Outcomes and predictors of response from an optimised, multidisciplinary intervention for chronic fatigue states.

BACKGROUND: Medically unexplained chronic fatigue states are prevalent and challenging to manage. Cognitive behavioural therapy (CBT) and graded exercise therapy (GET) are effective in clinical trials. The evaluation of delivery in a standard healthcare setting is rare. An integrated treatment programme with individualised allocation of resources to patients' needs was developed and implemented through an academic outpatient clinic. It was hypothesised that the programme would result in similar responses to those observed in the clinical trials.

AIM: To evaluate the outcomes of an integrated, 12-week CBT and GET programme delivered by exercise physiologists and clinical psychologists.

METHODS: Consecutive eligible patients (n = 264) who met the diagnostic criteria for chronic fatigue syndrome or post-cancer fatigue were evaluated with self-report measures of fatigue, functional capacity and mood disturbance at baseline, end-of-treatment (12 weeks) and follow-up (24 weeks). A semi-structured interview recording the same parameters was conducted pre- and post-treatment by an independent clinician. Primary outcome was analysed by repeated measures analysis of variance and predictors of response were analysed by logistic regression.

RESULTS: The intervention produced sustained improvements in symptom severity and functional capacity. A substantial minority of patients (35%) gained significant improvement, with male gender and higher pain scores at baseline predicting non-response. A small minority of patients (3%) worsened.

CONCLUSION: The manualised protocol of integrated CBT and GET was successfully implemented, confirming the generally positive findings of clinical trials. Assessment and treatment protocols are available for dissemination to allow standardised management. The beneficial effects described here provide the basis for ongoing studies to optimise the intervention further and better identify those most likely to respond.