Course: Health Behavior Change at the Individual, Household and Community Levels

  • Course Home
  • Syllabus
  • Schedule
  • Lecture Materials
  • Readings
  • Assignments
  • Want to stay in touch?

    2 + 4 =  


Topic Activities
Module 1: Culture, health and illness
Class 1: Introduction

Hour 1: Course introduction, disciplinary perspectives informing the course

I will start by explaining how the course is structured, and what is expected in the assignments.  I will then explain how I see different academic disciplines and fields of study informing the design and evaluation of behavior change interventions in public health: 1) Anthropology, 2) Sociology, 3) Psychology, 4) Health Education, and 5) Health Communication.

Hour 2: Do we understand each other? Translation problems in behavior change interventions

This hour marks the start of Module 1: Culture, health and illness.  This Module presents key concepts from medical anthropology that will be referred to throughout the remainder of the course.  In many behavior change models discussed in Module 2, constructs are included that measure people's assessment of the disease problem.  For example, the Health Belief Model includes constructs that measure perceived susceptibility to the disease, and the perceived severity of the person to the disease.  A major threat to the validity of such models is that communities may define the "disease problem", and classify it in relation to other conditions, in a very different way from medical and public health professionals.  In this hour I introduce terminology employed when examining local terminology for illnesses, and discuss the implications of different systems of terminology for behavior change interventions.

Lecture 1: PDF, MP3

Class 2: Illness etiologies, levels of causality, standards of efficacy

Hour 1: Illness etiologies, levels of causality

Different perceptions of illness etiology (causation) have long been identified as an important factor affecting health-related behaviors. For example, if a person thinks that lung cancer is caused by poor nutrition, while epidemiologists contend it is caused by cigarette smoking, we might expect that this person would be less receptive to messages exhorting him or her to quit smoking in order to reduce her/his risk of developing lung cancer.

I will start by discussing a classic article by the anthropologist George Foster. Foster draws the distinction between naturalistic and personalistic illness etiologies, and describes how some illnesses may have several levels of causality. The idea that entire medical systems are either naturalistic or personalistic, as presented by Foster in this article, is no longer accepted by most anthropologists. Rather, they would say that any medical system will display both naturalistic and personalistic aspects. This article also has some other serious problems, which we will discuss in class (See which ones you can identify). In spite of this limitation, the article is a useful introduction to the complex issue of etiology.

Unfortunately, it is difficult for us to discuss local understandings of illness etiology due to the terminology commonly employed to describe it. Local understandings of illness etiology are commonly characterized as "irrational beliefs", while biomedical understandings are characterized as "scientific knowledge". The article by Pelto and Pelto challenges this dichotomy, and suggests alternative ways to label these constructs.

Hour 2: Standards of efficacy

This article by Young is one of my favorite articles, because it addresses a crucial question which I encounter repeatedly: Why do people persist with practices that "obviously" do not work? Why do many physicians prescribe antibiotics for the common cold, when there are no signs of bacterial infection? Why do people think practices such as divination and exorcism "work", even when they fail to improve the condition of the patient? Allan Young employs the term "efficacy" to refer to people's assessment that the treatment or practice "works", or produces beneficial effects. He contends that in most cases practices persist because they appear to work. When we as public health professionals feel that a given practice does not work, but the communities we are working with feel that the practice does work, Allan Young explains that the practice is being evaluated by different standards of efficacy. He describes and provides examples for three standards of efficacy: scientific, empirical and symbolic efficacy.

This article is fairly dense. You may need to read it a couple of times to get all the points Young is making, but it is worth the effort.

Lecture 2: PDF, MP3

Class 3: Meanings of medications

Hour 1: Meanings of medications, metonymy

Sjaak van der Geest and Susan Reynolds White are anthropologists who both spent many years in the field examining what happens when modern medicines are introduced into low-income countries. Van der Geest carried out much of his field work in Ghana and Cameroon, while White conducted a series of studies in Uganda. Their explanation for why people take medications stand in contrast to the ideas of Young discussed in the previous lecture.

Hour 2: Discussion session #1 - Standards of efficacy

In this discussion group session you will apply concepts from the first three lectures, concentrating on the articles and chapter by Allan Young. Much has changed in Brazil since the article by Nations and Rebhun was written, but it provides a good example of different kinds of efficacy in practice.

Discussion questions

  1. How would you apply the concepts of scientific, empirical and symbolic efficacy to the article by Nations and Rebhun?
  2. How do the arguments presented by van der Geest and White for why people take medications fit with Young's model for why people take treatments/medications?
  3. What is symptom perceptualization (Young 1980, starting bottom of page 108), and how does it help us understand how people respond to the symptoms of chronic infectious and non-infectious diseases (schistosomiasis, syphilis, AIDS, cancer) and global threats such as climate change?
  4. How can concepts of empirical efficacy presented by Young be extended to analyzing the public and policy debate on responding to climate change?

Quiz #1

Lecture 3: PDF, MP3

Discussion Session #1


Module 2: Health behavior at the individual level

Class 4: Individual-level models

Hour 1: Health Belief Model

In this hour I introduce the concept of a behavior change model, and describe in detail what is in many ways the prototypical model: The Health Belief Model (HBM). Among psychologists, HBM is classified as an Expectancy-Value Theory. The HBM is the most commonly employed model because it incorporates constructs that are more intuitive and easy to understand for health professionals without any formal training on health behavior and behavior change interventions. The "default model" employed by such professionals specifies that transferring knowledge about the disease, its severity and its mode of transmission is critical to promoting behavior change, and should be the focus of health education and communication activities. The HBM builds on this embryonic and incomplete model (consisting of perceived susceptibility and perceived severity only) by adding the following constructs: perceived benefits, perceived barriers, cues to action and self-efficacy. Self-efficacy, which occupies a prominent place in Social Cognitive Theory, was a late addition to HBM.

Finally, I have posted examples of applications of the Health Belief Model to dental flossing (Buglar et al.), sexual behavior (Downing-Matibag and Geisinger), dengue fever in northern Thailand (Phuanukoonnon et al.) and recycling (Lindsay and Strathman). Select the topic of most interest to you, and read the article to get a sense of how the Health Belief Model is applied.

Hour 2: Theory of Reasoned Action, Theory of Blanned Behavior

I complete the introduction to behavior change models by discussing the Theory of Reasoned Action (TRA), and its variant the Theory of Planned Behavior (TPB). Many people find these models much less intuitive that the HPM, and consequently they are less commonly found in the literature. However from the perspective of social psychologists, they provide a much more powerful description of the influence of attitudes and norms on behavior. Also, these models are focused more on the behavior in question, rather than on the disease threat. We will discuss several examples of application of TRA and TPB.

Lecture 4: PDF, MP3

Class 5: Social cognitive theory, limitations of individual-level models

Hour 1: Social learning theory/social cognitive theory

Social cognitive theory built upon previous work on social learning theory, and focuses on the process of learning to perform a behavior. One of the key constructs in social cognitive theory, self-efficacy, has subsequently been incorporated into several other behavior change models including HBM.

Hour 2: Discussion session #2 - Applicability of behavior change models to other cultures

Behavior change models such as the Health Belief Model and the Theory of Reasoned Action have been criticized for being of limited utility in other cultures where people have a less individualistic, and more collective, orientation than is found in the United States. The article by Markus and Kitayama describes how concepts of the self differ between the United States and Asian cultures. This is summarized in Figure 1 on page 226 and in Table 1 on page 230.

Discussion questions

  1. Identify what other cultural contexts people in the discussion group are familiar with through their family background or work experiences. Choose a couple of these contexts, and discuss whether each one is closer to the "independent" or "interdependent" self-construals in Table 1 on page 230.
  2. Look at the constructs in the Health Belief Model in Table 2 on Page 14 of Theory At A Glance. To what degree will each of the constructs in the Health Belief Model (except self-efficacy) vary between cultures with an independent and an interdependent construal of self?
  3. Look at the constructs in TRA/TPB in Table 2 in Figure 3 on Page 18 of Theory At A Glance. To what degree will each of the constructs in TRA and TPB vary between cultures with an independent and an interdependent construal of self?
  4. Consider Voluntary Counseling and Testing " in which a counselor tells the client the result of the HIV test and discusses different options they have for prevention or management of HIV infection" and Community-Based Participatory Research "where we elicit a list of community priorities and work with the community to develop a plan to address them. How might these approaches differ between cultures where interdependent and independent self-construals are more prominent?

Quiz #2

Lecture 5: PDF, MP3

Discussion session #2

Class 6: Concepts of risk

Hour 1: Psychological perspectivces on risk

In this class we discuss people's perceptions of risk, why personal perceptions of risk frequently are inaccurate (why people are poor at predicting objective risk), and how these perceptions affect the degree to which people engage in risky behaviors. The Wikipedia entries are informative and comprehensive on this topic. There is a huge and growing literature on the psychology of risk perception, in particular for two topics:

Risk compensation is a hot topic in the HIV/AIDS prevention literature. There is concern that when people receive interventions such as male circumcision, they may engage in higher-risk behaviors, on the assumption that circumcision confers a high degree of protection from HIV transmission.
There is concern that people are completely unable to judge and respond to the global threats such as climate change and peak oil. The article by Kristof in the New York Times discusses risk perception in relation to global environmental threats.

Hour 2: Anthropological perspectives on risk

In the second hour of this lecture we discuss anthropological perspectives on risk, and the concept of risk discourse. The late anthropologist Mary Douglas is well known for her Cultural Theory of Risk, a framework that explains how different societies/culture define and perceive risks. She classified societies along two dimensions: Group and Grid. This idea is explained further in the Wikipedia entry on Cultural Theory of Risk.

Lecture 6: PDF, MP3
Module 3: Social networks, diffusion of innovations and social marketing

Class 7: Networks and diffusion of innovations

Hour 1: Social networks

This lecture is a quick overview of social network analysis, and introduces some key terms necessary for reading the literature on social networks, and also understanding diffusion of innovations and interventions such as social marketing.

Hour 2 Diffusion of innovations

Diffusion of innovations is a concept that underlies the design of many behavior change interventions in public health, especially in social marketing. The classic book on the topic is Diffusion of Innovations by Everett Rogers.

Response paper due on discussion 1 or 2 and/or lectures from first two modules

Lecture 7: PDF, MP3

Class 8: Social marketing, HIV/AIDS prevention

Hour 1: Social marketing and mass media interventions

The mass media are an integral part of many behavior change interventions. Public health aims to intervene at the population level, and the mass media are one of the most effective tools for reaching entire populations. There are a number of documented successes in large-scale behavior change in public health that would have been unimaginable without the mass media. At the same time, use of the mass media raises a number of questions, some of which we will discuss today. These include:

  • Through the mass media, can we truly inform, educate and empower people and thereby bring about lasting improvements in their lives, or are we merely manipulating people to produce short-term changes in indicators for the sake of the projects we are working on at the moment?
  • Media can be expensive (production of television and radio ads or programs, air time, production and distribution of print materials), and therefore campaigns tend to last as long as external funding is available. Are changes in behavior brought about through the mass media sustained once the campaigns are over?
  • Through what mechanism do the mass media produce changes in behavior? Is it direct (television ad says, "do this," then the person goes and does it) or indirect (third person effects)? The third-person effect hypothesis in mass media studies "states that a person exposed to a persuasive communication in the mass media sees it as having a greater effect on others than on himself or herself (Davison, 1983)", and "also argues that people are compelled themselves to take action after being exposed to a persuasive message but this action might not be due to the message itself but to the anticipation of the reaction of others." (Citations from Wikipedia).

Social marketing is one of the most common strategies for promotion of health behavior change. It draws on concepts developed in commercial marketing and applies them to the promotion of socially-desirable health-related products/commodities (condoms, oral rehydration solution, mosquito nets etc.) or behaviors (handwashing). Social marketing is not synonymous with mass media campaigns, as social marketing programs can and do incorporate many other channels of communication such as village meetings, counseling by health workers etc. A notable strength of social marketing in public health is that it introduces to the public health community ways of promoting health-related products and behaviors that do not rely on messages about disease severity, modes of transmission and risk, as is typical of many educational efforts implemented by health institutions and health workers.

Hour 2: Discussion session #3 - HIV prevention for serodiscordant couples

Discussion questions

  1. Bunnell et al., (2005) states that few clients or counselors could state accurate information about why HIV discordance among couples exists.  What are the common reasons people use to explain discordance?  What implications do these have on sexual behavior?  What implications do these have for behavioral interventions attempting to prevent transmission?
  2. The Bunnell study never fully defines what it means to be a "couple."  What do you think defines a couple?  Based on your definition, how would you define your target population for a couples-based intervention?  Could there be individual and/or group components as well?
  3. The opening paragraph of The Lancet editorial entitled "HIV treatment as prevention - it works" makes a bold statement.  Do you think we can treat our way out of the HIV epidemic? Why or why not? 
  4. HPTN 052, as discussed in The Lancet editorial, demonstrated that using treatment as prevention among sero-discordant couples was effective at reducing transmission of HIV to the uninfected partner.  Based on these results, if you were the Minister of Health in Uganda, would you try to scale-up treatment as prevention for sero-discordant couples?  Why or why not?
  5. Seale's response to The Lancet editorial says that we should "not hastily abandon non-biomedical elements of HIV prevention" even though there is less rigorous evidence available on the effectiveness of non-biomedical interventions. Do you agree with Seale's statement?  Why or why not?  If you do agree, what non-biomedical elements would you consider including in a couples-based intervention?
  6. The Karim comment discusses recent results from several pre-exposure prophylaxis (PrEP) trials.  What are some of the benefits of implementing PrEP for sero-discordant couples? What are some of the drawbacks? 
  7. The Karim comment discusses that it may be necessary to implement both treatment-as-prevention and PrEP in hyper-endemic settings, like Uganda.  Do you agree or disagree?  Are there any possible behavioral or structural interventions that could be combined with these biomedical interventions?

Quiz #3

Lecture 8: PDF, MP3

Discussion session #3

Module 4: Households and couples

Class 9: Households, public and private domains

Hour 1: Public and private domains, household structure

In this lecture I inaugurate Module 4 on households and couples by relating different ways household has been defined.  I then discuss the related concepts of Public Domain and Private Domain, and illustrate how distinction affects the implementation of public health interventions at the household level. 

The definitive article on definitions of household is by Jane Guyer, an anthropologist at Homewood Campus.  This is a long article, not for those who are operating under severe time constraints (and thus is optional).  Guyer challenges whether the concept of household is relevant to rural Africa, and proposes that lineage provides more analytical power and insight in many settings.  More is said about lineage in the second hour of today's class.

Hour 2: Kinship systems

Understanding and working within kinship systems is a crucial element of behavior change interventions in many countries.  The cornerstone of this lecture is for you to do a few topics in the University of Manitoba Kinship and Social Organization tutorial.  I would like you to be familiar with the kinds of kinship terminology that exist, but not necessarily memorize it.  If you know that this terminology and associated definitions are out there, you will be able to access them when you come to read the anthropological literature, or design an intervention for a population where the kinship system is an important consideration.

Lecture 9: PDF, MP3

Class 10: Counseling, peer education

Hour 1: Counseling

Counseling by health care workers, social workers and other professionals is a key behavior change intervention.  There has been surprisingly few well-designed studies to evaluate the impact of counseling interventions.  In this hour I will talk about two important counseling interventions: Voluntary Counseling and Testing (VCT) for HIV/AIDS and counseling of parents at part of Integrated Management of Childhood Illness (IMCI).  Difficulties in implementation of these two counseling interventions will be illustrated with examples from Mali (IMCI) and Tanzania (VCT).

Hour 2: Peer group education, peer counseling

In this hour I discuss peer education.  The key article by Turner and Shepherd describes the difficulties with establishing a theoretical basis for peer education.  Although the article was published in 1999, the problems described remain unresolved.

Lecture 10: PDF, MP3

Class 11: Intimate partner violence

Hour 1: Intimate partner violence, Transtheoretical model (Stages of Change)

One of the most critical public health problems, particularly at the household level, is gender-based violence.  In this lecture I start by reviewing the causes (Figure 3 in article by Jewkes) and intervention options (Page 1428 in article by Jewkes) for gender-based violence.  I then introduce Prochaska's Transtheoretical Model as one behavior change model that researchers have drawn from to design interventions to reduce gender-based violence.

Hour 2: Discussion session #4  - Intimate partner violence (IPV)

Discussion questions
  1. Identify one or two settings from the experience of the members of the group where IPV is common, and discuss what group members have observed.  Looking at Figure 3 in Jewkes 2002, which of the various factors in the diagram do you think made the greatest contribution to IPV?  Why?
  2. Referring to the article by Murray et al., how do you think IPV is related to relationship intimacy in this population?  Will IPV be common both in couples with high relationship intimacy and low relationship intimacy?  Why? 
  3. Again referring to the article by Murray et al., do you expect IPV to take a different form, or have different underlying causes, in couples with high and low relationship intimacy?  How would you relate Figure 3 of the article by Jewkes 2002 with the concepts of public and private domain?
  4. For the article by Lary et al., which of the intervention options listed on page 1428 of the article by Jewkes appear most appropriate?  Why?
  5. For the article by Burke et al., which of the intervention options listed on page 1428 of the article by Jewkes appear most appropriate?  Why?

Quiz #4

Lecture 11: PDF, MP3

Discussion session #4

Module 5: Community, ecological models and multi-level interventions

Class 12: Community, ecological models and multi-level interventions

Hour 1: Concepts of community

Community as a concept is a cornerstone of many behavior change interventions, but it is notoriously difficult to define.  I will start by reviewing concepts elaborated in writings by two early thinkers in the field of sociology, Ferdinand van Tonnies and Emile Durkheim, including Gemeinschaft and Gesellschaft ("Community" and "Society") and organic and mechanical solidarity.  I will then discuss the work of the anthropologist Victor Turner, and his conception of communitas as a transient phenomenon when the normal rules of hierarchy are temporarily suspended during a period of "liminality" or transition.  I will then discuss how these "classical" concepts are viewed currently, and use an article by Benjamin Paul on a participatory project in Guatemala to illustrate the consequences of employing an inaccurate or inappropriate model of community when intervening at the community level.

Hour 2: Interventions at the community level

Practitioners in a variety of disciplines have recognized for years that some changes are best promoted at the community level, rather than the individual or household levels.  This is thought to be more effective for many reasons, including the potential to harness resources only available at the community level, and have a more direct impact on social norms, a key determinant of behavior in many models.  In this lecture I review some of the terms to describe community-level interventions, and processes such as the Community Action Cycle that are integral to these approaches.

Lecture 12: PDF, MP3

Class 13: Social capital

Hour 1: Social capital

Social capital is the most current approach to analyzing the characteristics of communities and the types of intervention approaches appropriate for different communities.  It incorporates and subsumes many of the concepts and theories presented in Class 12.  Many public health interventions are being designed that incorporate elements drawn from theories of Social Capital.  In this lecture I introduce theories of Social Capital described in the article by Wakefield, and set the stage for the discussion group session that follows.

Hour 2: Discussion session #5 - Community-level interventions, social capital

Discussion questions

  1. Identify settings from the experience of the members of the group where you feel social capital is high or low.  What factors do you think contribute to the high or low social capital?  Do you think the high or low social capital has any effects on health?
  2. Summarize 1) Putnam's construction of social capital, and 2) Navarro's critique of Putnam's construction and the strengths and limitations of Putnam's and Navarro's ideas.
  3. Does San Pedro la Laguna have high or low social capital?
  4. What behavior change models, if any, do you feel describe the components of the interventions described in the articles by Jana et al. and Kerrigan et al.?
  5. The Kerrigan article suggests that "psychological and material factors were associated with participation in community-building activities." Do theories of social capital address psychological or material factors? If you were going to implement a follow-up intervention to the one implemented by Kerrigan et al., what elements would you include?
  6. What are the strengths and weaknesses of the interventions described in the articles?  Are there any potential negative consequences of interventions that aim to increase social capital?
  7. How sustainable are the interventions described in the articles?

Quiz #5

Lecture 13: PDF, MP3

Discussion session #5

Class 14: Ecological Models

Hour 1: Ecological/multi-level models

In this lecture I present the concept of an ecological or multi-level model.  Such a model incorporates levels of analysis reviewed earlier in the course (individual, household, social network, community), but includes additional levels.

Hour 2: Interventions based on multi-level models

In this lecture I discuss how interventions can be designed based on Ecological or Multi-Level model.  Work by Deanna Kerrigan and colleagues on 100% condom interventions in commercial sex work establishments in the Dominican Republic is taken as an illustration of the approach.

Question #1 of final assignment due along with outline of questions #2 and #3

Lecture 14: PDF, MP3

Class 15: Behavior of health workers

Hour 1: Behavior of health workers

To end the course I talk about a category of people who are increasingly a focus for behavior change interventions: health workers.  I discuss the different sets of issues that affect salaried and voluntary health workers, and the major categories of behaviors that are of concern.

Hour 2: Discussion session #6 - Behavior of health workers

You will analyze and discuss a current debate in the literature about community health workers, their place in the health system, and appropriate forms of motivation for them. 

Discussion questions

  1. What health care worker behaviors were intended to be addressed by the Glenton et al. study? How well do you think the study addressed those behaviors?
  2. Who was the intended audience of the original Glenton et al. article? What do you expect this audience would have gleaned from the article?
  3. After reading the original article, the critique by Maes et al., and the reply by Glenton et al., which (if any) of the critiques by Maes et al. do you consider valid? Describe the critique(s) and why the reply by Glenton et al. does or does not address the critique(s) in a manner that assuages your doubts.
  4. How would you compare the assumptions that went into each set of authors' articles? How do these assumptions manifest themselves in the conclusions each set of authors make about the study design, execution, and conclusions?
  5. After reading these articles and considering their respective arguments, what changes (if any) would you have made to the original design of this study or in its analysis? In what ways would the study and/or results have been improved by these changes?


Quiz #6

Lecture 15: PDF, MP3

Discussion session #6

Class 16: Career paths related to community and behavioral interventions

During the first 30 minutes of this lecture we will discuss points brought up in the discussion group on health workers in the previous class.  I then will discuss career paths related to community and behavioral interventions in low and middle-income countries.  I will finish by providing comments on courses related to social and behavioral interventions offered in the third and fourth terms.

Lecture 16: PDF, MP3

Final assignment due