Published  |  Last updated

Quick Answer: The COM-B model says a behaviour happens when a person has the capability, the opportunity and the motivation to do it. Michie, van Stralen and West proposed it in 2011 after finding that none of 19 earlier frameworks covered the full range of intervention functions or policies.1 In a survey of young adults, the model explained 31% of the variance in physical activity and 23% in eating behaviour.2 Use it to ask which of its six sources is the barrier, then pick intervention functions that address that source.1

What Is the COM-B Model?

COM-B describes a ‘behaviour system’ of three conditions that interact to generate behaviour: capability, opportunity and motivation. Behaviour in turn influences all three.1 The model is both a way to understand why a behaviour does or does not happen and a basis for designing an intervention to change it.1

Swipe sideways to see every column.

ComponentSourceHow Michie et al. describe it
CapabilityPhysicalThe physical side of the capacity to engage in the activity; physical skill development is the focus of training.
PsychologicalThe capacity to engage in the necessary thought processes, such as comprehension and reasoning; includes knowledge and skills.
OpportunityPhysicalWhat the environment affords.
SocialWhat the cultural milieu allows, including the words and concepts that make up our language.
MotivationReflectiveProcesses involving evaluations and plans.
AutomaticEmotions and impulses that arise from associative learning or innate dispositions; includes habit.

Descriptions paraphrase Michie et al. 2011.1

The authors write that capability includes having the necessary knowledge and skills, that opportunity covers all factors lying outside the individual, and that motivation is not just goals and conscious decision-making but also habitual processes and emotional responding.1 They also state that all six components apart from reflective motivation are necessary for a given behaviour, and that the system gives equal status to intra-psychic and external factors.1

Where COM-B Came From

The 2011 paper started from a review of frameworks used to describe behaviour change interventions. A literature search plus consultation with experts produced 19 frameworks. They were judged on comprehensiveness, coherence and a clear link to an overarching model of behaviour. None covered the full range of intervention functions or policies, and only a minority met the criteria of coherence or linkage to a model of behaviour.1

To build the model, the authors drew on a 1991 US consensus meeting of behavioural theorists, which named skills, a strong intention and no environmental constraints as prerequisites for a volitional behaviour, and on the legal idea of means, opportunity and motive. They added non-volitional mechanisms such as habit.1

From COM-B to the Behaviour Change Wheel

COM-B forms the hub of the Behaviour Change Wheel. Around it are nine intervention functions aimed at deficits in one or more of the three conditions, and around those are seven categories of policy that could enable the interventions.1 Table 2 of the paper links each source to the functions that can address it. We redrew that table in R and counted the links.1

Matrix of the six COM-B sources against nine intervention functions from Michie 2011 Table 2: Enablement is linked to five sources, environmental restructuring to three and modelling to one
Links between COM-B sources and intervention functions, redrawn from Michie et al. 2011, Table 2 (CC BY 2.0). Counts computed in R by Editverse.

Reading the matrix: automatic motivation is linked to six functions, the most of any source, and reflective motivation to four. Enablement is linked to five of the six sources, and modelling to only one. The count is a count of links in the table, not a measure of how effective a function is.

The nine intervention functions

Swipe sideways to see every column.

FunctionDefinitionExample given by the authors
EducationIncreasing knowledge or understandingProviding information to promote healthy eating
PersuasionUsing communication to induce positive or negative feelings or stimulate actionUsing imagery to motivate increases in physical activity
IncentivisationCreating expectation of rewardUsing prize draws to induce attempts to stop smoking
CoercionCreating expectation of punishment or costRaising the financial cost to reduce excessive alcohol consumption
TrainingImparting skillsAdvanced driver training to increase safe driving
RestrictionUsing rules to reduce the opportunity to engage in the target behaviourProhibiting sales of solvents to people under 18
Environmental restructuringChanging the physical or social contextOn-screen prompts for GPs to ask about smoking
ModellingProviding an example for people to aspire to or imitateTV drama scenes involving safe-sex practices
EnablementIncreasing means or reducing barriers to increase capability or opportunityBehavioural support for smoking cessation

The seven policy categories are communication/marketing, guidelines, fiscal, regulation, legislation, environmental/social planning and service provision.1 The authors tested the wheel by coding two documents. Inter-rater agreement was 88% for the 2010 English tobacco control strategy and 79% for the NICE obesity guidance.1

A Real Case: COM-B Tested in Young Adults

Willmott, Pang and Rundle-Thiele surveyed young adults aged 18 to 35 online and tested COM-B with structural equation modelling in two behaviours. The physical activity analysis included 582 people (mean age 22.8 years, 80.3% female) and the eating analysis 455 (mean age 24.9 years, 80.8% female), two independent samples.2 Physical activity was measured with the IPAQ short form and eating with a diet quality score. The paper is open access under CC BY 4.0, and we redrew its model weights in R.2

Two path diagrams of the COM-B model in young adults: physical activity with capability to motivation 0.91, opportunity to motivation 0.11, motivation to behaviour 0.71, 31% variance; eating with opportunity to capability 0.82, capability to motivation 0.95, motivation to behaviour 0.48, 23% variance
Final structural models from Willmott et al. 2021, Tables 5 and 6 (CC BY 4.0), redrawn in R. Numbers on arrows are standardised weights.

The model explained 31% of the variance in physical activity and 23% in eating behaviour. In the physical activity model, capability and opportunity were linked to behaviour through motivation (weights 0.91 and 0.11, with motivation to behaviour 0.71). In the eating model, capability was linked to behaviour through motivation (0.95 and 0.48), and opportunity reached motivation through capability (0.82).2 In both models the authors removed the direct paths from capability and opportunity to behaviour after adjusting the structural model using modification indices.2

Multiplying the published weights along each route gives an indirect path to behaviour. We computed these in R.

Bar chart of indirect paths to behaviour through motivation: capability 0.65 and opportunity 0.08 for physical activity; capability 0.46 and opportunity 0.37 for eating
Indirect paths computed by Editverse as products of the published standardised weights (Willmott et al. 2021, CC BY 4.0). Approximate, because the inputs are rounded.

Capability had the stronger indirect path in both models (0.65 against 0.08 for physical activity, and 0.46 against 0.37 for eating). Treat these as a way to compare routes, not as shares of variance: in the physical activity model 0.71 squared is 0.50, which does not match the 31% the authors report, so we use the products only for comparison.

What the case shows and what it does not

  • It shows that a COM-B model built from TDF-based measures can be fitted to survey data in two behaviours and that it explained 23% to 31% of the variance.2
  • The authors state that this compares favourably with variance reported for the theory of planned behaviour (14% to 24%) and for several social psychological theories in adolescent physical activity (24% to 35%).2 That comparison is theirs, across different studies.
  • It does not show cause and effect. The design is cross-sectional, the samples are two convenience samples that were mostly female and tertiary educated, and the measures are self-reported.2
  • It does not test an intervention. Nothing in the paper shows that changing a COM-B source changes behaviour.

How to Use COM-B in Five Steps

  1. 1Specify the target behaviour. Name who does what, where and when. Michie and colleagues start the design task from the behavioural target.
  2. 2Ask what is missing. For the target behaviour, ask for each of the six sources whether there is a deficit. One behaviour may need a change in only one source, another in all three.
  3. 3Add detail where you need it. The Theoretical Domains Framework gives a more granular view of psychological capability and reflective motivation, and its guide describes how to apply it.
  4. 4Shortlist functions. Use Table 2 to see which of the nine intervention functions address each deficit you found.
  5. 5Plan the test. Decide how you will measure the change and whether a trial is feasible. COM-B describes and diagnoses; it does not show that an intervention works.

Sources for the steps.1,3

A worked example: from a deficit to intervention functions

This is an Editverse illustration of how to read Table 2, not a case from the paper. Suppose an assessment of a target behaviour finds a deficit in one source. Table 2 then lists the functions linked to that source, and the authors give an example of each function.1

Swipe sideways to see every column.

If the deficit is inTable 2 links it toAuthors’ example of each function
Psychological capabilityEducation, training, enablementEducation: providing information to promote healthy eating. Training: advanced driver training to increase safe driving. Enablement: behavioural support for smoking cessation.
Physical opportunityRestriction, environmental restructuring, enablementRestriction: prohibiting sales of solvents to people under 18. Environmental restructuring: on-screen prompts for GPs to ask about smoking behaviour. Enablement: behavioural support for smoking cessation.

Links from Michie et al. 2011, Table 2; examples from their Table 1.1

The table narrows the choice; it does not make it. The authors of the young-adult study state that the barriers and enablers they identified for physical activity and eating represent potential targets for future intervention design.2

Linking COM-B to the Theoretical Domains Framework

The TDF was built by behavioural scientists and implementation researchers from 128 theoretical constructs drawn from 33 theories. After a validation exercise the second version has 14 domains covering 84 constructs.3 Atkins and colleagues describe how the TDF has been linked to COM-B, and how it provides a more granular understanding of psychological capability and reflective motivational processes.3 The 2021 young-adult study also mapped TDF domains onto the three COM constructs to choose its survey measures.2

Need the Analysis Run or Checked? Consult Editverse

Editverse subject-matter experts work to COPE, ICMJE and GPP 2022 guidance, with institutional and country-specific ethics followed.

Editverse logo

Editverse

Study Design Consultation

Building a behaviour-change study around COM-B?

You send: your target behaviour, population and the COM-B or TDF constructs you plan to measure.

You get: a design, measures and analysis plan that links each construct to a measurable indicator.

When: You have a behaviour to explain or change and no analysis plan yet.

Study Design Consultation · $400

Plan my study →

Editverse logo

Editverse

Qualitative Data Analysis

Interviews to code against COM-B or TDF domains?

You send: your interview or focus-group transcripts and your coding framework.

You get: coded themes mapped to capability, opportunity and motivation, with a clear audit trail.

When: Your transcripts are collected and need systematic coding.

Qualitative Data Analysis · $600

Code my interviews →

Editverse logo

Editverse

Statistical Analysis

Testing a COM-B model with survey data?

You send: your survey data file and your hypotheses.

You get: analysis run and reported in SPSS, R or STATA, with tables and methods text ready for your paper.

When: Your data are collected and the model still needs fitting.

SPSS/R/STATA Statistical Analysis · $400

Analyse my data →

Free Research Tools from Editverse

Frequently Asked Questions

What is the COM-B model?

COM-B is a model in which behaviour (B) arises from three interacting conditions: capability, opportunity and motivation. Michie, van Stralen and West proposed it in 2011 as the hub of the Behaviour Change Wheel, after reviewing 19 frameworks of behaviour change interventions.

What do the letters in COM-B stand for?

C is capability, O is opportunity, M is motivation and B is behaviour. Michie and colleagues define capability as the individual’s psychological and physical capacity to engage in the activity, opportunity as all the factors outside the individual that make the behaviour possible or prompt it, and motivation as all the brain processes that energise and direct behaviour.

What is the difference between COM-B and the Behaviour Change Wheel?

COM-B is the centre of the wheel. Around it sit nine intervention functions that address deficits in capability, opportunity or motivation, and around those sit seven policy categories that could enable the interventions.

What are the nine intervention functions?

Education, persuasion, incentivisation, coercion, training, restriction, environmental restructuring, modelling and enablement. Table 2 of the 2011 paper links each one to the COM-B sources it can address.

Does the COM-B model explain real behaviour?

In one cross-sectional survey of young adults aged 18 to 35, the COM-B model explained 31% of the variance in physical activity (n = 582) and 23% of the variance in eating behaviour (n = 455). The samples were convenience samples, mostly female and tertiary educated, and the data were self-reported.

How is COM-B related to the Theoretical Domains Framework?

The TDF has 14 domains covering 84 theoretical constructs in its second version. Atkins and colleagues note that it gives a more granular understanding of psychological capability and reflective motivation than COM-B alone. The 2021 young-adult study mapped TDF domains onto the three COM constructs to choose its measures.

How do I use COM-B in a study?

Define the target behaviour, then ask for each of the six sources whether there is a deficit that stops the behaviour. Michie and colleagues describe the task as considering what the behavioural target is and which components of the behaviour system would need to change to achieve it. Then use Table 2 to shortlist intervention functions.

Does COM-B prove that an intervention will work?

No. The authors of the 2011 paper state that research is needed to establish how far the wheel can lead to more efficient design of effective interventions. COM-B helps you diagnose and describe; trials are needed to test whether an intervention works.

References

  1. Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6:42. Open access, CC BY 2.0; Table 2 redrawn and counted by Editverse in R. doi.org/10.1186/1748-5908-6-42
  2. Willmott TJ, Pang B, Rundle-Thiele S. Capability, opportunity, and motivation: an across contexts empirical examination of the COM-B model. BMC Public Health. 2021;21:1014. Open access, CC BY 4.0; Tables 5 and 6 redrawn, products computed by Editverse in R. doi.org/10.1186/s12889-021-11019-w
  3. Atkins L, Francis J, Islam R, et al. A guide to using the Theoretical Domains Framework of behaviour change to investigate implementation problems. Implement Sci. 2017;12:77. Open access, CC BY 4.0. doi.org/10.1186/s13012-017-0605-9

Editverse Research Team | Updated October 2026