soleil de Lunettes magnésium UV de Noir polarisants soleil Hommes Lunettes UV Anti Mengonee volant de protection aluminium 400 d'extérieur Lunettes au wqSpXnIBx | El Corredor Del Laberinto La C.. HDTV-Screener | Сексология. Взаимоотношения полов

Effectiveness of cognitive behavioral therapy for the treatment of fatigue in patients with multiple sclerosis: A systematic review and meta-analysis

Effectiveness of cognitive behavioral therapy for the treatment of fatigue in patients with multiple sclerosis: A systematic review and meta-analysis

Journal of Psychosomatic Research 90 (2016) 33–42 Contents lists available at ScienceDirect Journal of Psychosomatic Research Review article Effec...

590KB Sizes 0 Downloads 0 Views

Recommend Documents

Multiple sclerosis in Brazil: A systematic review
•Profile of MS in Brazil is similar to that observed in high-MS-prevalence areas.•Classic predictors of long-term disabi

Efficacy and acceptability of group cognitive behavioral therapy for depression: A systematic review and meta-analysis
Despite treatment guidelines for depression placing group cognitive behavioral therapy (group CBT) between low- and high

Predictors of Treatment Outcome in Contextual Cognitive and Behavioral Therapies for Chronic Pain: A Systematic Review
•Predictors of outcomes in contextual cognitive-behavioral therapy appear unclear.•Methodological limitations and a lack

Integrity of hypothalamic fibers and cognitive fatigue in multiple sclerosis
•The posterior hypothalamus plays a key role in the regulation of wakefulness.•We analyzed the relation between integrit

Journal of Psychosomatic Research 90 (2016) 33–42

Contents lists available at ScienceDirect

Journal of Psychosomatic Research

Review article

Effectiveness of cognitive behavioral therapy for the treatment of fatigue in patients with multiple sclerosis: A systematic review and meta-analysis Lizanne Eva van den Akker MSc a,b,c,⁎, Heleen Beckerman PhD a,b,c, Emma Hubertine Collette PhD d, Isaline Catharine Josephine Maria Eijssen PhD a,b,c, Joost DekkerProf a,b,e, Vincent de GrootProf a,b,c a

Department of Rehabilitation Medicine, VU University Medical Center, Amsterdam, The Netherlands EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands c MS Center Amsterdam, Amsterdam, The Netherlands d Department of Medical Psychology, VU University Medical Center, Amsterdam, The Netherlands e Department of Psychiatry, VU University Medical Center, Amsterdam, The Netherlands b

a r t i c l e

i n f o

Article history: Received 26 May 2016 Received in revised form 2 September 2016 Accepted 4 September 2016 Available online xxxx Keywords: Cognitive behavioral therapy Multiple sclerosis Fatigue Systematic review and meta-analysis

a b s t r a c t Background: Fatigue is a frequently occurring symptom of multiple sclerosis (MS) that limits social participation. Objective: To systematically determine the short and long-term effects of cognitive behavioral therapy (CBT) for the treatment of MS-related fatigue. Data sources: Pubmed, Cochrane, EMBASE, Psychology and Behavioral Sciences Collection, ERIC, PsychINFO, Cinahl, PsycARTICLES, and relevant trial registers were searched up to February 2016. In addition, references from retrieved articles were examined. Study selection: Studies were included if participants had MS, fatigue was a primary outcome measure, the intervention was CBT, and the design was a randomized controlled trial. The search was performed by two independent reviewers, three CBT experts determined whether interventions were CBT. Data extraction: Data on patient and study characteristics and fatigue were systematically extracted using a standardized data extraction form. Two independent reviewers assessed risk of bias using the Cochrane Collaboration risk of bias tool. In the event of disagreement, a third reviewer was consulted. Data synthesis: Of the 994 identified studies, 4 studies were included in the meta-analysis, comprising 193 CBTtreated patients and 210 patients who underwent a control treatment. Meta-analyses of these studies showed that CBT treatment had a positive short-term effect on fatigue (standardized mean difference [SMD] = −0.47; 95% confidence interval [CI] = −0.88; −0.06; I2 = 73%). In addition, three studies showed a long-term positive effect of CBT (SMD = −0.30; CI −0.51; −0.08; I2 = 0%). Conclusions: This review found that the use of CBT for the treatment of fatigue in patients with MS has a moderately positive short-term effect. However, this effect decreases with cessation of treatment. © 2016 Elsevier Inc. All rights reserved.

Multiple sclerosis (MS) is a chronic neurodegenerative disease of the central nervous system that damages myelin and axons [1]. This damage can result in a wide range of symptoms such as muscle weakness, poor balance, pain, cognitive impairment, depression and fatigue [1]. MS-related fatigue is a highly prevalent symptom in the early stage of the disease [1–3], with almost 80% of patients experiencing it in the first year of disease onset [4]. It can result in significant socioeconomic costs, including loss of work hours and in some instances loss of employment [5–7]. MS-related fatigue is often described as being more severe and more disabling than fatigue in healthy controls ⁎ Corresponding author at: Department of Rehabilitation Medicine, VU University Medical Center, PO Box 7057, 1007, MB, Amsterdam, The Netherlands. E-mail address: [email protected] (L.E. van den Akker).

http://dx.doi.org/10.1016/j.jpsychores.2016.09.002 0022-3999/© 2016 Elsevier Inc. All rights reserved.

[2,3,8], and although common in MS, it is rather difficult to define. Frequently-used descriptions include a sense of exhaustion, lack of energy or tiredness and a subjective lack of physical and/or mental energy [9]. In recent years, researchers have become increasingly interested in non-pharmacological approaches to treat neurological symptoms, including MS-related fatigue [5,10,11]. To gain more insight into MSrelated fatigue, van Kessel et al. [12] developed a cognitive-behavioral model that proposes that primary disease factors trigger the initial fatigue in MS. The extent to which fatigue influences daily life depends on cognitive, emotional, behavioral and physiological factors, which may perpetuate or even worsen the fatigue [12]. Insights derived from this model provided the basis for the development of cognitive behavioral therapy (CBT) for the treatment of MS-related fatigue [13].

34

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

‘In general CBT aims to influence cognitions and behaviors that are related to the experienced problems [14]. CBT has changed in time; three CBT generations can be distinguished. The first generation, behavioral therapy, was aimed at scientifically well-established principles, emphasizing the importance of measurable behavioral changes. No attention was paid to internal feelings whatsoever. Patients learned to change their behavior, by experiencing the consequences of these behaviors [15,16]. In the second generation of therapies more attention was paid to internal processes, thoughts and feelings, that are now dealt with in a more direct way. Central in the second generation is the modification of dysfunctional beliefs and faulty information processing [17]. Examine and reconsider distorted thinking patterns is reached with strategies like cognitive therapy [18], and rational emotive behavior therapy [19]. For the third generation multiple strategies are combined, like mindfulness exercises (e.g. mindfulness of senses mediation, body scan), acceptance of unwanted thoughts, feelings, and internal sensations and/or cognitive defusion (stepping back and seeing thoughts as just thoughts) to elicit change in the thinking process in order to reduce the effects of symptoms [20]. To date, no study has systematically reviewed the effectiveness of CBT in reducing MS-related fatigue. Therefore, we performed a systematic review to determine the effectiveness of CBT in the treatment of MS-related fatigue.

CBT is a therapy that can consist of multiple strategies and components, it was sometimes difficult to ascertain whether an intervention was CBT. The interventions that were classified as ambiguous by the reviewers were checked by three clinicians experienced in CBT. These experts, blinded for author, study and journal details, were asked to categorize the interventions based on their clinical judgment into three categories: 1) solely CBT, 2) containing CBT-like components, 3) not CBT. A discussion was organized to reach consensus. The articles with CBT-like components were included in a secondary analysis.

1. Methods

1.5. Methodological quality assessment

1.1. Search methods

Two reviewers (LA and HB) judged the quality of evidence using the Cochrane Collaboration ‘risk of bias’ tool [26]. This tool addresses selection bias, performance bias, detection bias, attrition bias, reporting bias and other bias. Judgments are categorized as ‘low risk’ of bias, ‘high risk’ of bias or ‘unclear risk’ of bias [26]. Consensus was reached by discussion; when disagreements persisted a third reviewer was consulted.

This meta-analysis used the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines for systematic reviews and meta-analyses [21]. The literature search concluded on February 8, 2016. The following databases were searched: Pubmed; Cochrane, EMBASE, Psychology and Behavioral Sciences Collection, ERIC, PsychINFO, Cinahl and PsycARTICLES. In addition, trial registers were searched to identify published and unpublished trials on the topic and, when necessary corresponding authors were approached (Appendix A). All reference lists of eligible articles were searched for further relevant articles. The complete study protocol is available as Appendix C: Supplementary material. To develop an appropriate search strategy, we collected keywords for diagnosis and fatigue by combining two search strategies developed for related reviews [22,23]. CBT-related keywords were mainly based on the recent Cochrane review of Hunot et al. [23]. The complete search strategy can be found in Appendix B. We aimed for a sensitive search strategy, with broad keywords that did not exclude interventions that used components of CBT. Subsequently, non-CBT studies are filtered out by the inclusion criteria and the study selection process. 1.2. Inclusion criteria Full-text articles were included if they met the following inclusion criteria: • Participants: patients with MS (≥18 years old) • Intervention: cognitive behavioral therapy, i.e. ○ Interventions stated by the authors to be cognitive behavioral therapy or recognizable as such from the description provided ○ Interventions focusing on cognitive and/or behavioral components • Type of studies: randomized controlled trials (RCTs) • Primary outcome measure: fatigue • Language: at least an English abstract 1.3. Study selection Two reviewers (LA and HB) independently screened the titles and abstracts of the studies that were identified by the search. Because

1.4. Data extraction Data were extracted into a data extraction sheet based on the PRISMA, TIDieR [24] and WIDER [25] guidelines that included: study design, patient characteristics (number, age, gender, type of MS, definition of diagnosis, years since diagnosis, EDSS, education level, marital status, country), patient inclusion and exclusion criteria, description and type of experimental intervention, description of the control intervention and the reported outcomes (including unit of measurement and interpretation of scores), and a statement as formulated by the authors of overall findings. Outcome data at baseline, short-term (end of intervention) and longest long-term follow-up were collected as reported in the article.

1.6. Data analysis and statistical analysis Data analysis was conducted using Revman 5.3 (The Nordic Cochrane Centre, The Cochrane Collaboration). Standardized mean differences (SMDs) and 95% confidence intervals (CI) were determined using a random-effects model with inverse variance methods, using short- term and (when available) long-term follow-up means and SDs. SMDs were interpreted according to Cohen's definitions: 0.2–0.5 is considered a small effect size; 0.5–0.8 a moderate effect and N0.8 is considered a large effect size [27]. Statistical heterogeneity was determined using I2 statistics [28] and interpreted according to the Cochrane guidelines. The guidelines distinguish the following categories: 0% to 40% is no heterogeneity, 30% to 60% is moderate, 50% to 90% is substantial and 75% to 100% is considerable heterogeneity [29]. Primary analysis included the short-term and long-term SMDs of interventions that were judged to focus solely on CBT. For the secondary analysis, the articles that included CBT components were added to the primary analysis. 2. Results The initial search yielded 994 articles, of which 833 remained after removal of duplicates. Five authors of unpublished articles were contacted about the progress of their research and preliminary results, however, this did not result in extra information. In total, 734 articles were excluded based on the title and abstract. Of the remaining 99 articles, 30 were excluded because CBT was not used, 16 articles did not use fatigue as a primary outcome measure, 24 studies did not use an RCT design and of 11 ‘articles’ only abstracts were available. These 11 abstracts were excluded since 2 were not about CBT, 5 did not use a RCT design and 4 were already included in a more recent publication. The interventions of the remaining 18 articles were masked, and experienced clinicians were asked to decide whether the intervention was CBT. Five interventions were considered solely CBT [13,30–34] and 1 study

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

contained CBT-components [35]. The study of van Kessel et al. [33] compared 2 study groups that received the same CBT treatment. However, one group received email support and the other did not. Because the between-group comparison only gives information about the effectiveness of e-mail support, no conclusions could be drawn regarding the effectiveness of CBT on fatigue. Therefore, this study was excluded from the meta-analysis on the effectiveness of CBT. The intervention that contained CBT-like components did not provide results suitable for the meta-analysis [35] and was therefore excluded from the metaanalysis. However, it was deemed suitable for inclusion in the systematic review. The complete study flowchart is displayed in Fig. 1.

35

low for the following items: random sequence generation, allocation concealment, selective reporting and other bias. However, the risk of bias was high in blinding of participants and personnel, and blinding of (patient-reported) outcome assessment [13,30–32,34]. Most studies in rehabilitation medicine show high risk scores on these items, due to the impracticality of blinding participants, therapists and outcome assessments. One study had a high risk of attrition bias (incomplete outcome data) [31]. The three studies [13,30,32,34] that provided longterm results had a low attrition rate, and scored as low risk on the item ‘incomplete outcome data’. 2.2. Characteristics of included trials

2.1. Assessment of risk of bias The characteristics of the 5 trials, truncated from the data extraction sheet, that were considered solely CBT (2 publications by Thomas et al. on the FACETS trial were combined [32,34]) and the study that contained components of CBT are shown in Table 1. All studies were published between 2008 and 2015. The 4 trials [13,30–33] included in

The right-hand section of Fig. 2 illustrates the methodological quality of the 4 studies included in the meta-analysis [13,30–32,34]. To rate the Risk of Bias of the FACETS study, 2 publications were combined [32,34]. For all 4 studies in the meta-analysis [13,30–32,34] the risk of bias was

Identification

Database search Pubmed Cochrane EMBASE Psychological and Behavioral Science Collection ERIC PsychINFO CINAHL PsycARTICLES

(n=680) (n=216) (n=61) (n=35) (n=2) (n=0) (n=0) (n=0)

Screening

994 Articles 833 Articles after duplicate removal

833 Titles and abstracts screened

81 Articles excluded

Eligibility

99 Assessed for eligibility

No CBT No RCT No fatigue Abstract

(n=30) (n=24) (n=16) (n=11)

18 Interventions were blinded and assessed by experienced clinicians

12 Full-text articles excluded:

6 Studies included in qualitative synthesis (review):

2 Full-text articles excluded:

CBT CBT-components

Included

734 Records excluded based on title and abstract

(n=5) (n=1)

No CBT

(n=12)

Design hampered (n=1) CBT conclusions No data Available

(n=1)

4 Studies included in quantitative analysis (meta-analysis) CBT CBT- components

(n=4) (n=0)

Fig. 1. Consolidated Standards of Reporting Trials statement: Flow diagram of review process.

36

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

Fig. 2. A. Short term comparison of CBT and control conditions on fatigue in patients with MS; B. Long term comparison of CBT and Control conditions on fatigue in patients with MS (+ = low risk; − = high risk).

the meta-analysis yielded data from 403 patients with MS, of whom 193 received CBT and 210 received a control condition. Short-term outcome measurements, that were mostly performed immediately after the end of treatment, ranged from 8 weeks [13] to 10 weeks [30–33]. Only Thomas et al. [32] performed the short-term outcome measurement 4 weeks post-treatment. 2.3. Effectiveness of CBT Fig. 2 shows the standardized mean differences and forest plot [13, 30–32,34] of the four RCTs included in the meta-analysis. All studies showed an improvement in fatigue, resulting in a moderately positive short-term effect (SMD = −0.47; 95% CI, −0.88 to −0.06; I2 = 73%). The high I2 value of 73% indicates that most of the variability in the results across the studies is due to heterogeneity rather than chance [28]. Three RCTs [13,30,32,34] provided long-term (8 months [13] to 12 months [30,32,34]) follow-up data. The SMD for these three studies was small (SMD = − 0.30; 95% CI, − 0.51 to − 0.08; I2 = 0%). An I2 value of 0% indicates no observed heterogeneity in study results [28]. Fig. 3 displays the SMDs on all measurement times (both short and long-term) per study. Of the three studies that provided long-term data, all three showed a different pattern: in one study the effect resided after treatment [13]; in another study the positive treatment effect remained [30], and in one study the effect became stronger after treatment finished [32,34]. 2.4. CBT intervention characteristics Table 2 summarizes the content of the CBT sessions for the different trials. Two studies used the same intervention: MS-invigor8 [31,33], the only difference was that both studies used a different support system (telephone [31], or email support [33]). All solely CBT interventions addressed activity scheduling in terms of budgeting rest and activity, how to identify and deal with stressors, how to identify unhelpful thoughts and how to deal with them; how to continue to employ the skills they had learned and how to deal with possible relapses [13,30–33]. Most of these interventions also addressed understanding MS-related fatigue [13,31–33], how to improve sleeping patterns [13,31–33], how to manage difficult emotions [13,30,31,33], and understanding MS symptoms

in terms of focusing on symptoms and their role on MS-related fatigue [13,31,33]. The intervention described by van Ehde et al. [30] also emphasized problem-solving and relaxation strategies. The study of Thomas et al. [32,34] uses CBT, as well as energy conservation management strategies. Energy conservation reduces fatigue through a systematic analysis of daily work, home and leisure activities in all relevant environments [11]. Energy conservation strategies are about reducing energy expenditures, managing work and rest and setting priorities, budgeting energy, using the body efficiently and optimizing workspaces [11], whereas CBT for MS-related fatigue aims to identify and where possible alter or manage any cognitive, behavioral, emotional and external factors that may be contributing to MS-related fatigue [12,13]. 2.5. Control conditions characteristics Two studies provided a control intervention that contained some type of therapy: relaxation training [13] and telephone-delivered education [30]. In the two remaining studies [31,32] the control condition consisted of patients continuing with their local care. 2.6. Dropouts, session adherence and adverse events The main reasons for CBT treatment dropout were: too busy, work commitments, relapses/unwell, changed my mind and unknown reasons. For the control treatments, the reasons for dropout were: nonresponders, personal reasons, additional illness diagnosed, too busy and unknown reasons. In one study [30], 2 patients quit the sessions because they considered the telephone sessions too scripted. Session adherence varied from 4.3% [31] to 100% [13] for CBT treatments (see Table 2) and from 5.0% [33] to 95% [13,35] for control treatments. Four studies described that no related adverse events were reported [30–32,35] and two studies did not mention whether any adverse event occurred [13,33]. 3. Discussion The results of this review and meta-analysis show that CBT has a positive moderate short-term effect (SMD = − 0.47; CI 95% − 0.88, − 0.06), which turns into a small long-term effect when treatment

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

37

Table 1 Study characteristics. Study CBT Van Kessel et al. [13]

CBT

Short description

N (F/M)

Mean age

Therapist

Modes of delivery

Individual/group Sessions

Session Adherence CBT

Challenge any behavioral, cognitive, emotional and external factors that may be contributing to MS fatigue

35 (28/7)

42.9 (9.3)

CBT trained clinical psychologist

Face to face and telephone

Individual

8 weekly sessions 50

100% completed all sessions

Face to face and telephone

Individual

min (3 face to face; 5 by telephone) 8 weekly sessions 50

95% completed all sessions

Online +

Individual

Control Relaxation training (RT)

Moss-Morris CBT et al. [31]

MSInvigor8: Internet self-management package for MS fatigue based on CBT principles: breaking the cycle of fatigue

37 47.0 (27/10) (9.5)

23 (16/7)

CBT trained clinical psychologist (same as CBT-group)

40.1 Assistant psychologist (5 h (17.8) basic training in the interventions +

telephone

min (3 face to face; 5 by telephone) 8 weekly sessions 25–50

4.3% completed all sessions; 61% completed more than half of the sessions One month later 47% completed all sessions; 83% completed more than half of the sessions N/A

min + 3 telephone support session 30–60

supervision from a registered psychologist)

min

Control Current local practice Thomas et al. [32,34]

CBT

FACETS: normalize fatigue experiences, learn helpful ways of thinking about it and use available energy more effectively

Control Current local practice Ehde et al. [30]

Van Kessel et al. [33]

CBT

T-SM: telephone delivered fatigue self-management intervention to effectively manage fatigue, chronic pain, and/or depression Control Telephone delivered MS education intervention CBT

MSInvigor8-Plus: Internet self-management package for MS fatigue based on CBT principles: breaking the cycle of fatigue with email support

Control MSInvigor8-Only: breaking the cycle of fatigue without email support

CBT–components Plow et al. CBT [35]

Group wellness Intervention (GWI): maximizing health in terms of physical activity, nutrition, setting health and priority goals and energy conservation

Control Individualized physical rehabilitation (IPR)

17 (16/1) 84 (61/13)

41.8 N/A (11.4) 48.0 Health professionals with (10.2) experience of working with MS patients (such as occupational therapists, nurses or physiotherapists) 80 50.1 N/A (58/22) (9.1) 75 51.0 Masters level social (67/8) (10.1) workers and doctoral level psychologists

N/A

N/A

N/A

Face to face

Group (6–12 people)

6 weekly sessions; 90 min

N/A

N/A

Telephone Individual

Telephone Individual

19 (11/8)

Email support by skilled clinical psychologist (with CBT experience)

Online +

N/A

Online

20 (18/2)

45.7 (8.4)

N/A

N/A

8 weekly sessions 45–60

77% completed all sessions

min

88 53.2 Same as CB-group (75/13) (10.0) 43.0 (8.2)

43% completed all sessions; 86% completed at least 4 out of 6 sessions

Individual

email support

Individual

8 weekly sessions 45–60

88% completed all sessions

min 8 weekly sessions 25–50

10.5% completed all sessions; 63% completed more than half of the sessions

min + weekly email support 10 min 8 weekly sessions 25–50

5% completed all sessions; 35% completed more than half of the sessions

min

20 (16/4)

48.5 (9.1)

N/R

Face to face

Group (sessions) + individual (exercise)

22 (17/5)

48.5 N/R (12.3)

Face to face + telephone

Individual

7 weekly sessions 120 min +

80% completed all sessions

home exercise program 5 days a week: 45 min 4 session every other week +

95% completed all sessions

phone call every other week (continued on next page)

38

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

Table 1 (continued) Study

Outcomes

Analysis

Follow up(s)

Measurement Result adherence (n) CBT

CBT van Kessel et al. [13]

Primary: 1) Fatigue severity (Fatigue scale) Secondary: 2) Fatigue-related impairment (Work and Social Adjustment Scale) 3) Anxiety and depression (HADS) 4) Perceived stress (Perceived Stress Scale) Moss-Morris et al. [31] Primary: 1) Fatigue severity (Fatigue scale) 2) Fatigue impact (MFIS) Secondary: 1) Anxiety and depression (HADS) 2) Economic evaluation (EQ-5d) 3) Service use (adapted version CSRI) 4) 15 participants were interviewed about their experiences of the intervention Thomas et al. [32,34] Primary: 1) Fatigue severity (GFS subscale of FAI) 2) Disease-specific quality of life (MSIS) 3) Self-efficacy for managing fatigue (multiple sclerosis-Fatigue Self-Efficacy Scale) Secondary (From the protocol [36]): 1) Subscales of the FAI 2) The Fatigue Symptom Inventory 3) Anxiety and depression (HADS) 4) Physical + psychological subscales (MSIS) 5) Short-form survey 6) Economic evaluation (eq-5d) 7) Self-reported health services resource utilization 8) Physical activity over a 48 h period (ActivPAL accelerometers) Ehde et al. [30] Primary: 1) Fatigue impact (MFIS) 2) Pain interference (BPI) 3) Depressive symptoms (PHQ-9) Secondary: 1) Pain intensity (NRS) 2) Negative affect (PANAS) 3) Self-efficacy (SES) 4) Activation (PAM) 5) Physical HRQoL (SF-8) 6) Mental HRQoL (SF-8) 7) Satisfaction with social role (PROMIS) 8) Resilience (CD-RISC) Van Kessel et al. [33] Primary: 1) Fatigue severity (Fatigue scale) 2) Fatigue impact (MFIS)

ITT + PP 1) baseline 2) 2 mona 3) 5 mon 4) 8 mon

ITT

1) baseline 2) 2.5 mona

ITT + PP 1) baseline 2) 2.5 mona,b 3) 5.5 mon 4) 12 mon [34]

C

Conclusion (citation of study authors)

CBT

C

Mean (SD)

Mean (SD)

Fatigue scale 1) 20.9 (4.3) 2) 7.9 (4.3)a 3) 9.0 (5.3) 4) 10.4 (6.4)

Fatigue scale 1) 20.3 (4.3) 2) 11.6 (5.3) a 3) 11.1 (4.6) 4) 12.5 (5.2)

“Both CBT and RT appear to be clinically effective treatments for fatigue in MS patients, although the effects for CBT are greater than those for RT. Even 6 months after treatment, both treatment groups reported levels of fatigue equivalent to those of the healthy comparison group.”

Fatigue scale 1) 21.5 (3.6) 2) 19.6 (5.2)a

“Internet CBT-based self-management appears to be a promising, acceptable and cost-effective approach for treating MS fatigue and improving broader outcomes such as distress.”

GFS 1) 5.6 (1.1) 2) 5.6 (1.2)a 3) 5.7 (0.9) 4) 5.7 (1.0)

“FACETS is effective in reducing fatigue severity and increasing fatigue self-efficacy” [32]. “Improvements in fatigue severity and self-efficacy at 4-month follow-up following attendance of FACETS were mostly sustained at 1

1) 35 2) 35 3) 35 4) 34

1) 37 2) 35 3) 35 4) 35

1) 23 2) 23

Fatigue scale 1) 17 1) 21.4 (4.3) 2) 17 2) 12.4 (6.8)a

1) 81 2) 71 3) 70 4) 62

1) 78 2) 75 3) 74 4) 69

GFS 1) 5.6 (1.0) 2) 5.5 (0.9)a 3) 5.3 (1.0) 4) 5.3 (1.0)

year with additional improvements in MS impact. The FACETS program provides modest long-term benefits to people with MS-fatigue” [34].

ITT

ITT

MFIS 1) 48.0 (14.7) 2) 38.6 (15.9)a 3) 37.3 (16.0) 4) 40.2 (16.5)

MFIS 1) baseline 2) 2.5 mona 3) 6 mon 4)12 mon

1) 64 2) 64 3) 62 4) 60

1) 81 2) 81 3) 79 4) 80

Fatigue scale 1) baseline 2) 2.5 mona

1) 19 2) 19

Fatigue scale 1) 20 1) 22.4 (4.4) 2) 20 2) 11.4 (6.2)

MFIS 1) 51.2 (12.7) 2) 42.4 (15.8)¶ 3) 41.7 (16.2) 4) 43.3 (15.8)

“Both interventions resulted in short- and long-term, clinically meaningful benefits. The study demonstrated that the telephone is an effective method for engaging participants in and extending the reach of care for individuals with MS.”

Fatigue scale 1) 22.4 (4.4) 2) 17.5 (6.4)

“The MSInvigor8-Plus condition resulted in significantly greater reductions in fatigue severity and impact compared with the

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

39

Table 1 (continued) Study

Outcomes

Analysis

Follow up(s)

Measurement Result adherence (n) CBT

C

Conclusion (citation of study authors)

CBT

C

Mean (SD)

Mean (SD)

Secondary: 1) Anxiety and depression (HADS)

CBT-components Plow et al. (2009) [35]

Primary: 1) Mental health inventory 2) Mental health (SF36 3) Modified Fatigue Impact Scale (MFIS) 4) Physical HRQoL (SF-8) 5) Mental HRQoL (SF-8) Secondary: 1) Physical activity 2) Resting heart rate 3) Bench press 4) Sit-to-stand 5) Estimated VO2max 6) Waist circumference

MSInvigor8-Only condition. Large between-group effect sizes for fatigue severity and fatigue impact were obtained. No effects were found on changes in anxiety and depression.”

N/R

GWI improves health and physical activity; patients benefit mentally.

MFIS 1) pre-baseline (6 weeks before baseline) 2) baseline 3) 2 mona 4) 4 mon

Abbreviations: BPI, Brief Pain Inventory; C, control condition; CD-RISC, Connor-Davidson Resilience Scale; CSRI, client service receipt inventory; Eq-5d, Euro-Qol; F, female; FAI, Fatigue Assessment Instrument; GFS, Global Fatigue Severity; GWI, Group Wellness Intervention; HADS, Hospital Anxiety and Depression Scale; HRQoL, Health related quality of life; IPR, Individualized Physical Rehabilitation; M, male; MFIS, Modified Fatigue Impact Scale; mon, months; MSIS, multiple sclerosis Impact Scale; NRS, numeric rating scale; PAM, Patient Activation Measure; PANAS, Positive and Negative Affect Scale; PHQ-9, Patient Health Questionnaire; PROMIS, Patient Reported Outcomes Measurement Information System RT, Relax Training; SES, Self-Efficacy Scale; SF-8/36, Short-Form Health Survey of 8 or 36 items. a End of intervention period. b Thomas et al. [32,34] performed the posttreatment measurement one month after the end of the intervention.

ends (SMD = − 0.30; CI 95% − 0.51, − 0.08). A note of caution is due since the way CBT was provided was quite heterogeneous in terms of the mode of delivery of CBT (e.g. internet guidance vs. phone contact; group vs. individual guidance), the intensity (i.e. number and duration of sessions) and the type of therapists (e.g. physiotherapists and qualified CBT-therapists). Furthermore, the control conditions differed markedly (e.g. varying from relaxation training to usual care). The differences in content of CBT, and the great variety in the way CBT was provided might explain the heterogeneity of studies, which was found (Fig. 2A, I2 = 73%). Due to the heterogeneity of the studies and results, we wish to highlight certain factors that should be considered when interpreting the calculated SMDs reported in the current review and meta-analysis.

First, in the study by van Kessel et al. [33] patients in both the intervention group and the control group received the same CBT program (MSInvigor8), with the difference that patients in the intervention group received additional email support. Due to this design the article had to be excluded from the meta-analysis. Although small betweengroup effects may be expected, van Kessel et al. [33] found a large positive effect for email support. Second, complete treatment adherence was low, with 10.5% in the CBT group and only 5% in the control group completing all 8 sessions (63% of the CBT group vs. 35% of the control group completed more than half of the sessions). Treatment adherence was also low in the MSInvigor8 study by Moss-Morris et al. [31]; 4.3% completed all 8 sessions and 60.8% complete more than half. Both of these pilot studies did lack a proper power calculation.

Standardized Mean Difference

Treatment period

1.4

Follow-up period 1.2

Moss - Morris et al.

Standardized Mean Differences (SMD)

(2012) 1.0

0.8

0.6 Thomas et al. 0.4

(2013) van Kessel et al. (2008)

0.2

Ehde et al.. (2015)

0.0 0

10

20

30

40

50

Nr of Weeks

Fig. 3. Standardized mean differences (non-weighted) on all measurement times per study.

40

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

Table 2 Summary of content of provided CBT sessions for treating MS-related fatigue. Solely CBT Session

CBT-components

van Kessel et al. [13]

Moss-Morris et al. [31] van Kessel et al. [33]

Thomas et al. [32,34,36]

1

MS-related fatigue

Understanding MS fatigue

What is MS-related fatigue?

2

CBT for MS fatigue

Fatigue Diary

Opening an “energy account” Budgeting energy and smartening up goals

Ehde et al. [30] Introduction to self-management Goal setting and identifying stressors

Energy management

Nutrition

Rest and Activity Patterns.

4

Improving your sleep

Improving Sleep

5

Understanding your MS symptoms

Understanding MS Symptoms

6

Changing your thinking

Recording Thoughts

Managing stress, determining a sense of control and coping with emotions Social support and preparing for the future

Managing Stress

Managing emotions

Emotions, Support and the future

Building resilience

8

The combination of low treatment adherence and a small sample size increases the risk of bias in standardized mean differences [37]. Third, in an earlier study by van Kessel et al. [13] the standardized mean differences might underestimate the effect of CBT, since the active control group (relaxation training) also showed major improvement. Using a placebo-controlled study design might have resulted in larger SMD's [37]. Fourth, Thomas et al. [32,34] found an improvement in MSrelated fatigue 4 months after the intervention ended. These authors concluded that changes may take longer to have an impact because behavioral changes are only slowly adopted in daily life. Finally, we should also emphasize the difference in the specific aim of the treatments. The study performed by van Ehde et al. [30] provided a CBT-based selfmanagement intervention to decrease fatigue, pain and depression. They found smaller SMDs on fatigue, which might be due to the fact that 20% of the included patients had no fatigue problems, or due to the fact that there was insufficient emphasis during treatment on MSrelated fatigue. This broader focus might dilute SMDs. Over- and underestimations are an inevitable danger in systematic reviews. When we take them into account, the results of this study still indicate that CBT has a moderately positive effect over the short term.

Promotion physical activity

Problem solving and relaxation

Activity scheduling

7

Maximizing health with MS

Setting health goals and prescribing exercise programs

3

The stress response; Introducing the cognitive behavioral model Putting unhelpful thoughts on trial Recapping and taking the program forward

Plow et al. [35]

Working with thoughts (Part 1) Working with thoughts (Part 2)

Stress and depression Energy conservation Priority setting

related in patients with MS [39], exclusion of patients with clinical depression helps isolate the effect of CBT on MS-related fatigue alone. Moreover, the CBT interventions were specific for treating fatigue, the interventions did not include modules to treat depressive symptoms. Some limitations should be considered when interpreting the results presented here. First, this review included only a limited number of RCTs, which hampered taking risk of bias scores into account. Furthermore, due to the small amount of studies, subgroup analyses to study the influence of, for instance, the type of therapist or the way in which CBT was provided (individual/group; face-to-face/telephone/internet, number of sessions etc.) were not possible. This information is important for the future optimization of CBT treatment. Besides, if more studies are performed it would be interesting to see whether low-cost options can be as effective as the more costly methods of delivering CBT treatment. Second, limited evidence was available to draw conclusions about long-term effects, because only 3 studies provided this data. Third, publication bias can not be ruled out, since no information became available after contacting authors of unpublished trials. And finally, one study was excluded because essential data was not reported in the original article and it was not possible to retrieve this information from the authors [35].

3.1. Study strengths and limitations 3.2. Implications for clinical practice and research A robust aspect of the current review was that only studies with an RCT design were considered. An RCT design is regarded as an important inclusion criterion for the mathematical synthesis of data [29,38]. Since no general definition of CBT exists, the decision was made to let experienced CBT psychologists make the final decision as to whether interventions were accepted as CBT. Since CBT can contain multiple components, a broad range of behavioral and cognitive interventions may be considered CBT. However, experienced therapists can reliably recognize whether studies qualify as CBT. The therapists were asked to categorize the interventions based on their own clinical experiences. The current review also went beyond effectiveness, and provides information on the content of the CBT interventions of the included trials. This information will aid in optimizing the content of CBT, and therefore enhance therapy in terms of finding treatment topics that help tackle MSrelated fatigue. A further strength was that 4 (80%) [13,31–34] of the solely CBT trials provided interventions that were specifically aimed at decreasing fatigue, which aids the homogeneity of the provided treatments. A final strength was that 4 (80%) of the included solely CBT studies excluded depressed patients using the following criterion: antidepressant treatments any anti-depressant treatments in the 3 months before inclusion. Since fatigue and depression can be closely

This is the first meta-analysis of the effectiveness of CBT for fatigue in patients with MS, and it provides new insights for clinical applications. The results indicate a moderately positive short-term effect of CBT of 8 to 10 weeks on fatigue in patients with MS. When patients experience severe fatigue problems, CBT should be considered an evidence-based therapy for MS-related fatigue. Furthermore, this review and metaanalysis is the first to provide insights into the content of CBT. Most interventions covered the topics: how to identify unhelpful thoughts and how to deal with them, activity scheduling, how to identify and deal with stressors, how to continue to employ learned skills, and how to deal with possible relapses, and might be part of future fatigue programs. From the calculated SMDs we conclude that the long-term effect slightly diminishes, which indicates that future research should study how these positive short-term effects can be preserved. A possible solution might be to provide booster sessions after treatment ends. Furthermore, it would be interesting to analyze whether the way CBT is provided (face-to-face, telephone or online) influences the positive effect. van Kessel et al. [33] partially studied this issue and found that email support during an online intervention positively influenced adherence.

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

4. Conclusion

(continued)

The results of the current systematic review and meta-analysis suggest a moderately positive effect of CBT for the treatment of fatigue in patients with MS. However, this effect declines after cessation of treatment. Since the short-term effect of CBT on MS-related fatigue is positive, more research is needed to develop interventions that maintain these short-term effects in the long term.

Literature search, 8-2-2016 EMBASE

Fatigue OR tired OR tiredness OR energy OR vitality OR physical AND fatigue OR cognitive AND fatigue OR mental AND fatigue OR ‘worn out’ OR lethargy OR chronic AND fatigue OR fatigability OR weariness AND

The TREFAMS-ACE study is funded by the Fonds NutsOhra (ZonMw 89000005). Acknowledgements This study is supported by the Fonds NutsOhra (grant no. ZonMW 89000005). This article is produced in the context of the Trefams-ACE study: treatment of fatigue in multiple sclerosis: aerobic training, cognitive behavioral therapy and Energy Conservation Management. Appendix A. Trial registers The following trial registers were searched. Australian and New Zealand Clinical Trials Registry, Brazilian Clinical Trials Registry (REBEC), International Clinical Trials Registry platform, Chinese Clinical Trial Register, Clinical Research Information Service - Republic of Korea, ClinicalTrials.gov, Current Controlled Trials metaRegister, Current Controlled Trials metaRegister, EU Clinical Trials Register, German Clinical Trials Register, Hong Kong University Clinical Trials Register, Indian Clinical Trials Registry, International Clinical Trials Registry platform (WHO), IFPMA portal, ISRTCN, Iranian Registry of Clinical Trials, Japan Primary Registries Network,Netherlands National Trial Registry, Pan African Clinical Trials Registry, Sri Lanka Clinical Trials Registry, South African National Trials Registry, UK Clinical Research Network, UK Clinical Trials Gateway, UMIN.

ERIC,

NOT PsycINFO, Animals Not humans PsycARTICLES, AND Psychology and Behavioral Sciences

Multiple sclerosis OR transverse myelitis OR disseminated sclerosis NOT animals [mh] NOT humans [mh]

AND Cognitive therapy OR CT OR rational emotive behavior therapy OR REBT OR rational emotive therapy OR RET OR behavior therapy OR BT OR cognitive behavioral therapy OR CBT OR psychotherapeutic treatment OR neo behaviorism OR acceptance and commitment therapy OR ACT OR relational frame theory OR compassionate mind training OR CMT OR functional analytic psychotherapy OR FAP OR metacognitive therapy OR MCT OR mindfulness based cognitive therapy OR MBCT OR dialectical behavior therapy OR DBT OR behavioral activation OR BA OR fatigue management OR energy conservation OR multidisciplinary rehabilitation OR self-management OR

Fatigue OR tired OR tiredness OR energy OR vitality OR physical fatigue OR cognitive fatigue OR mental fatigue OR worn-out OR lethargy OR chronic fatigue OR fatigability OR weariness AND

Literature search, 8-2-2016

Fatigue OR tired* OR tiredness OR energy OR vitality OR physical fatigue OR cognitive fatigue OR mental fatigue OR worn-out OR lethargy OR chronic fatigue OR fatigability OR weariness

Cognitive AND therapy OR ct OR rational AND emotive AND behavior AND therapy OR rebt OR rational AND emotive AND therapy OR ret. ORbehavior AND therapy OR bt OR cognitive AND behavioral AND therapy OR cbt OR psychotherapeutic AND treatment OR neo AND behaviorism OR acceptance AND commitment AND therapy OR act OR relational AND frame AND theory OR compassionate AND mind AND training OR cmt OR functional AND analytic AND psychotherapy OR fap OR metacognitive AND therapy OR mct OR mindfulness AND based AND cognitive AND therapy OR mbct OR dialectical AND behavior AND therapy OR dbt OR behavioral AND activation OR ba OR fatigue AND management AND multidisciplinary Multiple sclerosis OR transverse myelitis OR disseminated sclerosis

CINAHL,

Appendix B. Search strategy

AND

occupational therapy OR mindfulness Multiple AND sclerosis AND

Funding

Pubmed

41

Cochrane

Cognitive therapy OR CT OR rational emotive behavior therapy OR REBT OR rational emotive therapy OR RET OR behavior therapy OR BT OR cognitive behavioral therapy OR CBT OR psychotherapeutic treatment OR neo behaviorism OR acceptance and commitment therapy OR ACT OR relational frame theory OR compassionate mind training OR CMT OR functional analytic psychotherapy OR FAP OR metacognitive therapy OR MCT OR mindfulness based cognitive therapy OR MBCT OR dialectical behavior therapy OR DBT OR behavioral activation OR BA OR fatigue management OR energy conservation OR multidisciplinary rehabilitation OR self-management OR occupational therapy OR mindfulness Multiple sclerosis [tiab] or transverse myelitis or disseminated sclerosis NOT animals [mh] not humans [mh] AND fatigue or tired* or tiredness or energy or vitality or physical fatigue or cognitive fatigue or mental fatigue or worn-out or lethargy or chronic fatigue or fatigability or weariness AND cognitive therapy or CT or rational emotive behavior therapy or REBT or rational emotive therapy or RET or behavior therapy or BT or cognitive behavioral therapy or CBT or psychotherapeutic treatment or neo (continued on next page)

42

L.E. van den Akker et al. / Journal of Psychosomatic Research 90 (2016) 33–42

(continued) Literature search, 8-2-2016 behaviorism or acceptance and commitment therapy or ACT or relational frame theory or compassionate mind training or CMT or functional analytic psychotherapy or FAP or metacognitive therapy or MCT or mindfulness based cognitive therapy or MBCT or dialectical behavior therapy or DBT or behavioral activation or BA or fatigue management or energy conservation or multidisciplinary rehabilitation or self-management or occupational therapy or mindfulness Only trials

Appendix C. Supplementary data Supplementary material. References [1] A. Compston, A. Coles, Multiple sclerosis, Lancet 25; 372 (9648) (2008) 1502–1517. [2] L. Krupp, Fatigue is intrinsic to multiple sclerosis (MS) and is the most commonly reported symptom of the disease, Mult. Scler. 12 (4) (2006 Aug) 367–368. [3] L.B. Krupp, D.J. Serafin, C. Christodoulou, Multiple sclerosis-associated fatigue, Expert. Rev. Neurother. 10 (9) (2010) 1437–1447. [4] R.J. Fox, T.E. Bacon, E. Chamot, A.R. Salter, G.R. Cutter, J.T. Kalina, et al., Prevalence of multiple sclerosis symptoms across lifespan: data from the NARCOMS Registry, Neurodegen. Dis. Manag. 5 (6 Suppl) (2015 Dec) 3–10. [5] I. Induruwa, C.S. Constantinescu, B. Gran, Fatigue in multiple sclerosis - a brief review, J. Neurol. Sci. 323 (1–2) (2012 Dec) 9–15 (Dec 15). [6] V. de Groot, H. Beckerman, J.W. Twisk, B.M. Uitdehaag, R.Q. Hintzen, A. Minneboo, Vitality, perceived social support and disease activity determine the performance of social roles in recently diagnosed multiple sclerosis: a longitudinal analysis, J Rehabil Med. 40 (2) (2008 Feb) 151–157. [7] L. Leocani, B. Colombo, G. Comi, Physiopathology of fatigue in multiple sclerosis, Neurol. Sci. 29 (Suppl 2) (2008 Sep) S241–S243. [8] D.A. Trojan, D. Arnold, J.P. Collet, S. Shapiro, A. Bar-Or, A. Robinson, et al., Fatigue in multiple sclerosis: association with disease-related, behavioural and psychosocial factors, Mult. Scler. 13 (8) (2007 Sep) 985–995. [9] L.B. Krupp, L.A. Alvarez, N.G. LaRocca, L.C. Scheinberg, Fatigue in multiple sclerosis, Arch. Neurol. 45 (4) (1988 Apr) 435–437. [10] S.J. Hourihan, Managing fatigue in adults with multiple sclerosis, Nurs. Stand. 29 (43) (2015 Jun 24) 51–58. [11] L.J. Blikman, B.M. Huisstede, H. Kooijmans, H.J. Stam, J.B. Bussmann, J. van Meeteren, Effectiveness of energy conservation treatment in reducing fatigue in multiple sclerosis: a systematic review and meta-analysis, Arch. Phys. Med. Rehabil. 94 (7) (2013 Jul) 1360–1376. [12] K. van Kessel, R. Moss-Morris, Understanding multiple sclerosis fatigue: a synthesis of biological and psychological factors, J. Psychosom. Res. 61 (5) (2006 Nov) 583–585. [13] K. van Kessel, R. Moss-Morris, E. Willoughby, T. Chalder, M.H. Johnson, E. Robinson, A randomized controlled trial of cognitive behavior therapy for multiple sclerosis fatigue, Psychosom. Med. 70 (2) (2008 Feb) 205–213. [14] L.G. Öst, Efficacy of the third wave of behavioral therapies: A systematic review and meta analysis, Behav. Res. Ther. 46 (2008) 295–321. [15] B.F. Skinner, Some contributions of an experimental analysis of behavior to psychology as a whole, Am. Psychol. 8 (1953) 9. [16] J.B. Watson, Behaviorism, Nortan, NY, 1925. [17] A.T. Beck, Cognitive therapy: past, present, and future, J. Consult. Clin. Psychol. 61 (2) (1993 Apr) 194–198. [18] A. Beck, M. Braunstein, Cognitive Therapy and the Emotional Disorders, 1976.

[19] A. Ellis, Outcome of employing three techniques of psychotherapy, J. Clin. Psychol. 13 (4) (1957 Oct) 344–350. [20] J.D. Herbert, E.M. Forman, E.L. England, Psychological acceptance, in: J. Wiley (Ed.), General Principles and Empirically Supported Techniques of Cognitive Behavior Therapy, Hoboken, NJ 2009, pp. 102–114. [21] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P. Group, Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement, Open Med. 3 (3) (2009) 123–130. [22] P.W. Thomas, S. Thomas, C. Hillier, K. Galvin, R. Baker, Psychological interventions for multiple sclerosis, Cochrane Database Syst. Rev. 25 (1) (2006). [23] V. Hunot, T.H. Moore, D.M. Caldwell, T.A. Furukawa, P. Davies, H. Jones, et al., ‘Third wave’ cognitive and behavioural therapies versus other psychological therapies for depression, Cochrane Database Syst. Rev. 18 (2013) 10. [24] T.C. Hoffmann, P.P. Glasziou, I. Boutron, R. Milne, Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide, BMJ 348 (2014) g1687. [25] L. Albrecht, M. Archibald, D. Arseneau, S.D. Scott, Development of a checklist to assess the quality of reporting of knowledge translation interventions using the workgroup for intervention development and evaluation research (WIDER) recommendations, Implement. Sci. 8 (2013) 52. [26] J.P.T. Higgings, D.G. Altman, J.P.T. Higgings, Assessing risk of bias in included studies, in: S. G (Ed.), Cochrane handbook for systematic reviews of interventions, Wiley, Hoboken 2011, pp. 187–241. [27] J. Cohen, The statistical power of abnormal-social psychological research: a review, J. Abnorm. Soc. Psychol. 65 (1962) 145–153. [28] J.P. Higgins, S.G. Thompson, J.J. Deeks, D.G. Altman, Measuring inconsistency in meta-analyses, BMJ 327 (2003) 557–560. [29] J.P.T. Higgins, S. Green, Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011], The Cochrane Collaboration, 2011 www. cochrane-handbook.org (Available from). [30] D.M. Ehde, J.L. Elzea, A.M. Verrall, L.E. Gibbons, A.E. Smith, D. Amtmann, Efficacy of a telephone-delivered self-management intervention for persons with multiple sclerosis: a randomized controlled trial with a one-year follow-up, Arch. Phys. Med. Rehabil. 96 (11) (2015 Nov) 1945–1958. [31] R. Moss-Morris, P. McCrone, L. Yardley, K. van Kessel, G. Wills, L. Dennison, A pilot randomised controlled trial of an Internet-based cognitive behavioural therapy self-management programme (MS Invigor8) for multiple sclerosis fatigue, Behav. Res. Ther. 50 (6) (2012 Jun) 415–421. [32] S. Thomas, P.W. Thomas, P. Kersten, R. Jones, C. Green, A. Nock, et al., A pragmatic parallel arm multi-centre randomised controlled trial to assess the effectiveness and cost-effectiveness of a group-based fatigue management programme (FACETS) for people with multiple sclerosis, J. Neurol. Neurosurg. Psychiatry 84 (10) (2013 Oct) 1092–1099. [33] K. van Kessel, T. Wouldes, R. Moss-Morris, A New Zealand pilot randomized controlled trial of a web-based interactive self-management programme (MSInvigor8) with and without email support for the treatment of multiple sclerosis fatigue, Clin. Rehabil. 30 (5) (2016 May) 454–462. [34] P.W. Thomas, S. Thomas, P. Kersten, R. Jones, V. Slingsby, A. Nock, et al., One year follow-up of a pragmatic multi-centre randomised controlled trial of a groupbased fatigue management programme (FACETS) for people with multiple sclerosis, BMC Neurol. 14 (2014) 109. [35] M.A. Plow, V. Mathiowetz, D.A. Lowe, Comparing individualized rehabilitation to a group wellness intervention for persons with multiple sclerosis, Am. J. Health Promot. 24 (1) (2009) 23–26. [36] P.W. Thomas, S. Thomas, P. Kersten, R. Jones, A. Nock, V. Slingsby, et al., Multi-centre parallel arm randomised controlled trial to assess the effectiveness and costeffectiveness of a group-based cognitive behavioural approach to managing fatigue in people with multiple sclerosis, BMC Neurol. 10 (2010 Jun 16) 43. [37] N. Takeshima, T. Sozu, A. Tajika, Y. Ogawa, Y. Hayasaka, T.A. Furukawa, Which is more generalizable, powerful and interpretable in meta-analyses, mean difference or standardized mean difference? BMC Med. Res. Methodol. 14 (2014 Feb 21) 30. [38] E.M. Antman, J. Lau, B. Kupelnick, F. Mosteller, T.C. Chalmers, A comparison of results of meta-analyses of randomised controlled trials and recommendations of clinical experts, JAMA 268 (1992) 240–248. [39] S. Simpson, H. Tan, P. Otahal, B. Taylor, A.L. Ponsonby, R.M. Lucas, et al., Anxiety, depression and fatigue at 5-year review following CNS demyelination, Acta Neurol. Scand. (2016 Jan 12) 1–11.