Authors' objectives

To summarise the findings from empirical applications of the transtheoretical model (TTM) of behaviour change in the domain of physical activity.

Searching

MEDLINE, PsycLIT, SPORTDiscus, and UnCover were searched from 1983 to January 2000 for studies in the English language, using the keyword combinations of 'health behaviour, physical activity, exercise (and) stages, changes, readiness and transtheoretical'. Manual searches were also conducted of the 1998 to 2000 issues of the following: American Journal of Health Promotion, Annals of Behavioural Medicine, Health Psychology, Journal of Sport and Exercise Psychology, Journal of Sports Sciences, Preventive Medicine, and Research Quarterly for Exercise and Sport. The reference sections of narrative reviews and primary studies were examined. In addition, an online reference list provided by the University of Rhode Island Cancer Prevention Centre was reviewed. Finally, personal files were searched and investigators in the field were contacted.

Study selection

Assessment of study quality

The authors do not state that they assessed validity.

Data extraction

The authors do not state how the data were extracted for the review, or how many of the reviewers performed the data extraction.

The data extracted included: study design, sampling method, recruitment method, publication status, gender, age, country, number, criteria for action, stage, pros and cons, and processes of change.

Methods of synthesis

Results of the review

Seventy-one published studies containing 91 samples were included in the review. Of these, 54 used a cross-sectional design, 6 were longitudinal, 10 were quasi-experimental and one was a randomised controlled trial.

Of the 91 independent samples, 80 contained useable data for further analyses. Across all constructs and stage transitions (n=56), 413 effect sizes were computed for physical activity (n=46), self-efficacy (n=67), pros (n=50), cons (n=50), experimental processes of change (n=100), and behavioural processes of change (n=100). Across the total sample, 14% of the individuals were in precontemplation, 16% in contemplation, 23% in preparation, 11% in action and 36% in maintenance. The results on the moderator variables were tabulated in the paper. The stage of change and physical activity was consistent with the TTM; the level of physical activity increased as individuals moved to a higher stage of change. The largest effect was evident for preparation for action (d=0.85, 95% CI: 0.64, 1.07). Small to moderate increases in physical activity were also evident from precontemplation to contemplation (d=0.34, 95% CI: 0.14, 0.55; fail-safe k=7), suggesting that transitions between inactive stages are associated with changes in physical activity.

Self efficacy: the effect estimates across the stage transitions were all positive and significant, suggesting that confidence to be active increased with each stage of change, as proposed by the TTM.

Decisional balance: all effect sizes for behavioural pros were positive and significant with the exception of contemplation to preparation, suggesting that perceived benefits of change increase for every forward stage transition. The largest and most robust effect size was evident from precontemplation to contemplation (d=0.97; fail-safe k=50). Contemplation to preparation had the smallest and least robust effect estimate (d=0.01).

Decisional balance: all effect sizes for behavioural cons were small to moderate, significant and negative, suggesting that the perceived disadvantages of becoming physically active decrease across the stages.

Processes of change (findings tabulated in the paper): across all processes of change, the largest effects were evident from precontemplation to contemplation (d range: 0.55 to 1.18). For all five behavioural processes and three experiential processes, the smallest effects were evident from action to maintenance (d range: 0.03 to 0.07). Across all processes and stage transitions, the largest single effect size was for self-liberation from precontemplation to contemplation (d=1.18; fail-safe k=25).

Authors' conclusions

The results supported the application because core constructs differ across the stages, and most changes are in the direction predicted by the TTM. The authors offer three general conclusions. Firstly, existing data are unable to confirm whether physical activity behaviour change occurs in a series of stages that are qualitatively different, or along adjacent segments of an underlying continuum. Secondly, there is a growing need to standardise and improve the reliability of measurement of the stages of the TTM. Finally, the role of processes of change requires re-examination, as higher order constructs are not apparent in the physical activity domain and stage by process interactions.

CRD commentary

This review attempted to illustrate the application of a theoretical model to predict and explain the uptake and increase of physical activity, and as such is a useful topic area. The literature search was limited to published English language papers, and although several mechanisms in addition to searching electronic databases were employed, there remains the possibility that important papers have been missed. There were no details of the review process, such as the quality assessment, data extraction, and the number of reviewers involved. The details of the large primary studies were presented (albeit briefly), and the statistical methods used were described clearly.

The meta-analysis was appropriate, and the conclusions drawn from the data presented appear justified.

Implications of the review for practice and research

Practice: The authors did not state any implications for practice.

Research: The authors state 'Future research should examine the moderators and mediators of stage transition'.

Bibliographic details

Marshall S J, Biddle S J. The transtheoretical model of behavior change: a meta-analysis of applications to physical activity and exercise. Annals of Behavioral Medicine 2001; 23(4): 229-246. [PubMed: 11761340]

Other publications of related interest

Hunter JE, Schmidt FL. Methods of meta-analysis: correcting error and bias in research findings. Newbury Park (CA): Sage; 1990.

Indexing Status

Subject indexing assigned by NLM

MeSH

Behavior Therapy /statistics & numerical data; Decision Making; Exercise /psychology; Health Behavior; Humans; Models, Theoretical; Self Efficacy

AccessionNumber

12001006341

Database entry date

28/02/2003

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.