The Assessments of Music Therapy for Dementia Based on the Cochrane Review.

BACKGROUND: Research on music therapy for dementia has taken a variety of measures and has been slow to consolidate evidence. Examining the outcomes that are currently being investigated and the measures that have been used can be useful for future research on music therapy for dementia.
OBJECTIVES: This study used cited original papers from a review in the Cochrane Database of Systematic Reviews to determine if there are items that should be measured or scales that should be used in conducting research on music therapy for dementia. The rating scales used and the outcomes examined were extracted.
METHOD: We used Dodd's criteria to identify (1) the outcome domains examined in music therapy for dementia, (2) the measures used, and (3) the measures capable of detecting significant intervention effects. RESULT: A search for reviews was conducted, and 7 systematic reviews (78 articles) were identified. Among them, 30 articles met the inclusion criteria. The 30 articles examined 18 of the 38 items in Dodd's outcome categories, while 20 items were not examined, and 78 different survey methods were used. The items most frequently surveyed in the studies were psychiatric outcomes, cognitive functioning, and global quality of life general outcomes.
CONCLUSIONS: We found that many studies investigated cognitive function, behavioral and psychological symptoms of dementia (BPSD), and quality of life; compared to BPSD, various types of rating scales were used for cognitive function. By standardizing the rating scales, we can contribute to the accumulation of evidence for music therapy for dementia.

Introduction

The treatment of dementia involves a combination of pharmacological and nonpharmacological therapies. Nonpharmacological therapies may be prioritized, especially for behavioral and psychological symptoms of dementia (BPSD) [1, 2]. However, the status of the recommended grade for implementation of nonpharmacological therapies in Japan, including music therapy for dementia, is weak, and the overall evidence is considered inconclusive. Despite various studies on the therapeutic effects of music therapy, conclusive evidence for its use in treating dementia has not been established.

Kelly et al. [3] collected more than 4,200 responses from approximately 1,500 participants on the question, “What should be investigated in future research on dementia”? and published the top 10 responses in rank order. “Establishing evidence for pharmacological and nonpharmacological therapies to manage BPSD” was ranked fourth, and establishing evidence for music therapy was found to be desired [3]. However, some studies on music therapy for dementia do not have sufficient data to support the conclusions discussed in the articles [4], making the issue of evidence and quality of research on music therapy for dementia an urgent matter to be addressed.

A core outcome set (COS) helps to establish evidence in clinical research. A COS is a consensus-derived, standardized, and parsimonious collection of outcomes to be reported at minimum in all studies of a particular condition (“what” to measure) [5, 6, 7, 8]. A COS for music therapy for dementia would facilitate the aggregation of the effects of music therapy interventions for dementia, leading to the establishment of evidence. However, there is no COS for music therapy for dementia currently, and the outcomes have not been classified. If the researchers used different batteries to assess the intervention effects, we cannot compare the results between them, and it may prevent from establishing the evidence.

The present study examined the outcomes and measures investigated in music therapy for dementia, found in the Cochrane Database of Systematic Reviews (CDSR), as a preliminary step in the development of a COS. These outcomes and measures were categorized based on an outcome taxonomy created by Dodd et al. [9] that can be used in general medical clinical research. It consists of 38 items comprising mortality, physiological and clinical, impact on life, resource utilization, and adverse events. This is used in both clinical [10, 11] and basic research using animal experiments such as mice [12]. The collection of outcomes for the COS is essentially a systematic review in the CDSR. However, there were only a few reviews of music therapy for dementia in the CDSR; hence, we used the articles in the reviews, extracted the rating scales used, and the outcomes examined. Dodd's criteria were used to identify (1) outcome domains examined in music therapy for dementia, (2) measures used, and (3) scales able to detect significant intervention effects, to determine whether there were any items to be measured and any scales to be used in conducting research on music therapy for dementia.

Materials and Methods

The following words were used to search the CDSR: “dementia,” “music,” and “therapy,” in studies registered on or before September 1, 2020. The inclusion criteria were as follows: (1) articles reporting results of studies using music therapy, music intervention, or music stimulation, (2) articles were written in English, and (3) the target disease was dementia. The exclusion criteria were as follows: (1) articles reporting results of a study that did not use music therapy, music intervention, or music stimulation; (2) articles written in languages other than English; and (3) the subject of the study was not dementia. Only articles that met the inclusion criteria were included in the study. The items and scales examined in each article were extracted. The extracted items and scales were classified according to Dodd et al.'s [9] outcome taxonomy. All authors participated in the exclusion of articles, extraction of items, scales examined, and assignment to Dodd's classification; decisions were made by consensus in cases of differences.

The following methods were used to conduct the analysis: (1) identifying the items examined by each research article, categorized according to Dodd et al.'s [9] criteria, and identifying the applicable and nonapplicable outcome domains and (2) frequency of use of the measures used in each outcome domain and frequency of detection of significant differences. It was approved by the Research Safety and Ethics Committee of the Advanced Institute of Industrial Technology (No. 19009).

Results

Seven reviews [4, 13, 14, 15, 16, 17, 18] used 76 articles, and 30 met the inclusion criteria [19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48] (Table 1). The 30 articles examined 18 of the 38 items in Dodd's outcome categories, while 20 items were not examined, and 78 different survey methods were used. The items most frequently surveyed in the studies were psychiatric outcomes, cognitive functioning, and global quality of life general outcomes (Table 2). The frequency of use of the scales used in each outcome domain and the frequency of detecting significant differences are shown in Table 3.

Table 1

The 30 articles examined 18 of the 38 items in Dodd's outcome categories

	Outcome domain	Studies, n
Physiological/clinical	Cardiac outcomes	4
	Endocrine outcomes	1
	Ear and labyrinth outcomes	1
	General outcomes	24
	Injury and poisoning outcomes	1
	Nervous system outcomes	7
	Psychiatric outcomes	27
	Skin and subcutaneous tissue
	outcomes	1
	Vascular outcomes	1
Life impact	Physical functioning	4
	Role functioning	2
	Emotional functioning/wellbeing	5
	Cognitive functioning	24
	Global quality of life	8
	Perceived health status	3
	Delivery of care	1
	Personal circumstances	8
Resource use	Societal/carer burden	6

Table 2

Outcomes included in the 30 applicable studies

Core area	N	Outcome domain	Article no.
			number	27	34	29	36	33	24	45	28	39	43	30	25	22	47	41	31	35	38	32	37	23	46	48	40	44	42	20	21	26	19
Physiological/clinical	3	Cardiac outcomes	4	○									○	○									○
Physiological/clinical	5	Endocrine outcomes	1									○
	6	Ear and labyrinth outcomes	1					○
Life impact	9	General outcomes	24	○		○	○	○	○		○	○	○	○	○	○	○	○	○	○	○		○		○	○	○	○		○	○	○
	13	Injury and poisoning outcomes	1																												○
	17	Nervous system outcomes	7			○							○	○							○				○					○		○
	21	Psychiatric outcomes	27	○	○	○	○	○	○		○	○	○	○	○	○	○		○	○	○	○	○	○	○	○	○	○	○	○	○	○
	23	Skin and subcutaneous tissue outcomes	1										○
	24	Vascular outcomes	1	○
	25	Physical functioning	4	○				○						○																○
Life impact	27	Role functioning	2					○			○
Resource use	28	Emotional functioning/wellbeing	5							○			○				○						○				○
	29	Cognitive functioning	24	○		○	○	○	○		○	○		○	○	○		○	○	○	○		○	○	○	○	○	○		○	○	○	○
	30	Global quality of life	8			○	○										○	○				○				○		○				○
	31	Perceived health status	3			○								○				○
	32	Delivery of care	1	○
	33	Personal circumstances	8				○				○	○	○					○	○											○		○
	37	Societal/carer burden	6				○				○							○								○	○					○

Table 3

The frequency of use of the scales used in each outcome domain and the frequency of detecting significant differences

Outcome	Sub outcome	Scale/survey method, n	Scale/survey method	Pertinent	Detections, n
Cardiac outcomes	HR and variability	3	HR	2	1
			ECG Holter monitoring (HR variability)	1	1
	Actual prevalence of heart disease	1	Actual prevalence of heart disease	1
Endocrine outcomes	Stress	1	Salivary cortisol	1
Ear and labyrinth outcomes	Hearing skill	1	Hearing and perceptive-communicative and relational skills	1
General outcomes	Characteristics	70	Age	24
			Sex	24
			Education	13
			Marital status	5
			Main medicines	2
			Most recent occupation	1
			Religion	1
	Musical skill	3	Music Preference Questionnaire	2
			Musical Expertise Questionnaire	1
Injury and poisoning outcomes	Pain	1	Pain	1	1
Nervous system outcomes	Diagnosis	6	Type of dementia	6
Psychiatric outcomes	Global BPSD	14	NPI	9	8
			BEHAVE-AD	2	2
			Observance	3	1
	Anxiety	4	RAID	3	1
			Hamilton Anxiety Scale	1	1
	Depression	7	GDS	5	3
			CSDD	2	2
	Agitation	14	CMAI	13	7
			VDB	1	1
Skin and subcutaneous tissue outcomes	Skin conductance	1	Skin conductance (Q sensor)	1
Vascular outcomes	Blood pressure	1	Blood pressure	1
Physical functioning	ADL	3	Index of Independence in ADL	3	1
	IADL	1	IADL	1
	Physical functioning	2	Use of walker or wheelchair	1
			Lawton and Brody's PSMS	1
Role functioning	Recording during music therapy	3	GMP	1	1
			SVAM or Music Therapy Activity Scale	1	1
			Participant Engagement Observation	1	1
			Checklist
Emotional functioning/wellbeing	Anxiety	2	STAI-A	2	1
	Emotional facial expressions	3	Facial Action System Coding	1	1
			Faces Scale	1	1
			EFE	1	1
	Measure both positive and negative affect	1	PANAS	1	1
	Discourse content	1	Discourse content	1	1
	Mood	1	Recall, social interaction, and mood questionnaire	1	1
Cognitive functioning	Multiple cognitive functions	26	MMSE	19	3
			ADAS-cog	2	1
			Severe Impairment Battery	1	1
			BIMS	1	1
			BCRS	1
			Short Portable Mental Status	1
			Questionnaire
			Consortium to Establish a Registry for Alzheimer's Disease battery	1
	Severity of dementia	11	Clinical Developmental Stage of Dementia (FAST)	1	1
			CDR	6
			GDS	2
			Severity of dementia (no scale)	2
	Memory function	4	MPI and MPD	1	1
			WHO-UCLA AVLT	1	1
			WMSR	1
			Forward and reverse digit-span exercise	1
	Language function	3	Semantic verbal fluency test	1	1
			Boston Naming Test	1
			Western Aphasia Battery	1
	Frontal function	3	Attentional matrixes	1	1
			Trail Making Test	1
			Frontal Assessment Battery	1
	Intellectual function	1	WAIS	1
Global quality of life	QOL	14	Barthel Index	3
			QOL-AD	2	2
			CBSQoL	2	2
			DQOL	2	2
			DCM	1	1
			Alzheimer's Disease-Related Quality of Life	1
Perceived health status	Duration of disease	1	Duration of disease	3
Delivery of care	Assess the state of care	1	Assess the state of care	1
Personal circumstances	Length of stay	6	Length of stay at care facility	6
	Dyad partner	2	Dyad partner (spouse/child/sibling or other relative/nurse)	1
Societal/carer burden	Care burden	3	NPI	2
			Zarit Burden Interview	1
	Proxy level of care	1	Staff proxy level	1
		1	Level of care in facility	1
		1	Global Health Questionnaire	1

HR, heart rate; BEHAVE-AD, Behavior Pathology in Alzheimer's Disease Rating Scale; RAID, Rating Anxiety in Dementia Scale; GDS, Geriatric Depression Scale; CSDD, Cornell Scale Depression in Dementia; CMAI, Cohen-Mansfield Agitation Inventory; VDB, VERBALLY disruptive behaviors; ADL, Activities of Daily Living; IADL, Instrumental Activities of Daily Living; PSMS, Physical Self Maintenance Scale; GMP, Geriatric Music Therapy Profile; SVAM, Scala di valutazione dellattivita' musicoterapeutica; STAI-A, State Trait Anxiety inventory for Adults; EFE, emotional facial expressions; PANAS, Positive and Negative Affect Schedule; ADAS-cog, Alzheimer's Disease Assessment Scale; BIMS, Brief Interview for Mental Status; BCRS, Brief Cognitive Rating Scale; CDR, Clinical Dementia Rating scale; GDS, Global Deterioration Scale; MPI and MPD, Immediate and Deferred Prose Memory test; WHO-UCLA AVLT, World Health Organization University of California-Los Angeles, Auditory Verbal Learning Test; WMSR, Wechsler Memory Scale III; WAIS, Wechsler Adult Intelligence Scale; QOL-AD, Quality of Life-Alzheimer's Disease; CBSQoL, Cornell Brown Scale Quality of Life; DQOL, Dementia Quality of Life questionnaire; DCM, Dementia Care Mapping.

Discussion/Conclusion

Many of the general outcomes were items examined as demographic data, and most studies examined age, gender, and education. Since this present review focused on studies of music therapy, music intervention, or music stimulation for dementia and not on subtypes of dementia (e.g., Alzheimer's disease [AD], dementia with Lewy bodies, and cerebrovascular dementia), age, gender, and education were not included in the demographic data. In some studies, the classification of dementia type was also included in the demographic data.

The cognitive outcome revealed the use of a variety of measures of cognitive function in addition to general intelligence, including the Mini-Mental State Examination (MMSE). In dementia, multiple cognitive functions are impaired, reflecting the areas of functional decline in each disease. The main areas of cognitive decline include attention, executive function, memory, language, visuospatial cognition, action, and social cognition [49]. It was inferred that the wide range of symptoms expressed by patients of different subtypes of dementia led to the use of many scales to assess each of them. Särkämö et al. [41] used 9 cognitive function tests, the most used, in an attempt to assess overall cerebral function. While it is important to use scales according to the functions to be assessed, the burden on the target dementia patients must be considered. It is therefore necessary to reorganize the scales to assess the patients' conditions with the least possible number of assessment method.

The most frequently rated psychiatric outcome reflected the BPSD rating in dementia. According to previous reports, music therapy and other music interventions for dementia have been shown to improve BPSD, with effects on anxiety and depression reported by Ueda et al. [50] and those for agitation reported by Vink et al. [38, 42] and Lin et al. [31]. Many studies included in this review were concerned with BPSD, and many of them used the Cohen-Mansfield Agitation Inventory, a rating scale for agitation, although there were differences in the shortened or Chinese versions. Others used a rating scale for depression, a rating scale for anxiety, and a rating scale that can evaluate several BPSDs together. The Neuropsychiatric Inventory (NPI) evaluates the occurrence, frequency, and severity of BPSD, and there are various versions, such as that for clinicians and nursing homes. The NPI also assesses the burden of care for each item, and some studies have partially used the NPI to assess care burden. While many studies have assessed BPSD, the method of assessment has remained constant compared to that for other items. Moreover, the use of these rating scales in studies of music therapy and music intervention for dementia has been thought to be useful in accumulating evidence.

There is a wide variety of quality of life (QOL) scales for AD, and the present review revealed that many scales have been used in traditional music therapy and music intervention research. QOL assessment scales follow Lowton's QOL model and have a variety of response formats, including those that require responses to caregivers [51]. Further, there are QOL scales for the elderly that are not limited to AD [52]. The QOL scale itself is under review, and it is anticipated that an optimal scale will be found in order to accumulate evidence for music therapy and music intervention in dementia.

Finally, 20 categories were not investigated in the articles included in the survey. The outcome classifications developed by Dodd et al. [9] are a classification table that can be applied in general clinical research, and some categories were not applicable to dementia (e.g., congenital, familial, and hereditary outcomes; pregnancy and perinatal outcomes). However, economic outcomes have been examined in several previous studies [53, 54], and in particular, music therapy for BPSD in dementia has been found to be more economical and more effective than pharmacotherapy. Importantly, while the studies reviewed were intervention studies, adverse events were not reported. Further, although singing was used in the intervention, respiratory outcomes were not assessed, suggesting that there are several issues that need to be clarified in future intervention studies using music therapy and music for dementia.

Consequently, we extracted articles used in a systematic review in CDSR. In a normal review study, it is desirable to extract and analyze the data according to the PRISMA guidelines, but the content of this study did not meet the PRISMA guidelines. Therefore, we decided to proceed with the study in an exploratory manner. This is a limitation of this study.

In conclusion, we investigated the effects of music therapy on dementia from articles used in a systematic review housed in the CDSR. As a result, we found that many studies investigated cognitive function, BPSD, and quality of life; compared to BPSD, various types of rating scales were used for cognitive function. By standardizing the rating scales, we can contribute to the accumulation of evidence for music therapy for dementia.

Statement of Ethics

This study protocol was reviewed and approved by Committee of the Advanced Institute of Industrial Technology, approval number 19009.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was supported by JSPS KAKENHI Grant-in-Aid for Scientific Research (B) (Grant No.: 20H04303), Grant-in-Aid for Scientific Research (C) (Grant No.: 21K12194), and Grant-in-Aid for Young Scientist (B) (Grant No.: 17K17811).

Author Contributions

Conception and design of the experiments were contributed by M.A. Research studies were conducted by M.A., K.T., and M.S. Data analysis was contributed by M.A., K.T., and M.S. Writing of the paper was contributed by M.A. Supervision and interpretation of the data were contributed by K.T. and M.S. All authors read and approved the final version of the paper.