Experience of aggressive behaviour of health professionals in home care services and the role of persons with dementia

Abstract Aims To explore the view of health professionals on the form and frequency of aggressive behaviour of clients against health professionals in home care services. Design An explorative cross‐sectional survey was conducted. Methods We conducted a survey using the Survey of Violence Experienced by Staff German version Revised (SOVES‐G‐R) and the Impact of Patient Aggression on Carers Scale (IMPACS). A convenience sample of 852 healthcare professionals from German‐speaking Switzerland participated. Data collection was conducted between July–October 2019. Data were analysed descriptively using IBM SPSS Statistics. Results Of the health professionals, 78.9% (N = 672) experienced aggressive behaviour since they worked in home care services. The most frequent aggressive behaviour was verbal aggression (75.6%, N = 644), while the most common predisposing factor was restriction in cognitive ability (71.3%, N = 67). Fear, burden and impairment of nursing relationship were common consequences of aggressive behaviour.


| INTRODUC TI ON
Home care services are a growing sector in the healthcare system worldwide (Genet et al., 2012) owing to the preference of most persons to stay at home as long as possible, as well as to the cost-effectiveness of home care services (Genet et al., 2012). In the current study, home care is defined as: "care provided by professional carers within clients' own homes. Professional care that relieves informal caregivers (respite care) has also been taken into account" (Genet et al.,p. 9). However, while the home care sector is growing, there are several challenges that burden health professionals in these services: lack of knowledge and education, time pressure, inconsistent or tight work scheduling, recruitment of employees, dilemmas related to autonomy, and conflicts related to provision of social support while maintaining a professional distance or refusal of care (Johnson et al., 2018). Besides these issues, workplace violence is a further challenge to be dealt with (Hanson et al., 2015). Violence against home care nurses seems to occur particularly in persons with dementia (Fitzwater & Gates, 2000;Galinsky et al., 2010;Schnelli, Karrer, Mayer, & Zeller, 2020). However, there is little research on dementia-related aggression in home care services. Hence, this study examines workplace violence against health professionals perpetrated by clients in home care services and the influence of dementia. Aggressive behaviour in this study is defined as behaviour that makes a person feel threatened, attacked or hurt, such as biting, hitting, verbal abuse or cursing (Steinert, 1995).

| Background
Existing literature shows that health professionals in home care services experience aggression perpetrated by clients frequently. Hanson and colleagues (2015) found that within 1 year, more than half (50.3%) of the 1,214 female health professionals surveyed in home care services in Oregon had experienced verbal aggression. Schablon et al. (2018) found that in Germany 56.7% of health professionals in home care services experienced aggression from clients within twelve months. Corresponding data for home care services in Switzerland are not available currently. However, data from inpatient settings in Switzerland indicate that aggression against health professionals occurs frequently (Hahn et al., 2010;Zeller et al., 2013). Hahn et al. (2010) state that 72% of the nurses in a general hospital in Switzerland experienced aggressive behaviour from patients or visitors within 12 months. Zeller et al. (2013) found in their cross-sectional study that 80% of health professionals in nursing homes in Switzerland experienced aggression from a resident within 12 months. In view of these findings, it can be assumed that aggressive behaviour in home care services in Switzerland may also be an issue. However, specific data are missing.
The consequences of aggressive behaviour against health professionals are far-reaching for both affected health professionals and the perpetrating clients. For example, increased stress level at work, anxiety and depression are possible consequences of aggressive behaviour for health professionals (Magnavita, 2013(Magnavita, , 2014. Hanson et al. (2015) showed that health professionals in home care services who experienced aggressive behaviour of a client were more likely to suffer from sleep problems or depression. In addition, knowledge from inpatient settings suggests that aggressive behaviour against health professionals can lead to more frequent fixations and disrupt the nursing relationship Paschali et al., 2018;Richter & Berger, 2009). However, while there is a broad body of literature on consequences of aggression in inpatient settings, little is known on consequences of aggression in home care services.
Furthermore, not only are the consequences poorly investigated in the home care setting but triggering factors on aggressive behaviour in home care services have also hardly been studied so far. However, several triggering and predisposing factors on aggressive behaviour in inpatient settings like hospitals, nursing homes or psychiatric hospitals are known: for example, long waiting times, aggressive behaviour of other patients, conflicts about nursing activities, late evening shifts Paschali et al., 2018;Richter & Berger, 2001). One of the risk factors which is often mentioned in literature is dementia. Yu et al. (2019) found that persons with dementia have a significantly higher risk of displaying aggressive behaviour than persons without dementia (27.8%, p = .000, OR = 4.9, 95% CI = 1.8-13.2). Zeller (2013) identified that 80.3% of the reported cases of aggressive behaviour against health professionals in nursing homes experienced within seven working days were perpetrated by a person with dementia. In our scoping review, we found several additional factors that, in combination with dementia, might trigger aggressive behaviour (e.g. supporting during personal hygiene or unmet needs like urinary urgency, hunger or social needs, Schnelli et al., 2020).
However, in what way dementia influences the occurrence of aggressive behaviour and which are the relevant triggering factors is currently not known (Schnelli et al., 2020).
In summary, research is needed to point out (1) the frequency of experienced aggressive behaviour in home care setting in the context of care of persons with dementia, (2) influencing factors of aggressive behaviour and (3) consequences of aggressive behaviour.

| Aims
The primary aim of the study was to assess occurrence of aggressive behaviour against health professionals in home care services in Switzerland. Second, the study will highlight which factors might influence the occurrence of aggressive behaviour in addition to dementia. Third, consequences of aggressive behaviour will be assessed from the health professional's perspective. Therefore, we formulated the following research questions: • How often and in what form is aggressive behaviour of clients experienced by health professionals in home care services?
• What are the triggering factors for the occurrence of aggressive behaviour of clients in home care services and what role does the predisposing factor dementia play from the health professionals' perspective?
• What are the possible consequences of aggressive behaviours of clients in home care services from the health professionals' perspective?

| Design
We conducted an explorative cross-sectional survey with health professionals in home care services in the German speaking part of Switzerland. This design allowed an overview of the current situation in home care services.

| Sample/Participants
In 2018, around 52,000 persons worked in 2,200 home care services in Switzerland. A convenience sample (N = 852) of adult health professionals working in home care services in the German speaking part of Switzerland was surveyed. We excluded independently working nurses. We asked professional associations of home care services to invite the home care services to take part in our project. A total of 24 home care organizations agreed to participate. The contact person in each home care service organization was trained about anonymity and voluntary participation. The questionnaire contained a statement, that by returning the questionnaire, the participants agreed that the information they provided in it could be used anonymously for the study. We included all health professionals in our survey who were working with direct hands-on contact with clients in the participating home care services: registered nurses, health specialists (a threeyear apprenticeship that ends with a diploma. The focus of this education is on basic care. A health specialist does not have the competences of a nurse), nursing assistants and house aids. Health professionals under the age of 18 were excluded.

| Data collection
We provided a total of 1,923 hard copy questionnaires corresponding to the exact number of staff to the 24 organizations.
Each questionnaire was prepared with an information sheet with the contact information of the main author and a prepaid and addressed envelope, in which the questionnaire was to be returned by mail after answering the questions. Each organization had a specific identification-code on the front of the questionnaire. The participants had a period of two months to answer and return the questionnaire. After a month, the researcher contacted the contact person at the institution again, informing him/her about the actual response rate and advising him/her to remind the staff to respond to the survey. Data collection was conducted between July-October 2019. A research assistant transferred the data from the returned hard copy questionnaires into a SPSS file using a codebook and under supervision of the project team. A double entry check was made on 10% of the data input from the hard copy questionnaires: The error rate was 0.2%.

| Instruments
We administered the survey of violence experienced by staff German version revised (SOVES-G-R) (Hahn et al., 2011;McKenna, 2004) which includes sociodemographic data, as well as the Impact of Patient Aggression on Carers Scale (IMPACS) .

| SOVES-G-R
The SOVES is a frequently used instrument to assess workplace violence in the health sector with 65 questions in seven sections.
It was originally developed by McKenna (2004) and tested for content validity by the European Violence in Psychiatry Group (McKenna, 2004). Its psychometric properties were tested by The further sections of the SOVES-G-R (g-h) are not included in this manuscript and are therefore only described shortly: Section (g) assesses support at the workplace as well as documentation and reporting of aggression events; and Section (h) assesses training in aggression management.
We adapted the SOVES-G-R for the home care setting in a discussion with two Swiss clinical nurse specialists working in home care services. Further, we conducted a face validity test with three health professionals in the home care services (a nurse, a health specialist and a nursing assistant). The adaptations were related to setting specific wording (such as "clients" instead of "residents" or "patients") and answer options (setting specific nursing interventions and options for reaction on aggression) and concerned selection responses in domain (d). In this manuscript, in line with the research question, the domains (a)-(f) (44 questions) were considered.

| Ethical considerations
The study was reviewed and approved by the ethics committee responsible for the eastern part of Switzerland (Project ID: 2019-00502 EKOS: 19/041). Data collection was voluntary and anonymous, and the participants were instructed not to provide identifying information in the questionnaire.

| Data analysis
Explorative descriptive data analysis was conducted using IBM SPSS Statistics (version 25). Fisher's exact tests (bivariate) were conducted for section (d) of the questionnaire to investigate factors correlating with dementia. A level of significance of 0.05 was assumed. MAXQDA version 2018 was used for the descriptive analysis of the free text answers, which were content coded. The identified contents in the first answers were labelled with codes and the list thus created was continuously updated with new codes emerging.
Content fitting to existing codes was assigned to the matching code.
Results were reported descriptively.

| RE SULTS/FINDING S
The response rate was 45.4% with a total of 874 questionnaires returned. Twenty-two (2.5%) questionnaires were excluded from data analysis because the survey was not answered (N = 5), cover page with the institutional code was missing (N = 1), sociodemographic data were missing (N = 8), less than 50% of the questions were answered (N = 8). Finally, we included 852 questionnaires in the data analysis (44.3%)

| Description of the institutions and participants
The percentages in this section refer to the number of analysed questionnaires (N = 852).

| Frequency and form of aggressive behaviour against health professionals
Seventy-nine per cent (N = 672) of the surveyed health professionals experienced aggressive behaviour of a client at least once since they worked in home care services (Table 2)

| Description of aggressive behaviour
A total of 94 free text answers were given to describe the most impressive aggressive behaviour experienced in the last seven days.
The most described aggression was verbal aggression. Verbal aggression included cursing or screaming at the health professionals, insults such as disparaging or racial remarks, direct questioning of nursing competence, playing health professionals against each other or complaining about the nursing intervention or the healthcare services. Complaints about the health professional being too busy, acting too fast or not conducting interventions properly were also described as verbal aggression.
The participants described physical threats and verbal threats.
Physical threats were, for example, threats with a fist in direction of the face of the health professional, kicking in the direction of the health professional, as well as sexual abuse by inappropriate touching. Verbal threats included threats to sue the health professional and threats to injure or to kill the health professional.
Physical aggression was described as hitting the health professional with the fist in the face or upper body, kicking them during the nursing activity, pinching, holding the arm of the health professional inappropriately tight, pushing, spitting or biting.

| Triggering and predisposing factors of aggressive behaviour in the last 7 days
In the most impressive aggressive behaviour experienced in the last seven days, 91.0% (N = 81 of 89 who answered the question) of the aggressive behaviour was shown by clients and 9.0%

| Triggering factors of aggressive behaviour in the last 7 days
Triggering factors were explored based on two sets of answers: situational triggering factors and further triggering factors.
"Misunderstanding" and "overstrain of the client" were the most reported triggering factors overall, both in the group of clients with dementia and in the group of clients without dementia. The main differences between persons with dementia and persons without dementia were visible in the triggering factors "dissatisfaction with care" and "dissatisfaction with the therapy" (Table 4).
Furthermore, in some descriptions of the aggressive behaviour, hints of the triggering factors of aggression could be identified.
These reported triggers were "resistance to care," "lack of insight into the need of nursing support" or "lack of compliance."

| Reactions to aggressive behaviour
The health professionals chose different strategies to handle the aggressive behaviour quickly with situational measures or in a broader period with further measures (Table 5). "Maintaining a distance to the client" and "informal discussion with staff" were the most chosen measures. Of note was that measures such as "alert the police" or "cancelling the collaboration of the home care service with the client" were not reported, although available for choice.

| The predisposing factor dementia
"Dementia" was the most reported psychiatric diagnosis of perpetrators (54.3%, N = 51) who acted aggressively within the last 7 days.
"Confusion" (p = .000) (Table 4) was a significantly associated triggering factor with the diagnosis dementia in persons acting aggressively. The trigger factors "dissatisfaction with care" (p = .000) or "therapy" (p = .000) ( Table 4) were significantly decreased in the clients with dementia who showed aggressive behaviour within the last 7 days. In addition, contacting the physician was significantly less frequent (p = .001) in clients with dementia (5.9%, N = 3 of 51 PwD) after an aggression incident than in clients without dementia (46.2%, N = 6 of 13 clients without dementia) in those 64 persons who chose this option.

| Consequences of aggressive behaviour
The questions in this section refer to the aggression experienced in the last twelve months. Included are the experienced burden, fear and the items of the IMPACS, which measures negative feelings after aggression.

| Negative feelings after aggression -IMPACS
In Table 7, the results of the IMPACS are illustrated. The Item "experience a disturbance in the relationship to the patient" was found in 29.4%, with ratings of "often" or "always." This was found to be the most frequently experienced negative feeling after an aggressive incident. Some participants noted in the section "further comments" that they found it difficult to complete the IMPACS. This was also reflected in the quality of the completion of the IMPACS. Not all of those who filled out the IMPACS, answered all the 11 Items. *Other: Addiction (3.2%, n = 3), loss of autonomy (3.2%, n = 3), concerns for the relative (1.1%, n = 1), depression/isolation (2.1%, n = 2), character trait of the CR (11.7%, n = 10), no information (1.1%, n = 1). a Other: loss of autonomy (2.1%, n = 2), scheduling (3.2%, n = 3), error in care (1.1%, n = 1), situation of the CR (1.1%, n = 1), character trait of the CR (6.4%, n = 6), no information (1.1%, n = 1). b Multiple answers possible.

| D ISCUSS I ON
This is the first study to describe the magnitude of aggressive behav- In the present study, in 54.3% of the reported aggressive behaviour, the perpetrator had been diagnosed with dementia and 71.3% of the perpetrators had declined in cognitive abilities. It is therefore suggested that in home care services for persons with dementia or with cognitive restrictions, aggressive behaviour occurs more frequently than in the care of persons without decline in cognitive abilities or dementia. Hence, the assumption noted in the scoping review of Schnelli et al. (2020) that dementia or decline in cognitive abilities in the clients is a predisposing factor for the occurrence of aggressive behaviour in home care settings is substantiated.
Most of the aggressive behaviour occurs in the context of closeto-body activities such as support in personal hygiene (Richter & Berger, 2001;Zeller et al., 2013). The most reported triggering factors in incidents with persons with dementia were misunderstanding of the situation on behalf of the client, overstrain and confusion.
These insights are in line with the results of a current review on the topic (Schnelli et al., 2020). However, it is surprising that in the reported aggressive incidents with a person with dementia as perpetrator, dissatisfaction with the care or the therapy is significantly less often a triggering factor than when the perpetrating client had no dementia. Nevertheless "enforcing the interventions of the care plan" was the most commonly mentioned triggering factor in the reported aggressive incidents with persons with dementia. One reason for this discrepancy might be that health professionals do not take into consideration that persons with dementia could be dissatisfied. (25.8%) to "always" (6.9%) avoiding contact with the perpetrating client after an aggressive incident and 42.9% (N = 200 of 466) felt insecure at work "sometimes" (32.6%) to "always" (1.5%). Avoiding contact with a client or feeling a disturbance in the relationship might lead to further consequences for the client, for example, neglect or abusive behaviour by the health professionals (Carter, 2016;Rabold & Goergen, 2007). Moreover, uncertainty of the health professional might lead to further aggressive incidents because of its influence on their capacity for action. These findings become even more explosive because social control in home of the clients almost completely disappears. Further research is needed to sharpen our knowledge of consequences of aggressive behaviour in home care services and their influence on quality of care. However, there are hints that although aggressive behaviour is less in home care services than in acute hospitals or in nursing homes, its consequences might be not less severe, especially when facing the fact that health professionals work alone.

| Limitations
We used a convenience sampling strategy. It is possible that organizations with larger problems with aggression and poor resources did not participate. Underreporting is a phenomenon discussed in this topic (Hahn et al., 2010). However, the hints provided by the results about the frequency of aggressive incidents allow the assumption that the number is higher because there are barriers to reporting aggressive incidents, such as not recognizing aggression (Schnelli et al., 2020). Further, dementia is underdiagnosed in home care services (Genet et al., 2012). Although the response rate was satisfactory, this study may not provide representative data. On the one hand, our sample was similar to the nationwide average regarding gender distribution and age, but the educational level of home care professionals was higher in our study compared with the nationwide average (Bundesamt für Statistik, 2016).

| CON CLUS IONS
Our results imply that aggression against health professionals in home care services is a common phenomenon. Specific situations while supporting clients with dementia during their personal hygiene triggered by overstraining or misunderstanding of the situation by the client and enforcing the intervention of the care plan by the health professionals lead to aggressive behaviour. Aggressive incidents are likely to result in fear and negative feelings of health professionals. Therefore, in practice, primary and secondary prevention concepts should be implemented and further research should focus more on the development of aggression, especially in the care of persons with dementia and its consequences.

ACK N OWLED G EM ENTS
The authors thank the associations of home care services and the managers of the home care services for their support and interest in this project. Our greatest appreciation goes to the participants of the study which diligently answered our set of questions.

CO N FLI C T O F I NTE R E S T
The authors declared no potential conflicts of interest concerning the research, authorship, and publication of this article.

AUTH O R CO NTR I B UTI O N S
AS, AZ, HM and SO: Study design. AS: Data collection. AS and SO: Data analysis. AS, AZ, SO and HM: Manuscript preparation.

DATA AVA I L A B I L I T Y S TAT E M E N T
The full data set is available on request from the main author (Angela Schnelli).