Abstract
Background
What it means to live with dementia is changing. Autonomy, independence and continued community involvement are now recognised goals. As a result, new initiatives are required to support people with dementia to sustain their quality of life, update community understanding and reflect resultant change in community attitudes. Measuring the impact of such initiatives can help determine the extent of their success, inform needs for further intervention and, ultimately, shape policy.
Objective
To discover – or, if this could not be achieved, develop – a questionnaire measuring community attitudes towards people with dementia reflecting these newly recognised goals in terms of both content and expression.
Methods
A four-stage approach was used in this research: a scoping review of the literature, questionnaire development and expert review, questionnaire piloting, including with people living with dementia and their families, and preliminary psychometric testing.
Results
The review failed to retrieve a suitable existing questionnaire. A ten-item questionnaire, the Dementia Community Attitudes Questionnaire was developed, content validity was established by expert review and piloting led to refinements. Exploratory factor analysis (N = 92) generated an interpretable three-factor solution. Cronbach’s alpha coefficient for Factor 1, Engagement, was good (0.855); for Factor 2, Challenges, and Factor 3, Decision-Making, reliability was acceptable (0.785 and 0.709, respectively).
Conclusion
Supporting people with dementia to sustain their quality of life requires new initiatives and suitable measures to evaluate their impact. The Dementia Community Attitudes Questionnaire was developed with input from people with dementia, their families, and relevant experts. Items reflect current opportunities for people with dementia to retain their independence, autonomy and community engagement for as long as possible. Following further psychometric testing, this new questionnaire may be useful to evaluate such initiatives.
Introduction and background
Dementia is a global public health priority. In 2015, there were approximately 47 million people diagnosed with the condition worldwide. Estimates suggest that this number will increase to 75 million by the year 2030 (World Health Organisation, 2017). The World Health Organisation (2017) continues to promote the need for further dementia awareness raising initiatives to help overcome dementia-related stigma. People’s attitudes influence the ways in which they interact with those people living with the condition. These attitudes are based upon their knowledge of dementia, their observation of people living with dementia and how people with dementia are treated by others (Gerritsen et al., 2018). In turn, the way an individual experiences dementia is partly determined by their social environment and the attitudes of those people with whom they associate (Cheston et al., 2016).
People diagnosed with dementia are required to manage both their symptoms and the associated stigma attached to the diagnosis. Historically, dementia has been seen from a biomedical viewpoint, that is as a degenerative disorder resulting in a decline in a person’s functional capacity (Edvardsson et al., 2008). Emphasising a person with dementia’s loss of ability and increased dependency can influence how they are viewed by others, for example as being deemed unworthy and childlike (Edvardsson et al., 2008; Gerritsen et al., 2018). People with dementia also report experiencing discriminatory behaviours, such as avoidance and segregation (Milne, 2010). Negative language is often used when talking about dementia and during media representations (Brookes et al., 2018; Van Gorp & Vercruysse, 2012). The stigma associated with ageing and mental illness also compounds such negative views (Gerritsen et al., 2018; Kitwood, 1997; Phillipson et al., 2014).
Globally, over recent decades, there has been a shift in people’s understanding of dementia, driven in part, by people diagnosed with dementia speaking publicly about their experiences. These consumer advocates continue to speak for all people with dementia to be included in decisions that affect their lives (Bryden, 2004, 2015; Rohra, 2014; Swafer, 2016; Taylor, 2007). Assisting the autonomy of people with dementia by promoting their inclusion in decision-making has been reflected in consumer directed care policies worldwide, for example in Norway (Norwegian Ministry of Health and Care Services, 2015), the United Kingdom (United Kingdom Department of Health, 2013), Australia (Australian Department of Health and Ageing, 2012) and New Zealand (New Zealand Ministry of Health, 2013). Such consumer-focused policy is indicative of attempts made to reverse dementia-related stigma and discrimination. Furthermore, the World Health Organisation (2017) has endorsed a human rights–based approach for people with dementia and the move to view dementia as a disability (UN General Assembly, 2007). A human rights–based approach aims to safeguard rights for independence, social protection and an adequate standard of living for people with dementia and their care partners, providing new meaning to living with the condition (World Health Organisation, 2015, 2017).
People with dementia have reported wanting to maintain their independence, stay connected to their pre-diagnosis lifestyle and remain involved within their community (Read et al., 2016). This continued involvement becomes more feasible when the community is dementia-friendly. A dementia-friendly community is one that promotes the inclusion of people with dementia, empowers them to live the life they want and adapts the social and physical environment so that it is more easily accessible by people with dementia (Heward et al., 2017; Kaufmann & Engel, 2016; World Health Organisation, 2017). Alzheimer’s Disease International (2015) outlined the principles of dementia-friendly communities and stated that for people with dementia to remain included within their community local businesses need to consider the needs of people living with the condition. Central to the successful development of such communities is the participation of people with dementia (Buckner et al., 2019; Woodward et al., 2018), but to evaluate the effectiveness of the communities, further research is required (Buckner et al., 2019).
The changing context of living with dementia to one where people are offered post-diagnostic support that includes opportunities to live well, remain engaged and independent, requires initiatives to address broader (community) understandings about what it means to live well with dementia and a shift in societal attitudes to create sustainable dementia-friendly communities (McParland et al., 2017). An attitude is a person’s cognitive, behavioural and affective response to an object, person or situation, reflecting what they know; their beliefs and perceptions and how they respond (either positively or negatively), both physically and emotionally. These three components of attitude form the tripartite model or three-component view of attitude (Breckler, 1984; Rosenberg, 1960). Using questionnaires to measure attitude change can help guide the need for further interventions, but the relevance of these questionnaires to the current context is critical (Denscombe, 2003). In addition, the need to avoid language or terminology that is inadvertently offensive or stigmatising is critical, not least because people with dementia themselves, as valued members of community, may be asked to complete these measures (Denscombe, 2003).
A range of existing questionnaires evaluates people’s attitudes towards dementia and/or related care approaches (Aihara et al., 2016; Cheston et al., 2016; Kinney et al., 2017; Phillipson et al., 2014; Tan et al., 2012), including those of healthcare workers (Mason et al., 2016; Staples & Killian, 2012; Travers et al., 2013; Wang et al., 2018) and students (Kimzey et al., 2016; Lundquist & Ready, 2008). However, as we move into a new era of supporting people with dementia, ensuring that such questionnaires evaluate attitudes within this new context is required. The purpose of this study was to discover an existing questionnaire that had relevance in 2020 or – if this could not be identified – develop a questionnaire measuring community attitudes towards people living with dementia that aligned with the new era of support and, in doing so, used terminology unlikely to cause offence to anyone experiencing this condition.
The article explains the four-stage approach used to develop the questionnaire. Stage 1, scoping review, examined existing questionnaires. When no suitable questionnaire was found, the following processes were followed: Stage 2, questionnaire development, including establishing content validity; Stage 3, piloting; and Stage 4, preliminary psychometric testing. Ethics approval was obtained from the Curtin University Human Research Ethics Committee.
Stage 1 scoping review
Aim
The aim of Stage 1 was to review existing questionnaires used to measure community attitudes towards people with dementia.
Design
A scoping review of the literature was undertaken, given the authors were looking to: (a) discover specific elements of the topic and (b) establish the extent to which any appropriate measures existed in the published literature (Munn et al., 2018).
Methods
Inclusion and exclusion criteria
Articles eligible for inclusion in the scoping review were written in English and published in peer-reviewed journals between 1 January 1996 and 31 December 2019. The rationale for selecting this date range was based on work by Lintern (2001), who identified that there were no validated dementia-specific scales published before her work in 1996. Consistent with the process recommended in the COnsensus-based Standards for the selection of health Measurement Instruments procedure (Prinsen et al., 2018), only articles reporting scale development studies, specifically designed to establish the psychometric properties of questionnaires measuring people’s attitudes towards people living with dementia, were included. A person’s attitude to dementia encompasses their understanding of dementia and/or how they feel about and respond to people diagnosed with the condition (Breckler, 1984). Articles reporting studies that involved scale development designed to measure health professionals’ or health professional students’ attitudes toward people living with dementia were excluded. Furthermore, the authors chose not to consider literature that detailed scales of other stigmatising conditions, given the uniqueness of dementia-related stigma and the historical context that (a) surrounds the use of language that is demeaning and (b) the belief that dementia causes a person to experience a loss of self (Kitwood & Bredin, 1992).
Search strategy
Six electronic databases, CINAHL, Embase, MEDLINE, PsycINFO, Scopus and Web of Science were searched from 7 May 2019 until the 11 May 2019 to identify articles that met the inclusion criteria. A suitably experienced librarian was consulted to confirm appropriate use of search terms. The following search terms were used (dementia OR Alzheimer’s OR vascular dementia OR Lewy body OR frontotemporal) AND “scale development” AND (attitude OR knowledge).
The first and fourth authors identified which articles met the inclusion criteria. In the first instance, titles and abstracts of all articles retrieved were screened. Second, the complete text of each of the articles that met the inclusion criteria was screened. The authors screened these articles independently and then met to compare decisions. Any disagreements were resolved by returning to the article for further review and discussion.
Data extraction and review
Data were extracted and reviewed from each of the included articles to determine the questionnaire domains and how these were identified. A data collection form was created to record the following: article title, authors, study aims, questionnaire domains and domain development process. The two authors extracted data independently and met to discuss and resolve any differences by revisiting the article.
Consideration of the questionnaires for current use
In a process involving the first and fourth authors working independently, the questionnaires outlined in the six included articles were examined. There were three main considerations: Were the three components of attitude outlined in the tripartite model (Breckler, 1984; Rosenberg, 1960) (cognitive, behavioural and affective responses) addressed? Were items included that addressed the issues of the autonomy, independence and continued community involvement of a person with dementia? Were there any concerns that the language or expression used in the items of the questionnaire might be stigmatising or inconsistent with the current understanding of what it means to be living with dementia?
Results
The initial search retrieved 75 articles. Following removal of duplicates and review of each article’s title and abstract, 14 remained. Subsequently, a full text review of the 14 articles was undertaken with six articles identified as having met the inclusion criteria. Studies reported within the six articles were conducted in the United States of America (Carpenter et al., 2009; O’Connor & McFadden, 2010), Australia (Annear et al., 2015, 2017; Baker et al., 2018) and the United Kingdom (Griffiths et al., 2018). Figure 1 outlines the article selection process. Literature search flow – scoping review. *Please note that some articles were omitted during the screening process for more than one reason; therefore, the number of excluded articles does not equal the number of articles detailing their reason for exclusion.
There were five questionnaires described in the six articles: (1) the Alzheimer’s Disease Knowledge Scale (Carpenter et al., 2009), (2) the Dementia Attitudes Scale (O’Connor & McFadden, 2010), (3) the Dementia Knowledge Assessment Scale (Annear et al., 2015, 2017), (4) the Kids Insight into Dementia Survey (Baker et al., 2018) and (5) the Adolescent Attitudes towards Dementia Scale (Griffiths et al., 2018). None of the five existing questionnaires were considered suitable for current use. In particular, two questionnaires (Annear et al., 2015, 2017; Carpenter et al., 2009) only addressed knowledge of dementia, two were worded for use only with children or adolescents (Baker et al., 2018; Griffiths et al., 2018), and it was felt that the remaining one (O’Connor & McFadden, 2010) was not entirely consistent with the new era of support for people living with dementia. The reviewing team was concerned that consistent with the new era of support for people with dementia, items included in the questionnaire should be available for use within communities that include people living with dementia, those whose friends or relatives have been diagnosed with the condition and those who may be experiencing some symptoms (and therefore considering seeking a diagnosis). The reviewers were concerned that the item in the questionnaire developed by O’Connor and McFadden (2010) ‘I am comfortable touching people with ADRD’ was potentially offensive in this context. To a lesser extent, this was the case for the items: ‘It is possible to enjoy interacting with people with ADRD’ and ‘People with ADRD can feel when others are kind to them’. Use of the term ‘difficult behaviours’ was similarly of concern, with ‘responsive behaviours’ now widely recognised as more appropriate (Herron & Wrathall, 2018). However, a review of the established questionnaire domains provided the initial basis for further questionnaire development.
Stage 2 questionnaire development and initial review
Aim
The aim of Stage 2 was to develop a new questionnaire measuring community attitudes towards people with dementia.
Method
Content development
Relevant domains addressed in the reviewed questionnaires were considered in relation to the current context of living with dementia (as outlined in the background to the study) and the related use of language and terminology. The following domains were considered to be relevant to a new measure of community attitudes towards people living with dementia: knowledge and understanding (Baker et al., 2018; O’Connor & McFadden, 2010), personhood (Baker et al., 2018), treatment and management (Carpenter et al., 2009); care considerations (Annear et al., 2017), caregiving (Carpenter et al., 2009), stigma (Baker et al., 2018), communication and behaviour (Annear et al., 2017), life impact (Carpenter et al., 2009), social comfort (O’Connor & McFadden, 2010) and perception of dementia and empathy (Griffiths et al., 2018). Each of these domains was also identified as related to at least one of the three components of attitude (affect, behaviour and cognition) (Breckler, 1984; Rosenberg, 1960). Items addressing each of these domains were developed for inclusion in a draft questionnaire that: (a) was considered brief enough to be feasible for use when evaluating community-based initiatives, such as public forums; (b) addressed the issues of the autonomy, independence and continued community involvement of people living with dementia and (c) used language specifically intended to avoid stigma or the causing of offence to people living with dementia or their families. 12 items were included in the resulting draft questionnaire.
Response scale
A 10-item Likert-type response scale was used. While providing participants with this number of response options can mean completing the questionnaire is more time consuming, it provides participants with greater opportunities to accurately indicate their point of view and allows for enhanced statistical analysis with the potential for a greater spread of scores and greater detection of a change in scores over time (Preston & Colman, 2000).
Expert reference group
Five nursing or allied health Australian researchers experienced in scale development and with knowledge of dementia were invited by email to participate in an expert reference group to assist in the development of the questionnaire. An information sheet and consent form were also included. The participants each provided a signed consent form to the researchers. The expert reference group members were asked to comment on the newly developed questionnaire’s content validity and review each of the 12 questions for clarity and validity using an evaluation framework (Mastaglia et al., 2003). The expert reference group was also asked to comment on the appropriateness of scoring responses and whether the order of questions was suitable (Macnee, 2004; Mastaglia et al., 2003; Polit et al., 2001).
Results
Questionnaire item development.
aDenotes items reverse scored as statements designated as incorrect.
Stage 3 questionnaire piloting
Aim
The aim of Stage 3 was to establish the questionnaire’s face validity and ensure that the questionnaire items were easily understood by a range of people and acceptable to those living with dementia.
Method
Sampling
Ten members of the West Australian community, including two people with dementia, four care partners of people with dementia and four members of the community not directly related to someone with dementia, were invited to complete the draft questionnaire. Alzheimer’s WA, a locally based community organisation, assisted in the recruitment of people with dementia and care partners. Participants were asked to read an information sheet about the study, which detailed what they would be required to do, and sign a consent form to signify their agreement to participate.
Data collection: cognitive interview verbal probing technique
The researcher asked the participants to first complete the questionnaire and then, using cognitive interviewing (Willis, 1994, 2005), went through the items with each participant using the pilot questions. The cognitive interview verbal probing technique was used as it was more suitable and potentially less complicated for participants with dementia (Willis, 1994, 2005). Participants were asked to comment on whether the questionnaire items were easy to understand, if items measured what was intended in their development, if any items were leading (prompted a person to respond in a particular way) (Denscombe, 2003) or were difficult to answer and if response options were appropriate (Macnee, 2004; Polit et al., 2001). Items 5 and 6 were reported to have leading content, and Item 9 was reported to be difficult to answer. Feedback on the formatting of the questionnaire provided by one participant with dementia also led to changes. These changes enhanced the usability of the questionnaire for people with cognitive impairment by refining the wording and how items were spaced throughout the questionnaire.
Results
Questionnaire items included in the dementia community attitudes questionnaire after refinements following pilot testing.
aDenotes items reverse scored as statements designated as incorrect.
Final questionnaire items and selected relevant domains from existing questionnaires mapped against the three components of attitude.
Stage 4 preliminary psychometric testing
Aim
The aim of Stage 4 was to undertake preliminary psychometric testing of the DCAQ to help establish construct validity.
Method
Intervention
A facilitated dementia awareness raising public forum was held at an Australian University on 2 October 2019. The forum aimed to increase community awareness and understanding of living with dementia and, in this way, to have a positive impact upon attitudes towards people living with dementia. A description and evaluation of the forum is described elsewhere (paper under review).
Sampling
A convenience sample of participants was invited to attend the forum via email and social media.
Data collection
An information sheet and a copy of the questionnaire were provided to attendees upon their arrival at the forum. Using the newly developed 10-item questionnaire (paper and pencil format), data were collected from attendees prior to and following the forum. Attendees provided their consent to participate by completing the questionnaire.
Data analysis
Missing values were imputed for four participants (representing <1% of data) using the standard imputation procedure in the Statistical Package for Social Sciences (SPSS; IBM SPSS Statistics for Windows, version 25.0), which uses a regression-based approach. Exploratory factor analysis was conducted to identify constructs within the newly developed questionnaire. Pearson correlations were calculated from the data and items with a large number of low correlations |r| < 0.3 were excluded. Data were also screened for multicollinearity (i.e. high correlation between pairs of variables), with the determinant score (a check for multicollinearity) above 0.0001. Sampling adequacy (i.e. a test for determining whether the sample size is large enough) was assessed using the Kaiser–Meyer–Olkin test (KMO = 0.86, with KMO values >0.5 indicating a sufficient sample size), and Bartlett’s test of sphericity was statistically significant (p < 0.001) indicating the data had a patterned relationship. Sampling adequacy was assessed on the diagonal element of the anti-correlation matrix, with all values >0.5. Communalities for all items were >0.2. The number of factors to extract was based on the number of factors with eigenvalues >1 (indicating that the factor explains more variance than a single variable) and through examination of the scree plot (a plot of the eigenvalue of each factor). Both orthogonal (varimax) and oblique rotations were evaluated (i.e. constraining the final solution to uncorrelated factor structure or allowing correlation between the factors), with the final solution fitted using an oblique rotation using the ‘oblimin’ procedure in SPSS. Reliability (internal consistency) coefficients using Cronbach’s alpha were calculated for each of the extracted constructs.
Results
Questionnaire responses were from 92 out of a possible 112 people who attended the forum (82%). Of these 92 people, 75 were female and 17 were male and the age range was 22–82 years (M = 51.01, SD = 13.997).
Exploratory factor analysis
Correlation matrix for questionnaire items.
#Denotes questionnaire items that were reverse scored. *Denotes statistically significant correlations (one-tailed). * p < 0.05, ** p < 0.01.
Factor loading pattern matrix.
On review of item groups, Factor 1 was named Engagement and comprised items consistent with community engagement for people with dementia (recognition of the potential of people with dementia and of the role for the community to help ensure that this potential is reached). Factor 2 was titled Challenges and comprised items acknowledging that symptoms resulting from dementia require support and intervention. Factor 3 was termed Decision Making and comprised items relating to the desire of people with dementia to remain central to decision-making. Factor 3 contained fewer than three items but was retained as a factor because the pairwise correlations between the two items (Items 9 and 10) were greater than 0.5 (See Table 5). Reliability for each factor was acceptable (>0.7) to good (>0.8), with Cronbach’s alpha for Factor 1 = 0.855, for Factor 2 = 0.785 and for Factor 3 = 0.709, albeit Factor 3 only contained 2 items (Boateng et al., 2018).
Discussion
This article describes the early development of the Dementia Community Attitudes Questionnaire to measure community attitudes towards people living with dementia. The questionnaire is sufficiently brief to be useful when evaluating community-based initiatives and aligns with the current and evolving paradigm of living with dementia. Consistent with this paradigm, the questionnaire also uses language deemed acceptable to people with the condition. The work followed a review of the literature that identified five dementia attitude or knowledge questionnaires, all of which encompassed some items deemed unsuitable for current use with adults in the general population to measure attitudes towards people with dementia. However, the work underpinning the existing questionnaires provided established domains that were relevant and valuable to the process of developing a new measure. The consultation that occurred with members of the expert reference group during questionnaire development and piloting by community members means that the new questionnaire is contemporary and suitable for use within the current context of living with dementia. Preliminary psychometric testing of the questionnaire has identified three factors explaining 75% of the variance. The first factor was titled ‘Engagement’, Factor two ‘Challenges’ and Factor three, ‘Decision-making’. The items in the questionnaire were also consistent with the constructs of attitude (i.e. cognition, behaviour and affect) described in the tripartite framework (Breckler, 1984; Rosenberg, 1960).
Evidence suggests that adults in the general population tend to believe that only limited support can be offered to people with dementia once they are diagnosed and that doctors have been less inclined to diagnose dementia for this reason (Moore & Cahill, 2013). Policy makers and people working within the education sector who implement use of the questionnaire may develop a greater understanding of the extent to which this phenomenon (otherwise known as nihilism) exists (Pentzek et al., 2009) within the community, by noting participant responses to questionnaire items that suggest it is possible for a range of support to be offered to people once they are diagnosed. Broadening people’s understanding about this dementia-related issue impacts on people’s attitude to dementia and addresses the cognitive construct of attitude (Breckler, 1984).
In the longer term, the new questionnaire may be used as a tool to measure the impact of dementia-friendly communities on attitudes towards people living with dementia. An entire societal response is required to create dementia-friendly communities. Members of the community need to be aware that despite a dementia diagnosis, people diagnosed often retain functional capacity that is sufficient to enable their ongoing community participation, with support, if this is what they desire – at least for a lengthy period. Item 5 (I have the potential to improve the lives of people living with dementia) aims to measure people’s awareness of the personal contribution they can make (with support as required) to ensure that people with dementia are included within the community. Item 6 (people with dementia can contribute substantially to their community) directly reflects a change in the paradigm of how we support people with dementia, recognising their right and capacity to remain active participants within the community if that is their desire (Heward et al., 2017; World Health Organisation, 2017).
There are study limitations that require consideration. Replication of the factor analysis using a confirmatory factor analysis approach within a larger sample is required to test the factor structure of the questionnaire, given the low sample size of the study. Furthermore, establishing convergent validity against other measures that use the same (or similar) constructs to determine a positive correlation would provide additional confirmation (Streiner & Norman, 2008). While Item 1 (I have a good understanding of what it would be like to live with dementia) on the questionnaire was deemed to be important as it aligned with the pre-established domain of empathy, the item was removed from the scale, given that its scores had low correlation with remaining questionnaire items. Future development of the questionnaire may seek to explore the domain of empathy, given its relevance to providing effective person-centred support to people with dementia (Ratka, 2018). As the third factor, ‘Decision-making’ encompassed only two items, additional items addressing this construct would be considered for inclusion in the future, given its relevance to the current human rights–focused approach in the support of people living with dementia. As the final version of the DCAQ consists of nine items, there is scope to add items without making the questionnaire too lengthy.
People with dementia have expressed the desire for continued self-determination (de Boer et al., 2012; Dening et al., 2013; Karel et al., 2007). How we assist people with dementia to sustain their independence requires the creation of new initiatives and environments that support autonomy (Ryan & Deci, 2017). Ongoing questionnaire development and standardised measures are required to evaluate the impact of new initiatives (Baker et al., 2018; Griffiths et al., 2018; Silverstein & Sherman, 2010; Whitlatch et al., 2017). It is hoped that the DCAQ, following further psychometric testing and refinement, will assist in evaluating ongoing dementia awareness raising campaigns that aim to reduce dementia-related stigma and improve the attitude of people within the community towards people living with the condition.
Conclusion
The context of what it means to live with dementia is changing and, as such, a change in societal attitudes towards people with dementia is required. This study details the development of a Dementia Community Attitudes Questionnaire that is current, aligns with the new era of support for people living with dementia and aims to measure community attitudes to people living with the condition. Measuring the impact of awareness raising initiatives can help to guide what further information is required to continue to strive for attitude change and ensure that people with dementia are included within our communities.
Footnotes
Acknowledgements
We extend our thanks to members of the Expert Reference Group who kindly provided their time to review and assist in the development of the questionnaire. Associate Professor Anne Whitworth, Curtin University; Dr Claire Eccleston, University of Tasmania; Professor Lily Xiao, Flinders University; Dr Pam Nichols, Curtin University and Professor Yun-Hee Jeon, The University of Sydney. Similarly, our thanks also extend to those people within the West Australian community who kindly provided their time to assist in the piloting of the questionnaire and Diana Blackwood, faculty librarian for the School of Nursing, Midwifery and Paramedicine at Curtin University for her assistance throughout the study.
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.
