INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY Int J Geriatr Psychiatry 2005; 20: 446451. Published online in Wiley InterScience (www.interscience.wiley.com).

DOI: 10.1002/gps.1304

A pilot study examining the effectiveness of maintenance Cognitive Stimulation Therapy (MCST) for people with dementia
Martin Orrell1*, Aimee Spector1, Lene Thorgrimsen1 and Bob Woods2
1 2

Department of Mental Health Sciences, University College London, London, UK Dementia Services Development Centre, Department of Clinical Psychology, University of Wales, Bangor, UK

SUMMARY
Background A recent randomised controlled trial on Cognitive Stimulation Therapy (CST) identied the need to evaluate its more long-term benets for people with dementia. This study evaluates the effectiveness of a weekly maintenance CST programme for people with dementia in residential care. Method Thirty-ve people with dementia were included, following on from a seven-week twice-weekly study of CST. The maintenance CST sessions ran in two residential homes using a once a week programme of CST over an additional 16 weeks. Two control homes did not receive the maintenance intervention. Results Using repeated measures ANOVAS, there was a continuous, signicant improvement in cognitive function (MMSE) for those receiving MCST (CST maintenance CST sessions) as compared to CST alone or no treatment (p 0.012). There were no effects on quality of life, behaviour or communication following maintenance sessions. The initial cognitive improvements following CST were only sustained at follow-up when followed by the programme of maintenance CST sessions. Conclusions The cognitive benets of CST can be maintained by weekly sessions for around 6 months. A large-scale, multi-centre maintenance CST trial is required to clarify potential longer-term benets of maintenance CST for dementia. Copyright # 2005 John Wiley & Sons, Ltd. key words dementia; cognition; maintenance; cognitive stimulation; quality of life

INTRODUCTION The ndings of a Cochrane review and a systematic evaluation of the research on Reality Orientation therapy for dementia were used to develop and evaluate an evidence-based Cognitive Stimulation Therapy (CST) programme for people with dementia (Spector et al., 1998; 2001, 2003). The CST groups were based on the concepts of Reality Orientation (RO; Folsom,
*Correspondence to: M. Orrell, Department of Mental Health Sciences, University College London, Wolfson Building, 48 Riding House Street, London, W1W 7EY, UK. Tel: 020 7679 9452. Fax: 020 7679 9426. E-mail: m.orrell@ucl.ac.uk Contract/grant sponsors: Responsive Funding Programme of the Department of Health; Barking, Havering and Brentwood Community NHS Trust.

1966) with a particular emphasis on more recent work described as Cognitive Stimulation (Breuil et al., 1994) which was identied through the systematic reviews as having the best results. The programme consisted of 14, 45-minute sessions which ran twice weekly for groups of approximately ve people. Topics included; using money, word games, the present day and famous faces, and multisensory stimulation was used when possible. The programme included an RO board, displaying both personal and orientation information, including the group name (chosen by participants). Two hundred and one people were recruited for this single-blind, multi-centre RCT from 23 residential homes and day centres in greater London. The treatment group improved signicantly relative to no-treatment controls on the main outcome measures (cognition and
Received 13 October 2004 Accepted 14 December 2004

Copyright # 2005 John Wiley & Sons, Ltd.

mcst for people with dementia quality of life). In terms of Numbers Needed to Treat the results compared favourably with trials of cholinesterase inhibitors for dementia (Spector et al., 2003). However, the long-term effects of CST were not evaluated by Spector et al. (2003) and the Cochrane review found no clear evidence of any long-term effects of RO. Gerber et al. (1991) found that benets for cognition and behaviour gained from RO were lost ten weeks after stopping the programme. Wallis et al. (1983) found that one month after RO terminated, benets in cognition were lost yet behavioural functioning continued to improve. It therefore appears uncertain as to how long any benets of RO, or similar interventions, may remain after the programme nishes. Further, it is uncertain how far maintenance programmes of RO might continue to benet the participants. The studies included in the RO Cochrane review ranged from using programmes of four to 21 weeks. However, there did not appear to be a relationship between the duration of the intervention and the outcome, and the trial with the best results (Breuil et al., 1994) only had a ve-week intervention. Zanetti et al. (1995) cited an expected yearly decrease in MiniMental-State-Examination score (MMSE, Folstein et al., 1975) of 1.84.2 points in people with dementia. Therefore, it might be that pre-post comparisons in the studies which had used longer interventions (20 and 21 weeks) would have shown weaker results (Ferrario et al., 1991; Woods, 1979). A number of studies have looked at the effects of an extended RO programme. Zanetti et al. (1995), in a controlled trial, evaluated the effects of more longterm RO. Their intervention ran in four cycles of 20 session blocks with rest periods in between, lasting 8.2 months in total. They found that the effect of RO on cognitive performance (a small increase in MMSE score of 0.68 points) appeared to counteract the decline, observed in the control group, of 2.58 points. Metitieri et al. (2001) studied the more longterm effects of RO by assessing 74 people with dementia over a 30-month period, who completed at least one cycle of RO groups (20 sessions). They compared 46 people who completed from 210 cycles (840 weeks of training) with the 28 who only completed one cycle. They found that people receiving long-term treatment declined in cognitive function signicantly later, and remained at home longer than those receiving only one cycle of RO. Both studies concluded that more long-term RO was effective in slowing, at least temporarily, the dementia process. This exploratory pilot study investigates the effectiveCopyright # 2005 John Wiley & Sons, Ltd.

447

ness of a weekly maintenance programme following participation in the CST trial (Spector et al., 2003). The aim was to investigate whether benets in cognition and quality of life could be maintained through participation in 16 further weekly sessions, in comparison with a group of people who participated in the initial CST programme but did not receive maintenance CST. METHOD When the maintenance trial began, groups for the CST study were running in six residential homes, all part of the Quantum Care group. Managers were approached and invited to participate in the maintenance trial through either: (i) the group participating in the 16-week maintenance programme and followup assessments or (ii) the group not receiving maintenance sessions but participating in follow-up assessments after 16 weeks (i.e. control homes). Two homes agreed to participate in the maintenance trial and two to be control homes. Thirty-ve people with dementia according to DSM-IV criteria (APA, 1994) were recruited for this study [see Spector et al. (2003) for full details of inclusion/exclusion criteria]. They had previously been randomly allocated into treatment or control groups for the initial CST programme. Procedure Blind assessments were conducted at baseline and in the week following the main CST trial (see Spector et al., 2003). The 16, weekly maintenance sessions followed on immediately from the original CST programme. Participants were re-assessed in the week following the maintenance programme. There were three groups. The maintenance CST group (MCST) consisted of eight people participating in both the initial and maintenance CST groups. These eight comprised four from each of the two maintenance CST homes (one person had dropped out from both original CST groups). The CST only group consisted of twelve people only participating in the initial CST groups (two groups of ve from the control homes and one person in each home where maintenance groups ran, but who could not participate in maintenance CST for health reasons). The baseline control group consisted of 15 people who received no CST. Programme Like the main CST programme, maintenance sessions focused on themes, with a primary emphasis on
Int J Geriatr Psychiatry 2005; 20: 446451.

448
Table 1. Demographics by group at baseline Group MCST CST No CST No of participants 8 12 15 Gender (female: male) 7:1 12:0 15:0 Mean age 84.3 82.8 85.2

m. orrell

ET AL.

Cognitive function (MMSE) 13.6 (3.3) 12.6 (4.3) 13.5 (3.1)

Quality of Communication Life (QoL-AD) (Holden) 34.1 (5.1) 34.5 (4.3) 34.2 (5.0) 7.5 (5.9) 12.8 (4.7) 8.5 (5.3)

Behaviour (CAPE-BRS) 10.0 (3.1) 12.7 (4.7) 11.9 (4.8)

cognitive stimulation, yet incorporating the process of reminiscence therapy and multi-sensory stimulation. Group names and songs, an RO board and introductory exercises provided continuity between sessions. The content of sessions can be found in Appendix 1. Measures Cognition. Mini-Mental State Examination (MMSE), (Folstein et al., 1975). This is a brief, widely used test of cognitive function, with good reliability and validity. Quality of life. Quality of LifeAlzheimers Disease Scale (Qol-AD; Logsdon et al., 1999). This brief, self-report questionnaire has 13 items covering the domains of physical health, energy, mood, living situation, memory, family, marriage, friends, chores, fun, money, self and life as a whole. It has good internal consistency, validity and reliability (Logsdon et al., 1999; Thorgrimsen et al., 2003). Communication. Holden Communication Scale (Holden and Woods, 1982). This scale, completed by staff, covers a range of social behaviour and communication variables including conversation, attempts

at communication, awareness, pleasure, humour and responsiveness. Higher scores indicate more difculty in communication. Behaviour. The Clifton Assessment Procedures for the ElderlyBehaviour Rating Scale (CAPE-BRS; Pattie and Gilleard, 1979). This covers general behaviour, personal care and behaviour towards others, with higher scores indicating greater dependency. It has good reliability and validity, and was included to assess the overall level of functional impairment and dependency. RESULTS The clinical and demographic proles of participants at baseline in each group are presented in Table 1. The CST only group had more communication difculties on average than the other two groups. The groups were relatively similar at baseline. A breakdown by group of mean scores on each measure at the three assessment stages is given in Table 2. Results were analysed using repeated measures ANOVAs, with time as the within subjects factor and group as the between subjects factor. Table 2 shows the results of these analyses. Changes in cognitive

Table 2. Means (and standard deviations) for the three groups on each measure, at the three assessment phases including repeated measures ANOVAs Group MCST CST No CST MCST CST No CST MCST CST No CST MCST CST No CST MMSE1 13.63 (3.29) 12.58 (4.34) 13.53 (3.11) QOL1 34.13 (5.14) 34.50 (4.30) 34.20 (4.97) BRS1 10.00 (3.07) 12.67 (4.66) 11.87 (4.76) Holden1 7.50 (5.88) 12.83 (4.71) 8.47 (5.25) MMSE2 14.88 (4.12) 13.67 (3.94) 12.27 (4.08) QOL2 35.75 (4.83) 35.67 (4.66) 34.73 (6.77) BRS2 14.57 (4.89) 11.70 (4.06) 12.43 (5.14) Holden2 10.29 (6.13) 13.60 (3.24) 14.23 (6.57) MMSE3 15.50 (5.17) 10.63 (2.92) 12.80 (4.42) QOL3 35.67 (3.83) 29.25 (5.12) 34.33 (7.97) BRS3 11.71 (6.07) 15.75 (3.20) 14.13 (5.49) Holden3 9.29 (6.63) 20.75 (3.86) 15.20 (6.55) MMSE Wilks Lambda 0.53, F(4, 40) 3.72, p 0.012* QOL-AD Wilks Lambda 0.69, F(4, 38) 1.92, p 0.13 BRS Wilks Lambda 0.78, F(4, 32) 1.06, p 0.39 Holden Wilks Lambda 0.67, F(4, 30) 1.68, p 0.18

Copyright # 2005 John Wiley & Sons, Ltd.

Int J Geriatr Psychiatry 2005; 20: 446451.

mcst for people with dementia

449

Figure 1. Change in cognitive function (MMSE score) over time

impact of cognitive stimulation programmes and Spector et al. (2003) concluded that small changes in cognition were unlikely to have any impact on areas of functional dependence described in the CAPE-BRS, such as feeding and dressing. For those in the CST only group, at the nal assessment, the mean MMSE and QoL-AD scores dropped to levels lower than at baseline, suggesting that the benets gained in cognitive function and quality of life were lost 16 weeks after CST ended. This cognitive deterioration mirrors that found four and ten weeks post intervention in other studies (Wallis et al., 1983; Gerber et al., 1991), though neither measured quality of life. Limitations

function over time, as measured by the MMSE, were signicantly different between the groups ( p 0.012). The programme ran over a period of almost six months, with a mean improvement of 1.9 points on the MMSE for those receiving continuous treatment, compared to a mean decline of 0.7 points for those receiving no treatment. Figure 1 shows how people in the MCST group improved following CST and again following the maintenance programme. In contrast, those in the CST only group improved following CST yet deteriorated over the follow-up period. Those in the control group (no CST) performed worse at both follow-up assessments than at baseline. There were no differences between groups in quality of life, communication or behavioural function over time. DISCUSSION This study indicates that 16 weekly sessions, following the same principles of CST, can be successful in maintaining and even accentuating the cognitive benets of a more intensive seven-week CST programme. These ndings, like those of Zanetti et al. (1995) suggest that such interventions over a longer period can be highly benecial. Benets in quality of life, as found following CST, were not maintained, suggesting that weekly sessions might not be sufcient to impact on how QoL is experienced by people with dementia in residential care. The results of the initial CST trial showed no signicant changes in behaviour or communication (although there were positive trends in the latter) and so changes were not predicted following the maintenance programme. Zanetti et al. (1995) suggested that behavioural outcome measures are often not sensitive enough to detect the functional
Copyright # 2005 John Wiley & Sons, Ltd.

The sample was small and hence may not have been representative of the total CST sample, for example only 3% of the maintenance sample was male, compared to 21% of the total CST sample. The Holden and CAPE-BRS were completed by staff who were not blind to group allocation and it was often not possible for the same staff member to complete the assessments each time. However, due to the exploratory pilot nature of this study, such limitations would be expected. Homes were recruited through voluntary participation rather than randomisation. This may have resulted in bias, for example homes where groups were going well might have been more likely to volunteer for the maintenance programme than the two homes that did not agree to take part. Additionally, two people were included as part of the CST only group, as they were too unwell to participate in the maintenance CST sessions. This might have resulted in bias if their physical deterioration been associated with cognitive decline. Lastly, the small sample size lacks power to detect potential differences between the groups so there is the possibility of a type 2 statistical error, particularly for outcomes such as quality of life which had been found to signicantly improve with CST (Spector et al., 2003). Implications for research A large, multi-centre maintenance CST trial, following a similar design to the CST study, would provide a broader evaluation of the maintenance programme, the longer-term effects of CST and how outcomes change when maintenance CST is discontinued. It might be helpful to examine the effectiveness of an extended programme to match the duration of drug trials which may last as long as 52 weeks, adding a
Int J Geriatr Psychiatry 2005; 20: 446451.

450 KEY POINTS

*

m. orrell

ET AL.

Little is known about the long term effects of CST on cognition in dementia. This study compares maintenance CST, with short-term CST and a treatment as usual control group. Over 23 weeks people receiving maintenance CST continued to show improvements in cognition whereas cognition declined in both the CST only and the control groups.

more detailed measure of cognition such as the ADAS-Cog (Rosen et al., 1984), and to evaluate maintenance CST in combination with cholinesterase inhibitors. Further, the question of whether QoL can be maintained would be better assessed with a larger sample. It has been demonstrated that in the short-term at least CST results in signicant improvements in cognition comparable to those found following treatment with acetylcholinesterase inhibitors (Spector et al., 2003). We have run a number of one-day training courses across the UK and other centres have started to implement CST locally. Training of staff in residential and day facilities could lead to appreciable benets to people with dementia. This study suggests that benets in cognition can be maintained by weekly sessions of CST following a more intensive programme. Such ndings offer hope of potentially delaying cognitive deterioration in people with dementia. ACKNOWLEDGEMENTS This study was funded by grants from the Responsive Funding Programme of the Department of Health and Barking, Havering and Brentwood Community NHS Trust. We are extremely grateful for the input of residents and staff at the four residential homes, all part of the Quantum Care group. We also thank Dr Pasco Fearon for his statistical advice. REFERENCES
American Psychiatric Association. 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. APA: Washington, DC. Breuil V, De Rotrou J, Forette F, et al. 1994. Cognitive stimulation of patients with dementia: preliminary results. Int J Geriatr Psychiatry 9: 211217. Ferrario E, Cappa G, Molaschi M, Rocco M, Fabris F. 1991. Reality orientation therapy in institutionalized elderly patients: preliminary reports. Archives of Gerontology and Geriatrics 2: 139142. Copyright # 2005 John Wiley & Sons, Ltd.

Folsom JC. 1966. Reality Orientation for the Elderly Mental Patient. Read at 122nd Annual Meeting of American Psychiatric Association, May 1966. Folstein MF, Folstein SE, McHugh PR. 1975. Mini-mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12: 189198. Gerber GJ, Prince PN, Snider HG, Atchinson K, Dubois L, Kilgour JA. 1991. Group activity and cognitive improvement among patients with Alzheimers disease. Hosp Commun Psychiatry 42(8): 843846. Holden UP, Woods RT. 1982. Holden Communication Scale. In Reality Orientation: Psychological Approaches to the Confused Elderly. Churchill Livingstone: Edinburgh, UK. Logsdon R, Gibbons LE, McCurry SM, Teri L. 1999. Quality of life in Alzheimers disease: patient and caregiver reports. J Ment Health Aging 5: 2132. Metitieri T, Zanetti O, Geroldi C, Frisoni GB, De Leo D, Dello Buono M. 2001. Reality Orientation Therapy to delay outcomes of progression in patients with dementia: a retrospective study. Clin Rehab 15: 471478. Pattie AH, Gilleard CJ. 1979. Clifton Assessment Procedures for the Elderly (CAPE). Hodder and Stoughton: Sevenoaks. Rosen WG, Mohs RC, Davis KL. 1984. A new rating scale for Alzheimers disease. Am J Psychiatry 141: 13561364. Spector A, Orrell M, Davies S, Woods B. 1998. Reality Orientation for dementia: a review of the evidence for its effectiveness. In The Cochrane Library, Issue 4, 1998. Update Software: Oxford. Spector A, Orrell M, Davies S, Woods B. 2001. Can reality orientation be rehabilitated? Development and piloting of an evidencebased programme of cognition-based therapies for people with dementia. Neuropsychologic Rehab 11(3/4): 377397. Spector A, Thorgrimsen L, Woods B, et al. 2003. Efcacy of an evidence-based cognitive stimulation therapy programme for people with dementia: Randomised Controlled Trial. Br J Psychiatry 183: 248254. Thorgrimsen L, Selwood A, Spector A, et al. 2003. Whose quality of life is it anyway? The validity and reliability of the Quality of Life-Alzheimers Disease (QoL-AD) Scale. Alzheimers dis Assoc Disord 17(4): 201208. Wallis GG, Baldwin M, Higgenbotham P. 1983. Reality Orientation Therapy: a controlled trial. Br J Medic Psychol 56: 271277. Zanetti O, Frisoni GB, De Leo D, Buono MD, Bianchetti A, Trabucci M. 1995. Reality Orientation Therapy in Alzheimers disease: useful or not? A controlled study. Alzheimers Dis Assoc Disord 9: 132138.

APPENDIX 1MAINTENANCE CST SESSIONS (1) Childhood. Questions from the memory diaries were used as prompts for discussion. (e.g. describe your childhood bedroom). Use of childhood toys and games. (2) Current affairs (A). Duplicate copies of discussion-provoking articles from newspapers were used to generate opinion and debate. (3) Current affairs (B). As above. (4) Using objects (A). This involved making a chocolate cake. (5) Number game (bingo). (6) Quiz, involving two teams.
Int J Geriatr Psychiatry 2005; 20: 446451.

mcst for people with dementia (7) Music session. This involved the playing of musical instruments, singing along to old songs and a song completion game, where people are given the rst few words of a song and are asked to sing the remainder. (8) Physical games, such as hoopla, skittles, boules and football. The group was encouraged to calculate the scores. (9) Categorising objects. New odd one out sheets were used, in which four words were presented on a sheet and the group required to guess the odd one out. The topix game was used again (naming objects beginning with a particular letter in a certain category). (10) Using objects (B). The reminiscence kit and modern objects (such as a mobile phone) were presented and discussed.

451

(11) Useful tips. A book called what our grandmothers knew was used to generate a discussion of useful tips, e.g. soothing burns, treating milk. (12) Golden expressions cards (A). Cards asking discussion-provoking questions were passed around the group, e.g. what is your favourite charity? How are elderly people treated by society?. (13) Golden expressions cards (B). As above. (14) Opinions on different types of art, generated through the presentation of art ranging from classic to modern. (15) Famous faces (B). Pictures of people from the past were used to make comparisons and to generate discussion. (16) Word completion (B), for example completion of proverbs and famous couples.

Copyright # 2005 John Wiley & Sons, Ltd.

Int J Geriatr Psychiatry 2005; 20: 446451.