Nurses in long-term care fácilities are increasingly called upon to care for elderly residents with moderate to severe mental impairment; the incidence of cognitive impairment among residents in long-term care facilities can be as high as 94% (Chandler & Chandler, 1988). Caring for the demented elderly resident in a long-term care facility can be extremely difficult and stressful due to difficult behaviors such as yelling, wandering, abusive language, and physical aggression, e.g., hitting, biting, scratching and kicking. Studies of the incidence of aggressive behavior in long-term care have found it to be high (Pétrie, Lawson, & Hollender, 1982; Ryden, Bossenmaier, & McLachlan, 1991; Whall, Gillis, Yankou, Booth & BeelBates, 1992; Zimmer, Watson, & Treat, 1984), with one study noting that up to 84% of nursing home residents exhibited behaviors that were serious enough to endanger themselves or others (Winger, Schirm, & Stewart, 1987).
Typically, it is the nursing staff in long-term care centers - nurses and nurses' aides - who are the targets of aggressive behavior in elderly residents. Documentation of physical abuse of staff by elderly residents in long-term care facilities is scarce, and such abuse is traditionally under-reported by nursing staff who may feel it is just "part of the job" (Beck, Robinson, & Baldwin, 1992; Goodykoontz & Herrick, 1988; Mentes & Ferrarlo, 1989; Winger, Schirm, & Stewart, 1987). Nevertheless, there is a growing awareness among both management and nursing staff that the cost of ongoing physical abuse of staff by demented elderly patients is very high, in terms of workman's compensation claims, staff injury, sick days, increased absenteeism, low staff morale and high staff turnover. There is also a growing awareness that although a certain degree of patient aggression is unavoidable when caring for the cognitively impaired elderly, many of the episodes of aggression toward nursing staff may well be related to their lack of knowledge regarding the nature of dementing illness in the elderly. In most long-term care facilities, nurses' aides provide 80% to 90% of the care to residents, yet many aides have received little or no training in approaches for preventing or reducing behavioral problems in the cognitively impaired residents they care for (Feldt & Ryden, 1992).
Two primary studies in the nursing literature have described attempts to provide nursing staff with inservice education on how to prevent and manage behavioral problems (including aggression) in the cognitively impaired elderly. Mentes and Ferrano (1989) implemented an inservice program - focusing on preventing and managing aggressive episodes in cognitively impaired elderly - for nurses' aides in a 342-bed nursing home. By charting the number of incident reports related to patient aggression for 3-month periods before and after the inservice program, the authors noted a slight decline in the incidence of staff abuse following the program. Incident reports, however, require a nursing staff member to attach his or her name to a form, and the very process of having to fill out a formal report may result in a vast under-rating of the actual incidence of patient aggression toward staff.
In another study, Feldt and Ryden (1992) implemented an inservice education program for nursing assistants employed in a 147-bed nursing home. In their evaluation of the program, no differences were found between preand post-dementia knowledge test scores, but nurses' aides did view their task of caregiving as significantly more rewarding and significantly less frustrating as a result of the inservice training. No attempts were made to document the incidence of aggressive episodes, or to test the effect of the education program on the incidence of such episodes.
This study, therefore, was conducted as a continuation of what little research has been done in this area to date. This study hoped to obtain a more accurate picture of the incidence of resident aggression toward nursing staff and the effects of a staff education program on such incidents of aggression, by using a system of reporting resident aggression toward nursing staff members mat was anonymous and confidential (unlike incidence reports).
To evaluate any impact the staff education program may have had on the incidence of physical aggression by confused elderly residents toward nursing staff, a quasi-experimental design - the simple interrupted time series design - was employed (Campbell & Stanley, 1963). Ongoing measurements were made of elderly resident aggression toward nursing staff for a period of 8 days prior to the educational program, and for a period of 8 days after the educational program was completed. The education program took 3 months to offer, during which time no measurement of aggression occurred. Thus, the complete process, including the pre- and post-program measurements of aggression, and the education program itself, took approximately 31/2 months.
Figure 1. Checklist for aggression by elderly residents toward staff.
Setting and Sample
The long-term care facility for this study, Overlander Extended Care Hospital, is a 200-bed extended care facility in Kamloops, British Columbia, Canada (population 70,000). Approximately 70% of the residents are female, approximately 90% suffer from moderate or severe dementia, and the average age of residents is roughly 80 years. To qualify as residents in the facility, people must require considerable nursing assistance with personal care and mobility.
All of the approximately 120 nurses' aides and 30 registered nurses at Overlander Extended Care Hospital were invited to participate in the education program. Staff were encouraged to attend the inservice education program available to them during paid working hours. At the end of a 3-month period (one module per month), 134 out of the 154 total nursing staff (87%) had attended the three inservice modules, along with 21 out of the 31 total housekeeping staff (67%) and all 16 (100%) of the therapeutic services staff (i.e., recreational staff).
Measurement of Resident Physical Aggression
This study required some means by which the incidence and nature of physical aggression of residents could be measured in a manner which was anonymous, confidential, and convenient to use. The final data collection form which was used (Figure 1) was an amalgamation of Ryden's (1988) Aggression Scale and a form used by the Juan de Fuca Hospital in Victoria, British Columbia (Monahan, 1992).
Prior to the collection of data, information sessions were held on each unit. The investigators and head nurses explained the inservice program and the associated research project designed to evaluate it. During these sessions, nursing staff were asked to voluntarily participate in the research project by consenting to collect data on the incidence of aggressive episodes before and after the educational program. Staff were introduced to the data collection form, kept in the nursing unit's report book, which the staff were asked to use to document each incident of physical aggression directed toward them by elderly (i.e., age 65 and over) residents in their care. The staff were fully informed that their participation in this data collection process was completely voluntary, and that by using the data collection forms they were giving their implied consent to participate in the research project. To ensure confidentiality and anonymity, the data collection forms were designed such that no staff or resident names were recorded.
The staff were asked to use the forms to record each incident of physical aggression which occurred while caring for an elderly resident for an 8-day period prior to the beginning of the education program. At the end of each day/ the data forms were collected from all the units by the researchers and replaced with new ones, which also gave the researchers an opportunity to answer any questions the staff might have, and to encourage their ongoing participation in the study. An 8-day initial data collection period was chosen for the reason mat an initial pilot study suggested that an 8-day period would give us an ample number of incidents of physical aggression, and that the staff's full involvement and interest in the data collection process was likely to wane for any period longer than a week.
Figure 2. Content areas for three-module inservice program.
The same method of data collection was used for an 8-day period after all the educational inservices were completed. All efforts were made to ensure that the nursing staff were just as vigilant in using the data collection forms after the educational program as they were before the program, and the head nurses and the nursing staff themselves reported that nursing staff were equally conscientious and enthusiastic about using the data collection forms for the 8-day periods before and after the educational program.
THE EDUCATIONAL PROGRAM
One of the prime considerations of the educational program was to ensure that the program was concise and relatively brief, as both nursing staff and management agreed the best program would be a short one allowing the nursing staff to attend the inservices while on duty. Since everyone agreed that 30 minutes was the maximum practical time a nursing staff member could physically leave her/his nursing unit to attend an inservice, the authors and staff decided on a program based on three, 30-minute modules. This arrangement meant that the facility could avoid the cost of having to pay staff to come in on their free time to attend the inservice program, and that the program did not have to risk low turnouts by relying on staff to voluntarily attend the program on their free time. Knowing these time limitations, the authors strove to establish what the most important material was that could be incorporated into such a short inservice program.
Another factor which guided the development of our program was the concern that because the program's primary goal was to reduce resident aggression through the use of staff inservices, the nursing staff might somehow perceive that the resident aggression toward them was therefore their "fault." In order to alleviate this, the authors attempted to stress throughout the inservice modules mat staff were not responsible for resident aggression, but that certain knowledge and skills might help to prevent some aggression from occurring. By giving the clear message to staff that much resident aggression could not be prevented no matter what anyone did, the staff reported they felt less guilty and more motivated to learn how to reduce some of the aggression that was occurring.
Ultimately, the authors developed an educational program, using the model of an educational program developed by Feldt and Ryden (1992), and divided the material into three, 30-minute modules. The modules used a mix of teaching strategies; i.e., short lecture, case studies, group discussion, role plays and the use of pre-videotaped vignettes.
The Education Modules
An outline of the three, 30-minute inservice modules is presented in Figure 2.
The first session briefly reviewed dementia, particularly stressing how the various losses associated with dementia (e.g., reasoning, short-term memory) increase the possibility of an elderly demented individual becoming aggressive. This information led to a discussion of the various risk factors for aggression in the demented elderly, and how certain nursing strategies (e.g., trying to reason with a demented elderly resident) might increase the likelihood of becoming a target of physical aggression. Many of the nursing staff, particularly some of the nurses' aides, reported that this session was their first realization of what dementia actually was, and led to a new understanding of why some of the residents in their care might be so frightened and aggressive.
The second session, building upon the basic knowledge the staff now had of dementia, focused on how to prevent aggression from occurring. Using Ryden's (1992) five basic goals of care with the demented elderly as a framework, nursing interventions to promote these five goals were discussed, reviewed on videotaped vignettes, and roleplayed. The five basic goals of care for the demented elderly include: for the client to feel safe, for the client to feel physically comfortable, for the client to experience a sense of control, for the client to experience optimal stress/ and for the client to experience pleasure (Ryden & Feldt, 1992). Using these five basic goals of care, individual cases of aggressive residents were presented by nursing staff, and nursing strategies were brainstormed during this session which would help to achieve a particular goal of care for the aggressive individual. For example, nursing strategies were generated during this session for an elderly woman who always became fearful, combative and aggressive when placed in a mechanical lift during bathing. These strategies, which were aimed at the goal of helping the client to feel safe, were tried out the following week with considerable success.
The third and final session focused on what safe and effective responses are available to nursing staff once an elderly confused resident has become physically aggressive. This session presented four main strategies to de-escalate aggression once it has already occurred in a confused elderly person: 1) accepting instead of contradicting a confused elderly person's reality, 2) validating the feelings of the confused elderly person, 3) using a "cooling off" period, and 4) using distraction as a means of diffusing "catastrophic reactions." A videotaped vignette was presented to the staff, using the case example of a particularly combative and confused resident at this long-term care facility, demonstrating the use of these four strategies. Finally, to minimize the chance of injury to either staff or residents in the event of resident aggression, simple techniques for the non-violent management of physical aggression were demonstrated and practiced. This included techniques on how to respond safely to such situations as being grabbed, pinched, choked, bitten, or having one's hair pulled.
Delivery of Program
Each module was taught by the facility inservice coordinator/ instructor twice a day for a onemonth period, which allowed nursing units to send most of their staff to each module. The inservices were kept to 30 minutes each, and were offered during "down" times on the units (late morning and mid-afternoon) to facilitate staff attendance. Head nurses from each unit would ensure that only one or two staff were gone from each unit at any given inservice, and colorful poster boards which recorded staff attendance at the inservices were prominently displayed on the nursing units - which helped to foster a healthy competition among the staff to attend all inservices. As these inservices were non-compulsory, it appeared that the high attendance figures (i.e., 87% of all nursing staff) indicated a great interest in, and satisfaction with, the educational program. Indeed, many of the staff made comments to this effect during the inservice, and made requests for the program to continue on an annual refresher basis.
Figure 5. Incidents of aggression by time of day (N=260).
A total of 275 incidents of physical aggression toward nursing staff by elderly residents were reported during the 16 days (24-hour periods) that data were collected, or an average of just over 17 incidente per 24-hour period. In terms of type of aggression, "pinching or grabbing" was the most common type of physical aggression, comprising 42% of all incidents (Figure 3). This was followed by "hitting or punching" at 30% of all incidents, and "shoving or pushing" at just under 20% of all incidents.
When the authors analyzed the incidents of resident aggression by what the nursing staff were doing when the aggression occurred, nearly half of the incidents occurred while the resident was being "dressed or changed" (Figure 4). The second most common nursing activity during resident aggression was "transfer or turning. "When speaking with nursing staff, they commented that these two activities - "dressing /changing" and "transfer /turning" most often occurred during times of personal care, such as cleaning and changing an incontinent resident.
The fact that most resident aggression toward nursing staff during times of personal care was also reflected by analyzing the time that most aggression occurred (Figure 5). As the nursing staff themselves confirmed, all of the peak times for aggression occurred during times when the nursing staff give the majority of personal care: morning care between 0800 and 1100, pre-bed care around 2000, and turning or changing residents in the middle of the night at 0100. When the incidents of physical aggression were collated according to shift, it was noted that just over half of the incidents occurred during the day shift (Figure 6).
Furthermore, the incidents of physical aggression were evenly split between males (51%) and females (49%), although males only make up approximately 30% of the entire resident population. Finally, only 6% (14) of the incidents of physical aggression by residents were reported to have caused injury, although the nature of the injuries were not reported.
Evaluating the Impact of the Program
Quantitative Data - From the total of 275 incidents of physical aggression by elderly residents toward nursing staff that were recorded, 182 incidents occurred in the 8-day data collection period before the staff inservices, compared with 93 incidents in the 8-day period after the staff inservices. Figure 7 presents a breakdown of the incidents of physical aggression before and after the educational program. Reports of resident aggression dropped by approximately 50%.
Clearly, a drop of approximately 50% in the reported incidents of physical aggression toward staff is clinically significant. To test the statistical significance of the difference between pre- and post-inservice incidents of physical aggression, the data were broken down into incidents occurring by shifts. Thus, the incidence of aggression toward staff for each of the 24 shifts (8 days times 3 shifts) before the inservices was compared with each of the corresponding 24 shifts after the inservices. A paired sample t-test was then used, pairing each shift preand post-inservice (e.g., day 1 morning shift pre-inservice compared with day 1 morning shift post-inservice). The paired sample t-test, using the 24 shifts pre-inservice compared with the 24 shifts post-inservice, obtained a result of (=3.24, significant at the p=.004 level.
Qualitative Data - As we collected feedback and comments from participants in this program, it became evident that they felt pleased with the program. Specifically, participants felt the program had a positive impact on the nursing care they gave demented residents, which in turn resulted in fewer incidents of resident aggression occurring toward them. As one nursing staff member put it: "I think it is great that everybody is hearing this stuff, and not just the "good staff." I feel that some of them will listen and try a different way to do up Mr. XXX, for example." Nursing staff reported that the approximately 50% drop in reported aggression reflected actual changes in their approach to nursing care of demented residents. While staff felt the program information was anywhere from a "good review and reminder" to "totally new stuff which really helped me understand dementia," most staff seemed to feel it encouraged them to try different approaches with residents who were identified as being aggressive. One staff member also noted that "I think it (the program) really helped, because the staff felt that they weren't being blamed." Thus, giving the staff a sense of acceptance and validation of their difficult caregiving situations and frustrations seemed to be an important component of this program. As another staff member said, "it's about time somebody paid attention to what's really happening out there on the wards!"
Figure 7. Incidents of aggression by day: Pre- vs. post-inservice (N=275).
Discussion of Results
One of the more striking findings of this study was the sheer number of reported incidents of physical aggression toward staff by elderly residents; there were a total of 275 reported incidents in a period of 16 days. While the authors found this to be an alarmingly high number of incidents, the nursing staff in the study reported that these numbers were a realistic reflection of the extent to which they are physically "assaulted" by confused elderly residents daily. Thus, in the nursing staff's minds, this study validated their own perceptions and helped to communicate to the nursing management the severity of the problem of resident aggression. As a result of this study documenting the high frequency of resident aggression toward staff, and the positive feedback received from staff regarding this program, the management in this particular long-term care facility has taken steps toward implementing this program on a regular ongoing basis.
This study also served to validate the nursing staff's reports as to the most "dangerous" times and nursing activities with regard to resident aggression. In discussion with the nursing staff, there was general agreement with the study's results that times of personal care, such as changing, dressing and turning, were times when the resident was most likely to be physically aggressive. Thus, it was not surprising to find that incidents of physical aggression peaked during times of the day when personal care was most likely to be given; that is, early morning and early evening.
With regard to the reduction of roughly 50% in the reported incidents of physical aggression after the staff education program, the nursing staff felt this reduction accurately reflected a change in their perceptions and actions. Many of the staff reported that learning about the nature of dementia and some techniques for working with residents with dementia helped them to identify nursing situations where aggression was potentially preventable. For example, staff noted that if a resident became frightened during turning in the middle of the night, rather than proceeding with the turning (which might well lead to increased fear and physical aggression), the staff would turn a bedside light on, leave the situation, and reapproach the resident more slowly after 5 or 10 minutes had passed. Thus, while the staff could not prevent all incidents of aggression from occurring, they were beginning to learn how certain nursing actions could help to minimize the occurrence of fear responses, and hence certain kinds of physical aggression, in confused elderly residents.
Limitations of Study
While the primary intention of this study was not so much to establish causal relationships as to ensure the education program included an effective evaluation component, there are some methodological limitations within the study itself. First, a variety of pragmatic reasons - including requests for complete staff and resident anonymity - made it difficult to utilize any sort of comparison groups for this study. The lack of comparison group(s) makes it particularly difficult to state with any real confidence that it was the education program per se which was responsible for the 50% reduction in reported incidents of physical aggression. There are several alternate explanations for what might have caused the reduction in reported aggression, including:
1) A change in the proportion of aggressive residents in the facility. However, after reviewing the recent admissions in this 3-month period, the head nurses did not feel there was a significant decrease or increase in the proportion of aggressive residents.
2) A turnover in staff. Despite the fact that some turnover of staff occurred during the 3-month period between data collection periods, turnover was minimal.
3) Seasonal differences. There may have been some seasonal differences between the first (August) and second (December) data collection periods which have made residents less aggressive.
4) Differences in staff reporting behavior. Among both the nursing staff and the head nurses, it was a shared perception that the staff were reporting equally conscientiously and accurately for both the first and second data collection periods. Nevertheless, as the education program was seen very positively by the staff, mere is the possibility mat the nursing staff were under-reporting aggression during the second data collection period, so as to obtain research results which would reflect favorably upon the program's effectiveness - and help to ensure the program's continuation.
While an interrupted time-series design (Campbell & Stanley, 1963) was used to help increase the study's internal validity, it was felt that it was unrealistic for nursing staff to have the time and commitment to collect information for more than an 8-day period in a row. While it would have been preferable to collect data for longer periods before, during and after the education program, even with 8-day data collection periods where staff were reminded and encouraged daily about the project, data reporting decreased gradually each day, likely reflecting some kind of fatigue factor within the study.
A number of important nursing implications arise from this study. One obvious implication is the need for some kind of education and support program, to help nursing staff in long-term care centers deal with the difficult and pervasive problem of aggression in elderly residents with dementia. Both the quantitative and the qualitative evidence from this study show that such educational programs are not only sorely needed, but can also be highly effective in helping staff to prevent many incidents of physical aggression from occurring.
Another implication surrounds the kind of nursing actions most associated with physical aggression. While the nature of aggression may vary from one long-term care facility to another, the majority of physical aggression toward staff in this study occurred during personal care. Thus, for this particular facility, it may well prove most effective and efficient to direct nursing education inservices specifically toward nursing interventions involving personal care. Other facilities may need to focus on nursing interventions involving other aspects of resident care, such as wandering. An important first step for reducing aggression in any long-term care facility, therefore, is an accurate assessment of just when the majority of aggression seems to occur, and an analysis of the nursing actions associated with the aggression.
Finally, unexpectedly high support and attendance for this educational program - and its associated evaluation, was displayed by all levels of nursing staff. This may reflect the keen interest of nursing staff in the problem of resident aggression, as well as the considerable effort the authors used to try and ensure that nursing staff had input into the educational program, and its evaluation. Our experience has shown that nursing staff at all levels in long-term care facilities can be keen and able participants in the clinical problem solving and research process, providing their input is solicited, incorporated, and validated at all steps of the process.
- Beck, C.M., Robinson, C., & Baldwin, B. (1992). Improving documentation of aggressive behavior in nursing home residents. Journal of Gerontoiogical Nursing, 18(2), 21-24.
- Campbell, D.T., & Stanley, J.C. (1963). Experimental and quasi-experimental designs for research. Chicago: Rand McNaIIy College Publishing.
- Chandler, J., & Chandler, J. (1988). The prevalence of neuropsychiatrie disorders in a nursing home population. Journal of Geriatric Psychiatry and Neurology, 1, 71-76.
- Feldt, K.S., & Ryden, M.B. (1992). Aggressive behavior: Educating nursing assistants. Journal of Gerontologien! Nursing, 18(5), 3-11.
- Goodykoontz, L., & Herrick, C.A. (1988). Evaluation of an inservice education program regarding aggressive behavior on a psychiatric unit. The journal of Continuing Education in Nursing, 23(3), 129-133.
- Mentes, J.C., & Ferrano, J. (1989). Calming aggressive reactions·. A preventive program. Journal of Gerontological Nursing, 15(2), 22-27.
- Monahan, A.M. (1992). Resident Aggression Checklist. (Available from A.M. Monahan, Juan De Fuca Hospital, Victoria, B.C., Canada).
- Petrie, W., Lawson, E., & Hollander, M. (1982). Violence in geriatric patients. Journal of the American Medical Association, 248, 433-434.
- Ryden, M.B. (1988). Aggressive behavior in persons with dementia who live in the community. Alzheimer Disease and Associated Disorders, 2(4), 342-355.
- Ryden, M-B. (1992). Alternatives to restraints and psychotropics in the care of aggressive, cognitively impaired elderly persons. In: K.C. Buckwalter (Ed.) Geriatric mental health nursing: Current and future challenges (pp. 84-93). Thorofare, N.J.: Slack.
- Ryden, M.B., Bossenmaier, M., & McLachlan, C. (1991). Aggressive behavior in cognitively impaired nursing home residents. Research in Nursing and Health, 14, 87-95.
- Ryden, M.B,, & Feldt, K-S, (1992). Goaldirected care: Caring for aggressive nursing home residents with dementia. )ournaJ of Gerontological Nursing, 18(11), 35-41.
- Whall, A.L., Gillis, G.L., Yankou, D., Boom, D.E., & Beel-Bates, C.A. (1992). Disruptive behavior in elderly nursing home residents: A survey of nursing staff, journal of Gerontological Nursing, 18(10), 13-17.
- Winger, J., Schirm, V, & Stewart, D. (1987). Aggressive behavior in long-term care. Journal of Psychosocial Nursing and Mental Health Services, 25(4), 28-33.
- Zimmer, J.G., Watson, N., & Treat, A. (1984). Behavioral problems among patients in skilled nursing facilities. American Journal of Public Health, 74(10), 1118-1121.