Multilevel Theories and
the Ecological Perspective
Health Promotion: Individuals and Communities
Learning objectives for this lesson:
- Explain why theories confined to a single ecological level cannot account for how influences at different levels act together, and contrast the equation-like form of single-level theory with the path-like form of multilevel theory.
- Describe Bronfenbrenner's Ecological Systems Theory, classify influences into its five systems, and explain the bioecological amendment and the distinction between proximal and distal processes.
- Explain the two dimensions of Flay's Theory of Triadic Influence, levels of causation and streams of influence, and use the waterfall metaphor to choose a feasible point of intervention.
- Compare the problem-specific multilevel models of the CDC, WHO, and UNICEF with the general theories, and evaluate multilevel theory using its two critiques: complexity and the placement of factors at particular levels.
- Describe the three components common to any program planning process, and use importance and changeability ratings to set priorities among candidate factors.
- Describe the nine phases of PRECEDE-PROCEED, classifying factors as predisposing, enabling, or reinforcing and distinguishing process, outcome, and impact evaluation.
- Apply the risk and protective factors planning model to an adolescent health problem by sorting factors into its domains, and compare the model with PRECEDE-PROCEED.
This course was developed by Dr. Kiffer G. Card, to accompany Edberg, M. Essentials of Health Behavior: Social and Behavioral Theory in Public Health. Jones & Bartlett Learning. This lesson follows Chapters 7 and 8 of the text.
Glossary: Key Terms, People & Concepts
📚 Reference page, available throughout the lesson
This glossary collects the key concepts, people, and ideas you will meet in this lesson. Use it as a reference while you work through the material, or as a review before assessments. Type in the search box to filter entries.
Why Multilevel? Bronfenbrenner's Ecological Systems Theory
⏱ Estimated reading time: 16 minutes
Overview
Many theories of health behaviour work mainly at one ecological level: the individual, the social group, the organization, the culture, the political economy. The first half of this lesson asks what happens when we admit that those levels act on each other, and it introduces theories built to describe that interaction. The second half then asks the practical question: once you accept that influences are layered, where do you start? This first section makes the case for multilevel theory, describes the first and most influential multilevel theory, Urie Bronfenbrenner's Ecological Systems Theory, follows it through a story and into school health programs, and closes with the critique that follows every theory in this lesson.
Learning Objectives
- 1.1 Explain why theories confined to a single ecological level cannot account for how contributing factors at different levels work together.
- 1.2 Contrast the linear, equation-like form of single-level theory with the path-like form of multilevel theory.
- 2.1 Describe Bronfenbrenner's five systems (microsystem, mesosystem, exosystem, macrosystem, chronosystem) and classify real influences into them, using the participation test to separate the mesosystem from the exosystem.
- 2.2 Explain the bioecological amendment and distinguish proximal processes from distal processes.
- 2.3 Trace a case through Bronfenbrenner's systems, and explain how the level at which an intervention acts shapes its outcome.
- 2.4 Map the Coordinated School Health Program and Canada's Comprehensive School Health framework onto Bronfenbrenner's systems.
- 4.1 State the main critique of Ecological Systems Theory, complexity, and the two questions it raises for research and intervention.
1.1 Why One Level Is Not Enough
Start with Albert Einstein, who wrote that a theory is more impressive the simpler its premises, the more different kinds of things it relates, and the wider its area of application. Einstein's general theory of relativity revolutionized physics, but it dealt primarily with gravity, one of the fundamental forces understood at the time alongside electromagnetism. He spent years trying to build a unified theory that explained both forces together, and he did not succeed. The story is useful here because health behaviour theory faces a similar problem. Each of the theories reviewed so far in this course focuses on the drivers of behaviour at a single ecological level, in the same way that relativity focused mainly on one force. The question this raises is whether behaviour in the real world is actually produced one level at a time, and the honest answer is that it is not likely.
It helps to think of the ecological levels as a filing system. By this we mean that dividing explanatory theories by level (individual, social group, organization, culture, political economy) is a way of organizing them so that we can see what each theory refers to; it is not a claim that the levels operate separately in people's lives. The purpose of sorting theories this way is clarity. However, it is generally understood that factors at one level interact with factors at other levels, even though most theory does not account for that interaction. For example, suppose a theory connects the social dynamics of a family to the attitudes and behaviours of the children in it. That theory may explain the children's behaviour well, but it leaves open what accounts for the family dynamics themselves, which may be shaped by socioeconomic constraints, low community efficacy, racism, gender roles, and more. A theory confined to the family level cannot say how those outside factors and the family work together to produce what the children do. The question for this lesson is how to account for that combined effect. Theorists often shy away from the attempt, and that is no surprise, because it is a very difficult challenge. If, however, the ecological model is the field's basic understanding of the context within which behaviour occurs, then theory has to take the challenge on rather than stay pigeonholed in specific levels.
If a theory confined to one level cannot capture how the levels act on each other, the next question is what a theory that does capture it would look like.
1.2 Equations Versus Paths
The two kinds of theory differ in the form their explanations take. Single-level theory is typically framed in somewhat linear terms, almost like an equation. By this we mean that the theory lists a set of factors at one level and treats them as adding together to produce a likelihood of behaviour, with influence running in one direction only. For example, in the Health Belief Model a person's perceptions add up to a likelihood of action; the Transtheoretical Model and diffusion of innovations are framed in a similar way. Multilevel theory, in contrast, is more likely to be framed as a path, or multiple paths, through the ecological layers. In other words, it describes how an influence at one layer, such as the economy, travels through other layers, such as a workplace and then a family, before it reaches the person. Those paths may be complex, branching like trees or networks of dendrites, and they may run in more than one direction: a behaviour change at the individual level may reverberate back up through the broader levels and change the character of the influence pathway itself. Because this can get abstract, Figure 6.1 puts the contrast in a single picture. Its left half draws a single-level theory as two beliefs and a cue added together into behaviour; its right half draws branching paths from culture, law, and economy down through workplace, services, and institutions and through family, school, and peers to behaviour, with a dashed path that reverberates back up.
Figure 6.1. The contrast in one picture: single-level theories read like equations; multilevel theories read like paths, trees, or dendrites, and the paths can run back upward.
When you meet a new theory, the form of its explanation is a quick test of its type. If it adds a set of factors at one level into a likelihood of behaviour, it is a single-level theory; if it traces influence from one layer to another, allows the paths to branch, and lets a change in behaviour feed back upward, it is a multilevel theory. The case for multilevel theory is now in place, and the rest of this section examines the first and most influential example, Bronfenbrenner's Ecological Systems Theory.
2.1 Bronfenbrenner's Five Nested Systems
This has to be the first multilevel theory examined, because it contributed so much to the ecological model in public health. Urie Bronfenbrenner (1917 to 2005) was a Russian-born American psychologist who focused on child development and was influenced by theorists who regarded human development as an interactive social process (Bronfenbrenner). He also cofounded the Head Start program in the United States, one of the most notable efforts ever made to address the effect of poverty on children. Canada's Aboriginal Head Start programs, launched federally in the 1990s in urban, northern, and on-reserve communities, carry the same idea into an Indigenous context with components spanning culture and language, education, health promotion, nutrition, social support, and parental involvement: a multilevel program in practice.
Most important for this course, Bronfenbrenner developed Ecological Systems Theory, a multilevel systems approach for understanding child outcomes, including behaviour (Bronfenbrenner, 1977). The basic idea is that child development and behaviour can be understood as outcomes of the interaction between an individual and several levels or layers within the surrounding system. By this we mean that a child's behaviour is not produced by the child alone, nor by the environment alone, but by the ongoing exchange between the two, and that the environment itself has several layers, nested one inside the other and running outward from the child.
The innermost layer is the microsystem, the set of relationships and structures with which the child has the most direct, immediate contact. For example, the family, the school, the neighbourhood, and childcare are all microsystem settings, because the child spends time in them and deals with the people in them every day. The mesosystem is the connective tissue between those settings. It is not a new place but the set of links between places the child already belongs to, such as the relationship between a child's teacher and the child's parents. Further out is the exosystem, the level the child does not interact with directly but which still affects the child's well-being. For example, a parent's workplace, social services, and health services make decisions that reach the child, even though the child never takes part in them. The macrosystem is the soup or broth of cultural values, customs, and laws in which all the other levels sit; in addition to acting on the child, it shapes what happens in every inner layer. Finally, the chronosystem adds time. By this we mean that the other systems operate and change over the years, so the same setting can affect a child differently at different ages. Taken together, the five systems ask you to locate any influence by its relationship to the child rather than by its size. The five cards below give the full definition and examples, starting with the microsystem and working outward.
Microsystem
The relationships and structures with which a child has the most direct, immediate contact. Examples: family, school, neighbourhood, and the childcare environment such as daycare or primary school. This is the layer the child touches every day.
Mesosystem
The relationships and structures that provide the connections between components of a child’s microsystem, a kind of "connective tissue." Examples: connections between a child’s teacher and parents; among an organization like the YMCA, parents, and the neighbourhood; among faith organizations, families, and neighbourhoods.
Exosystem
The social system level surrounding the first two systems. An individual child does not interact directly with it, but the child’s development is directly affected, because the exosystem includes structures and units such as the workplace, social services, and health services that have an effect on the child’s well-being.
Macrosystem
A kind of soup or broth that provides a base for, and is integrated with, the other system levels. It includes cultural values, customs, and laws, which can have multiple effects on the levels below. Example: in a society with a pronounced gender hierarchy, this could affect the ability of women to work or to act in a policymaking capacity about what services should be offered in communities.
Chronosystem
The factor of time added to the other system levels: the multiple systems, operating over time. Example: a lack of school resources for young children (mesosystem, exosystem) may change the relationship between those children and the other system components when the children are older, both in what they need and how they behave.
The cards show that each system is defined by the child's relationship to a setting, not by how large or important the setting is. Two of the five systems are confused more often than the others, and the test that separates them is worth committing to memory.
Common confusion: mesosystem or exosystem?
Students regularly file a parent's workplace under the mesosystem because it connects to the family. The test is whether the child takes part in the setting. A mesosystem link joins two settings the child is in, such as a meeting between the teacher and the parents. An exosystem setting is one the child never enters, such as the parent's workplace, the social services office, or the health authority, yet whose decisions reach the child. A shift schedule that keeps a parent away at bedtime is exosystem; a phone call between the daycare and the parent is mesosystem.
Interactive 6.1 applies the five systems to three behaviours: a grade 9 student who vapes, an older adult who walks daily, and a grade 3 child who eats breakfast. Choose each behaviour in turn, then click every ring and the time bar beneath it to see what sits at that level for that behaviour.
📊 Interactive 6.1: The nested systems.
Choose a behaviour, then click each ring (and the time bar) to see what sits at that level for that behaviour. The examples are illustrations of the nested systems, not a fixed taxonomy.Click a ring
Start with the individual at the centre and work outward. Each level's examples change with the behaviour you choose.
Whichever behaviour you choose, the nested structure stays the same while the influences inside it change. That is what makes the five systems a general tool for analysing behaviour rather than a list tied to one problem: to classify any influence, ask where it sits relative to the person and whether the person takes part in it. The original theory placed the person at the centre of the rings but said little about what goes on inside the person, and Bronfenbrenner's later amendment fills that gap.
2.2 The Bioecological Amendment: Proximal and Distal Processes
In the early 1990s and beyond, Bronfenbrenner began to amend the original theory to account for the interaction of the multiple systems with the individual's own biological processes, including psychological, cognitive, and genetic elements (Bronfenbrenner and Ceci, 1994). The original theory had drawn the person as a point at the centre of the rings, acted on by the systems around it. The amendment treats the person as an active participant with an inner life of their own. The systemic process came to be understood as one that involves the interaction of individuals, with all the internal biopsychological processes occurring at a particular developmental stage, and the multilayered environment. By this we mean that the same setting does not have the same effect on everyone, because what a child brings to it (temperament, cognitive ability, genetic makeup, and the stage of development they have reached) shapes how they respond to it. For example, a strict classroom may steady one seven-year-old and frustrate another, and the same teenager may respond to that classroom differently at 13 than at 17. At this point the temporal element, represented by the chronosystem, received a formal place in the model. The purpose of giving time a formal place was to reflect the evolving pattern of interaction as individuals develop and engage in different ways over time with the surrounding layers: as a child grows, both the child and the settings change, so the interaction between them changes too. The amended version is called Bioecological Systems Theory, a name that adds biology to the ecological layers the theory began with.
Adding the individual's own processes led to a distinction you will meet again in Flay's theory in the next section. Proximal processes occur between an individual and their immediate environment. By this we mean the direct, face-to-face exchanges in which the person takes part, such as a conversation, a lesson, or a meal together. Distal processes, in contrast, occur within that wider environment and eventually affect the immediate, proximal interactions. In other words, the person is not part of a distal process, but its effects arrive later through the proximal exchanges it changes. For example, a conversation between a teenager and a parent about vaping is proximal, because the teenager takes part in it directly. A provincial decision to restrict flavoured vaping products is distal: the teenager plays no part in making it, and it reaches her only through what stores stock, what friends can buy, and what the parent has read in the news. The distinction therefore turns on where a process occurs relative to the person, not on how large its effect is or whether it eventually changes what the person does. A distal process can have a large effect, but only by way of the proximal interactions it reshapes, so when you classify a process, ask whether the person takes part in it rather than how much it matters.
With the five systems, the person's own biology, and the two kinds of process in hand, a story shows how they combine over time.
HEAL connection: HEAL beside Bronfenbrenner
The HEAL Framework for Wellbeing introduced in Lesson 1 is also a multilevel model. Like Bronfenbrenner's systems, it places the person inside wider settings; like the bioecological amendment, it includes biology, through alteration (the direct imprint of environments on physiological function) and sensitivity (accumulated disadvantage making the same exposure more harmful); and like the chronosystem, it has a time axis, the life course.
HEAL also names processes that run back up the system: selection (health and disposition sorting people into social positions), compounding consequences (ill health lowering social position, which worsens exposure and access), and social construction (collective behaviour reshaping the sociopolitical context). The main critique in Section 4 applies to HEAL as well: a model this complete is usually tested one pathway at a time.
2.3 Rodney in Ourlandia: The Theory in Motion
A story illustrates the theory in motion, with one warning: the behavioural outcome does not arrive until the end. Read the four stages in order, from the gender beliefs of Rodney's culture to two possible behavioural outcomes, and note which systems are acting at each step.
Stage 1: Gender in the broth (macrosystem acting on microsystem)
Rodney lives in a fictional country called Ourlandia, within a culture where gender beliefs cast males as independent and autonomous, and females as social and interdependent. When Rodney is little, his parents teach him that if he gets hurt he should not cry out for help but figure it out himself, and he meets the same approach at school. Girls can and do cry out for help, and they get it, without any sense that doing so is improper. The macrosystem is reaching Rodney through the microsystem of family and school.
Stage 2: The economy shifts (exosystem evolving within the chronosystem)
Ourlandia is a developing country whose economy is shifting from independent farming and lumber toward services, with job growth in tourism and social services. In the past, school attendance was higher for boys, but education beyond high school was not needed to start or inherit a farm or lumber operation; autonomy and independence mattered more. School involvement was lower for girls because school subjects were not viewed as relevant to women's roles. The new jobs emphasize customer service, relationships, empathy, and interaction, and some companies even help fund technical schools to train social workers and tourism staff. None of these changes involves Rodney directly. The economy, the employers, and their decisions about training are exosystem settings in which he takes no part, and the shift unfolds over years, which is why the chronosystem matters here. What has changed is which qualities pay off, and the qualities that now pay off are the ones Rodney's culture taught girls rather than boys.
Stage 3: Girls and school change together; boys disengage (with a bioecological push)
Because the female-gendered characteristics have become more valuable, the interaction between girls and school begins to change. More girls and their families see that finishing school matters, and some graduates go on to technical schools. Boys and young men are increasingly disaffected by what school offers and show little interest in the technical schools or the jobs they lead to. At a stage when the biology of adolescence supports some impulsivity, the bioecological element, boys are also more prone to act out and to be kicked out of school. Add a dwindling number of farming and lumber jobs, in part because of falling world prices in a globalized economy, and the result is changing gender roles and opportunities: some increases for females and, at least in the short term, decreases for males.
Stage 4: Two forks in the road (the behavioural outcome, at last)
Are there behavioural and health outcomes? Maybe. Suppose Ourlandia was already a minor transit point for drug trafficking. With many young men out of work, the aggregation of factors could lead to increased involvement, with health consequences related to violence, substance use, and possibly HIV/AIDS. Or suppose the Ourlandia Tourist Industry Association, an exosystem component, decided to address young men's employment and social role by establishing technical school programs in starting tourist businesses and web design. These might appeal to traditionally male-gendered characteristics, keep young men in school, and steer them away from the riskier activities, with better health outcomes. It's all clear now, right?
Notice what the story does. No single fact explains Rodney. Gender beliefs (macrosystem), a changing economy (exosystem over time, in the chronosystem), family and school practices (microsystem), and adolescent biology (the bioecological element) all feed into the same outcome, and the outcome that arrives depends on an intervention made at a level Rodney never touches directly. The Tourist Industry Association's technical school programs mattered because they changed the opportunity structure in a way that fit the qualities the macrosystem had taught young men to value.
Worked Example 6.1 applies the same reasoning to a Canadian case: a resource town in the northern interior of British Columbia after its sawmill closes, where two proposals respond to rising injuries and substance-related harms among young men. Watch for the participation test from 2.1 when sorting the institutions, because it decides which system each proposal acts in.
Worked Example 6.1: A Mill Town After the Mill
For decades the sawmill employed most of the men in town, and a grade 12 diploma was optional for a well-paid job on the green chain. When the mill closes, the regional economy tilts toward health care, tourism, and public services, and the local college expands its practical nursing and hospitality programs, where enrolment is mostly young women. Many young men from mill families drift between seasonal jobs, and the health unit notices rising emergency visits among men in their late teens and twenties for injuries and substance-related harms. A school counsellor proposes an anti-drug assembly. An economic development officer proposes a trades pre-apprenticeship stream at the college, recruited through hockey teams and the volunteer fire department.
Step 1, the behaviour and the population. The behaviours are the risky activities behind injuries and substance-related harms, and the population is young men in their late teens and twenties from mill families.
Step 2, sort the elements of the story into Bronfenbrenner's systems.
- Microsystem: the young men's families, the high school, the hockey teams, and the volunteer fire department, all settings they take part in directly.
- Mesosystem: connections between those settings, such as a recruitment drive that links the hockey team and the fire hall to a training opportunity.
- Exosystem: the regional economy, the college's decisions about which programs to offer, and the economic development office. The young men do not take part in these decision-making settings, yet the decisions shape their options.
- Macrosystem: beliefs that a man's work is physical and independent and that a diploma is optional for a good job.
- Chronosystem: the mill closure and the shift of the economy over time. At 10, a boy in a mill family needed basic schooling and a working parent; at 17, the same boy needs a credential the town never used to require and a path into an economy that now rewards qualities he was not raised with.
Step 3, place each proposal. The anti-drug assembly acts in the microsystem, on the students present at the school, and leaves the exosystem untouched: the regional economy and the college's offerings stay as they were after the assembly ends. The trades stream is an exosystem decision that changes the opportunities available, and it reaches the young men through the mesosystem by recruiting through teams and a fire hall they already belong to.
Step 4, judge which proposal is more likely to change behaviour. Ecological Systems Theory points to the trades stream. As in Ourlandia, it changes the opportunity structure in a way that fits what the young men were raised to value, and it offers physical, hands-on work at a developmental stage when impulsivity runs high, which is a bioecological fit. The assembly asks the young men to resist a risk without offering a replacement role. The two proposals could be combined rather than chosen between.
What the example shows. In Ecological Systems Theory, the level at which an intervention acts matters as much as its message. The common mistake to avoid is filing the college and the economic development office in the mesosystem; the young men do not take part in those settings, so they belong in the exosystem.
The example shows that the proposal acting on the exosystem, and delivered through settings the young men already belong to, is the one the theory expects to change behaviour. The same layered logic can also be built into a program from the start, as school health programs show.
2.4 Using the Theory in School Health Programs
Many health promotion interventions are ecological in nature and draw from Bronfenbrenner, which is no surprise, because public health is grounded in the ecological model. A good example is the Coordinated School Health Program (CSHP), an approach to school health with eight components. It was adopted by the Division of Adolescent and School Health at the Centers for Disease Control and Prevention (CDC) as its general framework, and then by the World Health Organization (WHO) in its global promotion of school health. However, the eight components have traditionally been drawn arrayed around a graphic of children, with no distinctions between components or ecological levels. By this we mean that the usual diagram shows every component as equally close to the student, as if a classroom lesson and a community partnership worked in the same way. It is reasonable to argue that this misses the ecological nature of the program and the complexity of implementing it, because components that act through different levels need different partners and reach students by different routes. The purpose of redrawing the diagram is to make those differences visible. For example, Lohrmann proposed a complementary ecological model in which parents (microsystem), schools, and community institutions (mesosystem, exosystem) are organized into levels of influence working outward from the students. Table 6.1 sorts the eight components in that spirit, giving for each what it involves and the level it mostly acts through; note that Lohrmann's model is an attempt to organize the program into levels rather than an exact replica of Bronfenbrenner's.
Table 6.1. The eight components of the Coordinated School Health Program, what each involves, and the level of Bronfenbrenner's systems it mostly acts through.
| CSHP component | What it involves | Level it mostly acts through |
|---|---|---|
| Comprehensive school health education | Curriculum on health topics delivered to students | Microsystem (classroom) |
| Physical education | Structured physical activity and skills | Microsystem (classroom, gym) |
| School health services | Nursing, screening, first aid, referral | Microsystem, linked to exosystem health services |
| School nutrition services | Meals, snacks, food environment | Microsystem, shaped by school board policy (exosystem) |
| Counselling, psychological, and social services | Support for mental health and social problems | Microsystem, linked to mental health agencies (exosystem) |
| Healthy school environment | Facilities, safety, psychosocial climate | Microsystem and school policy |
| School-site health promotion for staff | Employee wellness | Exosystem (adults whose well-being reaches students indirectly) |
| Family and community involvement | Parents, health councils, service clubs, faith communities, businesses | Mesosystem (connections) and exosystem (institutions) |
Read down the last column of Table 6.1 and most components sit in the microsystem of the classroom and school, while staff wellness and family and community involvement reach outward to the mesosystem and exosystem. Canada's version is Comprehensive School Health, promoted since 2005 by the Pan-Canadian Joint Consortium for School Health, a partnership of provincial and territorial health and education ministries. It organizes school health into four interrelated components: the social and physical environment; teaching and learning; healthy school policy; and partnerships and services. The four map readily onto Bronfenbrenner's layers, from classroom teaching in the microsystem to board policy and community partnerships in the exosystem, and the framework's insistence that the components be planned together is the ecological point in a sentence.
School health programs show the theory's strength: it gives planners a way to see that a classroom lesson, a board policy, and a community partnership are parts of one system. The same richness creates the theory's main weakness.
4.1 The Main Critique: Complexity
One critique stands out, and it is the biggest critique of all similar theories: complexity. By this we mean that the very feature that makes Ecological Systems Theory realistic, its many interacting levels and processes, also makes it hard to use. With so many levels acting on each other, how can you tell where the causal chain leading to a health behaviour actually lies? Almost as in chaos theory, a small change in one component within one level is likely to cause reverberations through multiple levels, and those reverberations may in turn change the environment at every level with which the individual was interacting in the first place. For example, in Rodney's story a shift in the economy changed which qualities paid off, which changed how girls and boys engaged with school, which in turn shaped the paths open to young men later; tracing any one link in that chain is difficult because every other link is moving at the same time. Two questions follow from this critique and shape the rest of the lesson. The first is a research question: what kind of research could identify such causal chains? The second is a practical question: how could you set up an intervention to create a causal chain that leads to better health?
When you are asked to evaluate a multilevel theory, complexity is the critique to name first, and these two questions are the way to explain why it matters: realism about interacting levels comes at the price of being hard to test and hard to turn into a single intervention. Flay's theory, in the next section, is one attempt to make the chains traceable.
This section has made the case for multilevel theory and has set out Bronfenbrenner's Ecological Systems Theory: its five nested systems, the bioecological amendment with its distinction between proximal and distal processes, the theory in motion in Rodney's story and the mill town, its use in school health programs, and its main critique, complexity. The key takeaways and knowledge check that follow review this material.
Key Takeaways
- 1.1 Sorting theories by ecological level is a useful filing system, but it is not how the world works: factors at one level interact with factors at other levels, and a theory confined to one level cannot say how they work together. A theory of family dynamics, for example, leaves open what shapes the family itself, such as socioeconomic constraints, racism, and gender roles. When a question describes a single-level explanation, look for the factor at another level that it leaves unexplained. The trap to avoid is treating the filing system as if the levels were separate in real life.
- 1.2 Single-level theories read like equations, with factors at one level adding up to a likelihood of behaviour, as in the Health Belief Model. Multilevel theories read like paths through the layers, which can branch like trees or dendrites and can run back upward when a change in behaviour reverberates through the broader levels. To classify a theory on an exam, look at the form of its explanation: adding factors at one level points to a single-level theory, and tracing influence across layers with feedback points to a multilevel theory.
- 2.1 Bronfenbrenner's Ecological Systems Theory places the developing person inside five systems: the microsystem of direct contact, the mesosystem that connects those settings, the exosystem the person never enters but is affected by, the macrosystem of culture, customs, and law, and the chronosystem of time. To classify an influence, ask where it sits relative to the person and whether the person takes part in it. The most common error is filing a parent's workplace or a health authority in the mesosystem; if the child does not take part in the setting, it is exosystem, and a link between two settings the child does take part in, such as a teacher and parent meeting, is mesosystem.
- 2.2 In the 1990s Bronfenbrenner added the person's own biological, psychological, cognitive, and genetic processes and gave time a formal place, producing Bioecological Systems Theory. The amendment brought a distinction you will use again: proximal processes occur between a person and the immediate environment, and distal processes occur within the wider environment and eventually reach the person through those proximal interactions. A parent's conversation with a teenager is proximal; a provincial restriction on flavoured vapes is distal. Classify a process by where it occurs, not by how large its eventual effect is.
- 2.3 Rodney's story shows that no single fact explains a behaviour: gender beliefs, a changing economy, family and school practices, and adolescent biology all feed one outcome, and which outcome arrives depends on a decision made at a level Rodney never touches. The mill town example applies the same logic, showing why an exosystem change delivered through settings young men already belong to is more likely to change behaviour than a one-off assembly. When you analyse a case, sort each element into a system, place each proposed intervention at the level where it acts, and ask what each leaves untouched.
- 2.4 The Coordinated School Health Program has eight components that were traditionally drawn around children with no distinction between levels, and Lohrmann redrew them as levels of influence working outward from students. Canada's Comprehensive School Health organizes school health into four interrelated components (the social and physical environment, teaching and learning, healthy school policy, and partnerships and services) that map onto Bronfenbrenner's layers. If a question asks what the ecological view adds to school health, the answer is that components at different levels must be planned together rather than listed side by side.
- 4.1 The biggest critique of Ecological Systems Theory, and of every multilevel theory, is complexity: with so many interacting levels, a small change at one level can reverberate through the others, so it is hard to tell where a causal chain lies. The critique raises two questions, what research could trace such a chain and how an intervention could build one that leads to better health. When asked to evaluate a multilevel theory, name complexity and explain it through those two questions, rather than saying the theory is wrong.
1. A colleague reviewing a program for children argues that a theory linking family dynamics to children's eating habits is all the program needs, because the theories in this course are sorted by level and the family is the right level for this problem. Which response reflects the case for multilevel theory made in this lesson?
2. Single-level theories and multilevel theories differ in their form. Which statement best captures that contrast?
3. A grade 5 student's mother works rotating night shifts at a regional hospital, so the child is often alone at bedtime. Under Bronfenbrenner's Ecological Systems Theory, the hospital's shift-scheduling policy belongs to which system?
4. A provincial ministry restricts the sale of flavoured vaping products, and over the following year a grade 10 student finds that the store near her school no longer stocks the flavour her friends used to share. In the bioecological version of Bronfenbrenner's theory, how should the ministry's decision be classified?
5. In the Rodney story, the economy of Ourlandia shifts over the years from farming and lumber toward tourism and social services, changing which qualities the job market rewards. Which pair of systems correctly labels this shift?
6. In the mill town of Worked Example 6.1, a school counsellor proposes an anti-drug assembly and an economic development officer proposes a trades pre-apprenticeship stream at the college, recruited through hockey teams and the volunteer fire department. Which statement best applies Ecological Systems Theory to the two proposals?
7. A health promotion team argues that the Coordinated School Health Program's traditional diagram, with eight components arranged around a picture of children, should be redrawn. What is the main reason given in this lesson?
8. A research team plans to use Ecological Systems Theory to pin down exactly which factor caused a rise in injuries among young people in a town. A colleague warns them about the theory's main critique. What is the colleague most likely to say?
✦ Pass the knowledge check with 100% to continue
The Theory of Triadic Influence and Other Multilevel Models
⏱ Estimated reading time: 16 minutes
Overview
Bronfenbrenner gave us layers. Brian R. Flay and his colleagues set out to give us a map of how influence flows through them. The Theory of Triadic Influence (TTI) is the second general multilevel theory in this lesson and, because it was built by researchers working in applied settings, it comes with a clearer instruction for practitioners: find the stream, follow it upstream, and intervene where you can. This section sets out why the theory was built, its two organizing dimensions, the waterfall metaphor, its finer elements, one study that tested it, and two critiques, and then describes three multilevel models built for specific problems by the CDC, WHO, and UNICEF.
Learning Objectives
- 3.1 Explain the problem the Theory of Triadic Influence was built to solve: fragmented, narrowly targeted theory drawn from many disciplines.
- 3.2 Define the theory's two dimensions, levels of causation and streams of influence, and name the proximal predictor at the end of each stream.
- 3.3 Use the waterfall metaphor to trace a behaviour upstream and choose a feasible point of intervention.
- 3.4 Describe the theory's finer elements (placement of existing theories, substreams, feedback loops, related behaviours) and apply the rule that proximal causes predict behaviour while distal and ultimate causes explain it.
- 3.5 Summarize how the Washington State smoking study tested the theory and what its mixed results showed.
- 4.2 Describe the two critiques of the Theory of Triadic Influence: complexity, and the unclear rationale for placing factors at particular levels.
- 4.3 Describe three problem-specific multilevel models from the CDC, WHO, and UNICEF, and explain the conclusion they share with the general theories.
3.1 Fragmented Theory and a Broken Car
The TTI came out of a view that the most commonly used theoretical approaches, and the health promotion interventions built on them, address only pieces of the puzzle. Because each theory covers only some of the factors, an intervention built on it tends to be narrowly targeted at those factors and to leave the rest untouched. This also makes it difficult to confirm the validity of any one theory. If a program built on one theory fails, it is hard to tell whether the theory was wrong or whether factors the theory never considered were working against it; if the program succeeds, it is equally hard to tell whether the theory deserves the credit. In addition, for some health issues the theories in use come from several disciplines, such as psychology, sociology, economics, and anthropology, each with its own terminology and assumptions. By this we mean that two researchers may be describing the same influence on behaviour under different names, or using the same word to mean different things. This fragmentation is a real problem: how can we discuss or address a problem with any consistency when the people working on it are not speaking the same language?
A car that will not start makes the point. You enlist three friends. One insists it is all about temperature, because cars are made of metal and everything has to be at a medium temperature before you can fix anything. The second says it is always the electrical system. The third tells you to sit by the car and feel it, because nothing will be fixed until you share a vibe with it. The chances of getting the car fixed are not good. The TTI is an attempt to schematize the full spectrum of factors and processes involved in health-related behaviour so that there is a common point of reference, even when a given intervention, as most must, addresses only some of them. Like Bronfenbrenner's bioecological model, it holds that factors at multiple levels are linked in continuous, interactive pathways, and that some act in immediate, proximal ways while others act as long-term, distal or underlying causes.
To provide that common point of reference, the theory gives every factor two coordinates.
3.2 Two Dimensions: Levels of Causation and Streams of Influence
The model is organized around two basic dimensions (Flay, Snyder, and Petraitis, 2009). A useful way to picture them is as the two coordinates of a grid: one tells you how far a factor sits from the behaviour, and the other tells you what kind of influence it is.
Levels of causation. The first dimension sorts contributing factors by how close they sit to the behaviour in the causal chain. Causally proximal or immediate factors are the ones closest to the behaviour, such as a person's intentions and skills. Causally distal or predisposing factors sit one step further back and shape those immediate factors; examples are family relationships, social capital, and community laws or policies. Finally, underlying or ultimate causes sit furthest back, such as structural poverty, gender roles, and cultural goals and values. By this we mean that an ultimate cause rarely acts on a behaviour directly; it works through the distal factors, which in turn work through the proximal ones. The terms proximal and distal echo Bronfenbrenner's processes in 2.2, and in both theories they describe closeness to the person rather than size of effect.
Streams of influence. The second dimension describes the kind of influence a factor carries. The factors operate across the causal levels via three streams, and each stream ends in a different proximal predictor of behaviour. The intrapersonal stream carries a person's own characteristics, and it ends in self-efficacy, the person's confidence that they can carry out the behaviour. The interpersonal or social stream carries the influence of the people around the person, and it ends in social normative beliefs, what the person believes others do and expect. The cultural-environmental stream carries the influence of the wider culture, institutions, and economy, and it ends in attitudes toward the behaviour. Three streams therefore run through three levels, and the three streams are the triad in the theory's name. Open each of the three tabs to follow one stream from the ultimate level down to the proximal level; each tab closes with an illustration from youth vaping.
Intrapersonal influences, ending in self-efficacy
The personal stream carries intrapersonal characteristics that contribute to self-efficacy regarding specific behaviours. At the ultimate level, biological and genetic dispositions and personality; at the distal level, sense of self and personal control; at the proximal level, skills, self-determination, and self-efficacy or behavioural control.
Vaping illustration: a fifteen-year-old's impulsivity (ultimate), her sense that she can handle herself (distal), and her confidence that she could turn down a vape at a party without embarrassment (proximal self-efficacy).
Interpersonal and social influences, ending in social normative beliefs
The social stream carries the social situation, context, or microenvironment that contributes to social normative beliefs about specific behaviours. At the ultimate level, the social situation itself; at the distal level, interpersonal bonding, the behaviours and attitudes of others, and motivation to comply; at the proximal level, perceived norms and social normative beliefs.
Vaping illustration: a friend group that hangs out behind the arena after school (ultimate), an older sibling she admires who vapes (distal), and her belief that most people her age vape and expect her to (proximal normative belief).
Cultural-environmental influences, ending in attitudes
The environmental stream carries sociocultural and macroeconomic factors that contribute to attitudes about specific behaviours. At the ultimate level, the cultural environment; at the distal level, interactions with social institutions, information and opportunities, values and evaluations; at the proximal level, knowledge, expectancies, and attitudes toward the behaviour.
Vaping illustration: a market in which nicotine products are legal, marketed, and sold in flavours (ultimate), the price set by excise duties and the Health Canada campaigns she has seen (distal), and her attitude that vaping is harmless fun (proximal).
The factors at the various levels flow within each stream and act through the proximal predictors. By this we mean that an ultimate or distal factor does not reach behaviour by its own separate route; it reaches behaviour by changing the self-efficacy, normative belief, or attitude at the bottom of its stream. For example, in the vaping tabs, the price set by excise duties does not make the student vape or stop vaping directly. It changes her attitude toward vaping and what she expects from it, and that attitude, together with her normative beliefs and her self-efficacy, feeds her decision. This is why the three proximal predictors are the gateway through which every stream passes. Flay and colleagues draw the streams as arrows running from the social context, the cultural environment, and biology and personality, through social normative beliefs, attitudes, and self-efficacy, into decisions and intentions and then behaviour. A later version of the diagram, developed with specific reference to adolescent smoking, adds the causal levels as horizontal bands that the streams cross.
Common confusion: levels and streams are different dimensions
Proximal, distal, and ultimate describe how close a factor sits to the behaviour; intrapersonal, social, and cultural-environmental describe what kind of influence it is. Every factor needs both coordinates. Biology and personality are not a fourth stream: they are the ultimate level of the intrapersonal stream. A belief about what friends expect is the proximal end of the social stream, and community laws or policies sit at the distal level of the cultural-environmental stream.
Locating a factor by level and stream tells you where it sits. The waterfall metaphor tells you what to do with that location.
3.3 The Waterfall: Tracing Influence Upstream
If the two dimensions are hard to hold in mind, a picture helps. Behaviour at the individual level is like what happens at the bottom of a waterfall: the culmination of a cascade of influences that interact and mingle on their way down through the levels of causation to the point where the water hits the rocks or the pool below. You can look at the immediate determinants of a decision, but these can be traced back up the waterfall to their source. To influence behaviour, your task is to trace the factors back upstream and focus an intervention at a point where it is feasible and where you think it can make a difference in the chain. The purpose of the metaphor is to make two separate judgements explicit. The first is feasibility: whether your organization has the authority and resources to act at that point. The second is effect: whether acting there would change what flows down to the behaviour. For example, a school district can act on refusal skills, which sit near the pool, but only a province can act on flavoured products, which sit further upstream. The best point of intervention is therefore not always the one closest to the behaviour, nor the one furthest upstream, but the one where feasibility and effect meet.
Interactive 6.2 lets you practise the tracing for a grade 9 student who vapes. Its columns are the three streams of influence, its rows are the three levels of causation, and its bottom row is the waterfall's pool. Click every cell, working from the top row down, to see what the theory places there and where a Canadian program could act.
📊 Interactive 6.2: Trace it upstream.
Each cell is one level of causation in one stream of influence, applied to a grade 9 student who vapes. Click a cell to see what the theory places there and where a Canadian program could act. The bottom row is the waterfall's pool.| Intrapersonal (personal) stream | Interpersonal (social) stream | Cultural-environmental stream | |
|---|---|---|---|
| Ultimate / underlying causes | |||
| Distal / predisposing influences | |||
| Proximal / immediate predictors | |||
| The pool | |||
Click a cell
The labels in the cells are the theory's own, taken from the version of its diagram that was drawn for adolescent smoking.
Working down the grid shows the waterfall in action: each upstream cell feeds the proximal predictor beneath it, the cells a school can reach on its own sit mostly near the bottom, and the cells further up need partners at other levels of government. Choosing a point of intervention therefore means asking two questions about each cell, whether acting there is feasible for you and whether it would make a difference to the chain below it.
3.4 Finer Elements of the Theory
The TTI does more than sort factors. It acknowledges and integrates other social and health behaviour theories by causal level and stream. By this we mean that the TTI does not ask you to abandon the theories you already know; instead, it gives each of them an address on its grid, showing which part of the full picture that theory explains. For example, a theory about self-efficacy explains the proximal end of the intrapersonal stream, while a political-economic theory explains the ultimate level of the cultural-environmental stream. Placed this way, the two theories are no longer competing explanations of the same behaviour but explanations of different links in one chain. This is what makes the TTI a common point of reference rather than a rival, and it answers the fragmentation problem of 3.1. Table 6.2 collects the placements the theory's developers list.
Table 6.2. Existing social and health behaviour theories placed by causal level and stream of influence in the Theory of Triadic Influence.
| Causal level | Intrapersonal stream | Interpersonal / social stream | Cultural-environmental stream |
|---|---|---|---|
| Ultimate (underlying cause) | Biological and genetic theories; psychoanalytic theory | Family systems theory; peer-related theories | Class, social disorganization, political-economic, and other broad theories |
| Distal (predisposing) | Theories of self-esteem, personal control, and similar factors | Social cognitive theory; social development theory | Cultural, values, and motivational theory |
| Proximal (immediate) | Social skills, self-efficacy, self-regulation | Social normative beliefs; conformity | Expectancy or attitude patterns |
In addition to placing other theories, the TTI includes three finer elements. First, each stream has two substreams: an informative substream of cognition and thoughts, and a control substream of affect and feelings. Both flow from the most distant causes to the causes closest to the behaviour. By this we mean that, within the social stream for example, what a young person knows about friends' behaviour and how strongly she feels about belonging to the group travel side by side toward the same normative belief. The interactions among all of these are not easy to pin down. In addition, the streams do not run in isolation from each other, and interactions between streams can be negative or positive influences. For example, a student with strong refusal skills (intrapersonal) may find those skills weakened if everyone in her friend group vapes (social), which is a negative interaction; the same skills may be strengthened if her school and family both treat vaping as unacceptable, which is a positive one. Second, the theory includes feedback. Once a behaviour occurs there is a feedback loop, based on the external and internal reaction to it, that causes new behaviours and new loops. By this we mean that how a person feels after the behaviour (the internal reaction) and how others respond to it (the external reaction) become new influences in their own right. For example, a student who tries a vape and is welcomed by an older group may come to believe that vaping is normal among people she admires, and she may spend more time with that group; both changes then push toward vaping again. The pool at the bottom of the waterfall can therefore change the streams that feed it. Third, related behaviours have similar causes. For example, smoking and substance use share distal causes almost entirely and differ somewhat in proximal causes, while smoking and skiing, a pair of less related behaviours, have fewer causes in common. The practical consequence is that a program aimed at shared distal causes may affect several related behaviours at once. The practical logic of the theory compresses into one rule.
Rule to remember: proximal causes predict, distal and ultimate causes explain
Proximal causes predict most behaviour, and distal and ultimate causes explain it. If you want to forecast which students in a class will vape next month, measure their intentions, self-efficacy, normative beliefs, and attitudes. If you want to understand why the rate in that school is twice the rate in another, look upstream to family patterns, the local retail environment, price, and the cultural meaning of nicotine. An intervention can be aimed at either, but a program that changes only proximal factors while the upstream conditions remain in place is working against the current.
Worked Example 6.2 applies the grid, the waterfall, and this rule to a planning decision in a Fraser Valley school district. Watch how each proposal's position in the grid tells you when its effect would show up and who would have to carry it out.
Worked Example 6.2: Youth Vaping and Three Levels of Government
A school district in the Fraser Valley wants to reduce vaping among students in grades 8 to 10. Five proposals are on the table. (A) A classroom program that builds refusal skills and corrects the belief that most students vape. (B) A parent evening and a text-message series for families. (C) A district policy that treats vaping on school property as a health issue with a cessation referral rather than a suspension. (D) A letter supporting provincial restrictions on flavoured vaping products in convenience stores. (E) Support for the federal excise duty on vaping products and the advertising restrictions in the Tobacco and Vaping Products Act. The district can fund A, B, and C itself; D and E are advocacy.
Step 1, the behaviour and the population. The behaviour is vaping, and the population is students in grades 8 to 10 in the district.
Step 2, place each proposal by stream and causal level.
- (A) works at the proximal level in two streams: refusal skills build self-efficacy in the intrapersonal stream, and correcting the belief that most students vape targets social normative beliefs in the social stream.
- (B) works at the distal level of the social stream, through interpersonal bonding, the behaviour and attitudes of others, and motivation to comply.
- (C) works at the distal level of the cultural-environmental stream, because a district policy is an interaction with a social institution that shapes information, opportunities, and values around the behaviour.
- (D) and (E) act furthest upstream in the cultural-environmental stream, at the distal and ultimate levels: they change the laws, prices, and marketing that make up the environment in which nicotine products are sold. In the tab illustration, the flavoured, marketed market sits at the ultimate level and the price set by excise duties at the distal level.
Step 3, apply the rule that proximal causes predict and distal causes explain. Proposal A works on the proximal predictors, so it is the proposal most likely to shift next month's vaping rate. Proposals D and E, and to a degree B and C, act on the causes that explain why one school's rate is twice another's, and their effects would show up over years as price, availability, and marketing change.
Step 4, decide where intervention is feasible. The district can act at the pool (A) and at two upstream points within its reach (B and C), but it cannot change D or E itself. Running A alone treats the water after it has fallen: feasible, and the right place to start, but working against a current the upstream levels keep supplying. The waterfall therefore points to a combined plan: A, B, and C funded by the district, with D and E carried to the province and the federal government as advocacy.
What the example shows. The grid turns a list of proposals into a plan that says what each can change and when. The common mistake is to treat the proposal most likely to move next month's numbers as the whole program.
The example shows that the TTI is a planning tool as well as an explanation: it sorts proposals by what they can change, predicts when each would show its effect, and makes visible what a single agency cannot do alone.
3.5 Testing the Theory: Adolescent Smoking in Washington State
All of these levels, paths, and directions pose challenges for anyone collecting and analyzing data, whether to guide the choice of an intervention or to evaluate one said to be based on the TTI. The difficulty is that a simple comparison of one factor with one behaviour cannot show how factors at different levels act on each other along the way. For that reason, complex statistical methods that can trace the effect of multiple variables on each other and on behaviour are very likely to be necessary; path analysis and structural equation modelling are the usual candidates. In addition, the developers suggest thinking in terms of causal bundles related to specific behaviours. By this they mean groups of causes that tend to travel together and produce related behaviours, an idea that resembles syndemic health conditions, in which issues such as HIV/AIDS, substance use, and violence are related by similar causal circumstances.
Because the TTI was developed by researchers working in an applied context, it has been used in intervention studies, in research on predictors of health risk behaviour, and in program planning. One example is a predictor study of adolescents' transition from initiating smoking to regular smoking, using a large sample from Washington State (Bricker and colleagues, 2009). The team assessed three kinds of factor: (1) risk factors based on an individual's psychological makeup, such as a thrill-seeking tendency, at the ultimate level in the individual stream; (2) psychological risk factors related to the social world, such as a tendency to do what friends do, at the distal level in the individual stream; and (3) the behaviours of others, specifically parent and friend smoking, at the distal level in the social stream. According to the TTI, the more proximal second and third types should influence smoking decisions more than the individual characteristics.
The results were mixed but gave some support. Adolescents with individual psychological risk factors were more likely to move through the transition stages, but for trying a first cigarette thrill-seeking mattered more than a parent's or a close friend's smoking, the reverse of the theory's prediction, and at the later transitions the distal factors did not clearly outweigh it. Only 2 of the 30 interactions tested were significant, and neither involved thrill-seeking; in both, parents' or friends' smoking mattered more for adolescents who tended to comply with those parents or friends. Interactions of this kind are what the theory expects, and the many possible interactions are also what make the theory hard to test cleanly. If you are asked what the study showed, the accurate summary is partial support: the theory's structure was useful for organizing the factors, but its prediction about which level matters most did not hold for a first cigarette. That difficulty leads directly to the theory's critiques.
4.2 Two Critiques of the Theory of Triadic Influence
Critique 1: complexity. As with Ecological Systems Theory, the biggest issue is complexity. Tracing and tracking down the multiple influence pathways and interactions is not easy, and for that reason the theory is not easy to demonstrate. By this we mean that there are so many possible influence chains, across three streams, three levels, two substreams, and feedback loops, that an intervention or a research effort faces significant difficulties in trying to investigate them all. The Washington State study in 3.5 shows the problem in practice: it tested 30 interactions among its factors, and only 2 were significant. Despite the difficulties, approaches like this are important, because they keep the whole picture in view even when a single study can examine only part of it.
Critique 2: where things are placed. The second critique is narrower. It is not always clear what the rationale is for positioning some factors at particular levels, and a placement matters because it implies which factor causes which. For example, social competence is positioned at a causal level above social skills, so the diagram implies that competence acts further upstream; however, skills could be argued to be a necessary precursor to competence, which would put the order the other way round. This is partly an issue of presentation, and Flay and colleagues acknowledge that their diagrams are simplified and do not always capture the scope of interactions.
Keep the two critiques distinct when you evaluate the theory: complexity is about how many chains there are to trace, while the placement critique is about whether a particular factor belongs at the level where the theory puts it. The Theory of Triadic Influence, then, sorts the factors behind a behaviour by level of causation and stream of influence, directs the practitioner upstream to a point where intervention is feasible, and has drawn mixed but supportive evidence along with these two critiques. The last part of this section turns from the general theories to models built for particular problems.
4.3 Problem-Specific Multilevel Models
Beyond the two general theories, there are multilevel frameworks developed with reference to a particular health problem or population, and they belong in the same family as the risk and protective factors model you will meet in Section 4. By this we mean that, unlike Bronfenbrenner's theory or the TTI, which aim to describe any behaviour, these models were built to organize the factors behind one problem, such as violence, or the well-being of one group, such as adolescents. Three are described here. The first, the CDC model of violence, later adopted by WHO, sorts contributing factors into four domains: individual, relationship, community, and societal. The second, a CDC model of macroeconomic factors and youth violence, links the economy to youth violence through community, situational, and individual levels and adds a time dimension; its key finding was that family and school interventions were often undercut by community conditions. The third, the UNICEF Adolescent Well-Being Framework, takes a positive perspective: rather than listing risks, it sets out domains describing what needs to be in place, at multiple levels, for adolescents to do well. Each model is described below with how it organizes its levels and what it recommends.
The CDC and WHO model of violence: individual, relationship, community, societal
Originally developed by the CDC to understand the multiple levels of contributing factors for youth violence, this model was adopted by WHO in its World Report on Violence and Health and later applied to intimate partner and sexual violence. Its four domains: individual, biological and personal factors such as age, education, income, substance use, or a history of abuse; relationship, close relationships with peers, partners, and family; community, the settings such as schools, workplaces, and neighbourhoods where social relationships occur, and their characteristics; and societal, the norms and the health, economic, educational, and social policies that create a climate in which violence is encouraged or inhibited and maintain inequalities between groups. The model guides prevention toward building strengths to offset risk factors at each level. By this we mean that, rather than only trying to remove risks, a program can add something at the same level that counterbalances them, such as a supportive mentor at the relationship level for a young person whose family relationships are a source of risk. The CDC's view is that interventions are best designed and implemented at multiple levels. The same four-level layout is familiar in Canadian family violence and youth violence prevention work.
The CDC model of macroeconomic factors and youth violence
A review of research and expert input commissioned by the CDC in 2008 produced a model of how macroeconomic factors relate to youth violence, and the scholars involved published their papers as a book (Rosenfeld, Edberg, Fang, and Florence, 2013). It has a temporal dimension, long-term versus short-term effects, and holds that macroeconomic factors operate in two ways. They operate directly, through societal-level domains such as social spending, employment, or income inequality. They also operate indirectly, through conditions affecting youth such as family dysfunction, neighbourhood conditions, concentrated poverty, under-resourced schools, and street markets. By this we mean that an economic downturn can affect youth violence both by removing jobs young people might have held and by putting strain on the families, schools, and neighbourhoods in which they grow up. The factors run from the top down: macroeconomic; community-level (neighbourhoods, schools, services); situational (family violence, youth gangs, drug market volatility, lack of employment); and individual (perceptions, attitudes, behaviours such as gun carrying). The temporal dimension matters because the factors do not all move at the same speed. Inequality changes slowly, so its effects build gradually over years. Some factors act only past a threshold that varies by community; in other words, a factor such as poverty may have little visible effect until it reaches a certain level, and that level is not the same everywhere. Unemployment, in contrast, can change rapidly. A key finding was that family-level and school-level interventions, even when successful, were often reduced or minimized by community-level factors. For example, a school program may succeed in changing students' attitudes toward violence, yet the gain can be eroded when those students return each day to a neighbourhood with few jobs and an active street market. For this reason, the review concluded that single-level interventions should be paired with community or broader-level components.
The UNICEF Adolescent Well-Being Framework
In global health, multilevel models have flowed from the social determinants approach and an emerging interest in well-being. Since the mid-2000s the United Nations Children's Fund (UNICEF) has paid increasing attention to adolescents, establishing an Adolescent Development and Participation office at its headquarters, and several regions have developed models of resilience: what adolescents need in order to avoid health risk behaviours. The Latin America and Caribbean office's Adolescent Well-Being Framework identifies what to monitor from a positive perspective rather than only negative outcomes such as HIV rates, violence, dropout, substance use, and early pregnancy. In the version taught in this lesson, the framework has eight domains: health status, knowledge, and skills; subjective well-being; a supportive environment for identity and equity; a protective environment of legal protections and enforcement; educational opportunity and performance; access to supportive services and relationships; socioeconomic opportunity, actual and perceived; and participation. The list of domains changed as the framework was developed, and Edberg's background report for the regional office (2008) began from eleven. The domains rest on a working definition of adolescent well-being, which covers both the individual's ability to acquire knowledge, skills, values, relationships, and services, and the ability of family, peers, community, and institutions to support those aspects of well-being. This two-part definition is what makes the framework multilevel. By this we mean that well-being is not treated as something an adolescent achieves alone; it depends as much on whether the people and institutions around the young person provide the supports as on what the young person does with them. For example, an adolescent's educational opportunity depends on her own effort, but also on whether schools are available, whether her family can afford for her to attend, and whether the law protects her right to do so. Progress is measurable within and across the domains, so a region can see not only whether adolescents are doing better but in which domains the supports are still missing.
Notice what these three share with the general theories and with each other. Each sorts contributing factors into levels that run from the person outward to society, each assumes that the levels interact, and each concludes that action at one level alone is unlikely to hold. The CDC's macroeconomic review put the conclusion in its most practical form: single-level interventions, such as family or school programs, should be paired with community or broader-level components. That conclusion is the bridge to the second half of this lesson, where the question becomes which level, and which factor, to act on first.
This section has set out the Theory of Triadic Influence, from the fragmentation it was built to overcome through its two dimensions, the waterfall, its finer elements, the Washington State study, and its two critiques, and has described three problem-specific multilevel models from the CDC, WHO, and UNICEF. The key takeaways and knowledge check that follow review this material.
Key Takeaways
- 3.1 The Theory of Triadic Influence was built because the commonly used theories address only pieces of the puzzle, often in the vocabularies of different disciplines, which leaves interventions narrowly targeted and theories hard to validate. Like the three friends arguing over a broken car, specialists who do not share a language struggle to fix the problem. The theory schematizes the full spectrum of factors so there is a common point of reference, even when one program addresses only some of them. If asked what problem the theory solves, answer "fragmentation", not a gap in any single theory.
- 3.2 Every factor in the theory has two coordinates. Its level of causation says how close it is to the behaviour (proximal, such as intentions and skills; distal, such as family relationships and community laws; ultimate, such as structural poverty and cultural values), and its stream of influence says what kind of influence it is. The intrapersonal stream ends in self-efficacy, the interpersonal or social stream ends in social normative beliefs, and the cultural-environmental stream ends in attitudes. A frequent trap is inventing a fourth "biological" stream; biology and personality are the ultimate level of the intrapersonal stream.
- 3.3 The waterfall pictures behaviour as the pool at the bottom of a cascade of influences that mingle as they fall through the levels of causation. Its instruction is to trace the factors upstream from the immediate determinants to their source and intervene at a point that is feasible for you and can make a difference in the chain. In a case question, say both parts: where the factor sits, and why acting there is within reach of the agency in the case (a school district can teach refusal skills; only a province can restrict flavours).
- 3.4 The theory places existing theories by level and stream (Table 6.2), gives each stream a cognitive and an affective substream, includes feedback loops after a behaviour occurs, and expects related behaviours, such as smoking and substance use, to share most distal causes. Its practical rule is that proximal causes predict most behaviour, while distal and ultimate causes explain it. Use the rule to judge timing: a program on intentions or norms should move next month's numbers, while changes in price, availability, or culture explain differences between groups and show their effects over years.
- 3.5 The Washington State study placed thrill-seeking at the ultimate level of the individual stream, a tendency to do what friends do at the distal level of the individual stream, and parent and friend smoking at the distal level of the social stream, expecting the two distal factors to matter more. The results gave mixed but partial support: thrill-seeking mattered more than parents' or friends' smoking for trying a first cigarette, and only 2 of 30 interactions were significant. Summarize the study as partial support and remember that the many possible interactions are what make the theory hard to test.
- 4.2 The theory faces two critiques. Complexity means there are so many influence chains that no study or intervention can trace them all, and the placement critique asks why some factors sit where they do, such as social competence above social skills when skills might come first. Keep them separate on an exam: complexity is about the number of chains, and placement is about whether a particular factor belongs at its level. Flay and colleagues acknowledge that their diagrams are simplified.
- 4.3 Three problem-specific models apply multilevel thinking to particular problems: the CDC and WHO model of violence (individual, relationship, community, societal), the CDC model of macroeconomic factors and youth violence (macroeconomic, community, situational, and individual levels over time), and UNICEF's Adolescent Well-Being Framework with its domains of adolescent well-being. All reach the same conclusion as the general theories, that action at one level alone is unlikely to hold, so single-level programs should be paired with community or broader-level components. Learn the four violence domains by name, since they are easily confused with Bronfenbrenner's systems.
1. A study measures, among grade 10 students, how strongly each believes that most of their friends expect them to vape. In the Theory of Triadic Influence, this construct is the proximal end of which stream?
2. A planner places two factors in the Theory of Triadic Influence: structural poverty in a neighbourhood, and a teenager's confidence that she could turn down a cigarette at a party. Which placement is correct?
3. In Worked Example 6.2, suppose the Fraser Valley district decides to fund only proposal A, the classroom program on refusal skills and norms correction. What does the waterfall metaphor say about this choice?
4. The practical logic of the Theory of Triadic Influence compresses into one sentence about proximal and distal causes. Which is it?
5. Using the rule that proximal causes predict behaviour and distal and ultimate causes explain it, which Fraser Valley proposals from Worked Example 6.2 would you expect to shift the district's vaping rate next month, and which over the next five years?
6. In the Washington State smoking study, thrill-seeking was placed at one level, a tendency to do what friends do at another, and parent and friend smoking at a third. Which placement matches the study's design?
7. A reviewer of the Theory of Triadic Influence asks why social competence is positioned at a causal level above social skills, when skills could be argued to be a necessary precursor to competence. Which critique of the theory does this illustrate?
8. A CDC review of macroeconomic factors and youth violence found that family-level and school-level interventions, even when they showed some success, were often reduced or minimized by community-level factors. What recommendation followed?
✦ Pass the knowledge check with 100% to continue
Doing Something About It: Picking Your Battles and PRECEDE
⏱ Estimated reading time: 18 minutes
Overview
This section turns from explaining behaviour to changing it. Now the fun begins. Or rather, not so fast. The transition from theory to practice is one of the most difficult things to accomplish, and how to do it well must be one of the most discussed topics in public health journals, meetings, and policymaking agencies. This section covers the first half of that transition: why it is hard, how to choose what to work on, the three components every planning process shares, and the assessment phases of the best-known planning framework, PRECEDE-PROCEED. Two running cases carry the section: a hypothetical rust-belt community, North Hill County, and a Canadian health unit working on falls among older adults.
Learning Objectives
- 5.1 Explain why the transition from theory to practice is difficult, and name the three things most planning frameworks combine.
- 5.2 Use importance and changeability to sort candidate factors into four groups, and explain what can override the ratings.
- 5.3 State the three components of any program planning process and explain how they form a continuous feedback loop.
- 6.1 Explain what the name PRECEDE-PROCEED says about the approach and why it is a useful planning framework.
- 6.2 Describe the five PRECEDE assessment phases, the terms of the epidemiological assessment, and the sources of data each phase uses.
- 6.3 Classify factors from a case as predisposing, enabling, or reinforcing, and distinguish them from the behavioural and environmental risk factors of phase 3.
5.1 Why the Move from Theory to Practice Is Hard
You know about the complexity of identifying the causes of health behaviour. You have learned individual, social and group, organizational, political-economic, and cultural theories, and, in the previous two sections, theories that try to connect them. Most importantly, you have learned that each theoretical approach is a piece of an interrelated puzzle. The transition to practice means grappling with that complexity and arriving at a specific intervention, program, policy, communication strategy, message, or coordinated approach that could have an effect on the behaviours relevant to a health problem.
The wide range of factors that affect what people do can be pictured as the layers of an onion, and the ecological perspective remains the most useful framework for understanding problems as they exist in the real world. However, understanding a problem in all its layers runs straight into a practical constraint. Unless you have unlimited resources and unlimited time, you cannot address every factor you identify with a program component, so you have to pick and choose. For example, for the young women in the sex trade in Thailand, would you tackle the political economy, urbanization, class, and poverty, or individual knowledge and skills? For blues musicians and smoking, would you work on social support and networks, or on individual awareness? Neither question has an obvious answer, which is exactly why the choice should not be made by instinct. If you must pick, do it with a planning framework in mind, because a framework gives you a rationale for the choice that you can explain to funders and partners. Most frameworks combine three things. The first is an assessment of the problem and the factors behind the risk behaviours. The second is a means of selecting from those factors and using theory to develop program components. The third is an assessment of the resources available, meaning funds, people, materials, and time. In other words, a framework asks what is causing the problem, which causes you will act on and how, and what you have to work with.
The second of those three, the means of selecting, is where a planner first meets the problem of too many factors, and Green and Kreuter offer a simple tool for it.
5.2 Picking Your Battles: Importance and Changeability
Green and Kreuter, the authors of PRECEDE-PROCEED, give planners a simple discipline for the selecting step. Each candidate factor is rated on two things. The first is importance: how strongly is the factor linked to the health problem, and how widespread is it? By this we mean that a factor is important when changing it would make a real difference to the problem across many people. The second is changeability: is there evidence that a program can move the factor within the time and resources available? A factor can be very important and still hard to change, so the two ratings must be made separately. Plotting the ratings on a two-by-two grid then sorts the candidates into four groups, and each group calls for a different response. Factors high on both importance and changeability become the program's focus, because acting on them is both worthwhile and achievable. Factors that are important but hard to change call for an innovative program with evaluation built in, so that the planners learn whether the new approach works. Factors that are changeable but unimportant are a low priority, except where a visible early win matters politically. Finally, factors low on both lead to no program at all.
Interactive 6.3 lets you try the grid on a falls case. A health unit has identified six factors behind falls among adults over 75. Move the two sliders for each factor to rate it from 1 to 5 on importance and changeability, press Plot the matrix, and read the verdict; then change a rating or two and plot again to see how the verdict shifts.
📊 Interactive 6.3: Pick your battles.
A health unit has identified six factors behind falls among adults over 75. Rate each on importance and changeability from 1 to 5, then plot them. There is no single right answer; the verdict tells you what your ratings imply.What your ratings imply
The matrix is a discipline for thinking rather than a formula. By this we mean that it organizes a planner's judgement but does not replace it, and two things can override it. First, the community's own priorities, gathered in the first PRECEDE phase, may put an item the epidemiology rates as minor at the top of the list. Second, a factor that scores low on changeability for a health unit acting alone may score high once a municipal council, a pharmacy chain, or a provincial ministry joins in, which is the point the multilevel theories in the first half of this lesson were making. When a factor is important but lies wholly outside the planner's authority, such as a tax set by another level of government, the usual response is to make it an advocacy target for the partners who can change it rather than a program component.
Picking battles, then, means rating candidate factors on importance and changeability, with the community's priorities and the partners who join in able to change the result. The selection step sits inside a larger process that every planning model shares.
5.3 Three Components of Any Planning Process
Every planning approach, whatever its name, involves three steps. Figure 6.2 draws them (assessment, intervention, and evaluation) as a continuous feedback loop.
Figure 6.2. The three-part process found in every planning model, drawn as a continuous feedback loop.
The first component is assessment, or diagnosis, of the health problem and its causal and supporting factors. The second is the development of an intervention based on the assessment, targeting selected factors and using appropriate theory to guide the components. The third is evaluation, which determines whether the intervention was implemented as planned and whether it actually affected the factors it was meant to affect. This sequence is common sense, but one point deserves stress: the three parts work as a continuous feedback loop rather than a straight line with an end. What you assess directs which theories apply, what intervention to undertake, and what variables to evaluate, while the way the program works or fails in the real world feeds back into the assessment and modifies the program in turn. In a case question, a program that revises its assessment because of what its evaluation found is the loop in action. The best-known framework that builds on this three-part process is PRECEDE-PROCEED.
6.1 PRECEDE-PROCEED: What the Name Says and Why It Is Useful
PRECEDE-PROCEED is one of the most well-known approaches for planning, implementing, and evaluating health promotion programs (Green and Kreuter, 1999). Its title says a lot about what it entails: a process of assessment and planning before putting a program in place (PRECEDE), followed by implementation and evaluation of the program (PROCEED). Read the name as before and after, which is the useful meaning; the sequence is the three-part process of 5.3 laid out in nine phases. For the record, PRECEDE is an acronym for Predisposing, Reinforcing, and Enabling Constructs in Educational and Environmental Diagnosis and Evaluation, and PROCEED for Policy, Regulatory, and Organizational Constructs in Educational and Environmental Development. The three constructs in the first acronym are the heart of phase 4, described in 6.3.
Among the many planning frameworks, PRECEDE-PROCEED is useful in part because it is so comprehensive. It is designed to work within an ecological framework, which means it expects factors at several levels to contribute to a health problem, and it helps guide the selection of the factors to target among them. In addition, it is oriented toward the community or local level. The reason for this orientation is that the community is where the factors behind a health problem are easiest to identify, because planners can talk to residents, gather local data, and see the conditions for themselves. It is also where, on a reasonable assumption, resources are best organized, because local agencies, businesses, and volunteer groups can be brought together around a shared problem in a way that is difficult to arrange at the provincial or national level. Green and Kreuter put the orientation in a sentence: health behaviour "is seen increasingly not as isolated acts under the autonomous control of the individual, but rather as socially conditioned, culturally embedded, economically constrained patterns of living." Their definition of lifestyle carries the same view, that health behaviour is not really a distinct domain from behaviour in general.
Definition to remember: lifestyle, as Green and Kreuter define it
Lifestyle is "a complex of related practices and behavioral patterns in a person or group, maintained with some consistency over time. It includes conscious health-directed behavior as well as unconscious health-related behavior and practices pursued for non-health purposes but with health consequences or risks." An everyday habit such as toothbrushing fits every clause.
With the name and the orientation in place, the framework's first five phases, the assessments, can be followed one at a time.
6.2 The Five PRECEDE Assessment Phases
The framework runs from the broadest assessment of the problem to the more targeted and specific, then to the administrative and community resources for the program, then to implementation and evaluation. By this we mean that each assessment phase narrows the focus of the one before it, so that the planner moves step by step from a community's general situation to the specific factors a program will target. In brief, the five assessment phases are these. Phase 1, the social assessment and situational analysis, asks how the health problem fits into the community's larger situation and which problems the community considers most important; it is rarely completed in full, because funders usually define the problem first. Phase 2, the epidemiological assessment, establishes the nature and extent of the problem, who is affected, and the trends, and it is often where planning actually begins. Phase 3, the behavioural and environmental assessment, identifies behavioural risk factors (what people do) and environmental risk factors (conditions around them). Phase 4, the educational and ecological assessment, goes back one more causal level to the predisposing, enabling, and reinforcing factors behind those risks, and it is where theory enters. Phase 5, the administrative and policy assessment, takes stock of the organizations, politics, barriers, assets, staff, and funding the program will need.
The descriptions below follow a hypothetical rust-belt community, North Hill County, through the phases, with a Canadian parallel: a health unit in a small British Columbia interior city, call it Cedar Valley, working on falls among adults over 75, the leading cause of injury-related hospitalization among older adults in Canada. Each phase below gives a fuller description of the phase, the North Hill County illustration, and the Cedar Valley parallel.
Phase 1: Social assessment and situational analysis
What it is. The broadest assessment, and one you will not always have the luxury of doing. It asks how the health problem relates to what else is happening in the community, and at what problems the community should direct its resources. In reality you will rarely complete it in full, because the health problem is usually already defined by federal, provincial, local, or foundation funders: you respond to an agency's priorities, or look for an agency that funds the problem you consider important.
North Hill County. Over 20 years the county has suffered the decline typical of rust-belt communities: the industries that fuelled the economy have gone, leaving high unemployment and few resources. It has high rates of alcohol misuse and comorbid conditions such as liver disease and drunk-driving crashes. The social assessment asks whether the alcohol misuse is related to unemployment, to a lack of services, or to young people who have little to do and see little future. You may find the alcohol problem is inseparable from the larger situation, which shapes what kind of program can work.
Cedar Valley. An aging population on fixed incomes, long winters, and a seniors' centre whose members say what worries them most is losing independence: that, and not injury statistics, is the language the community uses.
Phase 2: Epidemiological assessment
What it is. More specific, and often where planning actually begins: the nature and extent of the problem, who is affected, and the trends, which narrows the target in both conditions and populations before decisions about theory. For example, if local data showed high rates of smoking-related cancers among women and of respiratory problems among young children, you would have three populations to consider: women, especially women who smoke; mothers who smoke and expose their children; and children in smoking households.
Cedar Valley. Hospital and emergency department data show the rate and trend of fall-related admissions among residents over 75; death registrations show fall-related deaths. The target narrows to community-dwelling adults over 75, particularly those living alone.
Phase 3: Behavioural and environmental assessment
What it is. Knowing what and whom you will target does not tell you how. This phase identifies risk factors of two kinds. Behavioural risk factors are internal or related to persons: they are what people do. Environmental risk factors are external to persons: conditions or situations around them, such as a factory upstream polluting the drinking water. These data are less regularly collected than epidemiological data, but sources exist: national behavioural surveys such as the CDC's Behavioral Risk Factor Survey and Monitoring the Future in the United States, WHO and United Nations data on global problems, and local surveys and occupational, environmental, and housing agencies.
North Hill County. Alcohol use is the behavioural risk factor behind liver disease and drunk-driving crashes; smoking is the behavioural risk factor behind cancer in women and respiratory problems in children. The county's roads may be an environmental risk factor if they are narrow, have poor visibility, or are not well lit.
Cedar Valley. Behavioural: inactivity, several sedating medications, unsafe footwear, reluctance to use a walker. Environmental: icy sidewalks, poor street lighting, homes with loose rugs, no grab bars, and steep stairs. Canadian counterparts of the American surveys include the Canadian Community Health Survey from Statistics Canada and provincial health authority surveys.
Phase 4: Educational and ecological assessment
What it is. Go back one more causal level to the factors that contribute to the risks: attitudes, decision-making, social and cultural influences, and to a degree broader social and economic factors. By this we mean that phase 3 told you what people do and what conditions surround them, and phase 4 asks why: why people drink heavily, or why older adults do not use the supports available to them. This is where the theories in this course come in, because each of them is an explanation of exactly these reasons, from individual beliefs to social norms to economic constraints. Green and Kreuter sort the findings three ways. Predisposing factors: a population's knowledge, attitudes, beliefs, values, perceptions, and genetic predispositions. Enabling factors: the skills, resources, and barriers that help or hinder the desired behaviour. Reinforcing factors: the rewards received or available when people adopt a behaviour, healthy or unhealthy, such as reduced insurance rates for safe driving. Data are scarcer still, and more than for any other phase you may have to collect them yourself.
North Hill County. With few local jobs, young people either leave or carve out a social role where they are, and one way to do that is to drink a lot and do crazy things that are then talked about and glorified. This is a social act, because a group legitimizes the reputation. Those attitudes are predisposing factors. Easy availability of alcohol at bars unconcerned about driving, and the lack of programs and public information, are enabling factors. Preferential hiring of people with clean driving records would be a reinforcing factor for reducing abuse. Of course, it could just be about boredom, and even that would be useful to know.
Cedar Valley. Predisposing: the belief that falls are a normal part of aging, and pride that resists a walker. Enabling: no affordable strength and balance class, no pharmacist medication review, a long waiting list for home assessments. Reinforcing: adult children who praise a parent for staying active, and a doctor who asks about falls at every visit.
Phase 5: Administrative and policy assessment
What it is. By now you have a connected picture, the beginnings of a logic model tying individual, social, environmental, and behavioural factors to the health conditions you want to change. Phase 5 is more practical: after the brilliant design, you still have to implement it in an actual community, so you take stock of the organizations, committees, task forces, and local politics, the barriers and assets, and the staff, equipment, facilities, and funding you need. The relevant theories concern implementing policies, working with organizations and community structures, funding, and public involvement.
North Hill County. Bars are desperate for business. When the county council tries to pass a law holding bars liable for injuries from drinking and driving, the bar and restaurant association defeats it: a political barrier. The planners then convince local car dealers to lower interest rates on car payments for drivers with a DUI-free record and for documented designated drivers. The dealers become a community resource, because they believe it will help their business.
Cedar Valley. Assets: an active seniors' centre, a pharmacy willing to do medication reviews, an empty recreation hall on weekday mornings, and a volunteer fire department that already does home visits. Barrier: the council has just cut the snow-clearing budget, and the health unit has no authority over sidewalks. The council becomes the first advocacy target; the fire department's visits become the vehicle for home hazard checks.
Read across the five items and the narrowing is plain: each phase takes the findings of the one before and asks a more specific question, until phase 5 turns from what to change to what the community can bring. Two parts of the assessment need a closer look, the vocabulary of phase 2 and the sources of information for every phase.
6.2.1 The Vocabulary of the Epidemiological Assessment
Phase 2 relies on a small vocabulary that every planner needs. The first pair of terms describes two kinds of data. Morbidity data are data on the incidence and prevalence of disease, in other words on how much illness or injury there is; mortality data, in contrast, are data on deaths. For example, in Cedar Valley, fall-related hospital admissions are morbidity data and fall-related deaths are mortality data. A third term describes how conditions relate to each other: conditions that are related, such as obesity and diabetes, are said to be comorbid, because one raises the risk of the other or they tend to occur together. The last pair describes two ways of counting cases. Incidence is new cases, the current rate at which a problem is being reported. Prevalence, however, is existing cases, the spread or distribution of the problem in a population at a given time. By this we mean that incidence tells you how quickly a problem is arising, while prevalence tells you how much of it is present. The five cards give the full definitions; several draw their examples from the two running cases.
Morbidity data
Data on the incidence and prevalence of disease. In Cedar Valley, the count of fall-related hospital admissions and emergency visits among residents over 75 is morbidity data.
Mortality data
Data on deaths due to disease and other factors. Fall-related deaths recorded in the provincial death registry are mortality data.
Comorbidity
Some health conditions are related, or comorbid. For example, obesity and diabetes are comorbid because obesity is a risk for diabetes. In North Hill County, alcohol misuse, liver disease, and drunk-driving injuries travel together.
Incidence
New cases, or the current rate at which a disease or health problem is being reported. It may be higher or lower than the prevalence rate if the problem is on the rise or dropping.
Prevalence
Existing case rates of a disease or health problem in a population: the spread or distribution of the problem in a population or an area. A rising incidence with a stable prevalence tells a different story from the reverse.
Of the five terms, the pair most often confused is incidence and prevalence. A rising number of new cases with a stable number of existing cases means incidence is rising while prevalence holds steady, and the reverse pattern tells a different story about the problem.
6.2.2 Where the Data Come From
The phase descriptions above set out what each phase asks. Table 6.3 adds two things it does not carry for every phase: where the information comes from, and what it tells you.
Table 6.3. The five PRECEDE assessment phases, the sources of information for each, and what each tells the planner.
| Phase | Sources of information | What it tells you |
|---|---|---|
| 1. Social assessment | Social indicators: employment, health services, education, crime, housing, parks and recreation | The interplay between health and other factors, potential quality-of-life outcomes, and the importance the community places on the issue |
| 2. Epidemiological assessment | Local surveillance data from public health and other sources such as schools, emergency rooms, and police | The nature and extent of the problem, patterns and trends, and who is affected |
| 3. Behavioural and environmental assessment | Local surveys, focus groups, schools, community groups, unions; environmental, occupational, and public health agencies | Behaviours and environmental conditions you may need to target |
| 4. Educational and ecological assessment | Local surveys, focus groups, research institutions, NGOs, schools, work-related groups; data you collect yourself | The close-in factors, attitudes, norms, and support systems, that you will often need to address first |
| 5. Administrative and policy assessment | Key community leaders, focus groups, surveys of nonprofits, colleges, and universities | The practical details of putting a program into action |
Notice in Table 6.3 that data become harder to find as the phases go deeper. Public health surveillance holds the morbidity and mortality data of phase 2; behavioural surveys, such as the Canadian Community Health Survey from Statistics Canada, hold much of the risk factor data of phase 3; and for the attitudes, norms, and supports of phase 4 you may have to collect the information yourself. Phase 4 is also where the assessment meets theory, and its three-way classification deserves its own subsection.
6.3 Predisposing, Enabling, and Reinforcing Factors
Green and Kreuter sort the findings of the educational and ecological assessment three ways. The purpose of the sorting is practical: it shows how each factor acts on the behaviour, which in turn suggests what a program component would have to change. Predisposing factors are a population's knowledge, attitudes, beliefs, values, perceptions, and genetic predispositions. By this we mean what people bring with them to the behaviour, before they ever try it: what they know, what they believe, and how they feel about it. Enabling factors are the skills, resources, and barriers that help or hinder the desired behaviour. In contrast with predisposing factors, which sit in people's heads, enabling factors are about whether the behaviour can actually be done: whether people have the skills, whether a service or product is available, and what stands in the way. Reinforcing factors are the rewards received or available when people adopt a behaviour, healthy or unhealthy, such as reduced insurance rates for safe driving. They come after the behaviour and make it more or less likely to be repeated. For example, in North Hill County, attitudes that glorify heavy drinking as a way to build a reputation are predisposing, because they are beliefs and values people hold before they drink. Easy availability of alcohol at bars unconcerned about driving, and the lack of programs and public information, are enabling, because they make heavy drinking easier and a safer choice harder. Preferential hiring of people with clean driving records would be reinforcing for reducing abuse, because it rewards the healthier behaviour after it occurs. A quick test helps: predisposing factors come before the behaviour and sit in people's heads, enabling factors make the behaviour easier or harder to do, and reinforcing factors come after it as a reward or a cost.
Common confusion: phase 3 risk factors or phase 4 factors?
Phase 3 and phase 4 classify different things. Phase 3 sorts the risks themselves: a behavioural risk factor is something people do (drinking heavily, taking several sedating medications), and an environmental risk factor is a condition around them (icy sidewalks, narrow unlit roads). Phase 4 sorts the factors behind those risks into predisposing, enabling, and reinforcing. The same situation can appear in both phases: icy sidewalks are an environmental risk factor in phase 3, and the council's snow-clearing budget that leaves them icy is an enabling factor in phase 4. Check which phase a question asks about before you classify.
Worked Example 6.3 returns to Cedar Valley once its five assessments are complete. It applies both classifications, and the importance and changeability grid from 5.2, to the four components the health unit has drafted. Watch for components that work through more than one factor type.
Worked Example 6.3: Cedar Valley Chooses Its Components
The Cedar Valley health unit has finished its five assessments and drafted four components: (1) a free twice-weekly strength and balance class at the recreation hall; (2) pharmacist medication reviews through the community pharmacy; (3) home hazard checks and grab bar installation during volunteer fire department visits; and (4) a presentation to municipal council, backed by the seniors' centre, asking for restored sidewalk snow clearing on the routes to the pharmacy, the grocery store, and the centre. A regional manager questions whether the fourth component belongs in a health unit program at all.
Step 1, the behaviour and the population. The health problem is falls, and the population is community-dwelling adults over 75 in Cedar Valley, particularly those living alone.
Step 2, classify each component by its phase 3 risk factor and its phase 4 factor type.
- The strength and balance class targets a behavioural risk factor, physical inactivity, and works mainly through an enabling factor, since no affordable class existed; a free, social class can also become reinforcing once neighbours attend together.
- The medication review targets a behavioural risk factor, the use of several sedating medications, and is enabling, since no review was available before; it may also shift a predisposing belief that a long medication list is normal.
- The home hazard check targets environmental risk factors inside the home (loose rugs, poor lighting, no grab bars) and is enabling, because the barrier was a waiting list and the cost of installation.
- The council presentation targets an environmental risk factor, icy sidewalks, and works through an enabling factor at the municipal level: the budget decision that determines whether the routes are walkable in winter.
Step 3, apply importance and changeability to the fourth component. Icy sidewalks rate high on importance for a town with long winters and many residents over 75 living alone, and the phase 2 and phase 3 findings (fall-related admissions and the environmental pattern behind them) support that rating. Changeability is low for the health unit acting alone, which is the manager's point. The phase 5 assessment, however, found an asset that raises it: the seniors' centre, which can bring its members to council. That moves the component from the innovate-and-evaluate group toward program focus.
Step 4, name the evidence and decide. The case for keeping the component rests on phase 3 (the environmental risk factor) and phase 5 (the political barrier of the budget cut and the community asset that answers it). A program that teaches balance and fixes homes while the route to the pharmacy stays icy leaves the environment pushing against the behaviour, which is the lesson of the multilevel theories. The component should carry a process measure, such as whether the presentation was made and the routes were cleared, so its contribution can be evaluated.
What the example shows. The two classifications and the grid work together: phase 3 says what risk a component targets, phase 4 says how it works, and the grid says whether it is worth doing. The mistake to avoid is dropping a component because the health unit cannot change the factor alone; partners found in phase 5 change the changeability rating.
The example shows how the five assessments feed a program design. By the end of phase 5 the planner has moved from a health problem to a set of named factors, a rationale for choosing among them, the theories that speak to each, and an inventory of what the community can bring.
This section has moved from the difficulty of turning theory into practice, through the importance and changeability ratings used to pick battles, to the three components of every planning process, the name and orientation of PRECEDE-PROCEED, its five assessment phases, and the predisposing, enabling, and reinforcing factors of phase 4. The key takeaways and knowledge check that follow review the section, and the next section picks up at the moment the program goes live.
Key Takeaways
- 5.1 Moving from theory to practice is one of the hardest steps in public health, because every theory is a piece of an interrelated puzzle and no program has the time or money to address every factor an ecological assessment turns up. You have to pick and choose, and it is best to do so with a planning framework. Most frameworks combine three things: an assessment of the problem and the factors behind it, a means of selecting factors and using theory to build program components, and an assessment of resources (funds, people, materials, and time). If a question proposes addressing every factor at once, the lesson's answer is that limited resources force a choice.
- 5.2 Green and Kreuter's tool for choosing rates each candidate factor on importance (how strongly it is linked to the problem and how widespread it is) and changeability (whether there is evidence a program can move it with the time and resources available). High on both means program focus; important but hard to change means an innovative program with evaluation built in; changeable but unimportant means low priority; low on both means no program. Two things can override the grid: the community's own priorities, and partners who raise changeability. When a factor is important but outside the planner's authority, it becomes an advocacy target rather than a program component.
- 5.3 Every planning model shares three components: assessment of the problem and its factors, an intervention that targets selected factors using theory, and evaluation of whether the program was implemented as planned and changed those factors. They form a continuous feedback loop, because what you assess directs the theory, the intervention, and the measures, and what happens in the real world feeds back into the assessment. In a case, a program that redesigns itself after its evaluation is the loop at work.
- 6.1 PRECEDE-PROCEED's name describes its sequence: assessment and planning come before the program is put in place (PRECEDE), and implementation and evaluation follow (PROCEED). It is useful because it is comprehensive, works within an ecological framework, helps guide the selection of factors, and is oriented to the community level, where factors are easiest to identify and resources are best organized. Its authors see health behaviour as socially conditioned, culturally embedded, economically constrained patterns of living rather than isolated individual acts. You do not need the full acronyms; you do need the before-and-after meaning.
- 6.2 The five PRECEDE assessments narrow step by step: the social assessment places the problem in the community's larger situation (and is rarely completed in full because funders usually define the problem), the epidemiological assessment measures its extent using morbidity, mortality, incidence, prevalence, and comorbidity, the behavioural and environmental assessment finds the risk factors, the educational and ecological assessment finds the factors behind them, and the administrative and policy assessment inventories resources and politics. Data get scarcer as the phases go deeper, and phase 4 information you may have to collect yourself. For incidence and prevalence questions, remember that incidence is new cases and prevalence is existing cases.
- 6.3 Phase 4 sorts the factors behind risk behaviours into three types. Predisposing factors are knowledge, attitudes, beliefs, values, and perceptions; enabling factors are the skills, resources, and barriers that help or hinder the behaviour; reinforcing factors are the rewards that follow it. In North Hill County, attitudes that glorify heavy drinking are predisposing, easy alcohol availability is enabling, and preferential hiring of clean drivers is reinforcing. The trap is mixing the phases: behavioural and environmental risk factors belong to phase 3, so check which phase the question asks about, and remember that one component can work through more than one factor type.
1. A new planner says that because the ecological perspective identifies factors at every level, a good program must address all of them. What is the lesson's response?
2. Using Green and Kreuter's importance and changeability ratings, a factor is rated highly important to the health problem but, on current evidence, hard to change. Which response fits that group?
3. After its first year, a falls prevention program finds that its home hazard checks are reaching very few people who live alone. The health unit revisits its assessment, learns that people living alone rarely answer unscheduled visits, and redesigns the component around booked appointments. Which feature of program planning does this illustrate?
4. What does the name PRECEDE-PROCEED say about the approach?
5. In reality, most planners will not complete the phase 1 social assessment in its full form. What is the reason?
6. A health unit reports that the number of new fall-related hospital admissions among residents over 75 rose this year, while the total number of residents currently living with a fall-related disability stayed the same. In the vocabulary of the phase 2 epidemiological assessment, what changed?
7. A planner in North Hill County learns that bars sell alcohol with little concern about whether customers will drive, and that there are no public information programs about drunk driving. In the phase 4 educational and ecological assessment, how would Green and Kreuter classify these findings?
8. In Worked Example 6.3, Cedar Valley's volunteer fire department adds home hazard checks and grab bar installation to its visits. Which pair of classifications fits this component?
✦ Pass the knowledge check with 100% to continue
PROCEED and the Risk and Protective Factors Model
⏱ Estimated reading time: 16 minutes
Overview
This section finishes PRECEDE-PROCEED with the phases that follow the assessments: implementation and the three types of evaluation, illustrated with British Columbia's Take Home Naloxone program. It then introduces a second planning model built for a narrower purpose, youth risk behaviour: the risk and protective factors model synthesized by Hawkins and Catalano. The two models look different on the page, and the closing point of the section is that they share the same spine.
Learning Objectives
- 6.4 Describe phase 6 of PRECEDE-PROCEED, implementation, and explain when adopting an evidence-based program is preferable to designing your own.
- 6.5 Give the external and internal reasons for evaluating a health promotion program.
- 6.6 Distinguish process, outcome, and impact evaluation (phases 7, 8, and 9) by what each measures and over what time frame, and apply them to a Canadian program.
- 7.1 Explain the logic of the risk and protective factors planning model and its origins in the work of Hawkins and Catalano.
- 7.2 Sort factors into the model's risk and protective domains and explain the role of resilience.
- 7.3 Apply the model's three phases of assessment, implementation, and evaluation to a community case.
- 7.4 Compare the risk and protective factors model with PRECEDE-PROCEED.
6.4 Phase 6: Implementation
This phase does not need a long explanation. It means putting your program in place after using the assessment process to design it; that is what is meant by proceed. One reminder matters. In today's program environment there is often a requirement to use evidence-based or model programs that are responsive to the community situation. Once you have made an assessment, there may be programs that already show consistent evidence of effectiveness with a similar population and situation, and it is sometimes advisable to use one instead of developing your own; many agencies maintain registries to help. The advantage of adopting such a program is that someone else has already tested it, so you start with evidence that it can work rather than hoping that a new design will. The assessment still comes first, however. A model program adopted without one is the cookie-cutter mistake: a program that worked for one population and situation is assumed to fit another without checking whether the problem, the factors behind it, and the community's resources are actually similar. The assessment is what tells you whether a model program matches your community, or whether it would be aimed at factors that are not the ones at work there. Once the program is running, the question becomes how you would know whether it is working.
6.5 Why Evaluate a Program
The last three phases involve evaluation: the data you collect about whether your program is working, in the sense of changing the health problem it is meant to affect. There are two families of reasons to evaluate. The first is external accountability. Dollars are scarce, and programs must show evidence of success. If your program works but you have no data to show it, funding may not continue; other communities will lack the data to argue for adopting it; and an innovative approach may not get the attention it deserves. The second is internal: evaluation tells you whether the program is working, in what ways, and why, so that you can adjust it. Notice that neither family claims that evaluation itself changes participants' behaviour; its value lies in accountability to others and in learning for the program. The three types of evaluation that serve both purposes correspond to phases 7, 8, and 9.
6.6 Process, Outcome, and Impact Evaluation
The three types differ in what they measure and in how long after the program they look. Process evaluation (phase 7) asks whether you did what you said you would do. By this we mean that it assesses program implementation rather than effects on people, mainly through tracking records of activities, materials, and the people reached. Outcome evaluation (phase 8), in contrast, looks at what changed as a result of the program in the short term, the kind of change that might happen in a year or two, maybe three. For example, it measures changes in knowledge, in policies and practices, in community awareness and systems, and in the way people use services. These correspond to the factors identified in phases 3 and 4, so outcome evaluation checks whether the program moved the factors it was designed to move. Impact evaluation (phase 9) assesses long-term effects over several years or more: changes in the health condition itself, in quality-of-life indicators, or in substantial policy or system change. These are the factors usually identified in phase 1 or 2, so impact evaluation asks whether the problem that started the planning has actually improved. Notice that each evaluation phase answers to an earlier assessment phase, which is the feedback loop of 5.3 built into the framework. The tabs apply each type to British Columbia's Take Home Naloxone program, run through the BC Centre for Disease Control since 2012, which trains people likely to witness an opioid overdose and gives them a kit to reverse it. Open all three tabs and compare what each type would count.
Process evaluation: did you do what you said you would do?
An assessment of program implementation. If you said you would develop brochures and flyers, did you? Did you distribute them, to how many people, and to the people the program is trying to reach? If you set up a website, how many hits does it receive? The data are primarily tracking records of what you did, and they also help in assessing how much of the program led to the result.
Take Home Naloxone: the number of sites registered to distribute kits, people trained, and kits shipped, and whether the sites are in the settings, such as emergency departments, harm reduction sites, and corrections, where overdoses are most likely to be witnessed.
Outcome evaluation: short-term effects
An assessment of the short-term effects, the kinds that might happen in a year or two, maybe three. Every intervention's ultimate goal is to eliminate a condition and improve quality of life, but change of that magnitude does not happen quickly, so outcomes usually refer to intermediate changes expected to contribute to it: changes in knowledge, in policies and practices, in community awareness and systems, and in the way people use services, corresponding to what you identified in phases 3 and 4. You measure these factors before the intervention and follow up with the same measures.
Take Home Naloxone: whether trained participants can recognize an overdose and use the kit, the reversals reported when kits are replaced, and policy changes such as Health Canada's decision in 2016 to make naloxone available without a prescription.
Impact evaluation: long-term effects
An assessment of the long-term effects, the kinds that might happen in several years or more: changes in the health condition itself, in quality-of-life indicators, in health behaviour if significant, in utilization patterns, or in substantial policy or system change, the factors usually identified in phase 1 or 2 and sometimes 3. Many programs are not funded long enough to track this, so it may require a separate longitudinal study that follows a community or group for many years with the same measures.
Take Home Naloxone: the effect on overdose deaths in the province over years. Because the program operates inside an evolving drug supply and alongside other measures, isolating its impact requires modelling rather than simple before-and-after counts: the complexity critique from the first half of this lesson arriving in the evaluation phase.
Set side by side, the three tabs show that one program yields three kinds of evidence, each answering a different question on a different time scale: what was delivered, what changed soon after, and what happened to the health problem over years. To classify an evaluation finding, ask two things, whether it measures the program's own activities or a change in people and policies, and how long after the program the change would appear.
Caution: outcome and impact are labelled differently in some sources
The labels outcome evaluation and impact evaluation are not used consistently across the field. This lesson follows the usage in which outcome is the short-term measure and impact the long-term one. Green and Kreuter's own PRECEDE-PROCEED diagrams use the labels the other way around, with impact for the near-term changes in predisposing, enabling, reinforcing, behavioural, and environmental factors and outcome for the eventual change in health and quality of life. Whenever you read an evaluation report, check how it defines the two words before you compare it with anything else. The idea underneath, short-term versus long-term, is stable even when the labels are not.
In addition to the three types tied to PRECEDE-PROCEED's phases, two other general types of evaluation are worth knowing by name. Formative evaluation is used to develop a program. By this we mean that it gathers information while a program is still being designed or tried out, for example by testing draft materials with a few members of the intended audience, so that problems can be fixed before full launch. Summative evaluation, in contrast, is used to draw broad or in-depth conclusions from the evaluation data, usually once a program has run, to judge what it achieved overall. Some evaluations also try to analyze the costs and benefits of a program. With implementation and the three types of evaluation, the nine phases of PRECEDE-PROCEED are complete, and the section turns to the second planning model.
7.1 The Logic of the Risk and Protective Factors Model
The risk and protective factors model is the second example of a planning approach in this lesson. It is widely used by United States federal agencies, including SAMHSA, the CDC, and the Office of Juvenile Justice and Delinquency Prevention, to plan and fund programs on youth risk behaviour: HIV/AIDS risk, early sexual activity, adolescent pregnancy, tobacco, alcohol, and drug use, and violence. The same three processes of assessment, intervention, and evaluation remain. The key difference is that it is a theory-based planning model focused on certain health risk behaviours as the targets of change, built from a synthesis of research about factors correlated with those behaviours. By this we mean that, where PRECEDE-PROCEED is a framework within which the planner assesses the community and then selects the theories that fit, this model starts from factors that research has already linked to youth risk behaviour. For example, the research identifies risk factors in the family, such as child development problems, parent-child interaction, and family conflict and abuse; in the community, such as community poverty; in a young person's experience, such as previous victimization by violence or regular witnessing of it and the absence of a positive adult role model; and in the individual, such as differences in personality traits, affective states, and behavioural skills. However, most research points not to any one of these factors but to multiple causes, cumulative exposure, or interactions among individual, family, and community risk factors over the course of development, a pattern described as a web of risk factors.
Hawkins, Catalano, and colleagues synthesized this research into a widely used approach that functions as a planning tool (Hawkins, Catalano, and Miller, 1992). The model lays out an algorithm of factors, or forces, that over the youth development process increase or decrease the likelihood that a given youth will engage in problem behaviours: violence, delinquency, substance use, school dropout, HIV/AIDS risk behaviour, and others. By this we mean that the model treats a young person's chance of problem behaviour as the combined result of many forces acting over years, some pushing toward the behaviour and some pushing away from it. Risk factors increase the likelihood; protective factors reduce it. Because the forces accumulate, no single factor decides the outcome. For example, a young person whose family is in conflict may still avoid violence if they are closely bonded to a coach and attend a school with clear standards, whereas the same family conflict combined with violent peers and a high-crime neighbourhood tips the balance the other way. It helps to think of the process as a box with inputs and outputs. Figure 6.3 draws that box, with risk factors and protective factors as its inputs and the likelihood of problem behaviour as its output.
Figure 6.3. Youth development as a box: put in many risk factors and few protective factors and you are highly likely to get health risk behaviour as an output; put in few risks and many protective factors and you are less likely to.
The box speaks in likelihoods, not certainties. A young person with many risk inputs and few protective ones is more likely to show problem behaviour, but the model does not say that every such young person will, and the inputs, unlike the output, are what a program can change. The next step is to see how the model organizes those inputs.
7.2 Risk Domains, Protective Domains, and Resilience
Under the Hawkins and Catalano model, risk factors are organized into domains, and protective factors, which are not as well specified, into a smaller set of similar domains. Table 6.4 lists both sets with a description of what each domain covers.
Table 6.4. Risk factor domains and protective factor domains in the Hawkins and Catalano model, with what each domain covers.
| Type | Domain | What it covers |
|---|---|---|
| Risk | Individual | Biological and psychological dispositions, attitudes, values, knowledge, skills, problem behaviours |
| Risk | Peer | Norms, activities, attachment to specific peer groups |
| Risk | Family | Family function, management, bonding, abuse and violence |
| Risk | School | Bonding to school, school climate, policy, performance |
| Risk | Community | Bonding, norms, resources, poverty level, crime, awareness and mobilization |
| Risk | Society / environment (sometimes) | Norms, policy, sanctions |
| Protective | Individual | Gender, intelligence, temperament |
| Protective | Social bonding | Attachment and commitment to positive, prosocial individuals and groups |
| Protective | Healthy beliefs and clear standards | In families, schools, and communities |
Table 6.4 also shows why the model needs protective domains at all. Not every youth exposed to risk factors becomes involved in risk behaviour; many do not, and they are said to have protective factors that counter or buffer the risks. Programs and research have focused on these under the general term resilience. By this we mean a young person's capacity to avoid risk behaviour despite exposure to risk, a capacity the model attributes to the protective factors around and within that person. The practical consequence has been a shift in emphasis: in addition to reducing risk factors, the more recent trend has been toward programs that support resilience or build protective factors. The complex, fluid interaction between an individual and the risk and protective factors in one or more domains has been described as a web of influence, an idea that traces back to Bronfenbrenner's ecological perspective: the thread that ties the planning half of this lesson back to the theory half.
Interactive 6.4 gives practice with the domains of Table 6.4. It presents eleven findings from a community assessment. Click each finding, decide first whether it is a risk factor or a protective factor, then click the domain where the Hawkins and Catalano model would place it; the feedback explains each placement.
📊 Interactive 6.4: Sort the risk and protective factors.
A community assessment has turned up the findings below. Click a finding, then click the domain where the Hawkins and Catalano model would place it. Decide first whether it is a risk factor or a protective factor.Individual (risk)
Peer
Family
School
Community
Society / environment
Individual (protective)
Social bonding
Healthy beliefs and clear standards
The findings that are hardest to place are usually the protective ones, because the protective domains are fewer and less well specified than the risk domains. When in doubt, ask first whether the finding raises or lowers the likelihood of problem behaviour, and only then choose the domain. With the domains in hand, the model can be followed through a planning cycle.
7.3 Using the Model: Assessment, Implementation, Evaluation
How is the approach used to plan, implement, and evaluate programs for adolescents and young adults? Thinking about it as we did for PRECEDE-PROCEED, there are three phases. In assessment, an agency identifies which risk and protective factors are present in the community; in implementation, it designs a program, or adopts a model program, matched to those factors; in evaluation, it uses process, outcome, and impact measures, many of them available as ready-made surveys and tools for this model. The descriptions below follow a hypothetical community that wants to address adolescent violence and alcohol and drug use through assessment, implementation, and evaluation.
Phase 1: Assessment
An agency, committee, or other organization collects information to identify which risk and protective factors are relevant to the violence and drug use in the community. In our example the agency finds: in the individual domain, a high prevalence of attitudes that treat violence as a way to gain a valued reputation among peers; in the community domain, easy access to weapons and a high crime rate; in the school domain, disorganized enforcement of no-violence policies; and a lack of opportunities for positive youth involvement. It also finds one protective factor: a relatively large number of volunteer and civic organizations interested in helping youth. The information comes from surveys such as the CDC Youth Risk Behavior Survey, police data, and the schools; a Canadian planner would draw on the Canadian Student Tobacco, Alcohol and Drugs Survey, provincial student health surveys, police-reported crime data, and school district records. The theories that speak to the findings concern social norms and attitudes about violence, access to weapons, adherence to school policy, and the mobilization of community resources.
Phase 2: Implementation
Based on the assessment, the agency can either design a program that targets the identified risk factors or adopt a model, evidence-based program already developed to target just those factors; for this planning model there is a well-developed system for identifying such programs. The program may have several components matched to the findings: a media campaign at schools and in the community focused on attitudes about violence and reputation; advocacy for strict enforcement of local weapons laws; a school component that draws on community volunteers to increase the adult, supervisory presence; and a mobilization campaign to help community organizations increase opportunities for youth participation outside school hours. Notice how the protective factor, the willing civic organizations, becomes the engine of the last two components. Which theories are useful for each? Think of social norms and communications theory for the campaign, policy and organizational change theory for the advocacy, and social bonding and social development theory for the volunteer and participation components.
Phase 3: Evaluation
Evaluation involves the same three types as PRECEDE-PROCEED, but in this model many surveys and evaluation tools already exist to determine whether a program affected the identified risk factors and supported the protective factors. Process evaluation tracks program activities. Outcome, short-term, evaluation assesses whether attitudes about violence, enforcement of weapons laws, school violence policies, and opportunities for positive youth involvement changed. Impact, long-term, evaluation, if you are able to do one, asks whether levels of violence among adolescents actually changed.
Across the three items, notice how the protective factor found in assessment, the willing civic organizations, became the engine of two implementation components and then something to evaluate. Worked Example 6.4 applies the same three phases to a Canadian case: a coalition in a prairie city preparing an application to Public Safety Canada's National Crime Prevention Strategy, which asks applicants to identify the risk and protective factors their project addresses. Watch for the component that builds on a protective factor rather than reducing a risk.
Worked Example 6.4: A Prairie City Applies for Federal Crime Prevention Funding
A coalition in a mid-sized prairie city, led by a youth agency, a school division, and an urban Indigenous Friendship Centre, is preparing an application to reduce youth violence in two neighbourhoods. Its assessment found: attitudes that respect is earned through fighting; weak attachment to school among students who have changed schools often; a shortage of supervised evening activities; and family instability linked to housing insecurity. It also found two protective factors: a Friendship Centre with cultural programming many youth already attend, and former gang members willing to mentor. Four components are under consideration: an evidence-based school program on conflict resolution; mentoring through the Friendship Centre; expanded evening drop-in hours at two community centres; and advocacy to the city for a housing stabilization fund for families with school-aged children.
Step 1, the behaviour and the population. The behaviour is youth violence, and the population is young people in the two neighbourhoods.
Step 2, assessment: assign each finding to a domain.
- Attitudes that respect is earned through fighting: individual risk domain (attitudes and values).
- Weak attachment to school after repeated moves: school risk domain (bonding to school).
- A shortage of supervised evening activities: community risk domain (resources and opportunities for positive youth involvement).
- Family instability linked to housing insecurity: family risk domain (family function and management), with a root in the community domain's poverty level.
- The Friendship Centre's cultural programming and the willing mentors: protective factors in the social bonding domain (attachment to positive, prosocial individuals and groups); the cultural programming also carries healthy beliefs and clear standards.
Step 3, implementation: match each component to a factor. The conflict-resolution program addresses the individual attitudes; mentoring addresses weak bonding by supplying a prosocial adult; the drop-in hours address the community shortage of supervised activities; and the housing advocacy addresses family instability at its source. Mentoring is the component that builds on a protective factor rather than reducing a risk, which is the direction recent programs have shifted toward, and it uses assets the community already has. The housing component is the kind of community-level pairing the CDC macroeconomic review recommended, without which the school and mentoring components risk being undercut.
Step 4, evaluation: design one measure of each type. A process measure is the number of mentoring matches made and sustained for six months, or the number of drop-in evenings delivered and youth attending. An outcome measure is a change, on the model's pre-built youth surveys within one to two years, in attitudes about fighting and in reported attachment to school and to a caring adult. An impact measure is the rate of police-reported violent incidents involving youth in the two neighbourhoods over five years or more. The impact measure is the one the coalition is least likely to be funded long enough to collect, which is the standing warning that many programs end before long-term change can be tracked.
What the example shows. The model carries a case from findings to domains, from domains to components, and from components to measures. The mistake to avoid is treating every component as risk reduction; a protective factor, once found, can be built on directly.
The example shows the model doing its planning work in a Canadian setting, and its three steps mirror the assessment, implementation, and evaluation of PRECEDE-PROCEED. With both planning models described, the last part of this section compares them.
7.4 Two Planning Models, One Spine
These two models are not the only planning models for health promotion programs. They were chosen to give a sense of what planning such programs involves and how it opens the door to selecting and applying theory. By this we mean that a planning model does not itself tell you which theory to use; it creates the points in the process, especially the assessment, at which a theory becomes useful because it explains the factors you have found. The key is the assessment, implementation, evaluation process, which helps ensure that a program is tailored to the real-life situation of a community. Because the assessment comes first, the program is built around the factors actually at work in that community rather than around a general idea of what usually causes the problem, and because evaluation feeds back into assessment, the program can be corrected when the real-life situation turns out to differ from the plan. Table 6.5 sets out how the two models are similar and how they differ.
Table 6.5. PRECEDE-PROCEED and the risk and protective factors model compared on seven features.
| Feature | PRECEDE-PROCEED | Risk and protective factors model |
|---|---|---|
| Core process | Assessment, implementation, evaluation, in nine phases | Assessment, implementation, evaluation, in three phases |
| Scope | Any health problem, any population; comprehensive and generic | Youth and adolescent health risk behaviours specifically |
| Relation to theory | A framework within which you select the theories that fit your assessment | Itself theory-based: built from a synthesis of research on risk and protective factors |
| Starting point | Social and epidemiological assessment of the community's problems and priorities | Identification of which known risk and protective factors are present in the community |
| Where the factors live | Predisposing, enabling, and reinforcing factors behind behavioural and environmental risks | Individual, peer, family, school, community, and societal domains, plus protective domains |
| Ready-made tools | Few; you assemble the data and instruments | Registries of model programs and pre-built surveys and evaluation instruments |
| Ecological roots | Green and Kreuter's community ecological orientation | Bronfenbrenner's web of influence, by way of Hawkins and Catalano |
Read Table 6.5 one more time with the multilevel theories in mind. The models differ in scope (any problem and population versus youth risk behaviour), in their relation to theory (a framework within which you select theories versus a model that is itself theory-based), and in the tools they supply (few versus registries of model programs and pre-built instruments). Yet both ask you to look at more than one level, both admit that you cannot act on every level at once, and both make you write down why you chose the factors you chose. That written rationale is what turns a multilevel theory from an abstraction into a plan. When a case asks which model to use, start with scope: a youth risk behaviour points toward the risk and protective factors model, and any other problem or population points toward PRECEDE-PROCEED.
This section has completed PRECEDE-PROCEED with implementation, the reasons for evaluation, and the three types of evaluation, introduced the risk and protective factors model with its logic, domains, and three phases, and compared the two planning models. The key takeaways and knowledge check that follow review this material.
Key Takeaways
- 6.4 Phase 6, implementation, means putting the program in place after the assessment has designed it, which is what "proceed" means. Where an evidence-based or model program already shows consistent effectiveness with a similar population and situation, adopting it is sometimes wiser than building your own, and agencies keep registries to help. The condition is the order: assessment first, then the choice to adopt. A model program adopted without an assessment is the cookie-cutter mistake, which assumes a program that worked for one population will fit another.
- 6.5 Programs are evaluated for two families of reasons. Externally, dollars are scarce, so a program that works but has no data may lose its funding, other communities will lack the evidence to adopt it, and an innovation may not get the attention it deserves. Internally, evaluation tells you whether, in what ways, and why the program is working, so you can adjust it. If a question lists a supposed reason that evaluation itself makes participants change, it is not one the lesson gives.
- 6.6 Process evaluation (phase 7) asks whether you did what you said you would do, using tracking records such as kits shipped or people trained. Outcome evaluation (phase 8) measures short-term change within a year or two in knowledge, policies, practices, awareness, and service use. Impact evaluation (phase 9) measures long-term change in the health condition itself, often through a separate longitudinal study that many programs are not funded long enough to run. To classify a finding, ask whether it counts the program's activities or a change in people, and over what time frame; and because some sources, including Green and Kreuter, swap the labels outcome and impact, always check the definitions a report uses.
- 7.1 The risk and protective factors model, synthesized by Hawkins, Catalano, and colleagues from research on youth problem behaviour, is a theory-based planning model widely used by United States federal agencies. Its logic is a box: risk factors raise the likelihood of problem behaviour, protective factors lower it, and many risks with few protections make risk behaviour a likely output. Remember that the model speaks in likelihoods, not certainties, and that it targets specific youth risk behaviours such as violence, substance use, and early sexual activity.
- 7.2 Risk factors fall into individual, peer, family, school, community, and sometimes society and environment domains; protective factors, less well specified, fall into individual characteristics (gender, intelligence, temperament), social bonding to prosocial people and groups, and healthy beliefs and clear standards. Because many youth exposed to risk do not engage in risk behaviour, research on resilience has shifted programs toward building protective factors. When sorting a finding, decide first whether it raises or lowers the likelihood of problem behaviour, then choose the domain; a caring mentor is a social bonding protective factor, while the absence of one is a risk.
- 7.3 The model runs in three phases. Assessment identifies which risk and protective factors are present; implementation designs or adopts a program whose components match those factors; and evaluation uses process, outcome, and impact measures, helped by ready-made surveys and tools. In the prairie city example, mentoring through the Friendship Centre built on a protective factor rather than reducing a risk. In a case, link every component to a named factor in a named domain, and look for protective factors the program can build on.
- 7.4 The two planning models differ in scope (PRECEDE-PROCEED is generic for any problem and population; the risk and protective factors model is for youth risk behaviour), in their relation to theory (a framework within which you select theories versus a model that is itself theory-based), and in ready-made tools (few versus registries and pre-built instruments). Both rest on the same spine of assessment, implementation, and evaluation, and both draw on the ecological perspective. To choose between them in a case, start with the problem and the population.
1. After a thorough community assessment, a health unit finds a registered, evidence-based falls prevention program that has worked with a similar population in a similar setting. A colleague objects that adopting it would be the cookie-cutter mistake. What is the lesson's view?
2. The lesson gives three consequences of running a program that works without collecting data to show it. Which of the following is NOT one of them?
3. Two years into a community naloxone program, evaluators report that trained participants can correctly recognize an overdose, that reported reversals have increased, and that a provincial policy now allows kits to be distributed through pharmacies. In the terms used in this lesson, which evaluation phase do these findings belong to?
4. The prairie city coalition in Worked Example 6.4 plans three measures: the number of mentoring matches sustained for six months; the change within two years in youth survey scores on attitudes about fighting and attachment to school; and the rate of police-reported violent incidents involving youth over five years. Which measure is the coalition least likely to be funded long enough to collect?
5. Two adolescents live in the same high-crime neighbourhood. One also has frequent family conflict, friends who use drugs, and no adult mentor; the other has a coach who sets clear expectations and a close bond with a teacher. Using the box image of the risk and protective factors model, which prediction fits?
6. A youth agency's assessment finds that most adolescents in a neighbourhood belong to at least one sports team or cultural group led by adults who state clear expectations about behaviour. In the Hawkins and Catalano model, this finding is best described as which of the following?
7. Which of the prairie city coalition's four proposed components in Worked Example 6.4 builds on a protective factor rather than reducing a risk?
8. Which statement correctly describes a key difference between the risk and protective factors model and PRECEDE-PROCEED, as this lesson presents them?
✦ Pass the knowledge check with 100% to continue
Final Review & Assessment
⏱ Estimated time: 25 minutes
Bringing It All Together
This lesson joined two halves that belong together. The first half argued that the theories reviewed so far, each working at one ecological level, cannot account for the way factors at different levels act on each other, and it offered theories built for that purpose. Bronfenbrenner's Ecological Systems Theory nests the developing person inside a microsystem of direct contact, a mesosystem that connects those settings, an exosystem the person never enters but is shaped by, a macrosystem of culture and law, and a chronosystem of time; its bioecological amendment added the person's own biology and the distinction between proximal and distal processes. Flay's Theory of Triadic Influence turned the layers into a map, sorting every factor by level of causation (ultimate, distal, proximal) and stream of influence (intrapersonal, social, cultural-environmental), and gave practitioners the waterfall: trace the influences upstream and intervene where it is feasible, remembering that proximal causes predict behaviour while distal and ultimate causes explain it. Problem-specific models from the CDC, WHO, and UNICEF reached the same conclusion from different directions, and every one of them carried the same critique: complexity.
The second half accepted the complexity and asked the practical question. You cannot address every factor, so you pick your battles, and it is best to do that with a planning framework. Every framework shares three components, assessment, intervention, and evaluation, linked in a feedback loop. PRECEDE-PROCEED works through them in nine phases: five assessments that narrow from the community's quality of life to the predisposing, enabling, and reinforcing factors behind behavioural and environmental risks and then to the resources and politics of implementation; implementation itself; and three evaluations, process, outcome, and impact. The risk and protective factors model, synthesized by Hawkins and Catalano for youth risk behaviour, runs the same three processes with a theory built in, sorting factors into risk domains and protective domains and treating problem behaviour as the output of a box whose inputs can be changed.
One thread runs through both halves. Behaviour is produced by interacting levels; a program that acts at one level alone is working against a current; and the discipline of assessment, whether through Bronfenbrenner's rings, Flay's grid, a PRECEDE phase, or a risk domain, is what turns a theory of that current into a plan for changing it. The final assessment asks you to move between the two halves in the same way: to trace a factor from a Bronfenbrenner system or a Triadic Influence stream into a PRECEDE phase or a risk domain, and to choose an evaluation that could show it changed.
Key Takeaways from this lesson
- You can now explain why a theory confined to one ecological level cannot account for how factors at different levels work together, and tell the two forms of theory apart: single-level theories read like equations, while multilevel theories read like branching paths through the layers that can run back upward. The practical reminder: when a case offers a single-level explanation, look for the factor at another level that it leaves unexplained.
- You can now describe Bronfenbrenner's five systems, classify real influences into them, and explain how the bioecological amendment added the person's own biology and the distinction between proximal and distal processes. You can also trace a story such as Rodney's, or a program such as Comprehensive School Health, through the levels. The most useful reminder is the participation test: a setting the person takes part in is microsystem, a link between two such settings is mesosystem, and a setting the person never enters but is affected by is exosystem.
- You can now place any factor in the Theory of Triadic Influence by its level of causation and its stream of influence, name the proximal predictor at the end of each stream (self-efficacy, social normative beliefs, attitudes), and use the waterfall to choose a feasible point of intervention. Keep the rule in mind when judging timing: proximal causes predict next month's behaviour, while distal and ultimate causes explain differences between groups and change over years.
- You can now describe the problem-specific models from the CDC, WHO, and UNICEF and evaluate multilevel theory through its critiques. Complexity is the critique shared by every multilevel theory, and the Theory of Triadic Influence adds a second, the unclear rationale for placing some factors at particular levels. When asked what all these models conclude, the answer is that action at one level alone is unlikely to hold, so single-level programs should be paired with broader components.
- You can now name the three things most planning frameworks combine and the three components every planning process shares, assessment, intervention, and evaluation in a feedback loop, and use importance and changeability to pick your battles. Remember that the grid can be overridden: community priorities can raise an item, partners can raise a factor's changeability, and an important factor outside the planner's authority becomes an advocacy target.
- You can now walk through the nine phases of PRECEDE-PROCEED, from the social assessment to impact evaluation, classify factors as predisposing, enabling, or reinforcing, and distinguish process, outcome, and impact evaluation. The two traps to avoid are mixing phase 3 risk factors with phase 4 factor types, and forgetting that some sources swap the labels outcome and impact; classify an evaluation by what it measures and over what time frame.
- You can now apply the risk and protective factors model to a youth health problem: sort findings into its risk and protective domains, match program components to factors, look for protective factors to build on, and design measures for each type of evaluation. You can also compare it with PRECEDE-PROCEED, which differs in scope, relation to theory, and ready-made tools but shares the same spine; start any choice between the two with the problem and the population.
Final Knowledge Assessment
This assessment covers all material from this lesson. You must score 100% to complete the lesson. Review the feedback for any incorrect answers and try again.
1. One student describes a theory in which perceived susceptibility and perceived severity add up to a likelihood of taking action. Another describes how a teenager's decision to quit vaping leads her parents to lobby the school board, which then changes its policy for every student. Which statement correctly characterizes the two descriptions?
2. A grade 4 student's teacher and parents meet to plan support for the child's reading. In Bronfenbrenner's Ecological Systems Theory, this meeting is an example of which system?
3. In the bioecological amendment, what is the difference between proximal and distal processes?
4. Which problem was the Theory of Triadic Influence designed to address?
5. A provincial ban on flavoured vaping products changes what convenience stores can sell near a school. In the Theory of Triadic Influence, at which causal level and in which stream does this factor sit?
6. In the Theory of Triadic Influence, related behaviours such as smoking and substance use have similar causes, while smoking and skiing have fewer causes in common. At which causal level are related behaviours most similar?
7. A small change in one component within one level of an ecological system is likely to cause reverberations through multiple levels, which in turn may change the environment the individual was interacting with in the first place. This observation supports which critique?
8. The CDC model of violence, later adopted by WHO, organizes contributing factors into four domains. Which list is correct?
9. Most planning frameworks combine three things. Which set is correct?
10. A planner rates icy, uncleared sidewalks as highly important to falls among older adults but, for a health unit acting alone, hard to change. The phase 5 assessment then identifies a seniors' centre willing to lobby council. According to the lesson, what does this partnership change?
11. A regional health authority planning to reduce vaping among students in grades 8 to 10 rates the price of vaping products as highly important, but not changeable by the authority itself, because the federal excise duty and provincial taxes set it. Using importance and changeability, how should the plan treat price?
12. In PRECEDE-PROCEED, which phase asks the planner to go back one more causal level, beyond the behavioural and environmental risks, to the attitudes, beliefs, social norms, and supports that contribute to them?
13. In North Hill County, planners persuade local car dealers to lower interest rates on car loans for drivers with a DUI-free record and for documented designated drivers. In PRECEDE-PROCEED terms, this arrangement is best described as which of the following?
14. In the same regional vaping plan, the assessment finds several convenience stores selling flavoured vapes within a block of the urban school. In the Theory of Triadic Influence, the local retail environment is an upstream influence in the cultural-environmental stream. Traced into PRECEDE-PROCEED, where does this finding appear and how is it classified?
15. A youth violence program reports, after six years, that police-reported violent incidents among adolescents in the target neighbourhoods have declined. In the terms used in this lesson, this is an example of which type of evaluation?
16. The regional vaping plan includes classroom lessons that correct students' belief that most of their peers vape. Which of the following is a process measure for that component?
17. Under the Hawkins and Catalano model, which of the following is listed among the protective factor domains rather than the risk factor domains?
18. Not every youth exposed to risk factors becomes involved in risk behaviour, and many have protective factors that buffer the risks. What programming trend is linked to this observation?
19. A planner must design a falls prevention program for adults over 75. A colleague suggests using the risk and protective factors model because it comes with ready-made surveys and model programs. Which response best reflects the comparison in this lesson?
✦ Pass every section knowledge check with 100% before submitting