Supplemental Digital Content1 -Exercise
Study / Interventions / Study design.Follow-up / Number / Method of randomization / Blinding of raters / Outcomes relevant to apathy / Significance / Quality
rating
Cevasco and Grant, 2003, USA (1) (See SDC 2 for music component of this study) / Exercise-to-movement activity consisting of combinations of
continuous cueing vs single cueing and difficult condition (i.e., participating in designated movement) vs easy condition (approximating the movement) during 38 weekly or twice weekly 50 min sessions over 8 months / Interrupted time series without parallel control group.
No f/u / 14 assisted living residents with early to middle AD (10 females) / Order of movement activities randomized. Method not stated. Participants served as their own controls. / No / Responsiveness and participation in structured, designated movement or approximating the movement or not participating recorded at 30 sec intervals by trained data collectors. / Continuous cueing/easy condition resulted in significantly higher participation than the single cueing/difficult condition (F(3,52) = 2.98, p < .05). / 6
Cott et al, 2002, Canada (2) / Walk-and-talk sessions, i.e., conversation while walking in pairs vs talk-only sessions, i.e., conversation while sitting in pairs; both groups’ sessions for 30 min, 5 x weekly for 16 weeks vs usual program / Clustered RCT.
No f/u / 86 ITT and 74 IA residents with AD in 3 geriatric long-term care facilities (39 females) / Residents randomly assigned to group within each site with random number table. No further details of allocation provided. / Raters blinded to group membership but not study design / Engagement, helping behavior and relationships with others as measured by the disengagement subscale of the London Psychogeriatric Rating Scale (3, 4); social communication and overall communication as measured by the Functional Assessment of Communication Skills for Adults (5) / No significant between-groups differences found on outcome measures at posttest for walk-and-talk sessions. Group and level of cognitive impairment examined as between-subject factors. Main effect for group not significant but main effect for cognitive impairment significant for communication in those with moderate impairment (F(1,68) = 44.65, p = .00).Group x cognitive impairment interaction not significant. / 12
Hopman-Rock et al, 1999, Netherlands (6) / PAP vs
usual activities twice weekly over 6 months (min 15 sessions for inclusion) / Pseudo- RCT.
No f/u / 134 ITTNH residents with dementia from 11 homes for the elderly 92 IA, experimental group n=45 (41 females), control group n=47 (46 females) / Where possible randomly allocated within NH but where not possible, NH group randomly allocated as a whole. Method not stated. Participants also served as their own controls. / No / Non-social and apathetic behaviors as measured by these subscales of the BIP (also known as GIP) (7) as well as positive and negative group behaviors as measured by these subscales of the SIPO (6) / Nonsignificant improvement demonstrated in non-social and apathetic behaviors as well as positive and negative group behaviors for PAP group. When group results were separated by level of cognitive impairment, those with higher cognitive performance (Cognitive Screening Test-20 >6) (8) showed a significant increase in positive group behaviors on the SIPO (F = 4.46, p = ≤ .05). / 9
Notes: f/u: follow up; AD: Alzheimer’s disease; RCT: randomized controlled trial; ITT: intention to treat; IA: in analysis; PAP: Psychomotor Activation Program;NH: nursing home; GIP/BIP: Gedragsobservatieschaal voor de Intramurale Psychogeriatrie [Dutch Behavior Rating Scale for Psychogeriatric Inpatients]; SIPO: Social Interaction Scale for Psychogeriatric Older People
1
References
1.Cevasco AM, Grant RE: Comparison of different methods for eliciting exercise-to-music for clients with Alzheimer's disease. J Music Ther 2003; 40:41-56
2.Cott CA, Dawson P, Sidani S, et al: The effects of a walking/talking program on communication, ambulation, and functional status in residents with Alzheimer disease. Alzheimer Dis Assoc Disord 2002; 16:81-87
3.Reid DW, Tierney MC, Zorzitto ML, et al: On the clinical value of the London Psychogeriatric Rating Scale. J Am Geriatr Soc. 1991; 39:368-371
4.Hersch EL, Kral VA, Palmer RB: Clinical value of the London psychogeriatric rating scale. J Am Geriatr Soc 1978; 26:348-354
5.Frattali CM, Thompson CM, Holland AL, et al: The FACS of life ASHA facs-a functional outcome measure for adults. ASHA 1995; 37:40-46
6.Hopman-Rock M, Staats PG, Tak EC, et al: The effects of a psychomotor activation programme for use in groups of cognitively impaired people in homes for the elderly. Int J Geriatr Psychiatry 1999; 14:633-642
7.Verstraten PF: The GIP: an observational ward behavior scale. Psychopharmacol Bull 1988; 24:717-719
8.Maring W, Deelman BG: The cognitive screening test: Long and short versions. Tijdschrift voor Gerontologie en Geriatrie. 1999; 30:205-211
Supplemental Digital Content2 - Music
Study / Interventions / Study design.Follow-up / Number / Method of randomization / Blinding of raters / Outcomes relevant to apathy / Significance / Quality rating
Ashida, 2000, USA (1) / Two weeks of no music therapy followed by reminiscence music therapy in 5 x daily sessions of an average 43 mins, over 1 week / Interrupted time series without parallel control group.
No f/u / 20 residents with dementia from 2 residential care facilities, divided into 4 small groups and treated identically (17 females) / Participants served as their own controls / No / Level and characteristics of on-task active and passive participation as recorded on videotape / Behavioral observations demonstrated nonsignificant increase in active participation and decrease in passive participation in 3 of the 4 groups as treatment weeks progressed. / 6
Cevasco and Grant, 2003, USA (2)
(See SDC1 for exercise component of this study) / Exercise-to-music activity consisting of combinations of vocal music vs instrumental music and with instruments vs without instruments during 26 weekly or twice weekly 50 min sessions over 6 months / Interrupted time series without parallel control group.
No f/u / 12 assisted living residents with early to middle AD (11 females) / Order of movement activities randomized but method not stated. Participants served as their own controls / No / Responsiveness and participation in structured, designated movement or approximating the movement or not participating recorded at 30 sec intervals by trained data collectors / Exercise to instrumental music resulted in significantly higher participation than exercise with instruments to vocal music (t = 2.6, p < .05). / 6
Clair, 2002, USA (3) / Music therapy - singing or dancing in weekly sessions of 40 min over 6 weeks / Interrupted time series without parallel control group.
No f/u / 8 couples consisting of residents with advanced dementia in a SCU and their caregivers (4 female residents) / Participants served as their own controls / No / Engagement: interaction between caregiver and care receiver defined as touch, conversation or looking as well as singing, vocalising, moving or dancing to music as recorded by a trained observer. / Highly significant increase in mean, posttest engagement scores when compared with pretest engagement scores (t = 2.88, p = .024). / 5
Gaebler & Hemsley, 1991, UK (4) / Musical, auditory stimulation on one occasion only / Pretest-posttest design without parallel control group.
No f/u / 6 female, inpatients with severe dementia in a long-stay geriatric ward / Participants served as their own controls / No / Behavioral engagement as measured by number of animated facial expressions and facial gestures as well as occasions where observable stimuli were followed, interaction attempts were initiated and absorption or involvement in an activity was evident as recorded by a trained observer. / No significant changes in 5 residents. One participant showed a significant mean change score between pre- and post-test on looking, facial gestures, interaction and interest (p < 0.1). / 6
Groene, 2001, USA (5) / Music therapy consisting of 4 singalong group sessions of combinations of each of the 4 conditions: live (L) and recorded (R) presentation modes with simple (S) and complex (C) accompaniment styles. Total sessions 16. Duration and frequency not reported / Interrupted time series without parallel control group.
No f/u / 8 residents with dementia in a SCU (7 females) / Order of conditions (L/S, L/C, R/S, R/C) randomized by a latin square design. No further details of allocation provided. / Not stated / Participation and responding as measured by number of: affirmative verbalisations made, leaving the group, facing therapist with eye contact, readiness to sing, nods of affirmation before and after sessions + reading lyrics, singing or mouthing songs and applauding / Participation, as evidenced by affirmative verbalisations (t(238) = 2.66, p = < .01) and applause after (t(238) = 2.03, p = < .043) were significantly higher for live sessions, overall, than for recorded; leaving the group was significantly higher for R/S condition (p = ≤ .01); joining in with reading lyrics was significantly higher for L/C condition (p = .03); facing therapist with eye contact at end of songs was significantly higher for L/C and R/C conditions (p = .00); applause was significantly higher for L/C condition (p ≤ .02) / 10
Holmes et al, 2006, UK(6) / Live interactive music therapy vs pre-recorded music sessions vs silence during 1 x 1.5 hr session consisting of 30 min period for each condition / Interrupted time series without parallel control group.
No f/u / 32 NH residents with moderate to severe dementia (28 females) / Order of silence and musical periods randomized. Method not stated / Blinded independent, observer-raters / Reduced apathy through positive engagement as measured by DCM, Behavior Category Codes (7, 8) according to video recordings rated at 3 min intervals by a trained, independent rater / Positive engagement during live music sessions was significantly greater than during pre-recorded music (p < .01) and silent (p < .00) sessions; positive engagement during pre-recorded music was not significantly greater than during silent periods
. / 13
Lord & Garner, 1993, USA (9) / Music group activity vs puzzle exercises vs usual recreational activities of drawing, painting and TV during 6 x 30 min sessions per week over 6 months / Pseudo- RCT.
No f/u / 60 residents with AD in a nursing care facility (42 females) / Participants nonsystematically separated into 3 equal groups and served as their own controls / Blinded analyses but raters not blinded / Change in social coaction ratings on a four-point scale developed by the authors. Active participation measured in the music group by singing, humming, playing an instrument or moving to the music as observed on a total 72 occasions per resident. Similar definitions of participation developed for puzzle and usual recreational groups. / Significant difference in mean gain participation scores for pre and post observations in music group (M = 1.60, t19 = 8.72, p < .01) but not puzzle or usual recreational activities groups. Anecdotal evidence indicates residents in music group smiled, laughed, sang, danced, whistled and eagerly anticipated the sessions while those in other groups did not. / 5
Mathews et al, 2001, USA (10) / Exercise activities vs exercise activities with rhythmic music, 22 mins x 5 days per week over 25 weeks / Interrupted time series without parallel control group.
No f/u / 18 residents in a residential, dementia SCU (17 females) / Participants served as their own controls / No / Engagement as defined by participation in the exercise activities using weekly observations taken twice during each 22 min session / Mean group engagement scores increased in the exercise activities with rhythmic music condition (M = 68-69%) when compared with exercise activities alone (M = 41-53%). Statistical significance not reported. / 6
Pollack & Namazi, 1992, USA (11) / Individualised music activity vs other activity of choice in 6 x 20 min sessions, 3 per week over 2 weeks / Pretest-posttest design without parallel control group.
No f/u / 8 residents with probable AD in a specialist AD residential facility (5 females) / Participants served as their own controls / No / Change in social behaviors and participation, as recorded before and after music sessions, by frequency of interaction, gesturing, smiling, touching, humming, singing and whistling as well as change in passive, nonsocial behaviors. / Increase in social behavior (24%) and a decrease in nonsocial and passive behavior (14%) for the group at close of treatment period (x² = 14.2, df = 1, p < .001) / 4
Raglio et al, 2008, Italy (12) / Nonverbal music therapy using rhythmical and melodic instruments vs educational and entertainment activities in 10 x 30 min session over 16 weeks / Psuedo- RCT.
4 week f/u / 59 residents with AD or VaD from 3 NHs). Experimental group n = 30; control group n = 29 (25 females each group). / Residents allocated alternately to control or experimental group / Yes / Apathy change scores as measured by the NPI (13) as well as active participation as measured by empathetic behavior, smiles, singing and body movements synchronic with the music / Significant improvement in apathy scores between pretest and follow up (F1,57 = 8.10, p < .05) in treatment group but not control group. Active participation also increased in the treatment group over time: empathetic behavior (F3,87 = 10.37, p < .0001; Cohen d = .61), smiles (F3,87 = 8.14, p < .0001; Cohen d = .53), singing (F3,87 = 6.98, p < .0003; Cohen d = .62) and synchronic body movements (F3,87 = 12.41, p < .0001; Cohen d = .62). Participation data for control group and group x time interaction effect not reported. / 9
Sherratt et al, 2004, UK (14) / Live music (guitar playing and singing by semi-professional musician) vs taped commercial music vs taped recording of music played by the musician vs no music during 1hr sessions for each condition over 3 months. 4 conditions counter-balanced. / Interrupted time series without parallel control group.
No f/u / 24 participants with moderate to severe dementia as well as signs of social withdrawal and minimal engagement (16 continuing care ward residents, 7 day hospital attendees and 1 observed across both settings (10 females) / Participants served as their own controls. / No / Increased engagement defined as meaningful activity, engagement with music source, interaction with music source and specific responses to music (e.g. clapping, singing) as measured by DCM (7, 8) using continuous time sampling and direct observations by trained observers / Significant reduction in percentage of time spent in no meaningful activity for live music condition when compared with 3 other conditions (X2 (3) = 26.65, p = .01). Significant increase in percentage of time spent in engagement with music source (X2 (2) = 13.30, p = .01), interaction with music source (X2 (2) = 14.02, p = .01) and responses to music (X2 (2) = 12.90, p = .01) as well as significant reduction in percentage of time spent in no observable response to music (X2 (2) = 14.21, p = .01) for live music condition when compared with other music conditions. Significant reduction in percentage of time spent in passive behaviors for live music condition when compared with no music condition (X2 (3) = 7.95, p = .05). / 9
Notes: f/u: follow up; AD: Alzheimer’s disease; SCU: special care unit; NH: nursing home; DCM: Dementia Care Mapping; RCT: randomized controlled trial; VaD: vascular dementia; NPI: Neuropsychiatric Inventory;
1
References
1.Ashida S: The effect of reminiscence music therapy sessions on changes in depressive symptoms in elderly persons with dementia. J Music Ther 2000; 37:170-182
2.Cevasco AM, Grant RE: Comparison of different methods for eliciting exercise-to-music for clients with Alzheimer's disease. J Music Ther 2003; 40:41-56
3.Clair AA: The effects of music therapy on engagement in family caregiver and care receiver couples with dementia. American Journal of Alzheimer's Disease and other Dementias. 2002; 17:286-290
4.Gaebler HC, Hemsley DR: The assessment and short-term manipulation of affect in the severely demented. Behavioural Psychotherapy 1991; 19:145-156
5.Groene R: The effect of presentation and accompaniment styles on attentional and responsive behaviors of participants with dementia diagnoses. J Music Ther 2001; 38:36-50
6.Holmes C, Knights A, Dean C, et al: Keep music live: music and the alleviation of apathy in dementia subjects. Int Psychogeriatr 2006; 18:623-630
7.Sloane PD, Brooker D, Cohen L, et al: Dementia care mapping as a research tool. Int J Geriatr Psychiatry 2007; 22:580-589
8.Brooker D: Dementia care mapping: a review of the research literature. Gerontologist 2005; 45 11-18
9.Lord TR, Garner JE: Effects of music on Alzheimer patients. Perceptual and Motor Skills. 1993; 76:451-455
10.Mathews RM, Clair AA, Kosloski K: Keeping the beat: use of rhythmic music during exercise activities for the elderly with dementia. American Journal of Alzheimer's Disease & Other Dementias 2001; 16:377-380
11.Pollack NJ, Namazi KH: The effect of music participation on the social behavior of Alzheimer's Disease Patients. J Music Ther 1992; 29:54-67
12.Raglio AMT, Bellelli GMD, Traficante DPP, et al: Efficacy of Music Therapy in the Treatment of Behavioral and Psychiatric Symptoms of Dementia. Alzheimer Dis Assoc Disord 2008; 22:158-162
13.Cummings JL, Mega M, Gray K, et al: The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology 1994; 44:2308-2314
14.Sherratt K, Thornton A, Hatton C: Emotional and behavioural responses to music in people with dementia: An observational study. Aging and Mental Health. 2004; 8:233-241
Supplemental Digital Content 3 - Multi-sensory
Study / Interventions / Study design.Follow-up / Number / Method of randomization / Blinding of raters / Outcomes relevant to apathy / Significance / Quality
rating
Baker et al, 2003, UK, Netherlands & Sweden (1) / MSS vs control activity during 30 min sessions twice weekly over 4 weeks / RCT.
1 month f/u / 136 ITT and 127 IA older adults with dementia (94 community dwelling and 42 psychogeriatric inpatients); intervention group n = 65, control group n = 71. / Computer randomization with epidemiology software / No / Apathy subscale of the BRS section of the Clifton Assessment Procedures for the Elderly (2, 3), apathetic/withdrawn subscale of the BMD (4, 5), non-social behavior subscale of the GIP (6) as well as spontaneity, initiative and inactivity as rated with the ‘Interact’ form developed by the authors (1) / No significant differences between MSS and control activity groups from before to after sessions.
Significant main effects of time. Both groups related better to others [F(1,108) = 28.97, P < 0.0001] and were less bored /inactive [F(1,108) = 43.38, P < 0.0001] after sessions when compared with before. The severely cognitively impaired in the MSS group were significantly less apathetic on the BRS apathy subscale after sessions when compared with before [F(1,83) = 7.20, P < 0.01]. Improvement had deteriorated at follow-up. / 12
Buettner 1999, USA (7) / “Simple Pleasures” i.e., readily accessible, age- and stage-appropriate sensorimotor recreational items vs usual care over 12 months / Interrupted, cross-over time series without parallel control group.
No f/u, but 6 months post intervention data available for group1 due to crossover design / 55 NH residents with dementia in 2 dementia units (44 females) / Participants served as their own controls / No / Time spent by residents in purposeful activities with sensorimotor recreational items as measured by the “Scanning the Environment Tool” and the “Time engaged with Item Scale” developed by the research team / Decrease in mean number of instances of residents “not doing anything” during intervention period (site 1 M = 8.26, site 2 M = 10.85) when compared with control period (site 1 M = 17.52, site 2 M = 15.36). Observations indicate that residents were more involved with recreation items and other residents during intervention period. Statistical significance not reported. / 4
Minner et al, 2004, USA (8) / MSS sessions to the total of 324 over 1 year / Interruptedtime series withoutparallel control group.
No f/u / 19 NH residents with dementia and difficult behaviors / Participants served as their own controls / No / Positive vocalisation, smile, positive gestures, positive noise expression and positive interactions with people or objects as recorded using an observational checklist developed by first author. / Increase in mean number of positive behaviors per resident during (M = 5.3) and after sessions (M = 5.1) when compared with before (M = 3.7). Statistical significance not reported. / 2
Moffat et al, 1993, UK (9) / MSS sessions of 30 mins duration 3 days per week, over 4 weeks / Interrupted, cross-over time series without parallel control group.
No f/u / 6 male patients with dementia from a continuing care ward and 6 female patients with dementia from a day hospital / Participants served as their own controls / Raters not blinded / Willingness to participate, interaction with others, interest and energy level during sessions as measured by a rating scale (10) as well as before and after sessions using an observation system (11). Behavior was also rated by the apathetic/withdrawn subscale of the BMD (4, 5). / Significant increase in frequency of observed interest during the 10 min pre-session period (tau = .43, p < 0.02) and 10 min post-session period (tau = .603, p < 0.003) over 4 weeks of study but no significant difference in mean interest was found between pre-and post-session scores. No significant difference between baseline and post MSS session scores on the apathetic/withdrawn subscale of the BMD. / 9
Staal et al, 2007, USA (12) / MSBT for 6 sessions of up to 30min each + standard psychiatric care (i.e., pharmacological therapy, occupational therapy and structured hospital environment) vs standard psychiatric care + structured activity sessions (no time frame reported) / Psuedo- RCT.
No f/u / 24 inpatients with moderate to severe dementia and behavioral disturbances in an acute, geriatric, psychiatric unit (16 females); intervention group n = 12, control group n = 12. / Not stated / Apathy raters not blinded / Apathy as measured by the avolition-apathy, social-emotional withdrawal and affective blunting subscales of the SANS-AD (13) / MSBT group showed significantly greater improvement in apathy than control group when controlling for physical health and age (F(1, 20) = 4.47, p = .04). No main effect was found for time but time x group interaction effect was significant (F(6, 120) = 3.15, p = .01). / 9
Van Weert et al, 2005, Netherlands (14) / Individual MSS integrated in 24hr care vs usual care over 18 months / Clustered psuedo- RCT.
No f/u / 128 residents with dementia from 12 psychogeriatric wards of 6 NHs (61 completers included in both pre- and posttest) / Randomization at ward level. 4 wards randomized by ‘drawing lots’ and 2 were assigned. 61 participants served as their own controls / Independent observers rated video recordings however, caregiver observations unblinded / Apathetic behaviors i.e., attentiveness and responsiveness in relation to the environment, others and initiative as measured by the apathy subscale of the GIP (6) / Significant treatment effect for change score on apathetic behaviors in intervention group but not control group (p < .05) / 9
Notes: MSS: multi-sensory stimulation; RCT: randomized controlled trial; f/u: follow up; ITT: intention to treat; IA: in analysis; BRS: Behavioral Rating Scale; BMD: Behaviour and Mood Disturbance Scale; GIP/BIP: Gedragsobservatieschaal voor de Intramurale Psychogeriatrie [Dutch Behavior Rating Scale for Psychogeriatric Inpatients]; NH: nursing home; MSBT: Multi sensory behavior therapy; SANS-AD: Scale for the Assessment of Negative Symptoms in Alzheimer’s Disease;