“Catastrophic” pathways to smoking cessation: findings from national survey
BMJ 2006; 332 doi: https://doi.org/10.1136/bmj.38723.573866.AE (Published 23 February 2006) Cite this as: BMJ 2006;332:458
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Section 1 of Table 2: No. of current and ex-smokers by type
of attempt and success status
|
6mo+ |
|
Total |
|
U |
183 |
97 |
280 |
P |
140 |
191 |
331 |
Total |
323 |
288 |
611 |
OR= 2.6
Section 3 of Table 2: No. of current smokers only by type of
attempt and success status
|
6mo+ |
|
Total |
U |
59 |
96 |
155 |
P |
49 |
187 |
236 |
Total |
108 |
283 |
391 |
OR= 2.4
Back-calculated numbers (section 1 minus section 3) for
ex-smokers only
|
6mo+ |
|
Total |
% lasted 6mo+ |
U |
124 |
1 |
125 |
99.3 |
P |
91 |
4 |
95 |
96.2 |
Tot |
215 |
5 |
220 |
97.9 |
OR= 5.2
Salaheddin
Mahmud, MD, MSc
Department of Community Health Sciences,University of Manitoba
S111 – 770 Bannnatyne Avenue
Winnipeg, Manitoba R3E 0W3
Competing interests:
None declared
Competing interests: As professor Prochaska remarked in his Rapid Response theinclusion of ex-smokers in Table 2 does not make sense because by definitionall ex-smokers succeeded in their last attempt, planned or unplanned, in abstainingfor more than 6 months. In other words, by definition every ex-smoker has theoutcome under consideration (abstinence > 6 months). Professor West did notfind this argument convincing. To illustrate the point, I computed a 2-by-2table for the ex-smokers only using data given in Table 2. The first section inTable 2 gives the total number of current smokers and ex-smokers who made aquit attempt between 6 months and 5 years previously. The third section givesthe same figures for current smokers only. So by subtracting the numbers in thethird section from the corresponding numbers in the first section, we obtainthe required 2-by-2 table for the ex-smokers only (see tables below). Out of220 ex-smokers, only 5 (2%) fill in the category “lasted <6 months” withonly one individual in the “unplanned attempt” category. (This is probably dueto a programming error as this number should have been zero.) In any case, andas expected, the data for ex-smokers are not of much use in calculating an oddsratio (OR) for the association with type of attempt. The OR would have beenincalculable if it was not to the five “misclassified” cases.
Dear Editors,
There is another potential bias in "'Catastrophic' pathways to
smoking cessation" that could significantly affect the results of this
study. If a subject made an unplanned attempt at quitting smoking that
lasted for a short time, he may be unlikely to remember this attempt, and
even less likely to identify it as a "serious attempt". In contrast, if he
planned to quit for weeks or months he might be much more likely to
remember, even if he never stopped smoking. This recollection bias can
only be remedied by prospective studies that ask if a subject is currently
engaged in a "serious attempt at quitting smoking".
Competing interests:
None declared
Competing interests: No competing interests
The cross sectional household survey of West and Sohal hypothesizes
that[1]:
1) smokers need to go through the stages of Prochaska and diClemente
before stopping
2) most of the quit-attempts are unplanned
3) unplanned quit-attempts are more successful than planned ones
The results of West and Sohal confirm the hypotheses 2 and 3. However the
suggestion of the authors that smokers need to pass through the stages of
pre-contemplation, contemplation, preparation and action does not hold
true. These stages, as described by Prochaska and diClemente, are since
long valid tools that help us to decide which approach will be effective.
When the patient is a (pre)contemplator the doctor will rather talk about
nicotine replacement, bupropion or nortryptiline than prescribe those
drugs. Prescription may be justified during the stage of preparation or
action. In Dutch general practice 10% of the COPD patients who stop
smoking receive prescriptions[2].
Interestingly, West en Sohal support the relevance of recognizing the
patients’ motivational stage. Moreover, they clearly demonstrate that
doctors might underestimate the effectiveness of their patients with
regard to quit smoking. Patients with planned quit attempts select
themselves for visits compared while catastrophic stop-smokers do less.
As to West and Solal, catastrophic pathway stopping is driven by high
levels of ‘motivational tension’ and subsequent effects of small
‘triggers’. It seems likely that ‘catastrophic’ smoking cessation is
associated with influx of endorphins and other neurotransmitters in the
brain. This implies that a doctor when guiding a stop smoking attempt
sometimes touches the patients’ emotions and must deals with the patients’
mood. Who is afraid of doing so?
References
1) West R Sohal T. "Catastrophic" pathways to smoking cessation: findings
from national survey.
Br Med J 2006;332:458-60
2) Hilberink SR, Jacobs JE, Bottema BJ, de Vries H, Grol RP. Smoking
cessation in patients with COPD in daily general practice (SMOCC): six
months' results. Prev Med 2005;41:822-7
Competing interests:
None declared
Competing interests: No competing interests
Professor West's suggestions that people quit smoking spontenously
without going through the 'stages of change' appears to overlook the
unconscious processes of consideration, that they are likely to have
undergone.
Given the frequent, extensive, highly emotive advertising campaigns
preaching the evils of smoking, some of which border on emotional
blackmail, it is to say the least, highly unlikely that there are any
smokers who have not been affected to a greater or lesser degree. Indeed
research published in the American Journal of Preventive Medicine,(AJPM)
(1)compared the impact of televised advertising campaigns promoting
smoking cessation, with seven conventional types of cessation help,
including NRT.
The sudy showed that the televised ads were by far and away the most
influential in people deciding to quit, thereby confirming what the
Chinese discovered two thousand plus years ago, that if you tell someone
something often enough, they come to believe it.
Since it is almost impossible for us to act in any way but in
accordance with our beliefs, it follows that those who quit went through
the stages of 'precontemplation', 'contemplation', 'preparation' and
action', either immediately following the first advertisement, or more
likely the progress through the stages, came about with the frequency of
the campaigns; whether or not this was a conscious or unconscious process
is irrelevant. It certainly was not spontenous, or an 'impulse buy'.
No one relinquishes a habit or behaviour from which they perceive
they are receiving some benefits, without considering ('contemplating')
compensating factors. Indeed if that were the case we would not have the
horrendous problems of drug and alcohol addiction in our society.
I do not think it is the cycle of change that is flawed, nor does
that wheel need re-inventing.
Peter O'Loughlin,
Addictions Counsellor.
1 Biener,Lois; Reimer, Rebecca L; Wakefield, Melanie; Szczpka;
Rigotti, Nancy and Connelly, Gregory: AMJP: Vol. 30. Issue 3. (March 2006)
Competing interests:
None declared
Competing interests: No competing interests
Work that supports those of us at the “front line” of behaviour
change is always welcome. Simplifying motivational interventions to 3 T’s
is helpful to busy practitioners with a lot to do in a short time.
The theoretical model offered here has face validity but does not extend
the bidirectional nature (described as “tension”) of motivation between
change to the new behaviour or to relapse. Simple extension of this
bidirectional principle to the second 2Ts permits us to consider reducing
Triggers to relapse, for example in legislation to ban smoking in public
places and advising against immediate availability of relapse “Treatment”
by, for example disposing of cigarettes within the home of the quitter.
Furthermore (presumably because of our profession’s unscientific obsession
with cessation in nicotine addiction despite its advocacy for harm
reduction and substitution in other forms of addiction) only two outcomes,
cessation and relapse from cessation are measured and included in this
theoretical model.
A host of comentators has already made negative rapid responses based on
their own points of view (making one's living from anything to do with
smoking cessation is as much a competing interest as owning shares in a
tobacco company!) derived from subjects who probably self selected anyway.
This work has already been misrepresented in the media as implying that
behavioural change interventions are ineffective.
As a GP I find the message both pragmatic and welcome. Despite those
in the cessation "industry" with vested interests in making things
difficult, we now need simple tools for assessing motivational tension, be
supported by public policy that gives positive health & negative
relapse triggers and have better availability and funding of treatment,
particularly the use of long term nicotine substitution (patches) as harm
reduction intervention for secondary prevention in COPD and Cardiovascular
disease.
Competing interests:
None declared
Competing interests: No competing interests
I read with great interest the article on smoking cessation by West
and Sohal [1]. In Japan, health insurance coverage is decided to extend to
the treatment of nicotine addiction from April 2006. Tobacco smoking has
now come to be recognized as a disease that requires treatment at least in
part. The Ministry of Health, Labor and Welfare plans to cover doctors'
smoking cessation counselling fees with public health insurance schemes,
expecting to stem the nation's ballooning medical costs. Patients who want
to quit immediately must agree to being enrolled in a 12-week smoking
treatment program. The result of their study that unplanned quit attempts
were more likely than planned ones to be successful is encouraging.
Reference
[1]West R, Sohal T. "Catastrophic" pathways to smoking cessation: findings
from national survey. BMJ 2006;332:458-460.
Competing interests:
None declared
Competing interests: No competing interests
Prochaska's commentary assumes that the article makes a claim which
it explicitly does not do. The findings cannot possibly be used to support
the argument to plan a quit attempt is a bad idea and we explicitly say so
in the article. The results can be interpreted in a number of ways but the
hypothesis that we put forward was that sudden quits reflect a greater
unconflicted commitment on the part of many of those making them than ones
that are planned ahead. What is challenging to the stages of change
approach is that sudden quits are so common and that so many of them
succeed - the observation that they are statistically more likely to be
successful than planned ones is a pointer to future research regarding the
true processes by which this kind of change occurs.
As regards the interpretatation of the data by Prochaska, this is
confused. Of course it makes sense to include ex-smokers in calculating
the percentage of quit attempts that succeeded - the point is that the
denominator also includes people who did not succeed and the ratio of the
two is what is of interest. Secondly, the analysis involving smokers who
relapsed was included to demonstrate a particular point as stated in the
paper and of course lasting longer than 6 months, even if there is
subsequent relapse, reflects greater success than lasting less than 6
months. Finally, there is no doubt that quit attempts are forgotten and we
acknowledge this which is why we showed that the difference between
planned and unplanned quit attempts was robust to examination over
different time periods.
Competing interests:
As declared in the original artical
Competing interests: No competing interests
West and Sahol’s (2006) recent research addresses a challenging
question. Does planning a quit attempt increase the chances that the quit
attempt will last at least 6 months? The results of their retrospective
research with a sample of British smokers leads them to conclude that
planning quit attempts decreases the chances.
The authors suggest that their results are especially challenging for
the stages of change model. The fact is, if taken at face value, their
results would be particularly challenging for almost all evidence-based
treatments for smoking cessation. From their measures it would appear
that one of the best indicators of planning would be setting a quit date.
Helping participants set a quit date is an important part of most
cessation approaches, especially those designed for motivated smokers who
are prepared to quit in the next month. Given the results of their
research, it would seem that the authors would have to disclose to smokers
that setting a quit date would be likely to more than double the chances
that their quit attempt would fail to last at least six months.
It might surprise the authors to learn that in our TTM based
population cessation approach we do not try to get most participants to
set a quit date during treatment because they are not ready.
The authors recognize the conundrum their results seem to create when
they write: “These findings do not necessarily imply that planning quit
attempts is counterproductive, and use of behavioral support and nicotine
replacement therapy are known to improve the chances of success even
though they generally require planning ahead?” But the authors provide no
evidence or argument against the implications that planning quit attempts
is counterproductive. The researchers and the readers are caught between
hundreds of prospective RCT’s that strongly suggest that planning quit
attempts is productive and this single retrospective study that suggests
otherwise.
Let’s take a closer look at this study to see why their results
shouldn’t be taken too seriously. First of all it makes no sense to
include ex-smokers in Table 2, since by definition 100% of planned and
unplanned attempts of ex-smokers would have to be successful and to have
lasted at least six months. Second, amongst current smokers 100% of the
planned and unplanned quit attempts failed, teaching us nothing about the
long-term efficacy of each approach. Third, the data presented indicate
that the percentage of people remembering planned quit attempts ranges
from about half in Table 1 that includes quit attempts over 5 years to
about two-thirds from those having quit in 6 to 12 months in Table 2.
Finally, based on “catastrophe theory” the authors recommend that
with smokers who are on the cusp of change in their orientation to
smoking, treatment needs to be made available immediately. Remember that
planning for later the same day, the next day or in a few days would hurt
their quit attempt. Are we going to have smokers go immediately to the
emergency room to seek cessation treatment? That would be a catastrophe!
West keeps insisting that he has a solution that is superior to the
stage of change model. This article suggests that he needs to keep
searching.
Competing interests:
None declared
Competing interests: No competing interests
There seems to be a misapprehension creeping into some responses
here, that because something commonly happens in a particular way 'in the
real world' then this should be considered the best way. This is flawed
reasoning.
While it is true that most smokers who quit do so without any
specific behavioural support or pharmacological treatement, this does not
preclude the possibility that they would have quit in much greater numbers
had they received such support and treatment. Indeed, there is a wealth of
evidence from published studies to show that supported quit attempts are
much more successful than unsupported ones.
The fact that most quitters have had to manage it alone is cause for
concern not celebration.
Competing interests:
Founder of the NHS stop-smoking service ‘Help 2 Quit’ which treats 6,000 smokers per year. Have occasionally provided consultancy, and undertaken research within the NHS, for companies that develop smoking cessation treatments.
Competing interests: No competing interests
Re: Possible (non-catastrophic) errors in Table 2
I am sorry to say that the analysis by this author is nonsense. It
should be apparent from the definition of the denominator (participants
who made their most recent quit attempt between 6 months and 5 years ago)
that there cannot be ANY participants who are ex-smokers who lasted less
than 6 months! The fact that there were 5 ex-smokers who reported their
most recent quit attempt as being at least 6 months ago and it lasting
less than 6 months is just a reflection of the fact that this in the this
small number of cases they made mutually inconsistent responses to
different items on the questionnaire.
I can only repeat the point I made to Prochaska in the hope that this
author and Prochaska will be able to understand it. The denominator for
the calculation has got nothing to do with whether participants are
smokers or ex-smokers - it is simply the number of people who made their
most recent quit attempt at least 6 months ago. Whether these are planned
or unplanned is similarly defined completely independently of whether the
participants are smokers or ex-smokers. So the only point where the fact
that they are ex-smokers comes in is in the OUTCOME. Therefore there can
be no tautology here.
There are many other potential sources of bias in this kind of
retrospective data and so it will be important to check these findings
with prospective studies.
Competing interests:
As stated in the original article
Competing interests: No competing interests