Setting matters: Associations of nurses’ attitudes towards people with dementia

Abstract Aim Identify associations between attitudes of nurses towards people with dementia and their socio‐demographic (age and gender) and work‐related characteristics (employment, work experience, nursing degree, care setting). Design The study was designed in a cross‐sectional way, collecting self‐reported questionnaire data. Methods Nurses (N = 417) completed the 20‐item Dementia Attitude Scale questionnaire, including socio‐demographic and work characteristics. Data were analysed using descriptive statistics, group mean comparison, correlation and regression analysis. Results The attitudes of nurses towards people with dementia were only related to the care setting. Attitudes of nurses working in dementia‐specialized long‐term care institutions were significantly more positive compared with those working in mixed long‐term care institutions and in the homecare setting. The other variables, such as age, gender, employment, nursing degree and work experience, were not associated with the attitudes.

people with dementia. The challenge of integrated facilities is that nurses must cope with the needs of people with dementia and with the needs of people with other diagnoses, while in specialized long-term care, nurses can focus exclusively on the needs of people with dementia. In integrated facilities, 64.5% of the residents are diagnosed or suspected to suffer from dementia (Schweizerische Alzheimervereinigung, 2014b).

| BACKG ROU N D
Previous studies already indicated that well-being and quality of life of the persons suffering from dementia depend on the nursing staff involved (Kada, Nygaard, Mukesh, & Geitung, 2009). Quality of nursing is known to be highly related to the knowledge about and the attitudes towards dementia held by professionals (Norbergh, Helin, Dahl, Hellzén, & Asplund, 2006). This connection has already been intensely studied by social psychologists and has been considered as essential for the quality of nursing (Åström, Nilsson, Norberg, Sandman, & Winblad, 1991;Fessey, 2007;Stahlberg & Frey, 1992, Soderhamn, Lindencrona, & Gustavsson, 2001. Various studies showed that nursing home staff tends to have a more negative than positive perception of dementia patients (Asplund & Norberg, 1993;Brodaty, Draper, & Low, 2003).
Interestingly, however, the contradictory effects were also found (Kada et al., 2009;Moyle, Murfield, Griffiths, & Venturato, 2011;Spector & Orrell, 2006). Norbergh et al. (2006) for instance found that nurses had a positive to neutral perception of their dementia patients. In contrast to the studies conducted by Asplund and Norberg (1993) and Brodaty et al. (2003) where nurses had to rate fictitious patients, in the study conducted by Norbergh Helin Dahl Hellzén and Asplund (2006), the nurses had to rate real dementia patients. The reasons for the discrepancy in the results might be that the personal experiences and personal relationships and the upcoming adoption of the personhood approach (Kitwood, 1997(Kitwood, , 2004 in long-term care evoked more positive attitudes towards dementia patients. Researchers have further focused on the factors that influence nurses' attitudes towards people with dementia. Addressed were factors such as workplace characteristics, socio-demographic characteristics and dementia knowledge (Åström et al., 1991;Brodaty et al., 2003;Kada et al., 2009¸ Moyle et al., 2011Zimmerman et al., 2005).
Regarding socio-demographic characteristics, nurses aged 50+ and work experience of less than 10 years (Kada et al., 2009) were associated with a less positive attitude towards dementia. In contrast, Åström et al. (1991) found no differences in attitudes towards people with dementia with regard to the staff's age or time at present place of work. In the sample of Brodaty et al. (2003), age had no influence on nurses' attitudes towards dementia either.
Several studies found that Registered Nurses, who have better theoretical dementia knowledge, show more positive attitudes towards residents with dementia than nurses with lower levels of education (Åström et al., 1991;Mellor, Greenhill, & Chew, 2007;Kada et al., 2009). Furthermore, nurses with specialized training in geriatrics, psychiatry or dementia care are reported to have more favourable attitudes than nurses lacking this training (Kada et al., 2009;Zimmerman et al., 2005). Moyle et al. (2011) suggested that such training courses seek to move care staff away from the focus on deficits towards a greater appreciation of the person's abilities and therefore can positively influence attitudes and behaviour.

| AIM
In sum, nurses' attitudes towards people with dementia are of great importance. Nurses' attitudes influence their interactions with persons with dementia and subsequently the quality of the affected persons' life. Nevertheless, as results are contradictious, more research on the factors influencing these attitudes is needed. To change nurses' attitudes in a positive direction, knowledge of the influencing factors is needed. Therefore, the aim of this study was to examine the relations between various socio-demographic and workplace characteristics of nurses and attitudes towards people with dementia.

| Design
A cross-sectional research study was conducted involving a survey with a sample of Swiss nurses caring for people with dementia in home care and in long-term institutional care settings.

| Setting
Data were collected in the German-speaking part of Switzerland, which comprises approximately two-thirds of the Swiss population or 4.5 million people (https://www.bfs.admin.ch/bfs/de/home/ statistiken/bevoelkerung/sprachen-religionen/sprachen.assetdetail.2244919.html#; accessed 13 August 2018). There are nearly 1,300 institutions of long-term care in this part of Switzerland.
Either they are integrated facilities, which care, among others, for people with dementia, or they are special care units, which are exclusively for people with dementia. In integrated facilities, 64.5% of the residents are diagnosed or suspected to suffer from dementia (Schweizerische Alzheimervereinigung, 2014b).
Among 50% of people with dementia live at home and are cared for by informal or professional caregivers (Schweizerische Alzheimervereinigung, 2014b). The most widely used professional home care service in Switzerland is called Spitex. Spitex is organized locally with approximately 450 organizations in German-speaking part of Switzerland. The 55% of Spitex clients are diagnosed or suspected to suffer from dementia (Ecoplan, 2013).
Nursing staff in institutional and in the home care setting are put into three categories, according to their academic degree in nursing: nursing aides, licensed nurses, Registered Nurses.
Data were collected in two waves. The first wave was com-

| Procedures
Information about the study was presented to the nursing directors and managers of the institutions. The 50 institutions of long-term care and 18 Spitex organizations agreed to participate in the study.
The nursing directors of the respective institutions distributed either the paper version or the link to the online questionnaire to the nursing staff, consisting of Registered Nurses, licensed nurses and nursing aides. Participants could fill in the questionnaire during working time. The sole exclusion criterion was a lack of proficiency in the German language, which would keep participants from understanding and filling in the questionnaire. Each questionnaire was accompanied by an information letter about the study and, if required, a postage-paid return envelope. Participants were asked to complete the questionnaire in two weeks. After these two weeks, a reminder was sent to the nursing directors to increase response rate. Participants could participate for two more weeks. Participants did not receive monetary compensation for filling in the questionnaire. As we used a snowball sampling, the response rate cannot be determined.

| Attitudes
To assess nurse's attitudes towards older people with dementia, data were collected using the German version of the Dementia Attitudes Scale (DAS; Peng, Moor, & Schelling, 2011). The researchers selected this scale because of its empirically affirmed reliability (Cronbach's alpha for the total scale α = 0.87) and validity in a Swiss sample (Peng et al., 2011). The original English version was developed by O' Connor and McFadden (2010). The scale consists of 20 positively or negatively formulated items. The factors of the scale are labelled "dementia knowledge" (item example: "People with ADRD can enjoy life.") and "social comfort" (item example: "I feel confident around people with ADRD."). Participants are asked to give their degree of agreement on a 7-point Likert scale, ranging from 1 -7 (1 = strongly disagree, 7 = strongly agree). Higher scores reflect a more positive attitude. Maximum score is 140 points.

| Socio-demographic and work characteristics
The questionnaire included different socio-demographic and work characteristics variables. In detail, the continuous variables age, experience with people suffering from dementia and level of employment were integrated. Furthermore, the categorical variables gender (male/female), degree in nursing (nursing aides/practical nurses/licensed nurses) and care setting (special care units for persons with dementia/mixed long-term care institutions/home care setting) were collected.

| Ethics
Nursing directors and managers of the institutions permitted their nurses to take part in this study and to fill in the questionnaire during work time. Nurses were informed that their participation was voluntary, and anonymity was guaranteed. The regional Ethical Committee was informed about the study.

| Characteristics of the participants
The sample consisted of 468 nurses that completed the questionnaire. After exclusion of cases with missing data, a total sample of 417 could be achieved and was used for all subsequent analysis. A total of 376 (90%) study participants were women and 41(10%) were  (Table 1). Table 2 shows the correlations of the variables included. Two variables were significantly positively correlated with the DAS score: care setting V = 0.467 (p ≤ 0.001) and work experience with people with dementia ρ = 0.113 (p ≤ 0.05). The other variables such as gender, age, level of employment and degree in nursing, however, did not significantly correlate with nurses' attitudes towards people with dementia.

| Correlations of nurses' attitudes towards older people with dementia
Mean values of nurses' attitudes towards people with dementia, depending on the different care settings, are presented in Table 3. The Kruskal-Wallis test shows significant group differences (χ 2 = 87.5, df = 2, p<0.001). Dunn-Bonferroni tests were used for pairwise group comparisons. The mean attitudes score of nurses working in specialized long-term care institutions was significantly more positive than the attitudes scores of nurses working in home care setting (z = 7.6, p<0.001) or in mixed long-term care institutions (z = 7.4, p = 0.000). Although nurses in the mixed long-term care had a higher DAS score than home care nurses, this difference is not statistically significant (z = 1.9, p = 0.167).
Further Kruskal-Wallis tests with the independent variables work experience, degree in nursing and age showed no significant group differences in the mean DAS score. The results of the Mann-Whitney tests did not reveal significant group differences in the mean DAS score for the variables gender and level of employment. Table 4 shows the results of the regression analyses conducted.

| Regression analyses
Based on the results of the correlation matrix in Table 2, the care setting was entered in the last step of the regression analysis (assuming this variable has the highest predictive power). Control variables TA B L E 1 Socio-demographic and work characteristics and DAS score of the sample  Furthermore, the mean attitude score of nurses in mixed long-term care institutions was more positive compared with the mean score of nurses in the home care setting. Regression analysis showed that the care setting allows for a significant prediction of nurses' attitudes towards people with dementia.

| D ISCUSS I ON
Nurses working in the home care setting must deal with various degrees of physical and/or mental disabilities. People are mostly mildly to moderately limited in their activities of daily living (Wächter & Bommer, 2015). Thus, caring for persons suffering from dementia in an early or middle stage might be especially demanding for those nurses, as affected persons are often aware of their increasing losses and react with diverse symptoms such as grief, depression, anxiousness or aggression (Schweizerische Alzheimervereinigung, 2014c). Additionally, informal caregivers, such as relatives, are suffering from the strains of their caregiving (Ecoplan, 2013;Wilz, Adler, Gunzelmann, & Brähler, 1999;Zank, Schacke, & Leipold, 2007). Therefore, nurses working in the home care setting might get a rather negative view of the situations people with dementia live in. Furthermore, Wächter and Bommer (2015) showed that a high percentage of home care clients are not long-term clients but people who need timely limited professional support after hospitalization or acute illness. Thereby, home care nurses' experience in interactions with people suffering from dementia is limited and the so-called "challenging behaviour" of the affected people might be more difficult to handle for them. Training programmes for home care nurses should cover a wide range of health-related themes, whereof dementia is probably not the most prominent. Summing up, these points could explain why home care nurses have the lowest DAS score in this study.
In mixed long-term care settings, half to two-thirds of the residents are diagnosed with dementia (Schweizerische Alzheimervereinigung, 2014b). Nurses working in this setting must cope with the needs of people with dementia and with the needs of people with other diagnoses.
This is a challenge with which nurses in specialized long-term care are not confronted. This opportunity to focus only on the needs of people with dementia leads to higher job satisfaction in nurses in specialized longterm care in comparison with nurses in mixed long-term care (Oppikofer, Lienhard, & Nussbaumer, 2009;Pekkarinen et al., 2006). Dissatisfaction at work has consistently shown to be associated with nurses' negative attitudes towards people with dementia (Åström et al., 1991;Brodaty et al., 2003;Moyle et al., 2011;Zimmerman et al., 2005). In specialized long-term care, training programmes are chosen or designed based on the needs nurses have in caring exclusively for people with dementia. This leads to deeper dementia knowledge in nurses in specialized long-term care (Grant, Kane, Potthoff, & Ryden, 1996;Grant, Potthoff, Ryden, & Kane, 1998;Teresi, Grant, Holmes, & Ory, 1998). Training in mixed long-term care has to cover a wider range of themes.
There exists an alternative explanation for the results found.
According to various conversations the authors had with nurses of specialized long-term care institutions, the aspects of the setting do not influence nurses' attitudes towards people with dementia, rather nurses' attitudes are essential in the choice for a certain care setting.
Nurses who decided to work in a specialized long-term care institution often report to have experienced a high affinity for people with dementia when working with older people. These reports were also found online (https://www.derwesten.de/staedte/sprockhoevel/ tagespflege-fuer-an-demenz-erkrankte-id10571454.html?keepUrl-Context=true; accessed 13 August 2018). However, empirical literature supporting this hypothesis could not be found.

| Strengths and limitations
Similar studies included 49-291 participants. In comparison, the present study comprises a big sample which allowed to conduct all indicated statistical analyses with relatively big subsamples. Data were collected in different, urban and rural regions of German-speaking part of Switzerland. Institutions and organizations with various amounts of residents or clients were involved. All care settings relevant for people with dementia were represented. However, as the concrete number of the population, that is nurses caring for older people with dementia in German-speaking part of Switzerland is unknown, we cannot be sure that the sample is representative.
One possible limitation of this study is that self-assessment questionnaires underlie to social desirability bias. Unfortunately, the nationality of one fourth of the sample could not be assessed and could therefore not be included in the statistical analyses. Among the respondents, only ten percentage were not of Swiss nationality.
This proportion seems to be too low to be representative. Merçay, Burla, and Widmer (2016) report a proportion of 15%-19% of immigrated nursing staff in long-term care.
The paper-and-pencil test and the link to the online questionnaire were distributed via snowball. Therefore, the response rate could not be determined.

| CON CLUS ION
The presented results indicate that nurses' attitudes towards people with dementia were associated with the care setting. However, further research is needed to understand the underlying mechanisms.
Then, measures for practice intervention should be derived. The aim of such interventions would be to focus on enhancing nurses' attitudes towards people with dementia in all non-specialized care settings.
The presented results go line with existing studies showing that further dissemination of the person-centred care approach in every care setting is a promising way to positively influence nurses' attitudes. An equally promising way is to give dementia education and training for all nurses working with people with dementia, regardless of their degree in nursing and type of care setting.

ACK N OWLED G EM ENTS
The authors thank the long-term care and Spitex organizations who supported the study and the nurses who participated in the study.
We also thank the funders.

CO N FLI C T S O F I NTE R E S T
The authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N S
RB, JB: Study design. JB, RB: Data collection and analyses. RB, JB: Manuscript preparation. All authors have agreed to the final version of the manuscript.
• substantial contributions to conception and design, acquisition of data or analysis and interpretation of data; • drafting the article or revising it critically for important intellectual content.