In the long-term care setting, many cognitively impaired elderly patients require partial or total distance with activities of daily living (ADLs) (Beck, Heacock, Rapp, & Mercer, 1993; Berg, 1991). With demented patients, the need for physical care is progressive (Berg, 1991; Maas, Swanson, Specht, & Buckwalter, 1994; Mungas, Weiler, Franzi, & Henry, 1989). At the same time, aggressive behavior is more likely with patients who have impaired ADL and cognitive functioning (Cohen-Mansfield, Marx, & Rosenthal, 1990; Colenda & Hamer, 1992; Mentes & Ferrano, 1989). Aggressive behavior has adverse effects, not only on the patients, but lso on the physical and mental health of those who interact with them (Giancola & Zeichner, 1993). The behavior problems common with these patients interfere with giving care and are extremely burdensome for caregivers (Ray et al., 1992).
As many clinicians have observed, much of the aggression takes place during assistance with bathing and grooming. Recent studies show that the majority of physical aggression by patients is directed toward staff and occurs during assistance with personal care (BridgesParlet, Knopman, & Thompson, 1994; Dougherty, Böiger, Preston, Jones, & Payne, 1992; Mentes & Ferrano, 1989).
PROBLfMANDFOCUS OF THIS STUDY
in spite of the demanding nature of the task, most nursing staff have had little or no training to develop the needed skills. This is especially true for nurses' aides, who are responsible for 80% to 90% of the direct care given to the patient, but who receive the least training (Cohn, Horgas, & Marsiske, 1990). In mis study, a theory-based training program was designed to help nursing staff develop knowledge and skill in preventing or reducing aggressive behavior of cognitively impaired elderly patients during assistance with bathing and grooming. The methods and principles included in the program are applicable to all settings where assistance is provided to patients with dementia or other forms of cognitive impairment.
THEORETICAL AND CUNICAL BASISFORTHETUAINING
Patient aggression, as defined in this study, includes both verbal aggression (such as cursing and racist remarks toward staff) and physical aggression (such as hitting, kicking, scratching, or pushing).
Dementia patients and others with cognitive impairment have perceptual distortions and diminished ability to interpret environmental cues (Cohen-Mansfíeld, Marx, & Rosenthal, 1990; Letemendia, 1985). Thus, they are often bewildered by the sights, sounds, and physical stimulation surrounding bathing and grooming. This can trigger feelings of fear, helplessness, and frustration (Berg, 1991). When the elderly confused patient misinterprets the environment, the situation can escalate into a catastrophic reaction, where the individual becomes excessively violent (Struble & Sivertsen, 1987). Multiple factors, including environmental conditions (such as noise or lack of privacy) and ineffective procedures by staff in approaching the patient can contribute to aggression (Colenda & Hamer, 1992; Rader & Harvath, 1991; Wagnild & Manning, 1985). Aggression is often a response to intrusion into the patient's personal space, and may be more of a defensive reaction than an expression of anger (Bridges-Parlet, Knopman, & Thompson, 1994).
The following core concepts from established theories of nursing care provide a theoretical basis for training staff:
1) Orem's concept of self-care (Orem, 1980). Staff should help the patient attain and sustain self-care by: a) diagnosing the self-care deficits and abilities of the individual, b) selecting methods of helping, and c) working cooperatively with other staff to provide support and guidance.
2) Orlando's concept of deliberate actions to help the patient in distress (Orlando, 1961). Patients may react with distress to procedures intended to be therapeutic. Staff should focus on meeting each individual patient's needs by: a) ascertaining the specific reasons for the patient's behavior and distress, including examining how their own actions and reactions are helping or not helping the patient; b) identifying the specific help the patient needs to relieve the distress; and c) taking deliberate actions to provide the needed help, then assessing the results by observing the behavior that follows. Staff are cautioned to be aware of automatic actions, which are those taken for reasons other than the patient's immediate needs.
3) Peplau's concept of the therapeutic, interpersonal process (Peplau, 1952). The quality of the interpersonal interactions between nurse and patient is critical to positive outcomes. Staff should use these interactions to understand the patienf s individual needs and goals, and to help the patient achieve increasingly higher levels of independence.
As the nursing literature emphasizes, training for staff in this area must be relevant and experiential, and must provide skill practice (Beck, Heacock, Rapp, & Mercer, 1993; Cohn, Horgas, & Marsiske, 1990; Mentes & Ferrario, 1989). In order for staff to go beyond learning the skills to changing their approaches when providing assistance to patients, the training should incorporate a normative-re-educative strategy (Bernard & Walsh 1981). This requires that the trainer make the desired change clear to staff, allow them to participate actively during the training, and interact directly with staff following training to provide needed support and positive reinforcement. It can be expected that staff will have difficulties and need help with understand ing or application.
SITE AND SUBJEOS FOR THE STUDY
The study was conducted in a rural, 210-bed, state-operated geropsydiiatric hospital located in the southeast. This facility provides both acute and long-term treatment services to patients who are 65 years of age or older. The patient mix includes approximately 50% dementia, 30% schizophrenia, and 20% other mental illness. These patients suffer different levels of cognitive impairment as a result of various medical and psychiatric conditions.
Staff participating in the study consisted of all nursing staff on three hospital units: aides (n=71), LPNs («=12), and RNs («=14). Ages ranged from 22 to 75 years, with a mean o 41 and standard deviation of 11.4 education 12 to 18 years, mean=12.8 sd=l.l; experience 1 to 20 years mean=9.8, sd=6.2. These staff had not been previously trained to use the model outlined in this study.
DESIGN AND STRUCTURE OF THE TRAINING INTERVENTION
To construct typical case exam pies for use in the training, observa rions of staff with a sample o patients from each unit were com pleted using an observation check ist, and the behaviors of both staff patient were recorded. These observation not only provided baseline information about current practices, but also revealed patterns uch as the following:
1) Sometimes patients begin with nild resistance, such as by saying 'no" or gently pulling away. When this doesn't work, they then escalate o a stronger form of resistance, such is verbal or physical aggression.
2) Skilled caregivers sometimes espond to the patient in a way that will prevent the escalation, such as by using distraction, giving them more time and space, approaching with a different staff member, or allowing choices.
3) For some patients, staff sometimes provide total assistance just because they know the patients are cognitively impaired, assuming that they are not able to do any part of bathing and grooming themselves. Some of these patients, in spite of their impairment, actually still have the ability to perform part of the procedures for themselves and get upset with staff for trying to do it all for them. They then may become passive and refuse to comply with a specific request, or they may become generally resistive and uncooperative toward anything the staff attempted to do for them.
The training was designed to include four components: a) helping staff re-examine the approaches they were currently using; b) introducing the methods outlined in the model and explaining why they are useful in meeting the specific needs of these patients; c) demonstrating how to apply the methods and allowing staff to practice them; and d) emphasizing that the trainers will assist them in using the methods after training, and men following this up with observation and consultation on the units. This last component is particularly important because staff are reluctant to change current practices unless they are confident someone in authority will support them when they do.
The first part of the training consisted of two one-hour classroom sessions for all direct care nursing staff, including aides, LPNs, and RNs. Emphasis was given to 1) understanding the patient's perceptions and responses, 2) supportive behaviors to use, and 3) individualization of approach.
Two trainers jointly conducted each session. Both were registered nurses with advanced training in psychiatric and gerontological nursing; one was a staff clinical nurse specialist with a master's degree and the other a student nearing completion of the master's degree. Staff from three units were trained, with participants grouped by unit. The second group was trained approximately 2 months after the first, and the third group several months later.
Nursing administration was supportive of this training because it addressed an important problem, and cooperated with trainers to make scheduling adjustments to allow staff to attend while maintaining adequate coverage on the units. While some staff initially resisted attending the mandatory sessions, they became interested and active participants once the trainers explained the purpose and gave examples of how the skills would be useful.
The acronym R.E.S.P.E.C.T. was adopted to help staff remember key practices and concepts, and to emphasize the importance of conveying respect toward the patient at all times. This paradigm is shown in Table 1.
Case examples were included to illustrate how to apply the practices. During learning activities, staff alternately played the role of patient in order to improve empathy. Each participant demonstrated how the practices would be used in typical situations. Additionally, each developed a plan of how they would apply the methods with a patient who was considered especially difficult.
The second part of the training consisted of follow-up instruction with staff as they applied the methods on the unit. Staff chose to work with those patients on their units who were most aggressive during bathing and grooming. All of these patients had some form of dementia, with cognitive impairment ranging from moderate to severe. All had impaired ADL functioning; 40% required total assistance with ADLs, while 60% required a moderate level of assistance. The trainers conducted random, unannounced observations of all participants until each had been observed at least twice. The observations served to reinforce the key practices staff had been trained to use, as well as to provide feedback to the trainers. When staff had difficulty applying the methods, the trainers provided additional consultation or guidance. When staff used the methods as instructed, the trainers acknowledged their interventions and encouraged them to continue.
Key Practices fn Providing Assistance wifffi Baffling and Grooming
Pre;- and Postatesi Scores (Showing Number and Percentage of Right Answers) for the Three Groups Trained
A21-item test was written specifically for this study to assess staff knowledge of 1) factors related to pathing and grooming that can contribute to aggression, 2) factors related to cognitive impairment in elderly patients that can affect aggression, and 3) strategies that prevent or minimize aggression while providing assistance. Reliability of the test was assessed using internal consistency analysis; Cronbach's alpha was .69, indicating moderate reliability. No analysis of validity of the test was performed in this study.
The test was administered preand post-training to assess knowledge gain. In scoring the test, one point was given for each right answer, so that the maximum possible score was 21. Administration of the post-test was delayed until 4 months after the training for Group I, and 1 month after training for Group II, to provide some indication of how lasting the effects of the training were. With Group II, the posttest was administered immediately after the training.
A mixed-factor design, with one between-groups factor and one repeated-measures factor was used to analyze changes in scores following the training, applying the Analysis of Variance (ANOVA) procedure with adjustment for unequal group sizes. Comparison of scores of the three training groups was then completed using the Bonferroni method of multiple group comparisons.
Pre- and post-scores on the knowledge test for the three training groups (units) are shown in Table 2.
All three groups achieved meaningful gains in knowledge from preto post-test. The mean score for Group I increased by 14.3%; Group II increased by 15.7%; and Group III increased by 17.8%. In addition to the larger mean scores, the standard deviation decreased following training for all three groups, which tends to further confirm knowledge gains in that this indicates greater consistency in knowledge scores following training.
The one-factor, two-repeat-measures model revealed that the change in test scores following training for all participants was significant [F(I, 87)=78.51, p<0.0001]. It also revealed that there was a significant difference in scores among the three groups [F(2, 87)=46.04, p<0.0001]. Scores were compared for the demographic variables of job position, age, experience, and education. The only significant difference in scores by demographic variable was on the variable of job position, with RNs scoring significantly higher (p=.05) both on the pre- and post-tests.
The Bonferroni multiple t-test procedure (with alpha=0.05, 0.017 for each group) was then used to perform post hoc comparisons of differences among the groups. This analysis revealed that the amount of gain in scores was significantly different for each of the three groups. As shown in Table 2, the largest difference in gain was between Groups I and II, with Group I showing the smallest gain, and Group II showing the largest gain.
In addition to test results, observations of staff were helpful in assessing the effects of the training. Random, unannounced observations by trainers of staff while assisting patients showed an observable gain in application of skills. A sampie of eleven separate observations was documented, revealing that after training staff were doing more of the following to adapt to the individual needs of each patient: 1) offering more choices, 2) incorporating patient's interests or hobbies, 3) using a written bathing plan, 4) altering bathing schedules, 5) encouraging self-care, 6) making changes in the environment or procedures used, 6) offering praise and encouragement, and 7) sharing more ideas with each other. Unit nursing supervisors were also asked to report their observations; they indicated that post-training, aides were taking initiative to use more creative plans suited to the specific needs of individual patients, whereas before the training they waited for a supervisor to give them permission first. This was a result of learning both how to develop a plan and that they had the authority to do so.
Indica tors of change in the level of patient aggression during assistance with bathing and grooming were examined for one of the three units in this study, the one having the most patient aggression. Aggressive incidents went from 41 during the month immediately preceding training, to 19 during the month immediately following. For the same 1month comparison periods, use of PRN medications associated with bathing dropped from 31 to 16, and staff injuries from patient aggression during assistance dropped from an average of four per month to zero. In anecdotal reports, unit supervisors noted that less time was required for staff to provide bathing assistance after the training, because patients were less aggressive. They also saw a noticeable reduction in complaints by staff regarding their bathing assignments.
There was a significant and meaningful gain in knowledge and skill following completion of training, for all three groups. Even though the post-test was delayed until a month after training for Group ?, and 4 months after training for Group I, these two groups still showed significant gains. Thus, it appears that there was little deterioration in knowledge gains over time.
Group I had the lowest mean score of the three groups both preand post-training. No data are available from the current study to explain this. The mix of licensed and unlicensed staff was similar for each of the three groups; education levels were also similar. Since this group was from the unit with the most aggressive patients, it is possible that the unit environment, stress among staff, or other factors could have affected the scores. Further investigation would be useful in explaining this outcome.
Observations of staff on the units following training revealed that staff were applying what they had learned. The fact that unit staff and supervisors acknowledged that the saw a noticeable reduction in aggressive behavior following the training and that less time was required to assist patients with bathing and grooming, was helpful in getting support to continue using the model.
The design of this training program drew from a strong theoretical base and extensive clinical experi enee, so it was likely that it would be of some help in improving staff knowledge and practice. However this program included several elements, any one of which might account for the positive outcomes For example, it may be that the increased emphasis given by trainers and supervisors to reducing aggression had the greatest effect. Or perhaps the follow-up consultation be trainers was most influential. Further study would be helpful in separating and assessing the effects of these and other contributing factors.
Patient aggression during bathing and grooming is especially burdens some for staff and patient. Any gaint in reducing such aggression have multiple immediate benefits. Over time, these benefits contribute to the patient's general well-being, dignity, self-esteem, and quality of life. For staff, learning to handle situations that previously were frustrating can increase their sense of job competence and mastery. This can lead to reduced caregiver burden and burnout, and increased job satisfaction. Thus, the positive results achieved from this training program are encouraging.
Further study should focus on maintaining the gains over time specially on those factors that reinorce the application of learning. When possible, random assignment of staff to training groups and control groups should be used to allow more rigorous evaluation. It would also be heipful to evaluate other approaches that include more specific levels of assistance.
- Beck, C., Heacock, P., Rapp, C.G., & Mercer, S.O. (1993). Assisting cognitively Impaired elders with Activities of Daily Living. journal of Alzheimer's Care and Related Disorders & Research, pp. 11-20.
- Berg, M.D. (1991). Special care units for persons with dementia. Journal of the American Geriatrics Society, 39, 1229-1236.
- Bernard, L.A., & Walsh, M. (1981). Leadership: The key to the professionalization of nursing.. New York: McGraw-Hill.
- Bridges-Parlet, S., Knopman, D., & Thompson, T.A. (1994). Descriptive study of physically aggressive behavior in dementia ly direct observation, journal of the American Geriatrics Society, 42, 192-197.
- Cohen-Mansfield, J-, Marx, M.S., & Rosenthal, A.S. (1990). Dementia and agitation in nursing home residents: How are they related? Psychology and Aging, 5(1), 3-8.
- Cohn, M.D., Horgas, A.L., & Marsiske, M. (1990). Behavior management training for nurse aides: Is it effective? Journal of Gerontological Nursing, 16(11), 21-25.
- Colenda, C.C., & Hamer, R.M. (1992). Antecedents and interventions for aggressive behavior of patients at a geropsychiatric state hospital. Hospital ana Community Psychiatry, 42(3), 287-292.
- Dougherty, L.M., Böiger, J.P., Preston, D.G., Jones, S.S., & Payne, H.C. (1992). Effects of exposure to aggressive behavior on job satisfaction of health care staff. Journal of Applied Gerontology, tt, 160-172.
- Giancola, PR., & Zeichner, A. (1993). Aggressive behavior in the elderly: A critical review. Clinical Gerontologist, 13(2), 3-22.
- Letemendia, M. (1985, April 24). An ageold problem. Nursing Times, 30-32.
- Maas, M.L., Swanson, E., Specht, }., & Buckwalter, K.C. (1994). Alzheimer's special care units. Nursing Clinics of North America, 29(1), 173-195.
- Mentes, J.C, & Ferrano, J. (1989). Calming aggressive reactions: A preventive program. Journal of Gerontological Nursing, 15(2), 22-27.
- Mungas, D-, Weifer, P., Pranzi, C., & Henry, R. (1989). Assessment of disruptive behavior associated with dementia: The Disruptive Behavior Rating Scales. Journal of Geriatric Psychiatry and Neurology, 2(4), 196-202.
- Orem, D.S. (1980). Nursing: Concepts of practice (2nd ed.). New York: McGraw-Hill.
- Orlando, I.J. (1961). The dynamic nursepatient relationship: Function, process ana principles. New York: G.P. Putnam's Sons.
- Peplau, H.E. (1952). Interpersonal relations in nursing: A conceptual frame of reference for psychodynamic nursing. New York: G.P. Putnam's Sons.
- Rader, ]" & Harvath, T. 1991, September/October). How to document difficult behavior. Geriatric Nursing, 231-233.
- Ray, W.A., Taylor, J. ?., Lichtenstein, M.J., Meador, K.G., Stoudemire, ?., Lipzin, B., & Blazer, D. (1992). Managing behavior problems in nursing home residents. Geriatric Medicine, 1(4), 71-112.
- Struble, L.M., & Sivertsen, L. (1987). Agitation behaviors in confused elderly patients. Journal of Gerontological Nursing, Î3(ll), 40-44.
- Wagnild, G.M., & Manning, R.W. (1985). Convey respect during bathing procedures. Journal of Gerontological Nursing, Ï2(12), 6-10.
Key Practices fn Providing Assistance wifffi Baffling and Grooming
Pre;- and Postatesi Scores (Showing Number and Percentage of Right Answers) for the Three Groups Trained