Printer Friendly

Non-pharmacological interventions for aggression in persons with dementia: a review of the literature.

The Problem of Aggression

Studies examining aggression in persons with dementia typically define aggression in terms of a variety of physical (e.g., hitting, pinching, biting) and/or verbal (e.g., cursing, threatening) behaviors. Estimates of the prevalence of aggression in cognitively impaired individuals vary widely from study to study, likely due to variations in definition of aggression and how aggression was measured. These estimates range from 13-86% (Deutsch, Bylsma, Rovner, Steelt, & Folstein, 1991; Hamel, et al, 1990; Lyketosos, et al., 2000; Pavesa, et al., 1992; Ryden, Bossenmaier, & McLachlan, 1991; Swearer, Drachman, O'Donnell, & Mitchell, 1988; Zimmerman, Watson & Treat, 1984). Aggressive behavior is also strongly correlated with greater dependence during self-care (Schreiner, 2001). These numbers suggest that aggression in persons with dementia is a significant problem and becomes more likely as the disease progresses.

Aggressive behavior can have serious consequences for persons with dementia as well as their caregivers. Aggressive behavior increases distress and burden for caregivers, which can result in nursing home placement (Hamel, et al., 1990; Ryden & Feldt, 1992). In addition, approximately 50% of nursing assistants have been injured during resident assaults (Gates, Fitzwater, Telintelo, Succop, & Sommers, 2004). Residents may also be at risk of being injured when engaging in aggressive behaviors (Ryden & Feldt, 1992). Aggressive residents in long-term care facilities may be labeled as "difficult" or "combative", which can result in social isolation and modifications in caregiving that can exacerbate problems (e.g., having 2 or 3 caregivers dress a resident instead of just one). Furthermore, due to the aversive nature of working with patients that are aggressive, the relationship between caregivers and patients is compromised, thereby decreasing quality of life for both parties (Ryden & Feldt, 1992). In fact, aggression is a significant source of job-related stress and frustration for professional caregivers (Everitt, Fields, Soumerai, & Avorn, 1991; Hagen & Sayers, 1995).

Another consequence of aggression is the administration of medications to manage the problem (Sloane, Mathew, & Scarborough, 1991). Aggression has been most typically treated using conventional or atypical antipsychotic medications. Studies have indicated that these medications produce only modest benefits in persons with dementia and carry significant dangers such as increased risk of stroke, exacerbation of cognitive decline, and increased risk of death (Schneider, Dagerman, & Insel, 2005; Sink, Holden, & Yaffe, 2005). Excessive sedation is also a common side effect associated with anti-psychotic medication (Zarit & Zarit, 2007, p. 311). These adverse side effects can result in a reduction in the individual's behavioral repertoire in terms of impairing language, causing gait disturbance that can result in falls, reducing the ability to access preferred events/activities, and causing further confusion and cognitive decline. The difficulty is that individuals with dementia are already experiencing a gradual deterioration in their behavioral repertoire (e.g., language, self-care) due to the disease process itself. It can be argued that treating individuals with medications that can further limit their behavioral repertoire is questionable ethically. Therefore, it is clear that developing restraint-free interventions for managing aggression is very important. The following section will include a review of empirical studies that investigated non-pharmacological interventions for managing aggression in persons with dementia.

Procedures and Literature Review

An initial search for relevant studies was conducted using the PsychInfo and Ageline electronic databases. Searches involved using various combinations of the search terms "dementia", "aggression", "intervention", "treatment" and "behavior therapy." This search generated a total of 40 articles. Each of these articles was then reviewed according to the following criteria: 1) all participants were diagnosed with some condition that causes progressive dementia; 2) the study was empirical in nature and was not simply a review of the literature, a narrative case description or a description of an intervention; 3) the study utilized a non-pharmacological/restraint-free intervention; and 4) the study included a specific measure (e.g., direct observation, questionnaires) of verbal or physical aggression. This final criterion ruled out any studies that targeted the general class of "agitated" behavior, but did not specifically target or measure aggressive behavior. The term agitation is frequently used in gerontological literature to refer to a wide variety of topographies of disruptive behavior in persons with dementia (e.g., pacing, fidgeting, frequently asking questions, or disruptive verbalizations). For the current paper, however, the authors focused exclusively on studies that specifically targeted and measured aggression given the prevalence of aggression, the numerous negative consequences associated with aggression, and the distressing nature of aggression for nursing staff. Using these criteria, a total of 18 studies were included for review. These studies were grouped according to the type of intervention utilized.

Distraction-Based Interventions. Three of the reviewed studies utilized interventions that involved distracting residents who were aggressive during bathing. For example, Thomas, Heitman, and Alexander (1997) utilized familiar music (as identified by family) to reduce aggression during bathing in 14 persons with dementia of the Alzheimer's type. Preferred music was played prior to bathing and during the participant's entrance into the bathing room. An adapted version of the Cohen-Mansfield Agitation Inventory (CMAI) was used to measure a variety of behaviors including aggression. This adapted CMAI utilized a dichotomous scoring system that involved caregivers to indicate that a behavior either occurred or did not occur during a bathing session. Data was collected during a total of 9 bathing episodes--3 baseline sessions, 3 music therapy sessions, and 3 post-intervention sessions. Statistical comparisons of the frequency of aggressive behavior in each of the 9 sessions were conducted. Results indicated statistically significant reductions in the frequency of aggressive acts between some music therapy sessions and some post-music therapy sessions. No differences were found between baseline sessions and post-intervention sessions.

Clark, Lipe, and Bilbrey (1998) utilized a similar method of playing preferred music during bathing episodes. This study included 18 persons with dementia who reportedly had a history of aggression during bathing. Participants were randomly assigned to be observed during 10 consecutive baths with preferred music or 10 consecutive baths with no music. Following these 10 observations, the conditions were reversed such that all participants were observed during 20 baths--10 with preferred music and 10 without music. Direct observation was conducted and the frequency of 15 separate behaviors (e.g., yelling, abusive language, hitting) was recorded. When simply comparing the frequencies of aggressive behaviors between the music and the no music conditions, twelve of the fifteen behaviors occurred less frequently in music condition. Statistical analysis revealed that significant reductions occurred with regard to the variables of hitting and total number of aggressive behaviors.

Whall and associates (1997) implemented a distraction-based intervention to reduce aggression during bathing in a group of 31 individuals with dementia. Participants were assigned to one of two groups: a usual care control (n = 15) or the natural experience intervention group (n = 16). The intervention involved training a total of 10 nursing staff members to give showers using a variety of different distractors (i.e., "natural elements") such as sounds of birds or other animals, large bright pictures corresponding to the sounds, and offering food. Staff was also instructed to ask questions about the various natural elements and to discuss them if the patient appeared to respond positively. Physical aggression was measured using a variant of the CMAI that involved having trained research assistants observe all bathing episodes and make ratings of the severity of aggression. Although aggression decreased in the treatment group over time, these declines were not statistically significant.

Bright Light Therapy. Two different studies have examined the effects of light therapy on aggressive behaviors. Haffmans and associates (2001) studied the effects of bright light on 6 persons with dementia. Participants were exposed to 30-minutes of bright light each day for a period of two weeks. In addition, participants were randomly assigned to receive either 2.5mg of melatonin or a placebo. Aggression was measured using the Social Dysfunction and Aggression Scale (SDAS). Results indicated that although bright light therapy had a positive effect on motor restlessness, there was no effect on aggression. Furthermore, adding melatonin to a regimen of bright light therapy was no more effective than light alone.

Ancoli-Israel and colleagues (2003) investigated the effects of light therapy in a group of 92 individuals with severe Alzheimer's disease. Participants were randomly assigned to receive either morning bright light, morning dim red light or evening bright light for two hours each day for 10 days. Aggressive behavior was measured using the CMAI. Measurement occurred for 3 days of baseline, 10 days of treatment and 5 days of post-treatment follow-up. No significant effects were found on the "aggressive behavior" subscale of the CMAI.

Activity-Based Interventions. Five studies have investigated the effects of increasing activities for managing aggression. For example, Holmberg (1997) instituted a structured walking program for eleven severely demented individuals living in a dementia unit of a nursing home. The frequency of resident-to-resident aggression as reported in facility incident reports was measured over the period of one year. A 30% reduction in aggressive acts was observed on days where the walking program was implemented.

Sival, Vingerhoets, Haffmans, Jansen, & Hazelhoff (1997) also used a number of group, social, musical, and physical activities to manage aggression in three individuals with severe dementia. Activities occurred twice daily for four weeks. Aggression was measured using the SDAS. Compared to a four-week baseline, no changes in aggressive behavior were observed in any of the participants. In fact, two participants showed increases in aggression during the intervention phase and reductions during the 4-week post-intervention follow-up phase. The authors suggest that individualized activities may produce better results.

Svansdottir and Snaedal (2006) investigated the effects of music therapy group in a sample of 38 patients with moderate to severe dementia. Participants were randomly assigned to a music therapy condition and a control condition. Music therapy was conducted in groups that included 3-4 dementia patients and a music therapist. Music therapy sessions lasted 30 minutes and involved singing familiar songs while the therapist played a guitar and the patients played various instruments. A total of 18 music therapy sessions were conducted over a period of 6 weeks. The Behavior Pathology in Alzheimer's Disease Rating Scale (BEHAVE-AD) was completed by 2 nurses that were blind to the purposes of the study. BEHAVE-AD ratings were completed pre-treatment, post-treatment, and at a 4-week follow-up. Results indicated that no significant changes in aggressive behavior occurred at post-treatment or at follow-up.

Other investigators have examined whether activities intended to be calming and soothing might have more generalized effects on aggression. For instance, Deguchi and colleagues (1999) used twice weekly night-time spa baths in a group of 10 dementia patients, four of whom engaged in aggressive behavior. Aggression was measured using a 5-point scale that indicated the severity of aggressive acts, with ratings being taken 10 times per day. An A-B-A design was used, with the baseline phase consisting of daytime baths. The intervention was considered effective for reducing aggression in all 4 participants. Only summary data pertaining to aggression was presented, however, so an analysis of trends during each phase is not possible. Also no statistical analysis was conducted, making the results of this study difficult to interpret.

Synder and associates (2001) implemented a glider-swing intervention with 30 nursing home residents with Alzheimer's disease who displayed aggression. The intervention involved having the participant and a research assistant spend 20 minutes on a glider swing each day for 10 consecutive days. An A-B-A design was utilized, with each baseline phase lasting 5 days. Nursing assistants reported the number of aggressive acts per shift using the Ryden Aggression Scale. Although reductions in aggression were observed, pre and post-treatment differences were not statistically significant. Furthermore, benefits that were found (e.g., improved emotions) were most pronounced 10 minutes after the intervention, thus being relatively short-lived.

Caregiver Training Interventions. Six studies have implemented caregiver training programs as a means for reducing aggression. For example, Hagen and Sayers (1995) implemented a 3-session educational program with a group of 134 nursing staff in a large extended care facility. Each session lasted 30 minutes and included information about the nature of dementia, risk factors for aggression, strategies to prevent aggression, strategies to de-escalate aggression, and protective interventions. Aggression was measured using a checklist similar to the Ryden Aggression Scale. Data was collected concerning each incident of aggression (i.e., when it occurred, the topography of the aggressive act, whether staff was injured, and during what activity aggression occurred). All participating staff collected data for eight days prior to the educational intervention and 8 days following completion of the intervention. A significant difference in aggressive acts was found from pre to post-intervention (182 incidents pre-intervention and 93 post-intervention). Furthermore, it was found that aggressive acts were most likely to occur during times of day when personal cares were completed.

Hoeffer and colleagues (1997) also implemented a staff education intervention with three nursing assistants and ten dementia patients. The intervention involved bedside consultation and developing individualized bathing programs for each resident. Individualized bathing plans involved changing the physical environment (e.g., temperature of water) and psychosocial environment (e.g., attending to the needs and preferences of the resident). Three to six pre-intervention baths were completed for each resident and five post-intervention baths were completed (3 at one-month post-intervention 2 at six-month post-intervention). Physical and verbal aggression was measured using the Ryden Aggression Scale. Statistically significant reductions were observed in both physical and verbal aggression from pre to post- intervention.

Gormley, Lyons, and Howard (2001) completed a 4-week community-based training program for family caregivers designed to reduce aggressive behaviors. The training program involved four primary components: avoidance or modification of precipitating and maintaining factors, use of appropriate communication techniques, validating and accepting false statements, and using distraction when the patient was aggressive. A total of 62 individuals were randomly assigned to either the training program or a control group that consisted of discussions with caregivers and patients about caregiving issues. Aggression was measured using the Rating Scale for Aggressive Behavior in the Elderly. At follow-up there were no significant differences in aggression between the experimental and control groups.

Savage et al. (2004) trained psychiatric hospital staff to recognize agitated behaviors and implement "evidence-based psychosocial interventions" to address agitated behavior (these interventions were not specified further). It was reasoned that if agitated behaviors were adequately addressed, aggressive acts and assaultive behaviors could be reduced. A total of 10 individuals participated; however, only 8 completed the study. Aggression was measured using the CMAI and incident reports of aggression. No significant reductions in aggression were found on the CMAI, although there was a significant reduction in reported assaults during the 36 weeks after the start of the intervention compared to the 36 weeks prior to beginning the intervention.

One of the largest studies in the area of aggression and dementia was conducted by Sloan and colleagues (2004). They conducted a multi-site randomized control trial with 69 persons with dementia and 37 nursing assistants. The study included three groups: a usual care control group, towel-bathing, and person centered-showering. Nursing staff in the two experimental groups were trained in person-centered showering (or towel bathing), which is a set of procedures that involves developing an individualized plan using techniques such as providing choices, distraction, using preferred bathing products, modifying the shower spray, and maintaining warmth and privacy as much as possible. Staff training required approximately 8 hours and consisted of didactics, hands-on supervision, and watching videos to identify behavioral symptoms and their antecedents. After an initial intervention phase of 6 weeks, treatments were switched and implemented for additional 6 weeks (i.e., institutions using person-centered-showering switched to using towel bathing and vice versa). Verbal and physical aggression was measured using a computerized behavioral observation system called the Care Recipient Behavior Assessment. Significant reductions in aggression were found in the two interventions, but not in the control group. In addition, there were no significant differences between the two interventions, but both interventions were significantly better than the control group.

Baker, Hanley, and Mathews (2006) trained a certified nursing assistant in a special care unit to implement a functional analysis and function-based intervention with a 96-year-old woman with Alzheimer's disease who engaged in aggressive behavior (i.e., hitting) during toileting. The functional analysis revealed that aggression served an escape function. Therefore, the staff member was trained to use non-contingent escape (NCE) as an intervention. NCE involved providing the participant with a 10-second break every 20 seconds during toileting routines. Using an A-B-A-B design, the authors found that reductions in aggression were associated with implementation of the NCE intervention. This study represents one of the few attempts to experimentally assess the function of aggressive behavior in dementia patients and then implement a function-based intervention.

Other Interventions. Two other studies investigated interventions that did not fall into one of the previous categories. Wisner and Green (1986) implemented a set of cognitive-behavioral interventions with a 73-year-old male with multi-infarct dementia who engaged in verbal outbursts. The intervention involved four components: contingency managements (i.e., he could not eat meals in the dining room if he acted out), teaching the resident that he could go to his room when he was upset, self-monitoring of anger, and six 1-hour individual therapy sessions that involved discussing alternative reactions and solutions. When verbal outbursts occurred, staff recorded it and rated the severity of anger on a 10-point Likert-type scale. An A-B design was implemented with a 2-week baseline and a 6-week intervention period. Results showed that the number of outbursts declined from 11 in fourteen days during baseline to 4 in forty-two days during the intervention. The severity of outbursts also appeared to decline. A three-month follow-up revealed that only one outburst occurred since the end of treatment.

Moniz-Cook, Woods, and Richards (2001) reported the use of a variety of interventions with five individuals with dementia (four engaged in physical and/or verbal aggression). Interventions were based on a functional analysis of each participant's behavior. Functional analyses involved observation of the individual's behavior and an investigation of the individual's premorbid history (e.g., likes/dislikes, culture, and personality). Based on this assessment data, hypotheses about the function of the behavior were formed, appropriate function-based interventions were devised, and the effectiveness of interventions was tested using a variety of different single-subject research designs (e.g., A-B-A-B and A-B-C-D-A). In each case, the frequency of aggression was measured through direct observation. As an example of this process, one resident was observed to be aggressive when a staff member wore a green coat. Because the resident had been a local fisherman and green was considered unlucky among this community (it is associated with death), it was hypothesized that aggression was triggered by the color green and that "aggression" might have actually represented an effort to "protect" those who wore green from death. Several conditions were devised that involved having research assistants wear green and other colors when interacting with the participant and conducting activities with green objects. Results indicated that the participant only was aggressive when others wore green. Thus, it was recommended that staff not wear green and redirect the participant when other residents wore green in his presence. This same set of procedures was repeated with the other participants whose aggressive behavior all appeared to be related to superstitious beliefs that were strongly held prior to developing dementia (e.g., fears of broken mirrors, avoidance of the color black). This and other studies (Middleton, Richardson, & Berman, 1997) support the importance of assessing the individual's history as a means for better understanding possible antecedents of aggressive behavior and developing individualized interventions. This also speaks to the importance of involving family in treatment planning given their extensive pre-morbid history with patients.

Summary of Current Literature

Because of the small number of studies that have explicitly addressed aggression in this population, it is difficult to make definite conclusions about any of the types of interventions discussed above. However, some tentative conclusions can be drawn concerning the current literature. First, there is currently little evidence that bright light therapy is effective for reducing aggression. Also, simply engaging residents in social, physical or calming activities has generally produced disappointing results, although Holmberg's (1997) study utilizing a walking group is an exception.

On the other hand, it appears as if using distractors such as preferred music during hands-on cares is a promising intervention. It also appears that choosing distractors that are tailored to individual preferences may increase the effectiveness of distraction-based interventions. Distraction-based interventions have the advantage of being relatively low-cost and they may require little staff training.

Caregiver training programs also show promise, but additional research is necessary due to the relatively small number of studies that have been conducted that specifically target and measure aggression as opposed to the broader construct of "agitation". Comprehensive programs that involve teaching professional nursing home staff a wide variety of behavior management skills for preventing aggression have proven effective in four studies (Hagen & Sayers, 1995; Hoeffer, et al., 1997; Savage, et al., 2004; Sloan, et al., 2004). In addition, one study demonstrated that training a professional caregiver to conduct a functional analysis and implement a specific function-based intervention was effective in reducing aggression during toileting. No significant effects were found in one study that trained family caregivers in the use of a number of behavior management skills (Gormley, et al., 2001).

Finally, there is a limited amount of data to support other interventions. Teaching cognitive-behavioral skills has produced beneficial results in one study that involved one resident (Wisner & Green, 1986). This is a promising outcome, but teaching patients self-management skills is likely to be effective only for those in the early stages of dementia. Moniz-Cook and colleagues (2001) demonstrated that conducting an in-depth assessment of variables functionally-related to aggressive behavior and developing interventions based on these assessments reduced aggression in 4 individuals.

Critiques and Suggestions for Future Research.

Given the prevalence and importance of the problem of aggression, the general paucity of empirical research addressing this issue is unexpected. The current literature offers some encouraging findings and some innovative approaches for managing this difficult problem. Other studies not meeting criteria for inclusion in this review also describe some creative approaches that may be useful for practitioners (e.g., Middleton, Richardson, & Berman, 1997; Woods & Ashley, 1995). There are a number of limitations existing in the current literature, however, that must be addressed in future research so a more solid base of empirical work can be created in order to guide clinical practice.

Targets of interventions. Although it is surprising that such a small number of studies have been devoted to examining the effects of non-pharmacological interventions for aggression, it is even more surprising that only six studies have specifically targeted aggression during hands-on caregiving tasks (bathing, toileting, and dressing). Aggression is most likely to occur during these tasks in this population (Colenda & Hamer, 1991; Hagen, & Sayers, 1995) and is a significant source of stress for caregivers (Everitt, et al., 1991). It is also one of the most frequently noted problems faced by nursing staff (Hagen & Sayers, 1995; Whall, Gillis, Yankou, Booth, & Beel-Bates, 1992) and is a problem very likely to be presented to psychologists and/or behavior analysts that serve as consultants in long-term care facilities. Therefore, there is a clear need for additional empirical research addressing the problems associated with aggressive behavior that occurs during intimate caregiving situations.

Design limitations. A variety of experimental designed were utilized in these 18 studies including traditional between group or pre-post designs (n=11), single-subject experimental designs that involved aggregating data across all participants and comparing baseline and treatment condition (n=4), and single-subject experimental designs that included the presentation of data for each participant (n=3). Of the eleven group design studies, eight included adequate sample sizes, while three studies included small sample sizes of 8-10 participants. Many also utilized control conditions and random assignment to conditions (two did not use random assignment and one study did not specify whether participants were randomly assigned).

Furthermore, the three studies that utilized single subject designs with individual participants included adequate phase changes to demonstrate experimental control and although one study used an AB design, it was appropriate given the nature of the intervention (the cognitive-behavioral intervention used with the patient could not be withdrawn). However, further replications of the procedures used in these studies are necessary to determine if the effectiveness of these procedures generalizes across subjects and settings.

Methodological problems occurred in some studies that utilized single-subject designs that involved aggregating data across all participants. For example, all but one study included small sample sizes of less than 14 residents (one study had an n=30). One study conducted a baseline of only 2 observations, which does not provide an adequate baseline by which to compare treatment effects. In addition, all four of these studies used either an A-B or A-B-A design and then compared phases using statistical analyses. Although these are very useful designs and statistical analyses are appropriate methods for comparing conditions, adding another treatment phase to an A-B-A design allows for a more definitive demonstration of experimental control.

Limitations in measurement. Several measurement concerns are present in the literature reviewed here. First, over half (n=11) of the studies included only self-report measures of aggression. Sole reliance on self-report measures completed by nursing staff can be subject to recall biases and may not accurately capture different dimensions (e.g., frequency and intensity) of aggression, particularly during hands-on caregiving tasks. For example, staff naturally will selectively recall acts of aggression, even if these acts occur relatively infrequently compared to baseline. In other words, staff may overestimate the occurrence of behavior simply because they recall acts of aggression that do occur more easily, particularly with residents with whom they have had difficulties in the past. Therefore, measures relying on staff reports may be less sensitive to changes in aggressive behavior resulting from intervention programs. Direct observation measures using non-participant observers tend to remove this form of bias, but have been utilized infrequently in this literature (n=5). Although direct observation can be more resource-intensive, it provides a very useful adjunct to self-report measures and removes certain biases that are inherently present in self-report measures completed by those familiar with the patient. Overall, it is recommended that future studies utilize multiple measures using multiple methods to increase the likelihood of detecting meaningful treatment effects, offset problems associated with any one measurement method, and allow for more definitive judgments about the effectiveness of an intervention. For instance, future studies should consider using not only well-validated self-report questionnaires such as the CMAI, but also other measurement methods such as direct observation, institutional incident reports and rating scales designed to measure the severity/intensity of aggression (see Sloane, et al., 2004 for an excellent example of the use of multiple methods of assessment).

Lack of social validity measures. When a behavior or set of behaviors is challenging and potentially dangerous, it is particularly important to assess the social validity of intervention programs. Wolf (1978) explained that social validity involves the assessment of: 1) the social significance of treatment goals; 2) social appropriateness or acceptability of interventions; and 3) the social importance of the effects of interventions. It seems clear that aggression is a socially significant goal. Measurement of the acceptability of interventions and the significance of results, however, is rarely completed yet very important. For instance, if we want staff to continue implementing procedures once studies are completed, researchers need feedback about the acceptability of intervention procedures. If staff finds a procedure too time-consuming or laborious, they may stop using the intervention once the researchers are gone.

Related to this, if staff does not notice treatment effects, interventions may not be sustained. For example, results of a study can be statistically significant, yet not be clinically important or even noticeable to direct care staff. It would seem crucial to know if staff perceives changes in resident behavior because this change in behavior may be an important reinforcer for continued use of an intervention. The measurement of social validity seems particularly important in long-term care settings where staff is very busy, overworked, and underpaid. Intervention programs that are brief, easy to implement, and potent are much more likely to be accepted, implemented appropriately, and maintained over time. Relatively simple self-report measures completed by participating staff can provide valuable information.

Four studies measured variables related to social validity. For example, three studies showed that bathing interventions added relatively little time to the length of baths (Hoeffer et al., 1997; Sloan, et al., 2004; Whall, et al., 1997). Whall and associated also mentioned that "aides in the treatment group enjoyed the natural experience (the experimental intervention) and their affect also became more positive in the opinion of the study RA." Similarly, Hagen and Sayers (1995) reported qualitative data such as "they (staff) felt pleased with the program." Hoeffer and colleagues (1997) assessed social validity more formally. These researchers used a measure called the Assessment of Bathing Experience to assess nursing assistant's perceptions of the resident and the experience of caregiving during baths. They found that staff reported bathing as being less frustrating and frightening and they perceived residents as calmer, less upset, and less aggressive. More studies need to incorporate formal measures of treatment acceptability and staff affect as was done by Hoeffer (1997).

Lack of follow-up. Follow-up assessment is particularly important when intervening with clinically important behavior such as aggression. Related to the measurement of social validity, follow-up assessment allows researchers to determine if interventions continue to be implemented and if effectiveness is maintained over time. It is reasonable to expect that the effectiveness of interventions may change over time in this population given the progressive nature of conditions such as Alzheimer's disease. Only four of the studies reviewed here, however, mentioned any kind of follow-up assessment (Deguchi et al., 1999; Hoeffer, et al., 1997; Moniz-Cook, et al., 2001; Wisner & Green, 1986). These follow-ups tended to be in the ranged from 2 weeks to 2 years.

Lack of individualization. Studies utilizing distractors and activities have at times failed to systematically assess the preferences of individual patients (e.g., Sival, et al., 1997; Whall et al., 1997). In studies utilizing music as a distractor, preference is usually determined by asking family members. It is possible, however, that family opinions concerning preferences may not be accurate given the significant changes that occur in individuals as a result of having dementia. The failure to individualize interventions to a patient's specific preferences may account for negative findings in some studies. Future research in this area may benefit from incorporating stimulus preference methods (e.g., Pace, Ivancic, Edwards, Iwata, & Page, 1985) as a means for directly determining patient preferences. For example, future studies could examine if music/activities chosen by the patient by means of stimulus preference assessment are more effective than music/activities chosen by others.

There are additional possibilities for future research regarding the use of music or other distractors during hands-on caregiving tasks. For example, future studies could periodically repeat stimulus preference assessments (e.g., doing a brief stimulus preference involving two pieces of music prior to each caregiving session), which may reduce the likelihood of habituation and serve to maintain effectiveness over time. Also, only one study (Whall, et al., 1997) specified any staff training to accompany the use of music or other distractors. Although turning on a CD player to play music or other sounds requires essentially no training, how staff orient the patient to the music could potentially be very important. For instance, if staff were to explicitly orient the patient to music ("isn't that music beautiful"), sing, encourage the patient to sing or hum, or sway the patient's hands to the music, the intervention may be more effective than if music or sound is simply played in the background. In this sense, the use of music could significantly change the quality of the patient-caregiver interaction, making bathing less stressful for both parties. Overall, the use of music and other distractors during caregiving tasks is a promising intervention, but additional research is needed to determine its effectiveness and to further specify how to best implement this intervention.

Lack of functional analysis. From a behavioral perspective, interventions for behavioral problems will presumably be more effective if they address variables functionally-related to aggressive behavior. Interventions utilized in some of the studies reviewed here (e.g., use of glider swings, programs of diverse activities, music therapy groups, bright light) may have been unsuccessful because interventions did not address the function of aggression for a given resident. In fact, only two studies in the current review conducted an assessment of the function of aggressive behavior (Baker et al., 2006; Moniz-Cook et al., 2001). Baker, Hanley, and Mathews (2006), for example, nicely demonstrated how aggression during hands-on caregiving tasks can serve an escape function and that providing opportunities for escape on a regular basis could reduce aggressive behavior. Moniz-Cook, Woods, and Richards (2001) showed how antecedent variables such as the color of clothing worn by staff and residents can reliably evoke aggressive behavior and that understanding of an individual's pre-morbid history can provide clues as to the function of behavior. Aggression could likely be maintained by several variables (e.g., escape from aversive tasks such as hands-on care, attention from staff, increasing or decreasing environmental stimulation). Likewise, any host of antecedent variables could be related to aggression such as gender or ethnic background of staff or other residents, communication style of staff, temperature of bathing areas, high or low levels of environmental stimulation in a given environment, or the number of staff simultaneously providing hands-on care at a given time. Therefore, future studies should consider incorporating some form of functional analysis and more explicitly tie intervention strategies to the presumed function of behavior for an individual patient.

Summary and Conclusions

Physical and verbal aggression is prevalent problem in persons with advanced dementia and is associated with numerous negative consequences such as over-medication, injury, and caregiver stress. Traditional management strategies involving anti-psychotic medications have limited efficacy and can have dangerous side effects. Non-pharmacological approaches represent restraint-free alternative to medications, but a there is a relatively small literature devoted to empirically investigating these approaches. Distraction-based methods and teaching caregivers behavior management strategies appear to be promising approaches. Behavior analytic assessment and intervention strategies have rarely been utilized to address this problem, but the few available studies suggest this approach can be effective. Behavior analysts likely have much to offer in terms of developing individually-tailored care plans through the use of behavioral assessment techniques (e.g., functional analysis, stimulus preference assessment) and treatment approaches (e.g., noncontingent reinforcement, embedding). It is clear, however, that much more research is necessary to address how to best determine which intervention strategies will be effective with which patients and how these strategies should change as dementing conditions progress.


American Psychiatric Association. (2004). Diagnostic and statistical manual of mental disorders--text revision (4th ed.). Washington, DC: Author.

Ancoli-Israel, S., Martin, J.L., Gehrman, P., Shochat, T., Corey-Bloom, J., Marler, M., Nolan, S., & Levi, L. (2003). Effect of light on agitation in institutionalized patients with severe Alzheimer's disease. American Journal of Geriatric Psychiatry, 11, 194-203.

Baker, J.C., Hanley, G.P., & Mathews, R.M. (2006). Staff-administered functional analysis and treatment of aggression by an elder with dementia. Journal of Applied Behavior Analysis, 39, 469-474.

Clark, M.E., Lipe, A.W., & Bilbrey, M. (1998). Use of music to decrease aggressive behaviors in people with dementia. Journal of Gerontological Nursing, 24, 10-17.

Colenda, C.C. & Hamer, R.H. (1991). Antecedent and interventions for aggressive behavior of patients at a geropsychiatric state hospital. Hospital and Community Psychiatry, 42, 287-292.

Deguchi, A., Nakamura, S., Yoneyama, S., Hamaguchi, H., Kawamura, Y., Horio, S., Nishimoto, Y., Saito, Y., & Deguchi, K. (1999). Improving symptoms of senile dementia by a night-time spa bathing. Archives of Gerontology and Geriatrics, 29, 267-273.

Deutsch , L. H., Bylsma, F. W., Rovner, B. W., Steele, C., Folstein, M. F. (1991). Psychosis and physical aggression in probable Alzheimer's disease. American Journal of Psychiatry. 149(9), 1159-1163.

Everitt, D.E., Fields, D.R., Soumerai, S.S., & Avorn, J. (1991). Resident behavior and staff distress in the nursing home. Journal of the American Geriatrics Society, 39, 792-798.

Gates, D., Fitzwater, E., Telintrlo, S., Succop, P., & Sommers, M.S. (2004). Preventing assaults in nursing home residents: Nursing assistants' knowledge and confidence--a pilot study. Journal of the American Medical Directors Association, 5, S16-21.

Gormley, N., Lyons, D., & Howard, R. (2001). Behavioural management of aggression in dementia: A randomized controlled trial. Age and Ageing, 30, 141-145.

Haffmans, P.M., Sival, R.C., Lucius, S.A., Cats, Q., & van Gelder, L. (2001). Bright light therapy and melatonin in motor restless behavior in dementia: A placebo-controlled study. International Journal of Geriatric Psychiatry, 16, 106-110.

Hagen, B.F. & Sayers, D. (1995). When caring leaves bruises: The effects of staff education on resident aggression. Journal of Gerontological Nursing, 21, 7-16.

Hamel, M., Gold, D. P., Andres, D., Reis, M., Dastoor, D., Grauer, H., & Bergman, H. (1990). Predictors and consequences of aggressive behavior by community-based dementia patients. The Gerontologist. 30, 206-211.

Hoeffer, B., Rader, J., McKenzie, D, Lavelle, M., & Stewart, B. (1997). Reducing aggressive behavior during bathing cognitively impaired nursing home residents. Journal of Gerontological Nursing, 23, 16-23.

Holmberg, S.K., (1997). Evaluation of a clinical intervention for wanderers on a geriatric nursing unit. Archives of Psychiatric Nursing, 11, 21-28.

Lyketosos C. G., Steinberg, M., Tschanz, J.T., Norton, M. C., Steffens, D. C., Breitner, D. C. (2000). Mental and behavioral disturbances in dementia: findings from Cache County Study on memory in aging. American Journal of Psychiatry. 157, 708-714.

Middleton, J.I., Richardson, S.J., & Berman, E. (1997). Assessment and intervention study of aggressive behavior in cognitively impaired institutionalized elderly. American Journal of Alzheimer's Disease, 12, 24-29.

Moniz-Cook, E., Woods, R.T., & Richards, K. (2001). Functional analysis of challenging behaviour in dementia: The role of superstition. International Journal of Geriatric Psychiatry, 16, 45-56.

Pace, G. M., Ivancic, M. T., Edwards, G. L., Iwata, B. A., & Page, T. J. (1985). Assessment of stimulus preference and reinforcer value with profoundly retarded individuals. Journal of Applied Behavioral Analysis, 18, 249-255.

Paveza, G. J., Cohen, D. Eisdorfer, C., Freels, S., Semla, T., Ashford, W., Gorelick, P., Hirschman, R., Luchins, D., & Levy, P. (1992). Severe family violence and Alzheimer's disease: prevalence and risk factors. The Gerontologist. 32(4), 493.

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). Goal directed care. Caring for aggressive nursing home residents with dementia. Journal of Gerontological Nursing. 18(11), 35-42.

Savage, T., Crawford, I, & Nashed, Y. (2004). Decreasing assault occurrence on a psychogeriatric ward. Journal of Gerontological Nursing, 30, 30-37.

Schneider, L.S., Dagerman, K.S. & Insel, P. (2005). Risk of death with atypical antipsychotic drug treatment for dementia: Meta-analysis of randomized placebo-controlled trials. Journal of the American Medical Association, 294, 1934-1943.

Schreiner, A. S. (2001). Aggressive behaviors among demented nursing home residents in Japan. International Journal of Geriatric Psychiatry. 16, 209-215.

Sink, K.M., Holden, K.F., & Yaffe, K. (2005). Pharmacological treatment of neuropsychiatric symptoms of dementia: A review of the evidence. JAMA, 293, 596-608.

Sival, R.C., Vingerhoets, R.W., Haffmans, P.M., Jansen, P.A.F., & Ton, H.J.N. (1997). Effect of a program of diverse activities on disturbed behavior in three severely demented patients. International Psychogeriatrics, 9, 423-430.

Sloane, P. D., Mathew, L. J., & Scarborough, N. (1991). Physical and pharmacological restraint of nursing home patients with dementia. Journal of the American Medical Association, 265, 1278-1282.

Sloane, P.D., Hoeffer, B., Mitchell, M.C., McKenzie, D.A., Barrick, A.L., Rader, J., et al. (2004). Effect of person-centered showering and the towel bath on bathing-associated aggression, agitation, and discomfort in nursing home residents: A randomized, controlled trial. Journal of the American Geriatrics Society, 52, 1795-1804.

Snyder, M., Tseng, Y., Brandt, C., Croghan, C., Hanson, S., Constatine, R., & Kirby, L. (2001). Glider swing intervention for people with dementia. Geriatric Nursing, 22, 86-90.

Svansdottir, H.B. & Snaedal, J. (2006). Music therapy in moderate and severe dementia of Alzheimer's type: A case-control study. International Psychogeriatrics, 18, 613-621.

Swearer, J. M., Drachman, D. A., O'Donnell, B. F., & Mitchell, A. L. (1988). Troublesome and disruptive behavior in dementia: Relationship to diagnosis and disease severity. Journal of the American Geriatrics Society, 36, 784-790

Thomas, D.W., Heitman, R.J., & Alexander, T. (1997). The effects of music on bathing cooperation for residents with dementia. Journal of Music Therapy, 34, 246-259.

Whall, A.L., Black, M.E., Groh, C.J., Yankou, D.J., Kupferscmid, B.J., & Foster, N.L. (1997). The effect of natural environments upon agitation and aggression in late stage dementia patients. American Journal of Alzheimer's disease, 12, 216-220.

Whall, A.L., Gillis, G.L, Yankou, D., Booth, D.E., & Beel-Bates, C.A. (1992). Disruptive behavior in elderly nursing home residents: A survey of nursing staff. Journal of Gerontological Nursing, 18, 13-17.

Wisner, E. & Green, M. (1986). Treatment of a demented patient's anger with cognitive-behavioral strategies. Psychological Reports, 59, 447-450.

Wolf, M.M. (1978). Social validity: The case for subjective measurement.

Journal of Applied Behavior Analysis. 11, 203-214.

Woods, P. & Ashley, J. (1995). Simulated presence therapy: Using selected memories to manage problem behaviors in Alzheimer's disease patients. Geriatric Nursing, 16, 9-14.

Zarit, S.H. & Zarit, J.M. (2007). Mental disorders in older adults: Fundamentals of assessment and treatment (2nd ed.). New York: Guilford Press.

Zimmerman, J. G., Watson, N., & Treat, A. (1984). Behavioral problems among patients of skilled nursing facilities. American Journal of Public Health. 74, 1118-1121.

Author Contact Information:

Jeffrey A. Buchanan, Ph.D.

Assistant Professor of Psychology

23 Armstrong Hall

Minnesota State University, Mankato

Mankato, MN 56001


Angela M. Christenson, B.A.

Clinical Psychology Graduate Student

23 Armstrong Hall

Minnesota State University, Mankato

Mankato, MN 56001


Carly Ostrom, B.A.

Clinical Psychology Graduate Student

23 Armstrong Hall

Minnesota State University, Mankato

Mankato, MN 56001


Nicole Hofman, B.A.

Clinical Psychology Graduate Student

23 Armstrong Hall

Minnesota State University, Mankato

Mankato, MN 56001

COPYRIGHT 2007 Behavior Analyst Online
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Buchanan, Jeffrey A.; Christenson, Angela M.; Ostrom, Carly; Hofman, Nicole
Publication:The Behavior Analyst Today
Article Type:Report
Date:Sep 22, 2007
Previous Article:Behavior analysis in the space age.
Next Article:Behavior analytic grounding of sociological social constructionism.

Related Articles
Spouses confront mild cognitive impairment.
A review of behavioral treatments for persons with dementia.
Disability in Alzheimer's disease: causes, consequences, and economic considerations.
Drugs used to treat Alzheimer's patients effective against dementia.
Exercise slows functional decline in nursing home residents with Alzheimer's disease.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters