Snoezelen for dementia (English)

Rate this item
(0 votes)

Background:

Snoezelen, multi-sensory stimulation, provides sensory stimuli to stimulate the primary senses of sight, hearing, touch, taste and smell, through the use of lighting effects, tactile surfaces, meditative music and the odour of relaxing essential oils (Pinkney 1997). The clinical application of snoezelen has been extended from the field of learning disability to dementia care over the past decade.

 

The rationale for its use lies in providing a sensory environment that places fewer demands on intellectual abilities but capitalizes on the residual sensorimotor abilities of people with dementia (e.g. Buettner 1999, Hope 1998). Practitioners are keen to use snoezelen in dementia care, and some encouraging results have been documented in the area of promoting adaptive behaviours (e.g. Baker 1997, Long 1992, Spaull 1998). However, the clinical application of snoezelen often varies in form, nature, principles and procedures. Such variations not only make examination of the therapeutic values of Snoezelen difficult, but also impede the clinical development of snoezelen in dementia care. A systematic review of evidence for the efficacy of snoezelen in the care of people with dementia is therefore needed to inform future clinical applications and research directions.

 

Objective: This review aims to examine the clinical efficacy of snoezelen for older people with dementia.

Search strategy: "Snoezelen", "multi-sensory", "dement*", "Alzheimer*", "randomized control/single control/double control" were used as keywords to search seven electronic databases (e.g. MEDLINE, PsyLIT). The list of trials was compared with those identified from a search of the Specialized Register of the Cochrane Dementia and Cognitive Improvement Group.

Selection criteria: All RCTs in which Snoezelen or multi-sensory programmes were used as an intervention for people with dementia were included in the review. Trial data included in the review were restricted to those involving people aged over 60 years suffering from any type of dementia, except one subject of Baker 1997's study was aged below 60 years.

Data collection and analysis: Only two RCTs fulfill the inclusion criteria for this systematic review. Two reviewers independently extracted the data from these two inclusion studies. Quantitative synthesis of the comparable data from the two trials was performed.

Main results: Two trials were included. Both Baker 1997 ( and Kragt 1997 examined the short-term values of snoezelen on the behaviours of people with dementia. Although the pooled results were insignificant, the trend was in the direction of favouring treatment (hence a negative value of the SMD). The standardized mean difference (SMD) was -1.22, with a 95% confidence interval (CI) (-4.08, 1.64). Kragt 1997's result, weighted 47%, was significant in favour of treatment, with a SMD of -2.77 and a 95% CI (-4.24, -1.29). During the snoezelen session, Kragt 1997's subjects presented significantly fewer apathetic behaviours (t=-8.22, p<0.01), fewer restless behaviours (t=-3.00, p=0.01), fewer repetitive behaviours (t=-.822, p<0.01), and fewer disturbances (t=-4.91, p<0.01). Baker 1997's result was slightly not in favour of the treatment, with a SMD of 0.16 and a 95% CI (-0.41, 0.73). The control subjects touched objects/equipment more appropriately within the activity sessions than the subjects who participated in snoezelen sessions (F(1,47)=5.96, p=.001). Kragt 1997 did not examine the carryover and long-term effects of snoezelen, so only Baker 1997's results were analysed. Baker used the Behavioural and Mood Disturbance scale (BMD), the REHAB, the CAPE and MMSE to assess patients mood, behaviour and cognition after (but not immediately after ) four treatment sessions and eight treatment sessions. Some assessments were carried at home, some at day hospital. There were many subscores and mostly there were no differences between treatment and control. The following significant differences were found with benefit in favour of snoezelen compared with control after four sessions: apathy score of the BRS (CAPE) (MD -3.00, 95%CIs -5.87 to -0.13, P=0.04), after eight sessions: mood score of the BRS (CAPE) (MD -2.60, 95%CIs -4.92 to -0.28, P=0.03), total score of the BRS (CAPE) (MD -6.92, 95%CIs -13.13 to -0.7, P=0.03), speech skills of the REHAB (MD 1.46, 95%CIs 0.01 to 2.82, P=0.03), psychomotor subscore of the cognitive assessment scale of CAPE (MD -3.12, 95%CIs -5.31 to -0.93, P<0.01).

{mospagebreak}Reviewers' conclusions: Two trials were reviewed. Although both studies examined the short-term values of snoezelen on people with dementia, it is not feasible to draw a firm conclusion at this stage, for two main reasons. Firstly, very limited data were available for analysis, thus limiting data inference and generalization. Secondly, different methodology and control conditions were adopted in the two trials. Such variations not only require a careful interpretation of results but also make the comparison of results across studies less valid. Hence, there is an urgent need for more systematic and scientific research studies to examine the clinical value of snoezelen for people with dementia. To our knowledge, there are four RCTs currently in progress. It is hoped that the data and results of these trials will enrich the systematic review of snoezelen for dementia in the next update.

Background

Derived from two Dutch words, 'sniff' and 'doze', snoezelen was first introduced in the 1970s as an intervention for people with learning disabilities, based on the rationale of reducing the aversive effects of sensory deprivation. Owing to their reduced cognitive abilities, people with learning disabilities are less ready to explore their environments for sensory inputs, and consequently they are likely to be deprived of adequate sensory stimulation. The expression of negative emotions and behaviours, such as vocally disruptive, self-stimulating, and apathetic behaviours, has been found to be associated with sensory deprivation (Cariaga 1991, Cohen-Mansfield 1997, Hallberg 1993). Adopting a non-directive and enabling approach, snoezelen encourages people with reduced cognitive functions to engage with sensory stimuli in a positive and non-stressful environment (Baker 1997, Hope 1998, Hutchinson 1994).

Snoezelen has been described as a 'sensory cafeteria' or 'multi-sensory environment' because of its use of a variety of sensory-based materials and equipment. Pinkney 1997 describes snoezelen as a medium of providing sensory stimuli to the primary senses of sight, hearing, touch, taste and smell, through the use of lighting effects, tactile surfaces, meditative music and the odour of relaxing essential oils. Some researchers however, regard snoezelen as a 'multi-sensory therapy' in which people with dementia are encouraged to engage in a cognitively less demanding sensory environment (e.g. Burns 2000). The goals of such therapy are to promote positive behaviours and to reduce maladaptive behaviours (Baker 1997, Slevin 1999). Whether snoezelen should be considered simply as a multi-sensory environment or as a therapeutic medium has stimulated significant debate. Proponents of the former school of thought have pointed out that the value of snoezelen lies in its aesthetic quality, and its use as a therapy undermines this characteristic (e.g. Hutchinson 1994). Supporters of the therapeutic value of snoezelen are keen to explore its benefits for individuals with cognitive impairments (e.g. Hulsegge 1987, Kewin 1994). In this review, snoezelen is regarded as a multi-sensory based intervention with embedded therapeutic values.

Over the past decade, the clinical application of snoezelen has been extended from the field of learning disabilities to the care of people with dementia. To a certain extent these two groups of individuals share the common characteristics of reduced cognitive functions and diminished communicative ability. However, people with dementia generally experience a gradual deterioration in all aspects of cognitive functions as the disease progresses. This progressive loss of cognitive abilities makes this group less suitable to participate in interventions that demand cognitive functions and language ability. In addition, people with dementia are less competent and have a lower stress threshold for coping with environmental demands (Hall 1987, Lawton 1986).

{mospagebreak}Maladaptive behaviours and affect occur when environmental stimulation exceeds an individual's adaptive level. On the other hand, too little sensory stimulation may lead to a decline in both cognition and function, and an increase in behavioural symptoms (Kitwood 1992, Kovach 1997). Based on these two hypothesis of sensory overload and sensory deprivation, Kovach 2000 put forward the model of sensoristasis, in which an equilibrium of the sensory state can be attained by balancing the pacing of sensory stimulating or sensory-calming activity.

The value of multi-sensory interventions has been documented in promoting relaxation and positive behavioural changes (Deakin 1995, Hutchinson 1994). Through the provision of non-sequential and unpatterned sensory stimuli, multi-sensory interventions capitalize on the residual sensorimotor abilities of dementia sufferers and present few attentional and intellectual demands on them (e.g. Baker 1997, Beatty 1998, Buettner 1999, Hope 1998). Moffat and colleagues (Moffat 1993) pioneered the use of snoezelen for people with moderate to severe dementia and found that they enjoyed the sensory stimuli and remained calm during the sessions. These encouraging results promote the use of multi-sensory interventions in dementia care, and have provoked waves of clinical research on the examination of therapeutic values of snoezelen for people with dementia. A review of the literature shows that snoezelen is commonly employed as a therapeutic modality in dementia care in four areas: (1) reducing maladaptive behaviours and increasing positive behaviours (e.g. Baker 1997, Hope 1998, Long 1992), (2) promoting positive mood and affect (e.g. Baker 1997, Pinkney 1997), (3) facilitating interaction and communication (Spaull 1998), and (4) promoting a caregiving relationship and reducing caregiving stress (e.g. McKenzie 1995, Savage 1996).

Although snoezelen has become a popular clinical intervention for people with dementia, its application often varies in form, principles, duration, and subject groups. For instance, some researchers apply snoezelen in the form of structured procedures and sensory stimuli to groups of individuals with dementia, whereas others encourage dementia sufferers to explore sensory stimuli based on personal choices. Such variations in application make the assessment of the therapeutic value of snoezelen difficult, which in turn undermines the clinical development of Snoezelen in dementia care. A systematic review of evidence for the efficacy of Snoezelen in the care of people with dementia is deemed necessary to inform future clinical applications and research directions.

Objectives

This review aims to examine the clinical effectiveness of snoezelen as a therapeutic intervention for older people with dementia.

Criteria for considering studies for this review

Types of studies

All randomized controlled trials in which snoezelen or multi-sensory stimulation programmes were used as an intervention for people with dementia are included in this review. Clinical trials without randomization might be discussed but were not included for review.

{mospagebreak}Types of participants

People aged over 60 years, suffering from any type of dementia (e.g. Alzheimer's disease, vascular dementia), of any degree of severity. The operational definition of dementia is based on the criteria used in DSM-IV (APA 1994), ICD-10 (WHO 1993), or NINCDS-ADRDA (National Institute of Neurological and Communicative Disorders and Stroke - Alzheimer\'s Disease and Related Disorders Association, McKhann 1984).

Types of intervention

The review includes intervention programmes based on snoezelen (multi-sensory stimulation) principles that have been compared with other intervention programmes (e.g. music) or with no treatment.

Types of outcome measures

Outcomes related to the use of snoezelen for people with dementia are measured in terms of behaviour, mood, speech, and cognition.

Search strategy for identification of studies

See: Cochrane Dementia and Cognitive Improvement Group search strategy

The trials were identified from a search of the Specialized Register of the Cochrane Dementia and Cognitive Improvement Group on 5 April 2001 using the terms snoezelen, multi-sensory and "multi sensory".

  • The Specialized Register at that time contained records from the following databases:
  • CCTR/Central: January 2001 (issue 1);
  • Medline: 1966 to December 2000;
  • Embase: 1980 to December 2000;
  • PsycLit: 1887 to December 2000;
  • Cinahl:1982 to December 2000;
  • SIGLE (Grey Literature in Europe): 1980 to December 2000;
  • ISTP (Index to Scientific and Technical Proceedings): to May 2000;
  • INSIDE (BL database of Conference Proceedings and Journals): to June 2000;
  • Aslib Index to Theses (UK and Ireland theses): 1970 to March 2001;
  • Dissertation Abstract (USA): 1861 to March 2001;
  • ADEAR (Alzheimer\'s Disease Clinical Trials Database): to March 2001;
  • National Research Register (including the MRC Clinical Trials Directory): January 2001 (issue 1)
  • Alzheimers Society Trials Database: to March 2001;
  • Glaxo-Wellcome Trials Database: to March 2001;
  • Centerwatch Trials Database: to December 2000.

 

 

  • The search strategies used to identify relevant records in Medline, Embase, PsycInfo and Cinahl can be found in the Group's module in the Cochrane Library.
  • In addition the reviewers searched the Science Citation Index and Social Science Citation Index (1974-2001 Sept) using the search terms "Snoezelen", "multi-sensory", "dement", "Alzheimer", "randomized control/single control/double control".
  • Journals or articles related to snoezelen/ multi-sensory and Alzheimer/ dementia were selected for further screening.

{mospagebreak}Methods of the review

 

SELECTION OF STUDIES

Based on the search strategy, a total of 16 research papers were identified. Two reviewers independently studied the abstracts and the full text of these papers, and then selected those trials that met the inclusion criteria. Any disagreements were discussed and resolved between the reviewers. Twelve papers were screened out: one examined non-dementia clients, six were review or discussion papers, one did not employ snoezelen or multi-sensory programme, one reported an observational study, and three were case studies with no randomizations. Four papers fulfilled the inclusion criteria, but they only reported two studies, as each study had been published in two papers. Baker 1997 appeared in two papers, with the 1997 one reporting the preliminary results of the trial while the 2001 one reported the completed trial with a full data set of 50 subjects. Kragt 1997 appeared in two papers that were both written in Dutch and published in 1997. English abstracts were available in both papers. With the help of the editorial base of CDCIG, English translations from the Dutch of the key points of the trial and the outcome measure were obtained.

 

The search of the National Research Register generated four ongoing studies that are examining the clinical values of snoezelen or multi-sensory programs on people with dementia. The results of these studies are not yet ready for publication, and therefore they are included in the Table of Ongoing Studies.

 

DATA EXTRACTION

The investigators of the Baker 1997 study provided all the individual patient data for all assessments, together with papers and questionnaires explaining the rating scales used. Statistics required for the review were derived from analysis of these data. Data for the Kragt 1997 study were extracted from the published reports, but only for the first period of this crossover study.

 

 

Quality assessment of included studies

For the RCTs, the quality of the studies was assessed to minimize potential sources of systematic bias. The criteria assessed were as outlined in the Cochrane Reviewer' Handbook (Clarke 2000), and included allocation concealment, blinding, level of dropout at follow-up stage, and differences in outcome measures. Unlike pharmacological studies, blinding of psychosocial interventions is not always possible as subjects are generally aware of the nature of the intervention they are receiving. Studies in which an assessor is blind to the intervention were considered to be of higher quality than those in which the same person performed both intervention and assessment. Blinding of intervention was considered as far as possible, despite the fact that subjects were generally aware of the nature of intervention they were receiving.

 

 

Other forms of quality assessment of the studies included:

(a) The number and characteristics (e.g. level of cognitive functioning) of subjects involved in the study.
(b) The format and duration of intervention and control conditions.
(c) The levels of data utilized.
(d) How far the recommendations were derived from the results.

{mospagebreak}Description of studies

 

Study objectives:

  • Adopting a parallel group experimental design, Baker 1997 investigated

 

 

  • the immediate effects of multi-sensory stimulation (MSS; intervention) on mood, speech and behaviours of older people with dementia,
  • the carry-over effects of MSS on mood and behaviour to day hospitals and home environments, and
  • the maintenance effects of MSS on mood, behaviour, and cognition over time. Kragt 1997, using a randomized crossover design, examined the behaviours of the subjects with dementia during their participation in snoezelen sessions (i.e. the immediate effect).

 

 

Subject recruitment and assignment:

  • Fifty subjects except one recruited in Baker 1997's study were aged over 60 years, and their average age was 78 years. Kragt 1997 recruited 17 subjects who were all aged over 65 years, and their average age was 86 years. The subjects of both studies received a diagnosis of dementia. Baker 1997's subjects were classified as being in the moderate to advanced stage of dementia (mean MMSE of intervention group=10.96, SD=6.5; mean MMSE of control condition=6.08, SD=5.07) and attended day hospitals whereas Kragt 1997 recruited nursing home residents who were in the advanced stage of dementia (no data on their MMSE score were given).
  • Using a computer-generated randomization system, 50 subjects of Baker 1997's study were randomly allocated to the intervention group or the control group, in which a one-to-one activity program was provided. Using an automated randomization programme, 16 subjects of Kragt 1997's study were randomly assigned to two different intervention sequences AB or BA (A=Snoezelen; B=control).

 

 

Experimental and control condition:

  • Baker 1997: The subjects attended eight 30-minute sessions of their designated programme twice a week over a 4-week period. Both experimental and control programmes were designed with a similar structure, except that the former (multi-sensory stimulation) adopted a non-directed and enabling approach. The experimental subjects explored and received unpatterned and non-sequential sensory stimuli in a multi-sensory environment that placed no intellectual or physical demands on them. The control condition was a one-to-one activity programme that was based on individual subjects' preferences and abilities. To keep them distinct from the experimental intervention, objects that provided obvious sensory inputs were not used in the activity sessions.

 

Kragt 1997: According to their assignment sequence, the experimental subjects attended 3 sessions of snoezelen in the snoezel-room on three consecutive days, each at a different time and with a different activity supervisor. Each snoezelen session ran according to an individualized snoezel plan made on the basis of a 10-hour screening period, and lasted between 30 minutes and one hour. The control subjects attended 3 sessions in the living room on three consecutive days, using the same schedule as with the snoezelen sessions. During the 30-minute control sessions, the subjects moved freely among other residents in the living room and received 'normal' attention from carers. There was a washout period of 4 days before the subjects proceeded to their next intervention assignment.

{mospagebreak}Outcome measures:

  • Seven outcome measure was used in Baker 1997's study. The immediate effects, including within sessions and immediately after sessions, of both experiemental and control programmes were measured by the INTERACT and its short-form respectively (Baker 1995). The generalization effect and longer-term effects were measured by five outcome measure. The General Behaviour and Community Skills sub-scales of REHAB (Baker 1988) measured the carryover effect to day hospitals, and the Behaviour and Mood Disturbance Scale (BMD) and the Behaviour Rating Scale (BRS) of the Clifton Assessment Procedures for the Elderly (CAPE) measured the carryover effect to the home, at pre-trial, mid-trial and post-trial. The post-trial effect and the maintenance effect (at the 1-month follow-up) were measured by REHAB, BMD, the Cognitive Assessment Scale (CAS) of CAPE and MMSE.

 

Kragt 1997 used the Behavioural Observation Scale for Intramural Psychogeriatrics (Gedragsobservatieschaal voor de Intramurale Psychogeriatrie, GIP) to measure the behavioural problems presented by subjects during different interventions. Subjects' behaviours in the sessions were video recorded. The middle ten minutes of each recording were used to score the subjects' behaviours.

Methodological quality

 

SELECTION BIAS

  • Baker 1997: A computer-generated randomization system was used to assign subjects to the experimental group (multi sensory stimulation sessions) or the control group (activity sessions).
  • Kragt 1997: The subjects included in this study were known to respond positively to snozelen, and therefore, it would potentially bias the results in favour of the treatment. An automated randomization programme was then used to allocate the subjects to the intervention sequence AB or BA.

 

 

PERFORMANCE BIAS

  • Baker 1997: All staff, including keyworkers, hospital staff, research assistants, and carers, were aware that a study was being carried out in which two equally valid therapies were implemented for comparison. All keyworkers involved were asked to implement the programmes (both treatment and control) based on the written guidelines and standardized procedures. The subjects were not blind to the study because of their participation in the programmes.
  • Kragt 1997: The subjects were not blind to the study because of their participation in the programmes. Both the activity supervisors for the experimental sessions and the carers for the control sessions were blind to the study, but the presence of video cameras could have influenced them.

 

 

ATTRITION BIAS

  • Baker 1997: There were two dropouts in the experimental group but no explanation was given.
  • Kragt 1997: One subject died immediately after randomization, so only 16 subjects were included for data analysis.

{mospagebreak}DETECTION BIAS

  • Baker 1997: INTERACT was rated by the keyworkers. INTERACT short and REHAB were rated by day hospital staff. BMD was rated by family carers at home with the aid of a research assistant; BRS, MMSE, and the CAS of CAPE were also rated by three research assistants. All raters were aware of the study and both programmes were presented as two equally valid therapies for comparison.
  • Kragt 1997: The two researchers of this study used GIP to rate independently the subjects\' behavioural performance during sessions, so they were not blind to the study.

 

Results

Immediate effects:

 

  • Within sessions

 

Both Baker 1997 and Kragt 1997 examined the immediate effects of snoezelen (or multi-sensory stimulation) on behaviours during the subjects' participation in the sessions. The results of the two studies were entered for meta-analysis with a total of 64 subjects (31 in the experimental group, 25 in the control group). Analyses were adjusted to the random effects model due to the heterogeneity of the trials, and standardized mean differences (SMD) were used because different outcome measures were employed in the two trials (Baker 1997 used the INTERACT and Kragt 1997 the GIP). The pooled results of the total scores were insignificant, but the trend was in the direction of favouring treatment (hence a negative value of the SMD). The standardized mean difference (SMD) was -1.22 and a 95% CI (-4.08, 1.64) P=0.42. Kragt 1997\'s result was significant in favour of treatment, unlike that of Baker 1997. The control treatment was quite different in the two trials and this may be the cause of the heterogeneity. During the snoezelen sessions, Kragt 1997's subjects presented significantly fewer apathetic behaviours (t=-8.22, p<0.01), fewer restless behaviours (t=-3.00, p=0.01), fewer repetitive behaviours (t=-.822, p<0.01), and fewer disturbances (t=-4.91, p<0.01). The results from Baker 1997 show no differences between treatment and control for any subscore or the total of INTERACT.

 

  • Immediately after treatment sessions

 

 

  • Kragt 1997 did not examine the effect of snoezelen immediately after the sessions, so only Baker 1997's results were analysed. Behavior and mood were examined using the INTERACT rating scale. There were no significant treatment differences between snoezelen and control for any subscore.
  • Some time after treatment sessions (4 and 8 sessions).

 

 

  • Baker 1997 used the Behavioural and Mood Disturbance scale (BMD), the REHAB, the CAPE and MMSE to assess patients mood, behaviour and cognition after (but not immediately after ) four treatment sessions and eight treatment sessions. Some assessments were carried at home, some at day hospital. There were many subscores and mostly there were no differences bettwen treatment and control. The following significant differences were found with benefit in favour of snoezelen compared with control after four sessions: apathy score of the BRS (CAPE) (MD -3.00, 95%CIs -5.87 to -0.13, P=0.04), after eight sessions: mood score of the BRS (CAPE) (MD -2.60, 95%CIs -4.92 to -0.28, P=0.03), total score of the BRS (CAPE) (MD -6.92, 95%CIs -13.13 to -0.7, P=0.03), speech skills of the REHAB (MD 1.46, 95%CIs 0.01 to 2.82, P=0.03), psychomotor subscore of the cognitive assessment scale of CAPE (MD -3.12, 95%CIs -5.31 to -0.93, P<0.01).

 

 

  • Long-term effects
  • Kragt 1997 also did not investigate the long-term effect of snoezelen, so only the results of Baker 1997's study, assessed one month after treatment had finished, using the BMD and REHAB scores, were analysed. There were no significant differences between snoezelen and control.

{mospagebreak}Discussion

Owing to the limited data obtained from the two RCTs, it is not feasible to reach an affirmative conclusion in this review. However, they further our understanding of the clinical values of snoezelen (multi-sensory stimulation) for people with dementia. The subjects of both Baker 1997 and Kragt 1997 had moderate to severe level of cognitive impairments, which limited much of their participation in normal social and leisure activities. Snoezelen programmes were provided to these subjects with the aims of reducing their maladaptive behaviours (Kragt 1997) and promoting their mood and behaviours (Baker 1997). Although the pooled results of the two studies did not demonstrate a significant result in favour of snoezelen, they independently demonstrated significant results in favour of snoezelen. Regarding the short-term effects, Kragt 1997's subjects presented significantly fewer behavioural problems (e.g. apathy, restlessness) during the snoezelen sessions than the control sessions. Baker 1997's subjects were more responsive to their surrounding environments immediately after the sessions.

However, it is important to note that nature of the experimental and control conditions were different in these two RCTs, thus caution is required when interpreting these data, particularly the very encouraging results of Kragt's study. As for the control condition, Baker 1997 used one-to-one individualized activity sessions to compare with one-to-one multi-sensory stimulation sessions, whilst Kragt 1997 adopted usual care programs to compared against the one-to-one snozelen sessions. One major limitation of Kragt 1997 is the failure to provide a credible control condition because the usual care program (control condition) did not address the one-to-one attention that was given in the snoezelen sessions (experimental condition). It is possible that the positive benefits seen in Kragt 1997's study may not be entirely due to snozelen but also the one-to-one attention. Indeed, the one-to-one attention and the tailored-made activity programmes provided in Baker 1997's control condition might account for the the positive behaviours presented by the control subjects. As for the experimental condition, Baker 1997 provided the subjects with a non-structured exploration of the snoezelen equipment whereas Kragt 1997 developed individualized snoezelen plans for the subjects. The significant better behaviours presented by Kragt 1997's subjects during the snoezelen sessions might be due to the fact that their individual sensory perferences and needs were met.

Compared to the short-term effects, the generalization effects and long-term effects of snoezelen programmes were examined far less. Baker 1997 reported a significant positive carryover effect of the multi-sensory interventions (snoezelen) on subjects' mood and behaviours to home environments in which they presented fewer social disturbing behaviours and better mood than the control subjects. This contradicts the findings of Moffat 1993, where no generalization of behaviour to home environments was found. However, the small subject size and the non-randomized control nature of Moffat 1993 affect the power of the analyses. There was no carryover effect to day hospitals, nor any other longer-term effects (1-month follow up after the intervention) identified in Baker 1997. Moreover, the gains obtained through the snoezelen programmes were quickly lost after the sessions had ceased. In other words, the therapeutic effects of snoezelen or multi-sensory programmes on mood and behaviours were short-lived and not sustained over time. This is similar to the findings of some non-RCT studies (e.g. Moffat 1993). Baker 1997 suggested two reasons explaining the quick deterioration of gains obtained by the subjects during the trial. Firstly, staff and carers were aware of the cessation of the programme and might have unintentionally changed their behaviours and perceptions towards the subjects. Secondly, there might be a negative withdrawal effect in which the subjects demonstrated even more difficulty in coping with the cessation of the one-to-one attention that they had received during the trial.

{mospagebreak}In terms of programme design and implementation, both Baker 1997 and Kragt 1997 shared some similar characteristics: (1) keyworkers/supervisors' use of a nondirective 'enabling' approach', (2) placing no intellectual demands on the subjects, (3) adopting an one-to-one basis for implementation, and (4) about 30-minute implementation for each session.

Differences, however, were noted in the two trials and other non-controlled trials. Firstly, snoezelen has a varied definition and application. Some refer it as a general approach whilst some refer it as a specially designed environment with a particular protocol. Kragt 1997 developed individualized snoezelen protocols and did not limit snoezelen to a snoezelen room, but was integrated it in subjects' daily activities such as dressing and grooming. Indiviudalized snoezelen protocols may be able to capture the sensory preferences and needs of people with dementia (Hope 1998, Moffat 1993). Baker 1997, Long 1992 used the snoezelen room was the primary independent variable and provided the subjects witn a non-structured exploration of snoezelen equipment . Secondly, differences in duration and frequency of snoezelen sessions are noted in the two RCTs and other non-RCTs. For instance, Baker 1997 implemented two snoezelen session every week over a period of four weeks whereas Kragt 1997 carried out three snoezelen on three consecutive days, each at a different time. Lastly, the natures of control can be ranged from usual care programmes (e.g. Kragt 1997, Long 1992), to music relaxation groups (e.g. Pinkney 1997), and to individualized activity sessions (e.g. Baker 1997). With a variation of definition and implementation pattern of snoezelen, and the nature of control condition, any comparison with regard to the outcomes or effects of the intervention will naturally be subject to challenge.

Reviewers' conclusions

Implications for practice

Given the limited empirical data available for this review, there is little to evaluate the clinical effectiveness of snoezelen (or multi-sensory interventions) as a therapeutic intervention for older people with dementia. Thus, it is not feasible to reach an affirmative conclusion at this stage. However, this very first systematic review on snoezelen for dementia sheds light for future research and practice.

From the practice perspective, snoezelen programmes demonstrate positive immediate outcomes in reducing maladaptive behaviours and promoting positive behaviours, suggesting that it should be considered as part of the general dementia care programme. However, the limited carryover and long-term effects of snoezelen programmes suggest that a continuous and ongoing programme should be implemented. In addition, the selection criteria of participants and protocols of snoezelen (or multi-sensory interventions) should be further developed for two main purposes: firstly, to promote understanding and communication between practitioners on the use of snoezelen in dementia care; and secondly, because such information can be used for replication studies to explore the clinical value of snoezelen for people with dementia.

Implications for research

From the research perspective, the major work ahead is to conduct more empirical and scientific studies of snoezelen for people with dementia. First of all, the methodological vigour of snoezelen studies should be further enhanced, as most current snoezelen studies are non-controlled case studies that result in relatively weak data inference. Secondly, the natures of both interventions and controls were different in the two RCTs examined in this review, thus making the comparison and generalization of results less feasible. Further research should focus more on the design of snoezelen programmes (e.g. nature, frequency and duration) that are comparable to the existing RCTs, as well as the control conditions that are comparable with the intervention. Thirdly, the existing research provides little information regarding at which stage of dementia (or level of cognitive impairment) clients can benefit most from snoezelen programmes, thus it would be worth examining this aspect in future research. Fourthly, there have been limited studies examining the carry-over and long-term effects of snoezelen for people with dementia, so further research should be performed to examine these areas. Finally, the literature has suggested that snoezelen might promote therapeutic relationships and quality of care, but good quality empirical studies on these aspects are lacking (e.g. McKenzie 1995, Morrissey 1997, Savage 1996). Further research could be done to examine whether snoezelen promotes a therapeutic relationship between clients and staff, as well as looking at the quality of care.

The clinical application of snoezelen needs to be adequately supported by scientific evidences, despite its popular use in dementia care. Without a well-developed evidence-based practice, snoezelen will merely be used as a general programme to occupy people with dementia without a meaningful purpose. Moreover, resources such as the manpower and costs of setting up a Snoezelen environment cannot be justified without such evidence. In conclusion, there is an urgent need for more reliable and useful research-based evidences to inform and justify the use of snoezelen in dementia care.

Acknowledgements

The team would like to thank Professor Baker and his team, in particular Jane Holloway, in providing us the data file and copies of the rating scale and questionnaires for the meta analysis. Jacqueline Birks of the CDCIG must be thanked for her assistance and coaching in our process of data analysis. Also, Grace Lee, advanced occupational therapy practitioner in Psychogeriatrics, is thanked for her advice given to the protocol.

Potential conflict of interest

None known

{mospagebreak}Tables

Characteristics of included studies

 

 

Study Baker 1997 
Methods  Randomized controlled trial for a period of 4 weeks.Concealment of treatment was not reported.Two dropouts with no reasons given. 
Participants  50 subjects (25M, 25F) with separate diagnoses of Alzheimer's disease (N=33; N=15 MSS); N=18 Activity), vascular dementia (N=7; N=5 MSS; N=2 Activity) or a mixed diagnosis (N=10; N=5 MSS; N=5 Activity). The subjects had moderate to severe levels of cognitive impairment (MMSE: 0-17).Mean age was 78 but one was aged below 60 years. Informed consent was obtained from carers. Blinding of subject allocation was not discussed in the paper. Two dropped out of the experimental group, but no reason was given. 
Interventions  RCT design to compare the effects of eight standardized multi-sensory programmes (experiemental), with a credible control condition of eight standardized activity sessions. Both programmes were implemented twice a week, each of 30 minutes, for all subjects. Both were of equivalent structure except in the keyworker approach, multisensory experience, nature of stimuli, and demands on clients. 
Outcomes  Immediate effect:1. Within session effect was measured by INTERACT (22 items)2. Immediate after session effect was measured by INTERACT short(12 items) Carry-over and long-term effect: 1. Behaviours at day hospital: REHAB (general behaviour subscale and deviant behaviour subscale)2. Behaviour and mood at home: Behaviour and Mood Disturbance Scale (BMD) and Behaviour Rating Scale (BRS) of CAPE 3. Cognition: MMSE Cognitive Assessment Scale (CAS) of CAPEAssessments were done at four points: pre-trial, mid-trial, post-trial, and follow-up one month later 
Notes 
Allocation concealment 
Study Kragt 1997 
Methods  Crossover trial 
Participants  17 subjects, one died almost immediately after randomization, thus 16 subjects were left.16 females, and 1 male (unclear if the person who died was male or female).Mean age was 86 years old, range of 79-97.All were in a very advanced stage of dementia. Proxy informed consent was obtained. Owing to subjects\'very reduced cognitive abilities, blinding of subjects was not necessary. 
Interventions  Crossover trial to examine the effect of snoezelen in the snoezel-room as compared to usual care (control). Subjects were randomly allocated to either a AB or BA intervention pattern, with a 4 days washout in between. Each subject participated in 3 snoezel sessions in the snoezel-room on three consecutive days, each at a different time and with a different activity supervisor. Each snoezel session lasted between 30 and one hour. Similarly, each subject participated in three sessions in the living room (control) on three consecutive days using the same schedule as used for the snoezel sessions. 
Outcomes  Subjects' behaviours were videotaped and measured by two raters who were blind to the study. Four subscales of 'Gedragsobservatieschaal voor de Intramurale Psychogeriatrie' (GIP) was used to measure: consciousness disturbance, senseless repetitive behaviours, apathetic behaviours, and restless behaviours. 
Notes 
Allocation concealment 

 

Characteristics of excluded studies

 

 

StudyReason for exclusion
Pinkney 1997 Non-randomized methodological quality, and an absence of comparative data.

 

{mospagebreak}Characteristics of ongoing studies

 

 

StudyTrial name or titleParticipantsInterventionsOutcomesStarting dateContact informationNotes
Baillon 1999 A pilot study of the long-term and short-term effects of sensory treatment therapy (Snoezelen) on patients suffering from dementia, and associated agitation. 16 subejcts with a diagnosis of dementia, will be randomly assigned to either Snoezelen therapy or a control intervention. 30-minute Snoezelen sessions over 6 weeks. Pattern and duration are same for the control condition. Clinical Dementia Rating Scale
Mini-Mental State Examination
The Cohen-Mansfield Agitation Inventory
The Cohen-Mansfield Agitation Behavioural Mapping Instrument
Physiological arousal (heart-rate monitor)
31/01/1998 Sarah Baillon
Psychiatry for the Elderly, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW UK
Full paper titled 'A pilot study of the physiological and behavioural effects of Snoezelen in dementia' will be published in the British Journal of Occupational Therapy in early 2002.
Baillon 2000 A study of the long-term and short-term effects of Snoezelen and reminiscence therapy on patients suffering from dementia who have associated agitated behaviour problems No information
A crossover trial will be used.
Intervention: 3 Snoezelen sessions over 2 weeks.
Control: 3 reminiscence sessions over 2 weeks.
Heart rate
Cohen-MansfieldAgitated Behavioural Mapping Instrument, after each session.
Cohen-Mansfield Agitation Inventory, after each intervention and at 2-weel follow up.
01/02/2000 Sarah Baillon
Psychiatry for the Elderly, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW UK
Data collection is still ongoing.
Bell 2000 The use of Dementia Care Mapping as a tool for evaluating Snoezelen as a recreational activity for older people with dementia. No information
Multiple single case experiemental design will be used.
Snoezelen actiivity Dementia Care Mapping 01/04/1999 Maggie Bell
Professional Development Nurse, Wakefiled and Pontefract Community Health NHS Trust, Wakefield WF1 3SP UK
Researcher contacted but no reply given.
Creaney 2000 Comparison of specific sensory stimulation with multi-sensory stimulation of people with dementia in a community setting: A pilot study No information
Crossover design will be adopted
Two intervention, massage and multi-sensory stimulation INTERACT
CAMCOG
Behavioural and Mood Disturbance Scale
01/04/2000 Dr. William Creaney
Medical Office, Ailsa Hospital
Dealmellington Road, Ayr, Scotland.
Project not completed.

 

{mospagebreak}References

References to studies included in this review

Baker 1997 {published data only}

Baker R, Bell S, Baker E, Gibson S, Holloway J, Pearce R, Dowling Z, Thomas P, Assey J, Wareing LA. A randomized controlled trial of the effects of multi-sensory stimulation (MSS) for people with dementia. British Journal of Clinical Psychology 2001;40(1):81-96.

Baker R, Dowling Z, Wareing LA, Dawson J, Assey J. Snoezelen: Its long-term and short-term effects on olkder people with dementia. British Journal of Occupational Therapy 1997;60(5):213-218.

 

Kragt 1997 {published data only}

Holtkamp CCM, Kragt K, van Dongen MCJM, van Rossum E, Salentijn C. [Effecten van snoezelen op het gedrag van demente ouderen]. Tijdschrift voor Gerontologie en Geriatrie 1997;28:124-128.

Kragt K, Holtkamp CCM, van Dongen MCJM, van Rossum E, Salentijn C. [Het effect van snoezelen in de snoezelruimte op het welbevinden van demente ouderen]. Verpleegkunde 1997;12(4):227-236.

 

* indicates the major publication for the study

References to studies excluded from this review

Pinkney 1997

Pinkney L. A comparison of the Snoezelen environment and a music relaxation group on the mood and behaviour of patients with senile dementia. British Journal of Occupational Therapy 1997;60(5):209-18.

 

Ongoing studies

Baillon 1999

Sarah Baillon Psychiatry for the Elderly, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW UK. A pilot study of the long-term and short-term effects of sensory treatment therapy (Snoezelen) on patients suffering from dementia, and associated agitation.. Ongoing study 31/01/1998.

 

Baillon 2000

Sarah Baillon Psychiatry for the Elderly, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW UK. A study of the long-term and short-term effects of Snoezelen and reminiscence therapy on patients suffering from dementia who have associated agitated behaviour problems. Ongoing study 01/02/2000.

{mospagebreak}Bell 2000

Maggie Bell Professional Development Nurse, Wakefiled and Pontefract Community Health NHS Trust, Wakefield WF1 3SP UK. The use of Dementia Care Mapping as a tool for evaluating Snoezelen as a recreational activity for older people with dementia.. Ongoing study 01/04/1999.

 

Creaney 2000

Dr. William Creaney Medical Office, Ailsa Hospital Dealmellington Road, Ayr, Scotland.. Comparison of specific sensory stimulation with multi-sensory stimulation of people with dementia in a community setting: A pilot study. Ongoing study 01/04/2000.

 

Additional references

APA 1994

American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th (DSM-IV) Edition. Washington, DC: American Psychiatric Association, 1994.

 

Baker 1988

Baker R, Hall J. REHAB: A new assessment for chronic psychiatric patients. Schizophreniz Bulletin 1988;14:97-111.

 

Baker 1995

Baker R, Dowling Z. INTERACT. Research adn Development Support Unit, Poole Hospital, Dorset 1995.

 

Baker 1997

Baker R, Dowling Z, Wareing LA, Dawson J, Assey J. Snoezelen: Its long-term and short-term effects on older people with dementia. British Journal of Occupational Therapy 1997;60(5):213-8.

 

Beatty 1998

Beatty W, Zavadil K, Bailly R, Rixen G, Zavadil L, Farnham N, Fisher L. Preserved musical skill in a severely demented patient. International Journal of Clinical Neuropsychology 1998;10:158-64.

 

Buettner 1999

Buettner LL. Simple pleasures: A multilevel sensorimotor intervention for nursing home residents with dementia. American Journal of Alzheimer's Disease 1999;14:41-51.

{mospagebreak}Burns 2000

Burns I, Cox H, Plant H. Leisure or therapeutics? Snoezelen and the care of older persons with dementia. International Journal of Nursing Practice 2000;6:118-26.

 

Cariaga 1991

Cariaga J, Burgio L, Flynn W, Martin D. A controlled study of disruptive vocalizations among geriatric residents in nursing homes. Journal of the American Geriatrics Society 1991;39:501-7.

 

Clarke 2000

Clarke M, Oxman AD. In: Cochrane Reviewers' Handbook 4.1 (updated June 2000); Section 6. In: Review Manager (RevMan) (Computer Program). Version 4.1 Oxford, England: The Cochrane Collaboration, 2000.

 

Cohen-Mansfield 1997

Cohen-Mansfield J, Werner P. Typology of disruptive vocalizations in older persons suffering from dementia. International Journal of Geriatric Psychiatry 1997;12:1079-91.

 

Deakin 1995

Deakin M. Using relaxation techniques to manage disruptive behaviour. Nursing Times 1995;91:40-41.

 

Dowling 1997

Dowling Z, Baker R, Wareing LA, Assey J. Lights, sound and special effects. Journal of Dementia Care 1997;(January/February):16-18.

 

Ganong 1987

Ganong LH. Integrative Reviews of Nursing Research. Research in Nursing & Health 1987;10:1-11.

 

Greene 1982

Greene JG, Smith R, Gardiner M, Timbury GC. Measuring behavioural disturbance of elderly demented elderly patients in the community and its effects on relatives: A factor analytic study. Age and Ageing 1982;11:121-6.

 

Hall 1987

Hall GR, Buckwalter KC. Progressively lowered stress threshold: A conceptual model for care of adults with Alzheimer\'s disease. Archives of Psychiatric Nursing 1987;1(6):399-406.

{mospagebreak}Hallberg 1993

Hallberg IR, Edberg AK, Nordmark A. Daytime vocal activity in institutionalized severely demented patients identified as vocally disruptive by nurses. International Journal of Geriatric Psychiatry 1993;8:155-64.

 

Hope 1998

Hope K. The effects of multisensory environments on older people with dementia. Journal of Psychiatric and Mental Health Nursing 1998;5:377-85.

 

Hulsegge 1987

Hulsegge J, Verheul A. Snoezelen: Another world. Chesterfield: Rompa, 1987.

 

Hutchinson 1994

Hutchinson R, Hagger L. In: Hutchinson R & Kewin J, editor(s). Sensations and disabilty: Sensory environments for leisure, Snoezelen, education and therapy Chesterfield: Rompa, 1994:18-48.

 

Kewin 1994

Kewin J. In: Hutchinson R & Kewin J, editor(s). Sensations and disabilty: Sensory environments for leisure, Snoezelen, education and therapy Chesterfield: Rompa, 1994:6-17.

 

Kitwood 1992

Kitwood T, Bredin A. Towards a theory of dementia care. Aging and Society 1992;12:269-287.

 

Kovach 1997

Kovach CR. Late-stage dementia care: A basic guide. Washington DC: Taylor & Francis, 1997.

 

Kovach 2000

Kovach CR. Sensoristasis and imbalance in persons with dementia. Journal of Nursing Scholorship 2000;32(4):379-384.

 

Lawton 1986

Lawton MP. Environment and aging. Albany, NY: Centre for the Study of Aging, 1986.

{mospagebreak}Long 1992

Long AP, Haig L. How do clients benefit from Snoezelen? An exploratory study. British Journal of Occupational Therapy 1992;55:103-6.

 

McKenzie 1995

McKenze C. Brightening the lives of elderly residents through Snoezelen. Nursing Practice 1995;7:11-13.

 

McKhann 1984

McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer's disease. Neurology 1984;34:939-44.

 

Moffat 1993

Moffat N, Barker P, Pinkney L, Garside M, Freeman C. Snoezelen: An experience for people with dementia. Chesterfield: Rompa, 1993.

 

Morrissey 1997

Morrissey M, Biela C. Snoezelen: Benefits for nursing old clients. Nursing standard 1997;12(3):38-40.

 

Pinkney 1997

Pinkney L. A comparison of the Snoezelen environment and a music relaxation group on the mood and behaviour of patients with senile dementia. British Journal of Occupational Therapy 1997;60(5):209-18.

 

Savage 1996

Savage P. Snoezelen for confused older people: Some concerns. Elderly Care 1996;8:20-21.

 

Slevin 1999

Slevin E, McClelland A. Multisensory environments: are they therapeutic? A single subject evaluation of the clinical effectiveness of a multisensory environment. Journal of Advanced Nursing 1999;8(1):48-56.

 

Spaull 1998

Spaull D, Leach C. An evaluation of the effects of sensory stimulation with people who have dementia. Behavioural and Cognitive Psychotherapy 1998;26(1):77-86.

 

WHO 1993

World Health Organization. The ICD-10 classification of mental and behavioural disorders: Diagnostic criteria for research. Geneva: World Health Organization, 1993.

{mospagebreak}Graphs

Graphs and Tables

To view a graph or table, click on the outcome title of the summary table below.

 

01 Snoezelen versus control

 
Outcome titleNo. of studiesNo. of participantsStatistical methodEffect size
01 Change in behaviours during sessions 64  Standardised Mean Difference (Random) 95% CI  -1.22 [-4.08, 1.64] 

 

 

02 Snoezelen versus control (Baker study only)

 
Outcome titleNo. of studiesNo. of participantsStatistical methodEffect size
01 Change in INTERACT score immediately after treament session compared with score immediately before treatment Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
02 Change in INTERACT score during treatment session compared with score immediately before treatment session Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
03 Change in BMD score after 4 treatment sessions compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
04 Change in REHAB score after 4 treatment sessions compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
05 Change in BMD and CAPE score after 8 treatment sessions compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
06 Change in REHAB score after 8 treatment sessions compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
07 Change in cognitive test score after 8 treatment sessions compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
08 Change in BMD score one month after trial compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 
09 Change in REHAB score one month after trial compared with score at pre-trial Weighted Mean Difference (Fixed) 95% CI  Subtotals only 

 

{mospagebreak}Cover sheet

 

 

Snoezelen for dementia

 
Reviewer(s) Chung JCC, Lai CKY, Chung PMB, French HP
Contribution of Reviewer(s)

-Jenny Chung: all correspondence; selecting trials for inclusion/exclusion, extracting and interpreting data for review, drafting and updating review versions.
-Claudia Lai: selecting trials to be included for reivew, extracting and interpreting data, and assising in drafting the reivew.
-Peter French: searching and collection of trial studies for this review, and assisting in drafting and updating the review.
-Betty Chung: obtaining trial studies and updating the review.

-Contact editor: Mario Fioravanti
-Lay Editor: Grace Lee
-This review has been reviewed by two external peer reviewers

Issue protocol first published 2001 issue 3
Issue review first published 2002 issue 4
Date of last minor amendment 15 August 2002
Date of last substantive amendment 24 June 2002
Most recent changes Information not supplied by reviewer
Date new studies sought but none found Information not supplied by reviewer
Date new studies found but not yet included/excluded Information not supplied by reviewer
Date new studies found and included/excluded 05 April 2002
Date reviewers\' conclusions section amended Information not supplied by reviewer
Contact address Jenny Chung
Hung Hom, Kowloon
HONG KONG
Telephone: 852 27666752
Facsimile: 852 23308656
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Cochrane Library number CD003152
Editorial group Cochrane Dementia and Cognitive Improvement Group
Editorial group code DEMENTIA

 

 

Sources of support

External sources of support

  • No sources of support supplied

Internal sources of support

  • The Hong Kong Polytechnic University HONG KONG

Synopsis

No evidence of the clinical effectiveness of snoezelen or multi-sensory stimulation programmes.

Snoezelen (or multi-sensory stimulation) has become a popular intervention in dementia care, but only two randomized clinical trials were available for this review. Some short-term benefits have been documented in promoting adaptive behaviours in people with dementia during and immediately after their participation in the snoezelen sessions. However, carryover and longer-term effects of snoezelen were not evident. More scientific studies are needed to further examine the clinical value of snoezelen for people at different stages of dementia.

Keywords

Human; Aged; Complementary Therapies[methods]; Dementia[*therapy]; Middle Age; Randomized Controlled Trials; *Sensory Art Therapies

 

Read 23384 times

Random