CoChat
Sign Up
Cited by 680Behavioral Health and InterventionsHealth Policy Implementation ScienceMeta-analysis and systematic reviews

The DOI matches its Crossref record; 176 of the 178 references it cites were found in a registry or an index, and 2 are not checked yet; none of the 141 sources its claims cite is checked for support yet.

Behaviour change techniques: the development and evaluation of a taxonomic method for reporting and describing behaviour change interventions (a suite of five studies involving consensus methods, randomised controlled trials and analysis of qualitative data)

Open access
Michie, Susan, Wood, Caroline E, Johnston, Marie, Abraham, Charles, Francis, Jill J, Hardeman, Wendy·2015·Health Technology Assessment
DOI 10.3310/hta19990PMC PMC4781650
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Meeting global health challenges requires effective behaviour change interventions (BCIs). This depends on advancing the science of behaviour change which, in turn, depends on accurate intervention reporting. Current reporting often lacks detail, preventing accurate replication and implementation. Recent developments have specified intervention content into behaviour change techniques (BCTs) – the ‘active ingredients’, for example goal-setting, self-monitoring of behaviour. BCTs are ‘the smallest components compatible with retaining the postulated active ingredients, i.e. the proposed mechanisms of change. They can be used alone or in combination with other BCTs’ (Michie S, Johnston M. Theories and techniques of behaviour change: developing a cumulative science of behaviour change. Health Psychol Rev 2012; 6 :1–6). Domain-specific taxonomies of BCTs have been developed, for example healthy eating and physical activity, smoking cessation and alcohol consumption. We need to build on these to develop an internationally shared language for specifying and developing interventions. This technology can be used for synthesising evidence, implementing effective interventions and testing theory. It has enormous potential added value for science and global health. Objective: (1) To develop a method of specifying content of BCIs in terms of component BCTs; (2) to lay a foundation for a comprehensive methodology applicable to different types of complex interventions; (3) to develop resources to support application of the taxonomy; and (4) to achieve multidisciplinary and international acceptance for future development. Design and participants: Four hundred participants (systematic reviewers, researchers, practitioners, policy-makers) from 12 countries engaged in investigating, designing and/or delivering BCIs. Development of the taxonomy involved a Delphi procedure, an iterative process of revisions and consultation with 41 international experts; hierarchical structure of the list was developed using inductive ‘bottom-up’ and theory-driven ‘top-down’ open-sort procedures ( n = 36); training in use of the taxonomy (1-day workshops and distance group tutorials) ( n = 161) was evaluated by changes in intercoder reliability and validity (agreement with expert consensus); evaluating the taxonomy for coding interventions was assessed by reliability (intercoder; test–retest) and validity ( n = 40 trained coders); and evaluating the taxonomy for writing descriptions was assessed by reliability (intercoder; test–retest) and by experimentally testing its value ( n = 190). Results: Ninety-three distinct, non-overlapping BCTs with clear labels and definitions formed Behaviour Change Technique Taxonomy version 1 (BCTTv1). BCTs clustered into 16 groupings using a ‘bottom-up’ open-sort procedure; there was overlap between these and groupings produced by a theory-driven, ‘top-down’ procedure. Both training methods improved validity (both p < 0.05), doubled the proportion of coders achieving competence and improved confidence in identifying BCTs in workshops (both p < 0.001) but did not improve intercoder reliability. Good intercoder reliability was observed for 80 of the 93 BCTs. Good within-coder agreement was observed after 1 month ( p < 0.001). Validity was good for 14 of 15 BCTs in the descriptions. The usefulness of BCTTv1 to report descriptions of observed interventions had mixed results. Conclusions: The developed taxonomy (BCTTv1) provides a methodology for identifying content of complex BCIs and a foundation for international cross-disciplinary collaboration for developing more effective interventions to improve health. Further work is needed to examine its usefulness for reporting interventions. Funding: This project was funded by the Medical Research Council Ref: G0901474/1. Funding also came from the Peninsula Collaboration for Leadership in Applied Health Research and Care.

Checked 2026-10-06 · 2 references not checked yet

Overview

AI snapshot

Written from a shortened version of the paper’s full text (it is longer than we read).

TL;DR

This suite of five studies developed and evaluated Behaviour Change Technique Taxonomy version 1 (BCTTv1), a consensus-based taxonomy of 93 distinct behaviour change techniques (BCTs) hierarchically clustered into 16 groupings. BCTTv1 was shown to be usable reliably and validly by trained coders for identifying BCTs in intervention descriptions, but evidence that it improves reporting of observed interventions was mixed and not clearly supportive.

Key findings

  • Study 1/2 produced BCTTv1: 93 distinct, clearly labelled, well-defined BCTs hierarchically organised into 16 groupings. The open-sort task yielded 16 groupings; 59 of 87 BCTs were reliably allocated to 12 of 14 Theoretical Domains Framework domains, with 47 significant and 12 borderline. There was a significant association between the 16 “bottom-up” groupings and 13 “top-down” groupings (χ² = 437.80; p < 0.001), but only six of 208 pairings achieved satisfactory evidence of similarity.
  • Study 3 found training improved trainee agreement with expert consensus for both 1-day workshops and distance group tutorials (both p < 0.05), with the proportion achieving competence (PABAK > 0.60) doubling: 46% of workshop trainees and 78% of tutorial trainees achieved competence after training. Confidence in identifying BCTs improved in workshops (p < 0.001), but intercoder agreement did not improve for either method (workshops p = 0.08; tutorials p = 0.57).
  • Study 4 found 80 of the 93 BCTs were identified by at least one trained coder. Good intercoder reliability was observed for 80 BCTs: 64 of 80 (80%) achieved mean PABAK ≥ 0.70 and 59 of 80 (74%) achieved ≥ 0.80. Test-retest reliability was good (p < 0.001), and concurrent validity was good for 14 of 15 BCTs. Reliability was poorer for frequent BCTs: 9 of 16 frequently identified BCTs failed to achieve PABAK > 0.70.
  • Studies 5a–5c found unclear benefit of BCTTv1 for reporting observed interventions. Study 5a showed no difference with taxonomy alone (all p > 0.05). Study 5b found descriptions by trained writers with BCTTv1 were rated poorer in quality than untrained writers without BCTTv1 (all p < 0.05), but reliability of BCT identification was greater for trained writers with BCTTv1 (mean PABAK trained = 0.87; untrained = 0.84; p < 0.05). Study 5c found within-person quality was higher after training with BCTTv1 (all p < 0.05 except replicability of mode of delivery), with no differences in reliability (p = 0.50) or validity (p = 0.28).

Methods

  • The paper reports a suite of five empirical studies using Delphi methods with an International Advisory Board, open-sort and closed-sort tasks, training evaluation, reliability/validity coding, and within- and between-participant experimental designs. A total of 400 participants were involved across studies, with some taking part in more than one study.
  • Study 1 used a two-round Delphi exercise with 14 international behaviour change experts plus International Advisory Board and study team feedback to develop BCT labels and definitions. Study 2 used 18 open-sort participants and 18 closed-sort participants to generate and compare “bottom-up” and “top-down” hierarchical structures. Study 3 evaluated training with 161 trainee coders (1-day workshops and distance group tutorials). Study 4 used 40 trained coders to code 40 intervention descriptions at two time points 1 month apart. Studies 5a–5c involved 166 writers, 12 untrained raters, and 12 trained coders.
  • Main analyses included prevalence-and-bias-adjusted kappa (PABAK) for intercoder reliability, test-retest reliability, and agreement with expert consensus; cluster analysis and discriminant content validity methods for sort tasks; and t-tests, ANOVA, and repeated-measures ANOVA for training and reporting studies.

Limitations

  • The paper states that the Delphi exercise involved 14 individuals, with a predominance of European experts from a psychological background; that arbitrary cut-offs were used for amending BCT labels and definitions; and that training covered only 24 of the 93 BCTs. It also states that study 5’s video-observation writing task was artificial, constrained by time pressure, and that training condition was confounded with time of day; the quality measures’ psychometric properties were not established.
  • The paper states that samples were predominantly female and from the UK, limiting diversity; that tutorial groups were not directly compared for efficacy; and that more training is required to become expert in using BCTTv1. It also notes that reliability was poorer for frequently identified BCTs and that some BCTs (e.g. credible source, social support unspecified, instruction on how to perform behaviour) required further refinement.
  • The text makes evident that the reporting studies did not show a clear benefit of BCTTv1 for writing observed intervention descriptions, and that the negative findings in study 5b may reflect task demands, insufficient training in report writing, or the artificial nature of the task rather than the taxonomy itself.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3310/hta19990

Cites

176 of 178 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 of 141 cited sources checked · 0 supported · 0 flagged

Show all 246 claims

  • Probably not a substantial claim. 156
  • It may state two separate claims as one. 174
  • The quote only partly states this claim. 15
291 of 291

  • The paper relies on this without support.

    Lack of a methodology for unpacking the 'black box' of behaviour change interventions prevents progress in developing more effective interventions and in reliably implementing effective ones.

    “Lack of a methodology for unpacking this 'black box' prevents progress in developing more effective interventions and in reliably implementing effective ones.”

  • The paper relies on this without support.

    There is a need for consensus about labels and definitions that can be understood across disciplines, behavioural domains and countries.

    “There is a need for consensus about labels and definitions that can be understood across disciplines, behavioural domains and countries.”

  • The paper relies on this without support.

    The 93-item BCT list poses problems for easy recall of, and ready access to, the BCTs and thus its speed and accuracy of use.

    “The 93-item BCT list poses problems for easy recall of, and ready access to, the BCTs and thus its speed and accuracy of use.”

  • The paper relies on this without support.

    The process of coding interventions into BCTs is a highly skilled task requiring familiarity with labels and definitions and ability to make a series of complex interpretative judgements.

    “The process of coding interventions into BCTs is a highly skilled task requiring familiarity with labels and definitions and ability to make a series of complex interpretative judgements.”

  • The paper relies on this without support.

    This requires an effective programme of coder training to ensure sufficient level of skills to demonstrate reliability and validity of identifying BCTs.

    “This requires an effective programme of coder training to ensure sufficient level of skills to demonstrate reliability and validity of identifying BCTs.”

  • The paper relies on this without support.

    There was no prior agreed methodology for developing a taxonomy of behaviour change techniques.

    “There was no prior agreed methodology for this work and there are limitations to the methods we have used.”

  • The paper relies on this without support.

    The experts who participated in the sort tasks followed the instruction to group BCTs by similarity of active ingredients.

    “Instructions guided the experts to 'group together BCTs which have similar active ingredients i.e. by the mechanism of change, NOT the mode of delivery'.73”

  • The paper relies on this without support.

    The hierarchical structure and grouping of BCTs within the taxonomy will increase the speed by which BCTs can be recalled by users.

    “The hierarchical structure and grouping of BCTs within the taxonomy has practical use in that it is predicted to increase the speed by which BCTs can be recalled by users.”

  • The paper relies on this without support.

    Links between BCTs and theory can be used to inform the design and evaluation of BCIs.

    “It also has theoretical interest in that links between BCTs and theory can be used to inform the design and evaluation of BCIs.”

  • The paper relies on this without support.

    Expert coding, assessed as consensus between expert coders, is the nearest available objective standard of validity for BCT coding.

    “Expert coding, assessed here as a consensus between expert coders, is the nearest we have to an objective standard of 'validity'.”

  • The paper relies on this without support.

    The increasing use of the BCT approach, particularly for specifying interventions in evidence synthesis, has prompted high demand for training in the reliable and valid application of BCTTv1.

    “The increasing use of the BCT approach, in particular for specifying interventions in evidence synthesis, has prompted high demand for training in the reliable and valid application of BCTTv1.”

  • The paper relies on this without support.

    Training in the most frequently occurring BCTs is appropriate because training in 93 BCTs is not feasible within the proposed training intensity and duration, and because frequently occurring BCTs are more accessible and useful for trainees to learn.

    “The decision was made to train coders in the most frequently occurring BCTs as training in 93 BCTs was not feasible within the proposed training intensity and duration, and because frequently occurring BCTs would be more accessible to trainees as well as more useful for them to learn.”

  • The paper relies on this without support.

    The 40 intervention descriptions are representative of published BCI descriptions.

    “Forty descriptions of BCIs were used to test reliability of identification of BCT labels and definitions.”

  • The paper relies on this without support.

    The coders trained for this study are typical of those likely to want to use BCT labels and definitions to identify intervention components associated with greater intervention effectiveness.

    “The coders trained for this study seem typical of those likely to want to use BCT labels and definitions to identify intervention components associated with greater intervention effectiveness, for example when conducting meta-analyses.20,23,48,56”

  • The paper relies on this without support.

    Interventions are usually developed in an iterative process of reading and discussion, so using observed interventions is an artificial scenario.

    “However, this is a rather artificial scenario as usually interventions are developed in an iterative process of reading and discussion.”

  • The paper relies on this without support.

    The normal practice is not to describe an intervention after observing it on a video; it is to describe it after reading and discussion, without a strict time pressure.

    “The normal practice would not be to describe an intervention after observing it on a video, it would be to describe it after reading and discussion, without a strict time pressure.”

  • The paper relies on this without support.

    Intervention descriptions usually emerge from reading and discussions.

    “Intervention descriptions usually emerge from reading and discussions, and a better evaluation would be to investigate the impact on this process of using BCTTv1.”

  • The paper relies on this without support.

    Assessing validity of BCT identification is less well established than using indices of reliability of BCT identification.

    “While the use of indices of reliability of BCT identification is an accepted method, assessing validity of BCT identification is less well established.”

  • The paper relies on this without support.

    Consensus following independent judgements by experienced expert coders is the best available approximation to a criterion of validity against which judgements can be evaluated.

    “We used consensus following independent judgements by experienced, 'expert', coders as our best approximation to a criterion of validity against which judgements could be evaluated.”

  • The paper relies on this without support.

    When links between BCTs and theoretical constructs are made, this is usually the result of the researchers' judgement rather than being based on evidence, pilot work or a systematic methodology of establishing this.

    “When links are made, this is usually the result of the researchers' judgement rather than being based on evidence, pilot work or a systematic methodology of establishing this.”

  • The paper relies on this without support.

    Behaviour change techniques are technologies and in themselves are agnostic with regard to the role of theories.

    “Behaviour change techniques are technologies and in themselves are agnostic with regard to the role of theories.”

  • The paper relies on this without support.

    Dividing the assistance prompt into 'prescription given' and 'referral made' simplifies documentation and operationalizes the fourth A (Assist) by making clear that referral and prescription are the two primary options available to provider and patient.

    “This approach was meant to simplify the documentation process and operationalize the 4th A, Assist, to make it clear that referral and prescription were the two primary options available to the provider and patient.”

  • The paper relies on this without support.

    Brief educational messages attached to test result reports influence providers' future requests for the targeted tests.

    “The reminder messages were intended to influence future requests for the targeted tests.”
23 of 23

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Different BCTs are named as the four just below the reliability threshold

    Major

    Two passages identify four BCTs scoring 0.68 or 0.69 at time 1, but their lists differ; the later passage names demonstration and self-monitoring where the earlier one names action planning and behavioural practice/rehearsal.

  • Study 5c names different exceptions to its quality improvement

    Minor

    The summary identifies replicability of mode of delivery as the exception to significant quality improvements. The detailed results instead identify replicability of active ingredients as non-significant and include mode of delivery among the improvements.

  • Test-retest counts reverse the reliability thresholds

    Minor

    Among the same 32 coders, the number scoring at least 0.90 cannot exceed the number scoring at least 0.80. The reported counts are 18 and 14, respectively.

  • Listed study 5 workshop attendance does not total 166 writers

    Minor

    The attendance figures for the six listed workshops sum to 142, not the stated 166 writers.

    Reported 166 · recomputed 142

  • Conflicting counts of frequently identified BCTs with poor reliability

    Minor

    The summary says nine of 16 frequently identified BCTs failed the 0.70 reliability threshold. The detailed results say 16 BCTs failed it and that all 16 were frequently identified.

  • Theoretical-domain allocation counts differ from Table 2

    Minor

    The narrative reports six allocations for goals and environmental context and five for emotion. Table 2 gives five, five and four, respectively, in its column for allocations with p < 0.10.

  • Twenty-four BCTs minus 18 leaves six, not seven

    Minor

    The discussion says seven of the 24 most frequently identified BCTs remained after 18 were allocated, but the remainder is six. The earlier results also list six unlinked BCTs.

    Reported 7 · recomputed 6

7 of 7

Citations

Cites

Flagged first, then by citations

Show all 177 references

Research works printed but not linked: 15

  1. NICE, 2007, Behaviour Change: The Principles for Effective Interventions (PH6)
  2. Dixon, 2012, Health Behavior Change Competency Framework: Competences to Deliver Interventions to Change Lifestyle Behaviors That Affect Health
  3. West, 2013, Carbon monoxide verified 4-week quit rates in the English Stop Smoking Services before versus after establishment of the National Centre for Smoking Cessation and Training, Smok Br, 1, 1
  4. NCSCT, National Centre for Smoking Cessation and Training
  5. Michie, 2005, Behaviour change interventions: developing a classification system
  6. University College London, The Behaviour Change Technique Taxonomy Project. Collaborators
  7. World Health Organization, 1992, ICD-10 International Statistical Classification of Diseases and Related Health Problems
  8. University College London, The Behaviour Change Technique Taxonomy Project
  9. Michie, 2013, Encyclopeadia of Behavioural Medicine, 182
  10. British Websites, BCTTv1. Online Training
  11. Johnston, 2013, Specifying the active content of behaviour change interventions: the Behaviour Change Technique Taxonomy v1 (BCTTv1), Revista de Psicología de la Salud (New Age), 1, 131, 10.1007/s12160-013-9486-6 · DOI · Paper page
  12. Gwet, 2012, Handbook of Inter-Rater Reliability: The Definitive Guide to Measuring the Extent of Agreement Among Multiple Raters
  13. NICE, 2014, Behaviour Change: The Principles for Effective Interventions (PH6)
  14. NICE, 2014, Behaviour Change: Individual Approaches (PH49)
  15. NCSCT, Very Brief Advice Training Module

Cited by

Sorted by citations

Items 601–677 of 677

Related

See all

Metadata

Authors
Michie, S., Wood, C. E., Johnston, M., Abraham, C., Francis, J. J., & Hardeman, W.
Affiliations
  • University College London (GB)
  • University of Aberdeen (GB)
  • University of Exeter (GB)
  • City, University of London (GB)
  • University of Cambridge (GB)
Corresponding author
Michie, Susan
ORCID
Venue
Health Technology Assessment
Publisher
National Institute for Health and Care Research
Published
2015-11
Volume
19
Issue
99
Pages
1-188
ISSN
  • 1366-5278 (ISSN-L)
  • 2046-4924
DOI
10.3310/hta19990
PMC
PMC4781650
OpenAlex
W2186580260
Semantic Scholar
4a3d10d88ba6875c77917bf09b5ae8f9e27cd1f6
Cited by
680
References
177
Citations by year
2026: 52 · 2025: 55 · 2024: 76 · 2023: 60 · 2022: 87 · 2021: 76 · 2020: 114 · 2019: 43 · 2018: 50 · 2017: 45 · 2016: 17 · 2015: 2
FWCI
60.74
Type
journal article
Language
en
Open access
Licence
Locations
  • Health Technology Assessment (publishedVersion, cc-by)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
  • City Research Online (City University London) (acceptedVersion)
  • Aberdeen University Research Archive (Aberdeen University) (submittedVersion, cc-by)
  • UEA Digital Repository (University of East Anglia) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
  • Figshare (submittedVersion, cc-by)
Funders
Medical Research Council
Topics
  • Social Sciences › Psychology › Applied Psychology › Behavioral Health and Interventions
  • Health Sciences › Health Professions › General Health Professions › Health Policy Implementation Science
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
Keywords
behaviour change techniques, behaviour change interventions, taxonomy development, intervention reporting, intervention description, coding reliability, intercoder reliability, test-retest reliability, validity, Delphi consensus, complex interventions, training workshop, global health, mechanisms of change, theory-based intervention design
MeSH terms
Data Accuracy, Behavior Therapy, Classification, Delphi Technique, Humans, Technology Assessment, Biomedical, Global Health, Health Behavior, Randomized Controlled Trials as Topic, Qualitative Research

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review