Multiple Sclerosis Journal

Multiplehttp://msj.sagepub.com/ Sclerosis Journal

Evaluation of an adjustment group forpeople with multiple sclerosis and lowmood: a randomized controlled trial Nadina B Lincoln, Faye Yuill, Jessica Holmes, Avril ER Drummond, Cris S Constantinescu, Sarah Armstrong and Ceri Phillips Mult Scler 2011 17: 1250 originally published online 25 May 2011 DOI: 10.1177/1352458511408753 The online version of this article can be found at: http://msj.sagepub.com/content/17/10/1250

Published by: http://www.sagepublications.com

Additional services and information for Multiple Sclerosis Journal can be found at: Email Alerts: http://msj.sagepub.com/cgi/alerts Subscriptions: http://msj.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Oct 6, 2011 OnlineFirst Version of Record - May 25, 2011 What is This?

Downloaded from msj.sagepub.com at University of Nottingham on October 21, 2012

Research Paper

Evaluation of an adjustment group for people with multiple sclerosis and low mood: a randomized controlled trial

Multiple Sclerosis Journal 17(10) 1250–1257 Ó The Author(s) 2011 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1352458511408753 msj.sagepub.com

Nadina B Lincoln1, Faye Yuill1, Jessica Holmes1, Avril ER Drummond2, Cris S Constantinescu3, Sarah Armstrong4 and Ceri Phillips5

Abstract Background: Mood problems affect many people with multiple sclerosis (MS). The aim was to evaluate the effectiveness of a group treatment based on cognitive behavioural principles. Methods: People with MS were screened on the General Health Questionnaire 12 (GHQ-12) and Hospital Anxiety and Depression Scale (HAD). Those identified with low mood were invited to take part in a randomized trial comparing the effect of attending an adjustment group with a waiting list control. Patients allocated to the adjustment group received six 2 h group treatment sessions. Outcomes were assessed 4 and 8 months after randomization, blind to group allocation. Results: Of the 311 patients identified, 221 (71%) met the criteria for low mood and 151 (68%) agreed to take part. Hierarchical regression analyses were conducted to compare the two groups, correcting for baseline mood and disability. At 4 months, group allocation alone was a significant predictor of the primary outcome measure, the GHQ-12. At 8 months, group allocation alone was no longer a significant predictor for GHQ-12 scores, but it was when baseline GHQ12 and Guy’s Neurological Disability Scale scores were controlled for. Comparison of the area under the curve revealed significant differences between the groups for GHQ-12 (p ¼ 0.003), HAD Anxiety (p ¼ 0.013), HAD Depression (p ¼ 0.004), Beck Depression Inventory (p ¼ 0.001), MS Self-efficacy (p ¼ 0.037) and MS Impact Scale Psychological (p ¼ 0.012). Conclusion: Patients receiving treatment were less distressed and had less depression and anxiety. There was some evidence of improved self-efficacy and a reduction of the impact of MS on people’s lives. Keywords Anxiety, depression, group, multiple sclerosis, psychological therapy Date received: 24th December 2010; revised: 29th March 2011; accepted: 6th April 2011

Introduction Depression is a common problem in people with multiple sclerosis (MS) and rates are higher than in both healthy controls and patients with other neurological conditions.1 Estimates of the prevalence of depression in people with MS vary according to the sample studied and assessment methods used, but are generally about 50%.2–5 Anxiety is also a common feature,2 and the prevalence is higher than in healthy controls.5 It has been acknowledged that cognitive and behavioural factors have an important role in psychological adjustment in people with early MS,6 suggesting that cognitive

1 Institute of Work, Health and Organisations, University of Nottingham, UK. 2 Division of Rehabilitation and Ageing, University of Nottingham, UK. 3 Division of Clinical Neurology, University of Nottingham, UK. 4 Trent Research and Development Support Unit, University of Nottingham, UK. 5 University of Swansea, UK.

Corresponding author: Nadina Lincoln, University of Nottingham, International House, B Floor, Jubilee Campus, Wollaton Road, Nottingham NG8 1BB, UK Email: [email protected]


Lincoln et al.

1251

behaviour therapy approaches may be useful for reducing levels of distress. The Goldman Consensus group7 suggested that depression in people with MS can lead to a decrease in adherence to medication as well as a reduced quality of life. They recommended that people with MS should be treated using psychotherapeutic, psychopharmacologic or integrated approaches depending upon individual circumstances and personal preferences. The NICE guidelines in the UK have recognized that many patients with moderate depression and chronic health conditions, including patients with MS, do not wish to be treated with medication and that psychotherapy should also be offered.8 Reviews of psychological interventions9,10 have indicated that there is some evidence to support the provision of cognitive behaviour therapy (CBT) and group treatments, but the quality of the studies found was poor and sample sizes were small. Although there is some support for individual psychological treatments, many MS services do not have the resources to provide these for all patients with MS who have low mood, and therefore group treatments have been introduced into clinical practice. Some group treatments include any patients with MS and have not restricted attendance only to those with low mood. For example, Tesar et al.11 compared a group treatment programme, based on cognitive behavioural principles to teach patients coping strategies to deal with stress, with a usual care control group. Although there was a decrease in depressive symptoms over time, there were no significant differences between the groups. Rigby et al.12 compared MS patients randomly allocated to group psychological therapy, a social discussion support group and a booklet-only control group. They found that both group treatments reduced anxiety and improved self-efficacy, compared with the booklet-only control. In contrast, Forman and Lincoln13 selectively recruited patients with MS and low mood and conducted a pilot randomized controlled trial comparing a support group based on cognitive behavioural principles with a waiting list control. The intervention focused on examining the interrelationships between thoughts, mood and behaviour and how changing thoughts or behaviour can lead to a change in mood. Sessions covered specific topics including problem-solving, realistic target-setting, worry (anxiety), gloom (depression), relationships and the future. In each session, participants were taught strategies to identify sources of distress and skills to reduce current and future distress. The sessions were designed to increase awareness of the role of thoughts, emotions and behaviours and their influence on each other, in order to facilitate coping and adjustment. Forman and Lincoln13 found significant differences between the groups on the General

Health Questionnaire 12, a measure of psychological distress. The aim of the current study was to evaluate further the group programme developed by Forman and Lincoln.13

Method Patients with a confirmed diagnosis of MS were identified from a register of patients attending clinics for people with MS at Nottingham University Hospitals NHS Trust. They were contacted by letter and invited to take part in a study on MS and mood. In addition, referrals were invited from specialist MS nurses and advertisements placed in MS Society publications inviting those interested in being involved to contact the researchers. Potential participants were sent a screening pack containing information about the study and two mood measures to complete and return by post. These were the General Health Questionnaire 12 (GHQ-12),14 a measure of psychological distress, and the Hospital Anxiety and Depression Scale (HADS),15 a measure of anxiety (HADA) and depression (HADD). Patients were eligible for inclusion if they scored three or more on the GHQ-12 or eight or more on the HADA or HADD sub-scales. Participants who were eligible for inclusion were contacted by telephone and an appointment was made to visit them at home. Patients were excluded at this visit if they had had a diagnosis of MS for less than 12 months, were unable to speak and understand conversational English, were unable to attend group sessions if offered group treatment or were engaged in other psychological research. Those who met the criteria and who gave informed consent were included in the study. They were asked to complete further baseline measures: . Beck Depression Inventory-II (BDI),16 a measure of depression, which has been used in previous studies of psychological treatments for people with MS11,17 and therefore could be used for meta-analysis. . Multiple Sclerosis Impact Scale-29 (MSIS),18 a measure of quality of life. Participants were asked to rate from 1 (not at all) to 5 (extremely) how much specific symptoms, such as ‘heavy arms and/or legs’ or ‘worries about my MS’, were affecting them. The scale has two summary scores to reflect the physical and psychological effects of MS. . MS Self-efficacy Scale (MSSE),19 a measure of adjustment. Participants were asked to indicate their level of agreement or disagreement with 14 statements, such as ‘Sometimes I feel inadequate because of my condition’. . Guy’s Neurological Disability Scale (GNDS),20 to measure impairment specifically due to MS.


1252

Multiple Sclerosis Journal 17(10)

This assesses 12 aspects of life that may be affected by MS, such as memory, concentration and mobility. . EQ5D,21 a generic quality of life measure. . A service use questionnaire, to assess the use of NHS services and medication. This data was included in cost-effectiveness analysis, which will be included in a separate publication. Recruitment continued until eight participants had completed the baseline assessments and were all able to attend group treatment at the same location. These eight participants were then randomly allocated, as a cluster, to either intervention or waiting list. The random allocation was carried out using a web-based randomization program prepared in advance by the Clinical Trials Unit at Nottingham University. This ensured that the researcher recruiting the participants was blinded to allocation. Participants randomized to the intervention were invited to attend a group therapy programme for six sessions over 12 weeks. Treatment comprised the group programme developed by Forman and Lincoln.13 The treatment manual was used to ensure consistency in the delivery of the sessions. The groups were delivered by a research psychologist under the supervision of a clinical psychologist, who was experienced in working with people with MS. Participants allocated to the waiting list received all other rehabilitation routinely provided and were offered the opportunity to attend groups once they had completed the study. Four and 8 months after random allocation, participants were sent a booklet of outcome questionnaires to complete and return by post. This booklet included the GHQ-12 and HADS in addition to measures completed at the baseline visit. The GNDS was included to assist in the interpretation of results but was not an outcome measure. Those who were not able to complete the outcome measures themselves, such as people with visual problems, were visited by a researcher who was unaware of the group allocation, who completed the questionnaires with them. If participants failed to return outcome questionnaires they were telephoned once to remind them. If questionnaire items were missed a researcher telephoned to complete the missing items over the phone. All questionnaires were scored and entered onto a database by a researcher blinded to group allocation. These were subsequently checked by another researcher who was also blind to group allocation. The primary outcome measure was the GHQ-12 at 4 months’ followup, which was scored using Likert scoring (0, 1, 2, 3 for each item). High scores on the mood questionnaires (GHQ-12, HADS, BDI) and MSIS indicate worse outcomes. On the MSSE and EQ5D, high scores indicate better outcomes. Results were analysed using PASW statistics.21

The results of the pilot study were used to inform the sample size calculation. In this study a difference in means between the groups on the GHQ-12 of 4.64 and a combined standard deviation of 7.46 was observed. The sample size calculation was based on a smaller difference between groups, as it was expected that the difference might be less than in the pilot study and also that it would be worth detecting a smaller difference. Assuming a power of 80% and two-sided alpha of 5%, 63 participants per group were required to detect a difference in means of 3.75. The number of participants to be recruited was increased to 144 in total (72 per group) to allow for a drop-out rate of 10%, based on the pilot study, and recruitment in clusters of eight. Comparison of those who gave consent and those who refused was conducted using t-tests for continuous data and 2 for categorical data. Analysis of outcomes was undertaken on an intention to treat basis, using multiple regression analysis for outcomes at 4 and 8 months. Initially a complete case analysis was performed, and then a sensitivity analysis was conducted in which missing outcome data were replaced using the last value carried forward procedure. A series of models was fitted for each outcome measure. The first model was an unadjusted comparison of outcomes between groups. The second model compared outcomes between groups after adjustment for the outcome score at baseline and disability, as measured by the GNDS. Finally, the effect of receiving ‘treatment’ within a cluster was investigated by extending the second model to include cluster as a random effect. In order to assess the overall effects of intervention, the area under the curve was calculated for each variable and compared using a Mann–Whitney U test.

Results Between 1 June 2008 and 30 September 2009, 311 patients with MS were referred to the study. Of these 221 (71%) were found to have low mood (a score 3 on the GHQ-12 or 8 on either subscale of the HADS). Of those with low mood, 151 (68%) agreed to take part in the study. Of those with low mood who were not included, two had already been involved in the pilot study,13 six were too soon after diagnosis, two lived too far from the location where the groups were being held, and 60 declined. Comparison of those who consented to take part with those who refused showed no significant differences in gender, type of MS, age or HADA, but those who consented to take part were significantly more depressed on the HADD ( p ¼ 0.02) and had significantly higher levels of distress on the GHQ-12 ( p ¼ 0.009). The CONSORT diagram in Figure 1 shows the recruitment and follow-up of participants.


Lincoln et al.

1253

Screened for low mood (n = 311) Mood not low (n = 90)

Eligible for study (n = 221)

Recruited to study and randomly allocated n = 151

Intervention Group (n = 72)

Excluded Took part in pilot study (n = 2) Too soon after diagnosis (n = 6) Too far from group site (n = 2) Did not want to take part (n = 60)

Waiting List (n = 79)

Too ill (n = 2) Failed to return (n = 9)

Too ill (n = 1) Failed to return (n = 4) Returned 8 months but not 4 months (n = 2)

Withdrew (n = 1) Too ill (n = 1) Failed to return (n = 7)

4 month Outcome (n = 61)


Failed to return (n = 3) Returned 8 months but not 4 months (n = 4)



Figure 1. Recruitment and follow-up of patients in the trial.

There were 72 participants in the intervention group and 79 participants in the waiting list control group. Baseline characteristics are shown in Table 1. The groups were well matched in baseline characteristics. The distribution of scores on questionnaire measures at baseline is also shown in Table 1. The groups were well matched on all baseline variables. At the 4-month outcome, 19 participants failed to return or were too ill to complete their follow-up questionnaires, and one participant withdrew from the study shortly after being randomized. At the 8- month follow-up a further eight failed to return follow-up questionnaires, but six who had not returned questionnaires at 4 months did so at 8 months. The distribution of scores on the outcome measures is shown in Table 2. Participants allocated to intervention attended between zero and six group treatment sessions (mean 3.5, SD 2.5 sessions).

The results of the regression analyses on those who completed outcome measures are shown in Table 3. At 4 months group allocation alone was a significant predictor of the primary outcome measure, the GHQ-12. On the secondary outcome measures, allocation alone was also a significant predictor of outcome on the BDI and MSIS Physical. When baseline values and disability were controlled for, GHQ-12, BDI and MSIS Physical remained significant and the effect of allocation became significant for HADA, HADD, MSSE and EQ5D. There was no significant effect of allocation on MSIS Psychological. At 8 months, group allocation alone was no longer a significant predictor for GHQ-12 scores, but it was when baseline GHQ-12 and GNDS scores were controlled for. Group allocation alone significantly predicted the 8-month outcome scores on HADA, HADD, BDI, MMSE, MSIS physical subscale and MSIS psychological subscale but not the EQ5D.


1254


Table 1. Characteristics and baseline scores of participants Intervention n ¼ 72

Type of MS Relapsing Remitting Primary Progressive Secondary Progressive Benign Age Years since diagnosis Guy’s Neurological Disability Scale GHQ-12 HAD Anxiety HAD Depression BDI MSSE MSIS Physical MSIS Psychological EQ5D

Waiting List Control n ¼ 79

n

%

n

%

55 4 12 1 Mean 44.5 9.2 17.3 22.4 11.5 9.5 23.1 42.0 43.9 51.5 0.53

76 6 17 1 SD 11.1 7.8 7.8 6.7 4.1 4.0 12.2 11.2 23.2 22.4 0.30

48 11 18 2 Mean 47.5 10.5 16.7 20.8 11.3 9.2 21.9 41.1 47.4 50.2 0.53

61 14 23 2 SD 10.5 8.0 6.9 6.8 3.2 3.7 8.7 9.3 22.5 19.7 0.28

BDI: Beck Depression Inventory, EQ5D: a generic quality of life measure, GHQ-12: General Health Questionnaire 12, GNDS: Guy’s Neurological Disability Scale, HAD: Hospital Anxiety and Depression Scale, MS: multiple sclerosis, MSSE: MS self-efficacy scale, MSIS: MS impact scale, SD: standard deviation.

Table 2. Scores on outcome measures at 4 and 8 months Intervention

Waiting List Control

Measure

Time

n

Mean

SD

n

Mean

SD

General Health Questionnaire 12

4 8 4 8 4 8 4 8 4 8 4 8 4 8 4 8 4 8

61 58 61 58 61 58 60 58 61 58 61 58 61 58 61 58 61 58

16.3 15.4 9.2 8.3 8.2 7.6 17.3 16.3 44.6 46.7 44.5 40.8 45.2 40.6 0.57 0.56 16.1 16.8

7.2 6.4 4.4 4.5 4.0 4.0 10.1 9.9 12.3 11.2 25.1 26.5 25.1 25.1 0.30 0.29 7.9 7.4

69 69 70 68 70 68 70 69 70 67 70 68 70 67 70 67 70 66

19.2 17.5 10.2 10.7 9.5 9.3 22.4 20.9 41.0 41.8 53.3 52.6 49.9 52.6 0.49 0.52 17.0 17.8

6.7 7.0 3.7 3.7 3.8 4.1 9.1 10.4 10.0 9.9 21.9 21.4 20.0 22.6 0.31 0.31 7.0 7.0

HADS Anxiety HADS Depression Beck Depression Inventory MS Self-efficacy Scale MSIS Physical MSIS Psychological EQ5D Guy’s Neurological Disability Scale

HADS: Hospital Anxiety and Depression Scale, MSIS: MS Impact Scale.


Lincoln et al.

1255

Table 3. Regression analysis of outcome measures at 4 and 8 months on participants with outcome data 4 months B GHQ-12 Unadjusted difference Difference adjusted for HADS Anxiety Unadjusted difference Difference adjusted for HADS Depression Unadjusted difference Difference adjusted for BDI Unadjusted difference Difference adjusted for MSSE Unadjusted difference Difference adjusted for MSIS Physical Unadjusted difference Difference adjusted for and disability MSIS Psychological Unadjusted difference Difference adjusted for and disability EQ5D Unadjusted difference Difference adjusted for

8 months

Beta

95% CI

p

B

Beta

95% CI

p

2.9 baseline GHQ-12 and disability 3.6

0.21 0.26

5.8 to 1.4 5.8 to 1.4

0.019 2.1 0.001 2.7

0.15 0.20

4.4 to 0.3 4.9 to 0.5

0.088 0.016

baseline HADA and disability

1.0 1.2

0.12 0.15

2.4 to 0.4 2.2 to 0.2

0.160 2.4 0.028 2.5

0.28 0.30

3.8 to 0.9 3.7 to 1.3

0.002 0.001

baseline HADD and disability

1.3 1.5

0.17 0.20

2.7 to 0.0 2.7 to 0.4

0.052 1.6 0.008 1.9

0.20 0.23

3.1 to 0.2 3.1 to 0.7

0.026 0.003

baseline BDI and disability

5.1 5.8

0.26 0.30

8.4 to 3.4 8.3 to 3.4

0.003 4.6 0.001 5.3

0.22 0.26

8.1 to 1.0 8.1 to 2.5

0.013 0.001

baseline MSSE and disability

3.6 0.16 0.2 to 7.5 0.064 4.9 3.1 0.14 0.4 to 5.8 0.025 4.4

0.23 1.2 to 8.7 0.010 0.21 1.8 to 7.1 0.001

baseline MSIS Physical

8.8 6.5

0.19 0.14

16.9 to 0.6 12.3 to 0.8

0.035 11.9 0.026 9.7

0.24 0.20

20.3 to 3.5 15.1 to 4.4

0.006 0.001

baseline MSIS Psychological

4.7 5.6

0.10 0.12

12.5 to 3.1 11.8 to 0.6

0.237 12.0 0.077 13.7

0.25 0.27

20.4 to 3.5 19.8 to 6.4

0.006 0.001

baseline EQ5D and disability

0.1 0.14 0.1 to 0.2 0.1 0.14 0.0 to 0.2

0.117 0.04 0.07 0.1 to 0.2 0.041 0.05 0.08 0.1 to 0.1

0.423 0.127

BDI: Beck Depression Inventory, GHQ-12: General Health Questionnaire 12, GNDS: Guy’s Neurological Disability Scale, HADS: Hospital Anxiety and Depression Scale, MSIS: MS Impact Scale, MSSE: MS Self-efficacy Scale.

Comparison of the area under the curve revealed significant differences between the groups for GHQ-12 ( p ¼ 0.003), HADA ( p ¼ 0.013), HADD ( p ¼ 0.004), BDI ( p ¼ 0.001), MSSE ( p ¼ 0.037) and MSIS Psychological (0.012) but no significant differences in MSIS Physical ( p ¼ 0.149) or EQ5D ( p ¼ 0.219). As some patients did not complete outcome assessments, a sensitivity analysis was also conducted. Results are shown in Table 4. This showed that allocation alone did not significantly predict outcomes at 4 months. When baseline values and disability were controlled for, there was a significant effect of allocation on GHQ-12, HADD and BDI. When cluster effects were added to the model as a random effect, the results were the same as obtained in the second model. At 8 months there was a significant effect of allocation alone on HADA, MSIS Physical and MSIS Psychological. When baseline values and disability were controlled for, there was also a significant effect of allocation on

HADD and BDI. When cluster effects were controlled for, the results were the same. Comparison of the area under the curve revealed significant differences between the groups for GHQ-12 ( p ¼ 0.006), HADA ( p ¼ 0.010), HADD ( p ¼ 0.009), BDI ( p ¼ 0.005) and MSIS Psychological (0.039) but no significant differences in MSSE ( p ¼ 0.159), MSIS Physical ( p ¼ 0.097) or EQ5D ( p ¼ 0.387).

Discussion Allocation to intervention or a waiting list control group significantly predicted scores on mood questionnaires 4 and 8 months after randomization. The overall summary scores of area under the curve showed significant differences in mood (GHQ-12, HADA, HADD, BDI) and psychological impact of MS. This suggests that the group treatment was effective for reducing low mood in patients with MS, and supports the results of the


1256


Table 4. Regression analysis of outcome measures at 4 and 8 months for all participants, replacing missing outcome data 4 months

GHQ-12 Unadjusted difference Difference adjusted for HADS Anxiety Unadjusted difference Difference adjusted for HADS Depression Unadjusted difference Difference adjusted for BDI Unadjusted difference Difference adjusted for MSSE Unadjusted difference Difference adjusted for MSIS Physical Unadjusted difference Difference adjusted for and disability MSIS Psychological Unadjusted difference Difference adjusted for and disability EQ5D Unadjusted difference Difference adjusted for

8 months

B

Beta

95% CI

p

baseline GHQ-12 and disability

1.4 2.3

0.10 0.16

3.7 to 0.9 4.3 to 0.3

baseline HADA and disability

0.5 0.7

0.06 0.08

baseline HADD and disability

1.1 1.3

baseline BDI and disability

2.7 3.5

baseline MSSE and disability

2.4 0.11 1.1 to 6.0 1.9 0.09 0.5 to 4.2

baseline MSIS Physical

6.6 4.1

0.14 0.09

baseline MSIS Psychological

1.6 2.6

0.04 0.06

baseline EQ5D and disability

0.1 0.06 0.1 to 0.2 0.0 0.14 0.0 to 0.1

B

Beta

95% CI

p

0.227 0.7 0.027 1.4

0.05 0.10

3.0 to 1.6 0.548 3.5 to 0.6 0.161

1.9 to 0.8 1.7 to 0.2

0.455 1.7 0.144 1.8

0.12 0.22

3.1 to 0.3 3.0 to 0.8

0.14 0.17

2.3 to 0.1 2.3 to 0.3

0.084 1.2 0.010 1.5

0.15 0.18

2.5 to 0.1 0.067 2.5 to 0.4 0.007

0.13 0.17

6.1 to 6.1 5.8 to 1.2

0.109 2.3 0.003 3.1

0.10 0.14

5.8 to 1.2 0.202 5.6 to 0.6 0.015

0.176 2.6 0.120 2.0

0.12 0.10

0.8 to 6.0 0.140 0.3 to 4.4 0.088

14.3 to 1.1 0.091 8.9 9.2 to 1.0 0.114 6.3

0.18 0.13

16.8 to 1.1 11.0 to 1.6

0.025 0.009

9.0 to 5.7 8.1 to 2.9

0.662 8.1 0.353 9.2

0.17 0.19

15.8 to 0.4 15.2 to 3.3

0.039 0.003

0.459 0.01 0.280 0.01

0.02 0.02

0.1 to 0.1 0.792 0.1 to 0.6 0.771

0.014 0.001

BDI: Beck Depression Inventory, GHQ-12: General Health Questionnaire 12, GNDS: Guy’s Neurological Disability Scale, HADS: Hospital Anxiety and Depression Scale, MSIS: MS Impact Scale, MSSE: MS Self-efficacy Scale.

pilot study conducted by Forman and Lincoln.13 It also supports evidence suggesting that CBT approaches are useful for people with MS17 and that group interventions are beneficial.11,12 However, not all patients attended the group sessions, and the mean number of sessions attended was only 3.5. Although all participants agreed to take part in groups if invited, only 63% of those randomized to receive the intervention attended three or more sessions. There were many reasons for this, such as prior hospital appointments, work commitments and relapses. This suggests that those people who did attend must have benefited to a greater extent than indicated by the results, compensating for the fact that many patients received very little of the intervention. Subjective impressions of the therapists were that most participants who attended the groups found them useful and seemed to make progress. Of those eligible for the study, not all decided to take part. Comparison of those who agreed and those who

declined indicated that people who were more depressed were more likely to take part in the study. They may have had most to gain from the therapy. It is also possible that the criteria for including people in the study on the basis of their mood scores were quite low, so that we offered treatment to those who did not really feel they needed it. This selection process also accounts for the high rate of depression in this sample (71%) as compared with previous research. Those with depression are probably more likely to respond to an invitation to take part in a treatment trial for those with low mood. The usual first choice of treatment is anti-depressant medication, and most patients reported having previously tried antidepressants. Current use was not recorded at the time of random allocation. However, the findings suggest that psychological treatment was of benefit even in patients who had previously been treated with anti-depressants. A limitation of the study is that treatment was delivered almost exclusively by one therapist. Although this


Lincoln et al.

1257

raises the possibility that the findings may be specific to this therapist, these findings support the results from the study by Forman and Lincoln,13 in which treatment was delivered by a different therapist. The therapist in the present study was relatively inexperienced and was probably fairly representative of NHS assistant psychologists. In addition, all patients were recruited through one MS service. This service had very little input from clinical psychologists, and therefore patients in the waiting list control group may have received less support than occurs in other centres. A multi-centre trial with several therapists is needed to determine the generalizability of these findings. For pragmatic reasons, participants were not randomized individually but in clusters of eight people who could attend group treatment if allocated to it, at the same time and venue. In order to assess the effects of this, a regression model was used which included cluster as a random effect. However, the close relation between group and cluster led to the problem of overparameterization, which meant that cluster effects could not be fully explored. The results of this study suggest that group-based interventions based upon CBT principles were successful at improving the mood of patients with MS who had low mood. Moreover, they also had a wider effect, as they showed evidence of reducing the impact of MS in people’s everyday lives. These results are very encouraging, given the high levels of distress that occur in people with MS. Conflict of interest

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

None declared.

15.

Funding This work was supported by the Multiple Sclerosis Society (grant number 870/07).

16.

Acknowledgements

17.

We would like to thank Emma Ford and Angelica Attard for assistance with data checking and outcome assessments, the MS nurses for assistance with recruitment, Karen Treece for supervising the group treatment, and our MS Partners, who provided useful feedback from the service user’s perspective.

18.

References 1. Dalton EJ and Heinrichs RW. Depression in multiple sclerosis: a quantitative review of the evidence. Neuropsychology 2005; 19: 152–158. 2. Sa´ MJ. Psychological aspects of multiple sclerosis. Clin Neurol Neurosurg 2008; 110: 868–877. 3. Siegert RJ and Abernethy DA. Depression in multiple sclerosis: a review. J Neurol Neurosurg Psychiatry 2005; 76: 469–475. 4. Bamer AM, Cetin K, Johnson KL, Gibbons LE and Ehde DM. Validation study of prevalence and correlates of

19.

20.

21.

depressive symptomatology in multiple sclerosis. Gen Hosp Psychiatry 2008; 30: 311–317. Dahl O, Stordal E, Lydersen S and Midgard R. Anxiety and depression in multiple sclerosis. A comparative population-based study in Nord-Trøndelag County, Norway. Mult Scler 2009; 15: 1495–1501. Dennison L, Moss-Morris R, Silber E, Galea I and Chalder T. Cognitive and behavioural correlates of different domains of psychological adjustment in early-stage multiple sclerosis. J Psychosom Res 2010; 69: 353–361. Goldman Consensus Group. The Goldman Consensus statement on depression in multiple sclerosis. Mult Scler 2005; 11: 328–337. National Institute for Health and Clinical Excellence. Depression in adults with a chronic physical health problem: treatment and management. NICE Clinical Guideline 91. London: NICE, 2009. Thomas PW, Thomas S, Hillier C, Galvin K and Baker R. Psychological interventions for multiple sclerosis. Cochrane Database of Systematic Reviews 2009; (1)CD004431. Malcomson KS, Dunwoody L and Lowe-Strong AS. Psychosocial interventions in people with multiple sclerosis: a review. J Neurol 2007; 254: 1–13. Tesar N, Baumhackl U, Kopp M and Gunther V. Effects of psychological group therapy in patients with multiple sclerosis. Acta Neurol Scand 2003; 107: 394–399. Rigby SA, Thornton EW and Young CA. A randomized group intervention trial to enhance mood and selfefficacy in people with multiple sclerosis. Br J Health Psychol 2008; 13: 619–631. Forman AC and Lincoln NB. Evaluation of an adjustment group for people with multiple sclerosis: a pilot randomized controlled trial. Clin Rehabil 2010; 24: 211–221. Goldberg D. General Health Questionnaire (GHQ-12). Windsor, UK: NFER-Nelson, 1992. Zigmond AS and Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361–370. Beck AT, Steer RA and Brown GK. Beck Depression Inventory-II (BDI-II). San Antonio, TX: Psychological Corporation, 1996. Mohr DC, Boudewyn AC, Goodkin DE, Bostrom A and Epstein L. Comparative outcomes for individual cognitive-behavior therapy, supportive-expressive group psychotherapy, and sertraline for the treatment of depression in multiple sclerosis. J Consult Clin Psychol 2001; 69: 942–949. Hobart J, Lamping D, Fitzpatrick R, Riazi A and Thompson A. The Multiple Sclerosis Impact Scale (MSIS-29): a new patient-based outcome measure. Brain 2001; 124: 962–973. Domenech CL, Thornton EW, Young CA, Rigby SA and Tedman S. Development of a self efficacy scale for people with multiple sclerosis. Mult Scler 1999; 5: S144. Sharrack B and Hughes RAC. The Guy’s Neurological Disability Scale (GNDS): a new disability measure for multiple sclerosis. Mult Scler 1999; 5: 223–233. The EuroQol group. EuroQol – a new facility for the measurement of health-related quality of life. Health Policy 1990; 16: 199–208.
