A randomised controlled trial (RCT) of telephone delivered cognitive behaviour therapy (TCBT) and exercise in the management of chronic widespread pain (CWP): Identifying long-term outcome and who benefits from which treatment

Gary J. Macfarlane, Marcus Beasley, Philip Keeley, Karina Lovell, Philip Hannaford, Deborah P. M. Symmons, Steve Woby, Gordon J. Prescott, MUSICIAN Study team, Gordon James Prescott

Research output: Contribution to journalAbstractpeer-review

Abstract

Background/Purpose:
CWP is challenging for rheumatologists to
manage and results from long-term epidemiological studies demonstrate
that improvement in symptoms is uncommon. Recent reviews have,
however, suggested that non-pharmacological therapies such as behaviour
therapy and exercise may be effective, at least in the short-term. The aim
of the current study is to determine whether tCBT, exercise, or both
treatments combined deliver long-term health benefits to patients with
CWP in comparison to treatment as usual (TAU) and to characterise
patients who benefit from each specific treatment.
Methods:
A 2x2 factorial RCT. Patients with CWP were identified
from a population screening survey of 45, 994 adults in the UK. Eligible
individuals met the ACR definition of CWP and reported disability. They
had consulted their family physician in the last year with pain, and with no
cause identified which required specific treatment. Participants were
randomly assigned to a) 8 sessions of tCBT over 6 weeks, and refresher
sessions at 3 and 6 months, b) an individually tailored exercise programme
with monthly review over 6 months at a local fitness centre c) a
combination of these treatments, or d) TAU. Participants were followed
up at the end of treatment and 3 and 24 months later. The primary outcome
was self reported “change in health since entering the study” and a
positive outcome was at least 6 (“I felt much better”) on a scale from 0 (“I
feel very much worse”) to 7 (“I feel very much better”). Analysis was
intention-to-treat with longitudinal logistic regression using generalised
estimating equations (GEE). Results are presented as Odds ratios (OR)
with 95% Confidence Intervals (CI). Additional models were run to assess
the effect of baseline characteristics in predicting the response to the
specific treatments received.
Results:
A total of 442 persons (median age 57 years, 69% female)
entered the study and 361 persons (82%) provided information at final
follow-up. At 24 months post-treatment, the proportion of patients
reporting a positive outcome was: tCBT 35.4%, Exercise 29.4%, com-
bined treatment 31.2%, TAU 12.8%. Response, after adjustment for age,
sex, baseline psychological distress, pain intensity and disability was
significantly more likely for exercise (OR 2.5, 95% CI (1.2, 5.4), tCBT
(3.6; 1.7–7.6) and the combination (2.9; 1.4,6.0) compared to TAU.
Baseline characteristics associated with significantly greater response to
tCBT (compared to those not receiving tCBT) were: high psychological
distress, a passive coping style, high intensity and/or disabling pain and
moderate levels of fatigue. Older persons responded significantly better to
the exercise intervention, although this was evident at the end of treatment
but not subsequently.
Conclusion:
A six month programme of exercise or tCBT is associ-
ated with long-term improvements in the health of patients with CWP.
The size of effect was similar with each treatment, and there was no
advantage in subjects receiving both. However we identified specific
characteristics associated with response to tCBT which can potentially
allow future targeting of therapy.
Original languageEnglish
Pages (from-to)S1221
Number of pages1
JournalArthritis & Rheumatism
Volume65
Issue numberS10
DOIs
Publication statusPublished - Oct 2013
Event77th Annual Meeting of the American-College-of-Rheumatology / 48th Annual Meeting of the Association-of-Rheumatology-Health-Professionals - San Diego, Canada
Duration: 25 Oct 201330 Oct 2013

Keywords

  • rct
  • cbt
  • chronic widespread pain
  • exercise
  • pain management

Cite this