have lower educational attainment,

have lower educational attainment, have a partner that smoked,
work in manual occupations, and be younger (see table at http://
www.pcpoh.bham.ac.uk/primarycare/research/ttm/index.htm).
Results
Participants in the TTM arms were slightly but not significantly
more likely to make positive stage changes in the TTM arms than
the control arm when those whose stage was missing were
included as having made no positive stage change (Table 3).
However, when these people were excluded, the effect of the TTM
armwas slightly greater, though thiswas not statistically significant
with or without adjustment for potential confounders. The effect
size was small, with an increase in odds of 20% or so. The mean
change score in the controls was 0.39 and in the TTM was 0.46,
a difference (95% CI) of 0.08 (0.00–0.16), which was not statistically
significant (p ¼0.064). On adjustment for confounders, this difference
was reduced slightly to 0.07 (0.01 to 0.15). When those with
missing data were excluded, the unadjusted and adjusted mean
differences (95% CIs) were 0.12 (0.01 to 0.24) and 0.10 (0.03 to
0.22).
The second hypothesis was that the relative benefit of the
TTM-based materials would be greater for participants in precontemplation
or contemplation than for participants in preparation.
There was little evidence to support this hypothesis. The
change score difference between control and TTM was very slightly
but not significantly larger in precontemplation/contemplation
than in preparation (Table 4). However, the likelihood of making
a positive change if experiencing the TTM intervention rather than
the control one was slightly though not significantly greater for
those in preparation rather than in precontemplation/contemplation,
contrary to the hypothesis (Table 5). The likelihood of participants
stopping smoking in the TTM arms relative to the control
arm was greater for participants in precontemplation/contemplation
than for participants in preparation for most of the smoking
outcomes examined, although this difference in relative effect was
not statistically significant (Table 5). The effect size difference
between TTM and control was generally similar.
Discussion
Martin et al. (1996) propose that stage movement is a better
outcome to assess the effectiveness of smoking cessation interventions
because, among other reasons, it is more common than
stopping smoking and trials would have more power to detect an
effect. In this study, there were very small and not statistically
significant advantages for the TTM arm in the quitting outcomes
(Aveyard et al., 2003), and it was plausible that the benefit was
obscured because quitting was rare and the analysis lacked power.
These data show no advantage for the TTM-based intervention in
stage movement, implying that the negative results on stopping
smoking were not due to lack of power. Thus, the TTM-based stagematched
intervention was not more effective than the nonstage-
based control intervention. Given the evidence that one
stage-based intervention is as effective as six (Velicer et al., 1999),
the fairly low rates of second and third intervention delivery in the
trial are probably not the cause of this lack of effect.
The stage-based interventions should be particularly effective
for people in precontemplation and contemplation because standard
smoking cessation interventions are aimed towards people
who want to stop smoking soon (Prochaska & Velicer, 1997). Thus,
for participants in preparation, the standard materials are stagematched
and the advantage of stage matching should be slight in
preparation. This was not the case. In general, the effect size for the
TTM-based intervention was similar in precontemplation/
contemplation as in preparation and there was no evidence of
a statistically significant difference in relative effect. Thus, the
second hypothesis was also not confirmed.
Prochaska and Velicer (1997) wrote that current standard selfhelp
materials are geared towards people who are ready to quit.
The second hypothesis was based on this assertion, which has not
been tested in the TTM literature, but is widely accepted by
proponents and critics of the TTM. However, our content analysis
showed that this study, and perhaps by implication other similar
studies, that examine TTM-based materials versus standard
materials may not be testing the match/mismatch hypothesis
because standard materials may actually urge readers to use the
processes of change appropriate for their stage even if they are in
precontemplation or contemplation. However, in our control
intervention, the main process of change used for the earlier
stages was consciousness raising, with very little use of the other
processes of change. This was the most prominent process of
change for the TTM interventions, but the manual showed more
balanced use of the other processes of change for precontemplation
and contemplation. The TTM is not specific about
whether all appropriate processes of change must be used to
move stage, or whether under-use of one process of change can
be compensated by over-use of others, so whether this is theoretically
an advantage or not is unclear. Furthermore, in the TTMbased
intervention, participants were staged by the opening
chapter and then directed to the chapter appropriate for them.
The expert letter also instructed the participant to read the
appropriate chapter. This means that TTM arm participants were
directed to stage-appropriate processes, while those in the control
arm were not. Prochaska and Velicer (1997: 43) state that