Management of behavioral and psychological symptoms of dementia in long-term care facilities in Japan

Abstract

Background:  An increasing number of old people, and their medical requirements, cannot be managed by their families in their homes, which has been the traditional and prevalent practice in Japan. The number of people with dementia is increasing and behavioral and psychological symptoms of dementia (BPSD) make care difficult. In the present study, we investigated management techniques for BPSD in long-term care facilities in Japan by using the data mining method, which looks at the reported behaviors of care providers.

Methods:  First, interviews were conducted with 15 care providers to develop items for a questionnaire. These data were analyzed qualitatively and synthesized with criteria from the professional literature. The resulting self-report questionnaire on techniques used to manage different symptoms of dementia was completed by 275 care providers. We applied the association rule as a data mining method examining 15 management techniques related to 13 BPSD.

Results:  Analysis identified four types of management techniques: (i) emotional and behavior-concordance techniques; (ii) acceptance and supportive techniques; (iii) restraining techniques; and (iv) avoidance techniques. Different management techniques, and combinations of techniques, were found to vary in use and effectiveness with different BPSD.

Conclusions:  Good management techniques for many BPSD have been developed and are being implemented by care providers. The present study has the potential to inform researchers and care providers in Japan about the types of management techniques in current use, as well as areas of potential need for staff training.

INTRODUCTION

Japan is grappling with the consequences of an aging population and a low birth rate. An increasing number of old people, and their medical requirements, cannot be managed by their families in their homes, which has been the traditional and prevalent practice in Japan. Consequently, the Japanese government established Long-Term Care Insurance (LTCI) in April 2000, and revamped it in 2004 and 2006. According to the Ministry of Health, Labor and Welfare,¹ in 2006 the older population requiring long-term care numbered 4.3 million; 820 000 of these people were residents in long-term care facilities and an estimated 80% showed behavioral and psychological symptoms of dementia (BPSD). The Japanese Society for Dementia Care² advocated standardizing care for BPSD, including creating an environment that will not induce or increase BPSD in the resident population and properly supervising people with dementia (PWD). However, the Society did not offer guidelines for actual care models.

There are several approaches for managing BPSD, such as medication and cognitive rehabilitation. Treatment with donepezil not only improves the cognitive dysfunctuion of Alzheimer's disease patients, but was also found to relieve BPSD.³ Among cognitive rehabilitation approaches, reality orientation⁴ and validation therapy^5–7 are well-known in Japan. Some studies support the effectiveness of reality orientation, as assessed by changes in the frequency of BPSD.⁸ Previously PWD were treated at hospitals, but now, to control costs, they are transferred from hospitals to long-term care facilities or to home care. However, long-term care institutions had minimal experience caring for people with BPSD and people providing home care usually do not have any training in how to deal with BPSD. In Japan, care providers at long-term facilities for old people consist of nursing staff (registered nurses and assistant nurses) and care aides assisting people with activities of daily living, but not healthcare itself. There are two levels of care aides: certified care workers and care attendants. Care providers (CPs) in Japan have a diversity of educational backgrounds, licenses, and care abilities. We define CPs as both nursing staff and care aides. Some CPs have sophisticated techniques for managing BPSD in daily practice, even if they are not aware of this or trained in these techniques.

Behavioral and psychological symptoms of dementia increase work for staff and complications for patients.⁹ Behavioral and psychological symptoms of dementia such as wandering, resistance to care, agitation and aggression cause difficulties in caring for patients, as well as lower job satisfaction, contributing to burnout in CPs.^10–13 The emotions that CPs feel in interpersonal situations may influence their intentions and behavior in caring for PWD.¹⁴ Previous studies noted that these difficulties may lead to elder abuse¹⁵ and an increased use of physical restraints.^16–18 Tanaka et al. reported that there abuses occur in approximately 30% of nursing homes.¹⁹ Therefore, we clarified in our previous study that it is important that CPs should not view BPSD primarily as negative behavior, but should realize that these symptoms occur due to the disturbed peace of mind and body of PWD.²⁰

For patients, BPSD are apt to be aggravated by the catalytic action of dementia itself and the prevalence of BPSD in the social environment, so maintaining the peace of mind of PWD is necessary in order not to increase BPSD.²⁰ Good management should prevent aggravation²¹ and minimize the confusion of PWD.²⁰

In the present study, we investigated the techniques that CPs report using in daily practice that are effective in dealing with BPSD. Our aims were to: (i) identify management techniques used by CPs for BPSD care in long-term facilities in Japan; and (ii) suggest effective management techniques for BPSD by using the association rule data mining method. More effective approaches for managing PWD may decrease BPSD, reduce the stress felt by CPs, and diminish the burden in long-term care facilities.²² Our priority was to examine what CPs report they actually do that is effective in managing BPSD as part of the care strategy for PWD, because these techniques are likely to be able to be implemented in an efficient way in long-term care facilities.

This article presents the results of two related studies that examined techniques for managing BPSD within long-term care facilities. Initial interviews were conducted to identify the management techniques used for BPSD. A qualitative analysis of the data gathered through these interviews led to the development of a questionnaire that was used with a larger CP sample to clarify the relationship between these management techniques and different specific BPSD.

METHODS

Study 1: Qualitative data pilot study

Participants

Interviews were conducted with CPs to clarify the characteristics of management techniques for BPSD in long-term care facilities. Participants were recruited using purposive and snowball sampling. Fifteen CPs were interviewed: seven were certified care workers, four were care attendants, three were registered nurses, and one was an assistant nurse. The number of years of experiences in dementia care ranged from a 6 months to 8 years.

Method

Semistructured interviews were conducted by the authors from January to May 2005 in long-term care facilities in suburban areas throughout Japan. All interviews were recorded with the consent of the participants. The data were analyzed using a qualitative analysis with the Grounded Theory Approach.²³

Study 2: Questionnaire survey of CPs

Participants

An anonymous questionnaire was distributed to 633 CPs employed at 42 long-term care facilities for old people in suburban Japan from January to February 2006. Questionnaires were returned by 292 CPs from 26 facilities, which included 12 nursing homes (NH), five health services facilities (HSF), three sanatorium-type medical care facilities (MCF), and six group homes for PWD (GH). Three questionnaires were not completed and 14 were completed by respondents who were not CPs, so these were excluded from analysis. The data from the remaining 275 CPs were analyzed. Of these 275 CPs, 32 were registered nurses (11.6%), 21 were assistant nurses (7.6%), 91 were certified care workers (33.1%), and 131 were care attendants (47.6%). The mean age of CPs was 35.0 ± 12.5 years and the percentage of women was 74.9%. The average bed occupancy rate in MCF was higher than in other facilities (MCF = 100.0%; NH = 93.5%; HSF = 92.0% ; GH = 91.9%). The ratio of PWD in GH was higher than in other facilities (PWD ratio: GH = 100%; NH = 75.4%; HSF = 68.0%; MCF = 91.0%).

Measures

The International Psychogeriatric Association defines BPSD as ‘Symptoms of disturbed perception, thought content, mood, behavior frequently occurring in patients with dementia’.²⁴ Furthermore, based descriptions in previous studies, we defined BPSD as ‘symptoms that cause difficulty in care and hence increase caregiver burden, and as a result have a negative impact on the relationship between caregiver and patient’.²⁵

In the present study, the Behavioral Pathology in Alzheimer's Disease Rating Scale^26,27 and the Dementia Behavior Disturbance Scale^28,29 were used in the selection of items that characterize BPSD. We also considered the items of the LTCI, which is used to evaluate behavior of Japanese people with dementia. Using this information, we determined 13 prevalent symptoms that comprise BPSD, such as ‘PWD repeat the same story or ask the same things of others’ and ‘PWD fight or are physically aggressive’. The difficulty of managing each BPSD was evaluated by the CPs using a visual analog scale ranging from 0 to 100. A higher score indicated greater difficulty in managing the BPSD. The present data showed high internal consistency (Cronbach's α = 0.88).

Fifteen management techniques for BPSD were derived from the qualitative data analysis of Study 1 (see Table 1) and from the professional literature.

Table 1.  Management techniques for behavioral and psychological symptoms of dementia

General management strategy	Specific management techniques for BPSD	Short title
BPSD, behavioral and psychological symptoms of dementia; PWD, people with dementia.
Emotional and behavioral-concordance techniques	1. Listen to their stories and go along with them	1. Listening and going along
	2. Share in what they want to do, at their request	2. Sharing behavior
	3. Try to satisfy their requests and demands	3. Satisfying requests
	4. Observe what they say and do, and protect them from danger	4. Guarding
	5. Provide care in a calm voice and manner	5. Providing care in a calm manner
Acceptance and supportive techniques	6. Construct a relationship between PWD and care providers	6. Forming relationships
	7. Try to divert their attention and requests to other things, such as a walk or talk, to change their mood	7. Diverting attention
	8. Tell them that they can feel relieved and secure because you are with them	8. Reassure
	9. Correct what they say or do wrong based on common sense	9. Correction
	10. Control symptoms with psychotropic drugs	10. Psychotropic drugs
Restraining techniques	11. Restrict their motions to prevent them from standing up from wheelchairs or chairs or falling down from beds	11. Restriction of movement
	12. Shut them up in their room	12. Confining to room
	13. Use tools for restricting limb and finger motions, such as overall clothes and mittens	13. Restrictive clothing
Avoidance techniques	14. Involve other care providers in care or ask for a substitute to avoid conflict	14. Coping with other care providers
	15. Pretend to be unaware of and ignore what they say and do	15. Ignoring

Analytical procedure for the association rule in the present study

We analyzed the survey data using the association rule, which is a data mining method, to uncover relationships between the management techniques for BPSD and the difficulty of managing specific BPSD. The association rule can be construed as filters that can be used to generate hypotheses for a more rigorous statistical analysis. Association rules were analyzed using Clementine, Version 10.1.³⁰ Fifteen management technique items were included as possible responses to the BPSD indicators. The difficulty posed by each BPSD was arranged as the consequence of the rule. We standardized the difficulty of each BPSD (mean = 0, SD = 1) at each facility to control for differences in mean scores. The difficulty of the BPSD was divided into three groups according to the standardized score (Z) as low (Z < −0.5), moderate (−0.5 ≤ Z < 0.5), and high (0.5 ≤ Z). We set the rule range (the ratio to fill antecedent for the whole) at a minimum with 1%, confidence ratio 5% and a lift value of 1. The lift value is measured as the ratio of the probability of antecedent and consequent occurring together to the probability of antecedent and consequent occurring independently. A lift value indicates how the consequent happens independently. In general, a lift value that is much higher than 1 is of more interest than values close to 1.^31,32

We selected rules when they were related to the actual BPSD and excluded rules such as management techniques that may threaten the peace of mind and body of PWD and their environment. The rules for appetite/eating disturbances were a combination of the three types of management techniques: (i) listening and going along; (ii) guarding; and (iii) care provided in a calm manner. However, listening and going along and guarding may raise the risk of aspiration and suffocation in PWD with appetite/eating disturbances. Therefore, in this situation we excluded these types of techniques and used one rule for one type of technique: dealing with PWD in a calm manner.

Moreover, we regarded there were no actionable rules for behaviors where there were many rules with a lift value of ≥1. Thus, no actionable rules were extracted for four BPSD: (i) repetitive questioning (35 rules extracted); (ii) physical aggression (71 rules); (iii) aberrant toilet behavior (20 rules); and (iv) the delusion of items being stolen or persecution (61 rules). That such large numbers of rules were extracted suggests that considerable effort has been directed to these BPSD and no rules were remarkably and specifically effective. We therefore did not adopt actionable rules for these BPSD.

Our research procedure was reviewed by the Ethics Committee of the Osaka University Graduate School of Medicine, Division of Health Sciences.

RESULTS

Study1

We identified four types of management techniques for BPSD: (i) emotional and behavioral-concordance techniques; (ii) acceptance and supportive techniques; (iii) restraining techniques; and (iv) avoidance techniques.

Emotional and behavioral-concordance techniques

We characterized these techniques as the CP's attempt to understand the feelings of PWD feelings and the meaning behind their behavior, and to respond accordingly. Understanding the mental status of PWD leads to respect of their human rights. These management techniques appear similar to parent–child interactions. The emotional and behavioral-concordance techniques includes the following three techniques (see Table 1): listening and going along; sharing behavior; and satisfying requests. For example, one CP said, ‘If people with dementia say they have lost something from their belongings, I look for it with them, or I generally do what they want me to do.

Acceptance and supportive techniques

This group of techniques is characterized by respecting the spontaneous behaviors of PWD while providing for their safety. While using these techniques, PWD are free to act unless they disturb or cause trouble for other residents or put themselves at risk. The CPs allow and supervise the actions and behavior of PWD, although the CPs do not necessarily approach the PWD. Acceptance and supportive techniques includes six techniques (see Table 1) as follows: guarding; care in a calm manner; forming relationships; diverting attention; reassurance; and correction. One CP described the use of these techniques as follows:

When PWD start wandering, I let them do what they want to. When PWD wander and enter into another resident’s room and cause trouble for them, I make them stay around us and within our line of vision. When one man's dementia worsened dramatically, we decided not to ignore his dementia progression, but to allow him to do what he wants, under our observation and we told his family and obtained their consent. After changing our way of caring, his dementia improved and he returned to what he used to be.

Another CP said:

When PWD say something over and over again and appear in a bad mood, I agree with them each time because they are concerned at that point in time. I do not become angry at their repetition. If I ask the person to stop, the symptoms may worsen or the BPSD may be further prolonged.

Restraining techniques

The use of physical restraint is a care strategy that limits the action of PWD to preserve their peace of mind and body. In the present study, we identified four items related to physical restraints (see Table 1): (i) psychotropic drugs; (ii) restriction of movement; (iii) confining the PWD to his/her room; and (iv) restrictive clothing. Since 1999, Japanese law has prohibited the use of physical restraints. However, from our pilot study it was clear that physical restraints were being used based on the legal exceptions of ‘urgency’, ‘non-fungibility’ and ‘temporariness’. The general reason for using restraints was explained by the lack of manpower at long-term care facilities. For example, one CP said:

Sometimes, we are obliged to restrain PWD but it is not what we want to do. I give PWD physical restraints to protect them from the risk of falling because of a lack of manpower. It may look abusive, but for the person's family, it would be a big issue if the person experiences trouble.

Avoidance techniques

The technique of avoiding risk or involving other care providers is described as CPs meeting the basic needs of PWD, such as feeding and toileting, but also trying to lessen any risks in providing care by avoiding situations or involving other staff. It is characterized by not approaching PWD in situations where the CP's safety should take precedence. The CPs consider this a technique for avoiding the escalation or resonance in the group setting of BPSD. This care includes two items (see Table 1), which are illustrated in the following statement made by a CP:

When physical aggression arises in PWD, I try my best to evade and avoid their strokes. Once a man swung a stick against me and ran after me, but I could not cope with his behavior by myself. He could not control himself, so I asked another staff member to get involved in his care. Other residents around us get anxious when the care staff has difficulty taking care of PWD with such extreme BPSD.

Study 2

Mean scores of management difficulty for each of the BPSD indicators and frequencies of use of the management techniques are given in Table 2. It can be seen that the difficulty of managing physical aggression was the highest at 68.7. Generally, the frequency of use of physical restraints (10–13) was low; the use of restrictive clothing for aberrant toilet behavior was 13.0%.

Table 2.  Frequency of use of different management techniques for each behavioral and psychological symptoms of dementia reported by care providers (n)

Example of BPSD	Short title	Mean score of difficulties of BPSD†	Management technique (as defined in Table 1)
			1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Values (%) show the frequency of each of the management techniques listed in Table 1 (1–15) chosen for each behavioral and psychiatric symptoms of dementia (BPSD). ^† This value ranges from 0 to 100, with a high score meaning that the care providers rated this BSPD as having a high level of difficulty to manage. Mean scores were calculated by excluding no answers.
Repeats the same story or asks the same questions (n = 267)	Repetitive  questioning	30.3	73.0	12.4	31.1	17.6	60.2	12.4	37.8	20.2	2.6	0.4	1.5	0.0	0.7	19.1	1.9
Walks or moves in wheelchair aimlessly (n = 262)	Wandering	33.8	9.9	30.9	12.6	76.0	22.9	6.1	30.2	7.3	1.5	0.0	5.3	0.4	1.5	18.3	4.6
Complains or accuses others (n = 257)	Complaints	45.5	60.3	5.8	35.4	10.5	50.6	15.2	29.2	8.2	8.2	0.0	0.0	0.4	0.8	25.3	1.9
Wakes during the night (n = 242)	Sleep  disturbances	45.4	12.8	16.9	14.5	60.7	23.1	5.4	14.5	26.4	5.0	1.7	6.2	0.8	0.8	19.0	3.3
Inappropriate sexual behavior or says improper things (n = 237)	Sexual  disinhibition	49.4	22.8	2.5	3.0	10.1	32.1	7.2	38.0	2.1	18.1	0.8	0.4	0.0	0.4	38.0	9.3
Reviles others (n = 249)	Verbal  aggression	51.2	38.6	2.4	8.8	15.3	56.5	11.2	14.9	5.2	15.7	0.4	0.0	0.8	0.8	25.7	10.0
Puts inedible things into mouth, such as tissues (n = 255)	Appetite/eating  disturbances	42.5	3.5	3.5	5.5	61.2	18.0	4.3	30.6	3.1	18.4	0.0	0.8	0.4	0.0	21.2	0.0
Resists advice or care (n = 258)	Resistance to care	51.7	27.9	6.2	10.9	16.3	46.1	24.8	12.8	14.3	6.6	0.4	0.8	0.8	0.4	51.2	0.8
Collects various things or takes them (n = 244)	Hoarding behavior	29.8	17.6	7.0	9.4	38.5	29.5	7.0	23.4	3.3	10.7	0.0	0.4	0.8	0.4	19.7	13.1
Says that ‘I want to go back home’, ‘I go to work’, and wants to leave the facility, or try to go out (n = 263)	Demands to go home	50.4	43.0	22.4	20.9	47.5	36.5	16.0	57.4	17.1	5.7	0.4	1.1	1.9	1.9	30.8	0.4
Fights or is physically aggressive, hits, bites, scratches, kicks, spits (n = 258)	Physical  aggression	68.7	16.3	4.7	9.3	33.3	70.2	10.1	21.3	10.5	19.4	1.9	2.7	1.2	5.0	61.6	2.3
Socially unacceptable, unsanitary toileting behavior (n = 239)	Aberrant toilet behavior	59.9	5.9	3.3	3.8	20.9	46.9	5.4	14.6	6.3	11.3	0.4	1.7	0.4	13.0	36.4	0.4
Feels persecuted or believes their things have been stolen (n = 256)	Delusion of being victimized	43.0	58.2	8.6	7.8	12.5	45.3	24.6	33.6	21.9	7.4	7.4	0.8	0.4	0.8	14.8	0.4

Association rules between BPSD management techniques and difficulty for each BPSD

Table 3 presents the types of management techniques commonly chosen for each BPSD indicator, rules with a high value of support and lift, excluded rules, and BPSD indicators without effective rules. The highest support value was 60.3% for ‘listening and going along’ for complaints (Confidence = 39.4%). Sixty-one of 155 CPs chose this type of technique, indicating that the difficulty of managing complaints was low.

Table 3.  Number of rules obtained for each behavioral and psychological symptoms of dementia and variables of association rules and relationship to management techniques as defined in Table 1

	No. rules	Support (%)	Confidence (%)	Lift	No. of antecedents	No. simultaneous antecedents and consequents	Emotional and behavioral-concordance techniques				Acceptance and supportive techniques					Restraining techniques				Avoidance techniques
							1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Association rules were shown when Lift is 1%, Confidence is 5%. Association rules were sorted by value of Lift. o, selected antecedent of association rule; x, selected antecedent of association rules, but finally it was excluded from rules because it threatened the peace of mind and body of people with dementia (PWD) and their environment. The behavioral and psychiatric symptoms of dementia defined as Repetitive questioning, Physical aggression, Aberrant toilet behavior and Delusion of being victimized in Table 2 are not included in the table above because they had no actionable rules.
Wandering	4	5.7	46.7	1.24	15	7	o				o
		22.9	43.3	1.15	60	26					o
		9.9	42.3	1.12	26	11	o
		17.2	42.2	1.12	45	19				o	o
Complains	2	50.6	40.8	1.12	130	53					o
		60.3	39.4	1.08	155	61	o
Sleep disturbance	9	3.3	62.5	1.78	8	5					o		o	o
		2.9	57.1	1.63	7	4		o			o		o	o
		2.9	57.1	1.63	7	4		o		o	o		o	o
		4.6	54.6	1.55	11	6				o			o	o
		7.4	50.0	1.42	18	9				o	o			o
		4.1	50.0	1.42	10	5			o	o			o
		14.1	44.1	1.26	34	25				o				o
		23.1	41.1	1.17	56	23					o
		13.6	39.4	1.12	33	13				o	o
Sexual disinhibition	1	22.8	53.7	1.70	54	29	o
Verbal aggression	2	8.0	50.0	1.58	20	10					o	o
		38.6	39.6	1.25	96	38	o
Appetite/eating disturbances	4	9.8	52.0	1.41	25	13				x	x
		18.0	45.7	1.24	46	21					o
		3.5	44.4	1.21	9	4	x
		61.2	39.7	1.08	156	62				x
Resistance to care	3	1.2	66.7	2.18	3	2							o		o
		2.7	57.1	1.87	7	4	o								o
		27.9	36.1	1.18	72	26	o
Hoarding behavior	4	1.3	100	2.82	3	3					o										o
		1.3	100	2.82	3	3	o														o
		2.9	71.4	2.02	7	5				o											o
		17.9	46.5	1.31	43	20	o
Demands to go home	4	1.1	66.7	1.91	3	2			o	o					o
		1.1	66.7	1.91	3	2			o	o			o		o
		8.0	47.6	1.36	21	10		o		o	o		o
		43.0	39.8	1.14	113	45	o

The highest lift values of rules were 2.82. Results indicated that there were two rules for hoarding behavior. We combined three types of management techniques for this, namely listening and going along, providing care in a calm manner, and ignoring the behavior. This suggests that the probability of the desired outcome increased threefold when these techniques were combined. Four rules consisting of combinations of three types of techniques were obtained as the management techniques for wandering, namely to supervise, listen and go along, guard the PWD, and to provide care in a calm manner.

The results indicated that two rules consisting of two types of techniques, namely listening and going along and providing care in a calm manner, were chosen where the difficulty of the complaints was perceived to be low. Nine rules consisting of combinations of six types of techniques were chosen for the management of the low difficulty BPSD of sleep disturbances, namely sharing behavior, satisfying requests, guarding, providing care in a calm manner, diverting attention and reassuring the PWD. Three rules consisting of combinations of three techniques were chosen for managing the low difficulty resistance to care by the PWD, namely listening and going along, diverting attention, and correction. Four rules consisting of combinations of seven techniques were chosen for the low difficulty BPSD of demanding to go home, namely listening and going along, sharing behavior, satisfying requests, guarding, providing care in a calm manner, diverting attention, and correction.

Certain combinations of techniques were effective in managing BPSD. Combined emotional and behavior-concordance and acceptance and supportive techniques were effective for wandering, complaints, sleep disturbance, verbal aggression, resistance to care and demands to go home. Emotional and behavior-concordance techniques were effective with sexual disinhibition. Acceptance and supportive technique were effective for appetite/eating disturbances. Avoidance and involving other CPs along with techniques of emotional and behavior-concordance and acceptance and supportive techniques were effective for hoarding behavior.

DISCUSSION

This study presents management techniques for BPSD reported by CPs at long-term care facilities in Japan. The data mining method showed that management techniques differed for different BPSD and were related to the perceived difficulty of managing the specific BPSD.

Consider the techniques that are used with wandering. We had the impression that the CPs recognized through trial and error that watching PWD wandering without acting to control their behavior minimized the BPSD and that if they tried to stop the PWD from wandering, the BPSD would worsen. However, wandering also influenced the quality of life of other residents when PWD entered other residents' rooms when wandering.³³ Therefore, CPs prevented trouble between residents by managing wandering mainly when the PWD entered another resident's room.²⁰ Moreover, CPs paid close attention to protect PWD from falling, because PWD are unaware of the risk of falling.

Techniques for managing complaints suggest that CPs valued the ideas and behavior of PWD, which corresponds to the concept of person-centered care suggested by Kitwood.³⁴ Resistance to care could be the result of misunderstandings due to cognitive dysfunction.³⁵

The CPs cannot comply with demands to go home, therefore they used several techniques to manage PWD with this BPSD. Correction was commonly selected, along with other types of techniques, because correction should not be offered alone to PWD.

Emotional and behavioral-concordance techniques were a preferred strategy for managing difficult BPSD for most of the conditions that were examined in the present study, which suggests its importance and value among management techniques. However, these techniques require time and effort from CPs.

Needless to say, we do not accept physical restraint because it is obviously an abuse of human rights. Therefore, we think CPs should avoid this procedure and always seek an alternative when caring for PWD. However, the results indicate that physical restraints are being used in these facilities, even though this is legally restricted. In April 2004, the US Food and Drug Administration (FDA) issued a warning stating that the use of atypical antipsychotics in the elderly increased their mortality rate. In accordance with this warning, it was widely announced in Japan that antipsychotics were not indicated as a treatment for dementia.³⁶ We must recognize that the use of psychotropic drugs is ‘off-label use’, meaning the prescription of a drug without approved official authorization. Ihara and Arai pointed out that a trial of these drugs should always be performed under careful clinical monitoring.³⁷ Physical restraint never appeared in the active rules, suggesting that this technique does not reduce the difficulty of managing BPSD.

There were no consensus rules for managing repetitive questioning, sleep disturbances, hoarding behavior and physical aggression. This suggests that CPs tried various types of techniques for these BPSD through trial and error. Further research should evaluate the techniques used for BPSD from multiple perspectives looking at the incidence of BPSD, the daily activities of PWD, and the cost of time and effort.

The present study is limited by the lack of observational data, because the CPs reported what they would choose to do, but what they actually do was not verified independently. When people have symptoms similar to BPSD, we have to consider the background of the symptoms.³⁸ The BPSD differ according to age,³⁹ type of dementia,^40,41 and the severity of the dementia.⁴² Therefore, we need to consider these factors in future studies.

Although there were associations between the perceived difficulty of managing each BPSD and the associated techniques that were used, we were unable to demonstrate any cause-and-effect relationships. It is known that difficulty in managing the different BPSD does not concur with the frequency of the symptoms.⁴³ Therefore, we need to develop a better definition of what is ‘low difficulty’.

We used the data mining method to explore the techniques for low difficulty BPSD and obtained some active rules. However, these rules should be examined for their effectiveness in future intervention studies. We need to seek a better consensus on techniques effective for managing physical aggression, repetitive questioning, aberrant toilet behavior, and the delusion of property being stolen. In these areas, CPs' management behavior reflects trial and error, and the best choice of management techniques based on the difficulty of managing these BPSD has not been established.

The association rules of the data mining method clarified that the management techniques differ for different BPSD. This method appears to be an efficient tool for identifying the relationships between the techniques and BPSD, which traditional multivariate analyses are unable to do. We suggest that the techniques commonly used with certain BPSD are capable of minimizing the confusion felt by a PWD and maintaining their peace of mind and body, as well as facilitating a positive environment for other residents.

In conclusion, our analysis identified four types of management techniques: (i) emotional and behavior-concordance techniques; (ii) acceptance and supportive techniques; (iii) restraining techniques; and (iv) avoidance techniques. Different management techniques, and combinations of techniques, were found to vary in use and effectiveness with different BPSD. Good management techniques for many BPSD have been developed and are being implemented by care providers. The present study has the potential to inform researchers and care providers in Japan about the types of management techniques in current use, as well as areas of potential need for staff training.

ACKNOWLEDGMENTS

This study was funded by the Foundation of Total Health Promotion and Osaka Gas Group Welfare Foundation.