Good intervention design starts with good diagnosis

Imagine visiting your doctor with persistent headaches. You wouldn't be satisfied if they simply diagnosed "physical health problems" and prescribed a general treatment - you'd expect them to investigate whether your headaches stem from tension, migraines, eye strain, or other specific causes, because the right treatment depends on the precise diagnosis.

READ ARTICLE ON SUBSTACK

Yet in behaviour change, we often settle for broad diagnoses. Models like COM-B make it tempting to stop at high-level categories: identify whether the problem lies in people's capability, opportunity, or motivation, and design interventions to address it. But just as in medicine, the real challenge lies in diagnosing behaviour accurately enough to design treatments that work.

Successful behaviour change requires more than identifying barriers at a broad level. By digging deeper into subdomains like attention, memory, and self-regulation, we can uncover the real challenges that prevent people from making better choices. Using the familiar calorie labelling case, we'll explore why being specific matters and how mistakes in this foundational stage can derail interventions - just as surely as a misdiagnosed medical condition leads to ineffective treatment.

When frameworks oversimplify

Broad diagnoses are where many interventions stumble. COM-B is a great starting point, but staying at this high level risks oversimplifying the problem. Just as 'cardiovascular issues' could mean anything from high blood pressure to heart disease, 'capability barriers' can hide several distinct challenges.

Let's take Capability as an example: when we diagnose it as a barrier, what does that really mean?

• Is it about knowledge gaps - a lack of understanding?
• Or attention - whether people notice the information?
• Or memory - whether they recall it at the right moment?
• Or even self-regulation - whether they can act on their intentions?

These are fundamentally different challenges, each requiring different solutions - just as different heart conditions require different treatments. Treating "Capability" as a single problem might feel efficient, but it leads to generic solutions that don't target the real barriers.

This is where the Theoretical Domains Framework (TDF) becomes valuable. Building on COM-B's foundation, the TDF breaks these broad categories into 14 specific domains and 84 constructs. It's like moving from general medical specialties to specific diagnostic tools - helping us see the subtle but crucial differences between seemingly similar barriers.

From symptoms to treatment

The calorie labelling case shows why precision matters. Just as a general practitioner might miss a specific condition without detailed investigation, this intervention focused solely on knowledge (a subdomain of Capability), assuming that providing calorie information would change behaviour. It succeeded in raising awareness but missed other critical diagnostic clues:

• Attention: Calorie labels often go unnoticed in fast-paced, distracting dining environments. If diners don't see the information in the first place, no amount of education about healthy eating will help. As a solution, we might make calorie information harder to ignore by positioning it prominently on menus, using visual cues like icons or bold colours, and placing reminders at key decision points.
• Memory: Even when diners notice calorie information, they face a different challenge - remembering and applying it when ordering. Seeing a calorie count of 850 means little without remembering the context of daily requirements or previous meals. We could anchor calorie counts against meaningful reference points like "average meal calories" or "recommended daily limits" to help people process and retain the information they need at the moment of choice.
• Self-Regulation: Then comes perhaps the toughest barrier - even with perfect attention and memory, habits and impulses often override intentions, especially in indulgent dining settings. Here, we might need tools to support self-regulation, like goal-setting prompts ("Set a calorie goal before ordering"), pre-commitment options ("Pre-order from a healthier menu online"), or tracking tools to bridge the gap between intention and action.

At this point, you might be thinking "isn't this making things needlessly complex?"

It's a fair challenge! The appeal of broad frameworks like COM-B lies in their simplicity - they're easy to explain and quick to apply, but keeping things simple comes at a cost. Without precise diagnosis, we might cycle through multiple iterations: first making labels bigger (attention), then adding guidelines (memory), then introducing planning prompts (self-regulation) with each revision costing time and resources.

In contrast, precise initial diagnosis helps us target the right barriers from the start. The key is finding the right level of precision - detailed enough to guide effective design, but practical enough to implement.

Think of it like medical diagnosis: while we don't need to understand every biological process, we do need to distinguish between a viral and bacterial infection to prescribe the right treatment.

Different barriers, different solutions

Once we recognise that Psychological Capability is more than just knowledge, the next step is to reframe the barriers and design interventions that address them effectively. Let’s look at what the calorie labelling intervention missed and how we could potentially fix it:

• Attention: Calorie labels often go unnoticed in fast-paced, distracting dining environments. As a solution, we could make calorie information harder to ignore by positioning it prominently on menus, using visual cues like icons or bold colours, and placing reminders at key decision points.
• Memory: Even when diners notice calorie information, they may forget daily guidelines when choosing what to order. In this case, we could anchor calorie counts against meaningful reference points such as “average meal calories” or “recommended daily limits” to help people process and retain the information.
• Self-Regulation: Habits and impulses often override intentions, especially in indulgent dining settings. We could also create interventions to support self-regulation by giving people tools to manage their behaviour, like goal-setting prompts (“Set a calorie goal before ordering”), pre-commitment options (“Pre-order from a healthier menu online”), or tracking tools (e.g., apps that log calorie intake).

By reframing the barriers this way, we shift from diagnosing the problem to designing precise, actionable solutions. Awareness is only the start—attention, memory, and self-regulation are what turn knowledge into action.

Note: I have deliberately kept the example interventions simple to illustrate the ideas, even if they are not all feasible solutions in this case.

Making theory work in practice

While broad frameworks like COM-B give us a starting point for intervention design, getting specific about psychological barriers can mean the difference between success and failure. When we rely on general diagnoses, we risk designing solutions that miss crucial aspects of behaviour.

Our calorie labelling example shows the value of breaking down broad categories like Psychological Capability. While we know that providing information alone had limited impact, we've used this case to explore how different barriers might need different solutions. This is just a thought exercise, but it demonstrates why looking deeper than surface-level diagnosis could improve intervention design.

To put this into practice, we should:

• Look beyond broad categories to identify specific psychological mechanisms
• Consider how different barriers might interact and overlap
• Use frameworks like TDF to understand the subtleties of real-world behaviour

These steps might seem demanding, but they're an investment in getting intervention design right. Just as doctors don't prescribe treatment without a clear diagnosis, we shouldn't design interventions without understanding the specific barriers to change.

Good intervention design starts with good diagnosis.

Previous articles in this series:

1. Does knowledge change behaviour? Lessons from calorie labelling
2. Deconstructing the logic of calorie labelling as an intervention - How to use reverse engineering to refine behaviour change strategies
3. Why interventions fail: a guide to common pitfalls in behaviour change. Calorie labelling shows how behaviour change interventions can go wrong and how we can learn to anticipate problems.