- Email: [email protected]
MAXCOG—Maximizing Cognition: A Randomized Controlled Trial of the Efficacy of Goal-Oriented Cognitive Rehabilitation for People with Mild Cognitive Impairment and Early Alzheimer Disease
Accepted Manuscript Title: MAXCOG – Maximising Cognition: a Randomised Controlled Trial of the Efficacy of Goal Oriented Cognitive Rehabilitation for People with Mild Cognitive Impairment and Early Alzheimer's Disease Author: B. Regan, Y. Wells, M. Farrow, P. O'Halloran, B. Workman PII: DOI: Reference:
S1064-7481(16)30302-5 http://dx.doi.org/doi: 10.1016/j.jagp.2016.11.008 AMGP 721
To appear in:
The American Journal of Geriatric Psychiatry
Received date: Revised date: Accepted date:
16-3-2016 7-11-2016 9-11-2016
Please cite this article as: B. Regan, Y. Wells, M. Farrow, P. O'Halloran, B. Workman, MAXCOG – Maximising Cognition: a Randomised Controlled Trial of the Efficacy of Goal Oriented Cognitive Rehabilitation for People with Mild Cognitive Impairment and Early Alzheimer's Disease, The American Journal of Geriatric Psychiatry (2016), http://dx.doi.org/doi: 10.1016/j.jagp.2016.11.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Abstract: 187 Text: 3499 Tables: 4 Figure: 4 References: 34 MAXCOG – Maximising Cognition: A randomised controlled trial of the efficacy of Goal Oriented Cognitive Rehabilitation for people with Mild Cognitive Impairment and Early Alzheimer’s Disease
Regan, B., D.Psych1 Wells, Y., Ph.D.1 Farrow, M., Ph.D.2 O’Halloran, P., Ph.D.3 Workman, B. M.D.4
1
La Trobe University, Australian Institute for Primary Care & Ageing, Melbourne, Australia
2
University of Tasmania, Wicking Dementia Research & Education Centre, Hobart, Australia
3
La Trobe University, School of Psychology and Public Health, Melbourne, Australia
4
Monash University and Monash Health, Monash Ageing Research Centre, Melbourne Australia
Corresponding author Bridget Regan, D.Psych Australian Institute for Primary Care & Ageing La Trobe University Level 5, Health Sciences 2 Bundoora 3083 Victoria, Australia Email: [email protected] Telephone: +61 3 94795809 Fax: +61 3 94795977 Key words: Mild Cognitive Impairment, Dementia, Cognitive Rehabilitation, Clinometric, Translational Supported by a grant from Alzheimer’s Australia Dementia Research Foundation (Vic) and in kind support from Alzheimer’s Australia (Vic) and the Monash Ageing Research Centre. The trial was registered on the Australian New Zealand Clinical Trials Registry (ANZCTR: reg. no. ACTRN12612001149853). Acknowledgments: We thank Vanessa Smithies, Cameron Redpath and Kerith Sharkey for assistance in the study design and conduct. We also thank the Early Intervention Team including Jenny Philip, Christina Bell and Jessica Byers from Alzheimer’s Australia (Vic). No disclosures to report.
Page 1 of 24
ABSTRACT Objectives: To review the efficacy of a home-based four-session individualised face to face cognitive rehabilitation (MAXCOG) intervention for clients with mild cognitive impairment (MCI) or early dementia and their close supporters. Design: Randomised controlled trial comparing the intervention group (MAXCOG) with treatment as usual (control). Participants: Fifty-five client supporter dyads were enrolled in the study and 40 completed; Twenty-five client-supporter dyads completed MAXCOG and 15 completed treatment as usual. Both MAXCOG and control groups included more MCI cases than dementia (22/3 and 12/3 respectively). Intervention: Four weekly individual sessions of MAXCOG consisting of personalised interventions to address individually relevant goals, supported by the provision of the MAXCOG information resource. Measures: The primary outcomes were goal performance and satisfaction, assessed using the Canadian Occupational Performance Measure (COPM). Questionnaires assessing mood, illness adjustment, quality of life, and carer burden were also administered. Results: The intervention group displayed significantly higher performance and satisfaction with primary goals on the COPM post-intervention than the control group, using a per protocol analysis. Conclusion: The MAXCOG intervention is effective in improving goal performance and satisfaction in clients with MCI and early dementia.
Page 2 of 24
INTRODUCTION Psychosocial interventions that address cognitive difficulties and their impact on day-to-day life have been developed for a range of conditions (e.g., learning disabilities, stroke, and traumatic brain injury) [1, 2]. More recently they have been implemented with older adults at a preclinical or early stage of dementia [3, 4]. Emerging evidence suggests that such interventions have the potential to increase the duration of independence, improve quality of life, and save healthcare costs [5]. A range of approaches have been trialled. Remediation relies on repeated, targeted practice of cognitive tasks that are compromised in early dementia. In contrast, rehabilitation (or compensation) approaches focus on maximising physical, psychological, and social wellbeing through external aids (e.g., calendar or smart phone) or adapted strategies (e.g., mnemonic techniques) [5]. To date, only a limited number of studies of cognitive rehabilitation for clients with MCI and early dementia have been published. The bulk of these have been group-based and teach a specific predetermined range of cognitive management strategies (e.g., memory support system). Few are randomised controlled trials (RCTs) [3, 4, 6]. Two recent reviews [4, 6] identified six RCTs on cognitive rehabilitation for people with MCI and early dementia [7-12]. All found significant improvements in use of compensatory memory strategies or memory self-efficacy in the intervention groups, but only two reported improvements in objective cognitive performance [7, 10]. Clare et al. [13] in the UK published the first evaluation of an individualised cognitive rehabilitation intervention for clients with early dementia (N = 69). Their ‘individualised’ approach differed from much of the previous literature in that, rather than teaching a predetermined range of cognitive strategies, each client developed their own personally relevant goals as the focus of their intervention [5]. This single-site, single-blind randomised controlled trial compared an eight-session
Page 3 of 24
cognitive rehabilitation intervention with relaxation therapy, involving equivalent therapist time and attention, and a treatment-as-usual group. An innovative feature of this study was a ‘clinometric’ outcome measure, which used targeted goals specific to each individual to provide a standardised group-level comparison. The measure used was the Canadian Occupational Performance Measure [COPM: 14], reported to be reliable, valid, and responsive as well as more sensitive to clinically significant individual change than more standardised tests or questionnaires [15, 16]. The study showed that the intervention produced significant improvement in self-ratings of goal performance and satisfaction, whereas scores for the other two groups did not change. The current study took the approach of Clare at al. [13] and applied it within a real-world clinical setting. This study has a more translational focus than its predecessor, in that staff involved were not specifically employed by the study but co-opted from an existing early intervention team from the local Alzheimer’s Association. A manual was created and training conducted to encourage consistency between counsellors in implementing the approach. A set of information handouts provided easy access to a range of ideas and strategies that could be utilised as part of a face-to-face intervention. Clare et al.’s [13] individualised approach was replicated, including use of the COPM, but the intervention was shorter in duration, and clients with MCI were included as well as those with early dementia. MCI clients were included because people in this ‘prodromal’ group were thought to be more likely to be capable of taking up cognitive rehabilitation strategies than those with a diagnosis of dementia [8]. Comprising four sessions, MAXCOG is shorter than other interventions reported in the literature; Clare et al. [13] provided a total of eight sessions. With this relatively brief intervention, only one (or sometimes, two) personally relevant goals may be addressed. However, a shorter intervention is easier to implement in a real-world setting than a more extended one. Our aim was to assess the efficacy of this brief cognitive rehabilitation intervention in day-to-day life.
Page 4 of 24
METHODS Study Design. The study was a two-year multi-centre RCT. A 3:1 randomization intervention-tocontrol ratio was adopted (after a six-month period at 1:1) due to concerns regarding initially high attrition levels in the intervention group. A central allocation process was used to randomize participants, using a computer programme and a random block size algorithm to prevent imbalance between the allocation groups, stratified by diagnosis (MCI; early dementia). All sessions were conducted in participants’ homes. The study received ethics approval from Monash University (CF13/2697-2013001446) and each health care site involved in recruitment. Participants. Participants were recruited from six memory clinics across metropolitan Melbourne, Australia. Eligibility criteria were that participants were community-dwelling with a recent (i.e., ≤ 6 months) clinical diagnosis of MCI or early dementia (any type); were fluent in English; were aged ≥ 50 years; had a Mini–Mental State Examination [17] score >20; would receive “1” or less on the Clinical Dementia Rating Scale [18]; and had a primary supporter in close contact with the client and willing to participate in the study. Participants were excluded if they had severe somatic or psychiatric comorbidity, had hearing or vision loss (sufficient to impair their ability to complete the intervention), or were involved in other counselling or cognitive rehabilitation programs. Measures. Pre assessments were administered by a trained research assistant blinded to group allocation until the allocation envelope was opened at the end of the pre assessment. Post assessments were administered by a non-masked assessor. The Canadian Occupational Performance Measure [14] was used to assist clients to identify up to five personally relevant goals in areas relating to self-care, leisure, and productivity. Levels of performance and satisfaction with respect to the goals were elicited on a 10-point scale (1 = unable to perform/ not satisfied and 10 = fully able to perform/ extremely satisfied). Participants in the MAXCOG group selected one or two goals to be the focus of their intervention. Post intervention, all
Page 5 of 24
participants re-rated their performance on and satisfaction with the originally identified goals. The COPM has good test-retest reliability over a one–week period, exceeding 0.80 in studies with different clinical groups [16]. Secondary outcome measures completed at initial interview and post intervention included the Hospital Anxiety and Depression Scale [HADS: 19]; Illness Cognition Questionnaire [ICQ: 20]; Multifactorial Meta Cognition Questionnaire [MMQ: 21]; and Quality of Life – AD [QOL-AD: 22]. Outcome measures completed by a close supporter included Bayar Activities of Daily Living Scale [BADL: 23]; and Everyday Cognition Scale [ECOG: 24]. The Revised Memory and Behaviour Problem Checklist [RMBPC: 25], which has good test-retest reliability (0.89 for frequency and 0.74 for reaction – [see 26]), was also used. Supporters also rated their own mental status using the Zarit Carer Burden [27] and the Hospital Anxiety and Depression Scale for supporters [HADS-S: 19]. Intervention. MAXCOG involved four weekly one-hour sessions delivered to client-supporter dyads in the client’s home with a focus on an individualised intervention addressing personally meaningful goals (e.g., to learn the names of people at church/ bowls club; to be less dependent on the spouse for reminders about appointments; to take medication reliably; to develop a better system for bookwork; and to remember more about grandchildren’s activities). The MAXCOG intervention was delivered by experienced counsellors (including staff from the Alzheimer’s Australia Victoria Early Intervention Team and a neuropsychologist). Strategies could be selected from those outlined in information handouts recently developed by our project team at La Trobe University and the Kingston Centre [28]. While the basic structure of sessions was prescribed in a manual, their content could be adapted flexibly to meet specific client goals. [insert figure 1 about here] Clients and supporters were encouraged to help brainstorm and select the most appropriate strategies. The focus of strategies was on positive resources, intact functions, retained skills, and
Page 6 of 24
activities clients could still take part in (see [28] for examples). Clients were encouraged to practice techniques with assistance from their supporter in between sessions. Clients enrolled in the control condition had no contact with the research team between their initial and post intervention assessments. Procedure. Potential participants were identified by memory clinic clinicians as clients of the local Alzheimer’s Association. Participants who met study criteria and gave informed consent received an initial assessment. If the participant was randomised into the treatment group, a referral was made to a counsellor to conduct the MAXCOG intervention. The counsellor was provided with a preliminary intervention plan compiled by the research team outlining potential areas of focus (based on COPM results) and ideas for intervention strategies (see figure 1). Once the intervention was completed, or four weeks had passed in the case of the control group, the research assistant collected follow up data as soon as feasible (up to four weeks post-intervention) either via telephone (with mailed questionnaires) or in person. Data Analysis. Behavioural data were examined only after all participants had completed the postintervention assessment. To compare baseline performances independent samples t-tests were used. For each outcome measure, mixed-model repeated measures ANOVAs were conducted to determine whether groups differed in change from pre- to post-intervention. Sharpio-Wilk, Fmax and Levene’s test statistics were used to test assumptions of normality and homogeneity of variance. Missing data within scales were replaced by the mean value for that item. Effect sizes were calculated using Cohen’s D for t-tests and partial eta-squared for non-independent repeated measures analyses [29, 30].
RESULTS
Page 7 of 24
Participants. Clients were 40 individuals (22 women, 18 men; mean age 77.2 years, standard deviation [SD] 6.5, range = 60-93). Supporters were 40 individuals (13 men, 27 women): mean age was 66.8 (N = 36, SD 15.0, range = 30-87). Participant characteristics are summarized in Table 1. There were no statistically significant differences between the two groups at baseline on most measures, including function and cognitive status, suggesting that the randomization process was largely successful at creating equivalent groups. Nevertheless, relatively large effect sizes on two measures at baseline—the RMBPC and the E-COG proxy—suggested a trend for supporters in the MAXCOG group to report more behavioural and cognitive difficulties than those in the control group. [Insert table 1 about here]
[Insert table 2 about here] The CONSORT chart is provided in Fig. 2. Of the 55 randomised participants, 37 were allocated to the MAXCOG intervention and 18 to the control group, using a 3:1 ratio. Post-intervention data were collected for 25 MAXCOG and 15 control group participants. Four protocol deviations were identified, in which the clients were found to be too severely affected by dementia to participate in the intervention or complete questionnaires (rated > 1 on the Clinical Dementia Rating Scale). None of the other clients who withdrew were available for post-intervention assessment, and therefore the analysis is a per protocol analysis. Mean duration between the collection of the pre- and postintervention data was slightly longer for the intervention group (115 versus 95 days); however, this difference was not statistically significant F (1,35) = 1.82, p = > 0.05. [insert figure 2 about here] Primary Outcomes. The primary outcome measures were goal performance and satisfaction scores on the COPM. At initial assessment, all 40 participants included in the per protocol analysis
Page 8 of 24
identified at least one goal for improvement. Between them, these clients identified 143 goals. Rates of goal identification were similar for the two groups (MAXCOG mean 3.7, SD = 1.2, range 1-5; control mean 3.3, SD = 1.2, range 1-5). Common areas of focus included remembering names and keeping track of items around the home, as well as increased socialisation, improvements in using technology, and managing finances and bookwork. Mean COPM performance and satisfaction scores are summarised in figures 3 and 4. [insert figure 3 about here] [insert figure 4 about here] For the first performance goal, a significant interaction effect was obtained between group and time F(1,37) = 4.77, p = 0.035, partial η2 = .11, with an increase in scores for the MAXCOG group but not the control group. For the second performance goal, a significant main effect of time was found F(1,33) = 5.43, p=.026, partial η2=.14, showing that for both groups the mean score on the performance COPM measure was higher post intervention. On the satisfaction measure for the first goal, both a significant main effect for time (F(1, 36) =9.81, p = 0.003, partial η2=.21) and a significant interaction effect (F(1,36)=7.77, p=0.008, partial η2=.18) were observed; inspection of mean scores indicated that satisfaction increased dramatically for the MAXCOG group but remained stable for the control group. Results of analysis of satisfaction for the second goal was similar to those for the first goal with a significant interaction effect (F(1,32)=7.18, p = .012, partial η2= .18); inspection of the means again revealed an increase in satisfaction in the MAXCOG group and stability or a slight decrease in the control group. In summary, on the first goal, both performance and satisfaction increased significantly for the MAXCOG group versus the control group, whereas on the second goal where sample sizes were smaller a significant difference between groups was noted only for satisfaction. [insert table 3 about here]
Page 9 of 24
Secondary Outcomes. Several significant findings were noted on the secondary outcome measures. A significant main effect of time (F (1, 38) = 7.05, p= .012, η2= .156) indicated that both client groups rated their own memory abilities on the multifactorial metacognition questionnaire as decreasing between pre and post measures (see Table 4). Similarly, supporters from both groups rated clients as significantly more impaired on their overall ECOG score on post measures (F(1,35) = 11.81, p = .002, η2= .252; see Table 4). [insert table 4 about here] On the RMBPC, supporters from the MAXCOG group indicated a significantly greater number of problem behaviours both pre- and post-assessment than the control group (F (1,37) = 5.82, p = 0.027, η2= .125; see Table 4). Results on supporters’ ratings of how much the problem bothered the client were interesting. There was both a main effect of time, with both groups of supporters rating behaviours as less severe post-intervention (F(1,35) = 57.97, p =.00, η2= .624), and an interaction effect with MAXCOG group scores declining more than control group scores (F(1,35) = 5.11, p = 0.03, η2= .127). There was a trend for scores on the helplessness scale (Illness Cognition Questionnaire) to increase over time for the control group but to decrease slightly for the MAXCOG group (F(1, 37) = 4.03, p = .052, η2= .098; see Table 4). No significant differences between pre and post intervention were noted on any other measures, including measures of quality of life and mood.
DISCUSSION This RCT investigated the efficacy of the MAXCOG cognitive rehabilitation intervention versus a treatment-as-usual control group. The intervention utilised individualised intervention plans to address personalised goals in an everyday setting, and progress was assessed using a clinometric
Page 10 of 24
measure. Overall our findings on the impacts of the MAXCOG intervention were encouraging. Our main finding was that clients in the MAXCOG group reported higher levels of goal performance and satisfaction after receiving the intervention than those in the control group. This was the case for the first goal and partially the case for the second goal (where satisfaction but not performance increased for the MAXCOG group). These improvements had moderate effect sizes with partial etasquared values of between .11 and .21 [29]. Findings suggest that the MAXCOG intervention was effective at assisting clients to reach at least one goal in comparison to the control group. This finding is consistent with the findings of the Clare et al. [13] study, revealing the COPM to be a sensitive measure to clinical change associated with goal formation and achievement. The COPM is the most proximal measure of the central focus of the intervention, and is not surprising that more distal measures, such as overall quality of life and depression, which are unlikely to change in response to practical improvement in one goal area, remained stable between the two assessment time points. However, other measures suggested additional positive impacts for the MAXCOG intervention. At baseline and post-intervention, MAXCOG group supporters reported a higher overall number of behavioural difficulties than the control group. However, the ratings of the severity of these behaviours changed between pre and post measures; the MAXCOG group improved significantly more than the control group. This suggests that, although challenging behaviours were still present post intervention, the MAXCOG intervention may have been helpful in reducing their perceived severity (other intervention studies using the RMBPC have reported similar findings - see [31]). Another explanation for this finding could be ‘regression to the mean’. Both groups rated their own memory abilities worse post-intervention than at baseline, and their supporters rated both groups’ overall day-to-day functional abilities as more impaired. This could reflect the fact that all participants were declining. Alternatively, it may be due to decreased confidence or increased awareness brought about by participating in the study or because clients
Page 11 of 24
felt more comfortable with the study procedures post intervention than at baseline and so were more honest about their difficulties. Limitations. Several methodological issues need to be considered when interpreting the results of the study, including the fact that the study utilised a translational design with a ‘real world’ clinical team. One factor that may have enhanced our significant main finding is the relatively large number of withdrawals post allocation (15 in total; 27%). Although many studies with older adults report attrition for a range of reasons [32], this study sustained a much higher proportion of withdrawals than that reported in the Clare et al. [8] study (5.8%). In the MAXCOG group, although six withdrawals were largely unavoidable (illness or dementia severity), a further six were because the client, supporter, and counsellor were unwilling to develop or follow through with an intervention plan (for the areas of difficulty earlier identified on the COPM) or because after the first session a client decided to withdraw. Therefore, only the more insightful and willing participants completed the study and this group may be more likely to be committed to the counselling process and to report positive outcomes from it. Conceivably, utilising a ‘real world’ clinical team with broad counselling backgrounds, rather than an occupational therapist specifically employed and trained to be part of the study (as in the Clare et al. study), may have led to differences in counsellors’ capacity or willingness to encourage clients to identify goals and continue to participate. Again, the omission of clients who could not be persuaded to participate may have enhanced the main findings. Another challenge with the use of a ‘real world’ clinical team was that the client-focussed nature of their usual work may have led to inconsistencies in their approach to the intervention and therefore potential threats to the internal validity of the study. Due to privacy concerns, it was not possible to conduct any direct monitoring of their ‘fidelity to treatment’ during sessions.
Page 12 of 24
Another limitation of the study design was the fact that no active control group was employed (e.g., a support group intervention to determine the specificity of the cognitive rehabilitation in addition to any non-specific effects of therapeutic intervention). The use of a blinded research assistant at follow up would have further enhanced the internal validity of the study. Finally, this study included both clients with MCI and early dementia. Clients’ diagnoses included a mixture of vascular, Alzheimer’s, and other pathologies. Different pathologies and different stages of illness may be associated with different cognitive strengths and weaknesses. Indeed a subgroup of approximately 20% of clients with MCI may improve or stabilise without developing dementia [33]. Therefore clients may differ in the extent to which they retain the capacity to benefit from cognitive rehabilitation approaches. Further research is required to determine which subgroups may respond best to such approaches. Implications Given the limitations in the study design outlined, the findings from this study should be considered preliminary. However, the findings do suggest that even relatively short individualised cognitive rehabilitation interventions may benefit clients by helping them to achieve their goals. In a ‘real world’ setting, the brevity of the intervention is critical, due to resourcing and cost considerations. Clinometric measures such as the COPM show promise as sensitive measures of clinically relevant outcomes. However, a remaining question is the extent to which participants’ responses to the COPM, particularly when rating performance and satisfaction post assessment, are influenced by social desirability and acquiescent response set, rather than an honest assessment of progress towards their goals. Again an active control group may help mitigate these concerns. Another strategy worth considering, particularly for groups whose insight and cognitive capacity to understand the rating scales are reduced [15], is the use of specific informant- or clinician-rated goal attainment scales.
Page 13 of 24
Another issue for future research is how to reduce attrition, particularly for the group that struggled to develop an intervention plan. One possibility may be to integrate motivational interviewing principles into the initial goal setting interview, to explore and address any ambivalence that the client may express [34]. Investigating which subgroups of clients with MCI or early dementia are likely to complete and benefit from cognitive rehabilitation interventions such as MAXCOG will also be important. If future studies do confirm the success of this type of intervention, it will add weight to the view that it is possible to improve specific goal-oriented outcomes for people with MCI and early dementia by addressing relevant personal, social, and environmental factors. In other words, it is not necessary to address impairment per se in order to improve engagement in activity and social participation and increase wellness.
Page 14 of 24
REFERENCES 1.
Cicerone KD, Langenbahn DM, Braden C, et al: Evidence based cognitive rehabilitation:
updated review of the literature from 2003 through 2008. Arch Phys Med Rehabil 2008; 92(4): 519530 2.
Bayley MT, Tate R, Douglas JM, et al: INCOG guidelines for cognitive rehabilitation following
traumatic brain injury: methods and overview. J Head Trauma Rehabil. 2014, 29(4):290-306 3.
Bahar-Fuchs A, Claire L, Woods B: Cognitive rehabilitation and cognitive training for early-
stage Alzheimer's disease and vascular dementia. Cochrane Database of Syst Rev 2013; 6: Art. No.: CD003260, DOI: 10.1002/14651858.CD003260.pub2 4
Rodakowski J, Saghafi E, Butters MA, et al: Non-pharmacological interventions for adults
with mild cognitive impairment and early stage dementia: An updated scoping review. Mol Aspects Med 2015; 43: 38-53 5
Clare L, Bayer A, Burns A, et al: Goal-oriented cognitive rehabilitation in early-stage
dementia: study protocol for a multi-centre single-blind randomised controlled trial (GREAT). Trials 2013; 14:152 6
Huckans M, Hutson L, Twamley E, et al: Efficacy of cognitive rehabilitation therapies for mild
cognitive impairment (MCI) in older adults: working toward a theoretical model and evidence-based interventions. Neuropsychol Rev 2013; 23:63-80 7
Greenaway M, Duncan N, Smith G: The memory support system for mild cognitive
impairment: radomised trial of a cognitive rehabilitation intervention. Int J Geriatr Psychiatry 2013; 28:402-409 8.
Kinsella GJ, Mullaly E, Rand E et al: Early intervention for mild cognitive impairment: a
randomised controlled trial. J Neurol Neurosurg Neuropsychiatry 2009; 80:730-736 9.
Rapp S, Brenes G,Marsh A: Memory enhancement training for older adults with mild
cognitive impairment: a preliminary study. Aging Ment Health 2002; 6(1):5-11
Page 15 of 24
10.
Jelcic N, Cagnin A, Meneghello F, et al: Effects of lexical–semantic treatment on memory in
early Alzheimer disease: an observer-blinded randomized controlled trial. Neurorehabil Neural Repair. 2012;26(8):949-56. 11
Kurz A, Thöne-Otto A, Cramer B, et al: CORDIAL: cognitive rehabilitation and cognitive-
behavioral treatment for early dementia in Alzheimer disease: a multicenter, randomized, controlled trial. Alzheimer Dis Assoc Disord 2012;26(3):246-53. 12
Troyer AK, Murphy KJ, Anderson ND, et al: Changing everyday memory behaviour in
amnestic mild cognitive impairment: a randomised controlled trial. Neuropsychol Rehabil 2008; 18:65-88 13
Clare L, Linden DE, Woods RT et al: Goal-oriented cognitive rehabilitation for people with
early-stage Alzheimer disease: a single-blind randomized controlled trial of clinical efficacy. Am J Geriatr Psychiatry 2010; 18:928-939 14
Law M, Baptise S, Carswell A et al: Canadian Occupational Performance Measure 4th edition,
Ottowa, CAOT publications ACE, 2005 15
Carswell A, McColl MA, Baptise S et al: The Canadian Occupational Performance Measure: a
research and clinical literature review. Can J Occ Ther 2004; 71: 210-222 16
Cup EHC, Scholte op Reimer, WJM et al: Reliability and validity of the canadian occupational
performance measure in stroke patients. Clin Rehabil 2003, 17(4): 402-9 17.
Folstein M, Folstein S,McHugh P: "Mini-mental State". A practical method of grading the
cognitive state of patients for the clinician. J Psychiatric Res 1975; 12:189-198 18.
Hughes CP, Berg L, Danziger WL et al: A new clinical scale for the staging of dementia. The
British J Psychiatry 1982; 140:566-572 19.
Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr Scand
1983; 67:361-370 20.
Evers AW, Kraaimaat FW, van Lankveld W, et al: Beyond unfavorable thinking: the illness
cognition questionnaire for chronic diseases. J Consult Clin Psychol 2001; 69:1026
Page 16 of 24
21.
Troyer AK, Rich JB: Psychometric properties of a new metamemory questionnaire for older
adults. J Gerontol B Psychol Sci Soc Sci; 57:19-22 22.
Logsdon RG, Gibbons LE, McCurry SM et al: Assessing quality of life in older adults with
cognitive impairment. Psychosom Med 2002; 64:510-519 23.
Hindmarch I, Lehfeld H, de Jongh P et al: The Bayer activities of daily living scale (B-ADL).
Dem Geriatr Cogn Dis 1998; 9:20-26 24.
Farias ST, Mungas D, Reed BR et al: The measurement of everyday cognition (ECog): scale
development and psychometric properties. Neuropsychol 2008; 22:531 25.
Teri L, Truax P, Logsdon R et al: Assessment of behavioral problems in dementia: the revised
memory and behavior problems checklist. Psychol and Aging 1992; 7:622 26.
Fuh JL, Liu CY, Wang SJ et al: Revised memory and behavior problems checklist in Taiwanese
patients with Alzheimer's disease. Int Psychogertiatr 1999 11(02):181-9. 27.
Zarit SH, Reever KE, Bach-Peterson J: Relatives of the impaired elderly: correlates of feelings
of burden. Gerontologist 1980; 20:649-655 28.
Ryburn B, Varanelli L, Wells Y: How helpful are our handouts? An evaluation of information
handouts for clients with MCI and their family members. J of Int Neuropsychol Soc 17(S2): 46 [Abstract] 29
Richardson JTE Eta Squared and partial eta squared as measures of effect size in educational
research. Educ Res Rev 2011, 6(2): 135-147 30.
Cohen J: Statistical power analysis for the behavioral sciences, Academic press, 2013
31.
Gitlin LN, Winter L, Corcoran M, et al: Effects of the home environmental skill building
program on the caregiver-care recipient dyad: 6 month outcomes from the Philadelpia REACH initiative. Gerontologist 2002; 43(4):532-546 32.
Provencher V, Mortenson WB, Tanguay-Garneau L, et al: Challenges and strategies
pertaining to recruitment and retention of frail elderly in research studies: A systematic review. Arch Gerontol Geriatr 2014; 59:18-24
Page 17 of 24
33.
Roberts RO, Knopman DS, Mielke MM, et al: Higher risk of progression to dementia in mild
cognitive impairment cases who revert to normal. Neurology 2014; 82:317-325 34. 2012
Miller WR, Rollnick S: Motivational interviewing: Helping people change, Guilford Press,
Page 18 of 24
Figure 1: An outline of the development of the intervention plan and the sessions undertaken in the MAXCOG intervention Figure 2: Consort Chart Figure 3: Pre and Post Assessment Measure on the Canadian Occupational Performance Measure (COPM) – Goal One Figure 4: Pre and Post Assessment Measure on the Canadian Occupational Performance Measure (COPM) – Goal Two
Page 19 of 24
Table 1. Sample Characteristics at Initial Assessment: Age, Activities of Daily Living, and Behavioural Status MAXCOG Parameters
n
Mean (SD), Range
n
Control
Significance of
Effect size
Mean (SD),
group
(Cohen’s D)
Range
differences [t, (p)]
Age, years
25 77.2 (5.8),
15 77.5 (7.7),
65-93
60-88
25 46.4 (2.4),
15 47.3 (2.7),
(max 100)*
25-71
24-64
B-ADL-proxy
25 93.2 (9.1),
15 84.7 (13.6),
MMQ-ability
(max 250)† E-COG-proxy
39-206
31-208
25 81.0 (20.8),
15 71.3 (18.1),
50-118
46-107
(max 195)‡ RMBPC
25 8.1 (3.1),
15 5.6 (4.0),
3-16
0-13
occurrence (max
0.12 (0.90)
0.02
0.26 (0.79)
0.08
0.54 (0.59)
0.17
-1.48 (0.26)
0.49
1.90 (0.06)
0.62
0.37 (0.71)
0.23
24) RMCPC severity
24 12.8 (7.7),
(max 120)
1-31
14 14.2 (14.8), 0-39
*Multifactorial Metacognition Questionnaire – Ability, †Bayar Activities of Daily Living Scale, ‡Everyday Cognition Scale
Page 20 of 24
Table 2: Sample characteristics Parameters
MAXCOG
Control
(n= 25)
(n = 15)
N (%)
N (%)
Male
14 (56)
4 (27)
Female
11 (44)
11 (73)
Primary
2 (8)
5 (33)
Secondary
12 (48)
7 (47)
Diploma
5 (20)
1 (7)
Degree
6 (24)
2 (13)
First
English
22 (88)
13 (87)
Language
Other
3 (12)
2 (13)
Diagnosis
Dementia
3 (12)
3 (20)
MCI
22 (88)
12 (80)
AChEI
Aricept
3 (12)
1 (7)
medication
No drug
22 (88)
14 (93)
Supporter
Sig. other
17 (68)
9 (60)
Adult child
7 (28)
1 (7)
Other family
1 (4)
3 (20)
Friend
1 (4)
2 (13)
Co-resident
Co-resident
16 (64)
10 (67)
supporter
Non-resident
9 (36)
5 (33)
Gender
Education
Categories
Statistic
p
χ2(1) =3.68
.06
χ2(3) =5.49
.14
χ2(1) =0.03
.86
χ2(1) =0.14
.70
χ2(1) =0.68
.41
χ2(3) =0.66
.88
χ2(1) = .01
.93
Notes: SD: standard deviation
Page 21 of 24
Table 3. Participant Outcomes Post intervention Initial assessment
Name
Max score
Post intervention
MAXCOG
Control
MAXCOG
Control
n
n
Mean (SD)
n
Mean (SD)
n
Mean (SD)
Mean (SD)
Significance of Group Differences (interaction effect)
Effect Size (partial eta squared)
Mood/adjustment and quality of life QoLAD
52
25
37.6 (5.7)
15
36.4 (5.6)
25
27.6 (5.6)
15
35.1 (7.0)
F(1,38) = 1.95, p = 0.17
0.05
HADS
56
25
10.2 (5.7)
15
8.8 (4.8)
25
10.0 (5.6)
14
10.9 (7.3)
F(1,37) = 1.83, p = 0.18
0.05
25
10.5 (3.6)
14
9.5 (3.3)
25
10.0 (3.85)
14
11.1 (4.07)
F(1,37) = 4.03, p = 0.05
0.10
25
15.8(3.7)
14
17.7 (3.4)
25
15.5 (4.8)
14
15.6 (3.0)
F(1,37) = 1.80, p = 0.18
0.05
25
36.8 (8.7)
15
37.1 (10.3)
25
37.1 (12.9)
15
36.6 (12.5)
F(1,38) = 0.14, p = 0.71
0.00
25
30.0 (9.7)
15
29.2 (8.1),
25
34.0 (10.3)
15
30.6 (9.1)
F(1,38) = 0.94, p = 0.34
0.02
25
46.3 (12.3)
15
47.3 (10.7)
25
42.8 (11.8)
15
44.4 (12.6)
F(1,38) = 0.07, p = 0.78
0.00
ICS 24 Helplessness ICS 24 Acceptance MMQ 72 contentment Strategy use and ability MMQ 95 strategies MMQ ability 80
Page 22 of 24
Table 4. Supporter Outcomes at Post intervention Initial assessment
Name
Post intervention
Max MAXCOG score n Mean (SD)
Control
MAXCOG
Control
n
n
Mean (SD)
n
250
25 93.2 (45.6)
15 84.7 (52.6)
24
98.9 (47.7)
15 75.1 (42.5)
ECOGmemory
32
25 20.6 (4.4)
15 18.7 (4.4)
24
21.5 (5.2)
15 18.1 (4.6)
ECOGlanguage
36
25 16.5 (4.7)
15 15.5 (4.9)
24
19.1 (5.0)
15 16.2 (4.5)
ECOG- visuo spatial
28
25 12.0 (4.1)
15 11.2 (4.5)
24
15.1 (4.8)
15 14.6 (5.7)
ECOG20 Executive: planning ECOG24 Executive: organization ECOG16 Executive: Divided Attention
25 8.7 (3.4)
15 8.9 (3.2)
24
10.6 (4.2)
15 9.7 (4.8)
23 14.9 (5.1)
15 12.3 (5.1)
23S 13.3 (4.8)
15 9.8 (2.7)
23 6.9 (2.7)
15 7.1 (2.5)
23
15 7.7 (3.2)
Function Bayer ADL scale
Mean (SD)
9.6 (3.1)
Significance of Group Differences (interaction effect)
Effect Size (partial eta squared)
F(1,37) = 0.97, p = 0.33 F(1,37) = 1.75, p = 0.19 F(1,37) = 2.46, p = 0.13 F(1,37) = 2.39, p = 0.13 F(1,37) = 3.86, p = 0.06 F(1,36) = 0.23, p = 0.63 F(1,35) = 0.50, p = 0.48
0.03
Mean (SD)
0.03
0.06
0.06
0.10
0.01
0.01
Strategy Use MMQ – strategy Behaviour RMBPCoverall occurrence
76
25
31.4 (10.1)
15
29.8 (12.4)
24
31.0 (12.3)
15
29.7 (12.3)
F(1,37) = 0.12, p = 0.73
0.00
24
25
8.1(3.1)
15
5.6 (4.0)
24
7.4 (3.8)
15
5.1 (3.4)
F(1,37) = 0.09, p = 0.76
0.00
Page 23 of 24
RMBPC – 96 overall severity Quality of Life QOL- AD 52
25
42.8 (17.0)
13
30.9 (14.8)
24
12.8 (7.6)
14
14.2 (14.8)
F(1,35) = 5.11, p = 0.03
0.13
25
36.5 (4.8)
15
37.5 (7.7)
24
36.4 (4.4)
15
36.8 (7.4),
F(1,37) = 0.22, p = 0.64
0.01
Carer burden and distress Zarit 88 25 20.0 (12.5)
15
13.7 (3.1)
24
20.3 (14.0)
15
14.4 (14.1)
0.00
HADS
15
6.1 (6.3)
24
8.2 (4.6)
14
8.1 (7.7)
F(1,37) = 0.01, p = 0.94 F(1,37) = 1.02, p = 0.32
distress
42
25
8.2 (4.1)
0.03
Page 24 of 24
S1064-7481(16)30302-5 http://dx.doi.org/doi: 10.1016/j.jagp.2016.11.008 AMGP 721
To appear in:
The American Journal of Geriatric Psychiatry
Received date: Revised date: Accepted date:
16-3-2016 7-11-2016 9-11-2016
Please cite this article as: B. Regan, Y. Wells, M. Farrow, P. O'Halloran, B. Workman, MAXCOG – Maximising Cognition: a Randomised Controlled Trial of the Efficacy of Goal Oriented Cognitive Rehabilitation for People with Mild Cognitive Impairment and Early Alzheimer's Disease, The American Journal of Geriatric Psychiatry (2016), http://dx.doi.org/doi: 10.1016/j.jagp.2016.11.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Abstract: 187 Text: 3499 Tables: 4 Figure: 4 References: 34 MAXCOG – Maximising Cognition: A randomised controlled trial of the efficacy of Goal Oriented Cognitive Rehabilitation for people with Mild Cognitive Impairment and Early Alzheimer’s Disease
Regan, B., D.Psych1 Wells, Y., Ph.D.1 Farrow, M., Ph.D.2 O’Halloran, P., Ph.D.3 Workman, B. M.D.4
1
La Trobe University, Australian Institute for Primary Care & Ageing, Melbourne, Australia
2
University of Tasmania, Wicking Dementia Research & Education Centre, Hobart, Australia
3
La Trobe University, School of Psychology and Public Health, Melbourne, Australia
4
Monash University and Monash Health, Monash Ageing Research Centre, Melbourne Australia
Corresponding author Bridget Regan, D.Psych Australian Institute for Primary Care & Ageing La Trobe University Level 5, Health Sciences 2 Bundoora 3083 Victoria, Australia Email: [email protected] Telephone: +61 3 94795809 Fax: +61 3 94795977 Key words: Mild Cognitive Impairment, Dementia, Cognitive Rehabilitation, Clinometric, Translational Supported by a grant from Alzheimer’s Australia Dementia Research Foundation (Vic) and in kind support from Alzheimer’s Australia (Vic) and the Monash Ageing Research Centre. The trial was registered on the Australian New Zealand Clinical Trials Registry (ANZCTR: reg. no. ACTRN12612001149853). Acknowledgments: We thank Vanessa Smithies, Cameron Redpath and Kerith Sharkey for assistance in the study design and conduct. We also thank the Early Intervention Team including Jenny Philip, Christina Bell and Jessica Byers from Alzheimer’s Australia (Vic). No disclosures to report.
Page 1 of 24
ABSTRACT Objectives: To review the efficacy of a home-based four-session individualised face to face cognitive rehabilitation (MAXCOG) intervention for clients with mild cognitive impairment (MCI) or early dementia and their close supporters. Design: Randomised controlled trial comparing the intervention group (MAXCOG) with treatment as usual (control). Participants: Fifty-five client supporter dyads were enrolled in the study and 40 completed; Twenty-five client-supporter dyads completed MAXCOG and 15 completed treatment as usual. Both MAXCOG and control groups included more MCI cases than dementia (22/3 and 12/3 respectively). Intervention: Four weekly individual sessions of MAXCOG consisting of personalised interventions to address individually relevant goals, supported by the provision of the MAXCOG information resource. Measures: The primary outcomes were goal performance and satisfaction, assessed using the Canadian Occupational Performance Measure (COPM). Questionnaires assessing mood, illness adjustment, quality of life, and carer burden were also administered. Results: The intervention group displayed significantly higher performance and satisfaction with primary goals on the COPM post-intervention than the control group, using a per protocol analysis. Conclusion: The MAXCOG intervention is effective in improving goal performance and satisfaction in clients with MCI and early dementia.
Page 2 of 24
INTRODUCTION Psychosocial interventions that address cognitive difficulties and their impact on day-to-day life have been developed for a range of conditions (e.g., learning disabilities, stroke, and traumatic brain injury) [1, 2]. More recently they have been implemented with older adults at a preclinical or early stage of dementia [3, 4]. Emerging evidence suggests that such interventions have the potential to increase the duration of independence, improve quality of life, and save healthcare costs [5]. A range of approaches have been trialled. Remediation relies on repeated, targeted practice of cognitive tasks that are compromised in early dementia. In contrast, rehabilitation (or compensation) approaches focus on maximising physical, psychological, and social wellbeing through external aids (e.g., calendar or smart phone) or adapted strategies (e.g., mnemonic techniques) [5]. To date, only a limited number of studies of cognitive rehabilitation for clients with MCI and early dementia have been published. The bulk of these have been group-based and teach a specific predetermined range of cognitive management strategies (e.g., memory support system). Few are randomised controlled trials (RCTs) [3, 4, 6]. Two recent reviews [4, 6] identified six RCTs on cognitive rehabilitation for people with MCI and early dementia [7-12]. All found significant improvements in use of compensatory memory strategies or memory self-efficacy in the intervention groups, but only two reported improvements in objective cognitive performance [7, 10]. Clare et al. [13] in the UK published the first evaluation of an individualised cognitive rehabilitation intervention for clients with early dementia (N = 69). Their ‘individualised’ approach differed from much of the previous literature in that, rather than teaching a predetermined range of cognitive strategies, each client developed their own personally relevant goals as the focus of their intervention [5]. This single-site, single-blind randomised controlled trial compared an eight-session
Page 3 of 24
cognitive rehabilitation intervention with relaxation therapy, involving equivalent therapist time and attention, and a treatment-as-usual group. An innovative feature of this study was a ‘clinometric’ outcome measure, which used targeted goals specific to each individual to provide a standardised group-level comparison. The measure used was the Canadian Occupational Performance Measure [COPM: 14], reported to be reliable, valid, and responsive as well as more sensitive to clinically significant individual change than more standardised tests or questionnaires [15, 16]. The study showed that the intervention produced significant improvement in self-ratings of goal performance and satisfaction, whereas scores for the other two groups did not change. The current study took the approach of Clare at al. [13] and applied it within a real-world clinical setting. This study has a more translational focus than its predecessor, in that staff involved were not specifically employed by the study but co-opted from an existing early intervention team from the local Alzheimer’s Association. A manual was created and training conducted to encourage consistency between counsellors in implementing the approach. A set of information handouts provided easy access to a range of ideas and strategies that could be utilised as part of a face-to-face intervention. Clare et al.’s [13] individualised approach was replicated, including use of the COPM, but the intervention was shorter in duration, and clients with MCI were included as well as those with early dementia. MCI clients were included because people in this ‘prodromal’ group were thought to be more likely to be capable of taking up cognitive rehabilitation strategies than those with a diagnosis of dementia [8]. Comprising four sessions, MAXCOG is shorter than other interventions reported in the literature; Clare et al. [13] provided a total of eight sessions. With this relatively brief intervention, only one (or sometimes, two) personally relevant goals may be addressed. However, a shorter intervention is easier to implement in a real-world setting than a more extended one. Our aim was to assess the efficacy of this brief cognitive rehabilitation intervention in day-to-day life.
Page 4 of 24
METHODS Study Design. The study was a two-year multi-centre RCT. A 3:1 randomization intervention-tocontrol ratio was adopted (after a six-month period at 1:1) due to concerns regarding initially high attrition levels in the intervention group. A central allocation process was used to randomize participants, using a computer programme and a random block size algorithm to prevent imbalance between the allocation groups, stratified by diagnosis (MCI; early dementia). All sessions were conducted in participants’ homes. The study received ethics approval from Monash University (CF13/2697-2013001446) and each health care site involved in recruitment. Participants. Participants were recruited from six memory clinics across metropolitan Melbourne, Australia. Eligibility criteria were that participants were community-dwelling with a recent (i.e., ≤ 6 months) clinical diagnosis of MCI or early dementia (any type); were fluent in English; were aged ≥ 50 years; had a Mini–Mental State Examination [17] score >20; would receive “1” or less on the Clinical Dementia Rating Scale [18]; and had a primary supporter in close contact with the client and willing to participate in the study. Participants were excluded if they had severe somatic or psychiatric comorbidity, had hearing or vision loss (sufficient to impair their ability to complete the intervention), or were involved in other counselling or cognitive rehabilitation programs. Measures. Pre assessments were administered by a trained research assistant blinded to group allocation until the allocation envelope was opened at the end of the pre assessment. Post assessments were administered by a non-masked assessor. The Canadian Occupational Performance Measure [14] was used to assist clients to identify up to five personally relevant goals in areas relating to self-care, leisure, and productivity. Levels of performance and satisfaction with respect to the goals were elicited on a 10-point scale (1 = unable to perform/ not satisfied and 10 = fully able to perform/ extremely satisfied). Participants in the MAXCOG group selected one or two goals to be the focus of their intervention. Post intervention, all
Page 5 of 24
participants re-rated their performance on and satisfaction with the originally identified goals. The COPM has good test-retest reliability over a one–week period, exceeding 0.80 in studies with different clinical groups [16]. Secondary outcome measures completed at initial interview and post intervention included the Hospital Anxiety and Depression Scale [HADS: 19]; Illness Cognition Questionnaire [ICQ: 20]; Multifactorial Meta Cognition Questionnaire [MMQ: 21]; and Quality of Life – AD [QOL-AD: 22]. Outcome measures completed by a close supporter included Bayar Activities of Daily Living Scale [BADL: 23]; and Everyday Cognition Scale [ECOG: 24]. The Revised Memory and Behaviour Problem Checklist [RMBPC: 25], which has good test-retest reliability (0.89 for frequency and 0.74 for reaction – [see 26]), was also used. Supporters also rated their own mental status using the Zarit Carer Burden [27] and the Hospital Anxiety and Depression Scale for supporters [HADS-S: 19]. Intervention. MAXCOG involved four weekly one-hour sessions delivered to client-supporter dyads in the client’s home with a focus on an individualised intervention addressing personally meaningful goals (e.g., to learn the names of people at church/ bowls club; to be less dependent on the spouse for reminders about appointments; to take medication reliably; to develop a better system for bookwork; and to remember more about grandchildren’s activities). The MAXCOG intervention was delivered by experienced counsellors (including staff from the Alzheimer’s Australia Victoria Early Intervention Team and a neuropsychologist). Strategies could be selected from those outlined in information handouts recently developed by our project team at La Trobe University and the Kingston Centre [28]. While the basic structure of sessions was prescribed in a manual, their content could be adapted flexibly to meet specific client goals. [insert figure 1 about here] Clients and supporters were encouraged to help brainstorm and select the most appropriate strategies. The focus of strategies was on positive resources, intact functions, retained skills, and
Page 6 of 24
activities clients could still take part in (see [28] for examples). Clients were encouraged to practice techniques with assistance from their supporter in between sessions. Clients enrolled in the control condition had no contact with the research team between their initial and post intervention assessments. Procedure. Potential participants were identified by memory clinic clinicians as clients of the local Alzheimer’s Association. Participants who met study criteria and gave informed consent received an initial assessment. If the participant was randomised into the treatment group, a referral was made to a counsellor to conduct the MAXCOG intervention. The counsellor was provided with a preliminary intervention plan compiled by the research team outlining potential areas of focus (based on COPM results) and ideas for intervention strategies (see figure 1). Once the intervention was completed, or four weeks had passed in the case of the control group, the research assistant collected follow up data as soon as feasible (up to four weeks post-intervention) either via telephone (with mailed questionnaires) or in person. Data Analysis. Behavioural data were examined only after all participants had completed the postintervention assessment. To compare baseline performances independent samples t-tests were used. For each outcome measure, mixed-model repeated measures ANOVAs were conducted to determine whether groups differed in change from pre- to post-intervention. Sharpio-Wilk, Fmax and Levene’s test statistics were used to test assumptions of normality and homogeneity of variance. Missing data within scales were replaced by the mean value for that item. Effect sizes were calculated using Cohen’s D for t-tests and partial eta-squared for non-independent repeated measures analyses [29, 30].
RESULTS
Page 7 of 24
Participants. Clients were 40 individuals (22 women, 18 men; mean age 77.2 years, standard deviation [SD] 6.5, range = 60-93). Supporters were 40 individuals (13 men, 27 women): mean age was 66.8 (N = 36, SD 15.0, range = 30-87). Participant characteristics are summarized in Table 1. There were no statistically significant differences between the two groups at baseline on most measures, including function and cognitive status, suggesting that the randomization process was largely successful at creating equivalent groups. Nevertheless, relatively large effect sizes on two measures at baseline—the RMBPC and the E-COG proxy—suggested a trend for supporters in the MAXCOG group to report more behavioural and cognitive difficulties than those in the control group. [Insert table 1 about here]
[Insert table 2 about here] The CONSORT chart is provided in Fig. 2. Of the 55 randomised participants, 37 were allocated to the MAXCOG intervention and 18 to the control group, using a 3:1 ratio. Post-intervention data were collected for 25 MAXCOG and 15 control group participants. Four protocol deviations were identified, in which the clients were found to be too severely affected by dementia to participate in the intervention or complete questionnaires (rated > 1 on the Clinical Dementia Rating Scale). None of the other clients who withdrew were available for post-intervention assessment, and therefore the analysis is a per protocol analysis. Mean duration between the collection of the pre- and postintervention data was slightly longer for the intervention group (115 versus 95 days); however, this difference was not statistically significant F (1,35) = 1.82, p = > 0.05. [insert figure 2 about here] Primary Outcomes. The primary outcome measures were goal performance and satisfaction scores on the COPM. At initial assessment, all 40 participants included in the per protocol analysis
Page 8 of 24
identified at least one goal for improvement. Between them, these clients identified 143 goals. Rates of goal identification were similar for the two groups (MAXCOG mean 3.7, SD = 1.2, range 1-5; control mean 3.3, SD = 1.2, range 1-5). Common areas of focus included remembering names and keeping track of items around the home, as well as increased socialisation, improvements in using technology, and managing finances and bookwork. Mean COPM performance and satisfaction scores are summarised in figures 3 and 4. [insert figure 3 about here] [insert figure 4 about here] For the first performance goal, a significant interaction effect was obtained between group and time F(1,37) = 4.77, p = 0.035, partial η2 = .11, with an increase in scores for the MAXCOG group but not the control group. For the second performance goal, a significant main effect of time was found F(1,33) = 5.43, p=.026, partial η2=.14, showing that for both groups the mean score on the performance COPM measure was higher post intervention. On the satisfaction measure for the first goal, both a significant main effect for time (F(1, 36) =9.81, p = 0.003, partial η2=.21) and a significant interaction effect (F(1,36)=7.77, p=0.008, partial η2=.18) were observed; inspection of mean scores indicated that satisfaction increased dramatically for the MAXCOG group but remained stable for the control group. Results of analysis of satisfaction for the second goal was similar to those for the first goal with a significant interaction effect (F(1,32)=7.18, p = .012, partial η2= .18); inspection of the means again revealed an increase in satisfaction in the MAXCOG group and stability or a slight decrease in the control group. In summary, on the first goal, both performance and satisfaction increased significantly for the MAXCOG group versus the control group, whereas on the second goal where sample sizes were smaller a significant difference between groups was noted only for satisfaction. [insert table 3 about here]
Page 9 of 24
Secondary Outcomes. Several significant findings were noted on the secondary outcome measures. A significant main effect of time (F (1, 38) = 7.05, p= .012, η2= .156) indicated that both client groups rated their own memory abilities on the multifactorial metacognition questionnaire as decreasing between pre and post measures (see Table 4). Similarly, supporters from both groups rated clients as significantly more impaired on their overall ECOG score on post measures (F(1,35) = 11.81, p = .002, η2= .252; see Table 4). [insert table 4 about here] On the RMBPC, supporters from the MAXCOG group indicated a significantly greater number of problem behaviours both pre- and post-assessment than the control group (F (1,37) = 5.82, p = 0.027, η2= .125; see Table 4). Results on supporters’ ratings of how much the problem bothered the client were interesting. There was both a main effect of time, with both groups of supporters rating behaviours as less severe post-intervention (F(1,35) = 57.97, p =.00, η2= .624), and an interaction effect with MAXCOG group scores declining more than control group scores (F(1,35) = 5.11, p = 0.03, η2= .127). There was a trend for scores on the helplessness scale (Illness Cognition Questionnaire) to increase over time for the control group but to decrease slightly for the MAXCOG group (F(1, 37) = 4.03, p = .052, η2= .098; see Table 4). No significant differences between pre and post intervention were noted on any other measures, including measures of quality of life and mood.
DISCUSSION This RCT investigated the efficacy of the MAXCOG cognitive rehabilitation intervention versus a treatment-as-usual control group. The intervention utilised individualised intervention plans to address personalised goals in an everyday setting, and progress was assessed using a clinometric
Page 10 of 24
measure. Overall our findings on the impacts of the MAXCOG intervention were encouraging. Our main finding was that clients in the MAXCOG group reported higher levels of goal performance and satisfaction after receiving the intervention than those in the control group. This was the case for the first goal and partially the case for the second goal (where satisfaction but not performance increased for the MAXCOG group). These improvements had moderate effect sizes with partial etasquared values of between .11 and .21 [29]. Findings suggest that the MAXCOG intervention was effective at assisting clients to reach at least one goal in comparison to the control group. This finding is consistent with the findings of the Clare et al. [13] study, revealing the COPM to be a sensitive measure to clinical change associated with goal formation and achievement. The COPM is the most proximal measure of the central focus of the intervention, and is not surprising that more distal measures, such as overall quality of life and depression, which are unlikely to change in response to practical improvement in one goal area, remained stable between the two assessment time points. However, other measures suggested additional positive impacts for the MAXCOG intervention. At baseline and post-intervention, MAXCOG group supporters reported a higher overall number of behavioural difficulties than the control group. However, the ratings of the severity of these behaviours changed between pre and post measures; the MAXCOG group improved significantly more than the control group. This suggests that, although challenging behaviours were still present post intervention, the MAXCOG intervention may have been helpful in reducing their perceived severity (other intervention studies using the RMBPC have reported similar findings - see [31]). Another explanation for this finding could be ‘regression to the mean’. Both groups rated their own memory abilities worse post-intervention than at baseline, and their supporters rated both groups’ overall day-to-day functional abilities as more impaired. This could reflect the fact that all participants were declining. Alternatively, it may be due to decreased confidence or increased awareness brought about by participating in the study or because clients
Page 11 of 24
felt more comfortable with the study procedures post intervention than at baseline and so were more honest about their difficulties. Limitations. Several methodological issues need to be considered when interpreting the results of the study, including the fact that the study utilised a translational design with a ‘real world’ clinical team. One factor that may have enhanced our significant main finding is the relatively large number of withdrawals post allocation (15 in total; 27%). Although many studies with older adults report attrition for a range of reasons [32], this study sustained a much higher proportion of withdrawals than that reported in the Clare et al. [8] study (5.8%). In the MAXCOG group, although six withdrawals were largely unavoidable (illness or dementia severity), a further six were because the client, supporter, and counsellor were unwilling to develop or follow through with an intervention plan (for the areas of difficulty earlier identified on the COPM) or because after the first session a client decided to withdraw. Therefore, only the more insightful and willing participants completed the study and this group may be more likely to be committed to the counselling process and to report positive outcomes from it. Conceivably, utilising a ‘real world’ clinical team with broad counselling backgrounds, rather than an occupational therapist specifically employed and trained to be part of the study (as in the Clare et al. study), may have led to differences in counsellors’ capacity or willingness to encourage clients to identify goals and continue to participate. Again, the omission of clients who could not be persuaded to participate may have enhanced the main findings. Another challenge with the use of a ‘real world’ clinical team was that the client-focussed nature of their usual work may have led to inconsistencies in their approach to the intervention and therefore potential threats to the internal validity of the study. Due to privacy concerns, it was not possible to conduct any direct monitoring of their ‘fidelity to treatment’ during sessions.
Page 12 of 24
Another limitation of the study design was the fact that no active control group was employed (e.g., a support group intervention to determine the specificity of the cognitive rehabilitation in addition to any non-specific effects of therapeutic intervention). The use of a blinded research assistant at follow up would have further enhanced the internal validity of the study. Finally, this study included both clients with MCI and early dementia. Clients’ diagnoses included a mixture of vascular, Alzheimer’s, and other pathologies. Different pathologies and different stages of illness may be associated with different cognitive strengths and weaknesses. Indeed a subgroup of approximately 20% of clients with MCI may improve or stabilise without developing dementia [33]. Therefore clients may differ in the extent to which they retain the capacity to benefit from cognitive rehabilitation approaches. Further research is required to determine which subgroups may respond best to such approaches. Implications Given the limitations in the study design outlined, the findings from this study should be considered preliminary. However, the findings do suggest that even relatively short individualised cognitive rehabilitation interventions may benefit clients by helping them to achieve their goals. In a ‘real world’ setting, the brevity of the intervention is critical, due to resourcing and cost considerations. Clinometric measures such as the COPM show promise as sensitive measures of clinically relevant outcomes. However, a remaining question is the extent to which participants’ responses to the COPM, particularly when rating performance and satisfaction post assessment, are influenced by social desirability and acquiescent response set, rather than an honest assessment of progress towards their goals. Again an active control group may help mitigate these concerns. Another strategy worth considering, particularly for groups whose insight and cognitive capacity to understand the rating scales are reduced [15], is the use of specific informant- or clinician-rated goal attainment scales.
Page 13 of 24
Another issue for future research is how to reduce attrition, particularly for the group that struggled to develop an intervention plan. One possibility may be to integrate motivational interviewing principles into the initial goal setting interview, to explore and address any ambivalence that the client may express [34]. Investigating which subgroups of clients with MCI or early dementia are likely to complete and benefit from cognitive rehabilitation interventions such as MAXCOG will also be important. If future studies do confirm the success of this type of intervention, it will add weight to the view that it is possible to improve specific goal-oriented outcomes for people with MCI and early dementia by addressing relevant personal, social, and environmental factors. In other words, it is not necessary to address impairment per se in order to improve engagement in activity and social participation and increase wellness.
Page 14 of 24
REFERENCES 1.
Cicerone KD, Langenbahn DM, Braden C, et al: Evidence based cognitive rehabilitation:
updated review of the literature from 2003 through 2008. Arch Phys Med Rehabil 2008; 92(4): 519530 2.
Bayley MT, Tate R, Douglas JM, et al: INCOG guidelines for cognitive rehabilitation following
traumatic brain injury: methods and overview. J Head Trauma Rehabil. 2014, 29(4):290-306 3.
Bahar-Fuchs A, Claire L, Woods B: Cognitive rehabilitation and cognitive training for early-
stage Alzheimer's disease and vascular dementia. Cochrane Database of Syst Rev 2013; 6: Art. No.: CD003260, DOI: 10.1002/14651858.CD003260.pub2 4
Rodakowski J, Saghafi E, Butters MA, et al: Non-pharmacological interventions for adults
with mild cognitive impairment and early stage dementia: An updated scoping review. Mol Aspects Med 2015; 43: 38-53 5
Clare L, Bayer A, Burns A, et al: Goal-oriented cognitive rehabilitation in early-stage
dementia: study protocol for a multi-centre single-blind randomised controlled trial (GREAT). Trials 2013; 14:152 6
Huckans M, Hutson L, Twamley E, et al: Efficacy of cognitive rehabilitation therapies for mild
cognitive impairment (MCI) in older adults: working toward a theoretical model and evidence-based interventions. Neuropsychol Rev 2013; 23:63-80 7
Greenaway M, Duncan N, Smith G: The memory support system for mild cognitive
impairment: radomised trial of a cognitive rehabilitation intervention. Int J Geriatr Psychiatry 2013; 28:402-409 8.
Kinsella GJ, Mullaly E, Rand E et al: Early intervention for mild cognitive impairment: a
randomised controlled trial. J Neurol Neurosurg Neuropsychiatry 2009; 80:730-736 9.
Rapp S, Brenes G,Marsh A: Memory enhancement training for older adults with mild
cognitive impairment: a preliminary study. Aging Ment Health 2002; 6(1):5-11
Page 15 of 24
10.
Jelcic N, Cagnin A, Meneghello F, et al: Effects of lexical–semantic treatment on memory in
early Alzheimer disease: an observer-blinded randomized controlled trial. Neurorehabil Neural Repair. 2012;26(8):949-56. 11
Kurz A, Thöne-Otto A, Cramer B, et al: CORDIAL: cognitive rehabilitation and cognitive-
behavioral treatment for early dementia in Alzheimer disease: a multicenter, randomized, controlled trial. Alzheimer Dis Assoc Disord 2012;26(3):246-53. 12
Troyer AK, Murphy KJ, Anderson ND, et al: Changing everyday memory behaviour in
amnestic mild cognitive impairment: a randomised controlled trial. Neuropsychol Rehabil 2008; 18:65-88 13
Clare L, Linden DE, Woods RT et al: Goal-oriented cognitive rehabilitation for people with
early-stage Alzheimer disease: a single-blind randomized controlled trial of clinical efficacy. Am J Geriatr Psychiatry 2010; 18:928-939 14
Law M, Baptise S, Carswell A et al: Canadian Occupational Performance Measure 4th edition,
Ottowa, CAOT publications ACE, 2005 15
Carswell A, McColl MA, Baptise S et al: The Canadian Occupational Performance Measure: a
research and clinical literature review. Can J Occ Ther 2004; 71: 210-222 16
Cup EHC, Scholte op Reimer, WJM et al: Reliability and validity of the canadian occupational
performance measure in stroke patients. Clin Rehabil 2003, 17(4): 402-9 17.
Folstein M, Folstein S,McHugh P: "Mini-mental State". A practical method of grading the
cognitive state of patients for the clinician. J Psychiatric Res 1975; 12:189-198 18.
Hughes CP, Berg L, Danziger WL et al: A new clinical scale for the staging of dementia. The
British J Psychiatry 1982; 140:566-572 19.
Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr Scand
1983; 67:361-370 20.
Evers AW, Kraaimaat FW, van Lankveld W, et al: Beyond unfavorable thinking: the illness
cognition questionnaire for chronic diseases. J Consult Clin Psychol 2001; 69:1026
Page 16 of 24
21.
Troyer AK, Rich JB: Psychometric properties of a new metamemory questionnaire for older
adults. J Gerontol B Psychol Sci Soc Sci; 57:19-22 22.
Logsdon RG, Gibbons LE, McCurry SM et al: Assessing quality of life in older adults with
cognitive impairment. Psychosom Med 2002; 64:510-519 23.
Hindmarch I, Lehfeld H, de Jongh P et al: The Bayer activities of daily living scale (B-ADL).
Dem Geriatr Cogn Dis 1998; 9:20-26 24.
Farias ST, Mungas D, Reed BR et al: The measurement of everyday cognition (ECog): scale
development and psychometric properties. Neuropsychol 2008; 22:531 25.
Teri L, Truax P, Logsdon R et al: Assessment of behavioral problems in dementia: the revised
memory and behavior problems checklist. Psychol and Aging 1992; 7:622 26.
Fuh JL, Liu CY, Wang SJ et al: Revised memory and behavior problems checklist in Taiwanese
patients with Alzheimer's disease. Int Psychogertiatr 1999 11(02):181-9. 27.
Zarit SH, Reever KE, Bach-Peterson J: Relatives of the impaired elderly: correlates of feelings
of burden. Gerontologist 1980; 20:649-655 28.
Ryburn B, Varanelli L, Wells Y: How helpful are our handouts? An evaluation of information
handouts for clients with MCI and their family members. J of Int Neuropsychol Soc 17(S2): 46 [Abstract] 29
Richardson JTE Eta Squared and partial eta squared as measures of effect size in educational
research. Educ Res Rev 2011, 6(2): 135-147 30.
Cohen J: Statistical power analysis for the behavioral sciences, Academic press, 2013
31.
Gitlin LN, Winter L, Corcoran M, et al: Effects of the home environmental skill building
program on the caregiver-care recipient dyad: 6 month outcomes from the Philadelpia REACH initiative. Gerontologist 2002; 43(4):532-546 32.
Provencher V, Mortenson WB, Tanguay-Garneau L, et al: Challenges and strategies
pertaining to recruitment and retention of frail elderly in research studies: A systematic review. Arch Gerontol Geriatr 2014; 59:18-24
Page 17 of 24
33.
Roberts RO, Knopman DS, Mielke MM, et al: Higher risk of progression to dementia in mild
cognitive impairment cases who revert to normal. Neurology 2014; 82:317-325 34. 2012
Miller WR, Rollnick S: Motivational interviewing: Helping people change, Guilford Press,
Page 18 of 24
Figure 1: An outline of the development of the intervention plan and the sessions undertaken in the MAXCOG intervention Figure 2: Consort Chart Figure 3: Pre and Post Assessment Measure on the Canadian Occupational Performance Measure (COPM) – Goal One Figure 4: Pre and Post Assessment Measure on the Canadian Occupational Performance Measure (COPM) – Goal Two
Page 19 of 24
Table 1. Sample Characteristics at Initial Assessment: Age, Activities of Daily Living, and Behavioural Status MAXCOG Parameters
n
Mean (SD), Range
n
Control
Significance of
Effect size
Mean (SD),
group
(Cohen’s D)
Range
differences [t, (p)]
Age, years
25 77.2 (5.8),
15 77.5 (7.7),
65-93
60-88
25 46.4 (2.4),
15 47.3 (2.7),
(max 100)*
25-71
24-64
B-ADL-proxy
25 93.2 (9.1),
15 84.7 (13.6),
MMQ-ability
(max 250)† E-COG-proxy
39-206
31-208
25 81.0 (20.8),
15 71.3 (18.1),
50-118
46-107
(max 195)‡ RMBPC
25 8.1 (3.1),
15 5.6 (4.0),
3-16
0-13
occurrence (max
0.12 (0.90)
0.02
0.26 (0.79)
0.08
0.54 (0.59)
0.17
-1.48 (0.26)
0.49
1.90 (0.06)
0.62
0.37 (0.71)
0.23
24) RMCPC severity
24 12.8 (7.7),
(max 120)
1-31
14 14.2 (14.8), 0-39
*Multifactorial Metacognition Questionnaire – Ability, †Bayar Activities of Daily Living Scale, ‡Everyday Cognition Scale
Page 20 of 24
Table 2: Sample characteristics Parameters
MAXCOG
Control
(n= 25)
(n = 15)
N (%)
N (%)
Male
14 (56)
4 (27)
Female
11 (44)
11 (73)
Primary
2 (8)
5 (33)
Secondary
12 (48)
7 (47)
Diploma
5 (20)
1 (7)
Degree
6 (24)
2 (13)
First
English
22 (88)
13 (87)
Language
Other
3 (12)
2 (13)
Diagnosis
Dementia
3 (12)
3 (20)
MCI
22 (88)
12 (80)
AChEI
Aricept
3 (12)
1 (7)
medication
No drug
22 (88)
14 (93)
Supporter
Sig. other
17 (68)
9 (60)
Adult child
7 (28)
1 (7)
Other family
1 (4)
3 (20)
Friend
1 (4)
2 (13)
Co-resident
Co-resident
16 (64)
10 (67)
supporter
Non-resident
9 (36)
5 (33)
Gender
Education
Categories
Statistic
p
χ2(1) =3.68
.06
χ2(3) =5.49
.14
χ2(1) =0.03
.86
χ2(1) =0.14
.70
χ2(1) =0.68
.41
χ2(3) =0.66
.88
χ2(1) = .01
.93
Notes: SD: standard deviation
Page 21 of 24
Table 3. Participant Outcomes Post intervention Initial assessment
Name
Max score
Post intervention
MAXCOG
Control
MAXCOG
Control
n
n
Mean (SD)
n
Mean (SD)
n
Mean (SD)
Mean (SD)
Significance of Group Differences (interaction effect)
Effect Size (partial eta squared)
Mood/adjustment and quality of life QoLAD
52
25
37.6 (5.7)
15
36.4 (5.6)
25
27.6 (5.6)
15
35.1 (7.0)
F(1,38) = 1.95, p = 0.17
0.05
HADS
56
25
10.2 (5.7)
15
8.8 (4.8)
25
10.0 (5.6)
14
10.9 (7.3)
F(1,37) = 1.83, p = 0.18
0.05
25
10.5 (3.6)
14
9.5 (3.3)
25
10.0 (3.85)
14
11.1 (4.07)
F(1,37) = 4.03, p = 0.05
0.10
25
15.8(3.7)
14
17.7 (3.4)
25
15.5 (4.8)
14
15.6 (3.0)
F(1,37) = 1.80, p = 0.18
0.05
25
36.8 (8.7)
15
37.1 (10.3)
25
37.1 (12.9)
15
36.6 (12.5)
F(1,38) = 0.14, p = 0.71
0.00
25
30.0 (9.7)
15
29.2 (8.1),
25
34.0 (10.3)
15
30.6 (9.1)
F(1,38) = 0.94, p = 0.34
0.02
25
46.3 (12.3)
15
47.3 (10.7)
25
42.8 (11.8)
15
44.4 (12.6)
F(1,38) = 0.07, p = 0.78
0.00
ICS 24 Helplessness ICS 24 Acceptance MMQ 72 contentment Strategy use and ability MMQ 95 strategies MMQ ability 80
Page 22 of 24
Table 4. Supporter Outcomes at Post intervention Initial assessment
Name
Post intervention
Max MAXCOG score n Mean (SD)
Control
MAXCOG
Control
n
n
Mean (SD)
n
250
25 93.2 (45.6)
15 84.7 (52.6)
24
98.9 (47.7)
15 75.1 (42.5)
ECOGmemory
32
25 20.6 (4.4)
15 18.7 (4.4)
24
21.5 (5.2)
15 18.1 (4.6)
ECOGlanguage
36
25 16.5 (4.7)
15 15.5 (4.9)
24
19.1 (5.0)
15 16.2 (4.5)
ECOG- visuo spatial
28
25 12.0 (4.1)
15 11.2 (4.5)
24
15.1 (4.8)
15 14.6 (5.7)
ECOG20 Executive: planning ECOG24 Executive: organization ECOG16 Executive: Divided Attention
25 8.7 (3.4)
15 8.9 (3.2)
24
10.6 (4.2)
15 9.7 (4.8)
23 14.9 (5.1)
15 12.3 (5.1)
23S 13.3 (4.8)
15 9.8 (2.7)
23 6.9 (2.7)
15 7.1 (2.5)
23
15 7.7 (3.2)
Function Bayer ADL scale
Mean (SD)
9.6 (3.1)
Significance of Group Differences (interaction effect)
Effect Size (partial eta squared)
F(1,37) = 0.97, p = 0.33 F(1,37) = 1.75, p = 0.19 F(1,37) = 2.46, p = 0.13 F(1,37) = 2.39, p = 0.13 F(1,37) = 3.86, p = 0.06 F(1,36) = 0.23, p = 0.63 F(1,35) = 0.50, p = 0.48
0.03
Mean (SD)
0.03
0.06
0.06
0.10
0.01
0.01
Strategy Use MMQ – strategy Behaviour RMBPCoverall occurrence
76
25
31.4 (10.1)
15
29.8 (12.4)
24
31.0 (12.3)
15
29.7 (12.3)
F(1,37) = 0.12, p = 0.73
0.00
24
25
8.1(3.1)
15
5.6 (4.0)
24
7.4 (3.8)
15
5.1 (3.4)
F(1,37) = 0.09, p = 0.76
0.00
Page 23 of 24
RMBPC – 96 overall severity Quality of Life QOL- AD 52
25
42.8 (17.0)
13
30.9 (14.8)
24
12.8 (7.6)
14
14.2 (14.8)
F(1,35) = 5.11, p = 0.03
0.13
25
36.5 (4.8)
15
37.5 (7.7)
24
36.4 (4.4)
15
36.8 (7.4),
F(1,37) = 0.22, p = 0.64
0.01
Carer burden and distress Zarit 88 25 20.0 (12.5)
15
13.7 (3.1)
24
20.3 (14.0)
15
14.4 (14.1)
0.00
HADS
15
6.1 (6.3)
24
8.2 (4.6)
14
8.1 (7.7)
F(1,37) = 0.01, p = 0.94 F(1,37) = 1.02, p = 0.32
distress
42
25
8.2 (4.1)
0.03
Page 24 of 24