The effect of computerized online brain training on post-stroke fatigue, cognitive functioning, and daily living in stroke patients
Nicky Beugeling
Department of Psychology: Brain and Cognition, University of Amsterdam
Master thesis Clinical Neuropsychology & Health Psychology Student number: 10427473
Date: January 19th 2015 Supervisor: R.M. van de Ven Second supervisor: S. van Gaal
Table of contents
Abstract………..……2
Introduction………...….3
Post-stroke fatigue and cognitive functioning………...………4
Computer-based cognitive rehabilitation………..…...5
Training Project Amsterdam Senior and Stroke……….…...6
Methods & Procedure………....7
Sample characteristics….………..7 Procedure………..8 Material………...9 Questionnaires………..………..………..9 Neuropsychological assessment……….…...………..11 Training……….……….………..13 Statistical analyses……….………...14 Results……….……….16 Patient characteristics……….………...16
Relationship fatigue, cognitive functioning and daily life functioning………...17
Training effect on cognitive functioning………..…..………..………...18
Training effect on fatigue………..………...………...19
Training effect on objective and subjective daily life functioning………....……..20
Discussion………….………...21
Abstract
There are many contradicting findings about post-stroke fatigue and its relationship with cognitive impairment. Both post-stroke fatigue and cognitive impairment are seriously debilitating for stroke patients and influence the patient’s quality of life. Adjacent to this, post-stroke fatigue and cognitive impairment both interfere with the rehabilitation program and cause higher health-care costs. Finding an effective and efficient treatment would therefore be very meaningful. The main goal in the present study was to examine the influence of computerized cognitive flexibility training on post-stroke fatigue, cognitive functioning and the influence on daily life functioning. A secondary aim was to explore the relationship between post-stroke fatigue, cognitive impairment and the effect on daily life functioning. Over forty stroke patients did a twelve week computerized online brain training, of which half of the participants were assigned to do the cognitive flexibility training and half of the participants to a mock training. Questionnaires and neuropsychological assessments were administered to measure the change in extent of fatigue, cognitive functioning and daily life functioning. The results indicated that the extent of fatigue was not related to cognitive impairment. It was however associated with subjective daily life functioning. The computerized online brain training did not have any effect on the extent of fatigue or daily life functioning. Unfortunately it is still unclear whether computerized online brain training is an effective intervention for cognitive impairment. The participants improved on neuropsychological assessments in both groups. This was probably due to practice effects. Future studies should aim to include a placebo group and enquire patients with established post-stroke fatigue in a more homogenous sample.
Keywords: post-stroke fatigue, cognitive functioning, daily life functioning, computerized online brain training, cognitive flexibility training.
Introduction
Stroke is one of the leading causes of death and one of the most important causes of disability in the Netherlands. Every year 41.000 Dutch citizens are diagnosed with stroke, of which one in five dies within one year after being hospitalized (hersenstichting.nl, 2014).A common and well known consequence of stroke is cognitive impairment. The estimated prevalence ranges from 22% to 47.3% (Douiri et al., 2013; Jacquin et al. 2014), and persistent cognitive impairment can last up to 11-years post stroke (Schaapsmeerders et al., 2013). Cognitive impairment has an huge impact on the ability to perform activities of daily living, such as cooking, dressing and undressing, and bathing (ADL; Claesson, linden, Skoog & Blomstrand,2005). Not only is cognitive impairment very debilitating for the patient, it also causes higher rates of institutionalization (Pasquini, Leys, Rousseaux, Pasquier & Hénon, 2007), and greater health-care costs (Claesson et al, 2005).
Fatigue is also a very common, but often underestimated complaint among stroke survivors. The estimated prevalence of post-stroke fatigue (PSF) varies between 39% and 72% (e.g., Carlsson, Moller & Blomstrand 2003; Choi-Kwon, Han, Kwon & Kim, 2005; Ingles, Eskes & Philips, 1999). Forty to fifty percent of stroke patients even indicated that fatigue is their main complaint, or one of their most debilitating symptoms (Ingles, 1999; van der Werf, 2001). It seems that fatigue problems arise shortly after stroke, and can persist for a long period of time (van der Werf, 2001; Schepers, 2006; Crosby, 2012). PSF often interferes with rehabilitation and resuming work (Glader, Stegmayr & Asplund, 2002; Andersen, Christensen, Kirkevold & Johnsen, 2011). Despite its high prevalence and debilitating effects on the patient, health care workers pay little attention to the patients’ fatigue and its effect on the rehabilitation program (Crosby, Munshi, Karat, Worthington & Lincoln, 2012).
Post-stroke fatigue and cognitive impairment
Fatigue is a complex, subjective experience, which is difficult to objectify. Staub and Bogousslavsky (2001, p.1) defined subjective fatigue in stroke patients as “a feeling of early exhaustion developing during mental activity, with weariness, lack of energy, and aversion to effort”. Patients experience PSF as a new type of fatigue that comes suddenly and without any specific reason (Bendz, 2003). Lerdal et al. (2009) conclude in their review that PSF might have an impact on energy consuming activities of daily living, such as leisure situations and social contact. Unfortunately, today, there is still relatively limited knowledge on the etiology of PSF. However, one hypothesis is the so called Coping hypothesis. This hypothesis postulates that fatigue may be due to the additional compensatory effort expended by individuals with brain injury in meeting the demands of everyday life in the presence of cognitive impairments (DeLuca, 2005).
Some recent studies did indeed find an association between PSF and cognitive functioning. Hubacher et al. (2012) found that the fatigue questionnaires in their study showed correlations with short-term memory, mental speed, working memory and executive functions. Pihlaja, Uimonen, Mustanoja, Tatlisumak & Poutiainen (2014) also found that stroke patients with PSF had impaired processing speed and impaired memory performance compared with stroke patients without PSF. They did not find an association between executive functions and PSF. Impaired processing speed has also been linked to fatigue in other neurological conditions, e.g. traumatic brain injury (TBI; Johansson, Berglund, & Rönnbäck, 2009) and Multiple Sclerosis (MS; Andreasen, Spliid, Andersen & Jakobsen, 2010).
Even though the Coping hypothesis seems like a plausible explanation, some studies did not find any relation between experienced fatigue and cognitive impairments (Crosby,
2012; van Zandvoort, 1998). It was also found that PSF is a primary consequence of stroke rather than a secondary complication (Christensen et al., 2008). This would mean that PSF is not a consequence of cognitive impairment but a direct consequence of stroke. In defense Staub and Bogousslavsky (2001) argue that the compensatory efforts to meet the demands of normal life might be particularly relevant in the less impaired patients, because they may have a greater awareness of their deficits and because there might be a greater social pressure for them to resume previous activities.
In summary, there are many contradicting findings about PSF and its relationship with cognitive impairment. Both PSF and cognitive impairment are seriously debilitating for stroke patients and influence the patient’s quality of life. Adjacent to this PSF and cognitive impairment both interfere with the rehabilitation program and cause higher health-care costs. Finding an effective and efficient treatment would therefore be very meaningful.
Computer-based cognitive rehabilitation
The only currently evidence-based treatment for PSF is the Cognitive and Graded Activity Training (COGRAT; Zedlitz, Rietveld, Geurts, & Fasotti, 2012). This treatment assumes that PSF is a primary consequence of stroke, and its goal is to change patients’ behavior so that they learn how to manage PSF in a more efficient way. A Graded activity training (GRAT) is offered alongside the cognitive treatment. In this training the physical condition is improved.
Another possible intervention, instead of handing compensation strategies, is improving cognitive functions through computer-based cognitive rehabilitation (CBCR; Cha&Kim, 2013). A recent study by Björkdahl, Akerlund, Svensson & Esbjörnsson (2013) found an increase in working memory and a decrease in fatigue (FIS; Fatigue Impact Scale) in patients with acquired brain injury after a computerized working memory training (Cogmed QM training; Klingberg, Forssberg & Westerberg, 2002). An important advantage of computerized brain training over traditional cognitive compensatory training is the
cost-effectiveness of individualized treatment, which is not institution-based and can also be done by homebound patients (Cha & Kim, 2013; Kueider, Parisi, Gross, & Rebok, 2012).
Although this looks very profitable, some studies doubt that computerized cognitive brain training improves general cognitive functioning. Even though studies have shown cognitive improvement in the tasks that were trained, Owen et al. (2010) could not find transfer effects to untrained tasks, even when those tasks were cognitively closely related. Nevertheless a recent review proposed that computerized cognitive brain training in healthy seniors, with frequent switching between various training tasks, may result in cognitive improvement (Buitenweg, Murre, & Ridderinkhof, 2012). Karbach & Kray (2009) found that the results of such cognitive flexibility training can even be generalized to tasks more distant from the trained skill.
Training Project Amsterdam Senior and Stroke
The ‘Training Project Amsterdam Senior and Stroke’ (TAPASS study) aims, among other things, to examine whether computer-based cognitive flexibility training improves cognitive functioning after stroke. This master thesis is part of the TAPASS study. The main goal of this study was to examine the influence of computerized cognitive flexibility training on PSF, cognitive functioning, and the influence on daily life functioning. However to examine the effectiveness of computerized online brain training we also have to understand the relationship between PSF and cognitive impairment. For this reason, a secondary aim of this study was to explore the relationship between PSF, cognitive functioning and the effect on daily life functioning.
Before examining the effect of computerized cognitive flexibility training, the relationship between PSF, cognitive functioning and daily life functioning was explored. A negative correlation between PSF and cognitive functioning at baseline was expected, which means that high cognitive functioning is associated with low PSF. Thereby it was expected that
processing speed would most likely negatively correlate with PSF, since impaired processing speed has particularly been linked to fatigue in previous studies. It was further expected that daily life functioning negatively correlates with PSF at baseline, which means that low PSF is associated with high daily life functioning.
Overall it was hypothesized that computerized cognitive flexibility training increases cognitive functioning in stroke patients, and this subsequently has a positive effect on the extent of experienced fatigue and objective and subjective daily life functioning. To examine the effect of cognitive flexibility training it was first determined whether the training resulted in cognitive improvement. It was predicted that cognitive functioning would increase more in the intervention group than in the active control group. Second it was predicted that cognitive flexibility training would lead to a larger decrease of fatigue compared with those who received a mock training. Finally it was hypothesized that cognitive flexibility training has a positive effect on the patients’ objective and subjective daily life functioning.
Methods Sample characteristics
All subjects in this study were patients who were three months to five years post-stroke and were between 30 and 80 years old. Patients received rehabilitation as inpatient or outpatient; they had cognitive dysfunction, as demonstrated by a neuropsychological assessment or as judged by a neurologist, physiatrist, psychologist, or other experienced clinician and presently still have cognitive complaints. Exclusion criteria were serious physical impairments, through which computer training and testing is not possible or psychiatric illness, such as history of psychotic episode, psychosis, depression and epilepsy. Other exclusion criteria were severe cognitive impairment, resulting from any diseases other than stroke; drug or alcohol dependency; severe color blindness, vision and hearing problems,
and aphasia or neglect. It was also important that participants were able to use a computer, were capable of understanding the instructions, and were able to complete 12 weeks of training.
Patients from several Dutch healthcare facilities were recruited for this study. Patients were selected from patient databases by checking for in- and exclusion criteria. Potentially eligible patients were invited by means of a letter with information about the research. They were asked to return the agreement form attached to the invitation when they met the inclusion criteria and were willing to participate. The study was also advertised in Dutch newsletters and Dutch radio. So far 137 participants indicated that they wanted to participate in the study. A researcher contacted the patient and requested the participant to sign an online provisional informed consent and to fill in an online screening questionnaire, which assessed the participant’s mental capabilities. Subsequently a cognitive screening by phone (Telephone Interview Cognitive Status; TICS; Kempen, Meier, Bouwens, van Deursen, & Verhey, 2007) was administered to screen the participant’s mental capabilities. A score of 26 or lower would exclude the participant from the study. Eventually 84 participants were included. After inclusion, 17 participants refused further participation, mostly due to overburdening or personal circumstances (e.g. illness or losing a family member). Participants who met all requirements were randomly assigned to the cognitive flexibility training (intervention group) or the mock-training (active control group). Only 43 participants concluded the training so far and could be included in the present study.
Procedure
To study the hypotheses, the intervention group and the active control group were both tested in the psychological research lab of the Universiteit van Amsterdam (UvA) prior to training and at the end of the training. When a participant was unable to come to the UvA, the testing was sometimes performed at a location suitable for the participant. At their first visit
participants also received a detailed training instruction and signed an informed consent. During the training period, the participants were contacted seven times by phone, to enquire how the participant was doing, answer any questions, and to motivate the participant. Participants could also email if they had any questions. As compensation for participating in the study, the participants were able to keep training with their Braingymmer account (an online training account), the rest of their lives, after conclusion of the research participation. Anonymity was guaranteed and patients could withdraw from the study at any moment without any consequences. This research was approved by the Medical Ethical Committee of the Vrije Universiteit of Amsterdam.
Material
The active control group in this study can show if the training really has an effect. It shows that differences in effect are not due to, participants’ positive expectations of the training, the personal attention a participant experiences, or spontaneous recovery. Socio-demographic characteristics were recorded at baseline for every participant. Participants were asked to inform about age, gender, and educational level. Educational level was labeled according to the Verhage (1964) classification. Participants were also asked when they had their stroke, to calculate the ‘time since stroke’. Assessments were carried out by master students at two time points: prior to training (pre-test) and at the end of the training (post-test). To determine cognitive functioning, several cognitive domains were assessed, by means of several neuropsychological tasks. At these time points a fatigue questionnaire and a participation questionnaire were also completed by the participant at home via a digital link.
Questionnaires
Fatigue was assessed with the Checklist Individual Strength- Fatigue subscale (CIS-f; Vercoulen, et al., 1994). The fatigue subscale indicates the level of experienced fatigue over the past 2 -week period and contains eight fatigue questions. All items are scored on a
7-points Likertscale, from 1 (No, that’s incorrect) to 7 (Yes that’s correct). There are three reversed items (3, 5, 8) by which the scoring has to be reversed, from 1 (Yes, that’s correct) to 7 (No, that’s incorrect). An example of an item in the fatigue subscale is “I feel tired” or
“I am rested”. A total score is calculated by summing the scores of the items. Higher scores
indicate a higher level of fatigue. A score of 27 or higher indicates abnormal fatigue and a score of 37 or higher is considered indicative for severe fatigue (Bultmann, et al, 2000). The CIS-f is well validated in studies amongst stroke patients (Snaphaan, van der Werf & de Leeuw, 2011). The Cronbach’s alpha reliability for the fatigue subscale is high (.88, Bultmann, et al., 2000).
The Utrecht scale for Evaluation of Rehabilitation-Participation (USER-P; Post, et al.,
2012) will be used to asses objective and subjective daily life functioning. USER-P is a measurement for the effect of outpatient rehabilitation, comprising of 32 items in three scales. The Frequency scale (objective perspective) consists of two parts in which frequency of participation is measured. Part A comprises four items on vocational activities, measuring the number of hours the respondent has spent on paid work, unpaid work, education and housekeeping in a typical week. Each item is scored from 0 (not at all) to 5 (36 hours or more). Part B comprises seven items on the frequency of leisure and social activities such as sports, going out, indoor and outdoor activities, visiting family and social contact via phone or computer in the past four weeks. Each item is scored from 0 (not at all) to 5 (19 times or more). The Restrictions scale (subjective perspective) comprises 11 items on experienced participation restrictions in vocational, leisure and social activities as a result of the person’s health or disability. Each item is scored from 0 (not possible at all) to 3 (without difficulty). A “not applicable” option is available for each item and can be used in case the item is not relevant to the person or if experienced restrictions are not related to the person’s health status or disability. The Satisfaction scale (subjective perspective) comprises ten items on
satisfaction with vocational, leisure and social relationships. Each item is scored on a scale of 0 (very dissatisfied) to 4 (very satisfied). A “not applicable” option is available for the items on vocational activities and partnership relations. The sum scores of the Frequency, Restrictions, and Satisfaction scales are based on the items that are applicable to the person’s situation and each sum score is converted to a 0-100 scale. Higher scores indicate good levels of participation (higher frequency, less restrictions and higher satisfaction). The internal consistency of the USER-P is satisfactory (α 0.70–0.91) and discriminant validity was shown by significant differences in USER-Participation scores between participants with different levels of independence and between participants with different health conditions (Post, et al., 2012). The Intraclass Correlation Coefficient (ICC) of the USER-Participation was low (0.65) for the Frequency scale (0.65) and moderate for the Restrictions scale (0.85) and the Satisfaction scale (0.84) (van der Zee, et al., 2010).
Neuropsychological assessment
The neuropsychological battery consisted of the Digit Symbol Substitution Test (DSST), The Paced Auditory Serial Addition Task (PASAT), the Rey’s Auditory Verbal Learning Test (RAVLT) and the Letter Fluency test. The tests were administered in two fixed orders and every participant followed the same instructions. The participants were randomly assigned to one of the two possible orders, which was the same at pre-test and post-test.
Information processing speed was determined by means of the Digit Symbol Substitution test (DSST) (WAIS III; Wechsler, 1997). This test consists of nine digit-symbol pairs
followed by a list of digits. Under each digit the subject had to write down the corresponding symbol as fast as possible. The number of correct symbols within the allowed time (120 sec) was measured. The reliability of this test is high (.86) and the test-retest reliability is also high (.83; Uterwijk, 2006). The validity is satisfactory (COTAN-beoordeling WAIS-III, 2006).
Attention and working memory were determined by means of the Paced Auditory Serial
Addition Task (PASAT; Gronwall.1977). In this test participants got to hear a number
sequence, the participant had to add the last heard number to the previous heard number and pronounce the answer before they heard the next number. Participants got to hear two separate sequences of 60 numbers, the first version was 3.2 seconds between each number and the second version was 2.8 seconds between each word. The independent measure was the number of correct answers of both sequences. The reliability of this test is high (α = .90; Crawford, Obonsawin & Allan, 1998) and the validity is satisfactory (O’Donnell et. al., 1994; quoted in Mitrushine et. al., 2005)
Memory was determined by means of the Rey’s auditory verbal learning test
(RAVLT; Saan & Deelman, 1986). The participant heard a list of 15 nouns and was asked to recall as many words from the list as possible; this was repeated four more times. After 20 minutes delay, the participant was asked once more to recall as many words as possible. After this “delayed recall” task, a list of 30 words was presented, containing words from the previous list and new words. The participant had to recognize the words from the previous mentioned list. Two different versions were used at the different time points, to prevent learning effects. The independent measure used in this study was the delayed recall T- score corrected for the total score on direct recall, sex, age, and education. This is a reliable and validated memory test (Davis, 2012; Ryan, 1986). The Cronbach’s alpha is .80-.83 and the parallel form reliability is .72-.80 (Bouma, Mulder, Lindeboom & Schmand, 2012).
Executive functioning was determined by means of a Letter Fluency test, the Dutch version of the Controlled Word Association Test (Schmand, Groenink, & Van Den Dungen, 2008). Participants were asked three times to produce as many words as possible in 60 seconds, beginning with a certain letter (K-O-M or P-G-R). There were three rules the participants were not allowed to violate: do not mention any place names or numbers and do
not mention the same preposition directly after one another. The independent measure was the number of correct words per letter. Two different versions were used to prevent learning effects. The test has a high reliability (KOM: r = .82; PGR: r = 0.84; Schmand, et al., 2008).
Training
The participants followed the online training at home, for 12 weeks. Participants trained five times per week for 30-45 minutes via uva.braingymmer.nl, at home on their own computers. The training tasks were designed to be motivating and looked like computer games. After every task feedback was given by means of a three star rating scale. With higher performances, more stars were given.
The intervention group received cognitive flexibility training with nine tasks. These tasks were intended to train three cognitive domains: attention, reasoning and working memory. The cognitive flexibility training provided several tasks within one session and participants were asked to frequently switch between these tasks. Each day they trained on 10 tasks for approximately three minutes each. Tasks were presented directly after each other to assure that cognitive flexibility, i.e. switching from one task to the other, was required. There were 20 levels to play; the difficulty of the level was based on the participants’ performance.
The active control group received a mock training in which only three out of four selected computer tasks per day were trained. These tasks were not likely to enhance cognitive functioning. Each task took approximately 10 minutes. Only nine levels of the tasks were presented to the control group. These levels were challenging enough but not too difficult. The levels of the tasks were adapted according to a predefined schedule. The first six weeks, the participant trained on a higher level each week and after six weeks the participant trained on a higher level every other week.
Statistical analyses
Statistical analyses were performed using the program IBM SPSS statistics 20. A significance level of 0.5 (two-tailed) was adhered. To counteract the problem of multiple comparisons a Bonferroni correction was used were necessary. First it was examined if there were any differences in participants between the intervention group and the active control group at baseline, by means of Chi-square tests and ANOVA.
The relationship between the extent of fatigue, cognitive functioning and daily life functioning at baseline for both groups was examined, by means of multiple Spearman’s Rank Order correlations. It was assumed that there was a monotonic relationship between the variables. First the Spearman’s Rank Order correlations between the CIS-f scores and the neuropsychological assessment tests (DSST, PASAT, RAVLT, and Fluency) at baseline were calculated. Second the correlations between the scores of the USER-P scales (Frequency, Restrictions, and Satisfaction) and the CIS-f were calculated.
Further, several analyses were performed to look at the effect of training on the extent of fatigue, cognitive functioning and daily life functioning. Therefore, pre-test and post-test scores on the selected tasks and questionnaires were compared. The assumptions of multicollinearity, normality, homogeneity of variances and homogeneity of covariances were tenable.
Repeated measures MANOVA was used to determine if cognitive functioning increases due to training, and whether it increased more in the intervention group than in the active control group. The between subject factor was type of training (intervention group or active control group) and the within subject factor was time (pre-test and post-test). The independent variables were the test scores in the neuropsychological assessment (DSST, PASAT, RAVLT, and Letter Fluency). There were some outliers in the data, as assessed by the outlier labeling rule (Hoaglin & Iglewicz, 1987). The analysis was run with and without
the outliers; this did not change the outcome. The outliers were included in the following data, because they did not materially affect the results and because it’s a clinical sample, in which you expect to have deviating participants.
A mixed ANOVA was used to determine if computerized online brain training has an effect on the extent of fatigue and whether cognitive flexibility training has a larger effect on fatigue than the mock training. The between subject factor was type of training (intervention group or active control group) and the within subject factor was time (pre-test and post-test). The independent variable is the total score on the CIS-f. There were no outliers in the data as assessed by the Outlier Labeling Rule (Hoaglin & Iglewicz, 1987).
The effect of training on objective and subjective daily life functioning was determined by means of repeated measures MANOVA. The between subject factor was type of training (intervention group or active control group) and the within subject factor was time (pre-test and post-test). The independent variables were the scores on the Frequency scale, the Restrictions scale and the Satisfaction scale. There were no outliers in the data as assessed by the Outlier Labeling Rule (Hoaglin & Iglewicz, 1987).
Results
Forty three Participants finished the complete training, filled in the questionnaires and had a neuropsychological assessment prior and after the training. Missing values in the neuropsychological assessment are either because participants were not capable to perform the cognitive tests or because participants were too tired to finish the tests at either of the two time points. Missing values in the questionnaires might be due to participants who did not complete the online questionnaires at home or because of technical difficulties.
Patient characteristics
Fifteen women (35%) and 28 men (65%) joined the study. Men and women were equally divided over the two conditions (χ2 = 1.685, df = 1, p = 0.194). The participants were on average 60 years old (M = 60.56 , SD = 8.0, Max = 73, Min = 42). Age differences were equally divided over the conditions (F (1, 42) = 0.038, p = 0.846). The mean number of days since stroke until pre-testing was 909 days (M = 909.37, SD = 470.19, Max = 1996, Min = 144), which is approximately two and a half years. Even though the time since stroke varies between participants, it was equally divided over the conditions (F (1, 42) = 0.198, p = 0.658). Education level was also equally divided over the conditions (χ2 = 0.973, df = 1, p = 0.324). The average number of training sessions performed by the participants was 57 (M = 56.53, SD = 5.02, Max = 62, Min = 30). This was equally divided over the conditions (F (2, 42) = 0.556, p = 0.460). At baseline 13 (31%) participants had normal fatigue, 11 (26,2%) had abnormal fatigue and 18 (42,9%) participants had severe fatigue, as measured by the CIS-f. The extent of fatigue at baseline was equally divided over the groups (χ2 = 4.664, df = 2, p = 0.097).
Table 1.
Baseline Demographic and Clinical Characteristics of the 43 Participants.
Intervention group
n = 20
Active control group
n = 23 Demographic characteristics Sex (n [%]) Male 11 (55) 17(73.9) Female 9 (45) 6 (26.1) Age in years M (SD) 60.3 (8.7) 60.8 (1.6) Educational status (n [%]) LBO 1 (5) 1 (4.3) Mavo/MBO 4 (20) 8 (34.8) Havo/VWO/HBO 15 (75) 10 (43.5) University 0 (0) 4 (17.4) Clinical characteristics
Time from stroke onset to pre-test in days M (SD) 944 (537.1) 879.3 (413.4) Number of training sessions M (SD) 57.2 (3) 56 (6.3) Fatigue severity (n [%])
Normal 3 (15) 10 (43.5)
Abnormal 6 (30) 5 (21.7)
Severe 11 (55) 7 (30.4)
Missing 0.0 (0.0) 1 (2.3)
Relationship fatigue, cognitive functioning and daily life functioning
Analysis revealed a significant positive correlation between the DSST scores and the CIS-f scores (rs = 0.473, p = 0.002). Furthermore, PASAT scores (rs = 0.182, p = 0.262), RAVLT scores (rs = -0.95, p = 0.550), and Fluency scores (rs = 0.257, p = 0.100) did not significantly correlate with CIS-f scores. Further analysis revealed a significant negative correlation between CIS-f scores and the USER-P Restrictions scale (rs = -0.314, p = 0.046) and Satisfaction scale (rs = -0.325, p = 0.036). No significant correlation was found between CIS-f scores and the Frequency scale (rs = -0.021, p = 0.897).
Training effect on cognitive functioning
Repeated-measures MANOVA analyses found no multivariate effect for type of training (V = 0.090, F (4, 34) = 0.845, p = 0.507). Univariate between-group analyses showed that DSST scores (F (1, 37) = 1.904, p = 0.176, η2 = 0.049), RAVLT scores (F (1, 37) = 0.007, p = 0.935, η2 = 0.000), PASAT scores (F (1, 37) = 0.676, p = 0.416, η2 = 0.018), and Letter Fluency scores (F (1, 37) = 0.000, p = 0.989, η2 = 0.000) did not significantly differ between the intervention group and the active control group. This means that there was, according to the four dependent variables at pre-test and post-test, no difference in cognitive functioning between participants who did the mock training and participants who did the cognitive flexibility training.
The analysis did confirm that there was a significant multivariate effect for time (V = .437,
F (4, 34) = 6.587, p < 0.005). Within-group univariate analyses indicated that DSST scores (F
(1, 37) = 9.166, p = 0.004, η2 = 0.199), PASAT scores (F (1, 37) = 9.517, p = 0.004, η2 = 0.205), and Letter Fluency scores (F (1, 37) = 4.729, p = 0.036, η2 = 0.113) were significantly improved between pre-test and post-test. The RAVLT scores (F (1, 37) = 3.833,
p = 0.058, η2 = 0.094) did not significantly differ between pre-test and post-test. This means
that, as assessed by the four dependent variables at pre-test and post-test, processing speed, attention, working memory and executive functioning improved over time, and memory did not improve in both the intervention group and the active control group. The mean scores and standard deviations of the dependent variables, for both the intervention group and the active control group at pre-test and post-test, are registered in Table 3.
Table 3.
DSST, RAVLT, Fluency and PASAT Scores Prior to Training and After Training.
Intervention group
n = 19
Active control group
n = 20 DSST a M (SD) Pre-test 62 (12.8) 54.4 (16.4) Post-test 63.9 (14.2)* 58.3 (17.2)* PASATb M (SD) Pre-test 75.8 (15.03) 69.6 (19) Post-test 79.5 (17.9)* 77.3 (16.8)* RAVLT c M (SD) Pre-test 49.6 (12.8) 53.2 (10.8) Post-test 49.1 (11.4) 46 (11.1) Fluency d M (SD) Pre-test 33.7 (13.6) 33.2 (13.9) Post-test 34.8 (13.1)* 35.6 (14.6)* a
Digit Symbol Substitution test,bPaced Auditory Serial Addition Task,c Rey’s auditory verbal learning test, d Letter Fluency.
* Significant difference with pre-test with α < .05
Training effect on fatigue
The mixed ANOVA main effect of group showed that there was no statistically significant difference of fatigue between the intervention group and the active control group, F (1, 39) = 3.137, p = 0.084, η2 = 0.074. The main effect of time showed no statistically significant difference in fatigue at the different time points either, F (1, 39) = 4.076, p = 0.050, η2 = 0.095. This means that computerized online brain training has no effect on the extent of fatigue, and cognitive flexibility training does not have a larger effect on fatigue than the mock training. The mean scores and standard deviations of the dependent variable, for both the intervention group and the active control group at pre-test and post-test, are registered in Table 4.
Table 4.
CIS-f Scores Prior to Training and After Training.
Intervention group
n = 20
Active control group
n = 22
CIS-f M (SD)
Pre-test 38.4 (12.9) 29.8 (12.1)
Post-test 34.4 (13) 29.1 (13.9)
Note. CIS-f = Checklist Individual Strength- Fatigue subscale
Training effect on objective and subjective daily life functioning
Repeated-measures MANOVA analyses found no multivariate effect for type of training (V = 0.017, F (3, 38) = 0.221, p = 0.881) or time (V = 0.144, F (3, 38) = 2.134, p = 0.112). Univariate between-group analyses showed that the Frequency scale (F (1, 40) = 0.060, p = 0.807, η2 = 0.002), the Restrictions scale (F (1, 40) = 0.027, p = 0.870, η2 = 0.001) and the Satisfaction scale (F (1, 40) = 0.296, p = 0.589, η2 = 0.007) did not significantly differ between the intervention group and the active control group. Within-group univariate analyses indicated that Frequency scale (F (1, 40) = 1.659, p = 0.205, η2 = 0.040), the Restrictions scale (F (1, 40) = 2.815, p = 0.101, η2 = 0.066) and the Satisfaction scale (F (1, 40) = 0.203, p = 0.654, η2 = 0.005) did not significantly differ between pre-test and post-test. This means that computerized online brain training has no effect on objective and subjective daily life functioning, measured with the USER-P. The mean scores and standard deviations of the dependent variables, for both the intervention group and the active control group at pre-test and post-test, are registered in Table 5.
Table 5.
USER-P Frequency scale, Restrictions scale and Satisfaction scale Prior to Training and After Training.
Intervention group
n = 20
Active control group
n = 22 Frequency scale M (SD) Pre-test 34.3 (7.9) 34.4 (10.9) Post-test 35.5 (7.3) 36.6 (10.5) Restrictions scale M (SD) Pre-test 75.1 (14.8) 71.2 (17.1) Post-test 75.4 (13.5) 77.9 (13.0) Satisfaction scale M (SD) Pre-test 59.8 (15.6) 62.2 (16.3) Post-test 58.6 (14.7) 61.6 (20.8)
Note. USER-P = the Utrecht Scale for Evaluation of Rehabilitation-Participation.
Discussion
The main goal of the present study was to examine the influence of computerized cognitive flexibility training on the extent of PSF, cognitive functioning and the influence on daily life functioning. Before examining the effect of computerized cognitive flexibility training, the relationship between PSF, cognitive impairment and daily life functioning was explored.
It was hypothesized that cognitive functioning negatively correlates with the extent of experienced fatigue and with information processing speed in particular. It was also expected that daily life functioning negatively correlates with fatigue. The results indicate that, opposing the Coping hypotheses, there was no relationship between the cognitive functions; attention, working memory, memory, and executive functioning and the extent of experienced fatigue. Contrary to the expectations, a positive relationship was found between information processing speed and fatigue. The present study did indicate a relationship between the extent of fatigue participants experienced and subjective daily life functioning. Objective daily life functioning did not correlate with the extent of fatigue.
The positive relationship between information processing speed and the extent of fatigue found in this study is a remarkable finding, considering previous findings in which fatigue was associated with slow information processing speed (Pihlaja et al., (2014). It might be possible that Staub and Bogousslavsky’s (2001) argument is applicable. They argue that the less impaired patients experience greater demands of normal life and therefore experience a greater extent of fatigue. We would however expect to find this effect in other cognitive domains as well. Another explanation could be that this finding was coincidental in view of the small sample size. Overall, the results did indicate that the degree of fatigue does not negatively correlate with cognitive functioning. It might be possible that PSF is not necessarily a result of cognitive impairment. For instance, Christensen et al. (2008) found that few stroke patients, being free of fatigue three months post-stroke, subsequently developed fatigue. A study by van der Werf et al. (2001) found that extreme fatigue also occurred in ‘neuropsychological problem-free’ stroke patients. These findings suggest that fatigue is a direct consequence of stroke rather than a consequence of cognitive impairment. However, as mentioned earlier in this study, some studies did find a relationship between cognitive impairment and experienced fatigue. It might therefore also be possible that with use of another fatigue instrument, we would have found different results. For instance, a pilot study by Hubacher et al. (2012) showed that, in the same sample, some fatigue instruments correlated with cognitive measures and others did not. Nevertheless, one can admit that it is most likely that neuropsychological deficits are associated with mental fatigue and vice versa, only in the present study it appears that cognitive impairment is not required for PSF.
Further this study indicated a relationship between the extent of fatigue participants experienced and subjective daily life functioning. However, objective daily life functioning did not correlate with the extent of fatigue. It seems that more fatigued participants did experience restrictions and dissatisfaction in their daily lives, however higher fatigue did not
correlate with less activity. A study by Van Der Zee, Visser-Meily, Lindeman, Jaap Kappelle, & Post (2013) found that objective and subjective daily life functioning were both determined by physical and cognitive independence but subjective daily life functioning was also determined by fatigue and mood. This might explain the absence of correlation between objective daily life functioning and experienced fatigue in the present study. Taken together, it appears that in the present study the extent of fatigue was not related to cognitive impairment. Fatigue does however seem to be related to subjective daily life functioning.
Nevertheless, it was hypothesized that computerized cognitive flexibility training increases cognitive functioning in stroke patients, and this subsequently has a positive effect on the extent of experienced fatigue and objective and subjective daily life functioning. To examine the effect of cognitive flexibility training it was first determined whether the training resulted in cognitive improvement. It was predicted that cognitive functioning would increase more in the intervention group than in the active control group. The results indicate that the effect of cognitive flexibility training on cognitive functioning in stroke patients did not differ from the effect of mock training, in which cognitive flexibility is not trained. Both kinds of training had an effect on the measure for information processing speed, the measure for attention and working memory, and the measure for executive functioning. A measure for memory did not significantly change as a result of computerized online brain training. Second it was predicted that cognitive flexibility training would lead to a larger decrease of fatigue compared with those who received a mock training. In contrast to the expectation, both cognitive flexibility training and mock training had no significant effect on the extent of fatigue. Even though some cognitive functions improved over time, the extent of fatigue did not significantly change. This corresponds with the previous finding that cognitive functioning does not negatively correlate with fatigue. Finally it was hypothesized that cognitive flexibility training has a positive effect on the patients’ objective and subjective daily life functioning.
This study did not find an effect of computerized online brain training on objective and subjective daily life functioning. Meaning that the improvement in performance on the neuropsychological assessment over time did not have a positive effect on the patients’ daily life functioning.
Because the same effects on cognitive functioning have been found in both training types it is uncertain if cognitive functioning really improved due to computerized online brain training. A possible explanation for the increase in cognitive functioning in both the training types might be that the mock training had more positive influence on cognitive functioning than expected. It is also possible that both training types actually did not have any effect on cognitive functioning and that the improvement is a consequence of spontaneous recovery. Future research should therefore add a placebo group in which participants do not get any training at all, by doing so cognitive improvement as a consequence of time can be examined. A more plausible explanation is that we found higher scores on both training groups due to practice effects of the neuropsychological assessment. Multiple studies have found practice effects for the DSST, even when alternative forms were used (Eileen De Monte, Malke Geffen, Kwapil, 2014; Hinton-Bayre & Geffen, 2005). A review of Tombaugh (2006) showed that the PASAT is extremely sensitive for practice effects. According to this research a possible explanation for these practice effects is that due to the complexity of the test, participants require time and practice to develop an effective strategy. A subsequent administration would therefore result in higher scores. A second factor that might have an influence on the test scores is the heightened emotional state that the PASAT induces. Some participants experience anxiety and frustration during the first administration. With repeated exposure participants are less emotional and better able to concentrate during the test. For the Letter Fluency and the RAVLT, two different versions were used to prevent learning effects. For the RAVLT different words had to be remembered at pre-test and post-test. Therefore, it
is unlikely that any practice effects occurred. For the Letter Fluency participants had to name as many words as possible but with different letters at pre-test and post-test. It might be possible that the time pressure in the Letter Fluency test had an effect on the emotional state of the participants as well. A second administration might therefore be less stressful and result in higher scores.
Fatigue did not change over time in both training groups. It seems that computerized brain training had no effect on the extent of experienced fatigue. A possible explanation might be that daily training was so energy consuming that the change in extent of fatigue over time was overruled by fatigue inflicted by the training. Another, more plausible, explanation is that in the present study experienced fatigue was not related to cognitive impairment and therefore (the questionable) cognitive improvement did not influence fatigue. It is important to take into account that it was not predetermined in this study whether a participant had PSF. Participants with normal, abnormal and severe fatigue, were included. Furthrmore it was not clear if any of the participants had fatigue complaints pre-stroke or what caused the fatigue complaints at baseline. Future research should therefore only research stroke patients with PSF at baseline.
The participants did not experience any training benefits in objective and subjective daily life functioning. It appears that the improvement in performance on the neuropsychological assessment did not have a positive effect on participants’ objective and subjective daily life functioning. This corresponds with the previous finding that subjective daily life functioning is associated with the extent of fatigue. As mentioned before, fatigue might have an impact on energy consuming activities of daily living, such as leisure situations and social contact (Lerdal et al, 2009). It might also be possible that training was so time consuming that participants did not have the opportunity to expand their activities of daily life.
Another factor to take into account is the variety between the participants in this study. In this study the time since stroke, the severity of cognitive impairment and the stroke location varied quite a lot between participants. A study of Peretz et al. (2011) revealed that cognitive computer training had a greater effect on participants with lower baseline cognitive functioning than participants with high baseline cognitive functioning. Bentley et al. (2014) suggest that clinical outcome after stroke is also associated with lesion location at baseline. Even the extent of fatigue might be associated with lesion location. Hubacher et al. (2012) found that patients with subcortical lesions showed higher scores on motor fatigue, while cognitive fatigue was more frequent in patients with cortical lesions.
Taken together, it appears that in the present study the extent of fatigue was not related to cognitive impairment. It was however associated with subjective daily life functioning. Perhaps fatigue really is a primary consequence of stroke and rehabilitation should indeed focus on teaching stroke patients compensatory strategies to compensate for their limitations. Further examination of the etiology of PSF and the relationship with cognitive impairment might shed more light on how to treat this underestimated consequence of stroke. The computerized online brain training did not have any effect on the extent of fatigue or daily life functioning. Unfortunately it is still unclear whether computerized online brain training is an effective intervention for cognitive impairment. Future studies should aim to include a placebo group and enquire patients with established PSF in a more homogenous sample. Computerized cognitive brain training might become an efficient and profitable rehabilitation program in the future, but the exact form still has to be studied thoroughly. Most of all it is important that rehabilitation centers become more aware of PSF and its negative effect on the patients’ progress.
References
Andersen, G., Christensen, D., Kirkevold, M., & Johnsen, S. P. (2012). Post-stroke fatigue and return to work: a 2-year follow-up. Acta Neurologica
Scandinavica, 125(4), 248-253.
Andreasen, A. K., Spliid, P. E., Andersen, H., & Jakobsen, J. (2010). Fatigue and processing speed are related in multiple sclerosis. European Journal of
Neurology, 17(2), 212-218.
Bendz, M. (2003). The first year of rehabilitation after a stroke – from two perspectives. Scandinavian Journal of Caring Sciences, 17(3), 215-222. Bentley, P., Kumar, G., Rinne, P., Buddha, S., Kallingal, J., Hookway, C., ... &
Beckmann, C. (2014). Lesion locations influencing baseline severity and early recovery in ischaemic stroke. European Journal of Neurology.
Björkdahl, A., Arkerlund, E., Svensson, S., Esbjörnsson, E. (2013). A randomized study of computerized working memory training and effects on functioning in everyday life for patients with brain injury. Brain injury, 27(13-14), 1658-1665. Bouma, A., Mulder, J., Lindeboom, J., Schmand, B. (2012). Handboek neuropsychologische
diagnostiek. (pp. 276-282). Amsterdam: Pearson Assessment and Information
Buitenweg, J.I.V., Murre, J.M.J., & Ridderinkhof, K.R. (2012). Brain training in progress: A review of trainability in healthy seniors. Frontiers in Human Neuroscience, 6, 183. Brown, L.F.,Kroenke, K., Theobald, D.E., & Wu, J.W. (2011). Comparison of SF-
36 vitality scale and Fatigue Symptom Inventory in assessing cancer-related fatigue.
Supportive Care In Cancer, 19(8), 1255-1259.
Bultmann U., de Vries M., Beurskens A.J., Bleijenberg G., Vercoulen J.H., & Kant I. (2000). Measurement of prolonged fatigue in the working population: Determination of a
cutoff point for the checklist individual strength. Journal of occupational health
psychology, 5(4),411-416.
Carlsson, G.E., Moller, A., & Blomstrand, C. (2003). Consequences of mild stroke in persons <75 years-a 1-year follow-up. Cerebrovasculair Diseases , 16(4), 383-388.
Cha, Y. J., & Kim, H. (2013). Effect of computer-based cognitive rehabilitation (CBCR) for people with stroke: A systematic review and meta-analysis.
NeuroRehabilitation, 32(2), 359-368.
Choi-Kwon, S., Han, S.W., Kwon, S.U., & Kim, J.S. (2005). Poststroke fatigue: Characteristics and related factors. Cerebrovasculair Diseases, 19(2), 84-90. Christensen, D., Johnsen, S.P., Watt, T., Harder, I., Kirkevold, M., & Andersen, G. (2008).
Dimensions of post-stroke fatigue: A two-year follow up study. Cerebrovascular
Diseases, 26(2), 134-141.
Claesson, L., Lindén, T., Skoog, I., & Blomstrand, C. (2005). Cognitive impairment after stroke–Impact on activities of daily living and costs of care for elderly people. Cerebrovascular Diseases, 19(2), 102-109.
COTAN-documentatie (2005). COTAN-beoordeling WAIS-III Retrieved November 27, 2014 from http://www.cotandocumentatie.nl.proxy.uba.uva.nl:2048/ test_details.php?id=384.
Crawford, J.R., Obonsawin, M.C., & Allan, K.M. (1998). PASAT and components
of WAIS-R performance: Convergent and discriminant validity. Neuropsychological
Rehabilitation, 8, 255–272.
Crosby, G.A., Munshi, S., Karat, A.S., Worthington, E., & Lincoln, N.B. (2012). Fatigue after stroke: Frequency and effect on daily life. Disability & Rehabilitation, 34(8), 633-637.
Culbertson, W., & Zillmer, E. (2005), Tower of London Drexel University (TOL DX). Chicago: IL: Multi-Health Systems.
Dahm, J., Wong, D. & Ponsford, J. (2013). Validity of the depression anxiety stress
scales in assessing depression and anxiety following traumatic brain injury. Journal of
Affective disorders, 151(1), 392-396.
Davis, J.J., Millis, S.R., Axelrod, B.N. (2012). Derivation of an embedded Rey Auditory Verbal Learning Test performance validity indicator. The clinical
neuropsychologist, 26(8), 1397-1408.
DeLuca, J. (2005). Fatigue as a window to the brain (pp. 37-58). Cambridge: MIT press.
Eileen De Monte, V., Malke Geffen, G., & Kwapil, K. (2005). Test-retest reliability and practice effects of a rapid screen of mild traumatic brain injury. Journal of clinical
and experimental neuropsychology, 27(5), 624-632.
Glader, E., Stegmayr, B., & Asplund, K.M. (2002). Poststroke fatigue: A 2-year follow-up study of stroke patients in Sweden. Journal of the American Heart Association, 33(5), 1327-1333.
Gronwall, D.M.A. (1977). Paced auditory serial-addition task: A measure of recovery from concussion. Perceptual and Motor Skills, 44, 367-373. Hersenstichting (2014). “Cijfers Hersenstichting.nl”. Retrieved from
https://www.hersenstichting.nl/alles-over-hersenen/hersenaandoeningen/ beroerte. Hinton-Bayre, A., & Geffen, G. (2005). Comparability, reliability, and practice effects on
alternate forms of the Digit Symbol Substitution and Symbol Digit Modalities tests. Psychological assessment, 17(2), 237.
Hoaglin, D.C., & Iglewicz, B. (1987). Fine-tuning some resistant rules for outlier labeling.
Hubacher, M., Calabrese, P., Basetti, C., Carota, A., Stöcklin, M., & Penner, I.-K. (2012). Assessment of post-stroke fatigue: The fatigue scale for motor and cognitive functions. European Neurology, 67(6), 377-384.
Ingles, J.L., Eskes , G.A., & Philips, S.J. (1999). Fatigue after stroke. Archives of Physical
Medicine and Rehabilitation, 80(2), 173-178.
Jacquin, A., Binquet, C., Rouaud, O., Graule-Petot, A., Daubail, B., Osseby, G. V., & Béjot, Y. (2014). Post-stroke cognitive impairment: high prevalence and determining factors in a cohort of mild stroke. Journal of Alzheimer's Disease, 40 1029–1038
Johansson, B., Berglund, P., & Rönnbäck, L. (2009). Mental fatigue and impaired information processing after mild and moderate traumatic brain injury. Brain Injury, 23(13-14), 1027-1040.
Karbach, J., & Kray, J. (2009). How useful is executive control training? Age differences in near and far transfer of task-switching training. Developmental science, 12(6), 978-990.
Kempen, G. I. J. M., Meier, A. J. L., Bouwens, S. F. M., van Deursen, J., & Verhey, F. R. J. (2007). Telefonisch Interview Cognitieve Status (TICS): psychometrische
aspecten. Tijdschrift voor Gerontologie en Geriatrie, 2007,38(1), 34-40. Klingberg, Forssberg & Westerberg (2002). Training of working memory in children
with ADHD. Journal of Clinical and Experimental Neuropsychology, 24(6), 781-791. Kueider, A. M., Parisi, J. M., Gross, A. L., & Rebok, G. W. (2012). Computerized
cognitive training with older adults: a systematic review. PloS one, 7(7), e40588. Lerdal, A., Bakken, L.N., Kouwenhoven, S.E., Pedersen, G., Kirkevold, M., Finset, A.
Kim, H.S. (2009). Poststroke fatigue – A review. Journal of pain and symptoms
Luteijn, F., & Barelds, D.P.H. (2004). Herziening van de GIT. Handleiding bij de GIT-2. Amsterdam: Harcourt.
Mitrushine, M., Boone, K. B., Razani, J. & D’Elia, L. F. (2005). Handbook of Data for Neuropsychological Assessment. New York, Oxford University Press.
Owen, A. M., Hampshire, A., Grahn, J. A., Stenton, R., Dajani, S., Burns, A. S., & Ballard, C. G. (2010). Putting brain training to the test. Nature, 465(7299), 775-778.
Pasquini, M., Leys, D., Rousseaux, M., Pasquier, F., & Henon, H. (2007). Influence
of cognitive impairment on the institutionalisation rate 3 years after a stroke. Journal
of Neurology, Neurosurgery & Psychiatry, 78(1), 56-59.
Peretz, C., Korczyn, A. D., Shatil, E., Aharonson, V., Birnboim, S., & Giladi, N.
(2011). Computer-based, personalized cognitive training versus classical computer games: A randomized double-blind prospective trial of cognitive stimulation.
Neuroepidemiology, 36(2), 91-99.
Pihlaja, R., Uimonen, J., Mustanoja, S., Tatlisumak, T., & Poutiainen, E. (2014). Post-stroke fatigue is associated with impaired processing speed and memory functions in first-ever stroke patients. Journal of psychosomatic research, 77(5), 380-384.
Post, M.W., van der Zee, C.H., Hennink, J., Schafrat, C.G., Visser-Meily, J.M., van
Berlekom, S.B., (2012). Validity of the Utrecht scale for evaluation of rehabilitation-participation. Dishability and Rehabilitation, 34(6), 478-485.
Ryan, J., Geisser, M., Randall, D., (1986). Alternate form reliability and equivalency of the rey auditory verbal learning test. Journal of clinical and experimental
Saan, R.J., & Deelman B.G. (1986). De 15-Woorden Tests A en B. (Een voorlopige handleiding). Goningen: afd. Neuropsychologie,AZG.
Schaapsmeerders, P., Maaijwee, N. A., van Dijk, E. J., Rutten-Jacobs, L. C., Arntz, R. M., Schoonderwaldt, H. C., & de Leeuw, F. E. (2013). Long-term cognitive impairment after first-ever ischemic stroke in young adults. Stroke,44(6), 1621-1628.
Schepers, V.P., Visser-Meily, A.M., Ketelaar, M., & Lindeman, E. (2006). Poststroke fatigue: Course and it’s relation to personal and stroke-related factors. Archives of
Physical Medicine and Rehabilitation, 87(2), 184-188.
Schmand, B., Groenink, S.C., & van den Dungen, M. (2008). Letterfluency: psychometrische eigenschappen en Nederlandse normen. Tijdschrift voor
Gerontologische Geriatrie, 39, 64-76.
Snaphaan, L., van der Werf, S., & de Leeuw, F.-E. (2011). Time course and risk factors of post-stroke fatigue: a prospective cohort study. European Journal of Neurology, 18(4), 611-617.
Staub, F., & Bogousslavsky, J. (2001). Fatigue after stroke: A major but neglected issue.
Cerebrovascular Diseases, 12(2), 75-81.
Tombaugh, T. N. (2006). A comprehensive review of the paced auditory serial addition test (PASAT). Archives of clinical neuropsychology, 21(1), 53-76.
Uterwijk, J.M. (red.)(2005). WAIS III NL Handleiding, 2005. Amsterdam: Pearson. van der Linden, D., Frese, M., & Meijman, T.F. (2003). Mental Fatigue and the control of
cognitive processes: Effects on perseveration and planning. Acta Psychologica, 113(1), 45-65.
van der Werf, S.P., van den Broek, H.L., & Anten, H.W. (2001). Experience of severe fatigue long after stroke and it’s relation to depressive symptoms and disease characteristics.
van der Zee, C. H., Priesterbach, A. R., van der Dussen, L., Kap, A., Schepers, V. P., Visser-Meily, J., & Post, M. W. (2010). Reproducibility of three self-report participation measures: The ICF Measure of Participation and Activities Screener, the Participation Scale, and the Utrecht Scale for Evaluation of Rehabilitation-Participation. Journal of
Rehabilitation Medicine, 42(8), 752-757.
Van Der Zee, C. H., Visser-Meily, J. M., Lindeman, E., Jaap Kappelle, L., & Post, M. W. (2013). Participation in the chronic phase of stroke. Topics in stroke
rehabilitation, 20(1), 52-61.
van Zandvoort, M.J., Kappelle, L.J., Algra, A., & de Haan, E.H. (1998). Decreased capacity for mental effort after single supratentorial lacunar infarct may affect performance in everyday life. Journal of Neurology, Neurosurgery and Psychiatry, 65(5), 697-702. Vercoulen, J. H., Swanink, C., Fennis, J. F., Galama, J., van der Meer, J. W., &
Bleijenberg, G. (1994). Dimensional assessment of chronic fatigue syndrome. Journal
of psychosomatic research, 38(5), 383-392.
Verhage, F. (1964). Intelligentie en leeftijd: onderzoek bij Nederlanders van twaalf tot
zevenenzeventig jaar. Assen: Van Gorcum.
Wechsler, D. (1997). Wechsler Adult Intellegence Scale – Third Edition:
Administation and scoring manual. San Antonio, TX: Psychological Corporation.
Zedlitz, A.M., Rietveld, T.C., Geurts, A.C., & Fasotti, L. (2012). Cognitive and graded activity training can alleviate persistent fatigue after stroke: A randomized, controlled trial. Stroke, 43(4), 1046-1051.
Zigmond, A.S., & Snaith, R.P. (1983). The hospital anxiety and depression scale.
