fluoxetine and graded exercise for chronic fatigue ...

Randomised, double-blind, placebo-controlled treatment trial of fluoxetine and graded exercise for chronic fatigue syndrome.

A J Wearden, R K Morriss, R Mullis, P L Strickland, D J Pearson, L Appleby, I T Campbell and J A Morris BJP 1998, 172:485-490. Access the most recent version at DOI: 10.1192/bjp.172.6.485

References

Reprints/ permissions You can respond to this article at Downloaded from

This article cites 0 articles, 0 of which you can access for free at: http://bjp.rcpsych.org/content/172/6/485#BIBL To obtain reprints or permission to reproduce material from this paper, please write to [email protected] http://bjp.rcpsych.org/cgi/eletter-submit/172/6/485 http://bjp.rcpsych.org/ on April 28, 2012 Published by The Royal College of Psychiatrists

To subscribe to The British Journal of Psychiatry go to: http://bjp.rcpsych.org/site/subscriptions/

B R I T I S H J O U R N A L OF P S Y C H I A T R Y ( 1 9 9 1 ) , 111, 4 1 5 - 4 9 0

Randomised, double-blind, placebo-controlled treatment trial of fluoxetine and graded exercise for chronic fatigue syndrome? A.J.WEARDEN, R. K. MORRISS, R. MULLIS, P. L. STRICKLAND, D. J. PEARSON, L. APPLEBY. I.T. CAMPBELL and J. A. MORRIS

Background The Joint Working Group ofthe Royal Colleges of Physicians, Psychiatrists and General Practitioners (1996) recommendedgraded exercise and antidepressantsfor patients with chronic fatigue syndrome.We assessed efficacy and acceptability ofthese treatments.

Method Six-month prospective randomised placebo and therapist contact time controlled trial with allocation to one offour treatment cells: exercise and 20 mg fluoxetine, exercise and placebo drug, appointments only and 20 mg fluoxetine, appointments and placebo drug. Drug treatment was double blind and patients were blind to assignment to exercise or appointments.

Results

Ninety-six (71%)of I36

patients completed the trial. Patients were

The Joint Working Group of the Royal Colleges of Physicians, Psychiatrists and General Practitioners Report (1996) advocated the use of antidepressants in depressed chronic fatigue syndrome (CFS) patients and graded exercise in all CFS patients. A controlled trial of fluoxetine found no benefit in chronically ill CFS patients (Vercoulen et al, 1996) in contrast to some open trials of antidepressants (Butler et al, 1991; Behan et al, 1994; White & Cleary, 1997). Graded exercise improved fatigue and physiological outcomes in CFS patients without psychiatric or sleep disorders compared to flexibility exercises (Fulcher & White, 1997).The aim of the present study was to determine the efficacy and adherence to six months' treatment with fluoxetine and a graded exercise programme, prescribed and supervised by a physiotherapist, in CFS outpatients with or without depression. Both treatments are simple to administer and readily available in most hospitals.

more likely to drop out ofexercise than non-exercisetreatment (P=0.05). In an intention to treat analysis, exercise resulted in fewer patients with case level fatigue than appointments only at 26

METHOD The trial was approved by the South Manchester Medical Ethics Committee. Patients gave written informed consent.

weeks (12 (18%) v. 4 (6%) respectively

Design

P=0.025) and improvement in functional

Patients

work capacity at 12 (P=0.005) and 26

Patients over 18 years of age were recruited from consecutive referrals to a university department of medicine out-patient clinic drawing from across north-west England and north Wales between June 1993 and March 1995. All patients were medically assessed by a doctor (R.M., P.S. and D.P.) under the supervision of a consultant physician (D.P.). A full history, physical examination and laboratory investigations (full blood count, erythrocyte sedimentation rate, biochemistry, creatinine phosphokinase, c-reactive protein and immunological screen including thyroid antibodies) were carried out.

weeks (P=0.03). Fluoxetine had a significant effect on depression at week 12 only (P=0.04). Exercise significantly improved health perception (P=O.O12) and fatigue (P=0.028) at 28 weeks.

Conclusions

Graded exercise

produced improvements in functional workcapacity and fatigue, while fluoxet~ne improved depression only. 'see commentary pp. 491 -492, t h ~ sIssue.

All subjects met operationalised 'Oxford' research criteria (Sharpe et al, 1991) for CFS: (a) a principal complaint of fatigue of at least six months' duration, exacerbated by physical activity (and usually mental activity); (b) impairment in three out of four areas of activity (activities of daily living, occupational, social or leisure activities); (c) no medical cause of fatigue. Subjects with schizophrenia, bipolar disorder, eating disorder, alcohol or illicit drug misuse, those with current suicidal ideation, a history of ischaemic heart disease or an inability to read and write English were excluded. Pre-menopausal women were required to take precautions against pregnancy during the trial. Subjects taking antidepressant medication were required to stop and undergo at least a two week washout period before entering the trial. Antidepressants were not withdrawn (and patients were excluded from the trial) if patients were judged to have any significant suicidal risk. A pre-study power calculation of the sample size using a criterion of 5 0 % improvement in depression symptoms in 60% of a fluoxetine-treated group and 30% of a placebo-treated group, suggested that 50 subjects per group would be required to show a significant between-groups difference, assuming a drop-out rate of 25%, with power of 80% at the 5% significance level. A further power calculation was carried out when 50 subjects had reached 12 weeks. At 80% power and 5 % significance level, 35 subjects would be required in each group (140 in total) to detect a mean change in fatigue scale of 20% from a baseline mean score of 14 (s.d. 4).

A fixed daily dose of 20 mg fluoxetine and a graded exercise programme were placebo controlled and controlled for the amount of therapist contact. The drug treatment was double blind. The placebo to fluoxetine was a capsule of similar taste and appearance. The placebo to the exercise programme was a review of activity diaries by the physiotherapist. With respect to the exercise programme, subjects were told that they would receive one of two forms of lifestyle advice and were blind to the other type of advice. The design and patient flow into the study are shown in Fig. 1. No attempt was made to provide psychological treatment.

I

227 patients screened 165 were eligible

I

29 not randomised of whom 25 refused 3 could not come off antidepressants 1 required orthopaedic treatment

-

/

33 randomised

to graded exercise and fluoxetine I

I

14 dropped out from treatment of whom no. assessed at week 0=14 week 12=7 week 26=9 and 5 lost to follow-up


lscMlpletedtlial and all assessments of whom 9 complied fully with graded exercise 19 complied with fluoxetine for 6 months Fig. I

35 randomised to exercise placebo and fluoxetine

34 randomised to graded exercise and drug placebo

,

I

34 randomised to

exercise placebo and drug placebo I


1

I

23 completed trial and all assessments of whom 14 complied fully with graded exercise 20 complied with drug placebo for 6 months

25 completed trial and all assessments of whom 25 complied fully with exercise-placebo 24 complied with fluoxetine for 6 months

I

5 dropped out from treatment of whom no, assessed at week 12=3 week 26=3 and 2 lost to follow-up 29 completed trial and all assessments of whom 29 complied fully with exercise-placebo 24 complied with drug placebo for 6 months

Design and patient flow into treatment trial.

Assignment Immediately after initial assessment, subjects were randomised into a treatment group by computer generated random numbers, with groups being balanced in blocks of 10 to obtain roughly equal numbers. No stratification was employed. A list of subject numbers marked with the exercise group for each number was held by the physiotherapist. Pharmacy staff dispensed medication in accordance with the subject number assigned to each subject. Assessments and outcome measures Subjects were assessed on entry to the trial (week 0) and at weeks 12 and 26. Subjects completed three self-rated questionnaires. These were: (a) the 14-item fatigue scale (Chalder et all 1993) - we used a cut-off of four or more to designate caseness; (b) the Medical Outcomes Survey Short-Form Scales (MOS; Stewart et al, 1988) which produces a measure of general health status on the following six scales (cut-off scores for poor function in parentheses): physical function ( < 83.3), role or occupation function ( < =SO), social function ( < =40), pain ( < =SO), health perceptions ( < =70) and

mental health ( < =67); (c) the Hospital Anxiety and Depression Scales (HAD; Zigmond & Snaith, 1983) - cut-offs of 11 or more designated caseness. On completion of the questionnaires, subjects were interviewed by the psychologist with the revised Clinical Interview Schedule (Lewis et all 1992). Supplementary questions were asked to enable a diagnosis to be made in accordance with DSM-111-R criteria (American Psychiatric Association, 1987). Physiological assessment included measurement of height, weight, body fat, grip strengths and functional work capacity. The last was determined using a Bosch ERG 551 electronically braked cycle ergometer. Oxygen consumption and carbon dioxide production were determined using a P.K. Morgan Exercise Test System (P.K. Morgan, Rainham, Kent). Subjects' predicted maximum heart rate was calculated from the formula: 210-0.65 x (age in years), and heart rate was monitored throughout the test. Most subjects reached subjective exhaustion before they had reached their predicted maximum heart rate and before a plateau in oxygen consumption had been reached; consequently no extrapolation to theoretical

maximum oxygen uptake was attempted. Functional work capacity was calculated as the amount of oxygen (in millilitres) consumed in the final minute of exercise per kilogram of body weight, and was used as the physiological outcome measure. The main outcome variables were the changes in score on the fatigue scale and the percentages of subjects who scored below case level on the fatigue scale (Chalder et al, 1993) from baseline to weeks 12 and 26. Secondary outcome variables were changes in functional work capacity, and changes in scores on the MOS scales and the HAD scale.

Treatments All subjects attended the hospital on eight occasions for treatment by the physiotherapist on weeks 0, 1, 2, 4, 8, 12, 20 and 26. Subjects were assessed at weeks 0, 12 and 26, as specified in the study protocol. A check-list of drug side-effects was completed at each visit. Subjects randomised to graded exercise were instructed to carry out their preferred aerobic activity (usually walkingljogging, swimming or cycling), for 20 minutes, at least three times per week. The intensity of activity was initially

TREATMENT FOR CHRONIC FATIGUE S I N D U O M E

set at a level which utilised oxygen at approximately 75% of the subject's tested functional maximum (American College of Sports Medicine, 1995). Subjects monitored their prescribed exercise programmes on a chart along with pre- and post-exercise heart rates and perceived exertion (Borg, 1982). Exercise intensity was increased when there was a consistent recorded reduction of 10 beats per minute in postexercise heart rate for one week and two points on the perceived exertion scale (about three times in six months in an adherent patient). Subjects adhered to the exercise programme if their charts showed that they had performed the required activity, at the required intensity, at least three times per week. Subjects randomised to the exercisecontrol groups were not offered any specific advice on how much exercise they should be taking, but were told to do what they could when they felt capable and to rest when they felt they needed to. Subjects who attended the required appointments adhered to the non-exercise treatment. Every fourth week, subjects in all groups were required to complete a diary of activities over a seven day period. These were discussed with the physiotherapist at the next visit. Where possible, subjects who dropped out of the trial were assessed immediately and at the next scheduled assessment date. Statistical analysis Statistical analysis was carried out using SPSS for Windows, Release 6. Analysis was carried out on an intention to treat basis. When there were missing data at weeks 12 and 26, scores on the previous assessment were substituted. No data were available on 1 7 patients for the week 12 assessment, functional work capacity assessments at week 0, seven at week 12 and seven at week 26. Mean and 95% confidence interval

change scores were computed from the differences between scores at week 12 and baseline and week 26 and baseline (using Liken scoring for the fatigue scales). These change scores were analysed using factorial analyses of variance, with the two treatments entered as between subjects factors. Drop outs from the trial were analysed using Cox's proportional hazards model. Differences on categorical variables were examined using Pearson's x2 test or loglinear model analysis. RESULTS Table 1 shows demographic details and Table 2 baseline measurements of the patient sample on entry to the treatment trial. The four treatment cells showed no clinically significant differences on any demographic detail or baseline measurement. Sixty-two subjects (46%) fulfilled DSM-111-R criteria for a current psychiatric diagnosis: 14 (10%) had major depression, 32 ( 2 4 % ) either dysthymia or depressive disorder not otherwise specified, 14 (10%) various anxiety disorders and two subjects (2%) somatisation disorder. Drop-out rate and characteristics One hundred and fourteen (84%) subjects completed three months and 96 (71%) subjects completed six months of the trial. Subjects dropped-out at a greater rate throughout the six months with exercise than with non-exercise (25168 (37%) v. 151 69 (22%), Cox's proportional hazards, P=O.OS)). The difference in drop-out rate between subjects allocated to fluoxetine (24168 (36%)) or to placebo drug (16169 (24%)) did not reach significance. Eleven subjects dropped out because of medication side-effects (two taking placebo drug), 16 because they were not improving or feeling worse, and 13 gave other reasons or no reason for dropping-out.

Drop-outs were significantly more likely than trial completers to be members of a self-help organisation (15139 (39%) v. 20195 (21%), x2 (1)=4.34, P=0.04), to have changed or given up their occupation as a result of their illness (38140 (95%) v. 76/96 (79%), zZ (1)=5.22, P=0.02) and had significantly worse baseline scores on the MOS health perception scale (median (interquartile range) 5.0 (18.8) v. 15.0 (18.8) Mann-Whitney U=1433, P=0.02). There were no significant differences between drop-outs and trial completers on any other demographic feature nor baseline clinical measure. Intention to treat analysis of changes in scores Table 2 shows the effects of exercise and fluoxetine on fatigue, HAD depression and functional work capacity in all treatment cells. Changes in scores over time on all measures were significant at the 1% level. There was a significant effect of exercise on functional work capacity at both weeks 12 (n=132, mean change=2.0 (95% CI 0.60-3.49); F(1,28)=7.99, P=O.OOS) and 26 (n=132, mean change=1.9 (95% CI 0.15-3.69); F(1,128)=4.74, P=0.03). Fluoxetine had no significant effect on functional work capacity at either weeks 12 (n=132, mean change=-0.4 (95% CI -1.9 to 1.0)) or 26 (n=132, mean change=O.l (95% CI -1.7 to 1.9)). There were trends for exercise to improve fatigue scale scores at week 12 (n=136, mean change=2.l (95% CI -0.6 to 4.8); F(1,132)=2.36, P=0.13) and at week 26 (n=136, mean change=2.9 (95% CI -0.2 to 6.1); F(1,132)=3.35, P=0.07). Fluoxetine had no effect on the fatigue scale at week 12 or week 26. Improvement in fatigue scale scores between baseline and 26 weeks correlated with that in functional work capacity (Spearman r=0.34, P=0.003).

Table I Baseline patient characteristics,mean scores (s.d.) or number (%)

Gender, female (%) Age in years (s.d.) Duration of fatigue, median (IQR) months Changed occupation (%) Member self-helpgroup (%) IQR, incerquartile range.

Overall n=136

Exercise and fluoxetine n=33

97 (71) 38.7 (10.8) 28.0 (39.5) 1 14 (84) 35 (26)

22 (67) 38.2 (10.7) 29.5 (42.0) 29 (88) 7 (21)

Exercise and placebo Exercise placebo and Exercise placebo and drug n=34 fluoxetine n=35 placebo drug n=34 27 (79) 40.4 ( 1 1.9) 34.5 (42.5) 29 (85) 1 1 (33)

27 (77) 38.8 ( I 0.4) 30.5 (39.8) 28 (80) 10 (29)

21 (62) 37.6 (10.7) 22.0 (36.8) 28 (82) 701)

WEARDEN ET AL

Table 2

Baseline and change scores from baseline or fatigue scale, HAD depression and functional work capacity

Exercise and fluoxetine

Exercise and placebo

Exercise control and

Exercise control and

n=33

n=34

fluoxetine n=35

placebo n=34

Fatigue scale Mean (95% CI) week 0 score Mean (95% CI) week 0-12 score Mean (95% CI) week 0-26 score H A D depression Mean (s.d.) week 0 score Mean change (95%CI) week 0-12 score Mean change (95% CI) week 0-26 score fwc. ml 0, kg-' min-' Mean (s.d.) week 0 score Mean change (95% CI) week 0-12 score Mean change (95%CI) week 0-26 score HAD, Hospital Anxiety and Depression Scale depression subscale; fwc, functional work capacity (data missing for four patients).

Fluoxetine significantly reduced the HAD depression score a t week 1 2 (n=136, mean change=l.l (95% CI 0.032.2); F(1,132)=4.24, P=0.04) but not at week 26 (n=136, mean change=0.6 (95% CI -0.6 to 1.9); F(1,132)=1.17, P=0.28). There were no significant effects of exercise on HAD depression scores at week 12 (n=136, mean change=-0.4 (95% CI - 1.5 to 0.7)) and week 26 (n=136, mean change=0.1 (95% CI - 1.2 to 1.3)). There were no significant changes on any MOS scale nor on the HAD anxiety scale in the intention to treat analysis. In trial completers there was a significant improvement in fatigue with exercise at 12 weeks (n=96, mean change=3.8 (95% CI 0.002-7.5); F(1,92)=5.08, P=0.027) and 26 weeks (n=96, mean change=4.5 ( 9 5 % CI 0.3-8.7); F ( 1 , 9 2 ) = 4 . 9 8 , P=0.028). In trial completers who adhered to the exercise treatment for six months there was a significant improvement on the MOS health perception scale (n=77, mean c h a n g e = 1 2 . 8 ( 9 5 % CI 2.9-22.8); F(1,73)=6.6, P=0.012) at 26 weeks but not at 12 weeks. Otherwise the results of our study are similar in trial ~0mpleterSand the whole sample included in the intention to treat analysis.

Intention to treat analysis of fatigue scale and H A D depression scale caseness

The effects of exercise and fluoxetine on case level fatigue are shown in Table 3. There was a dose response effect between adherence to prescribed exercise and the

percentage of non-cases of fatigue at six months: 8/23 (35%) subjects who fully adhered to the exercise prescription were non-cases, 3/21 (14%) who partially adhered, and 1/23 (4%) prescribed but not complying with exercise compared with 41 69 (6%) who were not prescribed exercise ( X Z (1) test for trend=7.14, P=0.008). Among subjects who adhered to drug treatment (fluoxetine (43168) v. placebo (44/68)), there was no significant difference between the number of non-cases of fatigue at six months in the fluoxetine and placebo groups. There were no effects of exercise on HAD case level of depression while the effects of fluoxetine are shown in Table 4.

Outcome for drop-outs

Twenty-one drop-outs were reassessed at week 26 (Fig. 1). For these patients, when mean change scores (weeks 0-26) were calculated for each of the four treatment groups, there was no worsening of scores on the fatigue scale, functional work Table 3

capacity, HAD depression scale and MOS health perception scale (data not shown).

DISCUSSION

Two treatments, which are within the resources of most general hospitals, are safe and produce clinically important but modest benefits to patients with CFS within three to six months. The high number of trial non-completers, the conservative practice of substituting the previous value for missing data, and regression to the mean or non-specific treatment effects may have resulted in an underestimate of the effects of graded exercise on fatigue and health perception. There was a 12% reduction in the number of patients with case level fatigue by 26 weeks and a 10% improvement in functional work capacity at 12 and 26 weeks. The only benefit of fluoxetine was a modest overall reduction in the HAD depression score at 12 weeks which was not sustained at 26 weeks, a finding

Number (%) patients in non-case range for fatigue, weeks 0.12 and 26

Exercise and

Exercise and

N o exercise and

N o exercise

fluoxetine

placebo

fluoxetine

and placebo

n=33

n=34

n=35

n=34

0

0

0

0

0

l2

6 (18)

1 (3)

1 (3)

2 (6)

26

6 (18)

6(18)

2 (6)

2 (6)

Fatigue, < 4

Effect of exercise on case level fatigue, y2(I) test for trend=5.0. P=0.025.

TREATMENT FOR C H R O N I C FATIGUE SYNDROME

which does not contradict the absence of clinical effect of 20 mg fluoxetine reported by Vercoulen et al (1996). A bigger clinical response might have been seen if a flexible fluoxetine dosage regime tailored to each patient's response to the drug was employed and more CFS patients (only 10%) with major depression were recruited.

CLINICAL IMPLICATIONS

Graded exercise produces small but clinically significant improvements in case level fatigue and functional work capacity in CFS patients.

20 mg fluoxetine per day produces smalt lmprovtments in depression in CFS patients. 8

Graded exercise may rmt be adhered to by CFS patients who are particularly

impaired.

Methodological limitations The effects of graded exercise in our study were not as large as those reported by Fulcher & White (1997). There are five possible reasons for the differences between the studies: our patients were on average 20% more impaired on the Likertscored Chalder fatigue scale; our study did not exclude patients with psychiatric disorder and sleep problems which are associated with a worse prognosis or greater disability (Sharpe et al, 1992; Morriss et a[, 1993); our physiotherapist had less face-to-face contact with the studv patients; treatment started at 75% rather than 40% measured peak oxygen consumption; and our patients exercised for longer every other day rather than a short period each day. No systematic therapeutic attempt was made to change patients' beliefs about their illness or exercise at the start of the programme or during treatment. Dropouts from the trial may have been reduced and adherence to exercise improved if patients were given more specific psychological treatment such as cognitive-behavioural therapy (Sharpe et al, 1996; Deale et al, 1997). It is possible that the maximum effect of graded exercise was not seen in our study which was limited to six months' follow-up since larger treatment effects are reported in studies with 12 months' follow-up (Sharpe et al, 1996; Deale et al, 1997).

Table 4

8

Analysis of the data was hindered by a high drop-wt rate.

8

The follow-up period was restricted to six months.

Treatment of depression was restricted to a singte fixeu aose of fluoxecine.Greater i m p m m e n t may have occurred with variabte doses tailored to the patient's clinical response to the drug.

ALISON J.WEARDEN.MSc. Universityof Manchester.Department of Rychiatry,Withington Hospital. Manchester;RICHARD K. MORRISS. MD, Universityof Manchester Department of Community Rychiatry and Guild NHS Trust, Royal Preston Hospital; RlCKY MUUIS. MSc. DAVID J. PEARSON. MD, Universityof Manchester Department of Medicine.Withington Hospital.Manchester;LOUIS APPLEBY. MD. University of Manchester,Department of Psychiatry,Withington Hospital. Manchester;IAlN T. CAMPBELL. FRCA,University of Manchester,Department of Anae~thesia~Withington Hospital. Manchester;JULIEA. MORRIS, MSc, Universityof Manchester, Department of Medical Statistic~~Withington Hospital. Manchester Correspondence:Dr Richard Morriss.Guild NHS Trust. Royal Preston Hosp~tal.Sharoe Green Lane. Preston PR2 9HT. Fax: 01772 710772 (First received 16 October 1997, final revision 27 January 1998, accepted 27January1998)

Mechanism of action of graded exercise Relatively low functional work capacities in CFS patients at the end of graded exercise treatment reported in our study and by Fulcher & White (1997), and the weak but significant correlation between improved fatigue and functional work capacity in our study, suggest that the

Effects of fluoxetineon Hospital Anxiety and Depressionuse Iml depression

Weeks from baseline

Existing cases at baseline oweeks 26 weeks

New uses arising in trial 26 weeks

Ruoxetine

Placebo drug

x2 (I)

test for trend

tolerance of subjects to exercise had improved rather than a substantial increase in cardiorespiratory conditioning (Allied Dunbar, 1992; McArdle et al, 1994). Graded exercise may work by providing the reassurance to patients that activity (delivered at a controlled rate with a clearly defined maximal level at each time point) need not cause an exacerbation of fatigue symptoms. As a result patients become more confident and willing to attempt other everyday activities.

Clinical implications of the study Our study provides modest support for the Joint Working Group Report's (1996) recommendations for a graded exercise programme to treat fatigue and nonsedating antidepressants to treat depression in chronic fatigue syndrome patients. A considerable amount of support, explanation and encouragement should be given

to patients with chronic fatigue syndrome otherwise graded exercise, antidepressants or other treatments may not be adhered to.

ACKNOWLEDGEMENTS Thls study was funded by a grant from the Llnbury Trust We are grateful toTrudle Roberts for some clln~calassessments. Sue Scoble for pharmacy services and to Ell L~lly(Dlsta) for provldlng a supply of fluoxetlne and placebo

REFERENCES Allied Dunbar (1991) Natlonal Fltness Survey London Sports Council and Health Education Authority American College of Sports Medicine (1995) Gu~delrnesfar Exercrse Testrng and Prescription (5th edn) Baltimore, M D Wllllams and W~lklns Amwican Psychiatric Association (1987) Dragnost~cond Statlstrcal Manual of Mental Drsorders (3rd edn, revlsed) (DSM- 111-R) Washington. D C APA Behm. P. O., Haniffah. B. A. G.. Doogan. D. P,. et a1 (1994) A pllot study of sertrallne for the treatment of Infectrous Drsease. 18 chronlc fatlgue syndrome Cl~n~cal (suppl). SI I I

Borg, G. (1981) Psychophyslcal bas~sof perceived exertlon Medrcal Sclence In Sports and Exercrse. 14. 377-381 Butler. S., Chalder,T.. Ron. M. .t a1 (1991) Cognltlve behavlour therapy ~nchron~cfat~gue syndrome.journal of Neurology. Neurosurgery and Psychlatry. 54. 153- 158 Chlder.T., Berslorritz, G. Pawlikomka.T., .t a1 (1993) Development of a fatigue scale.journal of Psychosomatic Research, 37. 147-153. Deale. A.. Chldcr.1.. Marks. I...t a1 (1997) Cognltlve behav~ourtherapy for chronlc fatlgue syndrome a random~sedcontrolled tr~alAmerrcan Journal of Psychratry. 154, 408414

Morrirr, R. K.. Sharpe, M.. Sharpiey, A. L., .t a1 (1993) Abnormalltles of sleep In patlents wlth chron~c fatlgue syndrome Brltrsh Medrcol]ournal. 306 1161-1164 Sharp, M. C., Archard. L. C., Banatvala, J. E., .t 01 (1991) A report - chronlc fat~guesyndrome gu~dellnes for research journal of the Royal Soclety of Medrcrne 384. 118-121

- ,Hawton, K.,

Seagr0tt.V.. et a1 (1992) Follow up of patlents presenting wlth fatigue to an ~nfecttousdlseases cl~nlcBrrtrsh Medlcol journal 305. 309-315

- ,-,Simkln, S.,

et 01 (1996) Cognltlve behavlour therapy for chron~cfat~guesyndrome a random~sed controlled tr~alBr~trshMedlcol jovrnal. 312 22-26

Fulcher, K.Y. & White, P. D. (1997) Random~sed controlled tr~alof graded exerclse In patlents wlth the chronlc fatlgue syndrome Brltlsh Medrcal Journal. 314. 1647- 1657

Stewart, A. L.. Hays, R. D. & Ware. J. E. (1988) The Medlcal Outcomes Survey Short-Form General Health Survey Rellablllty and valldlty in a patlent population Medrcal Care. 26. 724-734

Joint Working Group of the Royai Cdleges of Physicians. Psychiatrists and General Practitioners (1996) Chronrc Fatrgue Syndrome Council Report CR54 London Royal College of Physlclans

Vwcouien. J. H., Smnink,C. M., Zitmm, F.G., et a1 (1996) Randomlsed double-bl~nd,placebo-controlled study of fluoxetlne In chronlc fatigue syndrome Lancet. 347 858-861

Lewis,G., Pelori, A. J., Araya, R., et a1 (1992) Measuring psych~atrlcdlsorder In the community a standardlsed assessment for use by lay lntervlewers Psychologrcal Medrclne. 22. 465-486

White, P. D. & Chary, K. J. (1997) An open study ofthe efflcacy and adverse effects of moclobemlde In patlents w ~ t hthe chronlc fat~guesyndrome lnternatlonal Clrnrcal Psychophormocology 12.47-52

McArdie,W. D., Katch, F. I.& Katch,V. L. (1994) Essentials of Exercrse Physrology Phlladelphla Lea and Feb~ger

Zigrnond, A. 5 . 1 Sruith, R. P. (1983) The Hosp~tal Anxlety and Depress~onScale (HAD) Acta Psychrotrrca Scondlnanca 67 361-370