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.

Who is a high-risk population, and what is the behaviour about?

Public health interventions, as a whole, target everyone who has or is at risk for a health problem. Sometimes, though, it is necessary to focus on particular populations, as opposed to individuals, who are said to be at high risk or who are difficult to reach. This section asks what that label means, what has to happen before a program can work with such a population, and how well the theories you have studied explain behaviour in these settings.

What "high risk" means

The ecological approach gives the first answer: people can be at high risk for many reasons. The usual list includes hazardous conditions at work or in the living environment, limited access to health care or prevention information, involvement in particularly risky behaviour (unprotected sex with multiple partners), and involvement in illegal or marginalized activities such as the commercial sex trade or gangs. Youth involved in substance use, tobacco use, and early sexual activity are said to be at high risk for HIV, sexually transmitted infections, delinquency, and violence; people who inject drugs and their partners are said to be at high risk for HIV. In each case the risk behaviour is viewed as a kind of abnormal behaviour, but only when it is viewed solely in the context of health.

That qualifier matters. For populations whose risk is particularly high, the risk is almost always related to specific circumstances that shape it. The standard examples read as a list of circumstances rather than behaviours: migrant workers who do not speak the majority language and have little access to care; youth from high-crime communities and histories of family dysfunction; rural populations with few jobs, low income, and limited care; youth who drop out of school; people returning from prison; people in high-poverty communities where street income (drug sales, sex for money, theft) is common; and historically excluded communities, including ethnic and religious minorities, sexual minorities, and in some cases Indigenous peoples in rural, reserve, or urban settings.

Seven things that almost always matter

Because of these circumstances, seven issues are almost always important in working with high-risk populations. Each is a precondition for any program, whatever the theory. Click each card.

The behaviour is usually about something else

Throughout the course, we have argued that much behaviour with implications for health is not motivated by health concerns, or at least not solely by them. This is typically true of high-risk behaviour too. Public health categorizes people by their risk behaviours, without always considering context or the other aspects of who they are, and their primary identity becomes "a high-risk population". If that is all we focus on, we limit our understanding of the risk and, as a result, the appropriate use of theory and the development of effective interventions.

Marginalized and socioeconomically segregated populations often engage in risk behaviour with goals that relate to situations of limit, threat, alienation, and mistrust, and so with a different understanding of what is and is not a risk. In one ethnographic study, many of the runaway youth interviewed believed they would not live beyond, say, age 25. Why should people who do not expect a long life worry about an infection that may take ten years to become AIDS, when there are more immediate concerns? Fieldworkers in such settings have listened to people describe extreme violence, drug use, exploitation, and transience as if they were as ordinary as eating a sandwich. To be more effective, it may be necessary to step back from the behaviour and understand the world it comes from.

This raises the central question: how well does current theory really explain high-risk behaviour? The person categorized as an IDU may also be a car mechanic, and a good one, or a cousin, an uncle, a very good cook. He may be helping to raise a child in a small apartment shared with two other families. Which is the most meaningful way to understand this person?

The story sets the task for this lesson: to expand thinking beyond health behaviour as such, toward the socioeconomic constraints, social and cultural factors, motivations, and meanings that shape it.

Why programs stay at the inner levels

The ecological model acknowledges broader influences on behaviour, yet most actual interventions have difficulty addressing factors at the outer levels (community, society) and the interaction of factors across levels. Measuring success there is hard, and the time needed exceeds the three to five years usually allotted to funded programs. The need for quick, measurable results has therefore favoured theories that meet the shorter-term need, creating a bias in the research literature. When a theory is called "well supported", the support may say as much about what is easy to fund and measure as about what explains behaviour.

Applying behavioural theory: output theories

Theoretical approaches to violence, HIV risk, substance use, and other risk behaviours typically conceptualize behaviour as the output of a set of factors (inputs) or thinking processes (stages of readiness, cost-benefit decisions). We can call these output theories and examine two families.

The first is the individual decision models. Under the Health Belief Model, people weigh perceived susceptibility, perceived severity, the costs and benefits of action, a cue to action, and self-efficacy. Under the Theory of Planned Behavior, decisions follow from attitudes toward the behaviour, subjective norms, and perceived control. In both, a relatively narrow, linear process takes place in which the contemplated behaviour and its health consequences are the primary object of attention. Here is the rub. What if a health-related behaviour is not the object of the person's attention at all, but one of many actions related to motives and needs unrelated to health? The example is gender roles. The large age difference in arranged and other traditional marriages in India, parts of Africa, and elsewhere raises HIV risk through the power difference between partners and the likelihood that the older man has, and will continue to have, multiple partners. To the people involved, the behaviour is about gender roles, not health, and a linear decision model about health has little to work with.

The second family is the risk and protective factors model of Hawkins, Catalano, and colleagues (1992), widely used in violence and substance use prevention. Risk factors are exposures and characteristics that predict high-risk behaviour, and youth face multiple, cumulative exposures over development. Exposure to risk factors raises the likelihood of problem behaviour; exposure to protective factors lowers it. Open the accordion.

The diagram draws the process implied by both families of output theory: the exposure model in the upper panel, the decision models in the lower. Both end in the same box.

Try the exposure model

The widget is a schematic version of the exposure model. Build a profile for a hypothetical fifteen-year-old, Jordan, in a mid-sized Canadian city, and read what the model can and cannot tell you.

Have output approaches worked?

One problem is built into the design. If a program targets specific risk factors, its evaluation measures change in those factors, and a substantial literature shows effects of this kind. Whether the behaviour changed is a separate question. For youth violence there is less evidence of direct links between risk factor reduction and less violence. The 2001 U.S. Surgeon General's report concluded that little was known about the effectiveness of the hundreds of violence prevention programs in use and that many were ineffective; a 2004 National Institutes of Health consensus conference agreed, while noting that some programs had reduced precursors and some had reduced violence and arrests in the short term. For HIV the record is better, and the table shows why.

Two responses: harm reduction and the generative approach

Two approaches respond to the limits of output theories. Harm reduction changes what a program asks for first; the generative approach changes what a program tries to understand.

Harm reduction: meeting people where they are

The harm reduction approach has proved effective with high-risk populations for HIV and substance use. It has stirred controversy, but experience and data show that it "meets people where they are": it focuses on the key health effects of a person's behaviour while not judging or overcategorizing the person as a whole. People at high risk usually have multiple problems they cannot change all at once, so harm reduction programs focus on the specific behaviours that pose the greatest public health threat, while not immediately addressing other behaviours that are also risky or unhealthy.

The worked example is injection drug use and HIV. Injecting drugs is extremely damaging to the person, but a serious public health consequence is HIV transmission through shared equipment, a leading cause of infection. The practices that transmit HIV become the primary focus, in recognition that some people just cannot quit in the near future. It must be stressed that this does not mean ignoring or accepting drug use, a criticism often made of the approach: the first priority is to cut off transmission, and addressing drug use often becomes much easier once a nonjudgmental relationship exists. Three components might be part of such a program:

• Needle exchange programs: used syringes and equipment exchanged for sterile ones. In the United States, a long-standing federal funding ban meant that federally funded programs could not support this.
• Risk reduction kits: prevention information, condoms, and bleach for disinfecting needles, with instructions.
• Drug substitution treatment: medically supervised methadone or buprenorphine, which reduce craving without most of the negative consequences.

Such programs typically employ people in recovery as community health outreach workers, because they can relate to people who inject drugs in a knowledgeable and nonjudgmental way. Their role draws on three theories from earlier in the course.

Harm reduction applies beyond injection: alcohol policy that accepts drinking will occur and targets the most dangerous related behaviours, such as drinking and driving; and, with people in the sex trade, skills for safer sex and personal safety before any attempt to end involvement, because stopping may be difficult and even dangerous as a short-term goal. Ending involvement remains a goal, understood as a longer-term process.

Harm reduction in Canada: a continuum

Canada has one of the longest harm reduction records anywhere. Vancouver opened a needle exchange in the late 1980s. Insite, the first legally sanctioned supervised injection site in North America, opened in the Downtown Eastside in 2003 under an exemption from the Controlled Drugs and Substances Act, and in 2011 the Supreme Court of Canada ordered it kept open (Insite). British Columbia launched take-home naloxone in 2012 and declared a public health emergency over overdose deaths in 2016. Every service in the widget answers two questions: what is the most pressing threat, and what is left for later?

The generative approach: the meaning behind behaviour

The second approach grew out of ethnographic fieldwork with high-risk populations. The focus shifts from behaviour as inputs and outputs to the organizing and motivating processes through which individuals generate behaviour over time, which anthropology, social psychology, and sociology call cultural models, narratives, scripts, discourses, frames, or schema. They remain largely untapped in public health, with exceptions in HIV and substance use research and in the work of Oyserman and Markus (1990) on possible selves. The diagram below shows the difference: a generative scheme, shaped by social and cultural influence and carried out through practical knowledge, produces a range of behaviours, some risky and some not.

Generative approaches assume that a very important human process is to make meaning out of what we do. We are geared toward behaving in ways that feel coherent and purposeful within a way of life, itself shaped by structural, social, and economic constraints. A way of life carries expectations about everything, from possible life paths to what it means to go on a date, and these take shape as models of how things are supposed to be that guide actions, reactions, and emotions. These are generative schema. Two examples: status constructs (what makes for high status and what it signifies) and gender constructs, where "being a man" or "being a woman" are bundles of meaning tied, through cultural practice, to behaviours said to represent them.

Two features follow. First, we may engage in many behaviours all related to pursuing and defending a gender role; we can call this performance of the schema. Second, a schema is generative of behaviour but not necessarily predictive in the positivistic sense, because individuals assemble their behaviour in varying ways to carry out an organizing theme. Models are internalized, interpreted, and used, and a person pulls from the life material available to perform them, so the pursuit of status takes one form today and another tomorrow. Finally, schema develop within socioeconomic constraints: a group that expects short, precarious lives develops values that support short-term gain and survival strategies, some of them risky. Pierre Bourdieu called this integration of constraints, life patterns, and values a habitus.

Three cases from the field

Each case ends with the same move: name the need, then ask what else satisfies it.

Putting a generative approach to work

The applications of a generative approach: research to identify generative schema and the socioeconomic setting shaping them; research on other identities, roles, and behaviours that could satisfy them; adding less risky behaviours and possible selves to an "inventory of behaviors" through social cognitive modelling and positive youth development; community mobilization and advocacy to generate the resources that make alternatives realistic; and social marketing and diffusion to spread them.

Why do some populations fare worse in health?

Are current theories relevant to culture, diversity, and health disparities? That question runs through the rest of this lesson. This section covers the first half of it: what disparities are, why they became a policy issue, and why they exist, with a Canadian counterpart for each idea.

How disparities became a focus

The call for "culturally competent" interventions has usually accompanied the larger issue of disparities among racial and ethnic minority populations. Eliminating disparities was the second major goal of Healthy People 2010 and remains one of four major goals of Healthy People 2020. Not all disparities involve minority populations: some relate solely to socioeconomic status or geography. But the connection between diverse minority populations and disparities is a primary aspect of the problem.

Two reasons explain the focus. One is demographic: the Asian and Hispanic populations of the United States each grew by 43 percent between 2000 and 2010. The other is that minority populations have historically fared worse in health, a truism that is not always true. Twice as many African American, American Indian, and Alaska Native babies die before age one as Caucasian American babies; the HIV/AIDS death rate among African Americans is more than seven times the Caucasian rate; Vietnamese American women have nearly five times the cervical cancer rate.

Disparities became a policy issue with the 1985 Heckler Report, which identified six causes of the gap: cancer, cardiovascular disease and stroke, chemical dependency, diabetes, homicide and accidents, and infant mortality, and found about a thirty-year lag in health improvement for African Americans. Twenty-five years later, the 2011 National Healthcare Disparities Report found quality improving for all groups but access and disparities not improving for most minority and low-income groups.

A trajectory of health

Why have minority populations fared worse? The reasons are hard to separate from each population's historical experience. Discrimination, exclusion, lack of access, environmental risk, and poverty have produced patterns of living in which health beliefs, attitudes, and behaviour evolved and cannot help but reflect that experience. One useful way of thinking about this: historical circumstances produce a trajectory of health, made up of a population's vulnerability and exposure to disease together with the systems of knowledge, attitude, and practice that developed in response. Vulnerability, circumstance, and response together produce disparities.

Migrant workers from Central America, likewise, have often arrived reluctant to trust government agencies after years of civil conflict at home. Add a history of dealing with health problems outside the mainstream system, and it is not surprising that African American and Latino men are less likely than Caucasian American men to see a doctor, even in poor health. Patterns of living matter too: African Americans and Hispanics are less likely to exercise, and the explanation is historical rather than moral, because exercise as a discrete activity apart from work is a recent, largely middle- or upper-class phenomenon.

Ethnomedical systems and the case of Lia Lee

Health knowledge is also organized by ethnomedical systems: cultural systems of knowledge and practice that define illnesses, their causes, appropriate treatments, and appropriate providers. Where these differ from biomedicine and populations adhere to them strongly, a gap in understanding and use of standard care can result.

Even with access, differential treatment inside the health system has been documented repeatedly: the Institute of Medicine's Unequal Treatment described such disparities, including a lack of culturally and linguistically competent care and provider bias.

Cultural competence and the CLAS standards

Lack of cultural competency in care has been identified as one cause of disparities, and one response was the national standards for Culturally and Linguistically Appropriate Services, the CLAS standards (U.S. Office of Minority Health), which define cultural competency in practice under three themes.

Here the critique enters. The CLAS standards may focus on organizational behaviour without enough emphasis on the role of culture in health. Kleinman and Benson (2006) argue that cultural competence should not be viewed as learning a list of traits that supposedly represent a culture. Individuals within cultures vary and cultures are not static, so competence should not be a technical skill or a checklist for clinician behaviour; what matters is how individuals subjectively experience the intersection of culture and health. Lia Lee's family were not "the Hmong"; they were particular people with a particular history.

Socioeconomic status, place, and the systems that are missing

Socioeconomic status is itself a key factor. Minority populations are overrepresented in lower socioeconomic groups, so the consequences of low status fall harder on them: poor nutrition, smoking, injuries, pollution, unemployment, psychosocial stress, community violence, and limited recreational space. Housing segregation by race and ethnicity, regardless of income, is associated with health risks, and neighbourhood characteristics affect obesity, violence, and substance use. Several organized efforts have focused on these factors.

Finally, there are the systems that do not exist. Before disparities can be addressed they must be identified, which requires data by population, and often they are not collected. Many populations are lumped under designations such as "Asian", which obscure large differences: people from Vietnam and from India both fall in that category. Once identified, the information has to reach someone designated to act, with leadership, a mission, coordination, and resources. Canada has had this problem in acute form: race-based health data were rarely collected until the COVID-19 pandemic. The next section asks which theories reach each level.

What to do: how theory connects to resolving disparities

Resolving the issues behind racial and ethnic disparities is not easy, and national strategies are often caught in politics (think of the administrative and policy assessment phase of planning), because the choice among social factors, economic factors, and individual behaviour is hard to separate from political viewpoints. The areas of action amount to far more than competent programs and translated materials.

Picking your battles

The general ecological model applies to disparities, because the causal factors are many and varied and form an interrelated web. It becomes necessary to pick your battles, and the choice depends on the population: a precise set of factors may matter more for one population than another. What follows is a tour of the theories that reach each part of the web.

Individual approaches: fill the constructs with the population

Because individuals are "behavers", there is always a role for individual-oriented approaches, even while individuals function within a larger social context. Many theories contain categories, such as social norms or perceived barriers, that are useful only when filled with specifics for a population. Which norms? What barriers? The Health Belief Model is the worked example. Click each barrier.

Attitudes and intentions work the same way. Under the Theory of Planned Behavior, a person must intend to change, believe there is a valued outcome, and believe they are capable of it. If a risk behaviour is tied to norms for "being a man", such as drinking or not talking to health professionals about personal matters, intention may be low, and that may become the focus of an effort. Notice the link to the second section of this lesson: the norm for being a man is also a generative schema, and a program that only argues against it will fare worse than one that offers another way to perform it.

Social, community, and group approaches

Contexts vary by community and health problem; three families of possibilities follow.

Multilevel programs and the pathway from determinants to outcomes

Several programs tie these strategies together. The CDC REACH program (Racial and Ethnic Approaches to Community Health) links community organizational capacity, targeted action, knowledge and behaviour change, and improved health outcomes. THRIVE (Toolkit for Health and Resilience in Vulnerable Environments), from the Prevention Institute, strengthens four community clusters: the built environment, social capital, services and institutions, and structural factors. Tobacco control is the best documented coordinated effort, combining policy, cessation services, changing norms, and mobilization. The interactive below presents the CDC sociocultural framework, the map from determinants to healthier communities. Click any box.

Organizational theory: putting the pieces together

Organizational development theory applies to problems in the health system that affect disparities: putting pieces together at a local, regional, provincial, or national level. Six characteristics are likely to make a difference.

The general barrier is coordination. So many organizations have roles that there is often overlap, or organizations working on the same problem without communication or a consensus framework, leaving gaps that nobody addresses. Canada's division of responsibility for Indigenous health among federal, provincial, territorial, and First Nations governments is a standing illustration.

Culturally appropriate, culturally adapted, culturally grounded

It helps to sort the kinds of cultural work a program can do; this distinction goes a step beyond the frameworks presented so far. A program can be culturally appropriate in its delivery: the CLAS standards, interpreters, respectful care. It can be a cultural adaptation of an existing program: Resnicow and colleagues (1999) distinguished surface structure adaptation, matching people, places, language, and materials to the audience, from deep structure adaptation, which engages the cultural, social, and historical forces behind the behaviour. Or it can be culturally grounded: designed from within a community's own knowledge, values, and governance, so that culture is the source of the program rather than a layer applied to it. Indigenous-led health promotion in Canada is increasingly of the third kind.

Two Canadian cases (simplified)

Bringing It All Together

This lesson paired two topics that ask the same question from different directions: are the theories of this course adequate when the people a program serves are not the people the theories were built around? The first half began with high-risk populations. The label is almost always a matter of circumstance (poverty, exclusion, hazardous conditions, illegal or marginalized activity), and seven working issues come before any theory: trust, access, communication, a willingness to learn, confidentiality, honesty, and shared benefits. The central claim is that risk behaviour is usually about something other than health, so the output theories (the Health Belief Model, the Theory of Planned Behavior, and the risk and protective factors model), which assume the health behaviour is the object of the person's attention, explain less than their measurability suggests. The man in the wheelchair and the runaway youth who did not expect to live past twenty-five are reminders of how much a label hides.

Two approaches respond. Harm reduction focuses on the behaviour that poses the greatest public health threat, without judging or overcategorizing the person, and the trust it builds precedes movement toward change; its Canadian record, from needle distribution and Insite to take-home naloxone and managed alcohol programs, forms a continuum from keeping people alive to changing the conditions of use. The generative approach asks what a behaviour is for, names the generative schema (mastery, independence, loyalty, being something, family obligation) and the need beneath it, and adds less risky behaviours and possible selves that satisfy the same need, with real resources where the schema formed under constraint, which is Bourdieu's habitus.

The second half widened the frame to whole populations that fare worse in health. Disparities arise from trajectories in which historical circumstance produces vulnerability, and populations develop knowledge, attitudes, and practices in response, so that mistrust rooted in Tuskegee, or in residential schools and Indian hospitals, is rational rather than a knowledge deficit. The causes span every ecological level, from ethnomedical systems and differential treatment to socioeconomic status, neighbourhood, and the data systems that are missing, and cultural competence, though necessary, cannot be a checklist of traits; cultural safety is the Canadian expression of that critique. The answer to the question is that current theories are relevant on one condition: their categories must be filled with the specifics of a population. Individual theories then reach mistrust and language; community mobilization reaches access and environmental risk; social marketing and networks reach norms; multilevel programs and the sociocultural framework trace the path from determinants to healthier communities; and organizational theory supplies the purpose, leadership, collaboration, knowledge, resources, evaluation, and coordination without which nothing scales.

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.