|Home | About | Journals | Submit | Contact Us | Français|
To more fully characterize the spectrum of RRA.
A focus group study of nursing home staff members and residents who could reliably self-report.
A large urban, not-for-profit long-term care facility in New York City
7 residents and 96 staff members from multiple clinical and non-clinical occupational groups.
16 focus groups were conducted. Content was analyzed with nVivo 7 software for qualitative data.
35 different types of physical, verbal and sexual RRA were described, with screaming and/or yelling being the most common. Calling out and making noise were the most frequent of 29 antecedents identified as instigating episodes of RRA. RRA was most frequent in dining and residents’ rooms, and in the afternoon, though it occurred regularly throughout the facility at all times. While no proven strategies exist to manage RRA, staff described 25 self-initiated techniques to address the issue.
RRA is a ubiquitous phenomenon in nursing home settings with important consequences for affected individuals and facilities. Further epidemiologic research is necessary to more fully describe the phenomenon and identify risk factors and preventative strategies.
Although the term “elder mistreatment” in the context of nursing home care invariably evokes images of resident abuse by staff, resident-to-resident aggression (RRA) may be a much more prevalent and problematic phenomenon. Cognitive impairment afflicts 80–90% of nursing home residents, often leading to behavioral disturbances and aggression. Behavioral disturbances are a well-known risk factor for nursing home placement. Collocating, nursing home residents with behavioral disturbances likely increases the potential for RRA.
Remarkably, there is scant literature describing the prevalence, key characteristics, and outcomes of RRA in nursing home settings. Instead, the extant literature predominantly focuses on abuse of nursing home residents by staff,[4–6] assaultive behaviors experienced by clinical nursing assistants (CNAs) rendering daily assistance to nursing home residents with dementia,[7–13] and non-specific agitated or disruptive behaviors of the residents.[14–18]
The goal of this research was to more completely characterize the spectrum of RRA and develop a brief instrument to measure the phenomenon that would be easy for staff to implement in residential settings. To that end, focus groups were conducted with stakeholders most likely to be familiar with aggressive episodes: nursing home residents and staff. For the purposes of this study, we define RRA as negative and aggressive physical, sexual, or verbal interactions between long-term care residents, that in a community setting would likely be construed as unwelcome and have high potential to cause physical or psychological distress in the recipient.
Nursing home staff members were recruited into the focus groups as informants with the following rationale. Staff members provide over daily care and supervision, and are responsible for most aspects of residents’ physical and emotional well-being. Since few official guidelines, policies, or prevention strategies for RRA exist, these staff members are responsible both for incident reporting and crisis intervention. Additionally, rather than focus exclusively on the clinical care staff (nurses and CNAs), we hypothesized that employees with non-clinical responsibilities (e.g., housekeepers) within the facility would have different experiences with RRA. These frontline caregivers will undoubtedly be the implementers of any intervention program designed to effectively prevent or control this phenomenon. In addition, one focus group was conducted with nursing home residents who could reliably self-report.
The study was conducted at a large non-profit nursing facility in Manhattan, New York City. The majority of residents were women (78%), non-white (52%), and lived in shared rooms (74%). A minority (8%) of residents lived on a subacute rehabilitation floor and a closed dementia unit (8%). Approximately 52% of participants carried a clinical diagnosis of dementia. The facility’s case mix index (CMI) based on New York State’s Resource Utilization Group (RUG) II reflected a resident population that was clinically complex and requiring moderate ADL assistance. The payer mix for residents was 7% private payment, 17% private insurance, and 76% Medicaid. For 2006, the average resident length of stay was 1.38 years, and the staff-to-resident ratio was 1.35:1.
Focus groups have become a widely accepted health research methodology.[20–23] They have been used to explore various aspects of life in nursing homes for more than a decade.[24, 25] Focus groups are ideally suited for an investigation of RRA given the dearth of rich descriptive data regarding its prevalence, key characteristics and outcomes.
In essence, the methodology effectively unearths the spectrum of participant experiences and feelings and elicits further reflection on one’s own perspective in the context of other convergent and divergent views in the group. Nursing home residents and staff members’ experiences and views on RRA allowed us to characterize broad themes and typologies of the phenomenon. Moreover, the focus groups shed light on the language that residents and staff members use to describe RRA. Extraction of information from transcripts is also useful for a secondary goal of this research: future development of survey instruments to measure and quantitatively characterize this phenomenon using language that is consistent with that of the respondents.
Nursing home resident focus group members were recruited by the Medical Director and Social Work Director at the facility. Staff participants were recruited by department managers. Groups ranged in size from 2 to 13, and each staff group only included participants of a single job category. The focus groups were moderated by two of the authors (TR, ML) and were conducted at various locations at the facility. The focus groups ended either when the questions were fully explored or when the participants needed to return to work. The study was approved by the Weill Cornell Institutional Review Board.
The moderator initially introduced the study and its objectives, and briefly described the focus group process. The focus group protocol consisted of two sections. The brief first section (approximately 5 minutes) asked residents and staff members to indicate whether they had ever witnessed specific types of RRA. By listing potential types of RRA, we were able to more fully and clearly define to participants what we understood of the phenomenon already and what types of behavior we were exploring. The second section used a semi-structured format, the type most commonly used during focus groups, which has previously been employed in qualitative violence research with caregiver informants. We used questions that were developed from our previous focus group research and two of the authors’ (ML and KP) extensive clinical experience in nursing home care. The questionnaire is available for viewing at: http://www.cornellaging.org/medical/rea_tools.html.
Each focus group was fully transcribed and reviewed by the authors, yielding 510 pages of transcripts. The narrative data from these transcripts was analyzed using content analysis.[26, 28] Transcripts were reviewed iteratively in order to identify individual conversation blocks where participants were discussing a specific resident or topic. These conversation blocks were designated as major themes that were then analyzed for semantic connections and grouped into larger categories. The data were analyzed using the nVivo 7 (QSR International, Doncaster, Victoria, Australia) software package.
We used these counts as a guide for further qualitative analysis, with the goal of developing grounded explanations for the observed patterns. Specifically, numeric counts of themes or characteristics were used as potential proxy indicators of their relative importance within the phenomenon. The counting of code frequencies was undertaken at the individual participant, focus group, and aggregate levels, since each may provide different information about the phenomenon.[26, 30] For analysis, the themes and characteristics were ordered for display by percentage of focus groups mentioning, then by percentage of participants mentioning, then by total number of conversation blocks in which mentioned.
We conducted 15 focus groups with a total of 96 nursing home staff members and one with 7 cognitively intact residents. The staff groups represented 11 different job categories: nurses, CNAs, medical staff, social workers, physical therapists / occupational therapists, dieticians, speech therapists, activities staff, front desk staff, housekeepers, and food service workers. For nurses and CNAs, we also conducted focus groups with those who worked during evening and night shifts. The group discussions ranged in length from 18 to 68 minutes with a mean of 45 minutes.
Table 1 lists the locations and times of day that participants described RRA occurring. Participants described it as occurring in nearly all public and private areas, and at all times of the day and night, but described the dining room (mentioned by 81% of groups and 47% of participants) and resident rooms (69% of groups and 48% of participants) as the most common sites of occurrence. Staff also described the afternoon as the most common time of occurrence (56% of groups and 24% of participants). Nineteen percent of focus groups, however, suggested that RRA could occur at any time of day.
Focus group participants described 35 different types of physical, verbal and sexual RRA that they had actually observed, allowing a fuller characterization of the phenomenon than possible in previous studies (Table 2). Representative dialogue from these discussions includes:
Front Desk Staff #1:…that escalates into, you know, fighting verbally. They start arguing with each other. Then it escalates to cursing at each other. Then it’ll escalate to, well, one will spit at the other, or they will slap the other…
Verbally aggressive behavior was mentioned more commonly than physically or sexually aggressive behavior. It was discussed during all focus groups (100%) and by 72% of participants, suggesting that it may be the most frequent RRA type. The most commonly mentioned of the 8 types of verbally aggressive behavior directed toward another resident were “screaming or yelling,” (88% of focus groups) and “telling another resident to ‘shut up,’” (81% of focus groups). Physical aggression was mentioned by 72% of focus groups and 66% of participants. Participants identified 23 different types of physically aggressive behaviors. “Pushing” and “punching or fighting” were the most commonly discussed, each mentioned by 44% of focus groups. Although less frequent, sexual aggression was discussed by 38% of focus groups.
Participants also discussed the reasons or triggers for RRA, identifying 29 different contexts and circumstances in which these aggressive episodes emerge. Triggers described included such diverse catalysts as communication challenges between residents who do not speak the same language to residents wandering into the private room of another resident. Staff members even proposed that the full moon might increase the likelihood of RRA, (a hypothesis that was actually tested, and not supported, for non-specific agitation among nursing home residents).
Many of the triggers were frequently mentioned, suggesting that certain common provocative circumstances may be identifiable. Table 3 displays the most commonly mentioned reasons. Notably, “calling out or making noise” was the most frequently mentioned trigger that was identified by 88% of groups and 51% of participants. Further, this was discussed in 43 conversation blocks, suggesting that several focus groups discussed the issue iteratively during the course of the session, underscoring its importance.
Another commonly mentioned trigger was territoriality and challenges of communal living, discussed in 75% of focus groups and by 39% of participants. Conflict may arise between residents due to competition for a preferred chair in a TV lounge or dining room, for example. The inability of roommates to communicate and compromise with each other’s preferences for television volume, heat and air conditioning, opening and closing of window and shades, lighting, or other environmental factors was reported in 69% of groups and by 36% of residents.
In addition to the common reasons or triggers described above, several discrete and notable clinical phenomena emerged from the focus groups. While likely not as widespread, these illustrate the potential complexity of RRA and the management challenges it poses for nursing home staff members.
Racism, ethnic stereotyping, and religious differences contributed to RRA. Many residents are not used to closely interacting with members of other races, ethnicities, and religious backgrounds before they enter nursing homes. Long-term care facilities are frequently multi-cultural, particularly in large urban areas. As cognitive impairment frequently includes frontal disinhibition, RRA can be motivated by these prejudices and/or can include racial slurs or other inappropriate remarks.
CNA - night shift #8
It’s not good. They call one another name if, for instance, one is black and one is white…Racial remarks towards one another. “You black whatever--you white whatever”
Housekeeping #5: The resident, he was white. His neighbor was--he was black…He said he stole the money. And I asked him what happened, you know, “What are you talking about?” So he said, “Oh, he stole my money.” I said, “Why are you saying that?” “Because he’s black.”
CNA - evening shift #6: We have one now she has a roommate. And she said, I don’t know, she said the roommate is colored…And when I went in there she said, “Don’t you ever do anything for that lady.” And I said, “Why?” “Because you are here to take care of me, and I’m an American.”
Staff, who are often from different cultures and subjected to this abuse themselves, must manage this behavior.
Six focus groups described incidents wherein one resident’s calling out behavior would trigger another resident to be verbally abusive, subsequently leading other residents to direct their displeasure and anger at those who initially engaged in the behavior. This may create what staff repeatedly described as a “chain reaction,” where many residents in a dining room or public setting yell at each other, and calming them often proves difficult:
Dietician #3: …if one resident is causing disturbance throughout, there may be one particular resident that may feel that they’re in charge, quote, unquote, and that will say “Shut up, stop it.” And then that might bring on the echoing of the other residents on the floor, you know…
CNA – day shift #11: Sometimes it could get physical. Like, if a group of them in the dining room, and one starts behaving badly, walking around and screaming, other ones pick up. “Oh, she better shut her mouth or I’m gonna punch her out.” And then after a chain reaction and all of them gonna start. For half an hour, you have to try to calm– take that one out of the dining room.
Another situation described by 3 focus groups suggests that much RRA may not include intent to harm other residents. On the contrary, some residents may be physically aggressive while actually trying to help their fellow residents. Staff discussed incidents where confused residents actually tried to take care of other physically dependent residents, often mimicking CNA behaviors:
Medical Staff #7: …one resident actually is kind of demented. And she was trying to take care of her roommate. And she think her roommate is not being well cared, and she was trying to take care of her. She’s so demented, she was trying to- like once she even pulled out her tube. And when the resident was getting I.V. and then she tried to take it out. And she thought she was taking care of her, she was helping, and she did not intend to hurt her.
Social Worker #8: She was ambulatory. She was able come and go--and she had a friend—who was her old roommate--who was physically impaired, cognitively impaired, couldn’t speak—bedbound, basically—and the resident would go into her room and say “She’s needs help. She’s calling out for me.” And she would undress her. The staff perceived it as inappropriate touching as well.
Territoriality was described as a common trigger for RRA. In response, staff would often describe trying to discourage residents from growing attached to chairs, walkers, or other items, and repeatedly reminding them that the facility is a communal environment where resources must be shared. At the same time however, establishing and maintaining a routine is a common strategy that is used by staff to prevent demented residents from becoming agitated. This routine includes sitting the resident in the same chair every day and reinforcing the very territorial behaviors that might create aggression. Representative dialogue includes:
CNA – day shift #7: …Some of them they have their special chairs.
CNA – day shift #1: All of them have their own seats…“I want my chair.” Except for the dementia. We try to sit them in the same place every day…for their routine. So we sit them in the same spot every day. So, it’s sometimes hard, ‘cause one’ll say “I’m not gonna -- why should I get up from this chair? I’m not gonna get up.” And the other one says, “That’s my chair.”
Evidence from the focus groups suggested that some nursing home residents may repeatedly engage in RRA. As one nurse noted:
Nurse - day shift #4: And right now I have two residents in the unit who we cannot put--these are male residents, both of them--and we cannot put them to sit next to female residents. Because they always find physical, they always try to touch, or to feel, or to--
Staff described 25 different strategies they used to attempt to manage RRA. Table 4 displays the most commonly mentioned strategies identified. Examples included allowing the argument to continue, simply watching residents vigilantly, talking calmly to settle them down, to installing nylon barricades with alarms, contacting security, or requesting a psychiatric consultation. Some interventions were innovative, such as giving one resident headphones to allow him or her to listen to the television without disrupting his or her roommate. Others involved encouraging a resident to be patient with and tolerant of those who are confused or intrusive. The most common approach is for staff members to contact social services and change the residents’ rooms, as mentioned by 50% of focus groups and 20% of participants. Thirty-eight percent of focus groups described having to physically intervene or separate residents who were aggressive; this included having to take away objects such as canes or walkers that were being used as weapons.
RRA was described as adding significant distraction and stress to nursing home staff, who are already working in a very challenging environment.
Speech Therapist #1: We probably tend to be distractable. I don’t know…I find once people start shouting and yelling and everything and things are going on, it’s kind of hard to stay focused on what you are trying to do.
This problem is particularly acute when staff members are drawn into a resident dispute and must act as “referees” to adjudicate it, an occurrence described by 4 focus groups (25%).
This is the first qualitative study of resident-on-resident aggression (RRA) in nursing home settings using focus groups of staff and residents to characterize the spectrum of associated behaviors. We found that all employee groups and residents had experience with RRA and that respondents are willing to discuss it. Our study further found that RRA is a common, varied, complex phenomenon. Consistent themes emerged from these in-depth focus groups regarding the types, triggers, circumstances, and management strategies.
Our strategy of including residents and employees with varied job responsibilities in focus groups proved to be informative. Additionally, different shifts and staff categories witnessed different types of RRA with in a host of provocative contexts. By including this variety of staff members with different experiences, we were able to integrate their insights to more fully understand and characterize RRA.
We are aware of only two previous studies of RRA; neither collected detailed information from staff in a prospective fashion. Shinoda-Tagawa et al (2004) conducted a case-control study comparing Massachusetts victims of RRA (identified through the state’s ombudsman program) to “violence free” controls. The study used only official reports to the State Department of Health for case finding and relied on the data in the Minimum Data Set (MDS) for background information. Thus, its results were subject to underreporting and non-standard reporting policies and practices in different institutions. Also, MDS data is unreliable and error-prone. Several Minimum Data Set (MDS) derived factors were more common in subjects experiencing RRA: male sex, behavioral disturbance (especially wandering), moderate functional dependency, and cognitive impairment. Lacerations, bruises, and fractures were common injuries. RRA occurred most commonly in patient rooms but was also common in the dining room and hallways.
Our research group demonstrated that 5.6% of members of an established elder cohort in New Haven, Connecticut had police contact after institutionalization, with 89% of these contacts resulting from RRA. Indirectly, this finding suggested that the problem was potentially significant. This study also suggested a preliminary typology for RRA including unprovoked assault, invasion of space, male unbonding, competition for resources, and breaking point.
Our study expands on this work; focus groups represent a rich source of primary data providing significant insights into RRA, including clinical detail not available in previous studies. Participants described afternoons as the most common time period for the occurrence of RRA and attribute this to fatigue of the daytime primary care staff, disruption during the shift change, as well as resident fatigue and potentially boredom / lack of activity. Participants described RRA as occurring in nearly all public and private spaces, suggesting that this phenomenon is not confined predominantly to intimate care settings (e.g., bathrooms) as suggested by much of the extant literature. The most frequent locations were dining rooms where residents congregate, and resident rooms, where many nursing home patients spend most of their time. A wide variety of verbally, physically and sexually aggressive behavior were described, suggesting that RRA is phenomenologically complex and may not represent a unitary syndrome.
Reactions to calling out or noise-making behavior were the most commonly described trigger. Patients with dementia can exhibit calling out behaviors such as singing, moaning, repetitive speech, or screaming, and this vocally disruptive behavior occurs in 11–30% of nursing home residents. It is among the most challenging behaviors for nursing home staff  and can trigger anxiety and agitation in other residents.[36, 37] Because our study also identified other triggers such as intolerance of cognitively impaired residents and aggressive behavior towards wanderers, it appears that the commingling of non-demented and demented residents may be a significant factor that contributes to RRA. Teresi and colleagues found that non-impaired residents living with or near demented residents had higher rates of living situation dissatisfaction. While RRA was not investigated in the study, features that seemed to contribute to dissatisfaction and demoralization included agitated behaviors, noise, and other disturbances reported to be caused by suite or unit mates.
Many of the most common reasons or triggers for RRA resulted from challenges with adjustment to a communal living environment, including territoriality, impatience, and jealousy. The frustration driving these aggressive episodes may derive from the lack of control and choice that many nursing home residents feel they have over their everyday lives. Territoriality, another frequently mentioned reason in this study, is a common behavior among institutionalized patients and was described in the literature nearly three decades ago. It is rooted in attempts of residents to exert control and convert public space into private space. This situation, named “competition for resources,” was identified as the most common typology for RRA requiring police intervention in our previous study. Empowering CNAs to personalize care and to customize daily routines to suit the desires of individual residents may reduce these behaviors.
One notable challenge of communal living that commonly leads to RRA is conflict between roommates. The challenges of adjusting to life in close quarters with a stranger are known to anyone who has lived in a college dormitory. This is exacerbated in a nursing home setting, where rooms may be very small with only curtains separating the occupants and roommate changes may be frequent. Evidence exists that older adults overwhelmingly prefer private rooms to shared rooms. Studies underscore the importance of this preference, with one finding that, for a cohort of assisted living and nursing home residents, “private room and bath” was rated most important of the 12 features evaluated, ahead of “a safe place to live,” “access to medical care,” and “good food.” A recently published analysis suggests that private rooms lead to better psychosocial and clinical outcomes for nursing home residents and that these factors should be considered in facility design and construction. Our findings support and provide further evidence for this argument, as RRA may be an element contributing to poor adjustment and satisfaction when residents are congregated.
Future studies may evaluate other potential triggers for RRA. Research suggests that visual and hearing[46,47] impairment may increase nursing home resident agitated behaviors, though the findings on vision are not unequivocal. Therefore, these impairments may have an effect on RRA. Sensory impairment, by increasing disorientation, may also alter the type of RRA in which residents engage, making verbal more likely than physical. It is important to evaluate longitudinal patterns of RRA to understand the phenomenon more completely. For example, weekends and holidays, with the reduction in structured events, may have more RRA occurrences. One participant noted:
Dietician #2: …maybe they’re sitting around with nothing to do, and that’s when they get on each other’s nerves.
Studies also have shown that residents with poor quality of relationships were more likely aggressive, suggesting that poor family relationships and relationships within the nursing home may be a predictor for RRA behaviors. Our study found that loneliness and a feeling of abandonment were among the triggers for RRA. Understanding the impact of these and other triggers on the phenomenon will assist in developing effective interventions.
Many of the clinical phenomena described highlight the potential complexity of RRA and the variety of issues that must be addressed to successfully prevent it. CNAs, while trying to discourage territoriality among residents to prevent altercations, also attempt to create routine for cognitively impaired patients. This routine often includes placing them in the same seat each day. Research indicates that cognitively impaired nursing home residents are less able to deal with a change in habits, as their coping skills may be depleted, leading to behavioral problems. This paradigm, ironically, pits two strategies for RRA management against each other.
Only limited guidelines and training materials exist for staff prevention, intercession, and management of RRA, so staff have developed a large variety of formal and informal strategies to attempt to manage this phenomenon. Many of these strategies are innovative, and several have been successful at reducing RRA. A first step to developing comprehensive evidence-based interventions involves examining these techniques in greater detail. This study also suggests that all staff types witness RRA and attempt to intervene, as these events can occur without warning and frequently happen when primary care staff are not immediately available to assist. Therefore, while successful interventions may primarily involve primary care staff, such as CNAs, nurses, and physicians, all staff with resident contact can assist if properly trained and should be included in strategies to protect residents.
This study has several limitations. The qualitative data preclude making conclusions about the quantitative aspects of the incidence, prevalence, or severity of RRA. Many of our conclusions are based on coding transcripts and counting code frequencies. Although this is an established research technique, counts and their interpretation may be affected by the subjective way in which the transcripts are divided into blocks and the relative length of different focus groups. Also, comments made most frequently by focus group participants may not necessarily represent the most important themes. Despite these limitations, by conducting a large number of focus groups with a significant cross-section of the facility’s employees, we reduced the possibility that results would be dramatically affected by a single focus group or methodological choice during coding.
As all our research was conducted at a single long-term care facility, our findings may not be generalizable to all nursing homes. Also, all staff and resident participants were recruited by facility administrators, thus this was not a stratified random probability sample. Further, as in any study of this kind, self-reporting of RREM incidents may be unreliable. Nursing home staff only witness a small sample of resident interactions, and staff may be biased informants, especially when they are the targets of aggression. Nevertheless, these groups are the front-line witnesses of the phenomenon with the greatest potential to provide insights into its characteristics.
Episodes of RRA are probably common and detract from quality of life in long-term care; incidents may also predispose nursing homes to state and federal sanctions and civil lawsuits. We hope that this report will encourage further study of RRA, ultimately leading to prevention and intervention strategies to minimize the adverse consequences of the phenomenon.
We are grateful to the American Federation of Aging Research (AFAR), whose Medical Student Training in Aging Research (MSTAR) fellowship program provided the funding for Tony Rosen’s participation. Karl Pillemer and Mark Lachs also acknowledge support from an Edward R. Roybal Center grant from the National Institute on Aging (1 P50 AG11711-01). Dr. Lachs is the recipient of a mentoring award in patient-oriented research from the National Institute on Aging (K24 AG022399).
Conflict of Interest Disclosures:
Dr. Lachs: Board member of the American Federation for Aging Research; has served as an expert witness in both criminal and civil cases of elder mistreatment.
Dr. Bahrucha: Speaker's bureau for Forest Pharmaceuticals; serves as a consultant for Forest Pharmaceuticals
Dr. Stevens:> Lsted on two patents related to CareMedia, a research technology that has been used to monitor behavior in nursing homes. These patents are: Patent Number: 5,835,667, “Method and Apparatus for Creating a Searchable Digital Video Library and a System and Method of Using Such a Library” and Patent Number : 5,664,227, “System and Method for Skimming Digital Audio/Video Data
Author Contributions: Tony Rosen participated in study concept and design, acquisition of subjects and/or data, analysis and interpretation of data, and preparation of manuscript.
Sponsor’s Role: none
Tony Rosen, Division of Geriatric Medicine and Gerontology, Weill Cornell Medical College, Cornell University, New York, New York.
Mark S. Lachs, Division of Geriatric Medicine and Gerontology, Weill Cornell Medical College, Cornell University, New York, New York.
Ashok J. Bharucha, Western Psychiatric Institute and Clinic, University of Pittsburgh, Pittsburgh, Pennsylvania.
Scott M. Stevens, Human-Computer Interaction Institute, Carnegie Mellon University, Pittsburgh, Pennsylvania.
Jeanne A. Teresi, Research Division, Hebrew Home at Riverdale, Columbia University Stroud Center and Faculty of Medicine, New York State Psychiatric Institute, New York, New York.
Flor Nebres, Amsterdam Nursing Home, New York, New York.
Karl Pillemer, Department of Human Development and Cornell Institute for Translational Research on Aging, Cornell University, Ithaca, New York.