Basic HTML Version
Addictive Behaviors,
Vol. 11, pp.
355-365, 1986
Printed in the USA. All rights reserved.
0306-4603/86 $3.00 + .OO
Copyright 0 1986 Pergamon Journals Ltd
TREATMENT EFFECTIVENESS OF HYPNOSIS AND
BEHAVIOUR THERAPY IN SMOKING CESSATION:
A METHODOLOGICAL REFINEMENT
G.J. HYMAN
Chisholm Institute of Technology, Caulfield East, Australia
R.O. STANLEY, G.D. BURROWS, and D. J. HORNE
University of Melbourne
Abstract--Studies in smoking cessation have generally failed to adequately control for active
treatment effects and have assumed that measures of smoking behaviour (i.e., estimated smok-
ing rate, self-monitoring and chemical analysis) are equally reliable measures. Sixty smokers
were randomly assigned to one of four different smoking cessation treatment groups: hyp-
nosis, focussed smoking, attention placebo and a waiting list control. Subjects were asked to
estimate and monitor their own smoking behaviour.
Bloodsampleswerealso taken for thio-
cyanateanalysisbeforetreatment.Smoking ratesweresimilarlymeasureddirectly,at 3months
and 6 monthsafter treatment.The resultsindicatethat the three measuresof smokingbe-
haviourwereall highlycorrelated.No significantdifferenceswerefoundbetweentreatments,
directlyafter treatmentor at the 3-and 6-monthfollow-ups.
These results suggest that active
treatment effects may not be responsible for behavioural change in a smoking cessation pro-
gram. The implications of these findings are discussed.
For all “problem” behaviours, the basis of behavioural change in the clinical setting is
assumed to be the treatment programme. Smoking cessation is one clinical problem in
which success has been held to be determined by the specific nature of the treatment ap-
proach. It has been over 20 years since the drive to find an effective smoking cessation
treatment began. Disagreements are still evident and no single approach has yet emerged
as outstandingly effective in modifying smoking behaviour (Raw, 1978).
Two specific influences appear responsible for the present situation. Firstly, studies
have generally adopted the approach of comparing different specific treatments in
order to find the one that is the most effective. Inherent in this strategy is the idea that
treatment will be more effective than no treatment at all. This assumption has been
supported by the research evidence (Bernstein & McAlister, 1976; Russell, Armstrong,
& Patel, 1976).
Also implicit is the belief that it is the active rather than coincidental elements of the
treatment procedure that contribute to the successful cessation of smoking. Given the
research to date, this assumption cannot yet be justified. It is possible that nonspe-
cific effects, rather than active treatment effects, could be responsible for behavioural
change. Previous research has largely ignored the experimental control of this area. For
this reason a number of authors (Bernstein, 1969; Raw, 1976) have recommended the
use of attention placebo groups (i.e., subjects receive all the procedures of the treat-
ment groups except for the active treatment component). In this way the active treat-
ment effects may be discriminated from the nonspecific factors, such as expectations
and faith in the clinician.
A second influence that may contribute to the current equivocal situation in smoking
Requests for reprints should be addressed to G.J. Hayman, Department of Applied Psychology, Chisholm
Institute of Technology, Caulfield East, 3145, Australia.
355