Psychological interventions for multiple sclerosis

Peter W Thomas; Sarah Thomas; Charles Hillier; Kate Galvin; Roger Baker

doi:10.1002/14651858.CD004431.pub2

. 2006 Jan 25;2006(1):CD004431. doi: 10.1002/14651858.CD004431.pub2

Psychological interventions for multiple sclerosis

Peter W Thomas ^1,^✉, Sarah Thomas ¹, Charles Hillier ², Kate Galvin ³, Roger Baker ¹

Editor: Cochrane Multiple Sclerosis and Rare Diseases of the CNS Group

PMCID: PMC8406851 PMID: 16437487

Abstract

Background

The unpredictable, variable nature of Multiple Sclerosis (MS), and the possibility of increasing disability, means that a diagnosis can have substantial psychological consequences.

Objectives

To assess the effectiveness of psychological interventions for people with MS.

Search methods

We searched the Cochrane MS Group Specialised Register (December 2004), Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 4, 2004 ), MEDLINE (January 1966 to December 2004), PsychINFO (January 1887 to December 2004), CINAHL (January 1982 to December 2004) and 14 others. We searched reference lists of articles, wrote to corresponding authors of the 13 papers identified by June 2004, and searched for trials in progress using 3 research registers.

Selection criteria

Randomised controlled trials of interventions described as wholly or mostly based on psychological theory and practice, in people with MS. Primary outcome measures were disease specific and general quality of life, psychiatric symptoms, psychological functioning, disability, and cognitive outcomes. Secondary outcome measures were number of relapses, pain, fatigue, health care utilisation, changes in medication, and adherence to other therapies.

Data collection and analysis

Pertinent studies were identified from abstracts by one author. Full papers were independently compared to selection criteria by four authors. Key details were extracted from relevant papers using a standard format, and studies scored on three dimensions of quality. The review is organised into four mini‐reviews (MR) dependent on the intervention's target population; people with cognitive impairments (MR1), people with moderate to severe disability (MR2), people with MS (no other criteria) (MR3), and people with depression (MR4).

Main results

Overall 16 studies were identified and included. MR1: three trials (n=145). Some evidence of effectiveness of cognitive rehabilitation on cognitive outcomes, although this was difficult to interpret because of the large number of outcome measures used. MR2: three trials (n=80). One small trial suggesting psychotherapy may help with depression. MR3: seven studies (n=688). Some evidence that cognitive behavioural therapy may help people adjust to, and cope with, having MS (three trials). The other trials were diverse in nature and some difficult to interpret because of multiple outcome measures. MR4: three trials (n=93). Two small studies of cognitive behavioural therapy showed significant improvements in depression.

Authors' conclusions

The diversity of psychological interventions identified indicates the many ways in which they can potentially help people with MS. No definite conclusions can be made from this review. However there is reasonable evidence that cognitive behavioural approaches are beneficial in the treatment of depression, and in helping people adjust to, and cope with, having MS.

Plain language summary

Psychological treatments to help improve the quality of life of people with multiple sclerosis

In many countries MS is the most common neurological disorder among young adults. Its impact can be overwhelming with the person facing the likelihood of reduced physical function and of disability, with consequent disruptions in education, employment, sexual and family functioning, friendships and activities of daily living. MS can have a considerable impact on the individual's sense of self, especially if they can no longer perform previously valued activities. Unpleasant side effects from medication may also occur. Mood disorders such as depression and anxiety are common in people with MS, and are often a result of finding it difficult to adjust to, and cope with, having the disorder. Cognitive functioning (the mental processes of memory, concentration, reasoning and judgement) can also be affected. Therefore a diagnosis of MS can have substantial psychological consequences.

The authors of this review wanted to assess the effectiveness of psychological interventions (such as those addressing cognitive functioning, thoughts, mood and behaviour) for people with MS. This was done by considering their effect on quality of life, mood, cognitive functioning and disability in particular, but also on pain, fatigue, and use of other health related services and treatments.

Sixteen relevant studies were identified and included in this review. They have researched a variety of different interventions, having different purposes, and so a single overall definite conclusion cannot be made. However the authors cautiously conclude that Cognitive Behavioural Therapy, a therapy that addresses thoughts and behaviours, can help people with MS adjust to, and cope with, having MS, and can help them if they get depressed.

Psychological interventions can potentially help people with MS in many ways, including the management of symptoms such as pain and fatigue. Additional studies are needed, particularly those that include larger numbers of people.

Background

Multiple sclerosis (MS) is the most common neurological disorder among young adults and affects approximately 85 000 people in the UK (Graham 2002). It is a chronic and often disabling disease that typically commences between the ages of 20 and 40 years. It is more common in females than males by a ratio of approximately 3:2. The cause and early development of the disease are not fully understood but the current belief is that it is an autoimmune disorder affecting genetically susceptible individuals, possibly triggered by environmental or other factors. MS is characterised by inflammation and demyelination of the central nervous system. Virtually all functions innervated by the CNS can be affected. Common symptoms include, but are not limited to, loss of function or feeling in limbs, loss of bowel or bladder control, sexual dysfunction, debilitating fatigue, blindness due to optic neuritis, loss of balance, pain, cognitive dysfunction and mood disorders (Mohr 2001a). There is currently no cure and only minimal symptomatic relief available.

MS is a disease with an unpredictable course and prognosis is difficult to predict. Three basic types of clinical course of MS can be distinguished:

1. Relapse‐remitting MS is characterised by stable phases alternated with relapses. Relapses tend to be unpredictable and their causes are unclear. They can last for hours, days, weeks or months and vary from mild to severe. During a relapse new symptoms may occur or previous symptoms may return. Remissions can last as long as several years. Approximately 25% of people with MS have this form of the disease.

2. Secondary progressive MS commences with an initial relapse remitting course but is then followed by a progressive phase which is characterised by a steadily worsening condition. Approximately 40% of people with MS develop the secondary form, usually about 15 to 20 years after the initial onset of MS.

3. In the case of primary progressive MS there is no distinct pattern of relapses and remissions. The disease commences with steadily worsening symptoms and progressive disability which may level off or continue to deteriorate. Approximately 15% of people with MS have this form of the disease.

In addition to the course of MS these types of MS differ on a number of characteristics such as age at onset, degree of disability, disease duration and progression rate. As MS advances, the nature and severity of an individual's symptoms become increasingly heterogeneous. Some may experience few exacerbations of the illness whereas others may experience a rapid decline of function and eventually become wheelchair dependent. Similarly, while some individuals may experience minimal cognitive impairment, in others there may be severe decline.

PSYCHOLOGICAL AND SOCIAL FACTORS

A diagnosis of MS has profound social and psychological consequences. Because MS usually strikes individuals in their most productive years its impact can be overwhelming (Scheinberg 1984). The unpredictable and variable nature of MS may make it particularly difficult to come to terms with. The individual is firstly faced with the impact of receiving a diagnosis of a disease which is chronic, has an unpredictable course and affects many spheres of functioning. They face the likelihood of reduced physical function and disability in the future. Disruptions in education, employment, sexual and family functioning, friendships and activities of daily living are likely to occur. The unpredictability of day‐to‐day health in relapse remitting MS may greatly impinge upon quality of life (Mullins 2001). Individuals may, in addition, experience unpleasant side effects from medication.

MS can have a considerable impact on the individual's sense of self (LaRocca 1993). Physical changes and functional limitations may lead to a sense of loss of identity or role strain especially when the individual can no longer perform previously valued activities (Mullins 2001). It may be necessary to redefine one's self‐image in order to incorporate the limitations imposed by MS. Each time the individual experiences a new loss of function this sense of loss may be renewed (LaRocca 1993). Because of increasing disability, individuals with MS may have problems with isolation (O'Brien 1993, Walsh 1989), perceived social support (Miles 1979, O'Brien 1993), and social contacts (Gilchrist 1994, McIvor 1984). Loss of social support and/or social role has also been shown to be associated with depression (Mohr 2004).

The individual's perceptions of the uncertainty and variability of the disease and the perceived intrusiveness on daily activities are all related to depression and adjustment to the illness (Mullins 2001). Although illness related factors (such as neurological deterioration and functional disability) may partially contribute to how well the person with MS adjusts, the psychological response to the highly stressful nature of this disease is also likely to be important.

Individuals with MS exhibit a higher prevalence of mood disorders relative to individuals with comparable degrees of physical disability (Rao 1992, Minden 1991, Pollock 1990, Schiaffino 1996; Schubert 1993). Mood disorders refer to a "sustained and pervasive emotion that influences perception of self, others and the world such as depression or anxiety" (Minden 2000). Research has tended to focus on depression; however people with MS have also been found to have high levels of anxiety (Maurelli 1992).

Depression is one of the more common and debilitating psychological symptoms in MS (Thompson 1996). Lifetime prevalence of major depressive disorder (MDD) is approximately 50% (Joffe 1987; Minden 1987; Sadovnick 1991). Point prevalence estimates are less clear, ranging from 14% to 57% depending on the assessment instrument, criteria, and patient samples used (Schiffer 1983; Schubert 1993; Surridge 1969; Whitlock 1980). Rates of depression are higher than in other chronic illness (Minden 1987; Surridge 1969) or neurological disorders (Rabins 1986; Whitlock 1980). The high prevalence of depression in MS may have multiple aetiologies (Mohr 2001a) including psychosocial factors such as loss of social support or social role (Barnwell 1997; Gilchrist 1994; Gulick 1997) and inadequate coping (Aikens 1997; Mohr 1997a; Pakenham 1997; Pakenham 1999). It may also be a concomitant of immune dysregulation associated with MS exacerbations (Dalos 1983; Fassbender 1998) and the development of brain lesions (Franklin 1988; Pujol 1997). Thus depression can be both a complication associated with MS as well as a symptom of MS. The psychoneuroimmunology literature suggests that depression may induce abnormalities of immune functions that are relevant to MS (Mohr 2001b). There is also evidence (albeit equivocal) for depression being an iatrogenic side effect of MS disease‐modifying drugs (Mohr 1996; Mohr 1998; Mohr 1999a). Another complicating factor is that many of the symptoms of depression are confounded with MS (e.g. fatigue, psychomotor agitation or retardation, changes in sleep patterns and diminished ability to concentrate). Thus the relationship of mood disorders to MS is multi‐factorial and complex, and the extent to which they are direct consequences of the disease process itself or psychological reactions to it remains unclear. Such issues have important implications for the therapeutic approach that is adopted.

The relationship between medical and psychological variables is likely to be complex: The psychological state of an individual may affect their adherence to medical regimens (Mohr 1996), thereby exerting an indirect effect on disease processes. Psychological states may also directly affect MS. Stress arising from interpersonal conflict and disruption in routine can increase the likelihood of developing new brain lesions (Mohr 2002) and depression may increase MS related auto‐immune activity (Mohr 1999b, Mohr 2001b).

Literature on other emotional problems in people with MS is scarce. Although it is recognised that in addition to depression, emotional responses such as anxiety and anger can occur, much less research has been undertaken in these aspects of psychological functioning. Existing research suggests that the point prevalence of problems with anxiety ranges from 19% to 34% (Minden 1991; Pepper 1993; Stenager 1994). A study by Feinstein (Feinstein 1999) found anxiety to be more common than depression. Co‐morbid anxiety and depression was associated with elevated rates of suicidal ideation compared to depressed patients with little or no anxiety (Feinstein 1999).

COGNITIVE FACTORS

Cognitive factors are those relating to the mental processes of memory, concentration, reasoning and judgement. Cognitive problems are common in individuals with MS and point prevalence estimates range from 43%‐72% (Prosiegal 1993). It has been acknowledged that the types of patients studied, duration of illness and disease courses probably account for the wide range of estimates (Nelson 1988; Rao 1991). Community based populations and those with relapse‐remitting MS have lower rates (Rao 1991; Heaton 1985), whilst those attending hospital, the highest. Research has shown that memory, learning, conceptual reasoning, speed of information processing and reaction time, attention, concentration and executive function are affected whereas recognition memory, implicit learning and speech comprehension remain intact (see Brassington 1998). The degree of cognitive impairment evident in individuals with MS appears to be unrelated to their neurological disability status or disease duration (Maurelli 1992; Penman 1991; Rao 1991). Cognitive impairment has been found to be related to poorer social and employment outcomes (Rao 1991), low mood (Gilchrist 1994), sexual dysfunction, greater functional impairment (Amato 1995) and psychopathology. These findings suggest that cognitive dysfunction is a major factor in determining the quality of life of people with MS. Severe cognitive impairment presents a major barrier to rehabilitation programmes because individuals may be unable to retain advice or have difficulty in acquiring new skills. Cognitive impairment in MS often goes undetected or is misattributed to other problems (Lincoln 2002). Rao et al. (Rao 1991) reported that families and carers often attribute cognitive impairments to depression or other forms of psychological dysfunction.

PSYCHOLOGICAL INTERVENTIONS

The purpose of this research is to undertake a systematic review of randomised controlled trials of psychological interventions for adults with MS. Psychological interventions are broadly defined and include those that address mood and those that address cognition.

The literature suggests that people with MS are not adequately treated for their mood disorders (Minden 1987). Effective treatment of mood disorders might improve functional status, self‐esteem, quality of life and compliance with medical treatment (Spitzer 1995; Mohr 1997b). Psychological interventions may improve the psychological and physical well being of individuals with MS by treating mood disorders such as anxiety and depression (with possible benefits to immune function (Mohr 2001b)), by improving self management and adherence, enhancing self efficacy and esteem, reducing stress, improving coping skills and general quality of life. Psychological therapy on an individual basis may help individuals develop skills to cope with emotions, thoughts and adjustment to MS diagnosis and symptoms (Minden 1992). Group therapy is often used to decrease feelings of alienation, facilitate expression of emotions related to the disease, and provide peer support.

Undertaking cognitive assessments may help individuals restructure aspects of their lives to maximise their cognitive strengths. This could lead to a decrease in the functional impact of the disease (Langdon 1996), help in the planning of other rehabilitation services, and decrease dependency. Evaluations of specific interventions for cognitive problems have demonstrated beneficial effects in other patient groups such as stroke and traumatic brain injury (Wilson 1999). However, to date, little research has been undertaken to assess the benefits of cognitive rehabilitation in individuals with MS.

There have been a number of intervention studies to examine the effectiveness of psychological interventions in individuals with MS (Mohr 1999c). A variety of psychological interventions have been used including group therapy (Barnes 1954; Bolding 1960; Day 1953; Hartings 1976), individual cognitive behavioural based therapies (Crawford 1987; Foley 1987; Larcombe 1984), cognitive behavioural therapy delivered via the telephone (Mohr 2000), insight‐oriented group psychotherapy (Crawford 1985), coping skills training compared with peer telephone support (Schwartz 1999), relaxation and imagery (Maguire 1996), and cognitive rehabilitation (Lincoln 2002). A recent study by Mohr et al. compared individual cognitive‐behaviour therapy, with supportive expressive group therapy and the anti‐depressant sertraline (Mohr 2001c). Some psychological interventions have focussed on individuals with MS and depression (e.g. Larcombe 1984), some on individuals with cognitive deficits (Benedict 2000), some have been administered on a one‐to‐one basis (Foley 1987), some in group settings (e.g. Crawford 1985), and some over the telephone (Mohr 2000).

Before we started this review a number of methodological challenges were evident. Firstly, in the context of all randomised controlled trials in MS, the measurement of disease and function can be problematic, particularly when the condition has a relapse‐remitting course and follow‐up periods are short. Secondly psychological interventions may be given at different times, for different purposes, to different subgroups of people, at different stages of their disease. The type of intervention, content, theoretical basis, intensity, duration, length of each session, whether one‐to‐one or in groups can vary, as can the profession and experience of the person delivering the intervention, and the location. This heterogeneity could make it difficult to combine the results from different studies. Thirdly, both in clinical practice and in randomised trials, psychological interventions often incorporate a mix of different components, for example blending cognitive rehabilitation components, say, with psychotherapy components. This overlap makes it difficult to do a series of separate Cochrane reviews that adequately covers and summarises the randomised trial evidence base. Fourthly, in trials, psychological interventions might be compared to other psychological interventions, pharmaceutical treatment, a placebo, or usual care. Each addresses different research questions and so implies a different interpretation of the results. Our solution to the latter three problems has been to conduct a review that is broad in scope, but that contains within it a series of mini‐reviews that focus on the different interventions, comparison treatments and population target groups. Meta‐analysis, if carried out, has been confined to within each mini‐review. This provides a flexible framework that easily accommodates new evidence as it arises.

Particular attention has been paid to giving the review a clear, logical structure and layout. This should enable the reader of the review, despite the complexity outlined above, to find their way to the evidence that most interests them (e.g. long‐term benefits of cognitive behavioural therapy in the management of depression). It also helps to highlight where there are gaps in the evidence base. For the authors, the structure will facilitate standardised updating of the review in future years, particularly if the rate of publication of relevant studies continues to increase.

Objectives

To assess the effect of psychological interventions on mental and physical wellbeing in people with multiple sclerosis

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials and the pre‐ cross over component of randomised cross‐over trials. Studies were not excluded on the basis of type of control group (i.e. care as usual or standard care, placebo, waiting list controls, other psychological intervention, other intervention).

Types of participants

People with a diagnosis of multiple sclerosis (e.g. using recognised criteria such as Poser 1983, Schumacher 1965). Studies were included in the review regardless of clinical course or the length of time since diagnosis of MS. Studies based on subgroups of people with MS such as those with depression, those with cognitive impairment, those with severe physical disability, were included, and were considered in separate mini‐reviews. If we had found studies that included people with MS together with people with other medical conditions, and the number of people in the study with MS had made it worthwhile to do so (i.e. exceeded 20), the authors would have been contacted to try and get the results for the MS subgroup.

Types of interventions

Interventions described as wholly or mostly based on psychological theory and practice. Interventions could have been delivered by psychologists, counsellors, medical staff, nurses, occupational therapists or other health professionals, and were included regardless of the number of therapy sessions or whether sessions were one‐to‐one or group. The following gives an idea of the types of interventions that either were included or could have been included:

Psychoanalytic therapy ‐ looking back at past events in order to gain insight   Behavioural therapy ‐ a variety of techniques that attempt to modify behaviour, including stress management and relaxation.   Cognitive therapy ‐ a variety of techniques that attempt to change attitudes, perceptions and ways of thinking.   Educational ‐ teaching of psychological theory in relation to the disease   Counselling ‐ Non‐directive approach involving talking through problems   Cognitive rehabilitation ‐ a variety of techniques that attempt to reduce the impact of cognitive impairments   Other ‐ e.g. family therapy etc.

Sometimes the interventions evaluated consisted of a combination of these techniques. Interventions that were substantively different have been considered under separate sub‐headings within each mini‐review.

Types of outcome measures

Primary:

1. Disease specific quality of life ‐ perspective of person affected by MS (e.g. MS Impact Scale (MSIS‐29) (Hobart 2001), MSQOL‐54 (Vickrey 1995), MS Quality of Life Inventory (MSQLI ) (LaRocca 1996), Health Related Quality of Life Questionnaire for Multiple Sclerosis (HRQOL‐MS) (Pfennings 1999)).

2. General quality of life ‐ perspective of person affected by MS (e.g. Medical Outcomes Study 36‐item Short‐Form Health Survey (SF‐36) (Ware 1993), Euro‐Qol (EQ5D) (EuroQoL Group 1990)).

3. Psychiatric symptoms including anxiety and depression (e.g. anxiety and depression subscales of the Delusion‐States‐Symptoms‐Inventory (Bedford 1978)) ‐ rated by person with MS or by clinician.

4. Psychological functioning including emotions (e.g. Profile of Mood States McNair 1981), self‐efficacy (e.g. Schwarzer 1995), self‐esteem (e.g. Rosenberg 1965), etc.) ‐ rated by person with MS.

5. Measures of disability (e.g. Kurtzke Expanded Disability Status Scale (EDSS) (Kurtzke 1983), Guy's Neurological Disability Scale (GNDS) (Sharrack 1999) ‐ clinician rated.

6. Cognitive outcomes including memory, language, concentration, higher executive function etc. ‐ rated by clinician or by person with MS.

Given the broad range of interventions encompassed within this review it was important to include a broad range of primary outcome measures. Clearly though, not all primary outcome measures have the same relevance for all interventions.

Secondary:

Relapses/exacerbations (e.g. number of relapses, time to first relapse)   Pain   Fatigue   Health care utilisation, and economic assessment   Change in need for medication   Adherence to other therapies (e.g. drug therapies)

Outcomes were classified according to whether they had been measured (i) within a month post‐treatment ("immediately post‐treatment"), (ii) between one and six months post‐treatment ("short‐term follow‐up"), (iii) between seven and 12 months post‐treatment ("medium‐term follow‐up"), and (iv) over 12 months post‐treatment ("long‐term follow‐up").

Search methods for identification of studies

To identify appropriate studies that included people with multiple sclerosis the search strategy developed by the Cochrane MS Group was used.

In the next update of the review we intend to add in additional search terms relevant to identifying studies of cognitive rehabilitation.

For research not published in English, we either sought help with translation or contacted the respective author.

Electronic searches

1. the Cochrane Multiple Sclerosis Group Specialised Register (December 2004)   2. the Cochrane Central Register of Controlled Trials (The Cochrane Library issue 4, 2004)(Appendix 1)   3. MEDLINE (January 1966 to December 2004)(Appendix 2)   4. EMBASE (January 1974 to December 2004)(Appendix 3)   5. PsychINFO (January 1887 to December 2004)   6. CINAHL (January 1982 to December 2004)   7. BNID (January 1994 to December 2004)   8. AMED (January 1985 to December 2004)   9. CareData (January 1920 to December 2004)   10. ASSIA (January 1987 to December 2004)   11. IBSS (January 1951 to December 2004)   12. Web of Science (January 1981 to December 2004)   13. CAB (January 1973 to December 2004)   14. PapersFirst (OCLC) (January 1993 to December 2004)   15. ProceedingsFirst (OCLC) (January 1993 to December 2004)   16. ASLIP (January 1970 to December 2004)   17. Zetoc (January 1993 to December 2004)   18. EBSCO HOST Academic Search Elite (January 1985 to December 2004)   19. LISA (January 1969 to December 2004)

Searching other resources

The reference lists of identified papers were searched for further relevant trials. In June 2004 we wrote to those who had already undertaken trials in this area to identify other published or unpublished trials. We searched registers of trials in progress (National Research Register, www.controlled‐trials.com and www.clinicaltrials.gov) and contacted the primary investigator.

Data collection and analysis

The review was carried out by four authors (one with methodological expertise, one with psychological expertise, one with clinical expertise in multiple sclerosis, and one with expertise in nursing research). To help get a broader view of the context, design, conduct and interpretation of the review, the authors were aided by an advisory group. The advisory group included the authors and a clinical psychologist/ researcher, a clinical neurophysiologist/ researcher, a senior librarian, and 2 consumers.

Abstracts of identified studies were sifted by one author to identify those that seemed pertinent: at this stage we excluded studies that were obviously not controlled trials or randomised controlled trials, those that had not included people with MS, or assessed interventions of a psychological nature. The full papers of these (including those where there was any uncertainty) were then compared to the above criteria by the four authors. Where there was disagreement between the authors or uncertainty, this was resolved by discussion followed by either consensus or a majority vote. If disagreement or uncertainty remained then the authors consulted other members of the advisory group. Authors would have been contacted if clarification or further information had been needed in order to reach a decision.

Each study was allocated to a mini‐review based on the target population. Different types of intervention were then addressed in sub‐sections within each mini‐review. This structure was chosen instead of one basing each mini‐review on a type of intervention, because of difficulties in classifying some interventions, the eclectic nature of some interventions, and the overall diversity of interventions. Currently the mini‐reviews are:

Mini‐review 1: People with MS with cognitive impairment   Mini‐review 2: People with MS with moderate to severe disability   Mini‐review 3: People with MS   Mini‐review 4: People with MS with depression

The structure of the review permits other mini‐reviews to be added as new research evidence becomes available. Other possibilities for target populations include, people with MS with fatigue, people with MS with pain, people with MS with anxiety, and so on.

Study quality

The main measure of trial quality was whether random allocation had been adequately concealed. Estimates of treatment effect have been shown to be inflated when studies with poor allocation concealment are included in meta‐analysis (Moher 1998). Allocation concealment was rated as described in the Cochrane Reviewers' Handbook (Clarke 2000); A (adequate: e.g. central or independent randomisation, opaque sealed envelopes), B (unclear: method not described or described but not clear), C (inadequate: open random number lists, alternation, date of birth or any other procedure that is transparent before allocation), or D (not done).

Other components of study quality were also assessed and recorded:

The method of randomisation generation; A (adequate, such as random number table, computer generated random numbers), B (partial, such as sealed envelopes but no further details given), C (inadequate, such as birth date, alternation) or D (unknown, with just the term randomised used).
The number of participants withdrawing or dropping out of each arm of the study, together with a quality score; A (adequate with number randomised and withdrawn from each group stated along with reasons), B (partial, numbers given but not reasons, or vice versa), C (inadequate, with numbers randomised not stated or not specified in each group), or D (no details given)

The nature of the interventions being considered made it unlikely that the researcher or clinician who had given the intervention could be masked. Masking of the researcher who made the assessments or, of the person with MS, was possible although difficult. Many of the primary outcome measures had been completed by the person with MS recording their own perspectives and so assessor masking could not have been achieved unless the person with MS had been masked. It may have been possible to mask the people undertaking the data analysis and interpretation of results.

Data extraction     Data for the review were extracted independently by the four authors with disagreement and uncertainty dealt with as before. The following information was extracted using a pre‐agreed data extraction tool:

Date, country and clinical setting of trial.   Description of target population   Recruitment procedures, inclusion and exclusion criteria   Flow chart showing flow of participants through all stages of the study   Participant characteristics (age, gender, years since diagnosis, type of MS, degree of disability, and psychiatric diagnoses).   Description of intervention, duration, frequency, how delivered, who delivered, format of delivery, training of person delivering, whether adapted for MS, and whether concomitant interventions were given.   Type of control group (s) and, if appropriate, description of the duration, frequency, how delivered, who delivered, format of delivery, training of person delivering, whether adapted for MS, and whether concomitant interventions were given.

Comparability of baseline characteristics between treatment and control groups.   Description of follow‐up   Outcomes measured, whether primary or secondary, and when they were recorded   Number enrolled in trial and in each group   Presence of sample size calculation   Number included at each follow‐up in each group   Attempts at masking   Description of randomisation and randomisation concealment   Number and reasons for drop‐out and withdrawal in each group   Whether intention‐to‐treat analysis undertaken   For nominal outcomes (denominator and numerator in each category for each group)   For interval and ordinal data (N, mean, SD for each group) or (N, median, IQR or range) as appropriate.

If both intention‐to‐treat and "per protocol" analyses had been conducted only the former data were extracted.

Statistical analysis

Meta‐analysis was only conducted within mini‐reviews and only when it seemed logical to do so; that is, when groups, interventions, control groups, outcomes and length of follow‐up were sufficiently similar. When meta‐analysis was appropriate, a test of heterogeneity was used to assess whether the results from the different studies were (statistically) sufficiently similar to combine them.

For outcomes measured on a dichotomous scale, results are presented as odds ratios and as risk differences, homogeneity was assessed using a chi‐squared test, and odds ratios/ risk differences combined using the Mantel‐Haenszel procedure.

For outcomes measured on an interval or ordinal scale, results are presented as means and standard deviations, a chi‐squared test was used to assess homogeneity and results from different studies combined using weighted mean differences. Standardised effect sizes using Hedges' adjusted have been used to summarise mean differences between groups. This enables comparison of effect sizes between different outcome measures. Unless otherwise indicated, positive standardised effect sizes indicate that the intervention is producing more benefit to participants than the comparison group.

If two or more studies measured outcome in the same dimension but using a different tool then we considered standardising the scores (using the pooled within group standard deviation) and then combining the studies as above.

By default, a fixed‐effects model was used for combining studies. If tests for heterogeneity were statistically significant and inspection of the individual results suggested that it still seemed logical to combine results, then a random effects model was used. Statistical precision was summarised using 95% confidence intervals.

Within a mini‐review no subgroup analyses were done.

Had the results from three or more studies within a mini‐review been combined, then a sensitivity analysis would have been considered. This would have taken two forms (a) Sequential elimination of trials from the analysis according to their level of allocation concealment (i.e. type D studies eliminated first, then type C studies, then type B) to see how the results were affected; (b) data reanalysed using worst case, best case, and random case scenarios for participants without outcome data to assess the robustness of the results.

The review was conducted using RevMan 4.2 software

Results

Description of studies

The search strategy identified 26 relevant trials, of which six were excluded after reading the full papers (see table of "Characteristics of excluded studies"), two, identified between January and May 2005, are awaiting assessment (see reference list of studies waiting assessment) and two are ongoing (see table of "Characteristics of ongoing studies"). The remaining 16 trials, reported in 17 papers (one trial was reported over two papers (Mohr 2001, Mohr 2003)), are all randomised controlled trials, are included in the review, and are described in detail in the table of "Characteristics of included studies". A further paper (Mohr 2004), listed in the table of "Characteristics of excluded studies", contained additional data for a study that was included in the review (Mohr 2001), but none of the outcomes reported satisfied the review's inclusion criteria. One of the six excluded studies appeared not to come within our definition of a psychological intervention, one was excluded because it did not include appropriate outcome measures, and four were excluded because they did not appear to be randomised. The 16 included studies have been grouped together in four Mini‐reviews. Only two of the studies in one of the mini‐reviews were sufficiently similar in terms of type of participant, type of intervention, type of comparison group, and outcome measure for a meta‐analysis to be conducted. The number of participants per study ranged from 15 to 240, median 37, with a total of 1006 over all 16 studies. Twelve studies had a two group design, and four studies a three group design (Crawford 1985, Lincoln 2002, Mohr 2001 (although one group was excluded because it wasn't randomised), Rigby 2003). Length of follow‐up from start of treatment ranged from five weeks to four years.

One study used a measure of MS specific quality of life, two measured general quality of life, 12 measured psychiatric symptoms, 10 measured psychological function, two measured disability, five measured cognitive outcomes, three studies measured fatigue, one study measured pain, one measured an aspect of health care utilisation (none undertook an economic assessment), one measured changes in medication usage, and one measured adherence to medication. None measured MS relapses. Fifteen studies measured outcome immediately post‐treatment, nine measured short‐term outcome, three measured medium term outcome and one measured long‐term outcome.

Risk of bias in included studies

Study quality is given in the "Methods" column in the table of "Characteristics of included studies" in the order (i) method of randomisation generation, (ii) method of randomisation concealment, and (iii) description of withdrawals.

Method of randomisation generation was not reported for 11 studies, adequate in three studies and partially adequate in two studies.

Allocation concealment was adequate in three studies, unclear in 12 studies and inadequate in one study.

Reporting of withdrawals was adequate in five studies, partial in seven, inadequate in three and not reported in one study.

Effects of interventions

MINI‐REVIEW 1: People with MS with cognitive impairment     (a) Neuropsychological counseling

(i) Neuropsychological counseling versus Psychotherapy     One study Benedict 2000; a two group study of 15 people with marked cognitive impairment and behaviour disorder (and their caregivers), comparing individualised neurological compensatory training (enhancing understanding of cognitive impairment, social skills training, and cognitive behavioural therapy to help regulate social behaviour) to non‐specific supportive psychotherapy.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:     Immediately post‐treatment:‐ In the Benedict study (Benedict 2000) the mean change in self‐reported depression between pre‐treatment and immediately post‐treatment did not differ between treatment arms (p>0.05, no post‐treatment means reported). However overall sample size was small (n=15).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(b) Cognitive rehabilitation+neuropsychotherapy

(i) Cognitive rehabilitation+neuropsychotherapy versus placebo

One study (Jonsson 1993); a two group study of 48 people comparing individual cognitive rehabilitation (compensation, substitution and training) and goal directed neuropsychotherapy (based on personal problems and test profile) to a placebo individual intervention.

Primary outcomes:

Disease specific quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:     Immediately post‐treatment:‐In the study of Jonsson et al (Jonsson 1993) the improvement in depression between pre‐treatment and immediately post‐treatment was significantly higher in the group receiving cognitive rehabilitation and neuropsychology (mean improvement 2.4 versus 0.0, p=0.04, no further details given). There were no statistically significant differences between the intervention and placebo groups in current anxiety (mean improvement between pre‐treatment to immediately post‐treatment 5.6 v 2.7 respectively, p=0.17, no other details given) or general anxiety (mean improvement 3.8 v 3.5 respectively, p=0.92, no other details given).

Short term follow‐up:‐ In the study of Jonsson et al (Jonsson 1993) the improvement in depression between pre‐treatment and short term follow‐up was significantly higher in the group receiving cognitive rehabilitation and neuropsychology (mean improvement 1.1 versus ‐2.7, p=0.03, no further details given). There were no statistically significant differences between the intervention and placebo groups in current anxiety (mean improvement between pre‐treatment and short‐term follow‐up 1.1 v ‐1.6 respectively, p=0.42, no other details given) or general anxiety (mean improvement 1.5 v 0.6 respectively, p=0.75, no other details given).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:     Immediately post‐treatment:‐The study of Jonsson et al (Jonsson 1993) used 20 measures of cognition combined into 11 factors for the purpose of analysis (see Table of included studies). All factors have been converted into age, sex and education status adjusted t‐scores based on data from a sample of healthy people without MS where mean t‐score = 50 and SD = 10. One of these 11 factors is a composite of four factors (memory span, verbal learning, visuo‐spatial memory and visuo‐motor speed). This composite factor showed no difference in improvement between pre‐treatment and immediately post‐treatment in the intervention group compared to the control group (mean improvement 1.8 v 1.1 respectively, p=0.53, no other details given). There were no significant differences in the four factors that made up this composite factor. There was a larger improvement in visual perception in the intervention group compared to the control group (mean improvement 2.0 v 0.6 respectively, p=0.04, no other details given), but not in verbal intelligence (mean improvement 1.0 v 0.7 respectively, p=0.91, no other details given), or WAIS performance (mean improvement 5.2 v 7.3 respectively, p=0.42). No results reported for recognition memory, concentration or verbal fluency.

Short term follow‐up:‐The study of Jonsson et al (Jonsson 1993) used 20 measures of cognition combined into 11 factors (see Table of included studies). All factors have been converted into age, sex and education status adjusted t‐scores based on data from a sample of healthy people without MS where mean t‐score = 50 and SD = 10. One of these 11 factors is a composite of four other factors (memory span, verbal learning, visuo‐spatial memory and visuo‐motor speed). This composite factor showed no difference in improvement between pre‐treatment and immediately post‐treatment in the intervention group compared to the control group (mean improvement 1.6 v ‐0.5 respectively, p=0.09, no other details given). Comparing the four factors which make up this overall measure, one was of borderline statistical significance (mean improvement in visuo‐spatial memory from pre‐treatment to short‐term follow‐up 2.7 v 0.2, p=0.05, no further details given). There was no difference between the intervention and control groups in visual perception (mean improvement 2.2 v 1.0 respectively, p=0.09, no other details given), verbal intelligence (mean improvement 1.5 v 2.1 respectively, p=0.81, no other details given), or WAIS performance (mean improvement 3.7 v 4.2 respectively, p=0.87). No results reported for recognition memory, concentration or verbal fluency.

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(c) Cognitive rehabilitation

(i) Cognitive rehabilitation versus placebo     One study (Solari 2004); a two group study of 82 people comparing computer aided retraining of memory and attention to sham retraining.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐ In the study of Solari et al (Solari 2004) data on mental health and cognitive subscales of MS specific quality of life were reported and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. The percentage improvement in the mental health subscale was significantly better in the cognitive rehabilitation group than the placebo group (mean (SD) = 27% (73%) v 9% (41%) respectively, p=0.04), but there was no significant difference for the cognitive subscale (36% (86%) v 43% (123%) respectively, p=0.51).

Short term follow‐up:‐In the study of Solari et al (Solari 2004) data on mental health and cognitive subscales of MS specific quality of life were reported and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. The percentage improvement in the mental health subscale was not significantly different in the cognitive rehabilitation group than the placebo group (mean (SD) = 16% (47%) v 23% (69%) respectively, p=0.84), and neither was there a significant difference for the cognitive subscale (43% (126%) v 56% (140%) respectively, p=0.59).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐In the study of Solari et al (Solari 2004) data on the mood subscale of the mood depression inventory were reported and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. The percentage improvement in this scale did not differ significantly between the cognitive rehabilitation and placebo groups (mean (SD) = 2% (19%) v 0% (20%) respectively, p=0.67).

Short term follow‐up:‐In the study of Solari et al (Solari 2004) data on the mood subscale of the mood depression inventory were reported and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. The percentage improvement in this scale did not differ significantly between the cognitive rehabilitation and placebo groups (mean (SD) = 6% (19%) v 5% (21%) respectively, p=0.87).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐In the study of Solari et al (Solari 2004) data on the seven subscales of the Brief Repeatable Battery of Neuropsychological Tests were presented and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. No significant differences were found between cognitive rehabilitation and placebo groups for Buschke Selective Reminding (verbal learning and recall) ‐ consistent long term recall (mean (SD) = 138% (310%) v 92% (218%) respectively, p=0.54), Buschke Selective Reminding ‐ delayed recall (mean (SD) = 160% (317%) v 25% (68%) respectively, p=0.78), Symbol Digit Modalities Test (sustained attention and information processing speed) (mean (SD) = 13% (25%) v 9% (29%) respectively, p=0.37), Paced Auditory Serial Addition Test (complex attention and concentration) (mean (SD) = 24% (56%) v 33% (85%) respectively, p=0.97), 10/36 Spatial Recall Test (visuospatial learning) ‐ immediate recall (24% (47%) v 28% (69%), respectively, p=0.69), and 10/36 spatial recall ‐ delayed recall (20% (48%) v 81% (146%) respectively, p=0.33). A significant difference was found for the Word List Generation Test (verbal fluency and sustained attention) with greater improvement in the cognitive rehabilitation group compared to the placebo group (mean (SD) = 6% (45%) v ‐17% (27%) respectively, p=0.02. Given the large number of cognitive tests presented, the possibility that this is a type 1 error cannot be discounted. The proportion of patients in the cognitive rehabilitation group and the placebo group who showed at least a 20% improvement in two or more of these measures was 45% and 43% respectively (p=0.88).

Short term follow‐up:‐In the study of Solari et al (Solari 2004) data on the seven subscales of the Brief Repeatable Battery of Neuropsychological Tests were presented and analysed as percentage changes from baseline, and statistical analyses adjusted for baseline measures, age, education, and study centre. No significant differences were found between cognitive rehabilitation and placebo groups for Buschke Selective Reminding ‐ consistent long term recall (mean (SD) = 160% (317%) v 143% (288%) respectively, p=0.67), Buschke Selective Reminding (verbal learning and recall) ‐ delayed recall (mean (SD) = 10% (40%) v 44% (99%) respectively, p=0.06), Symbol Digit Modalities Test (sustained attention and information processing speed) (mean (SD) = 15% (27%) v 17% (37%) respectively, p=0.57), Paced Auditory Serial Addition Test (complex attention and concentration) (mean (SD) = 16% (50%) v 39% (103%) respectively, p=0.19), 10/36 spatial recall (visuospatial learning) ‐ immediate recall (17% (54%) v 27% (68%), respectively, p=0.68), and 10/36 spatial recall ‐ delayed recall (12% (64%) v 77% (152%) respectively, p=0.30). A significant difference was found for the Word List Generation Test (verbal fluency and sustained attention), with greater improvement in the cognitive rehabilitation group compared to the placebo group (mean (SD) = 32% (47%) v 0% (30%) respectively, p=0.03. Given the large number of cognitive tests presented, the possibility that this is a type 1 error cannot be discounted. The proportion of patients in the cognitive rehabilitation group and the placebo group who showed at least a 20% improvement in two or more of these measures was 48% and 54% respectively (p=0.57).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

MINI‐REVIEW 2: People with MS with moderate to severe disability     (a) Psychotherapy

(i) Psychotherapy versus placebo

One study Crawford 1985; a three group study of 32 people comparing group‐based insight‐oriented psychotherapy to a "dummy" group‐based therapy and a care as usual group.

Primary outcomes:     Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ In the Crawford study (Crawford 1985) the group receiving psychotherapy scored significantly lower on the depression score than the placebo group (mean post‐treatment score 19.3 v 23.5 respectively, p<0.05 adjusted for baseline depression, no further details). There was no statistically significant difference in anxiety (p>0.05, no further details given).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:

Immediately post‐treatment:‐ In the Crawford study (Crawford 1985) there were no statistically significant differences in locus of control (mean post‐treatment score 28.3 in the psychotherapy group, 30.7 in the placebo group, p>0.05, no other details reported), or self‐esteem (p>0.05, no other details reported).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(ii) Psychotherapy versus care as usual     One study Crawford 1985; a three group study of 32 people comparing group based insight oriented psychotherapy to a "dummy" group based therapy and a care as usual group.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ In the Crawford study (Crawford 1985) the group receiving psychotherapy scored significantly lower on the depression score than the care as usual group (mean score 19.3 v 23.5 respectively, p<0.05 adjusted for baseline depression, no further details). There was no statistically significant difference in anxiety (p>0.05, no further details given).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:     Immediately post‐treatment:‐ In the Crawford study (Crawford 1985) the group receiving psychotherapy were more internally oriented post‐treatment than the care as usual group (mean locus of control 28.3 v 37.0 respectively, p<0.05 adjusted for baseline locus of control, no other details reported). There were no differences in self‐esteem (p>0.05, no other details reported).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(b) Social skills training

(i) Social skills training versus care as usual     One study (Gordon 1997); a two group study with 26 people comparing social skills training aimed at easing interaction strain between people with and without disability, to care as usual.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ In the Gordon et al. study (Gordon 1997) there were no statistically significant differences between intervention and control groups in mean (SD) self efficacy (79.4 (15.0) v 72.1 (28.8) respectively, standardised mean difference = 0.31 (95% CI; ‐0.54, 1.15)), self esteem (352 (16) v 337 (36) respectively, standardised mean difference = 0.52 (95% CI; ‐0.33, 1.37)), social distress (24.6 (8.3) v 26.5 (9.3) respectively, standardised mean difference = 0.21 (95% CI; ‐0.63, 1.05)), social avoidance (24.0 (8.4) v 24.3 (8.6) respectively, standardised mean difference = 0.03 (95% CI; ‐0.80, 0.87)), social anxiety using positive thoughts (57.8 (11.3) v 45.5 (17.7) respectively, standardised mean difference = 0.80 (95% CI; ‐0.08, 1.67)) or social anxiety using negative thoughts (25.8 (6.2) v 28.9 (7.9) respectively, standardised mean difference = 0.42 (95% CI; ‐0.43, 1.27)). It should be noted that in this study the authors found that baseline levels of social distress and self‐esteem were reasonably normal, and that this may be one explanation for the apparent lack of treatment effect.

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(c) Cognitive assessment and rehabilitation

(i) Cognitive assessment and rehabilitation versus care as usual     One study (Mendoza 2001); 20 people in a long‐term nursing facility, comparing a cognitive remediation strategy involving daily interviews with nursing staff and a memory notebook, to care as usual. .     Primary outcomes:

Disease specific quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:     Immediately post‐treatment:‐ Mendoza et al. (Mendoza 2001) found mean depression to be 5.5 in the intervention group and 8.6 in the control group (no further information reported).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:     Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:     Immediately post‐treatment:‐Mendoza et al. (Mendoza 2001) measured a number of cognitive outcomes (mental status, adult reading, and verbal learning). Unfortunately no data are presented.

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

MINI‐REVIEW 3: People with MS     (a) Cognitive behavioural therapy

(i) Cognitive behavioural therapy versus minimal psychotherapy

One study (Foley 1987) of 41 people comparing individual stress inoculation training (cognitive behavioural therapy and relaxation) with care as usual that incorporated two hours of supportive psychotherapy.

Primary outcomes:

Disease specific quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ In the study of Foley et al. (Foley 1987) mean depression was significantly lower in the intervention group than the control group (post‐treatment mean 13.2 (10.5) v 21.6 (14.2) respectively, standardised mean difference unadjusted for baseline depression = 0.66 (95% CI; ‐0.02, 1.33), p‐value after taking into account baseline depression < 0.001), as was general (trait) anxiety (post‐treatment means 37.2 (13.8) v 50.5 (13.0) respectively, standardised mean difference unadjusted for baseline anxiety = 0.97 (95% CI; 0.27, 1.67), p‐value after taking into account baseline anxiety = 0.015), and perceived distress (post‐treatment means 57.5 (37.6) v 89.2 (67.1) respectively, standardised mean difference not taking into account baseline distress = 0.57 (‐0.10, 1.24), p‐value after taking into account baseline distress = 0.02) . The difference between the groups in current (state) anxiety was not statistically significant (mean post‐treatment scores 46.2 (13.1) in intervention group, 51.9 (13.4) in control group, mean standardised difference not taking into account baseline anxiety = 0.42 (95% CI; ‐0.24, 1.08), p‐value after taking into account baseline anxiety = 0.08).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:

Immediately post‐treatment:‐ In the study of Foley et al. (Foley 1987) individuals in the intervention group tended to score more highly on the problem‐focussed coping score (mean post‐treatment score 16.2 (4.8) than in the intervention group, 11.8 (4.6) in the control group, mean standardised difference without taking into account pre‐treatment coping score = 0.92 (95% CI; 0.22, 1.61), p‐value after taking into account baseline coping = 0.01). There were no differences in average locus of control (p>0.05, no other details reported)

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(ii) Cognitive behavioural therapy versus placebo control     One study, Rigby 2003; a three group study of 147 people comparing group‐based coping‐focussed cognitive therapy, to a dummy social discussion group, and a care as usual group (all groups also received an information booklet).

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving placebo (5.5 (5.3) v 6.0 (4.0) respectively, p>0.05), and no difference in median (IQR) depression ( 5.0 (5.3) v 5.0 (5.0) respectively, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either anxiety or depression.

Short term follow‐up:‐The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving placebo (7.0 (7.0) v 6.0 (5.0) respectively at one month post‐treatment, p>0.05; 9.0 (6.5) v 5.0 (6.0) respectively at three months, p>0.05; 7.0 (7.0) v 6.5 (6.0) respectively at six months, p>0.05), and no difference in median (IQR) depression (5.0 (6.0) v 4.0 (5.0) respectively at one month post‐treatment, p>0.05; 6.0 (5.5) v 5.0 (5.0) respectively at three months, p>0.05; 5.0 (5.0) v 5.0 (7.3) respectively at six months, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either anxiety or depression.

Medium term follow‐up:‐The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving placebo (6.0 (7.0) v 5.0 (5.0) respectively, p>0.05), and no difference in median (IQR) depression (6.5 (6.3) v 4.0 (4.5) respectively, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either anxiety or depression.

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving placebo (median (IQR) for MSSS = 50.0 (14.0) v 50.0 (15.0) respectively, p>0.05; median (IQR) for MSSE (control subscale) = 530 (253) v 590 (280), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either self‐efficacy measure (MSSS and MSSE). She found no difference in total dispositional resiliency between the CBT and placebo groups, nor in two (Commitment and Control) of the three subscales of dispositional resiliency (median (IQR) total score = 86.0 (21.3) v 88.0 (16.3), p>0.05; median (IQR) Commitment score = 31.5 (10.0) v 34.0 (7.0), p>0.05; median (IQR) Control score = 31.0 (9.0) v 32.0 (5.0), p>0.05). The Challenge subscale was significantly higher in the group given CBT (median (IQR) = 27.0 (8.0) v 22.0 (7.3), p<0.05, suggesting greater resiliency) than placebo. A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in total dispositional resiliency, nor in the Commitment and Control subscales, but again the group given CBT tended to have better scores on the Challenge subscale.

Short term follow‐up:‐The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving placebo (median (IQR) for MSSS = 52.0 (11.3) v 53.0 (17.0) respectively at one month, 49.0 (13.0) v 51.0 (15.0) at three months, 53.0 (11.0) v 50.0 (16.0) at six months, all p>0.05; median (IQR) for MSSE = 590 (240) v 590 (320) respectively at one month, 510 (223) v 580 (230) at three months, 540 (205) v 580 (228) at six months, all p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either self‐efficacy measure. She found no difference in total dispositional resiliency between the CBT and placebo groups, nor in the three subscales of dispositional resiliency (median (IQR) total score = 89.0 (19.0) v 88.0 (18.0) respectively at one month, 86.0 (20.0) v 91.0 (15.0) at three months, 89.0 (22.5) v 90.0 (18.0) at six months, all p>0.05; median (IQR) Commitment score = 32.0 (10.0) v 32.0 (9.0) at one month, 32.0 (8.0) v 34.0 (9.0) at three months, 32.0 (10.5) v 32.0 (8.0) at six months, all p>0.05; median (IQR) Control score = 32.0 (8.0) v 32.0 (7.0) at one month, 31.0 (7.0) v 33.0 (6.0) at three months, 31.0 (8.3) v 33.0 (5.0) at six months, all p>0.05; median (IQR) Challenge score = 26.0 (6.0) v 24.0 (7.0) respectively at one month, 26.0 (5.0) v 25.0 (7.0) at three months, 26.0 (6.5) v 24.0 (7.0) at six months, all p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in total dispositional resiliency, nor in the Commitment and Control subscales, but the group given CBT tended to have better scores on the Challenge subscale.

Medium term follow‐up:‐The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving placebo (median (IQR) for MSSS = 47.0 (14.3) v 56.0 (11.0) respectively, p>0.05; median (IQR) for MSSE = 510 (200) v 640 (255), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in either self‐efficacy measure. She found no difference in total dispositional resiliency between the CBT and placebo groups, nor in any of the three subscales of dispositional resiliency (median (IQR) total score = 89.0 (20.5) v 90.0 (15.0), p>0.05; median (IQR) Commitment score = 33.0 (10.0) v 33.0 (8.0), p>0.05; median (IQR) Control score = 32.0 (6.5) v 34.0 (5.0), p>0.05; median (IQR) Challenge subscale = 26.0 (7.5) v 24.0 (5.0), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was no significant difference between the two groups in total dispositional resiliency, nor in the Commitment and Control subscales, but the group given CBT tended to have better scores on the Challenge subscale.

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(iii) Cognitive behavioural therapy versus no treatment

Two studies; Rigby 2003 (a three group study of 147 people comparing group‐based coping‐focussed cognitive therapy, to a dummy social discussion group, and a care as usual group) and Wassem 2003 (a two group study of 27 people comparing a programme aimed at increasing self‐efficacy for adjustment (to MS) behaviours, to care as usual).

Primary outcomes:     Disease specific quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving information (5.5 (5.3) v 6.0 (5.0) respectively, p>0.05), and no difference in median (IQR) depression ( 5.0 (5.3) v 3.5 (5.0) respectively, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there were significantly lower levels of anxiety (p<0.01) in the CBT group, but no difference in depression.

Short term follow‐up:‐The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving information (7.0 (7.0) v 7.0 (4.5) respectively at one month post‐treatment, p>0.05; 9.0 (6.5) v 6.0 (4.0) respectively at three months, p>0.05; 7.0 (7.0) v 6.0 (6.3) respectively at six months, p>0.05), and no difference in median (IQR) depression (5.0 (6.0) v 4.5 (3.5) respectively at one month post‐treatment, p>0.05; 6.0 (5.5) v 6.0 (4.0) respectively at three months, p>0.05; 5.0 (5.0) v 5.0 (5.3) respectively at six months, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there were significantly lower levels of anxiety (p<0.01) in the CBT group, but no difference in depression.

Medium term follow‐up:‐The study by Rigby 2003 found no statistically significant differences in median anxiety (IQR) between those receiving CBT and those receiving information (6.0 (7.0) v 6.0 (5.0) respectively, p>0.05), and no difference in median (IQR) depression ( 6.5 (6.3) v 5.0 (4.0) respectively, p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there were significantly lower levels of anxiety (p<0.01) in the CBT group, but no difference in depression.

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving information (median (IQR) for MSSS = 50.0 (14.0) v 49.0 (18.5) respectively, p>0.05; median (IQR) for MSSE = 530 (253) v 520 (260), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was a significant difference in MSSE such that the intervention group had higher self‐efficacy (p<0.01), but no difference in MSSS (p>0.05). She found no difference in total dispositional resiliency between the CBT and information groups, nor in the three subscales of dispositional resiliency (median (IQR) total score = 86.0 (21.3) v 89.0 (13.5), p>0.05; median (IQR) Commitment score = 31.5 (10.0) v 33.5 (6.0), p>0.05; median (IQR) Control score = 31.0 (9.0) v 30.0 (6.0), p>0.05; median (IQR) Challenge score = 27.0 (8.0) v 25.0 (7.0), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was greater dispositional resiliency in the CBT group than the information group, as measured by the total score (p<0.01), Commitment score (p<0.01), and Challenge score (p<0.05), but not the Control score (p>0.05).

Short term follow‐up:‐The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving information (median (IQR) for MSSS = 52.0 (11.3) v 51.5 (14.3) respectively at one month; 49.0 (13.0) v 46.0 (11.0) at three months; 53.0 (11.0) v 48.0 (11.0) at six months, all p>0.05; median (IQR) for MSSE = 590 (240) v 510 (250) respectively at one month; 510 (223) v 560 (210) at three months; 540 (205) v 510 (305) at six months, all p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was a significant difference in MSSE such that the intervention group had higher self‐efficacy (p<0.01), but no difference in MSSS (p>0.05). She found no difference in total dispositional resiliency between the CBT and information groups, nor in the three subscales of dispositional resiliency (median (IQR) total score = 89.0 (19.0) v 92.0 (20.0) respectively at one month, 86.0 (20.0) v 90.5 (12.5) at three months, 89.0 (22.5) v 91.0 (17.0) at six months, all p>0.05; median (IQR) Commitment score = 32.0 (10.0) v 33.0 (7.0) at one month, 32.0 (8.0) v 35.0 (6.0) at three months, 32.0 (10.5) v 33.0 (6.5) at six months, all p>0.05; median (IQR) Control score = 32.0 (8.0) v 32.0 (6.5) at one month, 31.0 (7.0) v 31.0 (6.3) at three months, 31.0 (8.3) v 32.0 (7.0) at six months, all p>0.05; median (IQR) Challenge score = 26.0 (6.0) v 23.0 (7.0) respectively at one month, 26.0 (5.0) v 23.5 (7.5) at three months, 26.0 (6.5) v 24.0 (7.5) at six months, all p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was greater dispositional resiliency in the CBT group than the information group, as measured by the total score (p<0.01), Commitment score (p<0.01), and Challenge score (p<0.05), but not the Control score (p>0.05).

The study by Wassem et al (Wassem 2003) reported on two measures of adjustment to MS (the Psychosocial Adjustment to Illness Scale and the summation of psychological, social and physical adjustment visual analogue scales), and a measure of self‐efficacy (Self Efficacy for Adjustment Behaviours scale) on 10 occasions from baseline to long term follow‐up at four years. The data from baseline and all follow‐ups were analysed together using a repeated measures analysis of covariance model, and the efficacy of the intervention tested using the group by occasion interaction in this model. Effects were adjusted for age at diagnosis and illness severity. The p‐values for the measures, in the same order as that described earlier in the paragraph, are p>0.10, p=0.72, and p=0.55 respectively. No detail is given of means (SD) at short‐term follow‐up. The results for the self‐efficacy measurement could potentially be combined with those from Rigby (Rigby 2003), but insufficient detail was provided to make this possible.

Medium term follow‐up:‐The study by Rigby 2003 found no significant differences in self‐efficacy between those receiving CBT and those receiving information (median (IQR) for MSSS = 47.0 (14.3) v 47.0 (15.0) respectively, p>0.05; median (IQR) for MSSE = 510 (200) v 480 (245), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was a significant difference in MSSE such that the intervention group had higher self‐efficacy (p<0.01), but no difference in MSSS (p>0.05). She found no difference in total dispositional resiliency between the CBT and information groups, nor in any of the three subscales (median (IQR) total score = 89.0 (20.5) v 92.0 (15.0), p>0.05; median (IQR) Commitment score = 33.0 (10.0) v 33.0 (8.0), p>0.05; median (IQR) Control score = 32.0 (6.5) v 32.0 (8.0), p>0.05; median (IQR) Challenge subscale = 26.0 (7.5) v 23.0 (8.0), p>0.05). A complex statistical analysis summarising all baseline and post‐treatment measurements up to 12 months using "area under the curve" suggested that over the complete period of follow‐up there was greater dispositional resiliency in the CBT group than the information group, as measured by the total score (p<0.01), Commitment score (p<0.01), and Challenge score (p<0.05), but not the Control score (p>0.05).

The study by Wassem et al (Wassem 2003) reported on two measures of adjustment to MS (the Psychosocial Adjustment to Illness Scale and the summation of psychological, social and physical adjustment visual analogue scales), and a measure of self‐efficacy (Self Efficacy for Adjustment Behaviours scale) on 10 occasions from baseline to long term follow‐up at four years. The data from baseline and all follow‐ups were analysed together using a repeated measures analysis of covariance model, and the efficacy of the intervention tested using the group by occasion interaction in this model. Effects were adjusted for age at diagnosis and illness severity. The p‐values for the measures, in the same order as described earlier in the paragraph, are p>0.10, p=0.72, and p=0.55 respectively. No detail is given of means (SD) at medium‐term follow‐up. The results for the self‐efficacy measurement could potentially be combined with those from Rigby (Rigby 2003), but insufficient detail is provided to make this possible.

Long term follow‐up:‐The study by Wassem et al (Wassem 2003) reported on two measures of adjustment to MS (the Psychosocial Adjustment to Illness Scale and the summated psychological, social and physical adjustment visual analogue scales), and a measure of self‐efficacy (Self Efficacy for Adjustment Behaviours scale) on 10 occasions from baseline to long term follow‐up at four years. The data from baseline and all follow‐ups were analysed together using a repeated measures analysis of covariance model, and the efficacy of the intervention tested using the group by occasion interaction in this model. Effects were adjusted for age at diagnosis and illness severity. The p‐values for the measures, in the order earlier in the paragraph, are p>0.10, p=0.72, and p=0.55 respectively. No detail is given of means (SD) at long‐term follow‐up.

Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐In the study by Wassem et al (Wassem 2003) fatigue and pain were measured using a visual analogue scale on 10 occasions from baseline to four year follow‐up. Data for each individual follow‐up are not presented, and the authors used a repeated measures analysis of covariance to test the efficacy of the therapy using all 10 occasions simultaneously , adjusting for age at diagnosis and illness severity. The interaction between occasion and treatment arm was not statistically significant for fatigue (p=0.09), nor pain (p>0.10).

Medium term follow‐up:‐In the study by Wassem et al (Wassem 2003) fatigue and pain were measured using a visual analogue scale on 10 occasions from baseline to four year follow‐up. Data for each individual follow‐up are not presented, and the authors used a repeated measures analysis of covariance to test the efficacy of the therapy using all 10 occasions simultaneously , adjusting for age at diagnosis and illness severity. The interaction between occasion and treatment arm was not statistically significant for fatigue (p=0.09), nor pain (p>0.10).

Long term follow‐up:‐In the study by Wassem et al (Wassem 2003) fatigue and pain were measured using a visual analogue scale on 10 occasions from baseline to four year follow‐up. Data for each individual follow‐up are not presented, and the authors used a repeated measures analysis of covariance to test the efficacy of the therapy using all 10 occasions simultaneously , adjusting for age at diagnosis and illness severity. The interaction between occasion and treatment arm was not statistically significant for fatigue (p=0.09), nor pain (p>0.10).

( b) Cognitive assessment and rehabilitation

(i) Cognitive assessment and rehabilitation versus care as usual     One study Lincoln 2002; a three group study of 240 people comparing a group who received no psychological assessment or rehabilitation to a group receiving cognitive assessment, (with feedback on results to patients and health professionals) and a group receiving cognitive assessment and targeted cognitive rehabilitation based on the results of the assessment.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐ No data

Short‐term follow‐up:‐Lincoln et al (Lincoln 2002) found no statistically significant differences in mean physical health score (general quality of life) between the intervention and control groups (31.5 (10.0) v 31.9 (9.4) respectively, standardised mean difference = ‐0.04 (‐0.36, 0.28)) or in mean mental health score (general quality of life) (means 45.4 (11.9) v 46.5 (13.2) respectively, standardised mean difference = ‐0.09 (‐0.41, 0.23)). There were no statistically significant differences in overall quality of life (median (IQR) = 6.0 (4 to 8) in the intervention group, 6.5 (5 to 8) in the control group, p>0.05), nor in satisfaction with quality of life (median (IQR) = 4.0 (3 to 5) in the intervention group, 5.0 (4 to 8) in the control group, p>0.05). In all instances higher scores indicate better quality of life.

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐In the study by Lincoln et al (Lincoln 2002) there was no statistically significant difference in mean mood between the rehabilitation plus assessment group and the care as usual group (mean 27.0 (15.7) v 24.9 (14.7) respectively, standardised mean difference = ‐0.14 (‐0.46, 0.18), p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐Lincoln et al (Lincoln 2002) showed no statistically significant differences between intervention and control groups in dysexecutive functioning (median (IQR) = 18.0 (10 to 29) v 16.5 (9 to 32) respectively, p>0.05), everyday memory (median (IQR) = 15.0 (6 to 32) v 14.0 (7 to 37) respectively, p>0.05), and compensation for memory impairment (median (IQR) = 10.0 (5 to 14) v 10.0 (7 to 14) respectively, p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(ii) Cognitive assessment and rehabilitation versus cognitive assessment     One study Lincoln 2002; a three group study of 240 people comparing a group who received no psychological assessment or rehabilitation to a group receiving cognitive assessment, (with feedback on results to patients and health professionals) and a group receiving cognitive assessment and targeted cognitive rehabilitation based on the results of the assessment.

Primary outcomes:     Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐ Lincoln et al. (Lincoln 2002) found no statistically significant differences in mean physical health score (general quality of life) between the rehabilitation plus assessment and assessment only groups (31.5 (10.0) v 33.0 (11.1) respectively, standardised mean difference = ‐0.14 (‐0.47, 0.18)) or in mean mental health score (general quality of life) (means 45.4 (11.9) v 45.4 (13.7) respectively, standardised mean difference = 0.00 (‐0.33, 0.33)). There were no statistically significant differences in overall quality of life (median (IQR) = 6.0 (4 to 8) in the rehabilitation plus assessment group, 6.0 (4 to 7) in the assessment only group, p>0.05), nor in satisfaction with quality of life (median (IQR) = 4.0 (3 to 5) in the rehabilitation plus assessment group, 4.0 (3 to 5) in the control group, p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychiatric symptoms:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:

Immediately post‐treatment:‐ In the study by Lincoln et al. (Lincoln 2002) there was no statistically significant difference in mood between the rehabilitation plus assessment group and the assessment only group (mean 27.0 (15.7) v 27.3 (15.7) respectively, standardised mean difference = 0.02 (‐0.31, 0.35), p>0.05).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐Lincoln et al. (Lincoln 2002) showed no statistically significant differences between rehabilitation plus assessment and assessment alone groups in dysexecutive functioning (median (IQR) = 18.0 (10 to 29) v 18.0 (7 to 31) respectively, p>0.05), everyday memory (median (IQR) = 15.0 (6 to 32) v 15.0 (5 to 31) respectively, p>0.05), and compensation for memory impairment (median (IQR) = 10.0 (5 to 14) v 9.0 (6 to 15) respectively, p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(c) Cognitive assessment

(i) Cognitive assessment versus care as usual

One study Lincoln 2002; a three group study of 240 people comparing a group who received no psychological assessment or rehabilitation to a group receiving cognitive assessment, (with feedback on results to patients and health professionals) and a group receiving cognitive assessment and targeted cognitive rehabilitation based on the results of the assessments.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐Lincoln et al (Lincoln 2002) found no statistically significant differences in mean physical health score (general quality of life) between the cognitive assessment and care as usual groups (33.0 (11.1) v 31.9 (9.4) respectively, standardised mean difference = 0.11 (‐0.21, 0.43)) or in mean mental health score (general quality of life) (means 45.4 (13.7) v 46.5 (13.2) respectively, standardised mean difference = ‐0.08 (‐0.40, 0.24)). Overall quality of life was significantly higher in the care as usual group than in the cognitive assessment group (median (IQR) = 6.0 (4 to 7) in the cognitive assessment group, 6.5 (5 to 8) in the care as usual group, p<0.05), as was satisfaction with quality of life (median (IQR) = 4.0 (3 to 5) in the cognitive assessment group, 5.0 (4 to 8) in the care as usual group, p<0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐In the study by Lincoln et al (Lincoln 2002) there was no statistically significant difference between the assessment only group and the care as usual group (mean 27.3 (15.7) v 24.9 (14.7) respectively, standardised mean difference = ‐0.16 (‐0.48, 0.17), p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐Lincoln et al (Lincoln 2002) showed no statistically significant differences between assessment only and control groups in dysexecutive functioning (median (IQR) = 18.0 (7 to 31) v 16.5 (9 to 32) respectively, p>0.05), everyday memory (median (IQR) = 15.0 (5 to 31) v 14.0 (7 to 37) respectively, p>0.05), and compensation for memory impairment (median (IQR) = 9.0 (6 to 15) v 10.0 (7 to 14) respectively, p>0.05).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(d) Imagery+relaxation

(i) Imagery+relaxation versus care as usual

One study Maguire 1996; a two group study of 33 people comparing group‐based biologically‐oriented imagery treatment together with relaxation, to care as usual.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:     Immediately post‐treatment:‐ Maguire (Maguire 1996) found state anxiety to be significantly lower in the group receiving imagery (mean (SD) = 52.4 (13.0) in the imagery group, 56.3 (11.8) in the control group, standardised mean difference=0.31 (‐0.38, 1.0), p<0.05 after adjusting for baseline anxiety), but not trait anxiety (mean (SD) = 55.6 (13.7) in the imagery group, 56.8 (17.2) in the control group, standardised mean difference = 0.07 (‐0.61, 0.76), p>0.05 after adjusting for baseline anxiety).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:     Immediately post‐treatment:‐ Maguire (Maguire 1996) found no statistically significant differences in any of the dimensions of mood states (tension‐anxiety, depression‐dejection, anger‐hostility, vigor‐activity, fatigue‐inertia, confusion‐bewilderment ‐ all p>0.05, no other details given).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(e) Coping skills training

(i) Coping skills training versus peer support     One study Schwartz 1999; a two group study of 136 people comparing group‐based directive coping skills sessions together with peer telephone support from within the group, to non‐directive peer telephone support from lay people with MS trained in active listening.

Primary outcomes:     Disease specific quality of life:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐Satisfaction with quality of life was measured by global satisfaction, satisfaction with health, satisfaction with family life, spiritual satisfaction, and satisfaction with socio‐economic status in the study of Schwartz (Schwartz 1999). Psychosocial limitations of role performance and physical limitations of role performance were measured by the Sickness Impact Profile. The method of statistical analysis was complex, involved consideration of all pretreatment, during treatment and post‐treatment data simultaneously, and two types of statistical analysis used in parallel (one of which adjusted for sex and change in EDSS over the course of treatment). It was therefore not possible to identify from the report whether there was any specific effect immediately post‐treatment. There was some evidence that over the course of the whole study there were greater improvements in global satisfaction with quality of life, and satisfaction with family life in the imagery group, but in each case this was statistically significant for only one of the two methods of analysis. Means and SDs were not reported.

Short term follow‐up:‐As above, for the Schwartz study (Schwartz 1999) it was not possible to tell whether there were specific short term differences.

Medium term follow‐up:‐As above, for the Schwartz study (Schwartz 1999) it was not possible to tell whether there were any specific medium term differences.

Long term follow‐up:‐No data

Psychiatric symptoms:     Immediately post‐treatment:‐ Schwartz (Schwartz 1999) used the Arthritis Impact Measurement Scale to measure depression and anxiety. The method of statistical analysis was complex, involved consideration of all pretreatment, during treatment and post‐treatment data simultaneously, and two types of statistical analysis used in parallel (one of which adjusted for sex and change in EDSS over the course of treatment). It was therefore not possible to identify from the report whether there was any specific effect immediately post‐treatment. However, over the course of the entire study there were no statistically significant differences between the two groups in improvements in depression and anxiety. Means and SDs were not reported.

Short term follow‐up:‐As above for the study of Schwartz (Schwartz 1999). Again it was not possible to specifically identify short term effects.

Medium term follow‐up:‐As above for the study of Schwartz (Schwartz 1999). Again it was not possible to specifically identify medium term effects.

Long term follow‐up:‐No data

Psychological functioning:     Immediately post‐treatment:‐ MS specific self efficacy (control and function subscales), locus of control, ways of coping (problem solving, social support, reframing, religiosity, blaming others, blaming self, avoidance, and wishful thinking subscales), and happiness (purpose in life, personal growth, environmental mastery, social relatedness, and self‐acceptance subscales) were all measured by Schwartz (Schwartz 1999). The method of statistical analysis was complex, involved consideration of all pretreatment, during treatment and post‐treatment data simultaneously, and two types of statistical analysis used in parallel (one of which adjusted for sex and change in EDSS over the course of treatment). It was therefore not possible to identify from the report whether there was any specific effect immediately post‐treatment. However, over the course of the entire study there were statistically significant differences in locus of control (greater trend towards external locus of control in the peer‐support group, but only in one of the two sets of analyses), and blaming others as a way of coping (greater trend towards not blaming others in the coping skills group, but only in one of the two sets of analyses). No other differences were statistically significant.

Short term follow‐up:‐See above for the study of Schwartz (Schwartz 1999). It was not possible to specifically identify short term effects.

Medium term follow‐up:‐See above for the study of Schwartz (Schwartz 1999). Again it was not possible to specifically identify medium term effects.

Long term follow‐up:‐No data

Measures of disability:     Immediately post treatment:‐EDSS was measured post‐treatment in the Schwartz study (Schwartz 1999). There was a statistically significant difference between the coping skills and peer support group, but the direction of the difference is not reported.

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐The Schwartz study (Schwartz 1999) reported no difference in overall scores of the battery of cognitive tests between the two groups (p>0.05 adjusted for baseline cognitive ability, sex and change in EDSS).

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐In the Schwartz study (Schwartz 1999) fatigue was measured. The method of statistical analysis was complex, involved consideration of all pretreatment, during treatment and post‐treatment data simultaneously, and two types of statistical analysis used side by side (one of which adjusted for sex and change in EDSS over the course of treatment). It was therefore not possible to identify from the report whether there was any specific effect immediately post‐treatment. Over the course of the entire study there did not appear to be any differences in fatigue improvement between the two groups.

Short‐term follow‐up:‐See above. It was not possible to tell from the study of Schwartz (Schwartz 1999) whether there were any specific short term differences in fatigue improvement.

Medium term follow‐up:‐See above. It was not possible to tell from the study of Schwartz (Schwartz 1999) whether there were any specific medium term differences in fatigue improvement.

Long term follow‐up:‐No data     (f) Health behaviour therapy

(i) Health behaviour therapy versus waiting list control     One study Stuifbergen 2003; a two group study of 142 people comparing a group‐based wellness programme for women based on health promotion and health psychology models of health behaviour and beliefs, to a waiting list control group.

Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐Stuifbergen 2003 measured nine dimensions of general quality of life. Higher scores equate to better health. The mean (SD) in the intervention group, the waiting list control group and the standardised mean difference between them was 52.6 (28.9), 41.9 (30.8), standardised mean difference= 0.36 (‐0.02, 0.73) for physical functioning: 47.2 (42.7), 29.0 (34.3), standardised mean difference= 0.47 (0.09, 0.85) for physical role: 64.9 (27.2), 56.0 (28.0), standardised mean difference= 0.32 (‐0.06, 0.70) for bodily pain: 53.6 (23.3), 57.0 (23.2), standardised mean difference= ‐0.15 (‐0.52, 0.23) for general health: 43.3 (23.3), 36.1 (23.7), standardised mean difference= 0.30 (‐0.07, 0.68) for vitality: 72.1 (26.9), 60.5 (28.4), standardised mean difference= 0.42 (0.04, 0.80) for social functioning: 72.3 (40.7), 57.7 (44.7), standardised mean difference= 0.34 (‐0.04, 0.72) for emotional role: 74.0 (21.4), 69.4 (23.1), standardised mean difference= 0.20 (‐0.17, 0.58) for mental health. Thus, the intervention group tended to have better quality of life as measured by physical role and social functioning. In the paper, complex multilevel modelling was used to analyse all follow‐up points simultaneously (including mid‐intervention measurements) and to adjust for baseline quality of life and impairment. The authors concluded that across time points the intervention group had a higher score on the bodily pain scale (i.e. less pain on average) and a higher score on the mental health scale (better mental health).

Short term follow‐up:‐Stuifbergen 2003 measured nine dimensions of general quality of life. The mean (SD) in the intervention group, the waiting list control group and the standardised mean difference between them was 51.0 (29.1), 40.2 (30.8), standardised mean difference= 0.36 (‐0.03, 0.75) for physical functioning: 46.9 (43.8), 41.4 (42.0), standardised mean difference= 0.13 (‐0.26, 0.51) for physical role: 66.7 (24.6), 63.8 (28.2), standardised mean difference= 0.11 (‐0.28, 0.49) for bodily pain: 57.1 (24.9), 60.4 (23.9), standardised mean difference= ‐0.13 (‐0.52, 0.25) for general health: 44.0 (22.3), 41.2 (21.5), standardised mean difference= 0.13 (‐0.26, 0.51) for vitality: 69.6 (25.9), 70.2 (24.4), standardised mean difference= ‐0.02 (‐0.41, 0.36) for social functioning: 76.2 (36.0), 65.5 (42.5), standardised mean difference= 0.27 (‐0.12, 0.66) for emotional role: and 74.6 (15.0), 71.7 (19.7), standardised mean difference= 0.16 (‐0.22, 0.55) for mental health. Thus, there were no statistically significant differences between the groups. However, in the paper, complex multilevel modelling was used to analyse all follow‐up points simultaneously (including mid‐intervention measurements) and to adjust for baseline quality of life and impairment. The authors concluded that across time points the intervention group had a higher score on the bodily pain scale (i.e. less pain on average) and a higher score on the mental health scale (better mental health).

Medium term follow‐up:‐ No data

Long term follow‐up:‐ No data

Psychiatric symptoms:

Immediately post‐treatment:‐No data

Short term follow‐up:‐ No data

Medium term follow‐up:‐ No data

Long term follow‐up:‐ No data

Psychological functioning:

Immediately post‐treatment:‐ Stuifbergen 2003 measured self‐efficacy. The mean (SD) in the intervention group, the waiting list control group and the standardised difference between them was 92.6 (14.8) , 83.1 (18.8), standardised mean difference=0.56 (0.18, 0.93). This suggests higher self‐efficacy in women in the intervention group. This result was also reflected in the complex multilevel modelling used in the paper, where the difference between the intervention and control groups, over all time points simultaneously, was statistically significant.

Short term follow‐up:‐Stuifbergen 2003 measured self‐efficacy. The mean (SD) in the intervention group, the waiting list control group and the standardised difference between them was 93.5 (14.4) , 83.6 (19.2), standardised mean difference=0.58 (0.20, 0.96). This suggests higher self‐efficacy in women in the intervention group. This result was also reflected in the complex multilevel modelling used in the paper, where the difference between the intervention and control groups, over all time points simultaneously, was statistically significant.

Medium term follow‐up:‐ No data

Long term follow‐up:‐No data

Measures of disability:

Immediately post treatment:‐ No data

Short term follow‐up:‐Mean incapacity was 14.1 (7.3) in the intervention group and 16.3 (8.9) in the waiting list control group (standardised mean difference= 0.27 (‐0.12, 0.65)).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:     Immediately post‐treatment:‐

Short‐term follow‐up:‐See above.

Medium term follow‐up:‐See above.

Long term follow‐up:‐No data

MINI‐REVIEW 4: People with MS with depression     (a) Cognitive behavioural therapy

(i) Cognitive behavioural therapy versus care as usual

Two studies; Larcombe 1984 (a two arm study of 20 people comparing group‐based cognitive behavioural therapy with a waiting list control group), and Mohr 2000 (a two arm study of 32 people comparing telephone‐based individual cognitive behavioural therapy to care as usual).     Primary outcomes:

Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ In the Larcombe et al study (Larcombe 1984) mean (SD) post‐treatment self‐reported depression was 8.1 (5.0) in the CBT group and 33.4 (9.7) in the control group (standardised mean difference = 3.07 (95% CI; 1.65, 4.49), p‐value adjusted for pre‐treatment depression <0.01). Mean clinician rated depression was 2.0 (1.5) and 17.4 (8.3) respectively (standardised mean difference = 2.40 (95% CI; 1.16, 3.65), p‐value adjusted for pre‐treatment depression <0.01).

In the Mohr et al study (Mohr 2000) mean mood on a depression‐dejection subscale was significantly lower in the CBT group (18.7 (13.8)) than in the control group (26.7 (13.7)) (standardised mean difference = 0.57 (95% CI; ‐0.14, 1.28), p=0.01 after p‐value adjusted for baseline mood).

Using meta‐analysis to combine the results from these two studies, the overall standardised mean difference is 1.07 (0.43, 1.70) using a fixed effects model. The heterogeneity test comparing the mean differences between the two studies is statistically significant (p=0.002), possibly due to differences in the way depression is measured. Using a random effects model changed the standardised mean difference to 1.74 (‐0.71, 4.19), p=0.16. These mean differences are not adjusted for baseline measures of depression; in the individual studies the differences are statistically significant once baseline measures are taken into account.

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Psychological functioning:     Immediately post‐treatment:‐ In the Larcombe study (Larcombe 1984) the mean "best mood" during the day in the CBT and control groups was 44.4 (6.0) and 30.0 (6.8) respectively (standardised mean difference = 2.15 (95% CI; 0.96, 3.33), p‐value adjusted for baseline mood >0.05), the "worst mood" during the day was 37.3 (6.5) and 19.6 (5.4) respectively (standardised mean difference = 2.84 (95% CI; 1.48, 4.20), p‐value adjusted for baseline mood <0.05), and "average mood" during the day was 42.2 (5.0) and 26.1 (5.8) respectively (standardised mean difference = 2.84 (95% CI; 1.48, 4.19), p‐value after adjusting for baseline mood >0.05).

Short term follow‐up:‐ No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data     Measures of disability:

Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:     Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐In the Mohr et al study (Mohr 2000), all patients had been approved for treatment with interferon beta‐1a. In the CBT group 12 out of 16 were taking interferon beta‐1a after 12 months compared to six out of 15 in the control group (p=0.07 using Fisher's Exact Test).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

(ii) Cognitive behavioural therapy versus antidepressants

One study was found, reported over two papers (Mohr 2001, Mohr 2003), a three arm study (one arm not included) of 63 people comparing group based cognitive behavioural therapy to a group taking serotonin‐specific reuptake inhibitor.

Primary outcomes:     Disease specific quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

General quality of life:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychiatric symptoms:

Immediately post‐treatment:‐ In the study of Mohr et al (Mohr 2001) mean self‐reported depression in the CBT and Sertraline groups was 12.9 (8.6) and 13.7 (10.4) respectively (standardised mean difference = 0.08 (95% CI; ‐0.59, 0.75)), and mean clinician rated depression was 12.4 (6.6) and 13.9 (5.5) respectively (standardised mean difference = 0.24 (95% CI; ‐0.43, 0.91)). At the end of the treatment period 50% (10/20) of people in the CBT group and 24% (5/21) in the Sertraline group had a clinically significant improvement in depression (p=0.08 using Fisher's Exact Test). In the CBT group 60% (12/20) had moderate depressive disorder as assessed by the Structured Clinical Interview for DSM‐IV Axis 1 Disorders, compared with 62% (13/21) in the Sertraline group (p=0.88 using Fisher's Exact Test).

Short term follow‐up:‐In the study of Mohr et al (Mohr 2001) mean self reported depression in the CBT and Sertraline groups was 12.1 (7.4) and 15.5 (6.9) respectively (standardised mean difference = 0.47 (95% CI; ‐0.16, 1.09)), and mean clinician rated depression was 11.3 (5.6) and 12.5 (5.3) respectively (standardised mean difference = 0.21 (95% CI; ‐0.56, 0.98)). Occurrence of moderate depressive disorder in the six months following treatment was not significantly different between the two groups (no further details reported).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Psychological functioning:

Immediately post‐treatment:‐ No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Measures of disability:     Immediately post treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Cognitive outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐No data

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Secondary outcomes:

Immediately post‐treatment:‐No data

Short term follow‐up:‐Health care utilisation was summarised by whether the participant needed to start psychotherapy in the six month period following CBT. In the Mohr et al study (Mohr 2001), 24% of the CBT group started psychotherapy during the six months after treatment compared with 13% (1/12) of the control group (p=0.25 using Fisher's Exact Test). Change in medication was measured by whether antidepressant therapy was started during this period. In the CBT group 18% (3/17) started antidepressant therapy compared with 33% (4/12) in the control group (p=0.30 using Fisher's Exact Test).

Fatigue was measured as part of the same study, but reported in a separate paper (Mohr 2003). Mean (SD) total fatigue in the CBT and Sertraline groups was 140.4 (22.4) and 139.0 (29.4) respectively (standardised mean difference = ‐0.05 (‐0.70, 0.59)), mean (SD) fatigue severity was 52.5 (12.5) and 51.9 (15.8) respectively (standardised mean difference = ‐0.04 (‐0.69, 0.60)), mean (SD) fatigue consequences was 16.4 (3.4) and 15.8 (2.8) respectively (standardised mean difference = ‐0.19 (‐0.83, 0.46)), mean (SD) situation‐specific fatigue was 31.2 (6.0) and 32.4 (6.8) respectively (standardised mean difference = 0.19 (‐0.46, 0.83), and mean (SD) responsiveness to rest/sleep was 10.1 (3.8) and 10.3 (2.5) respectively (standardised mean difference = 0.06 (‐0.59, 0.70)).

Medium term follow‐up:‐No data

Long term follow‐up:‐No data

Discussion

General

Although the search strategy detected 16 randomised trials, there was little overlap between them in terms of population, intervention, outcome and timing of outcome. Only two studies were similar enough to be included in a meta‐analysis, and then only for one outcome measure. Many of our conclusions are necessarily based on the results of individual trials. It is evident that the scope for using psychological interventions to help people with MS is broad, but that the randomised controlled trial evidence base for their effectiveness is relatively small. Further trials need to be conducted. The flexibility of the mini‐review format used in this review makes it relatively easy to incorporate the diversity of existing and future trials. It is envisaged that further mini‐reviews will be added as the evidence base grows.

Methodology

Psychological interventions are complex in that they usually consist of a number of different elements. Some of these elements will be active ingredients specifically included because they are based on psychological theory. Other elements may not be specific to psychological interventions and may be common to many different types of intervention (such as interacting with other people with MS in a group). Still other elements will be specific to individual therapists (for example the therapist's experience and enthusiasm, and the way the therapist interacts with the client). To help generalisability of findings, interventions should be described in detail and, ideally, manualised so that the same intervention can be transferred outside the study environment. The distinction should always be made between trials that have a pragmatic basis (where the focus is on whether the whole therapy works and measures the effect that would be seen if the intervention was introduced into clinical practice) and those that have a therapeutic focus (where the aim is to discover which elements of the intervention work).

Many of the individual trials had a small sample size, and this raises the possibility that their conclusions are flawed because of the high chance of making a type II error (incorrectly concluding that the intervention has no effect). When the study is small because it is a pilot study, this should be clearly stated in the title or abstract. When the study is not a pilot study it is highly desirable that the protocol includes a comprehensive justification of sample size, including formal sample size calculations, and this should be reported in detail in resultant publications. The results of underpowered studies that replicate the protocol of existing, completed, studies can be usefully combined using meta‐analysis. In all studies, presenting differences between treatment arms using confidence intervals will allow the reader to gauge the influence of sample size on the precision of the results.

Because of the complex nature of psychological interventions, researchers often use a large number of diverse outcome measures, measured on more than one occasion, to assess their effect. However, with multiple significance testing comes the increased risk of making a type I error (incorrectly concluding that the intervention has an effect). The best way of minimising the impact of this multiplicity is to specify a priori, in the protocol, one or two primary outcome measures; these would then be given particular prominence in the resultant publications. Another approach that can be used in parallel, or as an alternative, is to adjust for multiple significance testing in the statistical analysis. A common way of doing this is to use a Bonferroni adjusted critical significance level that reduces the chance of making a type I error. It should be noted that using a procedure like this will have a knock‐on effect of increasing the chance of making a type II error, and so sample size will need to be increased to take this into account.

For many of the studies identified in this review it was difficult to make clear judgements about quality because the method of randomisation, method of allocation concealment and description of withdrawals were not reported in sufficient detail. To some extent the problem is historical; three studies were published in the 1980s and four in the 1990s. Encouraging researchers and Journal Editors to consider the guidance in the CONSORT statement should help improve the reporting of new trials.

The methods of statistical analysis used in trials of psychological interventions are often quite sophisticated; for example measurements made on multiple occasions (repeated measures) are often analysed simultaneously, and baseline measurements are often included in the analysis. However the reporting of basic statistics is often inadequate and this sometimes makes it impossible to complete the data extraction tool, and would prevent the use of meta‐analysis techniques if they were otherwise appropriate. We suggest that appropriate summary statistics (e.g. mean, SD and n for interval or ordinal scales outcomes, % and n for dichotomous outcomes) should be presented for all outcome measures (regardless of whether they were statistically significant) at each time point on which they were measured. This would enable studies to contribute fully to the systematic review.

Mini‐review 1:‐People with MS with cognitive impairment.

Only three trials have been conducted (Benedict 2000; Jonsson 1993; Solari 2004). One very small study (Benedict 2000) (n=15) compared one measure of psychiatric symptoms (depression) immediately post treatment in a group having neuropsychotherapy with a group receiving supportive psychotherapy. No statistically significant differences were found, although the chance of making a type II error is high because of the small sample size. A second study (Jonsson 1993) (n=48) compared psychiatric symptoms and cognitive outcomes in a group receiving combined cognitive rehabilitation and neuropsychotherapy with an "attention" control group. Of three measures of psychiatric symptoms, one, depression, was of borderline statistical significance both immediately post‐treatment and at short term follow‐up. Neither would have been significant if compared to a Bonferroni adjusted critical p‐value. Of 11 cognitive outcomes, one was of borderline statistical significance immediately post‐treatment, and another was of borderline statistical significance at short‐term follow‐up. Again, neither would have been significant if compared to a Bonferroni adjusted critical p‐value. A third study (Solari 2004) (n=82) compared two measures of MS specific quality of life, one measure of psychiatric symptoms, and eight measures of cognitive function in a group receiving computer aided retraining of memory with a sham therapy. Of the two measures of MS specific quality of life, one (mental health subscale) was of borderline statistical significance immediately post‐treatment, but not at short‐term follow‐up. The one measure of psychiatric symptoms (depression) was similar between the two groups. Of the eight measures of cognitive function, one (Word List Generation test) was significantly different immediately post‐treatment and at short‐term follow‐up, but none of the other seven measures showed a difference on either occasion. Neither would have been significant if compared to a Bonferroni adjusted critical p‐value. One summary measure of cognitive function was prespecified by the authors as being the primary outcome measure: this did not improve significantly as a result of treatment. No data were available on other types of outcome measures and on medium or long‐term effects of treatment. Given the different aims of the three studies, no meta analysis was possible.

Cognitive assessment and treatment are important areas for psychologists working with people with MS. Randomised trials of their effectiveness are limited in number and in the variety of interventions. The available data are not particularly conclusive either way, and one explanation could be the relatively small sample sizes. Further trials in this area would be useful.

Mini‐review 2: People with MS with moderate to severe disability

Three trials were included in this mini‐review (Crawford 1985; Gordon 1997; Mendoza 2001). None of the interventions was sufficiently similar to the others for a meta‐analysis to be conducted. A small trial of psychotherapy versus placebo versus care as usual (Crawford 1985) (n=32 over three groups) compared two measures of psychiatric symptoms and two measures of psychological functioning immediately post‐treatment. People in the psychotherapy group were less depressed than those in the other two groups. The other two trials were also small; one of social skills training looking at psychological functioning (Gordon 1997), and the other on combined cognitive assessment and rehabilitation looking at psychiatric symptoms and cognitive outcomes (Mendoza 2001). For the former study, participants tended to score reasonably well on the relevant measures even at baseline, and there were no significant differences between the groups post‐treatment. For the latter study the data were not presented in a way that allowed comparison of the two groups.

Again the limited data available suggest that further trials in this area would be useful.

Mini‐review 3:‐ People with MS

There were seven trials in this mini‐review (Foley 1987; Lincoln 2002; Maguire 1996; Rigby 2003; Schwartz 1999; Stuifbergen 2003; Wassem 2003). Three of the studies evaluated interventions using cognitive behavioural approaches. One small trial (Foley 1987) (n=41) of individual cognitive behavioural therapy versus psychotherapy using four measures of psychiatric symptoms and two measures of psychological functioning measured immediately post‐treatment found lower depression, general anxiety, perceived distress and improved coping in people given CBT. Another larger trial (Rigby 2003) (n=147) comparing a brief group‐based cognitive behavioural therapy to a placebo social discussion group, and to a group receiving an information booklet, used two measures of psychiatric symptoms and six measures of psychological function measured immediately post treatment, short‐term (on three occasions) and medium term. The data were analysed using two different methods, one of which used a baseline adjusted summary measure of the entire follow‐up period. This latter analysis suggested that the group given CBT had lower anxiety, higher self‐efficacy (one of two such measures) and better dispositional resilience (three of four such measures) than the information booklet group, but not the placebo control group. One small study (Wassem 2003) (n=27) with four years of follow‐up, compared a group‐based cognitive behavioural approach to improving adjustment to MS, with standard care, using three measures of adjustment and measures of pain and fatigue. There were no statistically significant differences. The results of two of the cognitive behavioural therapy trials look encouraging, although with the group‐based study we are unable to exclude the possibility that it is the group environment that is having the effect rather than the therapy per se. The three studies have different emphases (stress inoculation; understanding emotions, and coping; and self‐efficacy) but also appear to have much in common.

A small study of guided imagery and relaxation (Maguire 1996) found a difference in one outcome (lower state anxiety) out of eight (two measuring psychiatric symptoms and six measuring psychological functioning). A medium size study comparing coping skills training with peer support (Schwartz 1999) looked at a variety of quality of life (seven), psychiatric symptom (two), psychological functioning (16), disability (one), cognitive (one) and fatigue (one) outcomes. The method of analysis was complex and analysed outcomes immediately post‐treatment, short‐term and long‐term follow‐up simultaneously and in two different ways. Of the 55 different analyses, five were statistically significant (two quality of life, two psychological functioning and the measure of disability). Another medium‐sized trial (240 people over three groups) compared a group receiving cognitive assessment and cognitive rehabilitation with a group receiving cognitive assessment only and a care as usual group. There were no differences in quality of life (four measures), psychological functioning (one measure) or cognitive outcomes (three measures) between any of the groups except for a poorer average quality of life in the assessment only group. One medium‐sized study (Stuifbergen 2003) (n=142) compared a health behaviour change therapy with a waiting list control group using nine measures of general quality of life, and one measure of disability. Two of the measures of quality of life were statistically significant. The trial of cognitive rehabilitation and assessment (Lincoln 2002) was of reasonable size but did not show benefit; indeed, there was some suggestion that cognitive assessment only without rehabilitation treatment might even be harmful. The other trials were often difficult to interpret because of the small number of differences found relative to the number of outcome measures looked at, and none pre‐specified a small number of outcomes as primary outcome measures.

Mini‐review 4:

There were three trials in this mini‐review (Larcombe 1984; Mohr 2000; Mohr 2001; Mohr 2003). Two trials looked at cognitive behavioural therapy versus control conditions (Larcombe 1984; Mohr 2000). Both studies looked at a small number of outcome measures and even though sample sizes were small, both showed significant improvements in depression, and mood in the CBT group. A fixed effects meta‐analysis of these two studies also showed significant improvement as a result of CBT (although there was significant heterogeneity between the two studies, and a reanalysis using a random effects models was not statistically significant). A comparison of CBT with anti‐depressants (Mohr 2001; Mohr 2003) showed no statistically significant differences, although sample size was small (n=41), and the possibility of a type I error cannot be excluded. This last study also included a non‐randomised group of patients who received supporting‐expressive group therapy. Post‐treatment levels of depression in this group were significantly worse than in both the CBT and anti‐depressant groups. Results from this mini‐review are encouraging despite the small sample size. CBT is an effective treatment for mild to moderate depression in the general population, and this adds weight to the evidence in this mini review.

Authors' conclusions

Implications for practice.

MS is a complex disease, and, potentially, there are many different ways in which psychological interventions can help. This is reflected in the diversity of approaches that have been identified in this review. Interventions can differ by their target population, purpose, whether group‐based or one‐to‐one, the psychological theories and models from which they are derived, and so on. The review shows that considerable creativity and innovation has gone into developing and adapting these interventions. The downside of this is that, in general, interventions have been evaluated in single studies only, and often with limited sample size.

Cognitive behavioural therapy was the most frequently used psychological approach. In all but one of these seven studies the purpose of the intervention was either to treat depression or (broadly) to give people skills to help them cope with having MS (albeit in different ways). In general these results were promising. Also they should not be interpreted in isolation from research that has been conducted in people without MS, or from non‐randomised evidence. For example evidence on the effectiveness of cognitive behavioural approaches in treating mild to moderate depression in the general population may be relevant to people with MS with depression. A related question is whether people with MS should be regularly assessed for depression and other psychiatric problems. Clearly effective treatments need to be available before regular assessment becomes worthwhile.

There were no studies focusing on psychological approaches to managing common symptoms of MS such as fatigue and pain. However, there are Cochrane systematic reviews suggesting that cognitive behavioural approaches may be useful in adults with chronic pain (Ostelo 2004) and people with Chronic Fatigue Syndrome (Price 1998). Clearly interventions like these would need to be adapted for, and evaluated in, people with MS.

The evidence for the effectiveness of interventions to help people with cognitive impairments was inconclusive, partly because of the large number of outcome measures that are frequently used in this type of study, and partly because of small sample size. Further, the small number of studies actually conducted (all three of which tested different interventions) means that it is difficult to generalise about whether psychological approaches to treating cognitive impairment are helpful.

Two of the studies qualifying for this review used interventions that were delivered by nurses. Psychologists working in the acute hospital setting in the NHS are in short supply. Training and supporting other health professionals to deliver psychology‐based interventions, or to deliver interventions that incorporate psychological principles, could be useful. This would need to be corroborated by research evidence.

Where a psychological intervention is found to be effective, consideration should be given as to how that intervention can be replicated elsewhere whilst maintaining effectiveness. Describing interventions in some detail (e.g. in a manual) would facilitate the standardisation of procedures and widespread adoption of the intervention.

Implications for research.

Although this review has identified 16 relevant studies, they are very diverse, and there are large gaps in the evidence base for the use of psychological interventions for people with MS. There are many different areas that could be researched, but this section focuses on those areas that would probably be fruitful in the short to medium term. The studies on cognitive behavioural approaches for depression are encouraging, but sample sizes are small, and a larger sufficiently powered study is needed. Research is also needed to evaluate whether screening for depression and other psychiatric problems, combined with psychological or pharmacological therapy, is effective.

We found no completed studies evaluating psychological treatments for MS symptoms such as fatigue and pain. We found one ongoing study evaluating one‐to‐one cognitive behavioural therapy in the management of fatigue. Symptom management is an area where research in other chronic diseases is encouraging, and where further work in MS would be useful.

The existing research on adjustment and coping looks encouraging. MS is characterised by either relapse‐remitting phases or progressive accumulation of disability, and adjustment and coping can be seen as an ongoing process throughout the course of the disease. More focussed approaches to adjustment and coping, such as looking at the period soon after diagnosis, or looking at different types of MS, would be useful.

There are a number of issues that need to be addressed in relation to the delivery of psychological treatment. For example, what are the relative merits of providing treatment in groups or on a one‐to‐one basis? What is the potential for professions other than psychologists, such as nurses and occupational therapists, to conduct psychology‐based interventions? When is it most timely to deliver interventions such as those about adjusting to, and coping with having MS? Is it possible to identify subgroups of people with MS who would benefit most from psychological approaches?

Some of this research will need to take the form of a randomised control trial, although this may not always be the most suitable methodology.

As well as having implications for the types of research that need to be conducted, there are also implications for the reporting of randomised controlled trials, and for methodology. There were noticeable improvements over time in the quality of the papers reviewed. Reporting of trials should follow CONSORT guidelines where possible. Clearer reporting of indicators of quality such as the method of randomisation is needed. Masking of participants or researchers is usually not possible given the nature of these interventions. However, it may sometimes be possible to mask the person assessing outcome, the data analyst, or the person interpreting the results, and this would help to minimise bias. Some of the studies in the review have compared a psychological intervention to two control groups; one a care as usual group, and the other a placebo group. This design is helpful in addressing the dual questions of whether the intervention is beneficial, and if so, why? More detailed reporting of basic summary statistics such as means and standard deviations is necessary so that results can be incorporated into meta analyses. Interventions need to have a clear purpose, and described in sufficient detail so that the reader can understand the underlying principles and have a clear picture of the structure and content.

Many of the studies in this review are limited by small sample size. They were rarely described as pilot studies, and none was followed up with a larger trial. Pilot studies are an important phase of developing and evaluating a psychological intervention, and should be described as such. Sample size for other studies should be based on a formal sample size calculation. Many studies used large numbers of outcome measures, but rarely identified a small number, a priori, as primary outcome measures. This would greatly facilitate the interpretation of individual trials. Studies should, where appropriate, include measures that are of direct relevance to the person with MS, such as measures of quality of life. MS specific measures are likely to be better at picking up changes in quality of life than generic measures. Researchers should try to measure outcomes (particularly primary outcomes) even when participants drop out of the intervention. This will enable a proper intention‐to‐treat analysis to be conducted.

What's new

Date	Event	Description
27 August 2008	Amended	Converted to new review format.

Open in a new tab

Acknowledgements

John Gill (Principal Librarian) and Susan Merner helped to develop and implement the search strategy. Geoff Linder and Kate Burrall provided a consumer perspective on the research question, design of review, and choice of outcome measures. Geoff Linder also commented on the protocol and the completed review. Dr Jonathan Cole commented on an early draft of the protocol. Nicola Thornton and Silvana Simi commented on the draft protocol from a consumer perspective. Dr Lindsay Vowels commented on an early draft of the protocol. The Cochrane MS Group, particularly Liliana Coco and Maura Moggia, provided advice and support in preparing the protocol, and conducting search updates. Silvana Simi produced the first draft of the Plain Language Summary. Sten Fredrikson was the Review Editor.

Appendices

Appendix 1. CENTRAL search strategy

1.   (DEMYELINATING‐DISEASES#.DE.) OR (OPTIC‐NEURITIS#.DE.) OR (MULTIPLE ADJ SCLEROSIS.TI,AB.) OR (MYELITIS ADJ TRANSVERSE.TI,AB.) OR (OPTIC ADJ NEURITIS.TI,AB.) OR (NEUROMYELITIS ADJ OPTICA.TI,AB.) OR (DISSEMINATED ADJ SCLEROSIS.TI,AB.) OR (ADEM.TI,AB.) OR (DEVIC.TI,AB.)   2.   (PT=RANDOMIZED‐CONTROLLED‐TRIAL OR CONTROLLED‐CLINICAL‐TRIAL OR CLINICAL‐TRIAL) OR (CLINICAL‐TRIALS#.DE.) OR (RESEARCH‐DESIGN#.DE.) OR (SINGLE‐BLIND‐METHOD#.DE.) OR (CROSS‐OVER‐STUDIES#.DE.) OR (CROSSOVER ADJ STUD$.TI,AB.) OR (CLINICAL ADJ TRIAL$.TI,AB.) OR ((SINGLE OR DOUBLE OR TREBLE) ADJ (BLIND$ OR MASK$).TI,AB.) OR (RANDOM$.TI,AB.)   3.   (PSYCHOTHERAPY#.DE.) OR (COUNSELING#.DE.) OR (PATIENT‐EDUCATION#.DE) OR (SELF‐EFFICACY#.DE.) OR (SELF‐CONCEPT#.DE.)   4.   (PSYCHO$ OR COUNSEL$ OR DEPRESSI$ OR INTERPERSONAL OR ART OR AVERSION OR BEHAVIO$ OR (COLO$ OR COLOU$) OR DANCE OR GESTALT OR MUSIC OR MILIEU OR PLAY OR REALITY OR SUPPORTIV$ OR BRIEF OR FAMILY OR GROUP OR COGNITIV$ OR NONDIRECT$ OR (PROBLEM ADJ SOLV$)) AND THERAP$.TI,AB.   5.   (BALINT OR BEHAVIO$ OR COPING OR (CRISIS ADJ INTERVENTION) OR (ASSERT$ NEAR TRAIN$) OR (SELF ADJ CONTROL) OR (PERSON ADJ CENT$) OR (CLIENT ADJ CENT$) OR PSYCHODRAMA$ OR (PARADOXIC$ ADJ TECHN$) OR (RATIONAL ADJ EMOTI$) OR (ROLE ADJ PLAY) OR (SOCIOTHERAP$) OR SOCIOENVIRONMENT$ OR TRANSACTIONAL OR ALEXITHYMI$ OR (ANGER ADJ MANAG$) OR BIOFEEDBACK OR (CONSTRUCT ADJ THEOR$) OR DEBRIEF$ OR (EMOTION ADJ FOCUS) OR (EXPERT ADJ PATIENT$) OR EXPOSURE OR FEEDBACK OR FLOODING OR FREUD$ OR (FUNCTIONAL ADJ ANALYSIS) OR HYPNOSIS OR JUNG$ OR KLEIN$ OR MINDFULNESS OR MOOD OR MULTIMODAL OR (OBJECT ADJ RELATION) OR PENNEBAKER OR REINFORCEMENT OR RELAX$ OR ROGERIAN OR (SCHEMA ADJ FOCUS$) OR (SELF ADJ MANAG) OR (SELF ADJ MONITOR$) OR (SELF ADJ TALK$) OR (SELF ADJ MONITOR$) OR (SELF ADJ DISCLOS$) OR (SELF ADJ NARRATIVE) OR SOCIOTHERAP$ OR (STRESS ADJ CONTROL$) OR (STRESS ADJ MANAG$) OR SYSTEMIC).TI,AB.   6.   3 or 4 or 5   7.   1 and 2   8.   7 and 6

Appendix 2. MEDLINE (dialog datastar) search strategy

Appendix 3. EMABSE (dialog datastar) search strategy

1   (MULTIPLE‐SCLEROSIS#.DE.) OR (MULTIPLE ADJ SCLEROSIS.TI,AB.) OR (MYELITIS ADJ TRANSVERSE.TI,AB.) OR (OPTIC ADJ NEURITIS.TI,AB.) OR (NEUROMYELITIS ADJ OPTICA.TI,AB.) OR (DISSEMINATED ADJ SCLEROSIS.TI,AB.) OR (ADEM.TI,AB.) OR (DEVIC.TI,AB.)   2   (RANDOMIZED‐CONTROLLED‐TRIAL.DE.) OR (CONTROLLED‐STUDY.DE.) OR (CLINICAL‐TRIAL.DE.)   3   (CONTROL$ ADJ TRIAL.TI,AB.) OR (CROSSOVER ADJ STUD$.TI,AB.) OR (CLINICAL ADJ TRIAL$.TI,AB.) OR ((SINGLE OR DOUBLE OR TREBLE) ADJ (BLIND$ OR MASK$).TI,AB.) OR (RANDOM$.TI,AB.)   4   2 or 3   5   PSYCHOTHERAPY#.DE.   6   PSYCHO$ OR COUNSEL$ OR DEPRESSI$ OR INTERPERSONAL OR ART OR AVERSION OR BEHAVIO$ OR (COLO$ OR COLOU$) OR DANCE OR GESTALT OR MUSIC OR MILIEU OR PLAY OR REALITY OR SUPPORTIV$ OR BRIEF OR FAMILY OR GROUP OR COGNITIV$ OR NONDIRECT$ OR (PROBLEM ADJ SOLV$)   7   6 and THERAP$.TI,AB.   8   (BALINT OR COPING OR (CRISIS ADJ INTERVENTION) OR (ASSERT$ NEAR TRAIN$) OR (SELF ADJ CONTROL) OR (PERSON ADJ CENT$) OR (CLIENT ADJ CENT$) OR PSYCHODRAMA$ OR (PARADOXIC$ ADJ TECHN$) OR (RATIONAL ADJ EMOTI$) OR (ROLE ADJ PLAY) OR SOCIOTHERAP$ OR SOCIOENVIRONMENT$ OR TRANSACTIONAL OR ALEXITHYMI$ OR (ANGER ADJ MANAG$)).TI,AB.   9   (BIOFEEDBACK OR (CONSTRUCT ADJ THEOR$) OR DEBRIEF$ OR (EMOTION ADJ FOCUS) OR (EXPERT ADJ PATIENT$) OR FLOODING OR FREUD$ OR (FUNCTIONAL ADJ ANALYSIS) OR HYPNOSIS OR JUNG$ OR KLEIN$ OR MINDFULNESS OR MOOD OR (OBJECT ADJ RELATION)).TI,AB.   10   (PENNEBAKER OR ROGERIAN OR (SCHEMA ADJ FOCUS$) OR (SELF ADJ MONITOR$) OR (SELF ADJ TALK$) OR (SELF ADJ DISCLOS$) OR (SELF ADJ NARRATIVE) OR SOCIOTHERAP$ OR (STRESS ADJ CONTROL$) OR (STRESS ADJ MANAG$)).TI,AB.   11   5 or 7 or 8 or 9 or 10   12   1 and 4 and 11

Data and analyses

Comparison 1. Cognitive behavioural therapy versus care as usual for people with MS with depression.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Self‐reported measure of depression	2	51	Std. Mean Difference (IV, Fixed, 95% CI)	1.07 [0.43, 1.70]

Open in a new tab

1.1. Analysis.

Open in a new tab

Comparison 1 Cognitive behavioural therapy versus care as usual for people with MS with depression, Outcome 1 Self‐reported measure of depression.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Benedict 2000.

Methods	A 12‐14 week parallel group unmasked RCT.   Quality :   Randomisation generation = D   Allocation concealment = B   Withdrawals = A
Participants	People with MS with marked cognitive impairment and behaviour disorder. Recruitment via referral to a Neurology Department for clinical evaluation or by responding to advert for research pertaining to cognitive and emotional problems in MS.   Inclusion: (a) Clinically definite progressive MS   Exclusion: (a) other neurological disease, (b) drug/ alcohol dependence, (c) "psychiatric disease other than that attributable to MS", (d) clinical MS relapse, (e) Depressive disorder (DSM‐1V).   Number screened not known, 15 randomised.   Mean age (pooled SD) = 45 (10), 10F, 5M, 1 with primary progressive MS, 14 secondary progressive MS, disease duration not stated, mean EDSS (pooled SD) = 5.0 (2.4).
Interventions	Intervention 1: (8 randomised, 8 analysed). Neurological compensatory training. Intervention involved individual and caregiver only. 12 consecutive weekly sessions, 1 hr per week. Conducted by clinical neuropsychologist, site not stated. Three components (i) use of neuropsychological, neuroradiological evidence and caregiver anecdotes to enhance patient and caregiver understanding of neurological basis of cognitive impairment, pathological affect and abnormal social behaviour. (ii) social skills training to improve individuals' capacity to appreciate the perspective of others (iii) cognitive behavioural strategies to enhance self‐control and behaviour regulation in order to diminish frequency of socially aggressive behaviour.   Intervention 2: (7 randomised, 7 analysed). Non‐specific supportive psychotherapy. Intervention involved individual and caregiver only. 12 consecutive weekly sessions of 1 hr. Intervention provided by counsellor trained in general psychotherapy, site not stated. Aim was emotional support and stress modulation by fostering personality growth and behaviour change.
Outcomes	Measured at pre‐treatment and immediately post‐treatment (within 2 weeks of end of treatment). Primary outcomes: Measure of psychiatric symptoms using the Beck Depression Inventory (self reported). Secondary outcomes: None. Outcomes not included: Caregiver reports of Social Aggression Score (modified from a brain injury scale), Hogan Empathy Scale (adapted), and revised NEO Personality Inventory (neuroticism, extraversion, openness to experience, conscientiousness, agreeableness subscales).
Notes	Mini‐review 1.   The specific criteria by which participants were selected as having cognitive impairment or behaviour disorder are not defined. However, compared to a healthy control group of volunteers (n=15), the participants with MS scored (statistically) significantly higher on most of a baseline battery of 10 tests of cognition and 7 tests of personality/ behaviour. Further all those with MS scored over 4 on the Social Aggression Score and all healthy controls scored less than 4.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Crawford 1985.

Methods	A 25 week, 3 group, unmasked RCT. Triads of participants matched for sex, baseline measures, and length of illness were randomised, one to each group.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = C
Participants	People with MS who were inpatients at a large metropolitan hospital. Exclusion criteria: Severe deficits in mental status.   41 people screened, 32 enrolled, randomised and completed.   Mean age 47 yrs (range 20‐63), 13M, 19F, mean disease duration 18 yrs (range 2‐40), all with moderate to severe physical disability.
Interventions	Intervention (10 or 11 analysed): Traditional psychotherapy: One hour, twice per week, for 25 weeks. Group therapy based in hospital. Insight oriented with active encouragement to verbalise and confront issues of conflict.   Control 1 (10 or 11 analysed): Current events discussions. One hour twice per week for 25 weeks. Group sessions in hospital. Discussion of newsworthy topics of a didactic or humorous nature. Verbalisations of a personal or emotional nature were discouraged.   Control 2 (10 or 11 analysed). No treatment.
Outcomes	Measured at baseline and immediately post‐treatment (at 25 weeks). Primary outcome: Psychiatric symptoms measured using the Minnesota Multiphasic Personality Inventory Depression‐30 scale, and the Institute for Personality and Ability Testing Anxiety Scale Questionnaire. Psychological functioning measured using the Rosenburg Self‐Esteem Scale and the Adult Nowicki‐Strickland Internal‐External Control Scale (measure of locus of control).
Notes	Mini‐review 2.   It is not clear from the paper whether participants were hospitalised for the entire duration of the study.   The number of participants was not divisible by 3 (as would be expected from the method of randomisation that was used) and no explanation is given.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Foley 1987.

Methods	A 5 week parallel group unmasked RCT.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = C
Participants	People with MS who are outpatients at a college of medicine.   Inclusion: (a) Confirmed MS diagnosis, (b) EDSS of 8 or less, (c) no major cognitive deficits.   Number screened not stated, 41 randomised, 36 analysed.   Mean age = 39 yrs, 35F 6M, type of MS and duration not stated, mean EDSS = 6 (range 1 to 8), 60% experiencing MS exacerbations at study entry.
Interventions	Intervention: (number randomised and number analysed not stated). Stress inoculation training. Individual intervention, 6 sessions over 5 weeks. Conducted by a supervised advanced clinical psychology graduate in a hospital setting. Manualised cognitive behavioural intervention incorporating progressive deep muscle relaxation. Aims to enhance coping by ameliorating affective distress and preventing maladaptive psychological response to stress. Specially adapted for MS. Techniques included daily self‐monitoring with feedback, relaxation, cognitive reinterpretation and role playing.   Control (number randomised and number analysed not stated). Current available care, consisting of at least 2 hrs supportive psychotherapy over 5 weeks in a hospital setting. In addition 2 received antidepressants, 2 family counselling and 3 individual counselling. Also all individuals were told they would receive stress inoculation training at end of study.
Outcomes	Measured at pre‐treatment and immediately post‐treatment (at the end of 5 weeks). Primary outcomes: Psychiatric symptoms (self‐reported) measured by Beck Depression Inventory, State‐Trait Anxiety Inventory (State and Trait subscales). Psychological functioning measured by Ways of Coping Checklist (Problem‐Focussed Coping subscore), and Internal‐External Locus of Control. Outcomes not included: Hassles Scale.
Notes	Mini‐review 3
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Gordon 1997.

Methods	An 8 week, parallel group, unmasked RCT.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = B
Participants	People with MS attending a rehabilitation clinic.   Inclusion: (a) People with MS, (b) 18‐65 yrs old, (c) proficient in English, (d) acceptable cognitive ability (educational level and modified Mini‐Mental State Examination).   Exclusion: (a) history of mental illness or retardation.   250 people contacted, 42 screened, 26 randomised.   Mean age = 45 yrs, 19F 7M, disease type and duration not stated, all but 1 required assistive aids to walk, "several" used wheelchairs.
Interventions	Intervention (13 randomised, 11 analysed). Social skills training programme. Group intervention of 8 consecutive weekly 1.5 hour sessions. Person conducting intervention and site not stated. Focus on easing "interaction strain" between people with and without disability. Including (i) assertiveness training, (ii) presentation of self, (iii) acknowledgement and disclosure of disability and privacy, and (iv) general interpersonal communication. Methods used included role play, modelling, videotaped feedback and homework.   Control (13 randomised, 11 analysed). "Care as usual".
Outcomes	Measured at pre‐treatment and immediately post treatment (at the end of 8 weeks). Primary outcomes: Psychological functioning (self‐reported) measured by the Tennessee Self‐Concept Scale (overall measure of self‐esteem), Social Avoidance and Distress Scale (social distress and social avoidance subscales with some modification by authors‐ measures social anxiety experienced in interactive situations), and the Self Efficacy Questionnaire for Social Skills (instructions modified by authors). Secondary outcomes: None. Outcomes not included: The Social Interaction Self‐Statement Test (positive statements and negative statements subscores with modifications by authors; assesses thoughts prior to, during and following social interaction).
Notes	Mini‐review 2.   The authors report that, at baseline, participants in the study had low levels of social avoidance and distress, and had levels of self‐esteem that were in "the average range"
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Jonsson 1993.

Methods	A 33 week parallel group RCT with attempted masking of some outcome assessments.   Quality:   Randomisation generation = B   Allocation concealment = A   Withdrawals = B
Participants	Inpatients in a MS rehabilitation centre with mild to moderate cognitive and behavioural impairment.   Inclusion: (a) Diagnosis of MS (Schumacher definition) (b) inpatient at MS Rehab Hospital (c) mild to moderate cognitive impairment.   Exclusions: (a) >60 yrs old (b) Signs or symptoms of other neurological or cerebral disease (c) unrelated psychiatric disease (d) severe visual impairment (e) severe motor dysfunction of arms or hands (f) history of alcohol or drug abuse (g) regular user of psychopharmacia or analgesics (h) foreign origin (i) short admission.   130 screened, 48 eligible and randomised. Mean age (SD) =45 yrs (8), 19F 21M, 9 with primary progressive MS, 25 with secondary progressive MS, 6 with relapse‐remitting MS, mean disease duration (SD) =15yrs (10), mean EDSS (SD)=5.6 (1.7),
Interventions	Intervention (24 randomised, 20 analysed): Cognitive rehabilitation and neuropsychotherapy. 1‐1.5 hrs 3 times a week for mean 6.5 wks (SD=1wk). Individual therapy with close relative where possible, conducted by neuropsychologist in hospital. Cognitive rehabilitation consisted of basic principles of compensation, substitution and direct training. Neuropsychotherapy was goal directed based on neuropsychological test profile and personal problems.   Control group (24 randomised, 20 analysed): Non specific mental stimulation for 1‐1.5 hrs 3 times a week for average (SD) 6.6 weeks (1 wk). Conducted individually by neuropsychologist in hospital with close relative where possible. General discussion (film, newspapers), playing games, and discussion of personal and disease problems.
Outcomes	Measured at pre‐treatment, immediately post‐treatment (average of 46 days later), and short‐term follow‐up (6 months after the end of treatment). Primary outcomes: Psychiatric symptoms (self‐reported) measured using the Beck Depression Inventory and the State‐Trait Anxiety Inventory (both state and trait scales). Cognitive function, with attempted masking of the assessor, measured using 3 memory span tests, 2 verbal learning tests, 3 visuo‐spatial memory tests, 3 visuo‐motor speed tests, the sum of these 11 tests, 1 test of verbal intelligence, the WAIS performance test, a recognition memory test, a visual perception test, a concentration test and 4 tests of verbal fluency. Secondary outcomes: None.
Notes	Mini‐review 1.   Masking reported to be mostly unsuccessful.   Mild to moderate cognitive impairment is not specifically defined. Mean pre‐treatment scores on cognitive tests tended to be significantly lower than expected (for sex, age and educational status) compared to a sample of brain healthy individuals (n=141).
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Open in a new tab

Larcombe 1984.

Methods	A 12 week, parallel group, mostly unmasked RCT   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = B
Participants	People with MS with moderate depression recruited from either a MS Community Centre, a specialised MS hospital treatment unit, or via a MS Society Newsletter.   Inclusion: (a) 20‐65 years old, (b) self‐reported depression for at least 3 months, (c) at least 20 on the Beck Depression Inventory, (d) no prior/concurrent treatment with tranquillisers/ lithium, (e) definite or probable depression (Feighner criteria), (f) no other major psychological disorder, (g) low suicide risk (Beck criteria), (h) normal cognitive assessment, (i) Neurologist diagnosis of MS.   20 recruited out of 54 screened. Data available on 19 completers. Mean age 42.5 years, 6M, 13F. Disease duration 10 years or less for n=8, 11‐30 years for n=11. 8 require a wheelchair.
Interventions	Intervention (10 randomised, 9 included in analyses) : Standard Cognitive Behavioural Therapy (CBT) . Groups of 4/5, 1.5 hrs per week for 6 weeks by a Clinical Psychologist at a specialist MS centre. In sessions 1‐2 the emphasis was on behavioural procedures and based on a behavioural theory of depression. The sessions aimed to increase the frequency, quality and range of activities and social interactions. This included deriving activity schedules, instigating and monitoring engagement in activities, and discussing social interactions. In sessions 3‐6 there was increasing emphasis on cognitive procedures aiming to identify depressogenic and distorted cognitions and irrational beliefs, subject them to analysis and empirical testing, and encourage realistic and positive thinking. Participants were trained to monitor positive and negative thoughts; recognise the connection between affect, cognition and behaviour; examine the evidence against negative thoughts and irrational beliefs; substitute more positive interpretations for the negative cognitions; to identify and alter irrational beliefs which predispose to distortion of experiences.   Control group: Start of CBT delayed for 6 weeks (waiting list control)
Outcomes	Measured at 1 week pre‐treatment and immediately post‐treatment (1 week after end of treatment). Primary outcome: Psychiatric symptoms measured by the Beck Depression Inventory (self‐reported), and the Hamilton Rating Scale (depression scale scored by masked assessor). Psychological functioning was measured using daily 10 point ratings of "best", "worst" and "average" depressed mood. Secondary outcomes: None. Outcomes not included: Significant‐other Rating scale (depression as rated by a "significant other").
Notes	Mini‐review 4
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Lincoln 2002.

Methods	A 39 week, 3 group RCT with masked assessor.   Quality:   Randomisation generation = A   Allocation concealment = A   Withdrawals = B
Participants	People with clinically definite or probable MS (Poser definition) recruited from a University Hospital MS management clinic. Inclusion: Able to co‐operate with assessments for 30 mins at a time.   240 randomised from 262 giving consent. Data available on 223 completers. Mean (SD) age 43 yrs (10), 70% female, type and duration of MS not stated, median Guy's Neurological Disability Scale = 16.
Interventions	Control group (82 randomised, 77 included in analyses): No psychological assessment or treatment.   Assessment group (79 randomised, 72 included in analyses): 3hr cognitive assessment of memory, attention and executive functioning by a supervised assistant psychologist. Further assessments done on the basis of patient's performance (e.g. everyday attention, recognition memory, verbal and spatial reasoning). Appropriate feedback sent to GP, hospital, patients and (where agreed) relatives.   Intervention (79 randomised, 74 included in the analyses): Assessment and feedback as previous group together with targeted cognitive rehabilitation for any deficits identified. Various techniques used including diaries, calenders, notebooks, lists, and visual mnemonics. Progress monitored with weekly diaries. Patients visited for up to 6 mths after assessment.
Outcomes	Measured at 4 months after assessment (not included in review as this could be before the end of treatment in the intervention arm) and at short‐term follow‐up (8 months after recruitment). Primary outcomes: General quality of life (self‐reported) was measured using the SF‐36 (physical health and mental health composite scores), and 2 items from the SF‐54 (overall quality of life and satisfaction with quality of life). Psychological functioning was measured using the GHQ‐28 (self‐reported mood). Cognitive function was measured using the self‐reported Dysexecutive Syndrome Questionnaire and Everyday Memory Questionnaire. Secondary outcomes: None. Outcomes not included: Self‐reported carer mood measured by the GHQ‐28, carer assessment of patient's cognitive function using the Dysexecutive Syndrome Questionnaire and the Everyday Memory Questionnaire, and patient completed Extended Activities of Daily Living Scale.
Notes	Mini‐review 3.   Although participants were not selected on the basis of having cognitive impairment, 95% either reported cognitive problems on the Guy's Neurological Disability Scale or had significant impairment on the Brief Repeatable Battery (evaluates verbal memory, visual memory, attention, and speed of information processing).   Although the assessor is described as masked, the primary outcomes are all described as patient reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Open in a new tab

Maguire 1996.

Methods	A 2 parallel group unmasked RCT of unknown duration.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = D
Participants	People with MS diagnosis recruited through local chapter of MS Society.   Inclusion: (a) "ambulatory", (b) adult   Number screened and number randomised not stated. 33 people analysed. Mean age 45 yrs, 25F, 8M, disease duration not stated, all "ambulatory"
Interventions	Intervention: (Unknown number randomised, 15 analysed): Imagery. Group intervention, 6 one hour sessions, period not stated. Person conducting intervention and site not clear. Session 1‐ general progressive relaxation training. Sessions 2‐6 biologically oriented imagery treatment: Imagining repair of damaged myelin and a positive immune system response. Audiotapes of imagery and relaxation protocols provided for daily practice. Use of MS associated imagery drawings.   Control (number randomised not known, 18 analysed): Continued standard medical treatment
Outcomes	Measured pre‐treatment and immediately post‐treatment (1 week after end of treatment). Primary outcomes: Psychiatric symptoms measured using the State‐Trait Anxiety Inventory (State and Trait subscales). Psychological functioning measured using the Profile of Mood States (tension, depression, anger, vigor, fatigue, confusion subscales). Secondary outcomes: None. Outcomes not included: MS Symptoms Checklist, and the Health Attribution test (beliefs about causes and cures of illness using the internal, powerful others, and chance subscales)
Notes	Mini‐review 3.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Mendoza 2001.

Methods	A 2 parallel group unmasked RCT of 2 months duration.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = A
Participants	People with MS resident in a long‐term nursing facility.   Exclusion: (a) primary admitting diagnosis not MS, (b) poor visual acuity, (c) diagnosis of co‐morbid mental disorder, (d) incomprehensible speech, (e) impaired performance on Kaufman Short Neuropsychological Assessment Procedure (Mental status subtest score).   Number screened not known, 58 excluded, 20 of the remainder randomly selected and randomised.   Mean age 60 years, 6M, 14F, disease duration not stated, all had advanced MS with mean length of time in nursing facility 5 years.
Interventions	Intervention (10 randomised and analysed): Cognitive remediation strategy; Several components including (1) Participants allocated a certified nursing assistant, (2) 4x1hr training of certified nursing assistants on neuroanatomy, project details, rationale for using memory notebooks, review of their allocated patients' treatment protocols, and their cognitive strengths and weaknesses, (3) battery of cognitive tests, (4) participants had memory notebook attached to wheelchair. Participant interviewed daily by nursing assistant to determine their needs and concerns. Other staff encouraged to read and write in notebook.   Control (10 randomised, 9 analysed): No change in treatment regime. Participants not assigned individual nursing assistants.
Outcomes	Measured at baseline and immediately post‐treatment (2 months after baseline). Primary outcomes: Psychiatric symptoms measured using Beck Depression Inventory (self‐reported 21 items). Cognitive functioning measured by the North American New Adult Reading Test, Hopkins Verbal Learning Test and Kaufman Short Neuropsychological Assessment Procedure Mental Status Subtest. Secondary outcomes: None. Outcomes not included: Number of activity events participated in during study.
Notes	Mini‐review 2
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Mohr 2000.

Methods	A 26 week parallel group RCT with one masked outcome.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = A
Participants	People with relapsing MS newly started on interferon beta‐1a and with depression. Inclusion: (a) 15 or more on Profile of Moods Depression‐Dejection Scale, and (b) if currently under medical or psychological treatment for depression then must have been so for 3 months or more and intend to continue treatment for duration of study.   Exclusion: (a) Dementia, and (b) other neurological disorder.   73 people screened, 32 randomised. Mean age 42 yrs (SD approx 11), 9M 23F, mean disease duration 6 yrs (SD 7 yrs), mean Guy's Neurological Disability Scale=18 (SD=9), 3 use wheelchair, 11 walk with aids, 18 walk unaided.   3 taking anti‐depressants, 2 receiving psychotherapy
Interventions	Intervention (16 randomised, 11 completed, 16 analysed): Cognitive behavioural therapy. 8 weekly 50 min one‐to‐one sessions over the telephone to participant's home. Conducted by 3 advanced Doctoral students in psychology and 1 post‐doc fellow. All had 2‐5 years of therapy experience and had previously seen MS patients. Based on CBT model originally intended for depressed older adults and modified for use with MS patients. Therapist manual and patient workbook produced. In the first 2 sessions patients identified 3 specific, measurable, achievable goals. In the next 3 sessions they were taught thought monitoring, identification of cognitive distortions and how to challenge them. Additional modules were then available, including increasing pleasant events (n=6 patients) and fatigue management (n=3 patients). The final session addressed the degree to which patients' goals were met, continued care and maintenance of gains.   Control (16 randomised, 12 completed, 16 analysed): Usual care. No additional mental health care.
Outcomes	Measured at baseline, immediately post‐treatment (8‐9 weeks after start of treatment) and short‐term follow‐up (4 months after end of treatment). Primary outcome: Psychiatric symptoms measured by telephone administered Profile of Mood States, Depression‐Rejection Scale (baseline and immediately post‐treatment only; authors provide justification for its use as a measure of depressive symptomatology). Secondary outcome: Adherence to other therapies measured by adherence to Interferon Beta‐1a for the 6 months post randomisation.
Notes	Mini‐review 4.   Non‐completers included in analyses by using their baseline data.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Mohr 2001.

Methods	A 42 week 3 arm parallel group unmasked RCT. Participants in one of the arms of the trial were given Supportive‐Expressive Group therapy. However, this was not a randomised group and has not been considered further.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = B
Participants	People with MS with moderate depression.   Inclusion: (a) Diagnosis of MS (Poser definition), (b) Relapse‐remitting or secondary progressive disease course, (c) Diagnosis of major depressive disorder (DSM‐IV), (d) 16 or more on Hamilton Rating Scale for Depression, (e) 16 or more on Beck Depression Inventory, (f) Willing to abstain from additional psychological or pharmacological treatment for depression.   Exclusions: (a) Other psychological disorder, (b) Dementia, (c) Severe suicidality, (d) Recent corticosteroid treatment or Interferon, (e) Current MS exacerbation, (f) Other CNS disorders, (g) Pregnancy.   63 patients recruited and randomised from 481 screened for depression.   Mean age (SD)=43.9 yrs (10.0) , 73% female, average disease duration (range) 7.7 years (0.3, 31.2), mean (range) EDSS 2.4 (0, 8.0). 41 patients randomised to the 2 eligible treatment arms.
Interventions	Intervention (20 randomised and included in analysis): Cognitive behavioural therapy adapted from manualised model used in older adults. Weekly 50 minute sessions for 16 weeks by one of four post‐doc psychologists. Included behavioural activation and cognitive restructuring together with specific skills for the management of MS related symptoms and problems (fatigue, mild cognitive impairment, pain, stress, skills for intimacy, communication and sexual dysfunction, and social difficulties).   Control (21 randomised and included in analysis): Serotonin‐Specific Reuptake Inhibitor (Sertraline), 50‐200mg per day for 16 weeks. Includes 15 min dose finding interview with post‐doc psychologist every 4 weeks.
Outcomes	Measured at baseline, during treatment at 4, 8, and 12 weeks (not included in review), immediately post‐treatment (at end of treatment), and short‐term follow‐up (6 months post‐treatment). Primary outcomes: Psychiatric symptoms measured by Beck Depression Inventory (self‐reported), Hamilton Rating Scale for Depression (HRSD) (clinician rated), reduction of 50% or more in HRSD (immediately post‐treatment only), Structured Clinical Interview for DSM‐IV Axis 1 Disorders (immediately post‐treatment only), and Longitudinal Interval Follow‐up Evaluation (during period of short‐term follow‐up only). Secondary outcomes: Healthcare utilisation measured by uptake of psychotherapy up to short‐term follow‐up. Change in medication usage measured by antidepressant use during the period of short‐term follow‐up.
Notes	Mini‐review 4.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Mohr 2003.

Methods	Same study as Mohr 2001
Participants	Same as Mohr 2001, although inclusion criteria (c) and (d) not applied (i.e. study includes a small number of participants with "sub‐syndromal depression")
Interventions	See Mohr 2001
Outcomes	Measured at baseline and immediately post‐treatment (at end of treatment). Secondary outcomes: Fatigue measured by the Fatigue Assessment Instrument, including Total Score, and 4 subscales; Global Fatigue Severity, Fatigue Consequences, Situation‐Specific Fatigue, and Responsiveness to Rest or Sleep.
Notes	This paper is based on the same study as Mohr 2001, and reports data on fatigue. As stated in the "Participants" section, the study includes a small number of people who were excluded from the original Mohr 2001 paper.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Rigby 2003.

Methods	A 55 week 3 arm parallel group unmasked randomised controlled trial .   Quality:   Randomisation generation = B   Allocation concealment = B   Withdrawals = A.
Participants	People with MS recruited via an MS Clinic, either using an MS research database or from clinic visits.   Inclusion: (a) Neurologist diagnosis of MS for at least 6 months, (b) age 20‐65 years.   Exclusion: (a) Study co‐ordinator or doctor felt person would be unable to participate in sessions or complete study questionnaires.   219 assessed for eligibility, 53 allocated to a non‐randomised group (not included here), 147 randomised.   Type of MS unknown, mean age 44 yrs (SD 10), 91F, 56M, mean disease duration 9 yrs (SD 8), median EDSS in the 3 groups between 5 and 6.
Interventions	Intervention (49 randomised, 43 analysed): Brief coping‐focussed cognitive therapy, based on coping theory and cognitive behavioural models, and "MS on your Mind" booklet (given during session 1). 3 x weekly 1.5 hour sessions run in groups of less than 6 by a chartered health psychologist in an MS Clinic setting. Intervention designed specifically for people with MS. Objectives were (a) provide a safe forum for discussing MS issues, (b) recognise and vent feelings, (c) create awareness of own coping strategies, (d) consider different types of coping strategies to help with problems, (e) increase perceived control of situations, and increase confidence, (f) to help understand that feelings experienced are "normal". Session 1: Meeting the group and sharing experiences. Session 2: Flexible coping. Session 3: Feeling good and developing a sense of control. Control 1 (46 randomised, 40 analysed): Social discussion group and "MS on your Mind" booklet (given during first session). 3 x weekly 1.5 hour sessions with up to 6 per group run by study co‐ordinator at MS clinic. No formal structure; discussion led by participants. The aim was to enable people with MS to meet others and give them the opportunity to talk about their condition. Control 2 (52 randomised, 48 analysed): Booklet "MS on your mind" (mental and emotional aspects of MS produced by the UK MS Society) sent via the post.
Outcomes	Measured at baseline, immediately post‐treatment (1 week post‐treatment), short‐term follow‐up (1, 3 and 6 months post‐treatment), and medium term follow‐up (12 months post‐treatment). Primary outcomes: Psychiatric symptoms measured using the Hospital Anxiety and Depression Scale (self‐reported anxiety and depression subscales). Psychological function measured using the Dispositional Resiliency Index (self‐reported total score, and commitment, control, and challenge subscales), the MS Self‐efficacy Scale (MSSS), and the MS Self Efficacy Scale (MSSE) (control subscale).
Notes	Mini‐review 3.   Not all participants replied to all follow‐ups; 133 provided at least some data. 106, 100, 98, 101, 90 supplied data at 1 wk, 1 month, 3 months, 6 months, 12 months respectively.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Schwartz 1999.

Methods	A 104 week parallel group unmasked RCT.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = B
Participants	People with MS. Inclusion: (a) Neurologist confirmed, (b) EDSS 1‐8.5.   Exclusion: (a) "Actively psychotic" during interview.   172 interviewed, 136 randomised, 124 completed interventions, 132 supplied data at 2 yr follow‐up, 132 analysed.   76 chronic progressing MS, 56 relapse‐remitting MS. Mean age 43 yrs (SD=9), 34M, 98F, 30% currently treated for depression, 11% previous psychiatric admission, 24% current emotional problems, 40% ever treated for depression. Mean disease duration 8 yrs (SD 7), mean EDSS 4.7 (SD=1.8, range 1‐8.5), and 46% require walking aids.
Interventions	Intervention 1 (number randomised not known, 64 analysed): Directive coping skills group sessions followed by telephone support from a matched coping partner identified from within the group. One 2 hr session weekly for 8 weeks followed by 1 telephone call per month for 10 months. Group hospital‐based therapy by health professional in public health and clinical psychology. Teaching sessions on emotional difficulties, goal setting, dealing with cognitive deficits, improving communication with caregivers and social network. Participants were requested to bring a spouse, close friend or relative to the sessions on social networks.   Intervention 2 (number randomised not known, 68 analysed): Non‐directive peer telephone support from lay people with MS trained in active listening. Based on Rogerian client‐centred psychotherapy. One phone call (15 mins on average) per month for 12 months. Supporters matched by age, interests and level of disability and instructed not to provide advice. Supporters received monthly supervision.
Outcomes	Measured at baseline, 2 months (during treatment and not included in the review), immediately post‐treatment (12 months after start of treatment), short term follow‐up (6 months after end of treatment), and medium term follow‐up (12 months after end of treatment). Primary outcomes: General quality of life measured using the Sickness Impact Profile (psychosocial and physical limitations subscales), and the Quality of Life Index (Satisfaction with health, family aspects, spiritual aspects, socio‐economic status, and global subscales). Psychiatric symptoms were measured using the depression and anxiety subscales of the Arthritis Impact Measurement Scale. Psychological functioning was measured using the MS Self Efficacy scale (control and function subscales), the Multidimensional Health Locus of Control (internal subscale), the Ways of Coping Checklist (problem solving, social support, reframing, religiosity, blaming others, blaming self, avoidance, wishful thinking subscales), and the Ryff Happiness Scale (Purpose in life, personal growth, environmental mastery, social relatedness, self‐acceptance subscales). Disability measured by the Expanded Disability Status Scale (immediately post‐treatment only). Cognitive function was measured using a composite score of cognitive tests (verbal and spatial memory, complex attention, verbal fluency etc.) adjusted for age, educational status and gender (immediately post‐treatment only). Secondary outcome: Fatigue measured by the Multidimensional Assessment of Fatigue Scale. Not included: Social activity subscale of the Arthritis Impact Measurement Scale.
Notes	Mini‐review 3.   Discrepancy of 1 case in description of drop‐outs.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Solari 2004.

Methods	A 24 week parallel group double masked RCT.   Quality:   Randomisation generation = A   Allocation concealment = A   Withdrawals = A
Participants	People with MS with mild to moderate cognitive impairment recruited via neurologists in 6 centres.   Inclusion: (a) Poser definition, (b) complained of poor attention or memory, (c) score <80th percentile in 2 or more components of Brief Repeatable Battery of Neuropsychological tests.   Exclusion: (a) <18 or >65 yrs old, (b) Mini‐Mental State Examination <24, (c) education of <8yrs duration,   (d) ongoing major psychiatric disorder, (e) exacerbation in past 3 months, (f) immunomodulant/ immunosuppressant treatment initiation in last 4 months, (g) cognitive rehabilitation in last 6 months.   203 assessed for eligibility, 82 randomised, 77 started treatment, 75 followed‐up at 8 and 16 weeks and analysed.   3 with primary progressive MS, 35 with secondary progressive, and 39 with relapse‐remitting. Mean age 44yrs (range 22‐65 yrs), 49F, 28M, people with major psychiatric diagnosis excluded, mean age at diagnosis 29 yrs (range 19‐53), median EDSS 3.5 (range 1.5‐7.0).
Interventions	Intervention (42 randomised, 40 received treatment, 38 analysed): Computer aided retraining of memory and attention. 16 sessions of 45 minutes duration over 8 weeks. Individual, outpatient‐based therapy conducted with a trained psychologist. Used the "memory" and "attention" modules of the RehaCom software package. A special keyboard was employed in order to limit the impact of motor co‐ordination on the results.   Control (40 randomised, 37 treated, 37 analysed): Sham computer aided retraining of memory and attention. Exactly the same as above except using (simplified versions of) the "visuo‐constructional" and "visuo‐motor co‐ordination" retraining modules. These modules are similar in appearance to the ones used in the intervention, but train mainly motor skills rather than memory and attention.
Outcomes	Measured at baseline, immediately post‐treatment (8 weeks after baseline), and short‐term follow‐up (16 weeks after baseline). Primary outcomes: Disease specific quality of life was measured using the MSQOL‐54 (mental health subscale and cognitive subscale). Psychiatric symptoms were measured using the Chicago Mood Depression Inventory (mood subscale). Cognitive function was measured using the Brief Repeatable Battery of Neuropsychological tests (Buschke Selective Reminding Test ‐ consistent long‐term retrieval and delayed recall scores (measure of verbal learning and recall), Symbol Digit Modalities Test (measure of sustained attention and information processing speed), Paced Auditory Serial Addition Test (measure of complex attention and concentration), Word List Generation (measure of verbal fluency and sustained attention), 10/36 Spatial Recall Test‐ immediate recall and delayed recall (measures visuospatial learning and recall), and an increase of 20% or more from baseline in at least 2 of these tests).
Notes	Mini‐review 1
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Open in a new tab

Stuifbergen 2003.

Methods	A 2 parallel group unmasked RCT of 32 weeks duration.   Quality:   Randomisation generation = A   Allocation concealment = C   Withdrawals = B
Participants	Women with MS. Recruitment via MS newsletters, posters in neurologists' offices, support groups, an MS adaptive fitness day, previous research participants. Inclusion: (a) Female, (b) physician diagnosis of MS of at least 6 months duration, (c) age 20‐70 years. Exclusion: (a) pregnancy (b) concurrent medical conditions making changes in diet or exercise contra‐indicated.   142 randomised, 121 started treatment (inclusion criteria applied and consent sought post‐randomisation), 113 analysed. Based on 113 participants, mean (SD) age = 46 (10) yrs, 113F, 0M, mean (SD) duration = 11 (7) yrs, mean (SD) Incapacity Status Scale = 16 (7) points.
Interventions	Intervention: (76 randomised, 56 analysed): Wellness programme in 2 parts. Part 1: Educational skills‐based change programme. Group intervention, 8 x 90 minute sessions over 8 weeks. Conducted by clinical nurse specialist assisted by 1 person with MS with a background in health promotion. Intervention site not stated. Aims to educate participants about health promotion strategies in MS; enhance and support strategies to increase self‐efficacy for health behaviours; guide participants in self‐assessment of resources, barriers and behaviours. Based on health promotion and health psychology models of health behaviour and beliefs. Topics included maximising health in chronic illness, lifestyle adjustment, stress management strategies, healthy eating, exercise and physical activity, intimacy and sexuality, women's health issues. Included homework assignments in goal setting, self‐assessment, and summary of class content. Specially designed for and with people with MS. Part 2: 3 months of telephone support, 2 times per month by nurse specialist. Positive feedback for accomplishments, verbal persuasion, sharing success of other women with participants. Aims to increase self‐efficacy and encourage progress towards goals.   Control group. Put on 8 month waiting list for the intervention.
Outcomes	Measured pre‐treatment, immediately post‐treatment (5 months after start of treatment), and short‐term follow‐up (3 months post‐treatment). All were self‐reported measures.   Primary outcome: General quality of life measured using the SF‐36 (9 scales: general health, physical function, physical role, bodily pain, vitality, social functioning, emotional role, and mental health). Psychological functioning was measured using the Self Rated Abilities for Health Practices Scale (a measure of self‐efficacy covering nutrition, physical activity/ exercise, psychologic well‐being, and responsible health practices). Measure of disability using the Incapacity Status Scale (adapted by authors and not measured immediately post‐treatment). Secondary outcomes: None. Not included: Barriers to Health Promoting Activities for Disabled Persons Scale, the Personal Resource Questionnaire (a measure of social support), the Health Promoting Lifestyle Profile II (measure of health promoting behaviours).
Notes	Mini‐review 3.   The application of eligibility criteria, and obtaining of informed consent was done after participants had been randomised.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	High risk	C ‐ Inadequate

Open in a new tab

Wassem 2003.

Methods	A 208 week parallel group unmasked RCT.   Quality:   Randomisation generation = D   Allocation concealment = B   Withdrawals = C
Participants	People with MS recruited via an MS workshop, physicians, and an MS newsletter advert.   Inclusion: (a) MS   Number approached not known, 27 randomised, number analysed not known.   Type of MS not stated, mean age (range) = 44 yrs (18‐54), 72% female, mean (range) disease duration = 3.5 yrs (0.5‐7), mean (range) Modified Disability Status Scale = 3.4 (0‐9),
Interventions	Intervention (number randomised and number analysed not stated): MS‐REHAB program: 2 hrs once per week for 4 weeks. Group‐based therapy run by a nurse and held at a College of Nursing. Therapy developed from the literature and experience, and based on Bandura's Social Cognitive theory with an emphasis on self‐efficacy. The programme aimed to increase self‐efficacy for adjustment behaviours, and increase outcome expectations. This, it is theorised, will lead to increased use of adjustment behaviours, and hence increased adjustment to MS, increased symptom management, well‐being and activities. Behaviour acquisition using verbal persuasion, role modelling, performance accomplishment and vicarious experience. Week1:‐disease process, influencing factors, stretching and exercise, pharmacological and non‐pharmacological managment of symptoms, communicating with health professional, stress and relaxation. Week 2:‐ relaxation, diet, strategies for managing fatigue, exercise, stretching. Week 3:‐psychosocial factors, reaction of others, employment, sources of help, sleep, exercise, stretching, cognitive dysfunction. Week 4: Cognitive dysfunction, memory, impact on life, medication. Homework tasks used throughout.   Control group (number randomised and number analysed not known): Not given MS‐REHAB program.
Outcomes	Measured at baseline, immediately post‐treatment, short‐term follow‐up, medium‐term follow‐up and long‐term follow‐up (10 occasions in total up to 4 years, although exact timing not clear). Primary outcomes: Psychological function measured using the total Psychosocial Adjustment to Illness Scale (modified by dropping 2 subscales), a summated total adjustment score (derived from 3 visual analogue scores of psychological, social and physical adjustment), and the Self‐Efficacy for Adjustment Behaviours Scale. Secondary outcomes: Pain using a visual analogue scale, fatigue using a visual analogue scale. Not used: a summated pain, fatigue and sleep score based on 3 visual analogue scales.
Notes	Mini‐review 3.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Open in a new tab

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Klarskov 1994	This study is a randomised controlled trial, and assesses an intervention that is eligible for the review. However none of the outcomes measured (severity of bladder symptoms, count of incontinence episodes, flowmetry, residual urine measurement, water cystometry etc.) was relevant to the review.
Mendozzi 1998	This study compared three arms; computer assisted memory retraining; non‐specific computer assisted retraining; and no computer retaining. The first 30 participants were randomised to the 3 groups. The next 30 were placed in groups by the principal investigator "to balance the groups as much as possible for sex, age and education". This group of 30 does not appear to have been randomised.
Mills 2000	This study compared mindfulness of movement as a coping strategy for MS symptoms with standard care. The intervention has psychological aspects to it, and is conducted by a psychologist. However, it is conducted within the context of Tai Chi, and incorporates elements of balance, stretching, posture, breathing and co‐ordination. Therefore the intervention falls outside our definition.
Mohr 2004	This paper reports on social support outcome measures from the study described in the Mohr 2001 paper (see list of included studies). Social support was not among the primary or secondary outcome measures for the review, and so this paper has been excluded.
Rogers 1996	Study was described as quasi‐experimental. The waiting list control group was not randomised and allocation was partly based on participants' willingness to be placed on a waiting list.
Schmid 2004	This study compared active music therapy with being put on a waiting list for therapy. This intervention had a strong psychological basis, but was not included because group allocation did not appear to be randomised.
Tesar 2003	This study evaluates a psychological intervention that incorporates cognitive/behavioural strategies, stress‐coping training, and exercises that exert an influence on body image. Participants were not randomised; the first 15 participants in the study were allocated to the psychological intervention and the next 15 to the control group. Therefore this study has been excluded.

Open in a new tab

Characteristics of ongoing studies [ordered by study ID]

Mohr.

Trial name or title	Telephone administered psychotherapy for the treatment of depression in MS
Methods
Participants	People with MS with depression.   Inclusion: (a) Beck Depression Inventory > 16.   Proposed sample size 128 in total
Interventions	Intervention: Cognitive behavioural therapy over the telephone. 16 weekly sessions delivered by a PhD psychologist.   Control: Supportive expressive therapy over the telephone. 16 weekly sessions administered by a PhD psychologist.
Outcomes	Primary outcomes: Psychiatric symptoms measured using the Beck Depression Inventory, the Structured Clinical Interview for DSM‐IV Axis 1 Disorders, and the Hamilton Depression Scale.
Starting date	2001
Contact information	David C. Mohr   dmohr@itsa.ucsf.edu
Notes

Open in a new tab

Moss‐Morris.

Trial name or title	A randomised controlled trial of cognitive behavioural therapy for fatigue in patients with multiple sclerosis
Methods
Participants	People with MS with fatigue.   Inclusion: (a) Definite diagnosis of MS, (b) EDSS of 5.5 or less, (c) Score of 4 or more on Chalder fatigue scale, (d) willing not to start new psychological or pharmacological treatment during study.   Exclusion: (a) if taking Interferon then excluded if for less than 3 months, (b) medication for depression for less than 2 months, (c) dementia, (d) serious psychological disorder, (e) other serious chronic disease that might contribute to fatigue.   Proposed sample size is 70 in total.
Interventions	Intervention: Cognitive behavioural therapy. 8 x one hour weekly sessions by a CBT therapist, and supplemented with information leaflets and homework. Aim to show patients that activity can be increased steadily and safely without exacerbating symptoms, and that increasing activity itself is helpful for fatigue. Session 1: Rationale for CBT, assess impact of fatigue, complete self‐monitoring diaries. Session 2‐3: Activity scheduling, activity diaries, increasing activity, importance of exercise, and target setting. Sessions 4‐5: Understanding MS symptoms and using cognitive strategies to help evaluate and interpret them. Sessions 6‐7: Managing controllable and uncontrollable stressors, and other issues of personal control. Session 8: Dealing with setbacks, action plans.   Control: Relaxation therapy. 8 x 1 hour weekly sessions supplemented by information leaflets and homework. Teaching of different methods of relaxation and meditation.
Outcomes	Measured at pre‐treatment, immediately post‐treatment, and short term follow‐up (3 and 6 months after treatment).   Primary outcome measures: General quality of life using the SF‐36, psychiatric symptoms using the Hospital Anxiety and Depression Scale.   Secondary outcome measures: Fatigue measured by the Chalder Fatigue Scale, the Pilot Alertness Test, and the Epworth Sleepiness Scale.
Starting date	1st April 2002
Contact information	Dr Rona Moss‐Morris, Health Psychology, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland.
Notes

Open in a new tab

Differences between protocol and review

Since the original publication of the protocol, the following change has taken place. One measure of study quality has been dropped. Originally studies were going to be classified according to whether they used intention to treat principles; A (no withdrawals), B (withdrawals excluded), C (withdrawals included in the analysis by estimating outcome), D (withdrawals followed‐up and included in the analysis). In practice this system proved difficult to apply because some studies could not be classified uniquely (for example, in a single study some withdrawals might have been included, and some excluded). Also, there were different types of withdrawals (e.g. withdrawal from providing data, withdrawal from treatment, permanent withdrawal, temporary withdrawal), again making classification difficult.

Contributions of authors

PWT conceived the idea, managed the review, and took the lead in writing the review. PWT and ST designed the review, prepared the protocol, and wrote the review. KG, RB, and CH contributed to the design of the protocol and reviewed the scientific content. PWT, ST, KG and CH have read and appraised the papers for the review. RB and ST provided specialist psychology advice and support to the review.

Sources of support

Internal sources

Dorset Research and Development Support Unit, Poole Hospital, UK.
Institute of Health and Community Studies, Bournemouth University, UK.
Poole Hospital NHS Trust, UK.

External sources

No sources of support supplied

Declarations of interest

None

Edited (no change to conclusions)

References

References to studies included in this review

Benedict 2000 {published data only}

Benedict RHB, Shapiro A, Priore R, Miller C, Munschauer F, Jacobs L. Neuropsychological counseling improves social behavior in cognitively‐impaired multiple sclerosis patients. Multiple Sclerosis 2000;6:391‐6. [DOI] [PubMed] [Google Scholar]

Crawford 1985 {published data only}

Crawford JD, McIvor GP. Group psychotherapy: Benefits in multiple sclerosis. Archives of Physical Medicine and Rehabilitation 1985;66:810‐3. [PubMed] [Google Scholar]

Foley 1987 {published data only}

Foley FW, Bedell JR, LaRocca NG, Scheinberg LC, Reznikoff M. Efficacy of stress‐inoculation training in coping with multiple sclerosis. Journal of Consulting and Clinical Psychology 1987;55:919‐22. [DOI] [PubMed] [Google Scholar]

Gordon 1997 {published data only}

Gordon PA, Lam CS, Winter R. Interaction strain and persons with multiple sclerosis: Effectiveness of a social skills program. Journal of Applied Rehabilitation Counselling 1997;28:5‐11. [Google Scholar]

Jonsson 1993 {published data only}

Jonsson A, Korfitzen EM, Heltberg A, Ravnborg MH, Byskov‐Ottosen E. Effects of neuropsychological treatment in patients with multiple sclerosis. Acta Neurologica Scandinavica 1993;88:394‐400. [DOI] [PubMed] [Google Scholar]

Larcombe 1984 {published data only}

Larcombe NA, Wilson PH. An evaluation of cognitive‐behaviour therapy for depression in patients with multiple sclerosis. British Journal of Psychiatry 1984;145:366‐71. [DOI] [PubMed] [Google Scholar]

Lincoln 2002 {published data only}

Lincoln NB, Dent A, Harding J, Weyman N, Nicholl C, Blumhardt LD, et al. Evaluation of cognitive assessment and cognitive intervention for people with multiple sclerosis. Journal of Neurology Neurosurgery & Psychiatry 2002;72:93‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Maguire 1996 {published data only}

Maguire, B. The effects of imagery on attitudes and moods in multiple sclerosis patients. Alternative Therapies 1996;2:75‐9. [PubMed] [Google Scholar]

Mendoza 2001 {published data only}

Mendoza RJ, Pittenger DJ, Weinstein CS. Unit management of depression of patients with multiple sclerosis using cognitive remediation strategies: A preliminary study. Neurorehabilitation and Neural Repair 2001;15:9‐14. [DOI] [PubMed] [Google Scholar]

Mohr 2000 {published data only}

Mohr DC, Likosky W, Bertagnolli A, Goodkin DE, Wende J, Dwyer P, et al. Telephone‐administered cognitive behavioral therapy for the treatment of depressive symptoms in multiple sclerosis. Journal of consulting and Clinical Psychology 2000;68(2):356‐61. [DOI] [PubMed] [Google Scholar]

Mohr 2001 {published data only}

Mohr DC, Boudewyn AC, Goodkin DE, Bostrom A, Epstein L. Comparative outcomes for individual cognitive‐behavior therapy, supportive‐expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. Journal of Consulting and Clinical Psychology 2001;69(6):942‐9. [PubMed] [Google Scholar]

Mohr 2003 {published data only}

Mohr DC, Hart SL, Goldberg A. Effects of treatment for depression on fatigue in multiple sclerosis. Psychosomatic Medicine 2003;65:542‐547. [DOI] [PubMed] [Google Scholar]

Rigby 2003 {published data only}

Rigby, S. Coping with MS. Liverpool: University of Liverpool, 2003. [Google Scholar]

Schwartz 1999 {published data only}

Schwartz CE. Teaching coping skills enhances quality of life more than peer support: Results of a randomised trial with multiple sclerosis patients. Health Psychology 1999;18:211‐20. [DOI] [PubMed] [Google Scholar]

Solari 2004 {published data only}

Solari A, Motta A, Mendozzi L, Pucci E, Forni M, Mancardi G, Pozzilli C. Computer‐aided retraining of memory and attention in people with multiple sclerosis: a randomized, double‐blind controlled trial. Journal of the Neurological Sciences 2004;222:99‐104. [DOI] [PubMed] [Google Scholar]

Stuifbergen 2003 {published data only}

Stuifbergen AK, Becker H, Blozis S, Timmerman G, Kullberg V. A randomized clinical trial of a wellness intervention for women with multiple sclerosis. Archives of Physical Medicine and Rehabilitation 2003;84:467‐76. [DOI] [PubMed] [Google Scholar]

Wassem 2003 {published data only}

Wassem R, Dudley W. Symptom management and adjustment of patients with multiple sclerosis. Clinical Nursing Research 2003;12:102‐17. [DOI] [PubMed] [Google Scholar]

References to studies excluded from this review

Klarskov 1994 {published data only}

Klarskov P, Heely E, Nyholdt I, Rottensten K, Nordenbo A. Biofeedback treatment of bladder dysfunction in multiple sclerosis. A randomised trial.. Scandanavian Journal of Urology and Nephrology 1994;Suppl. 157:61‐65. [PubMed] [Google Scholar]

Mendozzi 1998 {published data only}

Mendozzi L, Pugnetti L, Motta A, Barbieri E, Gambini A, Cazzullo CL. Computer‐assisted memory retraining of patients with multiple sclerosis. Italian Journal of Neurological Science 1998;19:S431‐8. [Google Scholar]

Mills 2000 {published data only}

Mills N, Allen J. Mindfulness of movement as a coping strategy in multiple sclerosis. A pilot study. General Hospital Psychiatry 2000;22:425‐31. [DOI] [PubMed] [Google Scholar]

Mohr 2004 {published data only}

Mohr DC, Classen C, Barrera M. The relationship between social support, depression and treatment for depression in people with multiple sclerosis. Psychological Medicine 2004;34:533‐41. [DOI] [PubMed] [Google Scholar]

Rogers 1996 {published data only}

Rogers D, Khoo K, MacEachen M, Oven M, Beatty WW. Cognitive therapy for multiple sclerosis: A preliminary Study. Alternative Therapies 1996;2(5):70‐4. [PubMed] [Google Scholar]

Schmid 2004 {published data only}

Schmid W, Aldridge D. Active music therapy in the treatment of multiple sclerosis patients: A matched control study. Journal of Music Therapy 2004;41:225‐40. [DOI] [PubMed] [Google Scholar]

Tesar 2003 {published data only}

Tesar N, Baumhackl U, Kopp M, Gunther V. Effects of psychological group therapy in patients with multiple sclerosis. Acta Neurologica Scandinavica 2003;107:394‐9. [DOI] [PubMed] [Google Scholar]

References to studies awaiting assessment

Chiaravalloti 2005 {published data only}

Chiaravalloti ND, DeLuca J, Moore NB, Ricker JH. Treating learning impairments improves memory performance in multiple sclerosis: a randomized clinical trial. Multiple Sclerosis 2005;11:58‐68. [DOI] [PubMed] [Google Scholar]

Mohr 2005 {published data only}

Mohr DC, Cox D, Merluzzi N. Self‐injection anxiety training: a treatment for patients unable to self‐inject injectable medications. Multiple Sclerosis 2005;11:182‐5. [DOI] [PubMed] [Google Scholar]

References to ongoing studies

Mohr {published data only (unpublished sought but not used)}

Telephone administered psychotherapy for the treatment of depression in multiple sclerosis. Personal Communication.

Moss‐Morris {published data only (unpublished sought but not used)}

A randomised controlled trial of cognitive behavioural therapy for fatigue in patients with multiple sclerosis. Personal Communication.

Additional references

Aikens 1997

Aikens JE, Fischer JS, Namey M, Rudick RA. A replicated prospective investigation of life stress coping and depressive symptoms in multiple sclerosis. Journal of Behavioral Medicine 1997;20:433‐45. [DOI] [PubMed] [Google Scholar]

Amato 1995

Amato MP, Ponziani G, Pracucci G, Siracusa G, Amaducci L. Cognitive impairment in early‐onset multiple sclerosis. Archives of Neurology 1995;52:168‐72. [DOI] [PubMed] [Google Scholar]

Barnes 1954

Barnes RH, Busse EW, Dinken H. The alleviation of emotional problems in multiple sclerosis by group therapy. Group Psychotherapy 1954;6:193‐201. [Google Scholar]

Barnwell 1997

Barnwell AM, Kavanagh DJ. Prediction of psychological adjustment to multiple sclerosis. Social Science and Medicine 1997;45:411‐8. [DOI] [PubMed] [Google Scholar]

Bedford 1978

Bedford A, Foulds G. Delusions‐Symptoms‐States‐Inventory. Windsor: NFER‐Nelson, 1978. [Google Scholar]

Bolding 1960

Bolding H. Psychotherapeutic aspects in the management of patients with multiple sclerosis. Diseases of the Nervous System 1960;21:24‐6. [PubMed] [Google Scholar]

Brassington 1998

Brassington JC, Marsh NV. Neuropsychological aspects of multiple sclerosis. Neuropsychological Review 1998;8:43‐77. [DOI] [PubMed] [Google Scholar]

Clarke 2000

Clarke M, Oxman AD. Cochrane Reviewer's Handbook Version 4.1.1 (updated December 2000); Section 6. The Cochrane Library. Oxford: Update Software, 2001, issue 1.

Crawford 1987

Crawford JD, McIvor GP. Stress management for multiple sclerosis patients. Psychological Reports 1987;61:423‐429. [DOI] [PubMed] [Google Scholar]

Dalos 1983

Dalos NP, Rabins PV, Brooks BR, O'Donnell P. Disease activity and emotional state in multiple sclerosis. Annals of Neurology 1983;13:573‐83. [DOI] [PubMed] [Google Scholar]

Day 1953

Day M, Day E, Herman R. Group therapy of patients with multiple sclerosis. Archives of Neurology and Psychiatry 1953;69:193‐6. [DOI] [PubMed] [Google Scholar]

EuroQoL Group 1990

EuroQoL Group. EuroQoL: a new facility for the measurement of health related quality of life. Health Policy 1990;16:199‐208. [DOI] [PubMed] [Google Scholar]

Fassbender 1998

Fassbender K, Schmidt R, Mößner R, Kischka U, Kühnen J, Schwartz A, et al. Mood disorders and dysfunction of the hypothalamic‐pituitary‐adrenal axis in multiple sclerosis. Archives of Neurology 1998;55:66‐72. [DOI] [PubMed] [Google Scholar]

Feinstein 1999

Feinstein A, O'Connor P, Gray T, Feinstein K. The effects of anxiety on psychiatric morbidity in patients with multiple sclerosis. Multiple Sclerosis 1999;5:323‐6. [DOI] [PubMed] [Google Scholar]

Franklin 1988

Franklin GE, Heaton RK, Nelson LM, Filley CM, Seibert C. Correlation of neuropsychological and MRI findings in chronic/progressive multiple sclerosis. Neurology 1988;38:1826‐9. [DOI] [PubMed] [Google Scholar]

Gilchrist 1994

Gilchrist AC, Creed FH. Depression, cognitive impairment and social stress in multiple sclerosis. Journal of Psychosomatic Medicine 1994;38:193‐201. [DOI] [PubMed] [Google Scholar]

Graham 2002

Graham J. What is MS?. Multiple Sclerosis Society, UK. 2002:3.

Gulick 1997

Gulick EE. Correlates among quality of life among persons with multiple sclerosis. Nursing Research 1997;46:305‐11. [DOI] [PubMed] [Google Scholar]

Hartings 1976

Hartings MF, Pavlou MM, Davis FA. Group counseling of MS patients in a program of comprehensive care. Journal of Chronic Diseases 1976;29:65‐73. [DOI] [PubMed] [Google Scholar]

Heaton 1985

Heaton RK, Nelson LM, Thompson DS, Burks JS, Franklin GM. Neuropsychological findings in relapse‐remitting multiple sclerosis. Journal of Consulting and Clinical Psychology 1985;53:103‐10. [DOI] [PubMed] [Google Scholar]

Hobart 2001

Hobart J, Lamping D, Fitzpatrick R, Riazi A, Thompson A. The Multiple Sclerosis Impact Scale (MSIS‐29). A new patient‐based outcome measure. Brain 2001;124:962‐73. [DOI] [PubMed] [Google Scholar]

Joffe 1987

Joffe RT, Lippert GP, Gray T, Sawa G, Horwath Z. Mood disorder and multiple sclerosis. Archives of Neurology 1987;44:376‐8. [DOI] [PubMed] [Google Scholar]

Kurtzke 1983

Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444‐52. [DOI] [PubMed] [Google Scholar]

Langdon 1996

Langdon DW, Thompson AJ. Cognitive problems in multiple sclerosis. MS Management 1996;3:5‐9. [Google Scholar]

LaRocca 1993

LaRocca NG, Kalb RC, Foley FW, Caruso LS. Assessment of psychosocial outcomes. Journal of Neurological Rehabilitation 1993;7:109‐16. [Google Scholar]

LaRocca 1996

LaRocca NG, Ritvo PG, Miller DM, Fischer JS, Andrews H, Paty DW. "Quality of life" assessment in multiple sclerosis clinical trials: current status and strategies for improving multiple sclerosis clinical trial design. In: Goodkin DE, Rudick RA editor(s). Multiple sclerosis: advances in clinical trial design, treatment and future perspectives. London: Springer‐Verlag, 1996:145‐60. [Google Scholar]

Maurelli 1992

Maurelli M, Marchioni E, Cerretano R, Bosone D, Bergamaschi R, Citterio A, et al. Neuropsychological assessment in MS: Clinical, neurophysiological and neuroradiological relationships. Acta Neurological Scandinavica 1992;86:124‐8. [DOI] [PubMed] [Google Scholar]

McIvor 1984

McIvor GP, Riklan M, Reznikoff M. Depression in multiple sclerosis as a function of length and severity of illness, age, remissions and perceived social support. Journal of Clinical Psychology 1984;40:1028‐33. [DOI] [PubMed] [Google Scholar]

McNair 1981

McNair DM, Lorr M, Droppleman LF. Profile of Mood States Manual. San Diego: Educational and Industrial Testing Service, 1981. [Google Scholar]

Miles 1979

Miles A. Some psycho‐social consequences of multiple sclerosis: problems of social interaction and group identity. British Journal of Medical Psychology 1979;52:321‐31. [DOI] [PubMed] [Google Scholar]

Minden 1987

Minden SL, Orav J, Reich P. Depression in multiple sclerosis. General Hospital Psychiatry 1987;9:426‐34. [DOI] [PubMed] [Google Scholar]

Minden 1991

Minden SL, Schiffer RB. Depression and mood disorders in multiple sclerosis. Neuropsychiatry 1991;4:62‐77. [Google Scholar]

Minden 1992

Minden SL. Psychotherapy for people with multiple sclerosis. Neuropsychiatry 1992;4:198‐213. [DOI] [PubMed] [Google Scholar]

Minden 2000

Minden SL. Mood disorders in multiple sclerosis: diagnosis and treatment. Journal of Neurovirology 2000;6 (Suppl 2):S160‐7. [PubMed] [Google Scholar]

Moher 1998

Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta‐analyses?. Lancet 1998;352:609‐13. [DOI] [PubMed] [Google Scholar]

Mohr 1996

Mohr DC, Goodkin DE, Likoksy W, Gatto N, Neilley LK, Griffen C, et al. Therapeutic expectations of patients with mutliple sclerosis upon initiating interferon beta‐1b: relationship to adherence to treatment. Multiple Sclerosis 1996;2:222‐6. [DOI] [PubMed] [Google Scholar]

Mohr 1997a

Mohr DC, Goodkin DE, Gatto N, Wende J. Depression, coping, and level of neurological impairment in multiple sclerosis. Multiple Sclerosis 1997;3:254‐8. [DOI] [PubMed] [Google Scholar]

Mohr 1997b

Mohr DC, Goodkin DE, Likosky W, Gatto N, Baumann MA, Rudick RA. Treatment of depression improves adherence to interferon beta‐1b therapy for multiple sclerosis. Archives of Neurology 1997;54:531‐3. [DOI] [PubMed] [Google Scholar]

Mohr 1998

Mohr DC, Likoksy W, Boudewyn AC, Marietta P, Dwyer P, Wende J, et al. Side effect profile and compliance in the treatment of multiple sclerosis with interferon beta‐1a. Multiple Sclerosis 1998;4:487‐9. [DOI] [PubMed] [Google Scholar]

Mohr 1999a

Mohr DC, Likoksy W, Dwyer P, Wende J, Boudewyn AC, Goodkin DE. Course of depression during the initiation of interferon beta‐1a for multiple sclerosis. Archives of Neurology 1999;56:1263‐5. [DOI] [PubMed] [Google Scholar]

Mohr 1999b

Mohr DC, Genain CP, Boudewyn AC. Relationship between treatment for depression and interferon‐gamma in patients with multiple sclerosis. Psychosomatic Medicine 1999;61:112. [DOI] [PubMed] [Google Scholar]

Mohr 1999c

Mohr DC, Goodkin DE. Treatment of depression in multiple sclerosis: Review and meta‐analysis. Clinical Psychology: Science and Practice 1999;6:1‐9. [Google Scholar]

Mohr 2001a

Mohr DC, Cox D. Multiple sclerosis: Empirical literature for the clinical health psychologist. Journal of Clinical Psychology 2001;57:479‐99. [DOI] [PubMed] [Google Scholar]

Mohr 2001b

Mohr DC, Goodkin DE, Islar J, Hauser SL, Genain CP. Treatment of depression is associated with suppression of non‐specific and antigen‐specific TH1 responses in multiple sclerosis. Archives of Neurology 2001;58:1081‐6. [DOI] [PubMed] [Google Scholar]

Mohr 2001c

Mohr DC, Boudewyn AC, Goodkin DE, Bostrom A, Epstein L. Comparative outcomes for individual cognitive‐behavior therapy, supportive‐expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. Journal of Consulting and Clinical Psychology 2001;69(6):942‐9. [PubMed] [Google Scholar]

Mohr 2002

Mohr DC, Goodkin DE, Nelson S, Cox D, Weiner M. Moderating effects of coping on the relationship between stress and the development of new brain lesions in multiple sclerosis. Psychosomatic Medicine 2002;64:803‐9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Mullins 2001

Mullins LL, Cote MP, Fuemmeler BF, Jean VM, Beatty WW, Paul RH. Illness intrusiveness, uncertainty and distress in individuals with multiple sclerosis. Rehabilitation Psychology 2001;46(2):139‐53. [Google Scholar]

Nelson 1988

Nelson LM, Franklin GM, Hamman RF, Boteler DL, Baum HM, Burks JS. Referral bias in multiple sclerosis research. Journal of Clinical Epidemiology 1988;41:187‐92. [DOI] [PubMed] [Google Scholar]

O'Brien 1993

O'Brien MT. Multiple sclerosis: the relationship among self‐esteem, social support, and coping behaviour. Applied Nursing Research 1993;6:54‐63. [DOI] [PubMed] [Google Scholar]

Ostelo 2004

Ostelo RWJG, Tulder MW van, Vlaeyen JWS, Linton SJ, Morley SJ, Assendelft WJJ. Behavioural treatment for chronic low back pain. Cochrane Database of Systematic Reviews 2004, Issue 3. [Art.No:CD002014. DOI:10.1002/14651858.CD002014.pub2] [Google Scholar]

Pakenham 1997

Pakenham KI, Stewart CA, Rogers A. The role of coping in adjustment to multiple sclerosis‐related adaptive demands. Psychology, Health and Medicine 1997;2:197‐211. [Google Scholar]

Pakenham 1999

Pakenham KI. Adjustment to multiple sclerosis: Application of a stress and coping model. Health Psychology 1999;18:383‐92. [DOI] [PubMed] [Google Scholar]

Penman 1991

Penman MF. Assessing the prevalence of cognitive impairment in multiple sclerosis: Implications for patient management. Axone 1991;13:45‐9. [PubMed] [Google Scholar]

Pepper 1993

Pepper CM, Krupp IB, Friedberg F, Doscher C, Coyle PK. A comparison of neuropsychiatric characteristics in chronic fatigue syndrome, multiple sclerosis and major depression. Journal of Neuropsychiatry and Clinical Neurosciences 1993;5:200‐5. [DOI] [PubMed] [Google Scholar]

Pfennings 1999

Pfennings LE, Ploeg HM, Cohen L, Bramsen I, Polman CH, Lankhorst GJ, et al. A health related quality of life questionnaire for multiple sclerosis patients. Acta Neurologica Scandinavica 1999;100:148‐55. [DOI] [PubMed] [Google Scholar]

Pollock 1990

Pollock SE, Christian BJ, Sands D. Responses to chronic illness: Analysis of psychological and physiological adaptation. Nursing Research 1990;39:300‐4. [PubMed] [Google Scholar]

Poser 1983

Poser CM, Paty DW, Scheinburg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research proposals. Annals of Neurology 1983;13:227‐31. [DOI] [PubMed] [Google Scholar]

Price 1998

Price JR, Couper J. Cognitive behavioural therapy for chronic fatigue syndrome in adults. Cochrane Database of Systematic Reviews 1998, Issue 4. [Art. No: CD001027. DOI: 10.1002/14651858.CD001027] [Google Scholar]

Prosiegal 1993

Prosiegal M, Michael C. Neuropsychology and multiple sclerosis: diagnostic and rehabilitative approaches. Journal of the Neurological Sciences 1993;115(Suppl):S51‐4. [DOI] [PubMed] [Google Scholar]

Pujol 1997

Pujol J, Bello J, Deus J, Marti‐Vilalta JL, Capdevila A. Lesions in the left arcuate fasciculus region and depressive symptoms in multiple sclerosis. Neurology 1997;49:1105‐10. [DOI] [PubMed] [Google Scholar]

Rabins 1986

Rabins PV, Brooks BR, O'Donnell P, Pearlson GD, Moberg P. Structural correlates of emotional disorder in multiple sclerosis. Brain 1986;109:585‐97. [DOI] [PubMed] [Google Scholar]

Rao 1991

Rao SM, Leo GJ, Bernadin L, Unversagt F. Cognitive dysfunction in multiple sclerosis. I. Frequency, patterns, and prediction. Neurology 1991;41:685‐91. [DOI] [PubMed] [Google Scholar]

Rao 1992

Rao SM, Huber SJ, Bornstein RA. Emotional changes with multiple sclerosis and Parkinson's disease. Journal of Consulting and Clincial Psychology 1992;60:369‐78. [DOI] [PubMed] [Google Scholar]

Rosenberg 1965

Rosenberg M. Self‐esteem scale. In: Rosenberg M editor(s). Society and the adolescent self‐image. Princeton, New Jersey: Princeton University Press, 1965. [Google Scholar]

Sadovnick 1991

Sadovnick AD, Eisen K, Ebers GC, Paty DW. Cause of death in patients attending multiple sclerosis clinics. Neurology 1991;41:1193‐6. [DOI] [PubMed] [Google Scholar]

Scheinberg 1984

Scheinberg LC, Pfennings L, Pouwer F, Cohen L, Ketelaer P, Polman C, et al. Psychological functioning in primary progressive versus secondary progressive multiple sclerosis. British Journal of Medical Psychology 1998;71:99‐106. [DOI] [PubMed] [Google Scholar]

Schiaffino 1996

Schiaffino KM, Shawaryn MS, Blum MA. Assessing the psychosocial impact of multiple sclerosis: learning from research on rheumatoid arthritis. Journal of Neurologic Rehabilitation 1996;10:81‐9. [Google Scholar]

Schiffer 1983

Schiffer RB, Caine ED, Bamford KA, Levy S. Depressive episodes in patients with multiple sclerosis. American Journal of Psychiatry 1983;140:1498‐500. [DOI] [PubMed] [Google Scholar]

Schubert 1993

Schuber DSP, Foliart RH. Increased depression in multiple sclerosis: a meta‐analysis. Psychosomatics 1993;34:124‐30. [DOI] [PubMed] [Google Scholar]

Schumacher 1965

Schumacher GA, Beebe G, Kibler RF, Kurland LT, Kurtzke JF, McDowell F, et al. Problems of experimental trials of therapy in multiple sclerosis: Report by the panel of the evaluation of experimental trials in multiple sclerosis. Annals of the New York Academy of Sciences 1965;122:552‐68. [DOI] [PubMed] [Google Scholar]

Schwarzer 1995

Schwarzer R, Jerusalem M. Generalised self‐efficacy scale. In: Weinman J, Wright S, Johnston M editor(s). Measures in health psychology: A user's portfolio. Windsor, UK: NFER‐Nelson, 1995. [Google Scholar]

Sharrack 1999

Sharrack B, Hughs RA. The Guy's Neurological Disability Scale (GNDS): a new disability measure for multiple sclerosis. Multiple Sclerosis 1999;5:223‐33. [DOI] [PubMed] [Google Scholar]

Spitzer 1995

Spitzer RL, Kroenke K, Linzer M, Hahn SR, Williams JBW, Verloin deGruy III, et al. Health‐related quality of life in primary care patients with mental disorders. Results from the PRIME‐MD 1000 study. Journal of the American Medical Association 1995;274:1511‐7. [PubMed] [Google Scholar]

Stenager 1994

Stenager E, Knudsen L, Jensen K. Multiple sclerosis: Correlation of anxiety, physical impairment and cognitive dysfunction. Italian Journal of Neurological Science 1994;15:99‐103. [DOI] [PubMed] [Google Scholar]

Surridge 1969

Surridge D. An investigation into some psychiatric aspects of multiple sclerosis. British Journal of Psychiatry 1969;115:749‐64. [DOI] [PubMed] [Google Scholar]

Thompson 1996

Thompson AJ. Multiple sclerosis: symptomatic treatment. Journal of Neurology 1996;243:559‐65. [DOI] [PubMed] [Google Scholar]

Vickrey 1995

Vickrey BG, Hays RD, Harooni R, Myers LW, Ellison GW. A health‐related quality of life measure for multiple sclerosis. Qual Life Res 1995;4:187‐206. [DOI] [PubMed] [Google Scholar]

Walsh 1989

Walsh A, Walsh PA. Love, self‐esteem and multiple sclerosis. Social Science and Medicine 1989;29:793‐8. [DOI] [PubMed] [Google Scholar]

Ware 1993

Ware JE, Snow KK, Kosinski M, Gandek B. SF‐36 Health Survey manual and interpretation guide. Boston (MA): Nimrod Press, 1993. [Google Scholar]

Whitlock 1980

Whitlock FA, Siskind MM. Depression as a major symptom of multiple sclerosis. Journal of Neurology, Neurosurgery, and Psychiatry 1980;43:861‐5. [DOI] [PMC free article] [PubMed] [Google Scholar]

Wilson 1999

Wilson BA. Case studies in neuropsychological rehabilitation. New York: Oxford University Press, 1999. [Google Scholar]

[CD004431-bib-0001] Benedict RHB, Shapiro A, Priore R, Miller C, Munschauer F, Jacobs L. Neuropsychological counseling improves social behavior in cognitively‐impaired multiple sclerosis patients. Multiple Sclerosis 2000;6:391‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0002] Crawford JD, McIvor GP. Group psychotherapy: Benefits in multiple sclerosis. Archives of Physical Medicine and Rehabilitation 1985;66:810‐3. [PubMed] [Google Scholar]

[CD004431-bib-0003] Foley FW, Bedell JR, LaRocca NG, Scheinberg LC, Reznikoff M. Efficacy of stress‐inoculation training in coping with multiple sclerosis. Journal of Consulting and Clinical Psychology 1987;55:919‐22. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0004] Gordon PA, Lam CS, Winter R. Interaction strain and persons with multiple sclerosis: Effectiveness of a social skills program. Journal of Applied Rehabilitation Counselling 1997;28:5‐11. [Google Scholar]

[CD004431-bib-0005] Jonsson A, Korfitzen EM, Heltberg A, Ravnborg MH, Byskov‐Ottosen E. Effects of neuropsychological treatment in patients with multiple sclerosis. Acta Neurologica Scandinavica 1993;88:394‐400. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0006] Larcombe NA, Wilson PH. An evaluation of cognitive‐behaviour therapy for depression in patients with multiple sclerosis. British Journal of Psychiatry 1984;145:366‐71. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0007] Lincoln NB, Dent A, Harding J, Weyman N, Nicholl C, Blumhardt LD, et al. Evaluation of cognitive assessment and cognitive intervention for people with multiple sclerosis. Journal of Neurology Neurosurgery & Psychiatry 2002;72:93‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD004431-bib-0008] Maguire, B. The effects of imagery on attitudes and moods in multiple sclerosis patients. Alternative Therapies 1996;2:75‐9. [PubMed] [Google Scholar]

[CD004431-bib-0009] Mendoza RJ, Pittenger DJ, Weinstein CS. Unit management of depression of patients with multiple sclerosis using cognitive remediation strategies: A preliminary study. Neurorehabilitation and Neural Repair 2001;15:9‐14. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0010] Mohr DC, Likosky W, Bertagnolli A, Goodkin DE, Wende J, Dwyer P, et al. Telephone‐administered cognitive behavioral therapy for the treatment of depressive symptoms in multiple sclerosis. Journal of consulting and Clinical Psychology 2000;68(2):356‐61. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0011] Mohr DC, Boudewyn AC, Goodkin DE, Bostrom A, Epstein L. Comparative outcomes for individual cognitive‐behavior therapy, supportive‐expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. Journal of Consulting and Clinical Psychology 2001;69(6):942‐9. [PubMed] [Google Scholar]

[CD004431-bib-0012] Mohr DC, Hart SL, Goldberg A. Effects of treatment for depression on fatigue in multiple sclerosis. Psychosomatic Medicine 2003;65:542‐547. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0013] Rigby, S. Coping with MS. Liverpool: University of Liverpool, 2003. [Google Scholar]

[CD004431-bib-0014] Schwartz CE. Teaching coping skills enhances quality of life more than peer support: Results of a randomised trial with multiple sclerosis patients. Health Psychology 1999;18:211‐20. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0015] Solari A, Motta A, Mendozzi L, Pucci E, Forni M, Mancardi G, Pozzilli C. Computer‐aided retraining of memory and attention in people with multiple sclerosis: a randomized, double‐blind controlled trial. Journal of the Neurological Sciences 2004;222:99‐104. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0016] Stuifbergen AK, Becker H, Blozis S, Timmerman G, Kullberg V. A randomized clinical trial of a wellness intervention for women with multiple sclerosis. Archives of Physical Medicine and Rehabilitation 2003;84:467‐76. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0017] Wassem R, Dudley W. Symptom management and adjustment of patients with multiple sclerosis. Clinical Nursing Research 2003;12:102‐17. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0018] Klarskov P, Heely E, Nyholdt I, Rottensten K, Nordenbo A. Biofeedback treatment of bladder dysfunction in multiple sclerosis. A randomised trial.. Scandanavian Journal of Urology and Nephrology 1994;Suppl. 157:61‐65. [PubMed] [Google Scholar]

[CD004431-bib-0019] Mendozzi L, Pugnetti L, Motta A, Barbieri E, Gambini A, Cazzullo CL. Computer‐assisted memory retraining of patients with multiple sclerosis. Italian Journal of Neurological Science 1998;19:S431‐8. [Google Scholar]

[CD004431-bib-0020] Mills N, Allen J. Mindfulness of movement as a coping strategy in multiple sclerosis. A pilot study. General Hospital Psychiatry 2000;22:425‐31. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0021] Mohr DC, Classen C, Barrera M. The relationship between social support, depression and treatment for depression in people with multiple sclerosis. Psychological Medicine 2004;34:533‐41. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0022] Rogers D, Khoo K, MacEachen M, Oven M, Beatty WW. Cognitive therapy for multiple sclerosis: A preliminary Study. Alternative Therapies 1996;2(5):70‐4. [PubMed] [Google Scholar]

[CD004431-bib-0023] Schmid W, Aldridge D. Active music therapy in the treatment of multiple sclerosis patients: A matched control study. Journal of Music Therapy 2004;41:225‐40. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0024] Tesar N, Baumhackl U, Kopp M, Gunther V. Effects of psychological group therapy in patients with multiple sclerosis. Acta Neurologica Scandinavica 2003;107:394‐9. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0025] Chiaravalloti ND, DeLuca J, Moore NB, Ricker JH. Treating learning impairments improves memory performance in multiple sclerosis: a randomized clinical trial. Multiple Sclerosis 2005;11:58‐68. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0026] Mohr DC, Cox D, Merluzzi N. Self‐injection anxiety training: a treatment for patients unable to self‐inject injectable medications. Multiple Sclerosis 2005;11:182‐5. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0027] Telephone administered psychotherapy for the treatment of depression in multiple sclerosis. Personal Communication.

[CD004431-bib-0028] A randomised controlled trial of cognitive behavioural therapy for fatigue in patients with multiple sclerosis. Personal Communication.

[CD004431-bib-0029] Aikens JE, Fischer JS, Namey M, Rudick RA. A replicated prospective investigation of life stress coping and depressive symptoms in multiple sclerosis. Journal of Behavioral Medicine 1997;20:433‐45. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0030] Amato MP, Ponziani G, Pracucci G, Siracusa G, Amaducci L. Cognitive impairment in early‐onset multiple sclerosis. Archives of Neurology 1995;52:168‐72. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0031] Barnes RH, Busse EW, Dinken H. The alleviation of emotional problems in multiple sclerosis by group therapy. Group Psychotherapy 1954;6:193‐201. [Google Scholar]

[CD004431-bib-0032] Barnwell AM, Kavanagh DJ. Prediction of psychological adjustment to multiple sclerosis. Social Science and Medicine 1997;45:411‐8. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0033] Bedford A, Foulds G. Delusions‐Symptoms‐States‐Inventory. Windsor: NFER‐Nelson, 1978. [Google Scholar]

[CD004431-bib-0034] Bolding H. Psychotherapeutic aspects in the management of patients with multiple sclerosis. Diseases of the Nervous System 1960;21:24‐6. [PubMed] [Google Scholar]

[CD004431-bib-0035] Brassington JC, Marsh NV. Neuropsychological aspects of multiple sclerosis. Neuropsychological Review 1998;8:43‐77. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0036] Clarke M, Oxman AD. Cochrane Reviewer's Handbook Version 4.1.1 (updated December 2000); Section 6. The Cochrane Library. Oxford: Update Software, 2001, issue 1.

[CD004431-bib-0037] Crawford JD, McIvor GP. Stress management for multiple sclerosis patients. Psychological Reports 1987;61:423‐429. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0038] Dalos NP, Rabins PV, Brooks BR, O'Donnell P. Disease activity and emotional state in multiple sclerosis. Annals of Neurology 1983;13:573‐83. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0039] Day M, Day E, Herman R. Group therapy of patients with multiple sclerosis. Archives of Neurology and Psychiatry 1953;69:193‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0040] EuroQoL Group. EuroQoL: a new facility for the measurement of health related quality of life. Health Policy 1990;16:199‐208. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0041] Fassbender K, Schmidt R, Mößner R, Kischka U, Kühnen J, Schwartz A, et al. Mood disorders and dysfunction of the hypothalamic‐pituitary‐adrenal axis in multiple sclerosis. Archives of Neurology 1998;55:66‐72. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0042] Feinstein A, O'Connor P, Gray T, Feinstein K. The effects of anxiety on psychiatric morbidity in patients with multiple sclerosis. Multiple Sclerosis 1999;5:323‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0043] Franklin GE, Heaton RK, Nelson LM, Filley CM, Seibert C. Correlation of neuropsychological and MRI findings in chronic/progressive multiple sclerosis. Neurology 1988;38:1826‐9. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0044] Gilchrist AC, Creed FH. Depression, cognitive impairment and social stress in multiple sclerosis. Journal of Psychosomatic Medicine 1994;38:193‐201. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0045] Graham J. What is MS?. Multiple Sclerosis Society, UK. 2002:3.

[CD004431-bib-0046] Gulick EE. Correlates among quality of life among persons with multiple sclerosis. Nursing Research 1997;46:305‐11. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0047] Hartings MF, Pavlou MM, Davis FA. Group counseling of MS patients in a program of comprehensive care. Journal of Chronic Diseases 1976;29:65‐73. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0048] Heaton RK, Nelson LM, Thompson DS, Burks JS, Franklin GM. Neuropsychological findings in relapse‐remitting multiple sclerosis. Journal of Consulting and Clinical Psychology 1985;53:103‐10. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0049] Hobart J, Lamping D, Fitzpatrick R, Riazi A, Thompson A. The Multiple Sclerosis Impact Scale (MSIS‐29). A new patient‐based outcome measure. Brain 2001;124:962‐73. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0050] Joffe RT, Lippert GP, Gray T, Sawa G, Horwath Z. Mood disorder and multiple sclerosis. Archives of Neurology 1987;44:376‐8. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0051] Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444‐52. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0052] Langdon DW, Thompson AJ. Cognitive problems in multiple sclerosis. MS Management 1996;3:5‐9. [Google Scholar]

[CD004431-bib-0053] LaRocca NG, Kalb RC, Foley FW, Caruso LS. Assessment of psychosocial outcomes. Journal of Neurological Rehabilitation 1993;7:109‐16. [Google Scholar]

[CD004431-bib-0054] LaRocca NG, Ritvo PG, Miller DM, Fischer JS, Andrews H, Paty DW. "Quality of life" assessment in multiple sclerosis clinical trials: current status and strategies for improving multiple sclerosis clinical trial design. In: Goodkin DE, Rudick RA editor(s). Multiple sclerosis: advances in clinical trial design, treatment and future perspectives. London: Springer‐Verlag, 1996:145‐60. [Google Scholar]

[CD004431-bib-0055] Maurelli M, Marchioni E, Cerretano R, Bosone D, Bergamaschi R, Citterio A, et al. Neuropsychological assessment in MS: Clinical, neurophysiological and neuroradiological relationships. Acta Neurological Scandinavica 1992;86:124‐8. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0056] McIvor GP, Riklan M, Reznikoff M. Depression in multiple sclerosis as a function of length and severity of illness, age, remissions and perceived social support. Journal of Clinical Psychology 1984;40:1028‐33. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0057] McNair DM, Lorr M, Droppleman LF. Profile of Mood States Manual. San Diego: Educational and Industrial Testing Service, 1981. [Google Scholar]

[CD004431-bib-0058] Miles A. Some psycho‐social consequences of multiple sclerosis: problems of social interaction and group identity. British Journal of Medical Psychology 1979;52:321‐31. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0059] Minden SL, Orav J, Reich P. Depression in multiple sclerosis. General Hospital Psychiatry 1987;9:426‐34. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0060] Minden SL, Schiffer RB. Depression and mood disorders in multiple sclerosis. Neuropsychiatry 1991;4:62‐77. [Google Scholar]

[CD004431-bib-0061] Minden SL. Psychotherapy for people with multiple sclerosis. Neuropsychiatry 1992;4:198‐213. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0062] Minden SL. Mood disorders in multiple sclerosis: diagnosis and treatment. Journal of Neurovirology 2000;6 (Suppl 2):S160‐7. [PubMed] [Google Scholar]

[CD004431-bib-0063] Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta‐analyses?. Lancet 1998;352:609‐13. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0064] Mohr DC, Goodkin DE, Likoksy W, Gatto N, Neilley LK, Griffen C, et al. Therapeutic expectations of patients with mutliple sclerosis upon initiating interferon beta‐1b: relationship to adherence to treatment. Multiple Sclerosis 1996;2:222‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0065] Mohr DC, Goodkin DE, Gatto N, Wende J. Depression, coping, and level of neurological impairment in multiple sclerosis. Multiple Sclerosis 1997;3:254‐8. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0066] Mohr DC, Goodkin DE, Likosky W, Gatto N, Baumann MA, Rudick RA. Treatment of depression improves adherence to interferon beta‐1b therapy for multiple sclerosis. Archives of Neurology 1997;54:531‐3. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0067] Mohr DC, Likoksy W, Boudewyn AC, Marietta P, Dwyer P, Wende J, et al. Side effect profile and compliance in the treatment of multiple sclerosis with interferon beta‐1a. Multiple Sclerosis 1998;4:487‐9. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0068] Mohr DC, Likoksy W, Dwyer P, Wende J, Boudewyn AC, Goodkin DE. Course of depression during the initiation of interferon beta‐1a for multiple sclerosis. Archives of Neurology 1999;56:1263‐5. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0069] Mohr DC, Genain CP, Boudewyn AC. Relationship between treatment for depression and interferon‐gamma in patients with multiple sclerosis. Psychosomatic Medicine 1999;61:112. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0070] Mohr DC, Goodkin DE. Treatment of depression in multiple sclerosis: Review and meta‐analysis. Clinical Psychology: Science and Practice 1999;6:1‐9. [Google Scholar]

[CD004431-bib-0071] Mohr DC, Cox D. Multiple sclerosis: Empirical literature for the clinical health psychologist. Journal of Clinical Psychology 2001;57:479‐99. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0072] Mohr DC, Goodkin DE, Islar J, Hauser SL, Genain CP. Treatment of depression is associated with suppression of non‐specific and antigen‐specific TH1 responses in multiple sclerosis. Archives of Neurology 2001;58:1081‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0073] Mohr DC, Boudewyn AC, Goodkin DE, Bostrom A, Epstein L. Comparative outcomes for individual cognitive‐behavior therapy, supportive‐expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. Journal of Consulting and Clinical Psychology 2001;69(6):942‐9. [PubMed] [Google Scholar]

[CD004431-bib-0074] Mohr DC, Goodkin DE, Nelson S, Cox D, Weiner M. Moderating effects of coping on the relationship between stress and the development of new brain lesions in multiple sclerosis. Psychosomatic Medicine 2002;64:803‐9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD004431-bib-0075] Mullins LL, Cote MP, Fuemmeler BF, Jean VM, Beatty WW, Paul RH. Illness intrusiveness, uncertainty and distress in individuals with multiple sclerosis. Rehabilitation Psychology 2001;46(2):139‐53. [Google Scholar]

[CD004431-bib-0076] Nelson LM, Franklin GM, Hamman RF, Boteler DL, Baum HM, Burks JS. Referral bias in multiple sclerosis research. Journal of Clinical Epidemiology 1988;41:187‐92. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0077] O'Brien MT. Multiple sclerosis: the relationship among self‐esteem, social support, and coping behaviour. Applied Nursing Research 1993;6:54‐63. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0078] Ostelo RWJG, Tulder MW van, Vlaeyen JWS, Linton SJ, Morley SJ, Assendelft WJJ. Behavioural treatment for chronic low back pain. Cochrane Database of Systematic Reviews 2004, Issue 3. [Art.No:CD002014. DOI:10.1002/14651858.CD002014.pub2] [Google Scholar]

[CD004431-bib-0079] Pakenham KI, Stewart CA, Rogers A. The role of coping in adjustment to multiple sclerosis‐related adaptive demands. Psychology, Health and Medicine 1997;2:197‐211. [Google Scholar]

[CD004431-bib-0080] Pakenham KI. Adjustment to multiple sclerosis: Application of a stress and coping model. Health Psychology 1999;18:383‐92. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0081] Penman MF. Assessing the prevalence of cognitive impairment in multiple sclerosis: Implications for patient management. Axone 1991;13:45‐9. [PubMed] [Google Scholar]

[CD004431-bib-0082] Pepper CM, Krupp IB, Friedberg F, Doscher C, Coyle PK. A comparison of neuropsychiatric characteristics in chronic fatigue syndrome, multiple sclerosis and major depression. Journal of Neuropsychiatry and Clinical Neurosciences 1993;5:200‐5. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0083] Pfennings LE, Ploeg HM, Cohen L, Bramsen I, Polman CH, Lankhorst GJ, et al. A health related quality of life questionnaire for multiple sclerosis patients. Acta Neurologica Scandinavica 1999;100:148‐55. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0084] Pollock SE, Christian BJ, Sands D. Responses to chronic illness: Analysis of psychological and physiological adaptation. Nursing Research 1990;39:300‐4. [PubMed] [Google Scholar]

[CD004431-bib-0085] Poser CM, Paty DW, Scheinburg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research proposals. Annals of Neurology 1983;13:227‐31. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0086] Price JR, Couper J. Cognitive behavioural therapy for chronic fatigue syndrome in adults. Cochrane Database of Systematic Reviews 1998, Issue 4. [Art. No: CD001027. DOI: 10.1002/14651858.CD001027] [Google Scholar]

[CD004431-bib-0087] Prosiegal M, Michael C. Neuropsychology and multiple sclerosis: diagnostic and rehabilitative approaches. Journal of the Neurological Sciences 1993;115(Suppl):S51‐4. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0088] Pujol J, Bello J, Deus J, Marti‐Vilalta JL, Capdevila A. Lesions in the left arcuate fasciculus region and depressive symptoms in multiple sclerosis. Neurology 1997;49:1105‐10. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0089] Rabins PV, Brooks BR, O'Donnell P, Pearlson GD, Moberg P. Structural correlates of emotional disorder in multiple sclerosis. Brain 1986;109:585‐97. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0090] Rao SM, Leo GJ, Bernadin L, Unversagt F. Cognitive dysfunction in multiple sclerosis. I. Frequency, patterns, and prediction. Neurology 1991;41:685‐91. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0091] Rao SM, Huber SJ, Bornstein RA. Emotional changes with multiple sclerosis and Parkinson's disease. Journal of Consulting and Clincial Psychology 1992;60:369‐78. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0092] Rosenberg M. Self‐esteem scale. In: Rosenberg M editor(s). Society and the adolescent self‐image. Princeton, New Jersey: Princeton University Press, 1965. [Google Scholar]

[CD004431-bib-0093] Sadovnick AD, Eisen K, Ebers GC, Paty DW. Cause of death in patients attending multiple sclerosis clinics. Neurology 1991;41:1193‐6. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0094] Scheinberg LC, Pfennings L, Pouwer F, Cohen L, Ketelaer P, Polman C, et al. Psychological functioning in primary progressive versus secondary progressive multiple sclerosis. British Journal of Medical Psychology 1998;71:99‐106. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0095] Schiaffino KM, Shawaryn MS, Blum MA. Assessing the psychosocial impact of multiple sclerosis: learning from research on rheumatoid arthritis. Journal of Neurologic Rehabilitation 1996;10:81‐9. [Google Scholar]

[CD004431-bib-0096] Schiffer RB, Caine ED, Bamford KA, Levy S. Depressive episodes in patients with multiple sclerosis. American Journal of Psychiatry 1983;140:1498‐500. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0097] Schuber DSP, Foliart RH. Increased depression in multiple sclerosis: a meta‐analysis. Psychosomatics 1993;34:124‐30. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0098] Schumacher GA, Beebe G, Kibler RF, Kurland LT, Kurtzke JF, McDowell F, et al. Problems of experimental trials of therapy in multiple sclerosis: Report by the panel of the evaluation of experimental trials in multiple sclerosis. Annals of the New York Academy of Sciences 1965;122:552‐68. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0099] Schwarzer R, Jerusalem M. Generalised self‐efficacy scale. In: Weinman J, Wright S, Johnston M editor(s). Measures in health psychology: A user's portfolio. Windsor, UK: NFER‐Nelson, 1995. [Google Scholar]

[CD004431-bib-0100] Sharrack B, Hughs RA. The Guy's Neurological Disability Scale (GNDS): a new disability measure for multiple sclerosis. Multiple Sclerosis 1999;5:223‐33. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0101] Spitzer RL, Kroenke K, Linzer M, Hahn SR, Williams JBW, Verloin deGruy III, et al. Health‐related quality of life in primary care patients with mental disorders. Results from the PRIME‐MD 1000 study. Journal of the American Medical Association 1995;274:1511‐7. [PubMed] [Google Scholar]

[CD004431-bib-0102] Stenager E, Knudsen L, Jensen K. Multiple sclerosis: Correlation of anxiety, physical impairment and cognitive dysfunction. Italian Journal of Neurological Science 1994;15:99‐103. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0103] Surridge D. An investigation into some psychiatric aspects of multiple sclerosis. British Journal of Psychiatry 1969;115:749‐64. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0104] Thompson AJ. Multiple sclerosis: symptomatic treatment. Journal of Neurology 1996;243:559‐65. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0105] Vickrey BG, Hays RD, Harooni R, Myers LW, Ellison GW. A health‐related quality of life measure for multiple sclerosis. Qual Life Res 1995;4:187‐206. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0106] Walsh A, Walsh PA. Love, self‐esteem and multiple sclerosis. Social Science and Medicine 1989;29:793‐8. [DOI] [PubMed] [Google Scholar]

[CD004431-bib-0107] Ware JE, Snow KK, Kosinski M, Gandek B. SF‐36 Health Survey manual and interpretation guide. Boston (MA): Nimrod Press, 1993. [Google Scholar]

[CD004431-bib-0108] Whitlock FA, Siskind MM. Depression as a major symptom of multiple sclerosis. Journal of Neurology, Neurosurgery, and Psychiatry 1980;43:861‐5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD004431-bib-0109] Wilson BA. Case studies in neuropsychological rehabilitation. New York: Oxford University Press, 1999. [Google Scholar]

PERMALINK

Psychological interventions for multiple sclerosis

Peter W Thomas

Sarah Thomas

Charles Hillier

Kate Galvin

Roger Baker

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Background

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Results

Description of studies

Risk of bias in included studies

Effects of interventions

Discussion

Authors' conclusions

Implications for practice.

Implications for research.

What's new

Acknowledgements

Appendices

Appendix 1. CENTRAL search strategy

Appendix 2. MEDLINE (dialog datastar) search strategy

Appendix 3. EMABSE (dialog datastar) search strategy

Data and analyses

Comparison 1. Cognitive behavioural therapy versus care as usual for people with MS with depression.

1.1. Analysis.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Benedict 2000.

Crawford 1985.

Foley 1987.

Gordon 1997.

Jonsson 1993.

Larcombe 1984.

Lincoln 2002.

Maguire 1996.

Mendoza 2001.

Mohr 2000.

Mohr 2001.

Mohr 2003.

Rigby 2003.

Schwartz 1999.

Solari 2004.

Stuifbergen 2003.

Wassem 2003.

Characteristics of excluded studies [ordered by study ID]

Characteristics of ongoing studies [ordered by study ID]

Mohr.

Moss‐Morris.

Differences between protocol and review

Contributions of authors

Sources of support

Internal sources

External sources

Declarations of interest

References

References to studies included in this review

Benedict 2000 {published data only}

Crawford 1985 {published data only}

Foley 1987 {published data only}

Gordon 1997 {published data only}