Key Principle
COM-B is a diagnostic framework that maps all possible root causes of a behavior gap into six sub-factors before any intervention is designed:
Capability
- Physical Capability: Physical skills, strength, or stamina required to perform the behavior
- Psychological Capability: Knowledge, cognitive skills, emotional regulation, reasoning ability
Opportunity
- Physical Opportunity: Environmental features, resources, time, and tools that make the behavior possible
- Social Opportunity: Cultural norms, social cues, language, and role models that enable the behavior
Motivation
- Reflective Motivation: Conscious intentions, beliefs about outcomes, identity-based commitments
- Automatic Motivation: Habits, emotional associations, impulses, and conditioned responses
A behavior occurs when all relevant sub-factors are sufficiently present. A behavior gap means at least one sub-factor is deficient. COM-B forces diagnosis before prescription.
The Picture or Video Test: Verify behavioral specificity before diagnosing. Ask "If I took a picture or video of someone doing this behavior, what would I see?" If the answer is fuzzy, the behavior is still an intention or outcome — not a specific, designable behavior.
Why This Matters
Training addresses only Physical and Psychological Capability. If the root cause is in Opportunity (environment is broken, no social norm, no time to perform the behavior) or Motivation (habit, identity conflict, emotional barrier), training will have no effect regardless of quality. The COM-B diagnostic prevents the reflex training prescription.
The differential diagnosis principle: before selecting any intervention, identify which of the six sub-factors is deficient. This is not intuition — it requires research with the target audience. The most common diagnostic error is assuming Psychological Capability is the bottleneck when Automatic Motivation, Social Opportunity, or Physical Opportunity is actually the constraint.
Good Examples
Behavior specification — "wash hands properly" (Chapter 6): This fails the Picture/Video Test. "Properly" is not observable. The observable behavior is: "Use soap, lather for 20 seconds covering all surfaces, rinse under running water, dry with a clean towel before leaving the patient room." Only after specifying this can you assess each COM-B sub-factor.
Patient communication in healthcare (Chapter 7): Clinicians who fail to explain diagnoses clearly are often diagnosed as lacking communication skill (Psychological Capability). But the actual diagnosis for experienced clinicians is frequently tacit knowledge: they have automated their own understanding to the point where they no longer perceive what is missing for the patient. This is a different COM-B sub-factor requiring a different intervention — not instruction on communication, but deliberate practice at the patient's mental model level.
Handwashing triage (Chapter 7): A hospital handwashing COM-B analysis might find: Physical Capability is present (everyone can wash hands); Psychological Capability is present (everyone knows why); Physical Opportunity is absent or marginal (sinks inaccessible, time pressure); Automatic Motivation is absent (no habit cue in the physical environment). The correct interventions are Physical Opportunity restructuring (sinks, gel dispensers at point of care) and Automatic Motivation support (environmental cuing) — not more hand-hygiene training.
Behavior prioritization with APEASE (Chapter 6): When a COM-B analysis produces a long list of candidate behaviors, filter by: Acceptability (will stakeholders accept this intervention?), Practicability (can it actually be implemented?), Effectiveness (evidence it works?), Affordability, Spill-over effects (positive or negative on other behaviors?), Equity (does it reach all affected groups?).
Positive Deviance research (Chapter 6): Identify performers who succeed at the target behavior with the same or fewer resources as average performers. Their strategies are the most transferable models because their success is not explained by exceptional advantages. Star performers may succeed through resources, relationships, or traits unavailable to others.
When Training Is NOT the Answer (Chapter 7 Decision Tree)
Training is not the correct intervention when the root cause is:
- Physical Opportunity: Broken process, missing resources, insufficient time, inaccessible tools → fix the environment
- Social Opportunity: No cultural norm, no role models, counter-normative peer behavior → change the social system
- Automatic Motivation: Habit conflict, emotional association, conditioned fear response → habit change BCTs, exposure, desensitization
- Values/Identity misalignment: Behavior conflicts with self-concept → identity-based BCTs, reframing
- Person-role mismatch: Person genuinely does not value the role's requirements → this may be a selection/fit problem ("wouldn't it have been easier to hire a squirrel?")
- Structural incentives: Organizational rewards actively reinforce the opposite behavior → policy/incentive change required
Counterpoints
"We asked them what they need and they said training" — people request training because it is the most familiar intervention, not because it is the correct diagnosis. Stakeholder-requested training is a symptom to investigate, not a prescription to fulfill.
"The training scores are high, so capability must not be the problem" — assessment scores measure Rider-level knowledge, not automatic behavioral execution. High scores with low performance means the gap is in Automatic Motivation, Physical Opportunity, or Social Opportunity — not knowledge.
"This is obviously a skills gap" — the most common diagnostic error. Fluency (automaticity under real-world conditions) is a different problem from knowledge. An experienced clinician who "knows" the communication protocol but cannot deploy it under time pressure or emotional load has a different gap than a new hire who never learned it.
Key Quotes
"Before we do anything else, we need to figure out why the people involved are not engaging in the target behavior." — Julie Dirksen, Chapter 6: Analyzing Behaviors
"If training is the answer, what is the question?" — Julie Dirksen, Chapter 7: Determining if It's a Training Problem
Rules of Thumb
- Never start with intervention design. Start with COM-B diagnosis.
- Apply the Picture/Video Test to every candidate behavior before conducting any analysis.
- Research positive deviants before designing any training solution.
- For each COM-B sub-factor, ask: "What evidence do we have that this is or isn't deficient?" Don't assume.
- If training has been prescribed already, run COM-B anyway. Use it to identify what the training will and won't fix.
- Use APEASE to prioritize when your behavior list is longer than you can address.
Related References
- Mapping COM-B to Intervention Types and BCTs — COM-B sub-factors map to specific intervention types and BCT clusters
- The Rider/Elephant Framework and Systems View — Why COM-B diagnosis is necessary (Rider/Elephant gap, systems view)
- Implementation Playbook: From Diagnosis to Deployment — How to conduct a COM-B diagnostic interview in practice