A town wants residents to use a free preventive service. It publishes another explanation of why the service matters. The brochure is accurate, yet appointments remain low because the clinic is open when many people work, the booking route is confusing and transport is unreliable. A question about behavior has been answered as if it were only a question about knowledge. Dolores Albarracín and colleagues’ 2024 review asks a larger question: which determinants of behavior are worth targeting, and what happens when interventions try to change them?
The review matters because popular behavior programs often begin with the most visible individual explanation. They ask people to know more, want more or try harder. The authors synthesize meta-analytic evidence spanning individual and social-structural determinants and the interventions aimed at each. Their ranking points toward practical barriers and access, while also revealing a central limitation of all rankings: an average across varied studies is not a diagnosis of one person or place.
What the 2024 review set out to compare
Albarracín, Fayaz-Farkhad and Granados Samayoa published their review in Nature Reviews Psychology. It draws together multidisciplinary meta-analyses rather than recruiting one new group of participants for an intervention. The authors examine individual determinants such as knowledge, beliefs and habits, and social-structural determinants such as norms, support and access. They ask which kinds of interventions show more behavioral impact across the available syntheses.
This is a difficult comparison. Different literatures measure different behaviors, populations, follow-up intervals and outcomes. One study may count vaccine uptake, another energy conservation, another attendance. A comparison across them can identify recurring patterns, but a single ordered list should not be read as a universal league table. Intervention labels are broad: one “knowledge” message may be clear and timely while another is irrelevant; one access policy may remove a genuine barrier while another reaches few people.
The review’s contribution is partly conceptual. It places targets of change beside the evidence that interventions aimed at them actually alter behavior. A factor may predict an outcome without being an effective intervention target. People who know more about a service may use it more, but giving everybody an information leaflet may do little when the appointment system remains inaccessible. Conversely, changing a practical opportunity may work even if attitudes barely move.
That distinction matters for any organization tempted to infer a solution from a correlation. An observed association between motivation and attendance does not reveal whether a motivational speech would change attendance. The 2024 authors therefore compare intervention evidence as well as determinants. Their synthesis cannot settle causality for every specific intervention, but it improves the question a team should ask before choosing one.
The pattern for individual targets
In the review’s cross-domain ordering, interventions aimed at knowledge appear toward the lower end of behavioral impact. General skills, general attitudes and beliefs follow, with stronger average patterns for emotions, behavioral skills, behavioral attitudes and habits. The point is not that accurate information is useless. People need information to make informed choices, especially when a service is unfamiliar or a claim is uncertain. The point is that information alone often leaves the act itself difficult.
Behavioral skills differ from general encouragement. A person may endorse a goal but lack the specific steps needed at the moment of action. A booking service can teach how to navigate the form, while an accessible redesign can reduce the skill burden altogether. An intervention should identify the actual behavior and its setting before deciding whether the missing ingredient is knowledge, practice, confidence or opportunity.
Habit-targeting interventions score relatively strongly in the review’s synthesis, but that does not mean every routine should be automated. Wood and Rünger’s habit review describes how repeated responses in recurring contexts can become efficient default actions. Lally and colleagues’ longitudinal field study showed variable growth in self-reported automaticity for selected new daily behaviors. These sources explain one possible mechanism behind habit interventions; they do not establish that a generic reminder creates a habit in every setting.
There is also an ethical question about whose habit is being built. A person may want a reliable medication routine or a regular language-practice slot. A platform may want daily use regardless of the person’s purpose. The same cue-design technique can serve different interests. The 2024 review compares behavioral impact; it does not certify that the target action is beneficial, consented to or fairly distributed.
The pattern for social and structural targets
The review also orders interventions aimed at social-structural determinants. In the authors’ cross-domain synthesis, policies increasing access to the behavior occupy the strong end, alongside social support provision and material incentives. Descriptive norms, monitors and reminders, injunctive norms, institutional trustworthiness programs, and legal or administrative sanctions appear elsewhere in the ordered pattern. These are meta-analytic summaries, not instructions to apply the strongest-sounding category indiscriminately.
Access has a plain meaning in the town’s clinic example. A service can extend hours, simplify booking, provide a phone alternative and make the building reachable. If the obstacle is practical opportunity, a message asking residents to care more misidentifies the problem. Access changes may also be more equitable, because they help people who already intend to act but cannot navigate the current route.
Support can matter differently. A trusted person may accompany someone to an unfamiliar service, translate its process or help with childcare. Social support is not identical to peer pressure. A program that publicly ranks people by attendance may produce visible conformity without making the service more feasible. The review’s categories help distinguish interventions; local evaluation still needs to ask what kind of relationship is created and who bears the work.
Institutional trustworthiness is especially easy to misread. A campaign that says “trust us” is not the same as making a service more worthy of trust. Clear eligibility rules, competent staff, accessible complaints and reliable follow-through may change whether an institution deserves confidence. The review includes trustworthiness among structural targets, but it does not imply that a better slogan can substitute for better institutional practice.
Material incentives and sanctions raise further tradeoffs. An incentive may reduce a real financial barrier, or it may make a voluntary action feel like a transaction. A sanction may compel compliance while hiding resentment, unequal burdens or avoidance. The review’s impact ordering cannot answer whether such tools are proportionate, just or sustainable. These judgments require domain evidence and participation by affected people.
Compare the review with the behaviour change wheel
Michie, van Stralen and West’s behaviour change wheel offers a complementary diagnostic model: capability, opportunity and motivation interact in the behavior under study. The 2024 review is not a replacement for diagnosis. It asks whether certain targets have tended to yield behavior changes across meta-analyses; the wheel asks what capability, opportunity and motivation conditions are present for a specific action. Neither can tell a service team what residents face without talking to them and observing the route.
If the clinic has inconvenient hours, the opportunity diagnosis points to schedules and transport. If people cannot understand the eligibility criteria, capability and communication matter. If the service has a history of disrespectful treatment, motivation may be shaped by experience and institutional trustworthiness. These conditions can coexist. A team should not choose one box because a pooled rank looks impressive while leaving other barriers untouched.
The review’s emphasis on circumventing obstacles is a useful corrective to individual blame. It can also be oversimplified into “knowledge never matters,” which the evidence does not say. Accurate knowledge may be necessary for informed consent even when it has a smaller average effect on behavior. A respectful program asks both whether a person can act and whether they understand enough to choose. The two outcomes should be measured separately.
Method and inference limits
This paper is a review of meta-analyses, so its unit of analysis is already a synthesis rather than a single randomized trial. That allows breadth across fields. It also compounds differences in how the underlying studies defined an intervention, selected participants, measured behavior and handled bias. An apparent ordering may depend on which literatures were available and which outcomes were compared. The effect of a category in one policy setting should not be assumed to transfer unchanged to another.
A challenge with any behavioral outcome is that action can change for reasons the program did not intend. A person may use a clinic because a relative can drive them once, not because the scheduling redesign works broadly. An intervention may increase sign-ups but not attendance. An incentive may alter short-term behavior but not persist when removed. Strong evaluation records the immediate outcome, later continuation, the conditions of participation and possible adverse effects.
The authors’ review is also about efficacy, not a complete ethical assessment. More use of a service can be good if the service is safe and desired. It can be harmful if it displaces a better option, collects unnecessary data or makes refusal costly. A behavioral target should be justified in public terms and open to revision. The Institute’s application is to join intervention evidence with an agency review, rather than letting behavior change itself become the sole definition of success.
Publications in this area may be affected by selective reporting and by the practical difficulty of studying long-term structural change. A brochure can be randomized and measured quickly; changing opening hours, staff practices or transport may require institutions to coordinate over months. A literature count can therefore reflect ease of study as well as social importance. The 2024 paper’s pattern is best used to prioritize investigation, not to end it.
What changed after this synthesis
The strongest implication is a shift in sequence. Instead of beginning with “How do we persuade residents?”, a team can begin with “What would make the desired action possible and worthwhile for residents who want it?” It can describe the action, map barriers, select an intervention that addresses a real constraint and compare outcomes with an appropriate alternative. That workflow is an Institute synthesis built from the review and the behaviour change wheel, not a protocol validated by the review itself.
For the clinic, a first test could compare an extended-hours pilot with the existing schedule while keeping information accurate in both. It could track appointments made, appointments completed, travel burden and who remains excluded. Qualitative interviews could reveal barriers the numbers miss. If the pilot helps one group but not another, the team can revise the route before expanding it. More appointments alone would not prove equitable access.
A second test might examine whether booking instructions are understood, especially for people using another language or assistive technology. If the instructions are clear but the appointment slots remain unavailable, another leaflet is unlikely to solve the barrier. If slots are available but the eligibility criteria are confusing, better information is not a trivial add-on; it is part of meaningful access. The ranking helps prompt a diagnostic question, not silence it.
Why this matters in Northern Europe
Public services in Northern Europe often combine health, education, transport and digital systems. A behavioral intervention in one part can be constrained by another: an online booking process, a transit timetable, a language policy or a work schedule. The 2024 review supports attending to access and support, but it did not directly test a Nordic service reform. Local teams must check which barriers actually exist and whether a proposed change reaches those most affected.
A human-centered service can make its assumptions visible. It can state whose behavior it wants to change, why the action matters, what alternatives people have and which costs might move to households or staff. A seemingly efficient digital prompt may save agency time while requiring the resident to make multiple calls. Evaluating both sides of that exchange brings the structural determinants into view.
Behavioral evidence is most useful when it improves institutions’ ability to respond, not merely their ability to steer people. A public body can remove a needless form field, adjust opening hours, provide a human route and test whether these changes help people complete the task they came to do. That is a policy and design choice inspired by the review, not a claim that its pooled rankings certify one specific reform.
What the evidence does not show
Albarracín and colleagues do not show that every knowledge intervention is weak, every habit intervention is strong or every access policy is appropriate. Their ordered categories summarize varied meta-analytic findings across domains. The ranking does not diagnose an individual, identify the cause of one town’s appointment gap or prove that a particular intervention will work in a new population.
The review does not establish that the most behaviorally effective option is the most ethical. It does not remove the need for informed choice, trustworthy institutions, fair workload or a route to refuse. Nor does it prove that structural change can always be delivered quickly or that all barriers are external. People differ in aims and circumstances. The stronger question is which combination of capability, opportunity, motivation and support matters for the people and action under study.
The cited habit and behaviour change wheel papers explain related mechanisms and diagnostics, but they are not direct replications of the 2024 review’s cross-domain ordering. The Institute’s suggestions about pilots, access checks and measuring unequal burdens are a practice framework. They need to be tested with residents and staff rather than presented as outcomes already demonstrated by this source.
Return to the clinic
The town can still publish a clear explanation. People deserve to understand the preventive service and its limits. But the explanation should accompany an honest look at what happens when someone tries to book. If the phone is unanswered, the bus does not run and the only appointment falls during work, the behavior has not been made possible by improving the brochure.
This 2024 review invites a change in professional attention: from the most convenient message to the determinant that actually constrains action. The result is not a universal recipe. It is a better starting question for design, research and public accountability, especially where the person has already decided that the action matters and needs an institution to meet them halfway.
Research references
- Albarracín, D., Fayaz-Farkhad, B., & Granados Samayoa, J. A. (2024). Determinants of behaviour and their efficacy as targets of behavioural change interventions. Nature Reviews Psychology, 3, 377–392. View source
- Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. View source
- Wood, W., & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology, 67, 289–314. View source
- Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. View source
- Hollands, G. J., Shemilt, I., Marteau, T. M., et al. (2013). Altering micro-environments to change population health behaviour: Towards an evidence base for choice architecture interventions. BMC Public Health, 13, 1218. View source