Article review Article Review Instructions You will be required to submit a paper analyzing an article provided in the Reading & Study folder of Module/Week 6. The article is a collection of brief commentaries about the Stages of Change model (also known as the Transtheoretical Model). Your paper should be 2–4 pages, double spaced (not including the title page and reference page). This assignment is due by 11:59 p.m. (ET) on Monday of Module/Week 6. After reading the article, write a synopsis that includes the following 4 components. Each section should begin with the heading that is provided below in bold type: 1. Thesis Outline the main thesis, objective, or “opinion” of the article. 2. Rationale Select at least 2 authors from the article and provide an explanation of their perceptions of the Stages of Change model. You must also provide supporting rationale to explain the authors’ perceptions. 3. Response Provide a clear explanation of your response to the commentaries. (Do not just agree or disagree. Please state why you feel specific findings were or were not legitimate.) 4. Strengths Outline 2–3 strengths of the model. You must also reference a professional journal article which support these findings and observations.
WHEN POPULARITY OUTSTRIPS THE
EVIDENCE: COMMENT ON WEST (2005)
Robert West (2005) has taken the bold step of asserting
that the Transtheoretical Model (TTM) is so flawed that it
should be discarded. Whether one agrees with West’s
conclusion or not, his editorial should stimulate longoverdue debate about the TTM.
A sharp divide of opinion about the TTM has surfaced
in recent years. On one side the model enjoys substantial
popularity in the form of a voluminous research literature and a large following among clinicians. On the other
side there is discontent among many scholars (e.g. Sutton
2001) who have closely scrutinized the scientific rigor of
the model. There can be no questioning the popular success of the TTM: it is established fact. But the scientific
merit of the model can be questioned, and West has furthered the debate with his provocative editorial. Consistent with West, this commentary will focus on smoking
as the model addiction because smoking has been the primary focus of the TTM.
The TTM became popular because it brought attention to the intuitive notion that some smokers are more
ready to quit than others. For this, the originators of the
TTM deserve credit, though as West and others (e.g.
Bandura 1998) have pointed out, the observation that
some smokers are riper for change than others is a confirmation of the obvious. As a scientific model, however,
the TTM got off to an inauspicious start. The lynchpin of
the TTM is, of course, the stages of change. Thus it would
seem essential to take great care in formulating how the
stages were to be conceptualized and measured. However,
there has never been a peer-reviewed account of the
developmental research that led to the creation of the
stages of change algorithm. In fact, it is not clear that any
systematic developmental research took place at all, and
the consequences of this omission plague the model to the
present time. Instead, the authors of the TTM essentially
decreed that readiness to quit smoking should be measured using the stages of change. For the most part, the
addictions research community then adopted the stages
of change, with few questions asked.
There would be little consequence to the omission of
developmental data on the stages of change if the stages
had subsequently been proved to be valid and effective.
However, as West (2005) and others (e.g. Etter & Sutton
2002) point out, this is not the case. The stage of change
algorithm is a magpie collection of questionnaire items
that do not cohere particularly well. Two of the questionnaire items rely on arbitrary timeframes and binary yesno response options. One need not be an expert in
questionnaire development to detect potential problems
with this instrument.
After a time the stages of change became something
analogous to a ‘brand name.’ Virtually all addictions
researchers became familiar with the stages of change
model, and it became common to include the stages as a
basic sample characteristic in studies. A pattern emerged
in the TTM literature whereby success was declared on
behalf of the TTM regardless of research outcomes. Exuberant interpretations of modest results became commonplace. Grandiose conclusions were extrapolated from
unremarkable findings. The popularity of the TTM had
come to outstrip the scientific evidence.
The popularity of the TTM came at the cost of reduced
scientific and clinical progress. Alternative models of
motivation to quit smoking were not pursued because the
TTM had ‘cornered the market’ on the topic. Research on
tailored interventions for smoking cessation became
dominated by the TTM, which diverted resources and
attention that could have been devoted to more promising methods of tailoring.
Twenty years after the introduction of the TTM, West
(2005) has concluded that the TTM should be discarded,
and this commentator concurs. However, the larger topic
of motivation to quit should not be abandoned. Rather,
researchers should renew efforts to understand and measure motivation to quit smoking. West has done precisely
this by introducing his new model of behavior change.
There’s much to be learned from the case of the TTM.
© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050
First and foremost, researchers should insist that scientific models are judged according to the standards of scientific merit and not by popular trends or intuitive
THADDEUS A. HERZOG
Tobacco Research and Intervention Program
H. Lee Moffitt Cancer Center and Research Institute
4115 East Fowler Avenue,
E-mail: [email protected]
Bandura, A. (1998) Health promotion from the perspective of
social cognitive theory. Psychology and Health, 13, 623–649.
Etter, J.-F. & Sutton, S. R. (2002) Assessing ‘stage of change’ in
current and former smokers. Addiction, 97, 1171–1182.
Sutton, S. R. (2001) Back to the drawing board? A review of
applications of the transtheoretical model to substance use.
Addiction, 96, 175–186.
West, R. (2005) Time for a change: putting the Transtheoretical
(Stages of Change) Model to rest. Addiction, 100, 1036–1039.
Blackwell Science, LtdOxford, UKADDAddiction0965-2140© 2005 Society for the Study of Addiction
THEORETICAL TOOLS FOR THE
INDUSTRIAL ERA IN SMOKING
CESSATION COUNSELLING: A
COMMENT ON WEST (2005)
With over 1400 citations (according to http://www.
scopus.com [accessed 12 April 2005]), Prochaska’s
paper summarizing the transtheoretical model is one of
the most widely cited papers in the psychological literature (Prochaska, DiClemente & Norcross 1992). Astonishingly, this success was achieved in spite of the early
recognition of major problems affecting this model
The concept of ‘stage of change’ is a haphazard mixture of current behaviour, intention to change, past quit
attempts and duration of abstinence. As West (2005)
points out, stages are defined by setting arbitrary cutpoints on continuous variables (time and intention) and,
contrary to what is often believed, this theory says nothing about the time people spend in the first three stages.
Furthermore, there is little empirical evidence of progression through the entire stage sequence (Littell & Girvin
2002). It is therefore hard to believe that these stages
reflect reality. It makes little sense to classify in the same
category (e.g. precontemplation) people with different
levels of dependence, or people who have never tried to
quit with those who achieved long periods of abstinence.
Similarly, using only abstinence criteria to define Action
is too reductive, and neglects important steps that people
take on their way towards abstinence (e.g. cutting down,
non-daily smoking). Finally, the model does not take into
account dependence level, withdrawal symptoms and
other key determinants of smoking, in particular environmental and social factors.
The core of the transtheoretical model is a description
of associations between variables, in particular between
stages and self-change strategies that are supposed to be
used in a sequential pattern, ‘doing the right thing at the
right time’ (Segan et al. 2004). However, it has never been
convincingly shown that distinct strategies are needed to
progress across distinct stages (Herzog et al. 1999; Segan
et al. 2004). In fact, stage mismatched interventions,
where all smokers, including precontemplators, receive
action-oriented advice may be as effective (Dijkstra et al.
1998) or even more effective than stage matched interventions (Quinlan & McCaul 2000), which seriously
questions the basic tenets of this model.
This model has been widely used to guide interventions and determine who gets what treatment. However,
interventions based on this model have not been consistently proven to be more effective than control interventions or than no intervention (Riemsma et al. 2003; van
Sluijs et al. 2004). Worse, labelling people ‘precontemplators’ (an awkward jargon) is stigmatizing and may lead
clinicians to deprive patients of effective treatments. The
risk of excluding precontemplators from effective treatments is a major liability of this model.
Valid measurement is the foundation of good science,
but measurement of the model’s constructs is problematic. There are many, incompatible ways of measuring
stages of change, and questionnaires measuring the
model’s other core constructs (processes of change, selfefficacy and ‘pros and cons’) were published 15–20 years
ago and have not been revised since. A constant development and adjustment of scales is nevertheless required to
achieve the best possible measurement in each population subgroup, and theory should be subsequently developed according to empirical findings. But the
transtheoretical model is far too rigid, it has not evolved
much in the past two decades, and proponents of the
model have been reluctant to take into account external
criticism and to develop their theory accordingly. Rather,
one has the impression that the model is often used rigidly, almost religiously.
There is however, a need for an integrative, comprehensive theory on which interventions can be based.
Such a theory is needed in particular for computertailored programs, which can reach huge numbers of
smokers over the internet. For many smokers, the internet is the only source of information and counselling.
One-to-one counselling is not cost effective on the internet, but computer-tailored programs can provide effec-
© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050
tive, individually tailored advice to large audiences (Etter
2005). In fact, the transition from one-to-one counselling in clinical settings and telephone helplines to mass
level, individually tailored counselling on the internet is
comparable to the industrial revolution, when craftsmen
working in small shops were replaced by huge plants
managed by engineers. To successfully handle this transition, the field needs a theory that tells us what to measure, how to measure it, what type of advice should be
given to each category of smokers, when to give each recommendation, and what outcome should be expected in
each subgroup. Interventions should be explicitly derivable from this new theory, in contrast with the loose and
questionable links between the transtheoretical model
and interventions. Researchers should join in a collaborative effort to develop a theory that reflects reality better
than the transtheoretical model, and to assess whether
interventions based on this theory are more effectivethan existing interventions.JEAN-FRANÇOIS ETTERIMSP-CMU1 rue Michel-ServetGeneva 4CH1211SwitzerlandE-mail: [email protected], R. (1992) Prochaska and DiClemente’s model ofchange: a case study? British Journal of Addiction, 87, 821–822.Dijkstra, A., De Vries, H., Roijackers, J. & van Breukelen, G.(1998) Tailored interventions to communicate stagematched information to smokers in different motivationalstages. Journal of Consult Clinical Psychology, 66, 549–557.Etter, J. F. (2005) Comparing the efficacy of two internet-based,computer-tailored smoking cessation programs: a randomized trial. Journal of Medical Internet Research, 7, e2.Herzog, T. A., Abrams, D. B., Emmons, K. M., Linnan, L. A. &Shadel, W. G. (1999) Do processes of change predict smokingstage movements? A prospective analysis of the transtheoretical model. Health Psychology, 18, 369–375.Littell, J. H. & Girvin, H. (2002) Stages of change. A critique.Behavior Modification, 26, 223–273.Prochaska, J. O., DiClemente, C. C. & Norcross, J. C. (1992) Insearch of how people change. Applications to addictive behaviors. American Psychology, 47, 1102–1114.Quinlan, K. B. & McCaul, K. D. (2000) Matched and mismatchedinterventions with young adult smokers: testing a stage theory. Health Psychology, 19, 165–171.Riemsma, R. P., Pattenden, J., Bridle, C., Sowden, A. J., Mather,L., Watt, I. S. et al. (2003) Systematic review of the effectiveness of stage based interventions to promote smoking cessation. BMJ, 326, 1175–1177.Segan, C. J., Borland, R. & Greenwood, K. M. (2004) What is theright thing at the right time? Interactions between stages andprocesses of change among smokers who make a quit attempt.Health Psychology, 23, 86–93.van Sluijs, E. M., van Poppel, M. N. & Mechelen. W. (2004)Stage-based lifestyle interventions in primary care: are theyeffective? American Journal of Preventive Medicine, 26, 330–343.West, R. (2005) Time for a change: putting the Transtheoretical (Stages of Change) Model to rest. Addiction, 100, 1036–1039.Blackwell Science, LtdOxford100Original ArticleCommentaryCommentaryCommentaryWEIGHING THE PROS AND CONS OFCHANGING CHANGE MODELS: ACOMMENT ON WEST (2005)West (2005) challenges researchers and clinicians toabandon the transtheoretical model (TTM) and to revertto common sense ideas about motivation for change or todevelop new models that better account for the complexities of the change process. Part of West’s argument isbased upon the continuing popularity of the modeldespite the healthy debate about its theoretical and empirical shortcomings that has occurred in this journal andothers. As Whitelaw et al. (2000) point out, the need forcritique is the greatest at the point that an idea becomes‘accepted’ and the TTM has gained this accepted status.An example of this acceptance is the influence of themodel on service delivery. There is little empirical evidence that TTM stage-based interventions lead to superioroutcomes over non-stage based interventions (Riemsmaet al. 2003; van Sluijs, van Poppel & van Mechelen 2004;Adams & White 2005). Nonetheless, and despite ourpledge toward evidence-based practise, many jurisdictions are developing and implementing stage based interventions in a variety of problem areas despite the lack of astrong evidence-base. Such interventions have the falseappearance of being evidence-based because they arebased upon scientific models although, in reality, theappeal of TTM for this purpose appears intuitive.A related unfortunate effect of the wide acceptance ofthe model’s validity is the potential that decisions aboutwho gets what type of service are made on stage ofchange assessments (Piper & Brown 1998). Peopleassessed as precontemplators might be excluded from service or offered less action-oriented interventions thanthose deemed to be more ready to change. The evidence,at least in the area of smoking cessation, is inconsistentwith this stance. In a recent study, smokers, regardless oftheir stage of change, were offered a smoking cessationgroup (Pisinger et al. 2005). Only 16% of those who wereultimately successful had serious intention to quit priorto the intervention. The authors argue that cessationsupport needs to be offered to all smokers regardless oftheir stage of change. Clearly, intentions or readiness arehighly fluid (Hughes et al. 2005).Commentaries 1043© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050For these ethical reasons, I agree with one aspect ofWest’s argument. A moratorium on the uncritical useof the TTM model in clinical service delivery is warranted. However, I disagree that the flaws of the modelare so large that it should be entirely abandoned. Thepurpose of model building is to stimulate testablehypotheses that will lead to enhancements of ourunderstanding of complex phenomena. The sheer bulkof the research conducted on the TTM is evidence thatit provides such a focus. The model can continue todirect interesting research queries that presumably willlead to model refinement. For example, one facet of themodel that has empirical support is the predictive validity of the stages—people who are further along the continuum are more likely to have changed their behaviourat a future follow-up point than those who are at anearlier stage. It is reasonable to hypothesize that therapeutic efforts that result in forward stage movement willresult in a greater likelihood of future change thaninterventions that do not result in forward movement.This feature of the TTM accounts for part of its appealto clinicians working with addictive behaviours—if Ican’t get action from my clients then at least I canimprove my clients’ readiness for change . . . Alas, thishypothesis has not been empirically tested. We don’tknow whether these ‘soft outcomes’, as West describesthem, are ultimately helpful for our clients (perhaps clients shift back to their initial stage post intervention) orwhether these outcomes simply help the therapist copewith the limited success we have getting actual behaviour change from our clients. Surely researchers need toattend to this gap in our understanding before the TTMis abandoned.Another interesting but understudied clinical development has been the use of the model as therapeutic toolwith clients. Littell & Girvin (2002) note that the model isbased upon a rational actor assumption of behaviourchange, the notion that change is based upon a rationalcognitive self-examination by the individual. This featureis undoubtedly another part of the model’s appeal to clinicians. Many of us have started to provide our clientswith a description of the stages of change portion of themodel with the request that clients determine where theyfit. Perhaps this process of education and self-staging provides a helpful change schema for our clients that helpsorganize their ambivalent thoughts about and actionstoward change. West argues that self-labeling is animportant aspect of maintaining behavioural change.Perhaps, a cognitive understanding of change as a process is important in initiating the change.In short, the model continues to be an important stimulus to theory and practise development and it will ultimately be usurped by reformulated models. I look forwardto learning more about West’s alternative model (West, inpress), a model that he admits has been inspired by thecurrent body of research and critique of the TTM.DAVID C. HODGINSDepartment of PsychologyProgram in Clinical PsychologyUniversity of Calgary2500 University Drive NWCalgaryAlbertaCanada T2N 1N4E-mail: [email protected], J. & White, M. (2005) Why don’t stage-based activitypromotion interventions work? Health Education Research, 20,237–243.Hughes, J., Keeley, J. P., Fagerstom, K. O. & Callas, P. W. (2005)Intentions to quit smoking change over short periods of time.Addictive Behavior, 30, 653–662.Littell, J. H. & Girvin, H. (2002) Stages of change. A critique.Behavior Modification, 26, 223–273.Piper, S. & Brown, P. (1998) Psychology as a theoretical foundation for health education in nursing: Empowerment orsocial control? Nurse Education Today, 18, 637–641.Pisinger, C., Vestbo, J., Borch-Johnsen, K. & Jorgensen, T. (2005)It is possible to help smokers in early motivational stages toquit. The Inter99 study. Preventative Medicine, 40, 278–284.Riemsma, R. P., Pattenden, J., Bridle, C., Sowden, A. J., Mather,L., Watt, I. S. et al. (2003) Systematic review of the effectiveness of stage based interventions to promote smoking cessation. BMJ, 326, 1175–1177.van Sluijs, E. M., van Poppel, M. N. & van Mechelen, W. (2004)Stage-based lifestyle interventions in primary care: are theyeffective? American Journal of Preventive Medicine, 26, 330–343.West, R. (2005) Time for a change: putting the Transtheoretical(Stages of Change) Model to rest. Addiction, 100, 1036–1039.West, R. (in press) Theory of Addiction. Oxford: BlackwellPublishing.Whitelaw, S. S., Baldwin, S., Bunton, R. & Flynn, D. (2000) Thestatus of evidence and outcomes in Stages of Change research.Health Education Research, 15, 707–718.Blackwell Science, LtdOxford, UKADD100Original ArticleCommentaryCommentaryCommentaryANOTHER NAIL IN THE COFFIN OF THETRANSTHEORETICAL MODEL? ACOMMENT ON WEST (2005)In a series of publications (Sutton 1996; 2000a, 2000b,2001, 2005), I have critically examined the transtheoretical model (TTM) and its associated assessment instruments and evaluated the evidence for the model. I cameto the conclusion that the TTM cannot be recommendedin its present form and that we need to go ‘back to thedrawing board’ (Sutton 2001). Thus, I reached a similar© 2005 Society for the Study of Addiction Addiction, 100, 1040–10501044 Commentariesconclusion to West (2005) but by a somewhat differentroute.Although I endorse West’s conclusion, I disagree tosome extent with his analysis of what the Rhode Islandgroup calls stage effects (Prochaska et al. 2004) and withwhat he says about intervention studies.STAGE EFFECTSA stage effect is observed when initial pre-action stage ofchange predicts being in action or maintenance at followup: those in the preparation stage at baseline are morelikely to be in action or maintenance at follow-up thanthose in contemplation, and those in contemplation atbaseline are more likely to be in action or maintenance atfollow-up than those in precontemplation. We need ameta-analysis to quantify these effects, but my reading ofthe literature on the TTM is that stage effects are a highlyconsistent finding. Of course, this is not surprising,because the pre-action stages are defined in terms ofintentions and past behaviour, and there is ample evidence that these predict future behaviour. Stage effectsmean that stage measures may be of practical value, forexample in measuring progress towards smoking cessation. This may or may not be ‘common sense’, but it is nottrivial. However, as West points out, other measures maydo better.It is important to appreciate that stage effects do notnecessarily provide strong evidence for a stage modelbecause ‘pseudostage’ models may yield similar effects.For example, continuous measures of intention predictfuture behaviour and if such an intention measure werecategorized into, say, three categories, one would expectto find a (pseudo)stage effect. Stage models make specificpredictions about the probabilities of different stage transitions that can be tested using longitudinal data (Weinstein et al. 1998; Sutton 2000a).INTERVENTIONSWest states that ‘Where interventions have been developed that are based on the model these have not provedmore effective than interventions which are based on traditional concepts’. Unfortunately, the systematic reviewsof stage-based interventions that have been published todate (e.g. Riemsma et al. 2003; van Sluijs et al. 2004)have included studies that were not proper applications ofthe TTM. For an intervention to be labelled as TTM-based,it should (1) stratify participants by stage and (2) targetthe model’s ‘independent variables’ (pros and cons, confidence and temptation, and processes of change), focusing on different variables at different stages. There is aneed for more focused reviews of TTM-based interventionstudies.The interventions that come closest to a strict application of the TTM are those developed by the Rhode Islandgroup. The group’s studies of smoking cessation interventions (e.g. Prochaska et al. 1993, 2001a, 2001b; Pallonen et al. 1998)—none of which were cited in West’s(2005) editorial—have yielded mainly positive findings.However, adaptations of these interventions evaluated byresearch groups in the UK and Australia have yieldedmainly negative results (Aveyard et al. 1999, 2001,2003; Borland et al. 2003; Lawrence et al. 2003).Process analyses showing that TTM-based interventions do indeed influence the variables they target in particular stages and that forward stage movement can beexplained by these variables have not been published todate. There have also been few experimental studies ofmatched and mismatched interventions, which couldpotentially provide the strongest evidence for or againstthe model (Weinstein et al. 1998; Sutton 2005).West states that ‘the model is likely to lead to effectiveinterventions not being offered to people who wouldhave responded’. This consequence would not be in thespirit of the model. The TTM implies that everyone,regardless of which stage they are in, should receive theappropriate stage-matched intervention designed tomove them to the next stage; this includes precontemplators. The issue of whether health professionals shoulddeliver interventions to ‘all-comers’ or to subgroupsselected on the basis of higher risk, greater motivation orsome other criterion is a complex one that deservesmuch more detailed consideration than could be givento it in the editorial.ALTERNATIVE MODELSDiscarding the TTM does not necessarily mean abandoning the idea that behaviour change, including smoking cessation, involves movement through a sequence ofdiscrete stages. Two promising alternatives to the TTMare the precaution adoption process model (Weinstein &Sandman 2002; Sutton 2005) and the perspectives onchange model of smoking cessation (Borland, Balmford& Hunt 2004). In contrast to the TTM, both thesetheories are based on a thoughtful analysis of the process of behaviour change, but neither has been testedextensively.Among existing non-stage models, the theory ofplanned behaviour (Ajzen 1991, 2002) has severalattractive features: (1) it is a general theory; (2) it isclearly specified; (3) there exist clear recommendationsfor how the constructs should be operationalized; (4) ithas been widely used to study health behaviours as wellCommentaries 1045© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050as other kinds of behaviours; and (5) meta-analyses ofobservational studies show that it accounts for usefulamounts of variance in intentions and behaviour (Sutton2004). The theory is able to capture the moment-tomoment balance of motives mentioned by West and totake account of situational influences (because differentsets of beliefs may be salient in different situations). It isalso consistent with the idea that the processes involvedin the formation and modification of beliefs, attitudes andintentions may be largely automatic (Ajzen & Fishbein2000). However, there have been few experimental testsof the theory (Sutton 2002) and few intervention studies(Hardeman et al. 2002), and it has not been widelyapplied to smoking cessation.Concluding commentThe TTM has proven remarkably resilient to criticism.The Rhode Island group has not so far responded to myown critiques of the model or to those by Carey et al.(1999), Joseph et al. (1999), Littell & Girvin (2002) andRosen (2000) among others. The model is still accepteduncritically by many in the health promotion field. I hopethat West’s (2005) editorial does finally put the model torest, but I am not optimistic.STEPHEN SUTTONUniversity of CambridgeInstitute of Public HealthForvie SiteRobinson WayCambridge CB2 2SRE-mail: [email protected], I. (1991) The theory of planned behavior. OrganizationalBehavior and Human Decision Processes, 50, 179–211.Ajzen, I. (2002) The theory of planned behavior. http://www.people.umass.edu/aizen [accessed 13 April 2005].Ajzen, I. & Fishbein, M. (2000) Attitudes and the attitudebehavior relation: Reasoned and automatic processes. European Review of Social Psychology, 11, 1–33.Aveyard, P., Cheng, K. K., Almond, J., Sherratt, E., Lancashire,R., Lawrence, T., Griffin, C. et al. (1999) Cluster randomizedcontrolled trial of expert system based on the transtheoretical(‘stages of change’) model for smoking prevention and cessation in schools. BMJ, 319, 948–953.Aveyard, P., Griffin, C., Lawrence, T. & Cheng, K. K. (2003) Acontrolled trial of an expert system and self-help manualintervention based on the stages of change versus standardself-help materials in smoking cessation. Addiction, 98, 345–354.Aveyard, P., Sherratt, E., Almond, J., Lawrence, T., Lancashire,R., Griffin, C. et al. (2001) The change-in-stage and updatedsmoking status results from a cluster-randomized trial ofsmoking prevention and cessation using the transtheoreticalmodel among British adolescents. Preventive Medicine, 33,313–324.Borland, R., Balmford, J. & Hunt, D. (2004) The effectiveness ofpersonally tailored computer-generated advice letters forsmoking cessation. Addiction, 99, 369–377.Borland, R., Balmford, J., Segan, C., Livingston, P. & Owen, N.(2003) The effectiveness of personalized smoking cessationstrategies for callers to a Quitline service. Addiction, 98, 837–846.Carey, K. B., Purnine, D. M., Maisto, S. A. & Carey, M. P. (1999)Assessing readiness to change substance abuse: a criticalreview of instruments. Clinical Psychology Science and Practice,6, 245–266.Hardeman, W., Johnston, M., Johnston, D. W., Bonetti, D.,Wareham, N. & Kinmonth, A. L. (2002) Application of thetheory of planned behaviour in behaviour change interventions: a systematic review. Psychology and Health, 17, 123–158.Joseph, J., Breslin, C. & Skinner, H. (1999) Critical Perspectiveson the Transtheoretical Model and Stages of Change. In:Tucker, J. A., Donovan, D. M. & Marlatt, G. A., eds. ChangingAddictive Behavior: Bridging Clinical and Public Health Strategies,pp. 160–190. New York: Guilford.Lawrence, T., Aveyard, P., Evans, O. & Cheng, K. K. (2003) Acluster randomized controlled trial of smoking cessation inpregnant women comparing interventions based on the transtheoretical (stages of change) model to standard care. TobaccoControl, 12, 168–177.Littell, J. H. & Girvin, H. (2002) Stages of change: a critique.Behavior Modification, 26, 223–273.Pallonen, U. E., Velicer, W. F., Prochaska, J. O., Rossi, J. S., Bellis,J. M., Tsoh, J. Y. et al. (1998) Computer-based smoking cessation interventions in adolescents: description, feasibility, andsix-month follow-up findings. Substance Use and Misuse, 33,935–965.Prochaska, J. O., DiClemente, C. C., Velicer, W. F. & Rossi, J. S.(1993) Standardized, individualized, interactive, and personalized self-help programs for smoking cessation. Health Psychology, 12, 399–405.Prochaska, J. O., Velicer, W. F., Fava, J. L., Rossi, J. S. & Tsoh, J. Y.(2001a) Evaluating a population-based recruitmentapproach and a stage-based expert system intervention forsmoking cessation. Addictive Behaviors, 26, 583–602.Prochaska, J. O., Velicer, W. F., Fava, J. L., Ruggiero, L., Laforge,R. G., Rossi, J. S. et al. (2001b) Counselor and stimulus controlenhancements of a stage-matched expert system interventionfor smokers in a managed care setting. Preventive Medicine, 32,23–32.Prochaska, J. O., Velicer, W. F., Prochaska, J. M. & Johnson, J. L.(2004) Size, consistency and stability of stage effects for smoking cessation. Addictive Behaviors, 29, 207–213.Riemsma, R. P., Pattenden, J., Bridle, C., Sowden, A. J., Mather,L., Watt, I. S. et al. (2003) Systematic review of the effectiveness of stage based interventions to promote smoking cessation. BMJ, 326, 1175–1177.Rosen, C. S. (2000) Is the sequencing of change processes bystage consistent across health problems? A meta-analysis.Health Psychology, 19, 593–604.van Sluijs, E. M. F., van Poppel, M. N. M. & van Mechelen, W.(2004) Stage-based lifestyle interventions in primary care: arethey effective? American Journal of Preventive Medicine, 26,330–343.Sutton, S. R. (1996) Can ‘stages of change’ provide guidance inthe treatment of addictions? A critical examination of© 2005 Society for the Study of Addiction Addiction, 100, 1040–10501046 CommentariesProchaska and DiClemente’s model. In: Edwards, G. & Dare,C., eds. Psychotherapy, Psychological Treatments and the Addictions, pp. 189–205. Cambridge: Cambridge University Press.Sutton, S. (2000a) A critical review of the transtheoreticalmodel applied to smoking cessation. In: Norman, P., Abraham, C. & Conner, M., eds. Understanding and Changing HealthBehaviour: From Health Beliefs to Self-Regulation, pp. 207–225.Reading: Harwood Academic Press.Sutton, S. (2000b) Interpreting cross-sectional data on stages ofchange. Psychology and Health, 15, 163–171.Sutton, S. (2001) Back to the drawing board? A review of applications of the transtheoretical model to substance use. Addiction, 96, 175–186.Sutton, S. (2002) Testing attitude-behaviour theories usingnon-experimental data: An examination of some hiddenassumptions. European Review of Social Psychology, 13, 293–323.Sutton, S. (2004) Determinants of health-related behaviours:Theoretical and methodological issues. In: Sutton, S., Baum,A. & Johnston, M., eds. The Sage Handbook of Health Psychology,pp. 94–126. London: Sage.Sutton, S. (2005) Stage theories of health behaviour. In: Conner,M. & Norman, P., eds. Predicting Health Behaviour: Research andPractice with Social Cognition Models, 2nd edn, pp. 223–275.Buckingham: Open University Press.Weinstein, N. D., Rothman, A. J. & Sutton, S. R. (1998) Stagetheories of health behavior: Conceptual and methodologicalissues. Health Psychology, 17, 290–299.Weinstein, N. D. & Sandman, P. M. (2002) The precaution adoption process model. In: Glanz, K., Rimer, B. K. & Lewis, F. M.,eds. Health Behavior and Health Education: Theory, Research, andPractice, 3rd edn, pp. 121–143. San Francisco: Jossey-Bass.West, R. (2005) Time for a change: putting the Transtheoretical(Stages of Change) Model to rest. Addiction, 100, 1036–1039.ctionA PREMATURE OBITUARY FOR THETRANSTHEORETICAL MODEL: ARESPONSE TO WEST (2005)The editorial by Robert West eulogizing the Transtheoretical Model (TTM) offers a provocative perspective (West2005). However, it is not clear why Dr West feels the needto bury something that still has life or why he cannot create a new model from his insights that would make theold one obsolete, dying a natural death rather than whatI consider a premature interment. Although his critiqueoffers some valid concerns, essentially he repeats ongoingcriticisms that have been addressed previously in thisjournal and in more recent publications (Prochaska &DiClemente 1998; Connors et al. 2001; DiClemente &Velasquez 2002; DiClemente 2003; DiClemente,Schlundt & Gemell 2004) and continues to overreact toexaggerated claims that have been made about the utilityand scope of the model. I would agree that some claimshave been exaggerated and that there are challengingdata and anomalies that need to be examined, exploredand explained. However, the basic premise of the editorialis flawed. A balanced assessment would be more usefulfor advancing our understanding of the human changeprocess and for exploring both the stage and state aspectsof this process.Dr West’s critique is really a criticism of the stages ofchange and not the entire model. As is true of many previous critiques, he focuses on assessment of the stages ofchange and issues about time frames and labels.Although they are closely related, it is important not toconfuse construct with assessment and confound operationalizing a construct with the phenomenon that theconstruct is supposed to help explain.The dimensions of the Transtheoretical Model offer aframework that makes explicit elements of a humanintentional behavior change process and answers thequestion: what does it take for individuals to accomplishsuccessfully sustained behavior change? In contrast to aprior view of change as an on/off phenomenon (unmotivated or motivated; action or inaction), the originalinsight underlying stages of change was that thereappear to be a series of identifiable and separable tasksinvolved in changing a specific behavior. Stages were away to segment the process into meaningful steps relatedto critical tasks, namely concern about the problem andconsideration of the possibility of change, risk rewardanalysis and decision making, planning and prioritization, taking action and revising action plans, and integration of the behavior change into the person’s life-style.The terms ‘precontemplation’, ‘contemplation’, ‘preparation’, ‘action’ and ‘maintenance’ were an attempt toidentify specific steps or stages in the process and isolatesubsets of people who had similar tasks to accomplish asthey move forward in the process of change. Stages havealways been considered states and not traits so they arequite unstable and individuals can move between themquickly, engaging and abandoning some of these taskseven within a single session of intervention. The exception seems to be the action to maintenance shift, whichappears to need the passage of time for the task of consolidation of change. Individuals can also become stuck in atask, such as considering change for long periods of timebefore taking action. The labels and attempts at makingstages operational for assessment were thoughtful butarbitrary ways of labeling these sets of tasks and subgroups of people. Early work with the model followedlarge numbers of smokers for 2 and 3 years with andwithout interventions tracking their progress or lack ofprogress through the process of change. This extensiveresearch supported stage differences and the importanceof processes of change in the transitions from one stage tothe next (DiClemente & Prochaska 1998).Making a concept operational so that one can assessthe phenomenon is always arbitrary, and simply anattempt to create a dividing line that could be useful inisolating a concept or construct. This is true for all ourCommentaries 1047© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050psychological concepts such as stress, depression, anxiety, addiction, etc. Constructs always differ from the phenomenon and operationalizing and assessing are alwayschallenging. Although few would deny the existence ofanxiety or depression, there are many different ways ofoperationalizing these constructs, assessing individualswho may or may not have these conditions, and understanding the phenomenon. Problems of operationalization make it more difficult to study the phenomenon butshould not be confused with the value of the concept orconstruct. The objective of the TTM and the researchexamining the model has been to enhance our understanding of the process of change and our ability tointervene in this process and not to identify a rigidlydefined set of stages and prove the existence of thosestages.The key questions are whether these tasks are definable and separable to some degree, whether it is helpful toseparate these tasks in order to better conceptualize andmanage change, and whether we can identify and assistindividuals or groups who seem to be engaged in thesesimilar tasks. These separate tasks are not unique to theTTM and have been identified in many current theories ofhealth behavior change. Both the health belief model andBandura’s social cognitive theory describe decisionalconsiderations and self-evaluations that precede takingaction (Bandura 1986). The theory of planned behavioridentifies implementation planning as an importantdimension of change that precedes action. Early actionappears to be different from successfully sustainingchange, as is discussed in Marlatt’s relapse preventionmodel (Marlatt & Gordon 1985). What the stages do is toorganize these tasks into a logical sequence of activitiesthat seem to build upon one another. Individuals uninterested in change or unconcerned about a currentbehavior should differ from those convinced of the needto change and preparing for action both in their view ofchange and what they are doing to create change. However, simply because specific tasks can be identified as distinct does not mean that they are discontinuous anddichotomous. These tasks are part of a larger process ofchange and build on one another. Critical stage tasksneed to be completed in a ‘good enough’ fashion to allowthe individual to move forward but in reality stage tasksare not completely accomplished until successfully completed change is achieved. It seems obvious that someonecan move into action without having completed theproper decision-making, planning or prioritizationneeded to make the change successful. Stages are notboxes from which individuals jump, one to the next, inorder to make change, but represent tasks that can beaccomplished to a greater or lesser degree. In fact, relapseseems to be related to the quality of the accomplishmentof the stage of change tasks and not simply whether onetakes action. Recycling through the stages and the multiple attempts that individuals make in order to successfully recover from addiction seem to support the role ofsuccessive approximation in completing the decisionmaking, the commitment, the preparation, the plan andthe implementation in such a manner that can supportsuccessfully sustained change.Dr West contends that this view obstructs the view ofthe role of ‘moment to moment balance of desire versusvalue’ and the role of circumstances. I would argue thecontrary, that the stages offer a way of viewing and studying how the momentary and the circumstantial interactwith the larger process of change. There are implicit andexplicit cognitions and a host of normative comparisonsand self-evaluations that are operative in the process ofchange. Motivations are often momentary. Changeattempts can be very spontaneous looking and responsiveto specific events. I remember my days as a smoker whenI would wake up and say to myself that this is it and throwaway the cigarettes, only to search for them later thatmorning and abandon my attempt. Certainly there aremomentary influences and actions, but they seem part ofa larger process of change. Not until I was able to be convinced and convicted about smoking cessation, created aplan that could work for me and was able to stick withthat plan did I successfully quit smoking. Momentaryevents are not contrary to a process perspective, but complement it.There is ample evidence of significant differencesamong subgroups of individuals classified into differentstages that do not simply mirror ‘common sense’ differences between people actively changing and those whoare not as was indicated by Dr West. Across multiplebehaviors (smoking, dietary behaviors, physical activity,alcohol consumption and drug abuse) there are interesting and consistent differences among subgroups onmeaningful process of change dimensions. In longitudinal studies there have been consistent findings that individuals in earlier stages have less success in sustainingbehavior change. Dr West ignores these data.The model has also assisted in exploring interestingphenomena, has contributed to changing how treatmentprofessionals approach individuals referred to treatmentand challenged the field to think in a more differentiatedand complex manner about health and addictive behavior change (Stotts et al. 1996; Carbonari & DiClemente2000; DiClemente et al. 2003). The claim that the modelis hindering advances and exploration seems clearly erroneous. Practitioners have made interesting and creativechanges in the way they offer services and approach clients. Individuals who are in the process of change havetold us repeatedly that the model seems to reflect theirexperiences of changing a health behavior. There isincreased emphasis on early interventions and how inter-© 2005 Society for the Study of Addiction Addiction, 100, 1040–10501048 Commentariesventions can influence even people who seem to lackmotivation. Although often problematic in how stagesare assessed and simplistic in approach, research studiesinto the process of change have grown exponentially. In arecent presentation at the meeting of the Association forthe Advancement of Behavior Therapy, colleagues and Iexamined measures of stage and process of change in adually diagnosed sample of seriously mentally ill withcocaine dependence to see if measures looked the sameand were influenced by neurocognitive status. This led toan interesting discussion of whether seriously mentally illpeople accessed an intentional process of change or weremore influenced by current considerations and socialinfluences. Dr West’s contention that the model must bejettisoned before alternative views can be explored issimply not true, or is true only for those treating themodel as a religion and not a heuristic model to explorethe change process.Interment of such a provocative and heuristic modelseems premature and unnecessary. It would be a mistaketo return completely to a state model resembling the on/off views of the past. Readiness to change is not a singleconstruct but a compilation of tasks and accomplishmentsthat can produce both momentary change and sustainedchange. Pitting state against stage does a disservice to theprocess of intentional behavior change. In fact, this process incorporates and can elucidate many of the issuesthat Dr West identifies in his closing paragraph. There isclearly much more to understand about the process ofchange and how individuals go about creating and stabilizing a new behavior or abandoning one that is well established. The process involves biological, psychological andsocial/environmental determinants that are momentaryand more stable. However, the process of change appearsto be a very productive way to try to integrate these dimensions (DiClemente 2003). Hopefully, a dialogue on howthe model does or does not fit the process of change andhow various new discoveries challenge the model or makeit obsolete can promote a more complete and scientificunderstanding of human intentional behavior change.CARLO C. DICLEMENTEDepartment of PsychologyUniversity of MarylandBaltimore CountyMDUSAE-mail: [email protected], A. (1986) Social Foundations of Thought and Action: ASocial Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall.Carbonari, J. P. & DiClemente, C. C. (2000) Using transtheoretical model profiles to differentiate levels of alcohol abstinencesuccess. Journal of Consulting and Clinical Psychology, 68, 810–817.Connors, G., Donovan, D. & DiClemente, C. C. (2001) SubstanceAbuse Treatment and the Stages of Change: Selecting and PlanningInterventions. New York: Guilford Press.DiClemente, C. C. (2003) Addiction and Change. How AddictionsDevelop and Addicted People Recover. New York: Guilford Press.DiClemente, C. C., Carroll, K. M., Miller, W. R., Connors, G. J. &Donovan, D. M. (2003) A look inside treatment. Therapisteffects, the therapeutic alliance, and the process of intentionalbehavior change. In: Babor, T. & Del Boca, F. K., eds. TreatmentMatching in Alcoholism, pp. 166–183. Cambridge, UK:Cambridge University Press.DiClemente, C. C. & Prochaska, J. O. (1998) Toward a comprehensive, transtheoretical model of change: Stages of changeand addictive behaviors. In: Miller, W. R. & Heather, N., eds.Treating Addictive Behaviors, 2nd edn, pp. 3–24. New York: Plenum Press.DiClemente, C. C., Schlundt, D. & Gemell, L. (2004) Readinessand stages of change in addiction treatment. American Journalon Addictions, 13, 103–119.DiClemente, C. C. & Velasquez, M. (2002) Motivational interviewing and the stages of change. In: W. R. Miller & S.Rollnick., eds. Motivational Interviewing, 2nd edn: PreparingPeople for Change, pp. 201–216. New York: Guilford Publications, Inc.Marlatt, G. A. & Gordon, J. R. (1985) Relapse Prevention. NewYork: Guilford Press.Prochaska, J. O. & DiClemente, C. C. (1998) Comments, criteriaand creating better models. In: Miller, W. R. & Heather, N.,eds. Treating Addictive Behaviors, 2nd edn, pp. 39–45. NewYork: Plenum Press.Stotts, A., DiClemente, C. C., Carbonari, J. P. & Mullen, P. (1996)Pregnancy smoking cessation: a case of mistaken identity.Addictive Behaviors, 21, 459–471.West, R. (2005) Time for a change: putting the Transtheoretical(Stages of Change) Model to rest. Addiction, 100, 1036–1039.Blackwell Science, LtdOxfo100••••Original ArticleCommentaryCommentaryCommentaryWHAT DOES IT TAKE FOR A THEORYTO BE ABANDONED? THETRANSTHEORETICAL MODEL OFBEHAVIOUR CHANGE AS A TEST CASEI am grateful to my fellow researchers for their thoughtfulcomments on my editorial arguing for abandonment ofthe Transtheoretical Model of Behaviour Change (West2005).Herzog (2005) notes that the popularity of the modeldoes not seem to derive from a close analysis of its scientific merits. Etter’s (2005) commentary elaborates onsome of the major conceptual problems with the TTM.Hodgins (2005) notes limitations in the evidence base insupport of the model and argues for a moratorium onuncritical use of the model. However, he argues that themodel has been and could continue to be a useful stimulus to research. I have regretfully to demur for the reasonsgiven in the editorial.Sutton (2005) has argued for many years that themodel cannot be recommended and my editorial does lit-Commentaries 1049© 2005 Society for the Study of Addiction Addiction, 100, 1040–1050tle more than reinforce that view. However, he takes issuewith me on two counts. First, he argues that the finding of‘stage transitions’ which represents the main evidence infavour of the model, though common sense, is not trivial.Yet surely showing that individuals who at time A arethinking about doing something, or indeed trying to dosomething, are somewhat more likely to have done it attime B is not a major contribution to understanding ofmotivation.Secondly, Sutton argues that the failure to show thatTTM-based interventions are more effective than moretraditional interventions might have resulted from nothaving used the model appropriately. This kind of argument is deployed a great deal when empirical studies failto support theories in psychology. It cannot be ruled outas an explanation in this case. The problem is that it cannever be ruled out. How far should we go in allowing thiskind of ‘get out clause’? There is no simple answer, but ifa model has conceptual flaws and counterexamplesranged against it and if, after many years of research,independent reviews conclude that it has not led to development of more effective interventions, perhaps that isenough.DiClemente (2005) acknowledges that there are validconcerns with the model and rightly points out that myeditorial says little of substance that is new. However, hebelieves that the basic premise of the editorial is flawed.He argues that stages have always been considered statesand not traits and they are quite unstable, even changingwithin a session. But he also argues that the model aimedto segment people into those that could benefit from different types of intervention. These two statements seemat odds because there would not be much purpose in segmenting people on the basis of a state that is in any eventunstable.He also argues that my editorial confused the operationalisation of stages with the underlying concept. Inbehavioural research operationalisations are neededwhen we cannot measure something directly; the question here is what exactly is it that we are trying to measure and whether it has any reality beyond the method ofmeasurement. It is obvious that some smokers who,when they think about it, have some kind of intention tostop at some point in the not too distant future while others do not. It is also obvious that some smokers (we do notknow how many) will make more definite plans to stop ona particular date which they may or may not put intoeffect. If one wishes to go beyond this and posit ‘stages’that represent some more abstract and stable entity, theoperationalisation must specifically address the coredefinition of a stage which involves (1) stability and (2)discontinuity. The use of arbitrary dividing lines does notdo this. Using arbitrary dividing lines to put people’smotivational condition into artificial categories encourages people to make statements such as ‘xx% of smokersare in the contemplation stage’ as though the figurereferred to some real quality attaching to those individuals when it does not.DiClemente argues that the stage approach provides aframework into which momentary influences and statescan be understood and harnessed. He cites his own experience of stopping smoking in which the formulation of aplan to which he was committed was a precursor for lasting change. I doubt whether anyone would dispute astatement that making a plan to which one is committedcan (but does not always) help to combat momentarywishes and urges. Put in those terms, it seems, againto be a statement of the obvious. Put in more rigidterms—that making a plan to which one is committedis needed for lasting change—it is contradicted bycounterexamples.DiClemente says that I have ignored data showingthat people in earlier stages are less likely to sustainbehaviour change. But I did not ignore this data; I andother commentators have merely said that this kind oftest is too weak and that where stage assessment has beencompared with a simple rating of desire or an addictionmodel, it has been less successful. The onus is very muchon the proponents of the TTM after many years and hundreds of research papers to point to evidence that theirapproach predicts actual behaviour better than simplealternatives.It seems like a very negative thing to call for the abandonment of something into which so much has beeninvested. However, I think there is an onus of those ofworking in the field to be rigorous and objective in ourevaluation of theories and models. In the end it is selfdefeating to persist with ideas that are misconceived. Theproblem of theories that are developed and pursued without adequate regard to counter-examples and to whetherthey provide better prediction of behaviour than existingaccounts is endemic in behavioural research. Perhaps thetime has come for a root and branch review of our methods of theory development, testing and application.Here is a simple common sense account of healthrelated behaviour. It contains no insights derived fromformal study but is merely an articulation of what anyintelligent lay person might propose:‘Engaging in or failing to engage in health-protectivebehaviour patterns depends on opportunities affordedby the one’s social and physical environment and thebalance of motivations when those opportunities arepresent. These motivations include desires, urges,needs, habits, evaluations and level of commitment toany prior resolutions. The specific problem for healthpromoting behaviours is that they are often less pleasant than the alternatives on offer at the time. Inter-© 2005 Society for the Study of Addiction Addiction, 100, 1040–10501050 Commentariesventions to reduce chronic unhealthy behaviourpatterns involve reducing opportunities for unhealthybehaviours, increasing opportunities for healthybehaviours, reducing the actual or perceived attractiveness of or need for the unhealthy option, increasing the actual or perceived attractiveness of thehealthy option, and prompts to make and keep to resolutions to make lasting changes.Addiction presents special problems because theaddictive behaviour (usually a drug) causes changesto the addict’s CNS and/or social and physical environment which undermine desires to remain abstinent and/or heighten desire to continue with theactivity. This means that more powerful and sustainedinterventions to bolster motivation to abstain and toreduce motivation to engage in the addictive behaviour are needed.’Any supposedly scientific model of behaviourchange should be able to do demonstrably better atprediction and development of interventions than anoperationalisation of this model than this. It should notbe hard.ROBERT WESTHealth Behaviour UnitDepartment of Epidemiology, Brook HouseUniversity College London2–16 Torrington PlaceLondon WC1E 6BTE-mail: [email protected], C. (2005) A premature obituary for the Transtheoretical Model: a response to West (2005). Addiction, 100,1046–1048.Etter, J.-F. (2005) Theoretical tools for the industrial era insmoking cessation counselling: a comment on West (2005).Addiction, 100, 1041–1042.Herzog, T. (2005) When popularity outstrips the evidence: comment on West (2005). Addiction, 100, 1040–1041.Hodgins, D. (2005) Weighing the pros and cons of changingchange models: a comment on West (2005). Addiction, 100,1042–1043.Sutton, S. (2005) Another nail in the coffin of the Transtheoretical model? A comment on West (2005). Addiction, 100,1043–1046.West, R. (2005) Time for a change: putting the Transtheoretical(Stages of Change) Model to rest. Addiction, 100, 1036–1039.