Content uploaded by Eric Kwame Adae
Author content
All content in this area was uploaded by Eric Kwame Adae on Nov 30, 2021
Content may be subject to copyright.
International Journal of Communication 15(2021), 5252–5271 1932–8036/20210005
Copyright © 2021 (Gregory Gondwe and Eric Kwame Adae). Licensed under the Creative Commons
Attribution Non-commercial No Derivatives (by-nc-nd). Available at http://ijoc.org.
Media Campaign Strategies in Communicating HIV/AIDS in Zambia:
Comparing Risk and Crisis Communication Strategies in Mitigating
Behavior Change Among Sex Workers
GREGORY GONDWE
1
California State University, San Bernardino, USA
ERIC KWAME ADAE
Drake University, USA
This study examined the effects of emotions on risk concerns and behavior change among
sex workers in Zambia. The aim was to investigate which health communication ad
campaigns elicit emotions that lead to behavior change. Two types of HIV/AIDS ads were
used for the analysis: Those focusing on eradicating the scourge by evoking negative
emotions versus those aimed at fighting stigma with positive messages. Findings suggest
that participants exposed to negatively framed ad campaigns were more likely to quit their
sex working profession at follow-up than those exposed to ads designed to fight social
stigma. In other words, negatively framed ad campaigns that invoked fear about HIV/AIDS
were more likely to encourage behavior change among female sex workers in Zambia.
The study speaks to the issue of risk versus crisis communication as they relate to how
the Global North and South respond to stigma.
Keywords: crisis communication, risk communication, negative emotions, HIV/AIDS,
behavior change, Zambia
The growing need among most African governments to combat HIV/AIDS is complicated by how
one should structure health promotion messages that are culturally acceptable and directed toward behavior
change (Kharsany & Karim, 2016). Since the West sponsors most public health communication campaigns,
several African governments are conflicted between pushing for an African agenda in health communication
strategies or following the dictates of Western-oriented messages that sometimes conflict with the local
cultural values (Adae, 2019; Campbell, Foulis, Maimane, & Sibiya, 2005). In other words, there is a schism
in public health communication between what the local people consider as an ideal and decent health
communication message versus what the donors prescribe and consider effective for the local populations
Gregory Gondwe: gregory.gondwe@csusb.edu
Eric Kwame Adae: eric.adae@drake.edu
Date submitted: 2021-07-05
1
This publication was made possible by the support received from the Department of Journalism at the
University of Colorado Boulder.
International Journal of Communication 15(2021) Media Campaign Strategies 5253
(Sastry & Dutta, 2017). Due to funding from Western organizations, most African governments feel
compelled to employ messages that push for foreign ideologies rather than locally relevant strategies and
in tune with local cultures (Campbell et al., 2005).
For example, in the African nation of Zambia, the government continues to grapple with how to
frame HIV/AIDS media campaign messages. As a “Christian nation” with strong affiliation to the Judeo-
Christian ethics and traditional African values, Zambia has in many instances expressed some dissatisfaction
over the sort of appeals employed in public health communication media messages (“Anger at Zambia,”
2001). For example, some clergymen have publicly weighed in to challenge the government over the health
campaign messages they deemed as attacking and weakening an increasingly threadbare of the Zambian
moral fabric. Accordingly, critics argued that the newly adopted ad strategies were promoting promiscuity.
Specifically, they censured the government ads on condom usage as explicit yet inept in effecting behavior
change among the youth (“Anger at Zambia,” 2001). Some clergymen and women argued that ads aimed
at comforting the runaway HIV/AIDS seroprevalence in Zambia encouraged sexual activities outside the
context of marriage. The social marketing of condoms also normalizes and promotes promiscuity, the spread
of HIV/AIDS, and other sexually transmitted diseases.
An outspoken clergyman, Fr. Mwebe, argued that “what would be helpful are advertisements giving
factual information about the dangers of the pandemic” (Hachonda, 2001, p. 133). In more ways than one,
this not-so-holy war of words reflects the raging debates between “risk communication” and “crisis
communication.” Central to the matter is the careful determination of the genre of health communication
strategy deemed suitable or ideal for a given sociocultural milieu (Choi, Shin, & Park, 2018; Kharsany &
Karim, 2016). In this case, we question whether messages advanced by Western donors should take
precedence over the sociocultural and local requirements of the target population.
We argue that designing an effective health promotion ad campaign requires a strong
understanding of the cultural context. Essentially, the campaign needs to reflect deep insights into the
environment and the prevailing sociocultural perceptions regarding illicit sexual activities. This argument
aptly echoes Gondwe and Muchangwe’s (2020) call for a cultural approach in setting the agenda for health
communication campaigns in the sub-Saharan African region, as it ought to be everywhere else. A message
designed with such perspectives would most likely find higher acceptability and effectiveness rates among
the target audience, even when the campaign strategy instills negative emotions such as fear, disgust, and
worry among the primary focal population.
Against this backdrop, the study examines the role of the two forms of communication—risk
communication and crisis communication—in influencing behavior change. Essentially, we explore how
negative emotions aroused by risk and/or crisis communication in HIV/AIDS ad campaigns in Zambia are
associated with key cognitive and behavioral responses. We investigate how and whether negative emotions
emerging from the two forms of communication translate into given behaviors or definable actions among
commercial sex workers in Solwezi, Kalumbila, and Lumwana communities located in the North-Western
province of Zambia. Pinned to this are the differences in culture. The three locales were selected as case
study sites because of their long-standing reputation of large-scale mining operations that have made them
social magnets for the sex trade and, consequently, noted as flashpoints of rising HIV/AIDS infections in the
5254 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
past decade (Negi, 2014). As a side note, in no way does this study attempt to glorify stigmatization nor
undermine the noble efforts put in place to fight stigma. Instead, the study compares the two strategies to
contribute to the theory, literature, and practical methods for fighting the HIV/AIDS pandemic.
Literature Review
Health Communication in Africa
In Africa, health communication strategies have often been criticized for taking a moderate and
sympathetic approach in their fight against HIV/AIDS (Sastry & Dutta, 2017). In the dawn of antiretroviral
drugs (ARVs) and the rise in Ebola cases, many governments and nongovernmental organizations have
shifted their attention by giving precedence to the fight against Ebola over the continued rising HIV/AIDS
caseloads (Hoos, El-Sadr, & Dehne, 2017; Parpia, Ndeffo-Mbah, Wenzel, & Galvani, 2016; Sastry & Dutta,
2017). For many, the argument that ARVs can extend the life span of HIV/AIDS carriers and improve their
quality of life is sufficient reason to shift attention and focus on something else (Kharsany & Karim, 2016).
Naturally, the sense of euphoria that has accompanied the advent of ARVs is proving evanescent in the fight
against HIV/AIDS.
When the HIV/AIDS pandemic exploded in most sub-Saharan African countries in the early 1990s,
efforts to curb the scourge were a priority on the government’s agenda (Kharsany & Karim, 2016). Public
service advertising and other forms of health communication strategies were designed with elements of
what Reynolds and Seeger (2005) referred to as “risk communication.” This form of communication has
been conceptualized as being associated with the identification of risks to public health and characterized
by efforts to persuade the public to adopt healthier and less risky behaviors (Freimuth, Linnan, & Potter,
2000). Spacey (2017) defines risk communication as the process of communicating potential losses and
how they might be prevented. Risk communication includes warning, disclosures, and two-way
communication aimed at managing risk (Spacey, 2017).
A particular feature of risk communication entails a focus on disasters, harms, catastrophes, or
calamities that are communicated through stringent messaging strategies. For instance, communicating the
fact that the audience must vacate a given area or risk perishing in a tsunami or some other dire
phenomenon. Other examples include health messages mandating the display of a warning label, such as
those communicating the injurious effects of tobacco use, carried on cigarette packages. Characteristically,
risk communication messages tend to be direct and often threatening to the user (Spacey, 2017). Figures
1a–c are examples of what would be considered as risk communication strategies. These messages were
circulated and are common in most sub-Saharan Africa.
International Journal of Communication 15(2021) Media Campaign Strategies 5255
Figure 1a. A famous HIV/AIDS billboard at the University of Zambia (Source: UNZA Networks,
2012).
Figure 1b. An ad campaign depicting the harsh truths of HIV/AIDS in sub-Saharan Africa
(Source: Barbier, 2016).
5256 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
Figure 1c. An HIV/AIDS ad in sub-Saharan Africa framed with messages that evoke fear
(Source: Lubanga, 2017).
At the outset of the HIV/AIDS pandemic in the early 1990s, issues of fighting stigma among victims
were less considered, as most health communication messages focused on highlighting the dangers and the
grotesque facets of the ailment. This was done with the idea of discouraging individuals from engaging in
illicit and unprotected sexual activities (Brown, Macintyre, & Trujillo, 2003). Owing to cultural and religious
factors, people who contracted HIV/AIDS were generally stigmatized and perceived as promiscuous
(Campbell et al., 2005). Even as early as the year 2001, when ARVs were being introduced, many
international organizations emerged to challenge the stigma (Campbell et al., 2005). Since then,
governments in sub-Saharan Africa embraced what Reynolds and Seeger (2005) branded as crisis
communication. This is conceptualized as a form of communication designed “to prevent or lessen the
negative outcomes of a crisis, thereby, protecting an organization, stakeholders, and/or industry from
damage” (Coombs, 1999, p. 41). Thus, crisis communication is a “dialectic between what an ostensibly
offending organization does and says, and the reaction various stakeholders have to that action (inaction)
and statement or lack of statement” (Coombs, 1999, p. 5). Heath and O’Hair (2020) assert that
crisis communication, as a niche discipline is multidisciplinary such that its approach is a
sort of mix and match rules of logic, i.e., the messages for a tsunami are not as blatant
as those in “Risk” communication . . . if a major storm is known to hit populous areas
along the Gulf Coast or savage storm are likely to hit the upper regions of the United
States, we prepare both infrastructures as well as emergence. (p. 5)
In Zambia, and many other sub-Saharan countries, crisis communication is mainly attributed to
Western designs, in the sense that most messages seem not to align with the cultural and embedded
religious values. Overtly, most messages have been criticized as encouraging sexual misconduct through
International Journal of Communication 15(2021) Media Campaign Strategies 5257
the promise of the protection of a condom and that one can still live a healthy life with AIDS. Figures 2a–c
are examples of what could be considered as crisis communication strategies.
Figure 2a. Many Zambians believe that the message above promotes promiscuity (Source:
Community Media Trust—UCT Digital Collections, 2002).
Figure 2b. Campaigns against stigma (Source: Community Media Trust—UCT Digital Collections,
2002).
5258 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
Figure 2c. Example of HIV/AIDS ads that are framed with positive messages to fight stigma in
sub-Saharan Africa (Source: Fröhlich & Gitta, 2020).
It is noteworthy that there is an idea of preparing for infrastructure and emergencies in crisis
communication in case people are affected. This is at variance with the primary focus in risk communication
where the government provides a mandatory message or prescription for people to behave in certain ways,
such as evacuating from a location or enforcing a mask mandate. Within the context of HIV/AIDS ads, risk
communication messages would be equated with those messages that focus on highlighting the ramifications
or nemesis of engaging in sexual activities per se, while crisis communication would be focused on offering
prescriptions for “safer sex,” such as giving guidance for proper condom use. Thus, the main thrust of crisis
communication is the minimization of harm, rather than a focus on eradicating the pandemic.
What is problematic in these two approaches is that cultural contexts are given less precedence in
comparison with individual theories and models that inform HIV/AIDS prevention messages (Choi et al.,
2018; Gondwe & Muchangwe, 2020). Some scholars argue that there is a need to make culture pivotal in
every aspect of HIV/AIDS ad campaigning; from thought to finish (Airhihenbuwa, Makinwa, & Obregon,
2000). According to this perspective, culture should not be perceived as a barrier or toxic component of
campaigns, but rather a tonic complement that energizes the effective contextualization of the messages in
ways that inform and ensure positive social change since “media are a part of our culture in the sense that
it indeed locates culture within its definitions” (Airhihenbuwa et al., 2000, p. 13). By doing so, a process
that Dutta (2015) referred to as communicative inversion, would be avoided. According to Dutta,
communicative inversion refers to the “deployment of communication to circulate interpretation that are
reversal of the material manifestations” (p. 13). This is not to imply that crisis communication is less
important, but rather that its presence in Zambia conflicts more with the cultural values when compared
International Journal of Communication 15(2021) Media Campaign Strategies 5259
with risk communication. In other words, it is the degree and not the important of the approach that we
emphasize in this study.
Theoretical Arguments Surrounding Risk and Crisis Communication
Reynolds and Seeger’s (2005) conceptualization of “risk and crisis” communication ideas are
influential in contemporary health communication studies. These scholars note that these two forms of
health communication strategies are at poles and are differentiated by diverse messaging goals. While crisis
communication seeks to inform, risk communication seeks to persuade. While risk communication strategies
are grounded in current scientific and technical understanding of a particular risk, they also seek to address
strategic cultural and social factors (Reynolds & Seeger, 2005). On the other hand, risk communication has
the goal of informing the target audience and is considered “more sender/event-oriented in the sense that
it seeks to respond to immediate public needs for information in a much more spontaneous, and less
controlled manner” (Reynolds & Seeger, 2005, p. 49).
Normally in ad campaigns against HIV/AIDS, an admixture of both risk and crisis communication
strategies are employed to promote positive behavioral change (Sellnow & Seeger, 2021). The power of
such an approach derives from leveraging on psychological theories that instigate negative emotions toward
a behavior believed to contribute to the rise in HIV/AIDS infection rates (Newman-Norlund et al., 2014;
Yong et al., 2014). Among others, such emotions include fear, disgust, and worry (Byrne, Byrne, Katz,
Mathios, & Niederdeppe, 2015; Newman-Norlund et al., 2014). Generally, fear is seen as an emotion that
is evoked when perceiving a serious and personally relevant threat; disgust as an emotion that functions as
a mechanism to avoid diseases and could be triggered by the sight of bodily excretions such as blood and
wounds; and worrying as a cognitively oriented emotion that can stimulate constructive problem solving
(Cho et al., 2018).
These three psychological states have been described as eliciting specific negative emotions that
Peters (2012) deems necessary for promoting desired responses to health warnings. However, the influence
of these emotions on behavior change is contingent upon the strength of competing affective responses,
implying that the effect of negative emotions on “behavior change may be inhibited were the behavior to
change also elicits competing positive affective responses” (Cho et al., 2018, p. 228). When applied to the
case of female commercial sex workers in Zambia, risk communication is measured in terms of ads that
focus on presenting targeted behavioral messages with negative emotional appeals, as exemplified by
traditional ads that were typically used before stigmatization became an issue in health communication
campaigns.
How Culture Drives the Effectiveness of Risk and Crisis Communication in Africa
The metaphor “spare the rod and spoil the child” continues to dominate in most African cultures.
The belief is that sometimes one has to be explicit in their message, especially when some consequences
usually follow one’s rational choice (Gondwe, 2018). For example, data suggest HIV/AIDS is usually spread
through promiscuity—thus, recent campaigns encourage an individual to stick to one sex partner. Therefore,
the disregard for such is perceived as a rational choice, and the consequences are treated as warranted. In
5260 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
such instances, most African cultures tend to perceive risk communication as an effective way of combating
the pandemic.
The strategy has been used for different pandemics, including Ebola and the ongoing COVID-19
pandemic. In most parts of the Global South, for example, the governments used physical force (i.e.,
whipping people) to discourage group gatherings and put individuals into quarantine during the COVID-
19 pandemic. Adebisi, Rabe, and Lucero-Prisno (2021) observe that the use of such a strategy, combined
with community engagement, yielded positive results. In studying the response behavior of 13 African
countries (Ethiopia, Ghana, Kenya, Algeria, Angola, Cote d’Ivoire, the Democratic Republic of the Congo,
Mauritius, Nigeria, South Africa, Tanzania, Uganda, and Zambia), the author reported that most countries
prioritized risk communication strategies in their quest to contain the transmission and spread of the
pandemic. The backlash reported in the findings is not linked to the strategy itself but to other variables
that presented the strategy as Western initiated, thus raising questions of mistrust among the
communities toward their governments.
Risk and Crisis Communication in Zambian Media Campaigns
In explaining the use of risk communication, Choi and colleagues (2018) describe a typical ad
campaign of the 1990s that focused on depicting shocking images where an HIV/AIDS patient was shown
as extremely skinny, frail, emaciated, and on the verge of death. These ads were geared toward
communicating the dark side of HIV/AIDS for those who contracted HIV, the virus that causes AIDS. This
type of ad presented HIV/AIDS as a “death sentence” and not to be tolerated in any form within a society.
Thus, terminally ill patients were typically shown in such ads, speaking of their promiscuous lifestyles that
had predisposed them to HIV/AIDS.
On the other hand, crisis communication entails messages that present the positive side of being
a victim of HIV/AIDS. These ads would include messages that focus on adaptations, coping mechanisms,
and pathways of healthy living for the HIV/AIDS victim, such as condom use and the taking of ARVs (Cho
et al., 2018). Crisis communication strategies may highlight the possibility of living a normal life, even when
infected with HIV. Such messaging strategies thus focus on offering hope to infected persons while ignoring
the impact that such messages might have on those that are not affected (Choi et al., 2018). Given the two
approaches, this longitudinal study aimed at examining how female sex workers in Solwezi, Kalumbila, and
Lumwana mining areas in Zambia respond to public health media strategies over time, and whether the
messages promote behavioral change. We hypothesize the following:
H1a: Female sex workers exposed to risk communication ads would be more likely to quit their jobs as
commercial sex workers during the study period.
H1b: Female sex workers exposed to crisis communication ads would be less likely to quit their jobs as
commercial sex workers during the period of study.
The rationale behind the above hypotheses is that a sex worker exposed to risk communication will
report stronger negative emotions of fear, worry, and disgust. Such negative emotions will subsequently
International Journal of Communication 15(2021) Media Campaign Strategies 5261
lead them into changing behavior. Compared with the control group, the two hypotheses were tested
separately and on two types of ads. Participants were divided into two groups and exposed to two separate
ad stimuli. We further hypothesize the following:
H2a: There will be a weaker relationship between negative emotion and quit attempts for sex workers
that have been commercial sex workers for a relatively long time.
H2b: Sex workers reporting stronger negative emotional responses will be more likely to pay attention
to the media messages and their perceived health risks.
For hypotheses H2a and H2b, we assume that commercial sex workers who have been plying their
trade for a relatively long time would be more resistant to behavioral change, compared with their less
experienced counterparts. Such relatively experienced commercial sex workers are more likely to engage in
defensive behaviors, such as avoidance, by not attending to the messages, denying their relevance, or
undermining message credibility (Witte, 1994).
Method
The experimental study was approved by the National Health Research Authority of Zambia, a
Research Ethics Committee that derives its mandate from the Health Research Act of Zambia (HRAZ) under
the “Act of Parliament” (No. 2 of 2013). Although our study was longitudinal, the data collected was treated
as cross-sectional. We consider the study longitudinal because the experiment observed our respondents’
behavior change across a one-year time period. But the data was cross-sectional because they were
collected at two different times.
Sampling
Several participants operating in areas considered as “areas to find prostitutes” in Zambia were
directly approached and asked if they wanted to participate in a study with a promise of an incentive. We
purposively selected subjects from a population of women that were perceived by the communities as sex
workers and who self-identified as commercial sex workers who ply their trade in the Solwezi, Kalumbila,
and Lumwana mining areas in Zambia’s North-Western province. Several were approached and requested
to participate in a study by the authors and the research assistants with the promise of an incentive. The
inclusion criteria included evidence that participants were at least 18 years of age who self-identified as
commercial sex workers. We required a national registration card of some sort to verify a participant’s age
eligibility. Those who agreed to participate were invited to visit an office location in Solwezi, Kalumbila, and
Lumwana townships.
All participants were scheduled to participate in the experiment, following receipt of their signed
consent forms. We also followed a snowball sampling procedure, where those who self-identified as
commercial sex workers in the community named or recommended other participants whom they believed
qualified for inclusion in the study. We did not collect any personally identifiable data on the participants.
5262 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
A total of 166 of the 371 approached showed up at the three experiment centers for the study.
This number was arrived at after aggregating the data from the three centers. The 166 who agreed to
participate were divided into three groups. The two experimental groups comprised 50 participants each,
with the control group having 66 members. The first 50 participants were exposed to a risk communication
type of message and the other 50 to a crisis communication type of message. The control group was not
exposed to any message type, but only allowed to fill out questionnaires with everyone else. The ads used
for our experiments were delivered in a single exposure for each of our experimental groups. Experimental
groups were blinded in the sense that participants were not informed about why they were assigned to a
group, and they were randomly assigned.
Procedures
Experiments for all participants were conducted over two months, between March and April 2018
and March and April 2019. While the entire process averaged 45 minutes per participant, the actual exposure
period ranged between 10 and 20 minutes for video ads. Print (hard) copy versions of the ads were gifted
to participants. Each participant that showed up for the study received an incentive, ranging from a packet
of rice to cash in Zambian currency that was equivalent to $2. Some 44 participants dropped out for various
reasons. While some had second thoughts, others could just not commit fully to the experimental process.
This left a net sample size of 122 participants, of which 48 were exposed to risk communication messages,
51 to crisis communication messages, and the control group had 23 subjects.
The participants were randomly assigned to each of the groups. Research instruments were
administered by research assistants at different centers located within the focal areas of the Solwezi,
Lumwana, and Kalumbila mining areas. The questionnaire was designed to gain a greater understanding of
the sociodemographic characteristics of the participants as well as of their sex working behaviors, behavioral
responses to HIV/AIDS media campaigns, and their perspectives on the dangers of the sex trade. The
questionnaires were designed in two languages: KiiKaonde, the prevalent language of the North-Western
province of Zambia, and in English, the official language of Zambia. The researchers also were willing to
help those who could not read.
Dependent Variable
Our dependent variable was characterized by the desire for our participants to change their
behavior. This was measured by the “quit attempts” that the participants made during the period of our
study. The “quit attempt” measure was standardized by the number of attempts that the respondents
showed in trying to quit the sex trade. Not working for a temporary period did not qualify as a quit attempt.
A Likert scale was used for the measure of the question, “On a scale of 1 to 7, where 1 is unlikely and 7
very likely, would you consider quitting being a commercial sex worker if other financial opportunities were
provided?” In the process, an adjustment was made to a Likert scale question to standardize it to how many
times participants wanted or tried to quit throughout the study process. This question was asked with the
assumption that some sex workers found pleasure in that kind of job besides only the financial or economic
incentives. Some research conducted in Zambia suggests that even after some sex workers were provided
International Journal of Communication 15(2021) Media Campaign Strategies 5263
with alternative financial opportunities, most still went back to sex working (Cheelo, 2008; Musonda, 2013;
Negi, 2014).
Independent Variables
Independent variables were determined by the levels of negative emotions measured on a scale
ranging from 1 (nothing) to 7 (extreme). Particularly, fear, disgust, and worry characterized the measure.
The responses recorded high reliability, with Cronbach’s alpha ranging from 0.84 to 0.95, and were averaged
to derive an overall negative emotions scale. We measured cognitive responses as our mediating variables.
We asked participants about how often they paid attention to the HIV/AIDS campaigns in the media. We
also inquired about the extent to which those messages made participants consider the risks of their
profession. These cognitive responses were also measured on a scale of 1 to 7, where 1 referred to no
attention and 7 referred to extreme attention. Similarly, we asked participants to indicate how often they
had avoided ads that evoked negative emotions.
Data Analysis
Our data analysis process sought to test the four hypotheses through a path model analysis
generated by Mplus while employing maximum likelihood estimations to account for missing data. We
performed inferential statistical analyses across the three groups of our experiment using χ2 and F tests,
followed by a Tukey’s post hoc test for comparisons of negative emotions scale. We evaluated model fitness
through χ2 tests and other supplemental goodness of fit indices such as the comparative fit index (CFI), the
standardized root-mean square residual (SRMR) index, and the root mean square error approximation
(RMSEA). Indirect effects were assessed using 5,000 bias-corrected bootstrapped resamples of the data as
participants kept dropping. Hypotheses H1a and H1b were tested through logistic generalized estimating
equation regression models. The two hypotheses sought to examine whether sex workers exposed to the
“risk” communication strategy were more likely to quit their jobs as sex workers than sex workers exposed
to “crisis” communication strategies. Hypotheses H2a and H2b were tested by examining the interaction
between negative emotion scales, how long one has been a sex worker, and the dependence on sex working
for survival.
Similar to Cho and colleagues’ (2018) approach, we fitted four models to regress subsequent quit
attempts on the negative emotion scale and moderating variables of dependence for survival. Further, and
particularly for hypothesis H2b, we tested the relationship between negative emotions and mediator
variables by cross-sectionally regressing attention to ad campaign messages, risk concerns of their lifestyles,
and avoiding messages that were presented with a negative touch (“risk” communication), on negative
emotions. We further measured multicollinearity effects of whether other confounding variables would skew
the independent variables from impacting the dependent variable. Our coefficient approach test yielded the
variance inflation factor values less than 2.5 for all potential mediating variables to indicate anecdotal
multicollinear effects. Finally, we separately analyzed the emotional indicators (fear, disgust, worry) to
examine whether discrete negative emotions had varied effects on “risk” communication ad campaigns. This
was regressed by re-estimating all our models using the mean for each emotion indicator. These patterns
on analysis were pretested after every six months for consistency.
5264 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
Results
Data Description
The aggregate mean for all our sample (166) was estimated at 5.31, with a standard deviation
(SD) of 1.87. The estimated model demonstrated strong data-model fit statistics, χ2 = 1.872, df = 2, p >
.768; CFI = 1.000; RMSEA = 0.000, 90% CI [0.000, 0.099]; SRMR = 0.002. The attempt by our participants
to quit their jobs as sex workers were recorded high (73%) among those exposed to “risk” communication;
25% for those exposed to “crisis” communication ad campaigns, and 2% for our treatment group that was
not exposed to any campaign strategy, but participated in the survey.
On average, 60% of the sex workers indicated that they had been operating for more than five
years. Only 2% indicated that they started less than a year ago. However, very few indicated that they have
been active workers in the commercial sex industry for at least 15 years, except for those who started at
very young ages. At most, many tended to get infected and either die or are not strong/attractive enough
to keep up with the standard appearances. The average age was 24, with the youngest being 18 and the
oldest being 38 years old. Table 1 presents the summary of the findings.
Table 1. Demographic Characteristics of Respondents.
Risk Experimental group
Crisis Experimental Group Control Group
Sol
Kal
Lum
Solw
Kal
Lum
Sol
Kal
Lum
Average age
25
21
23
27
22
24
26
22
26
Average year of operation
7
3
2
5
3
4
5
2
4
Average quit attempts
47
38
43
19
13
12
3
0
1
Note. “Sol” is for Solwezi, “Kal” for Kalumbila, and “Lum” for Lumwana towns of the North-Western
province of Zambia.
Hypotheses H1a and H1b strongly supported the argument that female sex workers exposed to
“risk” communication would report stronger negative emotions and thus attempt to change their behavior.
In particular, we observed significant relationships between participants who reported stronger negative
emotions after viewing the ad campaigns (b = 0.371, p < .001; b = 0.213) and the desire to change their
behavior (b = 0.289, p < .001; b = 0.221), relative to the experimental group exposed to a crisis
communication ad campaign. The adjusted odds ratio (AOR) = 1.03 to 1.14; p < .001.
Hypothesis H2a, which argued that there was a weaker relationship between negative emotions
and sex workers with long-term experience, was not supported (b = −0.033, p > .417; b = −0.094). This
implied that even people with long-term experience as sex workers were still affected by emotions and thus
had the proclivity to change. On the other hand, H2b reported strong statistically significant results to
indicate that sex workers who reported stronger negative emotions were more likely to pay attention to the
media messages and their perceived risks. Table 2 presents the findings.
International Journal of Communication 15(2021) Media Campaign Strategies 5265
Table 2. Associations of Negative Emotions with Subsequent Behavior Change.
Exp. Group
Attend to Warnings
Concerned About
Risks
Avoid Warning
Behavior Change
Attempts
b (95% CI)
b (95% CI)
AOR (95% CI)
AOR (95% CI)
Risk c. group
0.09 (0.11,0.14)*
0.31 (0.22,.29)*
1.44 (1.01,1.23)*
1.11 (1.02,09)*
Crisis c. group
0.17 (0.14,0.18)*
0.47 (0.19,0.33)*
1.19 (1.16,1.25)*
1.15 (1.11,1.21)*
No group
0.24 (0.16,0.21)*
0.21 (0.13,.18)*
1.39 (1.17,1.20)*
1.07 (1.03,1.11)*
*p < .05.
Direct and Indirect Effects on Negative Emotions
We further tested the cognitive confounding effects and short-term behavioral responses to ads
that advance a message of warning. We observed significant indirect effects for concerns about the dangers
of their profession among all the groups. The parameter estimates for the total indirect effects range from
0.05 in the risk communication group (b 95% CI [0.043, 0.069]) to 0.19 in the treatment group (b 95% CI
[0.053, 0.0728]). Table 3 presents the findings.
Table 3. Effect of Negative Emotions on Attempt to Change Behavior.
Risk Group
Crisis Group
No Group
Total Indirect effects
0.07**
0.09**
0.11***
Attend to warnings
0.13***
0.03*
0.04*
Risk concerns
0.07**
0.09**
0.07**
Avoiding ads
0.00
0.05*
0.01
Direct effects
0.17***
0.14***
0.04*
Total effect
0.24***
0.25***
0.15***
Note. Indirect effects refer to the effects of negative emotions on behavior change mediated by a
confounding variable, while direct effects are those effects that are unmediated. The total effect is calculated
by finding the sum of the direct effect and the indirect effects.
*p < .05. **p < .01. ***p < .001.
Discussion
The primary purpose of this study was to examine whether negative emotions aroused by media
campaigns in the fight against HIV/AIDS had the power to instigate behavior change among female
commercial sex workers who ply their trade in the North-Western province of Zambia. We tested Reynolds
and Seeger’s (2001) risk versus crisis communication frameworks and argued that risk communication had
a higher probability of arousing negative emotions. Based on this argument, we were able to demonstrate
that participants who exhibited stronger negative emotions during our experiments had a higher likelihood
of quitting their jobs as commercial sex workers and embracing a new lifestyle toward behavior change. In
other words, we provide some strong evidence that individuals exposed to risk communication strategies
exhibited higher levels of negative emotions that could easily be translated into a heightened likelihood to
embrace positive behavior change.
5266 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
This finding is consistent with Cho and colleagues (2018), Yong and colleagues (2014), and
Hammond, Fong, McDonald, Brown, and Cameron (2004), who all reported that media messages that
advance a narrative that brings out negative emotions have a higher probability of changing behavior.
Although the aforementioned scholars made their findings about the perception of smokers on grotesque
cigarette warnings, their findings are quite instructive in efforts to curb HIV/AIDS in Zambia.
A consistent feature relates to the arousal of specific stimuli of fear, disgust, and/or worry among
the participants in this study. However, the push for negative emotions in the fight against HIV/AIDS in the
media is mostly interpreted as the stigmatization of HIV/AIDS sufferers. Earlier HIV/AIDS media campaigns
in Zambia were criticized for promoting stigma. This is because such messages focused on addressing the
dangers of the virus through persuasive strategies accompanied by a push in the change of lifestyle—if you
do not change your behavior, you will perish like them.
As Witte (1994) asserts, such ads feature a well-established problem–solution messaging
structure and a basic form of persuasion, yet did not resonate with what donors perceived as ubiquitous.
Evidence suggests that negative emotions have a higher likelihood to elicit strong emotions and desires
for behavioral change.
Still, a question yet to be adequately answered relates to whether the fight against HIV/AIDS
should employ (i) risk communication that usually entails scary messages that might inflict stigmatization,
yet potent in producing behavioral change, or (ii) a focus on crisis communication that takes a sympathetic
style that focuses on fighting stigma, minimizing harm, but producing relatively limited behavioral change.
Put differently, at the heart of the matter is whether fighting stigma is relatively more important than
completely eradicating the problem.
The suggestion we make, that culture be considered, seems to be the likely alternative in that how
change occurs is also contextual. What could be considered grotesque in one culture might not be viewed
in the same terms in another. Thus, communication strategies to stem the spread of HIV/AIDS need to be
locally relevant. Local culture in this context includes the full gamut of nuances of the traditional African
culture, including the customs, traditions, mores, usages, philosophies, social controls, religious and spiritual
beliefs, as well as many other elements of the daily lives of the people.
As Reynolds and Seeger (2005) argue, crisis communication is mostly designed to reduce and
contain harm, and not necessarily eradicate a health problem. Our findings are sympathetic to this view, as
is seen in the experimental group of sex workers that was exposed to crisis communication strategies
reporting fewer negative emotions, and fewer intentions to change behavior. Because of the mildness of the
messages and the “hope” or prospect to live a longer life with ARVs, many sex workers exposed to crisis
communication strategies displayed having to deal with fewer negative emotions and a weaker desire to
change behavior. We found that many participants justified their continued involvement in the commercial
sex trade with the lack of financial support, while giving themselves hope that being an HIV/AIDS victim
was not a death sentence, but only a different style of life and living reality.
International Journal of Communication 15(2021) Media Campaign Strategies 5267
Limitations of the Study
This study features several limitations that ranged from the recruitment process of the participants
to the areas of study. These limitations were a result of the lack of funding. The researchers only managed
to provide token incentives that proved unattractive to most commercial sex workers within the catchment
areas of the study. Therefore, a well-funded project is needed to recruit a sample size that would make the
findings more generalizable. Nonetheless, our findings make important contributions to the literature on risk
and crisis communication. We did not create “before” and “after” scenarios of exposure to the ads, designed
to measure participants’ likelihood to quit/attempt to quit if other financial opportunities were provided.
Neither did we perform a follow-up survey to ascertain how exposures to ads may have changed participants’
likelihood of quitting or attempting to quit. These lines of inquiry could be necessary because financial
opportunities are likely to confound with other independent variables.
Despite this limitation, we were able to partially control for financial opportunities with the
argument provided by other scholars that some sex workers were driven by something else beyond financial
considerations. Here, studies show that some commercial sex workers were unlikely to quit, despite the
financial and other alternatives incentives (Negi, 2014). The possible argument would be that the
financial opportunities provided did not match what they were likely to make when they engaged in
the sex working business (Cheelo, 2008; Musonda, 2013).
Conclusion
The quest to design media messages that could help eradicate HIV/AIDS in Zambia, while not
stigmatizing infected individuals, remains elusive. However, this is not a uniquely Zambian problem, as the
quest for the containment of the rising HIV/AIDS prevalence rates has eluded most governments in Africa.
The schism that exists between what would be considered as a media campaign message of change and
that which only minimizes the problem still looms large. While appreciating the value of crisis communication
and its goal of protecting disease victims from stigmatization, it could be argued that risk communication
strategies are also necessary to completely deal with the HIV/AIDS scourge.
This is where a blend of the two ad appeals as suggested by Reynolds and Seeger (2005) becomes
relevant. This study holds some practical value for health communication and messaging strategies. While
there is still so much more to learn about this phenomenon in Zambia, given our findings in this study, we
recommend that risk communication is the most viable strategy for behavior change. By saying so, we are
not minimizing the importance of crisis communication, but noting that the strategies involved should be
tailored toward individuals who are already infected. Recent data suggest that 11.3% of the population has
either HIV or AIDS. This is to suggest that 8.7% are not infected. Therefore, communication strategists
must deliver a message that will minimize the infection rate through behavior change. Overtly, the crisis
and emergency risk communication model are ideal in the sense that “it offers a comprehensive approach
within which risk and warning messages and crisis communication activities can be connected into a more
encompassing communication form” (Reynolds & Seeger, 2005, p. 51). That is why Obregon’s (2000) call
for a new strategy that locates culture within a theoretical framework in communicating HIV/AIDS
prevention and care messages resounds even more strongly.
5268 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
However, when adopting this blended strategic route, there is the need to have a strong sense of
which of these approaches should play a leading role. Within the Zambian context, as suggested by our
findings, it is imperative to first understand the sociocultural underpinnings and how the social structure
interpenetrates with and drives changes in health behavior. Based on our findings, it could be strongly
justified that according to risk communication precedence when merging the two is likely to provide better
results than a focus on crisis communication strategies.
Overall, based on finding that participants exposed to a risk communication messaging strategy
were more likely to translate the negative emotions into action than those exposed to crisis communication
ads, more empirical research is needed to determine what would constitute an effective HIV/AIDS media
campaign message that can present negative emotions/warnings, without inciting stigma. Our study was
limited in the sense that the findings are based on the responses given by the participants within a controlled
experimental setting. Future research could employ qualitative methods such as focus groups, long
interviews, and ethnographic methods to investigate the lived experiences of commercial sex workers more
fully and to gain an interpretive understanding of real changes in health behavior.
References
Adae, E. K. (2019). Public relations capitalism: Promotional culture, publics and commercial democracy.
Mass Communication and Society, 22(4), 558–561. doi:10.1080/15205436.2019.1565516
Adebisi, Y. A., Rabe, A., & Lucero-Prisno III, D. E. (2021). Risk communication and community
engagement strategies for COVID-19 in 13 African countries. Health Promotion
Perspectives, 11(2), 137–147. doi:10.34172/hpp.2021.18
Airhihenbuwa, C. O., Makinwa, B., & Obregon, F.C. (2000). Toward a new communications framework for
HIV/AIDS. Journal of Health Communication, 5(sup1), 101–111. doi:10.1080/108107300406820
Anger at Zambia anti-Aids ads. (2001, January 9). BBC News. Retrieved from
http://news.bbc.co.uk/2/hi/africa/1108182.stm
Barbier, B. (2016). An ad campaign depicting the harsh truths of HIV/AIDS in sub-Saharan Africa. Alamy-
Zambia. Retrieved from https://www.alamy.com/
Brown, L., Macintyre, K., & Trujillo, L. (2003). Interventions to reduce HIV/AIDS stigma: What have we
learned? AIDS Education and Prevention, 15(1), 49–69. doi:10.1521/aeap.15.1.49.23844
Byrne, S., Katz, S. J., Mathios, A., & Niederdeppe, J. (2015). Do the ends justify the means? A test of
alternatives to the FDA proposed cigarette warning labels. Health Communication, 30(7), 680–
693. doi:10.1080/10410236.2014.895282
International Journal of Communication 15(2021) Media Campaign Strategies 5269
Campbell, C., Foulis, C. A., Maimane, S., & Sibiya, Z. (2005). “I have an evil child at my house”: Stigma
and HIV/AIDS management in a South African community. American Journal of Public Health,
95(5), 808–815. doi:10.2105/AJPH.2003.037499
Cheelo, K. H. (2008). Behind the economic figures: Large-scale mining and rural poverty reduction in
Zambia, the case of Kansanshi copper mine in Solwezi (Doctoral dissertation). Massey University,
Palmerston North, New Zealand.
Cho, Y. J., Thrasher, J. F., Yong, H.-H., Szklo, A. S., O’Connor, R. J., Bansal-Travers, M., . . . Borland. R.
(2018). Path analysis of warning label effects on negative emotions and quit attempts: A
longitudinal study of smokers in Australia, Canada, Mexico, and the US. Social Science &
Medicine, 197(3), 226–234. doi:10.1016/j.socscimed.2017.10.003
Choi, D. H., Shin, D. H., Park, K., & Yoo, W. (2018). Exploring risk perception and intention to engage in
social and economic activities during the South Korean MERS outbreak. International Journal of
Communication, 12, 3600–3620.
Community Media Trust. (2002) “HIV/AIDS protest: AIDS, political aspect, treatment action campaign,
and social movements in Africa. Cape Town, South Africa: UCT Digital Collections, University of
Cape Town. Retrieved from https://digitalcollections.lib.uct.ac.za/collection/islandora-1047
Coombs, W. T. (1999). Ongoing crisis communication: Planning, managing, and responding. Thousand
Oaks, CA: SAGE Publications.
Dutta, M. J. (2015). Decolonizing communication for social change: A culture-centered
approach. Communication Theory, 25(2), 123–143. doi:10.1111/comt.12067
Freimuth, V., Linnan, H. W., & Potter, P. (2000). Communicating the threat of emerging infections to the
public. Emerging Infectious Diseases, 6(4), 337–347. doi:10.3201/eid0604.000403
Fröhlich, S., & Gitta, A., (2020, March 3). Uganda finalizes research on injectable HIV drug. DW.Com.
Retrieved from https://p.dw.com/p/3Ynr2
Gondwe, G. (2018). News believability and trustworthiness on African online networks: An experimental
design. International Communication Research Journal, 53(2), 51–74. doi:10.35810/708g-5f82
Gondwe, G., & Muchangwe, R. (2020). Agenda-setting theory in African contexts: A Jekyll and Hyde in the
Zambian Presidential elections. International Journal of Multidisciplinary Research and
Development, 7(5), 93–100. doi:10.25810/707h-6d59
Hammond, D., Fong, G. T., McDonald, P. W., Brown, K. S., & Cameron, R. (2004). Graphic Canadian
cigarette warning labels and adverse outcomes: Evidence from Canadian smokers. American
Journal of Public Health, 9(4), 1442–1445. doi:10.2105/AJPH.94.8.1442
5270 Gregory Gondwe and Eric Kwame Adae International Journal of Communication 15(2021)
Hachonda, H. (2001). The challenges facing civil society in Zambia. Retrieved from
https://sarpn.org/documents/d0001195/6-Challenges_Civil_Society_Zambia-Hachonda.pdf
Heath, R. L., & O’Hair, H. D. (Eds.). (2020). Handbook of risk and crisis communication. New York, NY:
Routledge.
Health Research Act of Zambia. (2013). Enacted by the Parliament of Zambia. Vol. Acts of Parliament—
post 1997. Retrieved from https://www.parliament.gov.zm/node/3078
Hoos, D., El-Sadr, W. M., & Dehne, K. L. (2017). Getting the balance right: Scaling-up treatment and
prevention. Global Public Health, 12(4), 483–497. doi:10.1080/17441692.2016.1171887
Kharsany, A. B., & Karim, Q. A. (2016). HIV infection and AIDS in sub-Saharan Africa: Current status,
challenges, and opportunities. The Open AIDS Journal, 10, 34–48.
doi:10.2174/1874613601610010034
Lubanga, P. (2021, August 27). “HIV doesn’t only infect naughty people—Malama.” The MAST Newspaper
Online. Retrieved from https://www.themastonline.com/
Musonda, L. (2013). An evaluation of communicating and managing responses of HIV and AIDS among
youths at Solwezi trades training institute and Solwezi youth alive (Master’s thesis). University of
Zambia, Lusaka, Zambia.
Negi, R. (2014). ‘Solwezi Mabanga’: Ambivalent developments on Zambia’s new mining frontier. Journal of
Southern African Studies, 40(5), 999–1013. doi:10.1080/03057070.2014.946215
Newman-Norlund, R. D., Thrasher, J. F., Fridriksson, J., Brixius, W., Froeliger, B., & Hammond, D. (2014).
Neural biomarkers for assessing different types of imagery in pictorial health warning labels for
cigarette packaging: A cross-sectional study. BMJ Open, 4, e006411. doi:10.1136/bmjopen-
2014-006411
Obregon, C. O. A. R. (2000). A critical assessment of theories/models used in health communication for
HIV/AIDS. Journal of Health Communication, 5(Suppl.1), 5–15.
doi:10.1080/10810730050019528
Parpia, A. S., Ndeffo-Mbah, M. L., Wenzel, N. S., & Galvani, A. P. (2016). Effects of response to 2014–
2015 Ebola outbreak on deaths from malaria, HIV/AIDS, and tuberculosis, West Africa. Emerging
Infectious Diseases, 22(3), 433–441. doi:10.3201/eid2203.150977
Peters, E., Williams, S., Cooke, M., & Kuipers, E. (2012). It's not what you hear, it's the way you think
about it: Appraisals as determinants of affect and behaviour in voice hearers. Psychological
Medicine, 42(7), 1507–1514. doi:10.1017/S0033291711002650
International Journal of Communication 15(2021) Media Campaign Strategies 5271
Reynolds, B., & Seeger, M. W. (2005). Crisis and emergency risk communication as an integrative model.
Journal of Health Communication, 10(1), 43–55. doi:10.1080/10810730590904571
Sastry, S., & Dutta, M. J. (2017). Health communication in the time of Ebola: A culture-centered
interrogation. Journal of Health Communication, 22(Suppl.1), 10–14.
doi:10.1080/10810730.2016.1216205
Sellnow, T. L., & Seeger, M. W. (2021). Theorizing crisis communication. Hoboken, NJ: John Wiley & Sons.
Spacey, J. (2017). 13 examples of requirements risk. Retrieved from
https://simplicable.com/new/requirements-risk
UNZA Networks. (2012). “Graduate with As, not Aids” billboard. Facebook. Retrieved from
https://mobile.facebook.com/UnzaNetwork/
Witte, K. (1994). Fear control and danger control: A test of the extended parallel process model (EPPM).
Communications Monographs, 61(2), 113–134. doi:10.1080/03637759409376328
Yong, H. H., Borland, R., Thrasher, J. F., Thompson, M. E., Nagelhout, G. E., Fong, G. T., . . . Cummings,
K. M. (2014). Mediational pathways of the impact of cigarette warning labels on quit attempts.
Health Psychology, 33(11), 1410–1420. doi:10.1037/hea0000056
