| Study characteristics |
| Methods |
Risk of bias:
1. Sequence generation: uncertain ‐ allocated sequentially by blockwise randomisation separately for each centre (but method of sequence generation not specified)
2. Allocation concealment: low ‐ sealed envelopes opened in presence of patient before first session
3. Blinding: uncertain ‐ measures self‐report except for Karnofsky scale which was administered by 'independent clinical psychologist', but exact nature of blinding not specified
4. Incomplete outcome data: high ‐ considerable dropout rate in intervention groups and no reasons provided for dropouts provided; intention‐to‐treat analysis performed
5. Selective outcome reporting: uncertain ‐ no reference to study protocol
6. Other sources of bias: high ‐ no testing for pre‐treatment differences between comparison groups is reported; CBT group conducted singly but guided support conducted in groups; subjects in CBT group forbidden to have further physical investigations but control group free to do so; frequent non‐attendance in guided support group had no consequences for further treatment or dropout status and resulted in significant difference in mean hours of attendance between groups |
| Participants |
Setting: secondary care; specialist outpatient clinic at teaching hospital
Sample population: patients with primary complaint of fatigue referred to specialist outpatient clinic
Sample size: 278
Inclusion criteria: CDC criteria with exception of 4 of 8 additional symptoms being present
Exclusion criteria: greater than 90 minutes travelling time from study centre, previous/current participation in CFS research, pregnancy, fertility treatment, less than 18 years or over 60 years.
Baseline characteristics: demographic characteristics of subjects reported but no statistics analysing pre‐treatment differences between comparison groups on demographic, psychiatric, or outcome variables |
| Interventions |
1: 'cognitive behaviour therapy'; individual; 16 sessions of 60 minutes over 8 months; treatment protocol used; sessions audiotaped and treatment fidelity assessed by independent judge; therapists were psychologists, psychiatrists, health scientists (supervised throughout trial); therapy emphasised control rather than physician/prescription dependenc, challenging fatigue related cognitions; subjects encouraged to attain and maintain base level of physical activity needed to prevent bursts of activity and resultant extreme fatigue; final sessions dealt with relapse prevention and further improvement of self‐control
2: ' guided support group'; approximately 8 patients in each group given non‐directive and client‐centred counselling by social worker; no treatment protocol specified
3: 'natural course'; no other assessments or treatments (subjects free to have other examinations or treatments) |
| Outcomes |
1. MEASURES: number improving significantly on fatigue subscale of CIS, number improving significantly on Karnofsky scale, number improving significantly on self‐rated improvement, CIS, SIP, Karnofsky scale, EuroQol, mean hours working in job per day
2. FOLLOW‐UP TIMES: 8 months (immediately after treatment), 14 months |
| Notes |
Mean duration of fatigue approximately 66 months |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Allocation concealment (selection bias) |
Low risk |
sealed envelopes opened in presence of patient before first session |
