. The database search was completed on 24 August 2016 and included all articles published up to this date. We did not specify a lower bound for date of publication.
Development of search terms
Search terms were developed through team discussion of synonyms for the components of the research question. These were then mapped onto the corresponding subject headings for each database. Following Booth and Carroll’s (2015) guidance on the systematic search of theory, the term ‘model*, concept* OR theor*’ was included as a title or abstract search term, so that the search returned only those articles which described a theoretical account. The search terms used can be found in Table 1.
Search terms included in the systematic review. Please note that the specific combinations of terms depended on the subject headings used by each database.
Identifying relevant studies
Screening
Following electronic database search, the titles and abstracts of identified studies were screened. Studies were excluded by the following criteria:
No mention of communication, language or conversation No mention of dementia or Alzheimer’s disease No mention of caring professional or family member.
Unfortunately we also had to exclude studies which were not available in English due to translation costs.
b. Full text review
Articles were included if they described an explicit theory, model or framework for understanding communication between a person with dementia and a person without dementia.
This included:
Articles setting out a theory based on existing literature Articles setting out a theory based on qualitative work/clinical observations Applications of existing theory to a novel area.
And it excluded:
Theories applied non-specifically to people with dementia (e.g. theories of communication with older people which may include those with dementia but is not specifically applied in this way) Articles which described communication between more than two people simultaneously Articles which described the unilateral transmission of information. c. Hand search
Reference lists of papers retained at this stage (n = 21) were hand searched for studies not identified by the database search. The full text of each paper identified in this way was then reviewed as per the above criteria.
d. Contacting experts
The list of identified papers was circulated to experts in the area of dementia and communication to ask if they knew of other relevant studies which may not have been returned by the database search. Academic experts were selected on the basis that they had more than citation identified in the initial database search and an active email address. We contacted 13 people in this way and received seven responses. The academics who responded had backgrounds in linguistics, psychology, nursing, cognitive science and social work. The full text of every article identified in this way was reviewed according to the criteria set out above.
Data extraction
A table was created to record the relevant data from each retained paper. This included metadata such as study name and country of origin, as well as a description of the theory or model itself. These data are reproduced in Tables 2 and 3 of the ‘Results’ section.
Metadata for the studies included in the final synthesis.
Studies are listed alphabetically by first author surname. The ‘background’ column refers to the university departments affiliated with the author(s). The superscript numbers correspond to the theories described in Table 3.
Theories identified from the 21 articles included in the review.
The superscript numbers in the first column correspond to the superscript numbers found in Table 2. The “Perspective” column represents the perspective identified by the authors of each article. They are not directly comparable because authors did not necessarily identify with perspectives which are aspects of the same dimension. The decision was taken not to abstract these to a level on which they could be compared as this would involve some interpretation of the meaning that each author ascribed to the perspective with which they identified. This meaning could be highly idiosyncratic (e.g. the meaning of being ‘Relational’ or ‘Psychoanalytic’), and so to abstract from these interpretations could easily lead to misrepresentation of the authors’ views.
Thematic synthesis
Our approach to synthesising the theories identified in our review was informed by Pound and Campbell’s (2015) guidance on the synthesis of theory. To begin with, the theories were succinctly summarised with a brief narrative description. Themes were then identified from the theories. This was an attempt to succinctly represent the main concepts of each theory. In many cases this was aided by the authors themselves who often summarised the key components of their theory; e.g. ‘The main concepts of the model are clarity of cues, sensitivity, interpretation, responsiveness and synchrony’ (Athlin & Norberg, 1987, p. 1). If the main concepts were not explicitly stated by the authors, themes were derived from the narrative description of the theory. These themes were recorded in a table and then categorised according to their similarity. These categories were assigned a label which characterised all the themes within that category. The two largest categories were recorded in diagrams whilst the other categories were recorded in a table. These can be found in Figures 2 and 3 and Table 4 of the ‘Results’ section.
Other categories arising from the synthesis of themes.
Team debriefing
The lead author (LE) read the articles included in this review in totality. He was supported by the other authors via a series of team meetings held throughout the study. These meetings involved the discussion of interpretations, clarification of meanings and tracing interpretations back to the original texts. This process helped to identify the biases of the lead analyst (see ‘Reflexivity’ section) and also to ensure the coherence of interpretations.
Results
Reliability analysis
A second researcher was asked to screen 25% (166) of the studies identified by the database search in order to assess reliability of the screening criteria. Cohen’s kappa was calculated as .71 indicating adequate reliability.
Retained studies
Please see Figure 1 for a flow chart of identified studies for inclusion in this review. In total 706 studies were considered for inclusion. Twenty-one studies met the criteria for inclusion in the final synthesis. Table 2 describes the metadata for each of the included studies.

Flow chart for studies identified in this review.
Eighteen theories were derived from the 21 included studies. Information about the theories described in these papers can be found in Table 3.
Thematic synthesis
The largest category which emerged from the synthesis was labelled mutuality. Mutuality signifies the understanding shared by two people. Themes relating to mutuality were identified in sixteen of the articles included in the synthesis. Mutuality consisted of three subcategories: temporal attunement, cognitive attunement and emotional attunement. These subcategories and the means by which they might be achieved are presented in Figure 2.

The ways in which mutuality may be achieved have not been separated according to emotional, temporal or cognitive attunement because each process or behaviour may rely on all three aspects simultaneously. For example, shared enjoyment implies some synchrony between conversational partners in each of the emotional, temporal and cognitive subcategories.
The next largest category was labelled personhood. Themes relating to personhood were identified in twelve of the 21 articles included in the review. A diagram of the themes relating to personhood can be found in Figure 3.

Personhood was comprised of three subcategories. These were autonomy, identity and respect. Autonomy refers to the ability to initiate actions oneself, including conversation. Identity refers to the things which other people come to associate with being a particular person, such as maintaining hobbies and interests. Respect refers to the regard shown towards a valued person.
Five other categories also emerged from the synthesis of themes. These are presented in Table 4.
The largest of the other categories was anxiety reduction. Eight articles mentioned themes relating to this category. Themes relating to environmental considerations were mentioned in five articles. The three other categories were each mentioned by only one article.
Discussion
Summary
The current paper provides a systematic scoping review of theories of communication between a person with dementia and a person without dementia. The review identified 21 studies which met the inclusion criteria. Eighteen different theories were derived from these studies.
Thematic synthesis suggested that the four largest categories could account for the majority of themes derived from the included theories. These categories were mutuality, personhood, anxiety reduction and environmental considerations. These categories were often shared between theories, even though the theories originated from diverse perspectives such as social constructivism, linguistics and psychoanalytic theory.
Reflexivity
Given that the synthesis of the papers included in this review required some interpretation of the literature, some reflection on the position of the lead autor (LE) is necessary prior to discussing the results in detail. We hope this will allow the reader to contextualise the findings so they are able to make their own inferences as to their validity.
LE is a trainee clinical psychologist whose practice is mainly derived from the humanist perspective. His clinical work emphasises non-judgement, empathy and curiosity. The research presented here was completed as part of the requirements of his training. It has been written alongside a piece of research investigating the lived experience of communicating with a loved one who is living with dementia .
Study implications
a. Similarity with definitions of care
The four largest categories of themes we identified in this study were mutuality, personhood, anxiety reduction and environmental considerations. The most striking thing to note about these categories in context is their similarity to the concepts described in Kitwood’s theory of person-centred dementia care (Kitwood, 1993, 1998; Kitwood & Bredin, 1992).
Personhood and mutuality were central themes in Kitwood’s work. He believed that personhood was created in an infant through intersubjectivity/mutuality with their caregiver (Kitwood, 1993). This involved the caregiver responding appropriately to the cues of the infant, thus communicating to the infant that they were agentic in the world (Kitwood, 1993). Anxiety was also an important theme in Kitwood’s theory. Kitwood believed that infants lacked sophisticated psychological defences and were thus prone to anxiety. As a result, their emotions must be ‘held’ by external caregivers in intersubjective states (Kitwood, 1993). Kitwood (1993) also noted that care environments could be more or less facilitative of personhood-enhancing interactions. Kitwood transposed these ideas – taken from psychoanalytic developmental psychology – onto the care of people with dementia. Hence, positive environments were thought to facilitate mutual interactions which had the dual function of fostering personhood and offsetting the anxiety experienced due to reduced defences.
Clearly, there are very strong similarities between Kitwood’s ideas and the categories we identified. This implies that the same kinds of themes are identified with communication as with care. Perhaps this suggests a rather artificial distinction between the two concepts or their inherent symbiosis: good communication can be a caring process and caring itself communicates. On the other hand, this may reflect the vast influence of Kitwood’s work in the field of dementia care. This is considered in more detail below.
b. Similarities between theories
In addition to their apparent similarity with definitions of care, the theories we included in this review often shared similarities with each other. This was despite coming from a range of distinct professional and theoretical bases. This finding could be interpreted in a number of ways.
(I) Structural/realist explanation One explanation is that the categories identified represent some underlying structure or truth which each theory describes with more or less success. The large degree of commensurability between the theories would by this account suggest a fairly good correspondence between the theory and the way the world actually is (i.e. the ‘truth’). It is possible that some of these theories access the truth at different levels. Some may describe the ‘what’ (e.g. the ‘maternal thinking’ recommended by Haggstrom and Norberg (1996)) whereas others describe the ‘how’ (such as Luxmoore & McEvoy’s description of mentalisation, 2017). Some may even begin to broach the ‘why’ (e.g. Seifert, Flaherty, & Trill, 2013). A pluralist account such as this is antireductionist in nature. As such it is comparable with other antireductionist positions such as Velmans’ reflexive monism (in the sense that good communication may consist of complementary and mutually irreducible components; Velmans, 2008) and the multiple realisations accounts of Putnam and Fodor (in the sense that the same experience may be brought about through different routes; Fodor, 1974; Putnam, 1967). Given the multifaceted nature of communication, it is likely that multifaceted interventions will have the greatest impact. Some evidence already attests to this: in their systematic review of communication interventions, Eggenberger et al. (2013) noted that educational interventions were improved when combined with other components. The general agreement between theories of communication identified in our research suggests exciting opportunities for collaboration in the development of complex interventions for communication in the context of dementia. It might be possible to integrate theories in order to produce interventions which address different aspects of communication. For example, complete interventions may consist of diverse modules including personal reflection, perspective taking, anxiety reduction techniques, and so on. Some may even consider incorporating political or spiritual elements.
(II) Societal explanation
It is possible that our findings demonstrate a societal trend to look for certain things, and indeed some tacit ontological agreement. Theorists may feel as though they are investigating the world through their own lens, yet a much more powerful lens is in common with other researchers. Researchers come up with similar ideas because these ideas are all that we have access to at this particular time. In effect, our findings are signalling a contemporaneous zeitgeist. This view is consistent with the work of Thomas Kuhn (1962) and would indicate a period of ‘normal science’; currently, it seems as though few credible alternatives to the current paradigm exist.
Assuming this perspective, and in light of the similarities with Kitwood’s work, one might infer that the current paradigm in dementia care research is one of ‘person-centred care’. The widespread adoption of this paradigm has interesting implications for the field. One of the obvious implications is the relative neglect of individual and societal perspectives (see below). The adoption of the person-centred care paradigm also gives primacy to anxiety at the expense of other emotions such as sadness and anger. This may reflect the psychoanalytic interests of Kitwood himself, as anxiety is so central in much psychoanalytic psychology (e.g. Freud, 1926/1977). This is an interesting point to consider, because it implies that theories which do not identify with psychoanalysis at all may in fact trace their roots back to the discipline.
(III) Bias
The results that we present here might be an example of bias on our part. Of course, some bias is inevitable in all research and we can only aim to declare this so that the reader is able to judge the validity of our findings themselves. We do hope that our transparency regarding the lead researcher’s position and study methods inspires confidence in our findings. Nevertheless, it would be irresponsible for us to ignore this potentially less interesting explanation of our results.
c. Theories situated outside the prevailing discourse
One of the fascinating findings of this research is the situation of two theories more or less entirely outside of the four largest categories. Clare and Shakespeare’s (2004) social constructionist theory of resistance gives primacy to the concepts of power and societal discourse. These are societal-level explanations of interpersonal communication. On the opposite end of the spectrum is Wong et al.’s (2009) theory which they term the communicative effectiveness framework. This theory provides a mostly individualised account of interpersonal communication.
These theories seem to fit neither with each other, nor with the prevailing theoretical discourse. This perhaps suggests that the current epoch is one in which researchers are most accepting of an explanation of communication in terms of the two communicative actors present during the interaction. Neither a narrower focus such as Wong et al.’s (2009) nor a broader focus such as Clare and Shakespeare’s (2004) seems particularly popular. Again, a realist perspective would suggest that this indicates some truth as to the interpretation of communication as an interpersonal (as opposed to individual or societal) process. On the other hand, it may suggest nothing more than that this interpretation is currently in vogue.
Whichever position is adopted, our review shows that societal- and individual-level theories of communication are presently scant. The lack of theory in these areas will limit the availability of evidence-based interventions at each of these levels. This is reflected in two systematic reviews of communication interventions, both of which only identified interventions which were focused on communication at the interpersonal rather than the individual or societal level (Eggenberger et al., 2013; Vasse, Vernooij-Dassen, Spijker, Rikkert, & Koopmans, 2010). Without viable theory, there can be little understanding of the mechanisms upon which interventions are based (Booth & Carroll, 2015; Craig et al., 2008).
Limitations
This study cannot offer anything in the way of prevalence, utility or evidence for each of the theories here identified. We can simply say that they exist. This means that the applied utility of these findings, e.g. for healthcare commissioners or clinicians, may be limited. Our findings may be of greater interest to researchers attempting to understand dementia and communication. Perhaps the results presented here say something more generally about the nature of theory in this area and present directions for future research and collaboration.
The intention of this study was to identify theories of communication in a specific context. This may have meant the exclusion of some more general theories of communication which could be relevant, but which have not been directly applied to people with dementia. In connection with this point is the implication in our research question that communicating with a person with dementia is likely to be different to communicating with a person who does not have dementia. Although many of the studies identified in this review attest to the communicative changes that dementia brings about, certain others (notably Clare & Shakespeare, 2004) might suggest that studies such as ours construct or partially construct the differences we seek to understand. It is possible then that the phenomenon itself may be a partial product of the tools with which we investigate.
Given the systematic nature of this investigation, bodies of knowledge which fall outside of the scientific or quasi-scientific domain were excluded. A more complete study could include anthropologists, historians and theologians on the research team whose expertise could enrich the findings along other dimensions.
Finally, the synthesis of the theories included in this review involved a certain amount of interpretation. It is possible that our interpretations will differ from those of other researchers, and we recognise that our review will have been biased by the terms of reference and search strategy we employed. Having said that, theory synthesis is an interpretive endeavour (Pound & Campbell, 2015), and thus non-agreement between researchers would not necessarily discredit either synthesis (and by extension, agreement would not necessarily strengthen the synthesis either).
Future directions
Given the apparent similarities between theories which are purportedly distinct, researchers in the area of dementia and communication may find it beneficial to work together. Clearly there are pockets of knowledge distributed throughout a number of different professions and schools. It would certainly seem more efficient to collaborate in the pursuit of understanding, rather than to each pursue the same goal separately – especially considering that the destination often seems to end up the same. This could also help to enrich existing interventions along other dimensions with a view to increasing their effectiveness.
Next, our review has shown that there are relatively few individual- and societal-level theories of communication. More work must be done to populate these areas in order to develop corresponding interventions based on well-evidenced theoretical principles. Without the development of such theory, interventions that focus on the individual or society are likely to be based on poorly evidenced mechanisms or neglected entirely. Individual and societal components to multilayered interventions are also likely to be lacking without the development of such theory.
Finally, in entertaining the societal explanation for our findings, it would be very interesting to discover cross-cultural interpretations of communication and care in the context of dementia. Notions of personhood, for example, might conceivably be embedded in the high value placed on individuality in countries like the UK and USA. This kind of research would be valuable in assessing the universality of concepts such as personhood and mutuality, as well as potentially indicating new avenues for communication and care research in the context of dementia.
Final conclusions
The research we have presented here suggests that theories of communication in the context of dementia share many points of agreement, even though they originate from different perspectives. This presents opportunities for collaboration in the development of multicomponent interventions. Most current theory focuses on communication as an interpersonal process. Societal- and individual-level explanations are currently relatively impoverished in terms of theory and may therefore benefit from further investigation into the basic mechanisms which underlie this complex phenomenon.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
