Overview of Structural Interventions to Decrease Noncommercial Sex Risk

Abstract

Although the majority of case studies in this book center on programs for some of the populations most at risk for HIV, this portion focuses on structural interventions to reduce HIV in the general population by targeting the primary mode of transmission throughout the world—noncommercial heterosexual behaviors. Globally, 85% of HIV transmission is through heterosexual intercourse (Nettleman, 2011). Moreover, this category of HIV/AIDS cases continues to rise rapidly. In the United States, approximately one-third of new diagnoses appear to be related to heterosexual transmission, with male-to-male sexual contact still accounting for more than half of new diagnoses and intravenous drug use contributing to the majority of the remaining cases. However, because the HIV/AIDS diagnosis often occurs years after infection, it is likely that a higher proportion of recent infections in the USA are due to heterosexual transmission (Nettleman).

Appendices

Case Study 6: Louisiana’s Condom Social Marketing Program: Effect of Improved Availability and Access on Use

Original Program Developers and Evaluators

Deborah Cohen

Thomas Farley

Richard Scribner

William Ward

Carl Kendall

Steven Pinkerton

Ariane Lisann

Bedimo Jean

Roger Bedimo-Etame

Case Study Contents

• Abstract

• Program at a Glance

• Program Information and Implementation

• Original Program Evaluation

• Implications and Lessons Learned

• Supplementary Materials Available

Bibliography

Cohen, D. A., Farley, T. A., Bedimo-Etame, J. R., Scribner, R., Ward, W., Kendall, C., et al. (1999). Implementation of condom social marketing in Louisiana, 1993 to 1996. American Journal of Public Health, 89(2), 204–208.

Abstract

The Louisiana Department of Health and Hospitals developed a statewide social marketing campaign to increase condom use in areas with high rates of sexually transmitted infections (STIs). Condoms were provided to health clinics, small businesses, and physicians’ offices free of charge. The campaign implementers encouraged the office staff to make the condoms highly visible and freely available. In addition, they encouraged the office staff to take condoms home to give away to anyone who might need them.

The program was initiated in one area of New Orleans in 1993 (area A). The researchers also identified another similar sociodemographic area where the program was not implemented in the first year (area B). Implementers tried to limit condom distribution to area A, but the intervention was popular and spread across the state in 1994. By 1996, the number of local businesses that were participating in the condom distribution in area B nearly matched the number in area A.

The researchers assessed condom use with surveys conducted in 1994, 1995, and 1996. Women were asked by registration or administrative staff to fill out the surveys anonymously at public clinics. The clinics represented all nine geographic regions of Louisiana and included both rural and urban areas. The surveys were distributed from February 1994 through December 1996. The researchers interviewed men by intercepting them on 20 selected street corners in New Orleans, from both areas A and B.

The increased availability of condoms in Louisiana did appear to change condom-use behaviors of the targeted populations over time. A higher percentage of African American women and African American women with two or more sex partners reported using a condom during their last sexual encounter in 1996 as compared to 1994. In 1996, women were also more aware of the program and showed evidence of greater exposure to the program. The percentage of women who had condoms, had obtained free condoms, knew where to get free condoms, and reported that their friends used condoms all increased over time.

For men surveyed on the street, condom use increased from 1994 to 1996 in both areas A and B. Researchers combined data from the two areas and sought other indicators that the program was successfully influencing behavior. The percentage of participants who were using the brand of condom supplied by the program increased from 1994 to 1996, as did the percentage who reported they had received free condoms and the percentage who knew where to get free condoms. In addition, the percentage of participants who reported that they did not own any condoms decreased from 1994 to 1996.

Program at a Glance

Goal: To reduce HIV and STI transmission by increasing condom use among sexually active individuals in Louisiana

Target Populations: Sexually active individuals at high risk for STIs and HIV, both males and females, from teens to adults, of all races

Geographic Location and Region: State of Louisiana, United States

Establishment and Duration: 1993–1996

Resources Required and Goods and Services Provided: Condoms to distribute freely and without restriction to targeted populations and to health clinics, businesses and physician offices willing to distribute condoms free of charge. The intervention cost about $3 million to deliver over a 3-year period, including $972,000 in staff compensation; $213,000 in office space, travel, and supplies; and $1,815,000 in condoms.

Strategies and Components

• Utilized a statewide social marketing campaign

• Focused on increasing accessibility of condoms

• Targeted populations with high rates of STIs

Key Partners: Louisiana Department of Health and Hospitals

Key Evaluation Findings

Statistically Significant

• Increased condom use among African American women, especially among those with two or more partners during the previous year

• Increased access to condoms and knowledge of the program

• Increased the percentage of African American women who used condoms

• Increased the number of women who had condoms, obtained free condoms, knew where to get free condoms, and reported that their friends used condoms

• Increased the percentage of men who identified the brand of condom being distributed by the program as the one they last used

• Increased the percentage of men who said they had obtained free condoms and knew where to get free condoms

• Decreased the percentage of men who reported not owning any condoms

• Increased condom use among men

No Effect

• Did not change condom use among white women

• Did not change the percentage of men or women reporting two or more sex partners

Program Information and Implementation

Background, History, and Public Health Relevance

Condom use is one of the most effective ways to prevent the spread of HIV and other STIs. While at-risk individuals may be aware of the benefits conferred by consistent condom use, several barriers may prevent individuals from successfully obtaining and using condoms. The cost of condoms, low concentration of condom distributors, and lack of confidentiality or purchasing embarrassment are all valid concerns that may discourage condom use.

Increasing the accessibility to condoms is one method that may be used to increase condom use with at-risk populations. If condoms are provided free of charge and are readily available in many easily accessible locations, the number of condoms that individuals obtain could increase, along with the likelihood that the condoms will be used during sexual encounters. In addition, simply the increased presence and visibility of condoms in a community may increase the social acceptability of using condoms, which would increase their use by the population. At the time of the Louisiana study, there had been several condom distribution interventions implemented in countries suffering from HIV epidemics (e.g., Ghana, West Africa and Tanzania, East Africa). Condom distribution efforts in the United States were limited, however, because many people believed condoms to be already adequately available for sale in drugstores and supermarkets.

In response to the HIV epidemic and high rates of STIs in Louisiana (the ninth highest state AIDS case rate in 1997), the Louisiana Department of Health and Hospitals developed a statewide social marketing campaign in 1993 to increase condom accessibility to high-risk populations. More than 33 million condoms were made freely available throughout Louisiana between 1994 and 1996. The campaign was the first condom social marketing structural intervention in the United States to target and evaluate the general population in high-STI areas.

Theoretical Basis

The condom distribution campaign operated following the theory of harm reduction, which focuses on reducing the negative effects from unsafe behaviors rather than reducing the occurrence of unsafe behaviors. In addition, the diffusion of innovation theory supported the process by which knowledge of the free condom service was communicated and spread through the population.

Objectives

The campaign aimed to increase safer sex practices and reduce HIV and STI transmission by (1) providing free condoms to sexually active men and women, (2) using social marketing techniques to increasing awareness and acceptability of freely provided condoms, and (3) increasing the likelihood that condoms were available to individuals at the time of sexual encounters.

Class and Type of Outcome or Behavior Change Targeted

• ☐ Decrease IDU risk

• ☑ Decrease noncommercial sex risk

• ☐ Decrease commercial sex risk

• ☐ Increase health services utilization (exams, testing, and treatment)

Target Population and Venue for HIV Prevention

The campaign targeted sexually active members of the general population (both men and women) by providing free condoms in treatment centers, clinics, community health centers, physician offices, and private businesses. The campaign specifically targeted individuals at high risk for contracting STIs and HIV and program implementers selected neighborhoods with high rates of reported STIs for the intervention.

Pathways for Structural Change

• ☑ Changes in programs

• ☐ Changes in practices

• ☐ Changes in policies and laws

The social marketing campaign increased the availability, accessibility, and acceptability of male condoms in the target population. Providing condoms in the self-service manner employed by the campaign increased condom availability and reduced the embarrassment and need for planning associated with buying condoms. The large number of free condoms available in many different locations increased the accessibility of condoms and may have indirectly increased the acceptability of using condoms through social norm change.

Strategies and Tactics for Structural Change

The intervention provided free condoms in both public and private sector sites serving populations at high risk for HIV and STIs. In May 1993, the Louisiana Department of Health and Hospitals mandated condom accessibility to all clients of publicly funded health department clinics. Condoms were also made available at no charge in county public health clinics, community health centers, public substance abuse treatment centers, and community-based housing projects. In the private sector, a variety of businesses (including bars, dry cleaners, barbershops, liquor stores, convenience stores, and low-cost motels) were invited to distribute condoms at no charge to their customers. Community-based organizations already involved in HIV and STI prevention efforts were also supplied with large quantities of condoms for distribution. The program was piloted in one area of New Orleans in 1993 and then expanded statewide in 1994. More than 33 million condoms were made freely available throughout Louisiana between 1994 and 1996.

Core Components

The Louisiana Department of Health and Hospitals developed a statewide social marketing campaign to increase accessibility of condoms by providing them free of charge to Louisiana residents. Before the social marketing program began, health-care providers gave clients free condoms in small quantities. After the social marketing program was well underway, condoms were freely available in 93 public health clinics, 39 community mental health centers, 29 substance abuse treatment sites, and more than 1,000 businesses in neighborhoods with high rates of STIs. The staff members at businesses and clinics did not limit the number of condoms that clients could take, and they posted signs near the condoms indicating that they were free. The researchers tracked the number of condoms that were distributed and where they were distributed, and they also obtained data on the number of condoms that were sold commercially to 60 % of the Louisiana supermarkets during the time of the intervention.

Resources Required

The resources required were condoms as well as businesses and health centers willing to distribute the condoms free of charge. The intervention cost about $3 million to deliver over a 3-year period, including $972,000 in staff compensation; $213,000 in office space, travel, and supplies; and $1,815,000 in condoms. Because the condoms were purchased in bulk, they cost $0.05 each at the time the program was implemented. Approximately 33 million condoms were distributed during the 3-year study period, which was an average cost of $0.09 per condom distributed.

Management Structure

The Louisiana Department of Health and Hospitals managed the social marketing campaign. Staff from the Office of Public Health in New Orleans, the Louisiana State University Medical Center in New Orleans, and Tulane University evaluated the program. Towne and Oller, Associates, a marketing research firm, helped to obtain and analyze data on the number of condoms distributed commercially through wholesalers to Louisiana supermarkets and drugstores.

Implementation Themes

In addition to providing free condoms in many visible areas, the Louisiana Department of Health and Hospitals also provided training on condom use, condom efficacy, and increasing condom accessibility to the staff of publicly funded health departments. Staff were also encouraged to take condoms home and distribute them freely to anyone who might need them.

Main Challenges Faced

The researchers originally tried to identify a control area to compare to an area served by the condom social marketing campaign. The campaign was so popular, however, that control area organizations and businesses eventually became involved in the distribution of free condoms. That occurred about a year after the implementation of the intervention in other areas, so there was a time delay between the originally designated intervention and control areas.

Program Continuity and Present-Day Status

Condoms were provided for free until 1996, when Louisiana experienced budget issues associated with maintaining the campaign. As a solution, the state developed a brand of private-label condoms and switched from free distribution to low-cost distribution and sold condoms for $0.25 each. After a survey revealed a decrease in condom use associated with the new price, the state returned to the free distribution method. Currently, the Prevention Unit of the Office of Public Health spends a portion of a grant from the Centers for Disease Control and Prevention to purchase and distribute condoms for free.

Other Locations and Regions that Have Implemented Similar Programs

• The New York City Department of Health and Mental Hygiene (DOHMH) increased the availability and accessibility of condoms to high-risk groups under the Free Condom Initiative that began in June 2005. The department advertised a website to health and social service organizations where staff could order large batches of condoms as frequently as necessary. The organizations then made condoms freely available to customers and clients. The DOHMH also marketed and delivered large batches of condoms to small businesses such as beauty parlors, nail salons, and small hotels and motels in areas with high rates of HIV, and the businesses provided condoms freely to customers.

  http://www.nyc.gov/html/doh/html/condoms/condoms.shtml

• The HIV/AIDS, Hepatitis, STD and TB Administration (HAHSTA) in the District of Columbia Department of Health distributes free condoms to organizations such as health facilities and local business and to any DC resident. To market the free condoms and relevant health information, a social marketing campaign includes custom packaging, dispensers, information cards, T-shirts, and posters. In 2009, 3.5 million free condoms were distributed.

  www.doh.dc.gov/condoms

Original Program Evaluation

Study Design

Timeline and Duration

The program was initiated in one area of New Orleans in 1993 and expanded statewide in 1994. Questionnaires querying respondents about past behavior were distributed two to four times each year over a 1–2-week period from February 1994 through December 1996.

Cohorts

• ☑ Cross-sectional (snap shots in time)

• ☐ Longitudinal (same people followed over time)

Temporal Direction of Data Collection Relative to Intervention

• ☑ Prospective

• ☐ Retrospective

Data were collected prospectively although survey questions asked about past behavior.

Assessment Time Points (Temporal Comparison)

• ☐ Before and after intervention (baseline and follow-up measures)

• ☐ After only

• ☑ Serial (more than two measures taken over time)

Implementation Level (Geographic Comparison)

• ☐ Countries

• ☑ Regions

• ☑ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☐ Individuals

The program began in one part of the state and eventually expanded statewide.

Sampling Unit

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☑ Individuals

Recruitment Techniques

Researchers passed out surveys to women who visited clinics for family planning or prenatal visits or who brought their children for visits. They conducted interviewer-assisted street-intercept surveys of African American men age 15–45 in targeted areas of New Orleans.

Randomization

• ☑ No

• ☐ Yes

  • ☐ Random assignment

  • ☐ Random sampling

The original comparison and intervention areas were purposefully selected for their high rates of gonorrhea. Participant sampling was opportunistic in nature.

Study Type

Quasi-experimental

Methods

Data Collection

Data Sources

• ☑ Questionnaire or survey

• ☐ Chart information or surveillance

• ☐ Record of biological specimen (e.g., urine sample)

Interview

• ☑ Interviewer administered

• ☑ Self-administered

Women completed self-administered surveys; men completed interviewer-assisted surveys.

Instruments

• ☑ Paper and pencil (data entry after fieldwork)

• ☐ Computer (ACASI or direct data entry in the field)

Modality

• ☑ In-person

• ☐ Mail

• ☐ Phone

• ☐ Internet

Data Analysis

Exposure Variables Measured

Exposure variables included whether respondents had previously obtained free condoms and their knowledge of where to obtain free condoms and the number of condoms distributed through the program.

Outcome Variables Measured

• ☑ Knowledge, attitudes, and beliefs

• ☑ Behaviors and practices

• ☐ Biomarker and clinical data

For Women:

• 1. Number of sex partners in the last year (1 or 2 or >2)

• 2. Condom use at last sexual encounter

• 3. Direct and indirect measures of program exposure: percent reporting that they had received free condoms, percent reporting knowledge of where to obtain free condoms, percent reporting that they owned condoms, and percent reporting that their friends use condoms

For Men:

• 1. Number of sex partners in the last year (1 or 2 or >2)

• 2. Condom use at last sexual encounter

• 3. Direct and indirect measures of program exposure: percent who identified the brand of condom distributed through the program as the one they last used, percent reporting obtaining free condoms, percent reporting knowledge of where to obtain free condoms, and percent reporting not owning any condoms

Other Variables Measured

• ☑ Demographics

• ☐ Risk groups

• ☑ Behaviors

Researchers collected information on race, age, marital status, and education from participants. Women and men were included in the study if they reported having sex in the past 12 months.

Statistical Methods

The researchers analyzed data from each 12-month period between 1994 and 1996. When measuring changes in condom use and number of sex partners, the researchers used logistic regression and controlled for type of service received at the clinic, race, education, and marital status.

Strengths and Weaknesses of the Study Design and Methodology

• ☑ Cross contamination between intervention and comparison groups

• ☐ Concurrent interventions occurring in experimental and comparison areas

• ☐ Historical bias or trend due to historical factors

The researchers originally identified a control area to compare to an area served by the condom social marketing campaign. The intervention was so popular, however, that control area organizations and businesses soon became involved in the distribution of free condoms.

Results

Sample Size

	1994	1995	1996
Women	1,614	1,706	1,787
Men	695	584	497
Total	2,309	2,290	2,284

Retention and Loss to Follow-Up (Cohort Studies Only)

Not applicable to this study

Sample Demographics

Age

Women	1994	1995	1996
15–24	Not collected	813	877
25–34	Not collected	622	652
35 or >	Not collected	202	211

Men	1994	1995	1996
15–24	268	307	246
25–34	226	144	116
35 or >	210	159	138

Race or Ethnicity

Women	1994	1995	1996
African American	710	677	763
White	813	954	957
Other	61	51	50

All men surveyed were African American.

Gender

	1994	1995	1996
Women	1,614	1,706	1,787
Men	695	584	497

Sexual Orientation

Not reported

Outcome Measures

Measure	Finding
Condom distribution	In 1994, 8,735,000 free condoms were distributed; in 1995, 11,900,000 were distributed; and 13,360,000 were distributed in 1996 across the entire state of Louisiana. The number of condoms sold commercially in Louisiana (about two million per year) stayed stable during the peak period of the program, suggesting that the campaign benefited those who did not buy condoms before the intervention was implemented
Number of sex partners	For women, there was no significant change in the proportion of women reporting two or more sex partners during the study periods (17 % in 1994, 12 % in 1995, and 18 % in 1996). For men as well, the percentage reporting two or more sex partners did not change significantly during the survey periods
Condom use	From 1994 to 1996, self-reported condom use at the last sexual encounter did not change among white women (18 % in both 1994 and 1996) but increased among African American women (from 28 % to 36 %) and increased sharply (from 30 % to 48 %) among African American women who reported two or more partners in the previous year. When race, marital status, type of clinic visited, and education were controlled, condom use increased significantly from 1994 to 1996 among all women with two or more sex partners (p < 0.05).
	For men, condom use increased from 40 % to 56 % in the intervention area between 1994 and 1995 (p < 0.0001) and decreased slightly to 52 % the following year (p = 0.45). In the comparison area, condom use increased from 41 % to 48 % between 1994 and 1995 (p = 0.06) and increased to 55 % in 1996 (p < 0.003 for 1996 vs. 1994)
Program exposure	There were consecutive increases from 1994 to 1996 in the percentage of women who had condoms (p < .00001), obtained free condoms (p < 0.0001), knew where to get free condoms (p < 0.0001), and reported that their friends used condoms (p < 0.0001). Among men, there were increases from 1994 to 1996 in the percentage who reported using the brand of condom being distributed through the program (p < 0.0001), who reported obtaining free condoms (p < 0.001), and who knew where to get free condoms (p < 0.0001). There was also a decrease in the percentage of men who reported not owning any condoms (p < 0.0002)

Conclusions

The investigators were able to implement a successful statewide condom distribution program. The condom distribution program was associated with positive changes in the reported use of condoms by the targeted populations, without being associated with an increase in the number of sex partners. The percentage of men reporting two or more sex partners did not differ between the comparison and intervention areas and did not change significantly during the survey periods although condom use did increase.

The researchers concluded that this type of social marketing program is useful for stemming the spread of HIV because it reached large segments of the general population, was relatively inexpensive and relatively easy to implement, and did not involve any special training of facilitators or client recruiters.

Implications and Lessons Learned

Before the statewide condom social marketing program was initiated, approximately 323,000 free condoms were distributed in 1992. The social marketing program distributed more than 32 million free condoms between 1994 and 1996, most of which were distributed in health clinics, but in 1995 and 1996, more than two million were distributed by private businesses in high-risk neighborhoods. Initially, more than 50 % of businesses approached about participating in the program agreed to participate. As the program began its operations, many business owners who had heard positive things about the condom distribution program contacted program staff with requests to be included.

Through a separate analysis, the program was credited with preventing 170 HIV infections (63 primary and 107 secondary) and saving 1,909 quality-adjusted life years (QALYs). The number of QALYs saved for each averted case of HIV infection (assuming 26 years of age at infection) was estimated at 11.23. More than $323 million in direct medical costs were averted, for a total cost per discounted QALY saved of $15,809.

The results of the analysis indicated that the social marketing campaign prevented HIV infections and was a cost savings for all reasonable parameter assumptions. While the researchers noted that the Louisiana condom social marketing program was able to increase condom usage by 30 % for African American men and 29 % for African American women, the actual number of infections averted may have been higher due to the uncounted downstream infections prevented. They calculated that condom social marketing would provide cost savings even with an increase in condom use as small as 2.7 %.

Supplementary Materials Available

For more information about the Condom Media Campaign in New York:

• http://www.nyc.gov/html/doh/html/living/condoms-press.shtml

The website contains information and resources that New York City is using to market its brand of condoms, including logos, posters, TV spots, and subway ads.

Additional References

Bedimo, A. L., Pinkerton, S. D., Cohen, D. A., Gray, B., & Farley, T. A. (2002). Condom distribution: A cost-utility analysis. International Journal of STD & AIDS, 13, 384–392.

Case Study 7: Promoting Sexual Responsibility Among Youth in Zimbabwe: Effect of an Extensive Multimedia Campaign on Indicators of Safer Sex

Original Program Developers and Evaluators

Young Mi Kim

Adrienne Kols

Ronika Nyakauru

Caroline Marangwanda

Peter Chibatamoto

Case Study Contents

• Abstract

• Program at a Glance

• Program Information and Implementation

• Original Program Evaluation

• Implications and Lessons Learned

• Supplementary Materials Available

Bibliography

Kim, Y. M., Kols, A., Nyakauru, R., Marangwanda, C., & Chibatamoto, P. (2001). Promoting sexual responsibility among young people in Zimbabwe. International Family Planning Perspectives, 27(1), 11–19.

Abstract

To promote safer sex behaviors among Zimbabwean youth, the Zimbabwe National Family Planning Council (ZNFPC) developed the Promotion of Youth Responsibility Project campaign which used mass media (posters, leaflets, newsletters, and an educational and entertainment radio program), in addition to interpersonal communication strategies (launch events, dramas, peer educators, a hot line, and trained family planning providers). All of the media and communication strategies promoted themes of self-respect and self-control through messages such as “value your body” and “respect yourself.” The media campaign lasted for a total of 6 months with baseline and follow-up assessments conducted 3 months pre- and post-intervention. The researchers surveyed youth respondents in both a campaign area and a comparison area and contrasted changes in knowledge and sexual risk behaviors between the two.

The campaign reached the vast majority of the respondents (97 %) in the treatment area and effectively increased knowledge of all birth control methods except the implant. The campaign did not increase general health knowledge (such as whether people can get HIV the first time they have sex). It did, however, increase the likelihood of youth discussing health-relevant topics with others as well as the likelihood that respondents reported adopting safer sexual behaviors (such as saying “no” to sex and continuing to abstain from sex). Most notably, the campaign dramatically increased the likelihood that sexually active respondents reported remaining monogamous. Despite the short length of the campaign, youth reported evident changes as a result of this intervention.

Program at a Glance

Goal: To encourage young people in Zimbabwe to adopt behaviors that reduce the risk of pregnancy and STIs (including HIV) by encouraging abstinence for younger individuals with no sexual experience, while promoting condom use and fewer partners for those already sexually active

Target Populations: 10–24-year-olds

Geographic Location and Region: Zimbabwe, Africa

Establishment and Duration: Intervention campaign, July 1997–January 1998; baseline survey, April–May 1997; follow-up survey, April 1998

Resources Required and Goods and Services Provided: Campaign posters, leaflets, a newsletter, 26 episodes of an educational and entertainment radio program, a peer counselor or doctor on staff during the radio airings, launch events, community theater troupes, peer educators, and a hot line

Strategies and Components

• Utilized a mass media campaign and multiple interpersonal communication channels to reach different subsets of the target group

• Targeted rural and urban areas

• Received input from target population for design and implementation phases

• Trained family planning providers at designated clinics in youth counseling, followed by referrals to those clinics

Key Partners: The Zimbabwe National Family Planning Council (ZNFPC), The Johns Hopkins University Population Communication Services, and the U.S. Agency for International Development

Key Evaluation Findings

Statistically Significant

• Reached the target audience as well as members of the larger community

• Increased communication about sexual health issues

• Increased knowledge of most contraceptive methods

• Increased safer sex behaviors (abstinence, health center visits, using modern contraceptives, monogamy) among youth in a dose-response fashion

No Effect

• Did not increase general reproductive health knowledge

• Did not change young peoples’ ideas about gender roles

• Did not increase requests to partners to use condoms for youth with sexual experience

• Did not decrease sexual activity for youth with sexual experience

Program Information and Implementation

Background, History, and Public Health Relevance

With one of the highest HIV/AIDS prevalences in the world, Zimbabwe’s rates of transmission are especially high among women and people younger than 25. Although many young people in Zimbabwe know about HIV/AIDS, they may engage in unprotected sex for a variety of reasons. Gender norms (which make it difficult for women to insist on condom use) and societal norms (which discourage providing reproductive services to people under 16) both constitute significant barriers to safer sex behaviors. As a result, young people in Zimbabwe are generally under-informed about important sexual health issues, do not have the skills or self-efficacy to delay sex or insist on contraceptive use, and have limited access to health services.

Theoretical Basis

This intervention was based on the Steps to Behavior Change framework which posits that individuals pass through five stages as they change their behavior. The stages include knowledge, approval, intention, practice, and advocacy. A successful communication intervention should therefore determine the audience’s stage with respect to a given behavior and then focus its campaign accordingly.

Objectives

The campaign aimed to increase (1) knowledge of HIV/AIDS and other reproductive health issues, (2) acceptance of safer sexual behaviors, (3) use of reproductive health services, and (4) safer sexual behaviors.

Class and Type of Outcome or Behavior Change Targeted

• ☐ Decrease IDU risk

• ☑ Decrease noncommercial sex risk

• ☐ Decrease commercial sex risk

• ☑ Increase health services utilization (exams, testing, and treatment)

In addition to the mass media campaign to change the behavior of young people, the Zimbabwe National Family Planning Council (ZNFPC) designated 26 clinics in the campaign area as youth friendly. During a 1-week course, the ZNFPC trained one family planning provider from each of these clinics in interpersonal communication and youth counseling skills.

Target Population and Venue for HIV Prevention

The campaign aimed to influence youth, so the developers circulated mass media campaign materials to the general public and the schools. Radio programs targeted young people nationwide, while launch events targeted all community members. Two community theater groups performed educational and entertainment dramas in schools, churches, and town centers, while peer educators spoke with young people in these locations and in their homes. The developers established a hot line at a youth center to receive calls from youth all over the country. The campaign also aimed to influence parents, teachers, and health-care providers.

Pathways for Structural Change

• ☑ Changes in programs

• ☑ Changes in practices

• ☐ Changes in policies and laws

The intervention operated through changes to (1) practices, by promoting the acceptability of safer behaviors for youth on a societal level, and (2) programs, by increasing the availability and accessibility of appropriate services at clinics.

Strategies and Tactics for Structural Change

The intervention focused on the social and cultural environments of the target communities through changes to the mass media environment and the use of interpersonal communication strategies as follows.

Media	Descriptions
Posters	A series of eight posters carried messages such as “Value your body and a happy future lies ahead” and “You may think you are ready for sex, but are you ready for the consequences?” In campaign sites, 10,000 copies of each poster were distributed
Leaflets	Five leaflets—on abstinence, how to say “no” to sex, postponing sex, delaying parenthood, and STIs—were produced, and 19,000 copies of each were distributed
Newsletters	Peer educators and schools distributed 100,000 copies of Straight Talk, a four-page newsletter on reproductive health issues of importance to young people
Radio program	During the campaign, 26 episodes of Youth for Real, a 1-h radio variety show, were broadcast nationwide. This weekly program combined information and advice with entertainment such as music and mini-dramas and hosted a peer counselor and doctor to answer listener questions
Launch events	To mobilize community support for the campaign, local committees planned elaborate launch activities including speeches, dramatic performances, soccer games, and performances by popular musicians. Adults who influence youth attended the launches, and novelty items bearing campaign messages were distributed
Hot line	A hot line established at a youth center and staffed by peer educators operated 8 h a day and was used to disseminate campaign messages as well as to answer individual inquiries

The Messengers

In addition to the family planning providers trained in youth-friendly clinics, peer educators ages 18–24 years old were recruited from the community and trained to speak with young people at schools, churches, town centers, homes, and over the hot line.

The Messages

The developers promoted two main themes throughout the intervention: (1) respect yourself and value your body and (2) have self-control. All of the campaign materials reinforced these two messages and emphasized the consequences of unprotected sex, how to negotiate safer sexual behaviors, and discussion with friends, family, and providers.

To reach young people outside of schools, all media used an entertainment and education strategy. Designed to hold the attention of large audiences, the entertainment and education programs provided role models for social learning and aimed to generate emotional responses from listeners and viewers to heighten the impact of the messages. Additionally throughout the campaign, peer educators, drama groups, and print materials referred young people who needed reproductive health services to youth-friendly clinics.

Incentives

Novelty items bearing campaign messages were distributed at launch events.

Core Components

The youth campaign used two complementary communication strategies to convey information to its intended audience: (1) mass media, which can reach large audiences at a relatively low cost, and (2) interpersonal communication tactics, which can address individual concerns. In addition, the campaign used a variety of media to reach different subsets of the target population (e.g., radio, posters, leaflets).

Resources Required

The campaign required funds and materials for posters (10,000 copies), informational leaflets (19,000 copies), a four-page newsletter (100,000 copies), producing and airing a 26-episode educational and entertainment radio program, a peer counselor or doctor on staff whenever a radio program aired, hosting launch events (including activities such as musical and dramatic performances, speeches, and games), community theater troupes to perform daily, peer educators (four for each rural area and six for each city) to speak with groups of young people, and hotline staff.

Management Structure

Zimbabwe National Family Planning Council (ZNFPC) employed researchers and managers in their Evaluation and Research Unit and received input from researchers and evaluation advisors from The Johns Hopkins University Center for Communication Programs. Each of the provinces was managed locally by province managers and information, education, and communication officers. Management was decentralized to local committees that included representatives from local government, religious organizations, and educational, health, and business groups.

Implementation Themes

Building community support for behavior change helped to ensure that young people found approval for their actions and had access to services.

Main Challenges Faced

Because of budget limitations, the program developers created print materials and the radio program in English only (generally spoken throughout the country), whereas rural youth preferred the two main native languages (Shona and Ndebele). The campaign may not have reached some critical populations because of language issues. The primary social barriers were gender roles that encourage women to be submissive and societal norms discouraging sex education for younger teens.

Program Continuity and Present-Day Status

Funding for the campaign provided by the US Agency for International Development lasted 6 months. Community support for the project enabled the continuation of peer educator training, youth-friendly clinics, and the hot line after the program ended. Other programs initiated after the project ended have also targeted the sexual health of youth in Zimbabwe. The ZNFCP has continued to promote sexual health responsibility by partnering with the United Nations Population Fund. These organizations and others have initiated efforts to increase the availability, accessibility, and acceptability of reproductive health services for youth; promoted gender-sensitive reproductive health skills and knowledge; and encouraged the participation of youth in leadership positions as part of the Young People’s Network for HIV and AIDS.

Other Locations and Regions that Have Implemented Similar Programs

• The Jerusalem AIDS Project in Israel aimed to increase the discussion of HIV between parents and children, knowledge of HIV, and safer sexual behaviors among sexually active students. It provided AIDS education in and out of schools, a telephone hot line, outreach to the community, and training by teachers.

  http://israaid.org.il/member_page.asp?id=11

• The Delivery of Improved Services for Health Project in Uganda aimed to increase health service use and safer sexual behaviors through a mix of radio, television, and print messages as well as community education activities.

  http://www.ugandadish.org/

• The Sankha Wekha (It’s Your Choice) multimedia HIV youth campaign in Malawi targeted young people and attempted to change attitudes surrounding male promiscuity and female submission which contribute to the spread of HIV.

  http://www.storyworkshop.org/our_approach/radio/radio_magazines/index.html

Original Program Evaluation

Study Design

Timeline and Duration

Baseline assessments were conducted in April and May 1997, before the campaign began in July 1997. Lasting 6 months, the campaign activities ended in January 1998, and the follow-up survey was conducted in April 1998. That survey asked the same questions as the baseline assessments as well as additional questions about campaign exposure and the resulting actions.

Cohorts

• ☑ Cross-sectional (snap shots in time)

• ☐ Longitudinal (same people followed over time)

Temporal Direction of Data Collection Relative to Intervention

• ☑ Prospective

• ☐ Retrospective

Researchers collected data prospectively; however, the nature of the data was based on subject recall and may be considered retrospective.

Assessment Time Points (Temporal Comparison)

• ☑ Before and after intervention (baseline and follow-up measures)

• ☐ After only

• ☐ Serial (more than two measures taken over time)

Implementation Level (Geographic Comparison)

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☑ Cities

• ☑ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☐ Individuals

The campaign operated in five experimental cities and towns: Mutare (an urban area) and Maphisa, Nemanwa, Nzvimbo, and Tongogara (all small towns at the center of rural districts). It did not target Kwekwe (a city) or Mubaira (a small town at the center of a rural district) which served as the comparison areas, although residents may have received some campaign messages.

Sampling Unit

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☑ Households

• ☐ Couples, pairs, and dyads

• ☑ Individuals

Recruitment Techniques

Interviewers went door to door to selected homes.

Randomization

• ☐ No

• ☑ Yes

  • ☐ Random assignment

  • ☑ Random sampling

Although the researchers assigned treatment areas for practical reasons (matching on key characteristics where possible), they did randomly choose houses within a 30-km radius of each town or city center; within each household, fieldworkers randomly selected one youth between the ages of 10 and 24 (of the same gender as the fieldworker).

Study Type

Quasi-experimental

Methods

Data Collection

Data Sources

• ☑ Questionnaire or survey

• ☐ Chart information or surveillance

• ☐ Record of biological specimen (e.g., urine sample)

Interview

• ☑ Interviewer administered

• ☐ Self-administered

Instruments

• ☑ Paper and pencil (data entry after fieldwork)

• ☐ Computer (ACASI or direct data entry in the field)

Modality

• ☑ In-person

• ☐ Mail

• ☐ Phone

• ☐ Internet

Four teams each consisting of a supervisor and four interviewers conducted the fieldwork. In each home, an interviewer explained the reason for the research, described the survey, and then asked for the consent of the potential participant or the parent if the respondent was younger than 15.

Data Analysis

Exposure Variables Measured

Survey respondents self-reported whether or not they had seen or heard the campaign in each of the media employed.

Outcome Variables Measured

• ☑ Knowledge, attitudes, and beliefs

• ☑ Behaviors and practices

• ☐ Biomarker and clinical data

• 1. Coverage of target population—percent stating that they recognized the campaign slogans

• 2. Knowledge of HIV/AIDS and other reproductive health issues—percent giving spontaneous answers and prompted knowledge about contraceptive methods and percent with correct answers to a series of true-false questions

• 3. Acceptance of safer sexual behaviors—percent reporting conversations with others about sexual issues, HIV and AIDS, and physical growth and maturity

• 4. Use of reproductive health services—percent reporting seeking services at a health or youth center

• 5. Practice of safer sexual behaviors—percent saying “no” to sex, continuing abstinence, avoiding a “sugar daddy,” and, among respondents with sexual experience, the percent that stopped having sex, stayed with one partner, started to use condoms, or asked a partner to use condoms

Variable 1, above, was measured at follow-up only. Variables 2–5 were measured identically in the baseline and follow-up surveys.

Other Variables Measured

• ☑ Demographics

• ☐ Risk groups

• ☑ Behaviors

Age, sex, education, sexual experience, marital status, and urban-rural residence were all controlled for in the analyses.

Statistical Methods

The researchers analyzed the percentages of respondents in the campaign and comparison areas indicating exposure to each element of the campaign, followed by logistic regression analyses to determine the odds of knowledge or behavior change in the campaign versus comparison areas while controlling for background variables.

Strengths and Weaknesses of the Study Design and Methodology

• ☑ Cross contamination between intervention and comparison groups

• ☑ Concurrent interventions occurring in experimental and comparison areas

• ☐ Historical bias or trend due to historical factors

Several methodological complications were inherent in the study design. The comparison area received some elements of the media campaign, including the radio program, hot line advertisements, and posters. Several other interventions also occurred across Zimbabwe concurrent with this campaign, including a condom marketing program, peer educators and youth dramas sponsored by an HIV prevention program, and family life education in the schools.

Results

Sample Size

	Baseline	Follow-up	Total
Intervention area	973	1,000	1,973
Comparison area	453	400	853
Total	1,426	1,400	2,826

At baseline, the study had 973 individuals (ages 10–24) from the campaign area and 453 individuals from the comparison area. At follow-up, it had 1,000 from the campaign area and 400 from the comparison area.

Retention and Loss to Follow-Up (Cohort Studies Only)

Not applicable to this study

Sample Demographics

Age

	Baseline		Follow-up
Years	Campaign (%)	Comparison (%)	Campaign (%)	Comparison (%)
10–14	33	19.7	21.9	23.8
15–19	45.3	49.8	54.3	54.0
20–24	21.7	30.5	23.8	22.3

Race or Ethnicity

African, from Zimbabwe

Gender

	Baseline		Follow-up
Gender	Campaign (%)	Comparison (%)	Campaign (%)	Comparison (%)
Male	49.9	50.0	50.2	50.5
Female	50.1	50.0	49.8	49.5

Sexual Orientation

Not provided

Outcome Measures

Measure	Finding
Coverage of target population	Significantly higher proportions of youth in campaign than in comparison areas reported exposure to the campaign as a whole and to the individual media components separately (p < 0.001).
Knowledge of HIV/AIDS and other reproductive health issues	The percent increase in spontaneous answers and prompted knowledge about contraceptive methods in the campaign versus the comparison areas was statistically significant for all methods except the implant (p < 0.001). However, the low percentage of respondents with general reproductive knowledge did not increase significantly for five out of six questions in the campaign versus comparison areas.
Acceptance of safer sexual behaviors	The percent of respondents reporting conversations with others about sexual issues, HIV and AIDS, and physical growth and maturity in the campaign versus comparison areas was statistically significantly higher (p < 0.001) for any discussions (79.8 % vs. 20.2 %, respectively) and separately for discussions with friends, siblings, parents, teachers, and partners (all p < 0.001).
Use of reproductive health services	The percent of respondents reporting seeking services at a health or youth center was dramatically higher in the campaign versus comparison areas (33.5 % vs. 9.5 %, respectively, p < 0.001).
Practice of safer sexual behaviors	The percent of respondents in the campaign versus comparison area was statistically significantly higher for those saying “no” to sex, continuing abstinence, and avoiding a “sugar daddy” (all p < 0.001). The percentage of respondents with sexual experience reporting staying with one partner and starting to use condoms was significantly higher in the campaign versus comparison area (p < 0.001 and p < 0.05, respectively). In contrast, the percentage of respondents with sexual experience in the campaign versus comparison area that stopped having sex or asked their partners to use condoms did not appear to increase.

The researchers found evidence of a dose-response relationship between exposure and impact such that the more intervention materials and activities that young people were exposed to, the more likely they were to have discussions about sexual issues, adopt safer sexual behaviors, and seek health services. In addition, by controlling for respondents’ exposure to all other campaign components, the researchers determined the impact of each campaign component on self-reported knowledge, beliefs, attitudes, and actions. They found that launch events had the strongest impact, followed by leaflets and dramas.

Conclusions

The intervention campaign successfully reached its audience of 10–24-year-olds in the campaign areas and beyond, as well as members of the larger community, through the use of multiple media and communications strategies. The campaign increased specific knowledge about all but one contraceptive method, although it did not have much impact on general reproductive health knowledge. It also successfully spurred dialogue about important reproductive and health-related issues among youth and others, which in turn appears to have helped build support among community members and within the health-care system for reproductive health interventions for young people.

Implications and Lessons Learned

The youth campaign sought to change knowledge, attitudes, and behavior by altering the social environment (availability of, access to, and acceptance of family planning goods and services) through a multifaceted approach, emphasizing empowering and informative messages for young people, caretakers, and health-care providers.

The youth campaign reached a significant portion of its target audience and had an impact on their relevant knowledge, attitudes, and behaviors. By measuring exposure to the different intervention components, the researchers established a dose-response relationship between exposure and outcomes—a powerful indicator of impact. Clearly, the more exposure to intervention materials, the more likely youth were to change their beliefs and behaviors. The campaign also appears to have reached a secondary audience among adults including parents and providers, which the developers believe contributed to a shift in social norms making sex education for teens more acceptable.

The success of the campaign most likely resulted from two approaches taken. First, the developers solicited input from youth and adults in local committees as well as health-care providers during the design and implementation of the campaign, which likely increased its educational value and relevance within the target groups. Second, the developers employed multiple media and interpersonal communication strategies to convey messages, thus optimizing the likelihood of providing health knowledge to individuals with access to only some types of media or a preference for one communication style.

Supplementary Materials Available

Additional References

For more information about the current state of Zimbabwe youth and HIV prevention:

• http://countryoffice.unfpa.org/zimbabwe/2010/11/20/2891/adolescents_sexual_and_reproductive_health/

• http://www.prb.org/Articles/2003/STIRisksHighAmongZimbabwesYouth.aspx

Website for Zimbabwe Youth Council:

• http://zimbabweyouthcouncil.org/index.php?option=com_content&view=article&id=106&catid=35

Case Study 8: The 100% Jeune Social Marketing Campaign: Effect on Condom Use Among Youth in Cameroon

Original Program Developers and Evaluators

Dominique Meekers

Sohail Agha

Megan Klein

Andrea Plautz

Case Study Contents

• Abstract

• Program at a Glance

• Program Information and Implementation

• Original Program Evaluation

• Implications and Lessons Learned

• Supplementary Materials Available

Bibliography

Meekers, D., Agha, S., & Klein, M. (2005). The impact on condom use of the 100 % Jeune social marketing program in Cameroon. Journal of Adolescent Health, 36, 530–541.

Abstract

Since the mid-1990s, HIV rates in Cameroon have been steadily increasing, with HIV prevalence rates reaching about 12 % nationwide. Young people have the highest risk for acquiring HIV, due to early sexual debuts, low condom use rates, and the prevalence of other STIs. In response to a rising concern over increased HIV infection rates, the 100 % Jeune program was developed in 2000 to motivate urban at-risk youth in Cameroon to engage in healthier sexual behavior. The 100 % Jeune program, an integrated social marketing campaign, included peer education; a monthly magazine, a radio drama, and call-in show; and television, radio, and billboard campaigns, in addition to condom promotion. A survey of youth conducted 2 years after 100 % Jeune began indicated that exposure to the campaign was high. Furthermore, significant positive changes in condom use and also in perceived condom use self-efficacy, perceived condom attributes and access, and perceived social support correlated with exposure to the program.

Program at a Glance

Goal: To increase condom use and condom use predictors among Cameroonian youth

Target Populations: Cameroonian youth (aged 15–24 years) in the cities of Yaoundé and Douala

Geographic Location and Region: Cameroon, Africa

Establishment and Duration: The 100 % Jeune program was launched in 2000 and is still operating. An evaluation study was implemented in 18-month intervals between 2000 and 2003. The first wave was implemented from July 26 to August 10, 2000; the second wave from January 4 to January 19, 2002; and the third wave from June 19 to July 16, 2003. The main study compared the first two cross-sectional waves.

Resources Required and Goods and Services Provided: Staff and funding for peer educators; funding to produce a monthly magazine and 18-episode radio drama; funding to support a call-in radio show; funding for television, radio and billboard campaigns

Strategies and Components

• Utilized social marketing and mass media strategies to promote condom use, including a monthly magazine, radio dramas, and television, radio, and billboard campaigns

• Targeted at-risk youth through a variety of communication strategies

• Incorporated interactive components into the campaign, such as peer education and radio call-in shows

• Increased condom accessibility by creating youth-friendly condom outlets

Key Partners: Programme de Marketing Social au Cameroun (PMSC), an affiliate of Population Services International (PSI), and the Bill & Melinda Gates Foundation

Key Evaluation Findings

Statistically Significant

• Reached the target audience successfully

• Increased regular condom use with casual and regular partners, discussion of STIs and AIDS with friends and others, perceived social support from parents, confidence in how to use a condom correctly, and knowledge of nearby condom sources for both men and women

• Decreased shyness of obtaining condoms in both men and women

No Effect

For Women

• Did not increase beliefs that condoms were effective for family planning

• Did not increase self-efficacy to convince regular or casual partners to use condoms

• Did not increase discussion of family planning with others

For Men

• Did not increase self-efficacy to convince casual partners to use condoms

• Did not increase perceived support from friends about condom use or discussion of family planning with friends in the past year

• Did not increase perceived risk of HIV/AIDS or decrease the false belief that AIDS can be cured

• Did not increase condom use with regular partners

For Men and Women

• Did not increase knowledge that condoms are effective for HIV/AIDS prevention

Program Information and Implementation

Background, History, and Public Health Relevance

Since the mid-1990s, HIV rates in Cameroon have been steadily increasing, with HIV prevalence rates close to 12 % nationwide. Young people have the highest risk for acquiring HIV, due to early sexual debuts, low condom-use rates, and the prevalence of other STIs. Young Cameroonian women are particularly at risk of acquiring HIV and that population also experiences high rates of unwanted pregnancies, abortion, and pregnancy-related school dropouts.

Theoretical Basis

The 100 % Jeune program used elements of the Social Learning Theory, Theory of Reasoned Action, and the Health Belief Model to support behavior change program objectives and activities. The theoretical framework assumed that behavior change is a product of individual, environmental, and social factors, including perceived severity of sexual risks, perceived personal risks, perceived condom attributes and access, perceived social support, and self-efficacy related to condom use. The program aimed to change perceptions through social marketing and behavior change programs in order to ultimately change sexual behaviors.

Objectives

The program aimed to increase condom use and condom-use predictors among Cameroonian youth.

Class and Type of Outcome or Behavior Change Targeted

• ☐ Decrease IDU risk

• ☑ Decrease noncommercial sex risk

• ☐ Decrease commercial sex risk

• ☐ Increase health services utilization (exams, testing, and treatment)

Target Population and Venue for HIV Prevention

The program targeted unmarried youth aged 15–24 in the cities of Yaoundé and Douala, the two largest cities in Cameroon. Peer education and promotion teams performed shows in schools and other places where youth congregated (e.g., soccer matches). The 100 % Le Journal magazine was sold in youth clubs and by street hawkers. Youth-friendly condom outlets supported the 100 % Jeune campaign and were located across the two intervention cities.

Pathways for Structural Change

• ☑ Changes in programs

• ☑ Changes in practices

• ☐ Changes in policies and laws

Strategies and Tactics for Structural Change

The 100 % Jeune program was an intensive social marketing program conducted in Cameroon that promoted adolescent reproductive health by empowering youth to practice safe sex, encouraging dialogue about adolescent reproductive health in the community, and making condoms accessible to youth. The main communication themes highlighted previous sexual history as a risk factor for STIs and HIV, emphasized the need for young girls to take responsibility for their reproductive health, and encouraged couples to discuss sensitive issues such as abstinence and condom use. The program used a variety of mass media and interpersonal communication methods to diffuse its messages and encourage youth to practice and adopt new, healthy behaviors.

Core Components

The 100 % Jeune campaign used a multifaceted mass media and interpersonal communication campaign, including:

• Peer Education and Promotion (PEP) Teams: PEP teams were used to reach both in- and out-of-school youth with participatory, interactive, and entertaining shows conducted at schools and youth hangouts (e.g., soccer matches, movie theaters, concerts, and street corners). PEP shows lasted about 1 h and consisted of a sketch, role-plays, a condom demonstration, discussions, and contests designed to improve HIV and STI and condom skills and knowledge. The PEP teams conducted 50–80 shows per month.

• 100 % Jeune Le Journal Magazine: A 12-page monthly magazine was designed to entertain and inform youth through articles on reproductive health issues, sports, and music. The magazine also included a reproductive health-oriented comic strip, a pullout poster, and correspondence from both readers and peer educators. The 100 % Jeune program sold the newspaper to youth newspaper hawkers on commission, who in turn sold the paper to the target audience.

• Radio Dramas and Talk Shows: An 18-episode Solange Let’s Talk About Sex radio drama reinforced the main themes of the campaign and addressed a wide range of issues such as sex for economic gain, cross-generational sex, HIV testing, and communication with parents about reproductive health issues. In July 2001, a promotional campaign for the drama involving small billboards, radio spots, brochures, and print ads in the 100 % Jeune magazine began. In addition, a 100 % Jeune call-in radio show was broadcast weekly in both Yaoundé and Douala. The shows included interviews with a featured guest, street interviews with youth, call-in questions from listeners, and popular music.

• An Integrated Mass Media Campaign: Additional television, radio, and billboard campaigns covered topics such as condom negotiation between partners, sexual history as a risk factor, HIV and other STIs, and empowering women to buy condoms. Television and radio spots were broadcasted from December 2000 to January 2001, during national youth week in February 2001, and from August through September 2001.

• Condom Provision: A network of branded youth-friendly condom outlets called Vendeurs Amis des Jeunes supplemented the 100 % Jeune communication campaign. Identified by stickers featuring the mascot of the 100 % Jeune program, the condom sales were launched in September 2001 by means of a workshop for 50 vendors. 100 % Jeune staff promoted the network through peer education and promotion teams, radio shows, and 100 % Jeune Le Journal. In addition, network participants received weekly visits from the PEP teams to monitor performance and to reinforce the value of the network.

Resources Required

The program required staff and funding to conduct peer educator shows, funding to produce a monthly magazine and 18-episode radio drama, funding to support a call-in radio show, and funding for television, radio, and billboard campaigns.

Management Structure

Population Services International (PSI) headquarters in Washington, DC, provided structural support for the program by helping to define the key determinants of behavior change and the types of messages to be delivered and assisted with general logistics. On the local level, individual teams supported each intervention component, for example, a newspaper team worked exclusively on developing and selling the 100 % Jeune Le Journal. PSI team members hired and trained Cameroonian youth and young adults to manage and fulfill the needs of the project, so that they were ultimately responsible for the day-to-day operations and creative expression of the intervention.

Implementation Themes

The 100 % Jeune program conducted an integrated social marketing campaign that included peer education; a monthly magazine; a radio drama and call-in show; television, radio, and billboard campaigns; and condom promotion to encourage youth in Cameroon to adopt safer sexual practices. The integration of different social marketing techniques was key to the program’s success (D. Ward, personal communication, September 14, 2011).

Main Challenges Faced

Hiring, training, and working with a continually changing group of peer educators proved to be difficult. PSI organized a group of ten youth in each city to deliver safer sex messages to peers. Peer educator youth were employed full time to act in this capacity, and they were constantly trained, updated, and observed by project staff to ensure that they were delivering medically accurate information to peers. There was a lingering concern that peer educators would pass along inaccurate information, however, if they were asked questions about unfamiliar topics. Peer educators did not remain a part of the program indefinitely since they needed to be the same age as the target audience. Every few years, PSI had to retire the current staff and hire and train new youth community members (D. Ward, personal communication, September 14, 2011).

During an initial needs assessment, PSI determined that only about 10 % of youth read printed materials, which would have been a serious challenge to their plan of reaching youth through a monthly magazine. However, PSI suspected that youth might not be reading printed resources simply because most were aimed at and written for adults. Indeed, once the youth-oriented 100 % Jeune Le Journal was produced, it became very popular with its intended audience (D. Ward, personal communication, September 14, 2011).

The program implementers also found it difficult to reach youth who were not enrolled in school, and they had to disperse education efforts into the community. Even reaching peers in schools was occasionally challenging, since school authorities had to be convinced of the benefit of teaching safer sex information to students (D. Ward, personal communication, September 14, 2011).

Because the findings from this study were based on cross-sectional data and since individuals could not be randomly assigned to the mass media campaign exposure or not, the causation of findings cannot be confirmed.

Program Continuity and Present-Day Status

The 100 % Jeune program was initiated in 2000, and components of the program, such as the website, newspaper, and online forum, are still in operation.

Other Locations and Regions that Have Implemented Similar Programs

Population Services International has implemented similar social marketing programs to decrease HIV transmission in more than 60 countries across the world, especially in regions of Southeast Asia and Africa.

Original Program Evaluation

Study Design

Timeline and Duration

The 100 % Jeune program was launched in 2000, and an evaluation study was implemented in 18-month intervals between 2000 and 2003. The first wave was implemented from July 26 to August 10, 2000; the second wave from January 4 to January 19, 2002; and the third wave from June 19 to July 16, 2003. The study reviewed here compared the first two cross-sectional waves.

Cohorts

• ☑ Cross-sectional (snap shots in time)

• ☐ Longitudinal (same people followed over time)

Temporal Direction of Data Collection Relative to Intervention

• ☑ Prospective

• ☐ Retrospective

Assessment Time Points (Temporal Comparison)

• ☑ Before and after intervention (baseline and follow-up measures)

• ☐ After only

• ☑ Serial (more than two measures taken over time)

Implementation Level (Geographic Comparison)

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☑ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☐ Individuals

The program operated in Yaoundé and Douala, the two largest cities in Cameroon.

Sampling Unit

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☑ Individuals

Recruitment Techniques

In each city, neighborhoods were selected with probability of selection proportional to population size. Within the selected neighborhoods, a total of 30 areas were identified, and households with at least one household member aged 15–24 were randomly selected within this area. Three interview attempts were made in each selected household, and no replacements were made for those who could not be reached.

Randomization

• ☐ No

• ☑ Yes

  • ☐ Random assignment

  • ☑ Random sampling

Random sampling was used to select households within previously selected neighborhoods.

Study Type

Quasi-experimental

Methods

Data Collection

Data Sources

• ☑ Questionnaire or survey

• ☐ Chart information or surveillance

• ☐ Record of biological specimen (e.g., urine sample)

Interview

• ☑ Interviewer administered

• ☐ Self-administered

Instruments

☐ Paper and pencil (data entry after fieldwork)

☐ Computer (ACASI or direct data entry in the field)

☑ Unknown

Modality

• ☑ In-person

• ☐ Mail

• ☐ Phone

• ☐ Internet

Data Analysis

Exposure Variables Measured

To measure exposure to the 100 % Jeune program, the researchers computed a composite indicator based on prompted recall of exposure to campaign elements.

Outcome Variables Measured

• ☑ Knowledge, attitudes, and beliefs

• ☑ Behaviors and practices

• ☐ Biomarker and clinical data

Outcome variables fell into five clusters drawn from the theoretical models informing the design of the program.

• 1. The perceived severity of the health threat was measured through questions such as “Do you believe that AIDS can be cured?”

• 2. To measure perceived risk, researchers asked questions such as “If you would not use condoms, would you say your risk of contracting HIV/AIDS would be high, moderate, low, or that there would be no risk?”

• 3. Perceived condom attributes and access was measured by asking participants whether condoms were effective for pregnancy prevention and for HIV/AIDS prevention.

• 4. The researchers measured several components of condom-use self-efficacy by asking questions such as “Would you be shy buying condoms in a shop near your home?”

• 5. Perceived social support was measured by asking questions such as “Do your parents support condom use by youth?”

Condom-use indicators included whether the respondent reported ever having used condoms, using a condom in the last sex act with a regular partner, or using a condom in the last sex act with a casual partner. In addition, respondents reported how often they used condoms with their regular and casual partners, which formed the basis of a consistent condom-use indicator.

Other Variables Measured

• ☑ Demographics

• ☐ Risk groups

• ☑ Behaviors

Age, gender, level of education attained, school enrollment status, socioeconomic status, and number of sexual partners were controlled for in the analyses.

Statistical Methods

The researchers used logistic regression analyses to examine trends in predictors and indicators of condom use. They analyzed data separately for male and female respondents, and because few females reported having casual partners, they limited the analyses of condom use with casual partners to males. In addition, they created logistic regression models to study the association between program exposure and predictors and indicators of condom use.

Strengths and Weaknesses of the Study Design and Methodology

• ☐ Cross contamination between intervention and comparison groups

• ☑ Concurrent interventions occurring

• ☐ Historical bias or trend due to historical factors

Two other reproductive health programs may have had some influence on the target population during the study. The Ministry of Public Health produced and distributed brochures that described the AIDS incidence among Cameroonian youth and encouraged protective practices. In 2002, the Ministry of Public Health distributed an estimated 2000 brochures nationwide. The women’s association Cercle des Amis du Cameroun (CERAC) began an AIDS prevention campaign for youth in Cameroon in 2001. CERAC disseminated AIDS prevention messages to young people through peer education efforts and by distributing notebooks and other materials containing campaign messages.

Results

Sample Size

	Yaoundé	Douala	Total
First wave	996	960	1,956
Second wave	1,583	1,654	3,237
Total	2,579	2,614	5,193

Retention and Loss to Follow-Up (Cohort Studies Only)

Not applicable to this study

Sample Demographics

Age

Years	First wave (%)	Second wave (%)
15–19	57	61
20–24	43	39

Race or Ethnicity

African, from Cameroon

Gender

	First wave (%)	Second wave (%)
Male	54	54
Female	46	46

Sexual Orientation

Not reported

Outcome Measures

Both study waves included more males than females, and the gender distribution did not vary across survey waves. Study staff anticipated the excess of males because they restricted the working sample to youth who had not been married or cohabitated (since females marry earlier than males, females were more likely to be omitted from the working sample). The percentage of youth aged 15–19 increased from 57 % to 61 % (p = 0.007) across the two surveys, indicating that the 2002 sample was slightly younger than that of 2000. The 2002 sample also had a lower percentage of high socioeconomic status youth (26 % vs. 34 %, p < 0.001). The percentage of youth with secondary education declined between the two surveys (from 90 % to 85 %, p < 0.001), as did the percentage enrolled in school (66–57 %, p < 0.001). Overall, the second wave was younger, less educated, less likely to be a student, and of lower SES status than the first wave.

The percentage of youth who reported hearing of at least one reproductive health program increased from 34.8 % in July–August 2000 to 41.9 % (p < 0.001) in January 2002. During the same period, spontaneous recall of the 100 % Jeune program increased from 1.3 % to 25.9 % (p < 0.001). About one in eight youth (11.9 %) had attended at least one of the 100 % Jeune peer education sessions, and more than one in twelve youth (7.6 %) reported having spoken personally with a peer educator.

The 100 % Jeune radio call-in show had good exposure, nearly half of all youth (47.3 %) had heard the call-in show in the 3 months before the survey, 11.2 % of all youth reported often listening, and 5.1 % reported always listening to the show. Exposure to the television and radio spots was high, with 64.5 % having heard the radio spots in the past 3 months. The Solange Let’s Talk About Sex radio drama had somewhat less reach; only 26 % of all youth reported having heard the drama. Exposure to the 100 % Jeune Le Journal magazine was high; three out of four youth (73.9 %) had read at least one issue. In addition, 18.4 % of all youth reported often reading it, and 15.9 % reported always reading it. A third of youth had heard of the “Vendeurs Amis des Jeunes” youth-friendly outlets in the past 3 months, but only 5.5 % reported having visited them in that same period.

The percentage of females who believed that an HIV-positive person could survive the disease increased from 14 % in 2000 to 35 % in 2002 (p < 0.001); for males, this was the case for 25 % and 39 %, respectively (p < 0.01). The percentage of females who believed they had a moderate to high personal risk of HIV infection increased from 52 % in 2000 to 66 % in 2002 (p < 0.001); among males, the percentage stayed constant at nearly 75 %. Knowledge of a nearby condom source improved considerably from 2000 to 2002. Among females, the percentage who knew of a condom source within 10 min travel increased from 62 % in 2000 to 68 % in 2002 (p < 0.001); for males, it increased from 80 % to 84 % (p < 0.001). There was also a significant, albeit small, increase in the percentage of males who believed that condoms were effective for family planning (80 % to 83 %, p < 0.05).

The percentage of females reporting not being shy about obtaining condoms increased from 43 % to 56 % (p < 0.001); for males, it increased from 61 % to 68 % (p < 0.001). The percentage that expressed confidence that they knew how to use a condom correctly increased from 39 % to 47 % for females (p < 0.001) and from 66 % to 72 % for males (p < 0.001). The perception that youth can convince their partners to use condoms was already high at the first survey, and no further improvement was noted during the course of the program.

Perceived parental support for adolescent condom use increased for both genders, from 59 % to 68 % for females (p < 0.001) and from 64 % to 75 % for males (p < 0.001). Discussion of HIV and other STIs with friends increased significantly among both females and males but decreased significantly with others for both females (37.6 % to 31.0 %, p < 0.05) and males (45.7 % to 35.9 %, p < 0.001).

Most indicators showed that condom use increased significantly. Between 2000 and 2002, the percentage of youth whoever used condoms increased from 51 % to 62 % for females (p < 0.001) and from 58 % to 65 % for males (p < 0.001). The percentage that used a condom at last intercourse with a regular partner increased from 32 % to 45 % for females (p < 0.001) and from 44 % to 61 % for males (p < 0.001). Among females, there was a corresponding increase in reported consistency of condom use with regular partners. For both males and females, the consistency of condom use increased with casual partners (p < 0.05 for males and females).

Both males and females who had high exposure to the 100 % Jeune program experienced lower barriers to condom use. Exposure to the program was associated with significantly higher levels of self-efficacy and perceived social support for condom use. The data showed a strong association between program exposure and the respondents’ reported knowledge of correct condom use. Among females, the percentage who reported feeling confident that they knew how to use a condom correctly varied from 38 % for those with low exposure to 51 % for those with medium-high exposure (p < 0.01, compared to 2000 data) to 64 % for those with high exposure (p < 0.01, compared to 2000 data). Similarly, high program exposure was associated with lower levels of shyness to obtain condoms for both males and females.

Reviewing indicators of condom use showed that among males, exposure to the 100 % Jeune program was associated with a significantly higher level of ever having used condoms (p < 0.05), as well as use at last intercourse with a regular partner (p < 0.01). Those associations supported the idea that the 100 % Jeune program contributed to the observed increase in condom use for males. In contrast, for females, exposure to 100 % Jeune did not appear to contribute to the observed increase in condom use.

Conclusions

The multifaceted mass media and community outreach activities implemented by the 100 % Jeune social marketing program proved to reach and influence the target youth population. Exposure to the different program elements was associated with reductions in barriers to condom use (such as perceived social support and condom-use self-efficacy) and increased use of condoms.

A key element of the program was that youth were targeted repeatedly through many different channels, including television, radio, and billboard campaigns; peer education; a magazine; and radio drama and call-in shows, in addition to the provision of condoms through youth-friendly condom outlets. Many of the positive changes in behavior and attitudes were associated with high exposure to at least two different campaign elements, suggesting that program efficacy increased by reaching the target audience in varied and repeated ways.

Implications and Lessons Learned

While the 100 % Jeune program effectively increased indicators of safer sex among Cameroonian youth, the impact of the program varied across groups. For example, the program appeared to be more effective for males than females. The program developers reasoned that the difference was because the program had a limited impact on young women’s risk perception and on their perceived ability to convince their partners to use condoms, which are both key predictors of condom use by females.

In response to the varied impact, the program developers refined the program messages and activities following the survey wave in 2002. They encouraged parents to discuss reproductive health issues with their children, revised program materials and messages to appeal to females, and integrated participatory approaches into the peer education sessions.

PSI chose the two largest cities in Cameroon to reach as many young people as possible and employed mass media and social marketing. The program was able to foster a certain amount of interactivity into its components. The radio show allowed listeners to call in with questions or comments, and a doctor on staff answered medical, sexual, or health questions from callers. The 100 % Jeune Le Journal published questions from readers and staff members wrote responses to the questions. By using a variety of methods, PSI designed a cost-effective method of reaching youth and changing behavior (D. Ward, personal communication, September 14, 2011).

Supplementary Materials Available

Additional References

The official website of the 100 % Jeune project can be found at this address:

http://www.reglo.org/.

Plautz, A. & Meekers, D. (2007). Evaluation of the reach and impact of the 100 % Jeune youth social marketing program in Cameroon: Findings from three cross-sectional surveys. Reproductive Health, 4(1), 1–15.

Case Study 9: Twenda na Wakati (Lets Go with the Times): Effects of a Radio Soap Opera on HIV/AIDS Prevention in Tanzania

Original Program Developers and Evaluators

Peter Vaughan

Everett Rogers

Arvind Singhal

Ramadhan Swalehe

Case Study Contents

• Abstract

• Program at a Glance

• Program Information and Implementation

• Original Program Evaluation

• Implications and Lessons Learned

• Supplementary Materials Available

Bibliography

Vaughan, P. W., Rogers, E. M., Singhal, A., & Swalehe, R. M. (2000). Entertainment-education and HIV/AIDS prevention: A field experiment in Tanzania. Journal of Health Communication, 5, 81–100.

Abstract

Tanzania forms part of the AIDS belt, where 2 % of the world’s population lives, but nearly 50 % of the world’s AIDS cases are found. The number of AIDS cases in Tanzania approached 450,000 by 1996, with heterosexual contact as the main mode of transmission. In 1993, Radio Tanzania, in collaboration with the Ministry of Health and the Ministry of Community Development, Women Affairs, and Children, developed and broadcasted a radio soap opera, Twende na Wakati (Let’s Go with the Times), designed to help prevent the spread of HIV/AIDS by increasing audience members’ HIV/AIDS knowledge and preventive behavior. Airing twice per week for 30 min from July 1993 through 1999 in Tanzania with the exception of the Dodoma region, Twende na Wakati addressed four main HIV/AIDS prevention themes. The Dodoma region served as a comparison area from 1993 to 1995 and subsequently received the intervention.

To evaluate the efficacy of the program, project staff conducted surveys in 15 randomly selected wards in the comparison area and 27 randomly selected wards in the treatment area, surveying female citizens aged 15–49 and male citizens aged 15–60 once per year starting in 1993. Surveys measured personal characteristics, exposure to and perceptions of Twende na Wakati and other HIV/AIDS information sources, knowledge of HIV/AIDS, relevant attitudes, and HIV/AIDS prevention behaviors. Analyses showed improvements in the intervention area relative to the comparison area in HIV/AIDS: (1) knowledge over time (as measured by the HIV/AIDS knowledge scale), (2) attitudes such as the personal risk of HIV, and (3) preventive behaviors such as a decrease in the number of partners.

Program at a Glance

Goal: To use a radio soap opera to increase knowledge about HIV/AIDS and promote HIV prevention behaviors

Target Populations: All community members in Tanzania with a focus on men and women of reproductive age

Geographic Location and Region: All of Tanzania (except for the Dodoma region, which received the intervention 2 years later)

Establishment and Duration: July 1993 through 1999 in the designated areas

Resources Required and Goods and Services Provided: Resources to create a culturally relevant entertainment-educational radio program and enough funds to develop, produce, and record the program, as well as means of paying radio stations

Strategies and Components

• Utilized an educational-entertainment strategy to disseminate health information

• Connected with listeners by depicting negative, transitional, and positive role models

• Modeled behavior change and HIV/AIDS prevention strategies

Key Partners: Helen Lang Charitable Trust; Weyerhaeuser Family Foundation; Rockefeller Foundation; United Nations Population Fund; Radio Tanzania; the Ministry of Health; the Ministry of Community Development, Women Affairs and Children; and the Population Communications International of New York

Key Evaluation Findings

Statistically Significant

• Increased HIV/AIDS knowledge

• Increased perceptions of personal risk for HIV infection

• Decreased the percentage of people who felt they were not at risk for HIV but actually reported having one or more risk factors

• Increased individuals’ sense of efficacy with respect to HIV/AIDS

• Decreased the number of sexual partners for sexually active men and women

• Increased condom use for individuals with more than one sex partner

No Effect

• Did not decrease the belief that one can get HIV from condoms

• Did not change beliefs about the number of acceptable sexual partners for men and women

• Did not change sharing of razors and needles

Program Information and Implementation

Background, History, and Public Health Relevance

At the outset of the Twende na Wakati program, Tanzania had one of the highest HIV infection rates in the world. In 1996, the number of AIDS cases in this East African country approached 450,000, and an estimated 1.4 million individuals were living with HIV. In contrast to more developed countries where homosexual contact and IV drug use contribute significantly to the transmission of HIV, in Tanzania, heterosexual intercourse causes about 90 % of HIV infections. The widespread practice of extramarital sex, especially by men engaging in casual sex, contributes dramatically to the problem. Believing they are in monogamous relationships, many women unknowingly risk contracting HIV from a partner who has been infected by another casual sex partner. In addition in Tanzania, the prevalent misunderstandings about HIV/AIDS (such as the belief that the condom lubricant can cause HIV), an irregular availability of condoms, and poor knowledge of where to obtain them compound the spread of HIV.

Radio Tanzania, in collaboration with the Ministry of Health, and the Ministry of Community Development, Women Affairs, and Children, promoted HIV/AIDS knowledge and prevention behaviors in Tanzania by forming the Twende na Wakati program—an intervention strategy combining entertainment with education, using role models to teach the benefits of HIV prevention behaviors and the consequences of practicing risky behaviors.

Theoretical Basis

Elements of the Health Belief Model, Theory of Reasoned Action, Social Cognitive Theory, Diffusion Theory, Social-Movement Theory, and Staged Model of Behavior Change shaped the design and evaluation of the program. The developers integrated several concepts, such as the stages of behavioral change, self-efficacy, role modeling, homophily, self-relevancy, and interpersonal communication in planning the mass media program and assessing its effects, paying particular attention to entertainment-education concepts and strategies.

Objectives

The program aimed to increase knowledge about HIV/AIDS and promote HIV prevention behaviors through a radio soap opera.

Class and Type of Outcome or Behavior Change Targeted

• ☐ Decrease IDU risk

• ☑ Decrease noncommercial sex risk

• ☐ Decrease commercial sex risk

• ☐ Increase health services utilization (exams, testing, and treatment)

Target Population and Venue for HIV Prevention

The program targeted men and women in the general population in their homes or public venues that had radios.

Pathways for Structural Change

• ☐ Changes in programs

• ☑ Changes in practices

• ☐ Changes in policies and laws

Strategies and Tactics for Structural Change

The program used an educational-entertainment strategy to disseminate health information to the general population through a radio soap opera. It connected with listeners by depicting negative, transitional, and positive role models and modeled behavior change and HIV/AIDS prevention strategies.

Core Components

The Twende na Wakati soap opera used role modeling of characters to encourage and help listeners adopt safer HIV prevention practices in their own lives. It used negative, transitional, and positive role model characters in order to present HIV prevention behaviors to its audience. Negative role models showed the consequences of engaging in risky sexual behaviors. Positive characters provided wise advice and counsel to others. Transitional characters, to whom listeners could relate, served as self-efficacious models for behavior change.

Resources Required

The program required radio stations willing to broadcast the programs and funds to write, produce, and air the show.

Management Structure

Not reported

Implementation Themes

Radio Tanzania used a popular radio format to broadcast Twende na Wakati in Swahili, the national language of Tanzania. The radio show presented four HIV/AIDS prevention themes: (1) STIs should be medically treated, (2) condoms help prevent HIV, (3) AIDS is an incurable disease spread by sexual contact, and (4) various rumors about HIV/AIDS are false. In addition, the radio show addressed family planning, gender equity, and other health issues.

Main Challenges Faced

The Twende na Wakati team found that one of their questions in the first round of the survey prompted a bias response, so they changed the question in the following years. In addition, they could not randomly select the treatment and comparison areas for a variety of practical reasons, they could not easily measure the degree of exposure to the intervention, they could not prevent travel between the comparison and intervention areas, and other HIV/AIDS programs occurred concurrently.

Program Continuity and Present-Day Status

In Tanzania, Twende na Wakati was only scheduled to run until 1997. The evaluation of the program was so positive that in 1997 UNFPA renewed the broadcasts for four more years.

Other Locations and Regions that Have Implemented Similar Programs

A similar radio soap opera called “Banadda Twegande” (Together We Will) was implemented in central Uganda by UNFPA, the Population Secretariat, Central Broadcasting Station, and AHADI. It aimed to reduce STIs including HIV and promote condom use by increasing knowledge of sexual health issues and awareness of reproductive rights. This educational-entertainment strategy specifically targeted farmers, plantation and factory workers, and community service motor cyclists during the radio soap opera and in a weekly comic strip in the newspaper.

Original Program Evaluation

Study Design

Timeline and Duration

The Twende na Wakati radio program aired from July 1993 through 1999 (6½ years). From 1993 to 1995, the Dodoma region of Tanzania did not receive the broadcasts and served as a comparison area for the program evaluation. Beginning in 1995, the first 2 years of Twende na Wakati were rebroadcast in Dodoma, and the researchers not only had a comparison group for the 1993–1995 period, but they also had an opportunity to measure the effects of the radio program in a new area.

Cohorts

• ☑ Cross-sectional (snap shots in time)

• ☐ Longitudinal (same people followed over time)

Temporal Direction of Data Collection Relative to Intervention

• ☑ Prospective

• ☐ Retrospective

Assessment Time Points (Temporal Comparison)

• ☐ Before and after intervention (baseline and follow-up measures)

• ☐ After only

• ☑ Serial (more than two measures taken over time)

Implementation Level (Geographic Comparison)

• ☐ Countries

• ☑ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☑ Households

• ☐ Couples, pairs, and dyads

• ☐ Individuals

Within the intervention and comparison areas, units were selected randomly from each of the prior units in the following order: districts, wards, villages, ten-cell units, household leaders and households, and eligible household members.

Sampling Unit

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☑ Households

• ☐ Couples, pairs and dyads

• ☑ Individuals

Recruitment Techniques

Not reported

Randomization

• ☐ No

• ☑ Yes

  • ☐ Random assignment

  • ☑ Random sampling

The researchers used cluster sampling, both designated and random. The first step was nonrandom to designate intervention and comparison areas. Subsequent steps used cluster sampling with random selection.

Study Type

Quasi-experimental

Methods

Data Collection

Data Sources

• ☑ Questionnaire or survey

• ☐ Chart information or surveillance

• ☐ Record of biological specimen (e.g., urine sample)

Interview

• ☐ Interviewer administered

• ☐ Self-administered

Not reported

Instruments

• ☐ Paper and pencil (data entry after fieldwork)

• ☐ Computer (ACASI or direct data entry in the field)

Not reported

Modality

• ☑ In-person

• ☐ Mail

• ☐ Phone

• ☐ Internet

Data Analysis

Exposure Variables Measured

The exposure variable was listening to the Twende na Wakati radio program.

Outcome Variables Measured

• ☑ Knowledge, attitudes, and beliefs

• ☑ Behaviors and practices

• ☐ Biomarker and clinical data

The surveys measured (1) personal characteristics, (2) exposure to and perceptions of Twende na Wakati, (3) knowledge of HIV/AIDS, (4) relevant attitudes toward HIV/AIDS, and (5) HIV/AIDS preventive behaviors.

Other Variables Measured

• ☑ Demographics

• ☐ Risk groups

• ☐ Behaviors

Control variables included listenership to other HIV/AIDS educational programs, access to local family planning services, age, religion, radio ownership, marital status, parity, income, gender, electricity in the home, formal education, and rural/urban residence.

Statistical Methods

To examine the effects of Twende na Wakati, the researchers computed three different statistical tests on each dependent variable. First, using logit linear models for categorical dependent variables and analysis of variance models for continuous dependent variables, they tested for a significant treatment-by-year interaction term using data from 1993 to 1995. Those tests measured whether there were greater changes in the dependent variables in the implementation area versus the comparison area and did not control for other independent variables that might have influenced the results.

Second, the researchers used either logistic regression for categorical dependent variables or ANOVA models for continuous dependent variables to test for a significant treatment-by-year interaction term while controlling for eight other independent variables and radio ownership.

Third, they aggregated the individual respondents’ data to the level of the 35 wards of the study. They calculated change scores for the dependent and the independent variables and were able to control in part for geographical differences in initial conditions and see the effects of ward-level exposure to Twende na Wakati as a dose response created by geographic variation in exposure to the program.

Strengths and Weaknesses of the Study Design and Methodology

• ☑ Cross contamination between intervention and comparison groups

• ☑ Concurrent interventions occurring in experimental and comparison areas

• ☐ Historical bias or trend due to historical factors

Initial 1994 analyses showed a large percentage of Twende na Wakati listeners responding that they had adopted HIV/AIDS prevention behaviors as a result of listening to the program. However, the independent Tanzania Demographic and Health Survey from 1996 showed that a similarly large percentage of people in the general population also changed their sexual behaviors to protect themselves against HIV during that time, suggesting that the survey question may have prompted positive responses from individuals who would have changed behaviors even in the absence of Twende na Wakati. To account for this, the researchers changed their questions in 1996.

Three different phenomena threatened the internal validity of the study. First, the comparison area was chosen nonrandomly, which led to some differences in the initial conditions between the intervention and comparison areas. The researchers used multivariate statistical tests to minimize the impact of the regional differences. Second, not everyone in the intervention area heard Twende na Wakati, and some people in the comparison area did. Most likely due to the difficulty of travel and the low incomes of people in Tanzania, however, only 2 % of the sample in the comparison area reported having heard the program prior to its airing there. Ward-level analysis also helped to control for the different levels of treatment exposure among the different ward populations by testing for a dose response. Third, the 5-year study took place in a large and diverse country, where other development initiatives were underway at the same time. To limit historical contamination, they controlled for exposure to other radio programs with HIV/AIDS content, radio ownership, and household electricity.

Results

Sample Size

	1993	1994	1995	1996	1997	Total
Intervention area	1,793	1,924	1,940	1,919	1,933	9,509
Comparison area	859	861	861	831	624	4,036
Total	2,652	2,785	2,801	2,750	2,557	13,545

Retention and Loss to Follow-Up (Cohort Studies Only)

Not applicable to this study

Sample Demographics

Age

Not reported

Race or Ethnicity

African, from Tanzania

Gender

Not reported

Sexual Orientation

Not reported

Outcome Measures

Exposure to Twende na Wakati

Although more respondents in the comparison area owned radios than in the intervention area, radio ownership increased in both areas over the course of the study. About half of all survey respondents reported radio as the most important source of HIV/AIDS information.

Concurrent to Twende na Wakati, several other radio programs with HIV/AIDS educational content aired across Tanzania. Reported listening to Twende na Wakati in the intervention area reached 47 % in 1994 and increased to 58 % by 1997, and about 60 % of those listening reported listening regularly (at least once a week). In the comparison area between 1993 and 1995, exposure to Twende na Wakati reached only 2 %. When the program aired in the comparison area in 1995, exposure increased to 51 % in 1996 and 75 % in1997.

HIV/AIDS Knowledge

After the first year of the radio broadcast, 73 % of Twende na Wakati listeners in the intervention area reported learning about AIDS from the radio program, increasing to 85 % in 1997. In the comparison area (after introducing the Twende na Wakati broadcast in 1995), 77 % of listeners reported learning about AIDS from the radio program in the first year of broadcast, increasing to 89 % by the second year.

The researchers also found changes over time as measured by their HIV/AIDS knowledge scale. Between 1993 and 1995, residents in the intervention area showed an increase in HIV/AIDS knowledge (0.7 point increase, from 10.0 to 10.7), while residents in the comparison area showed a decrease in this knowledge (0.5 point decrease from 10.9 to 10.4). Listeners to Twende na Wakati generally scored 1.5 points higher on the knowledge scale than non-listeners.

HIV/AIDS Attitudes

The percentage of listeners who reported talking to someone about the AIDS educational content of Twende na Wakati in the intervention area increased steadily between 1994 and 1997.

Respondents in the intervention area also showed significant changes in attitudes about HIV/AIDS, with increases in the perceptions of (1) personal risk for HIV, (2) personal risk of HIV/AIDS while also reporting one or more risk factors, and (3) efficacy with respect to HIV/AIDS. The percentage of people in the intervention area who could provide an efficacious response to the question, “What would you do if a doctor told you that you had HIV/AIDS?” also increased significantly in contrast to the comparison area.

HIV/AIDS Prevention and Risk-Taking Behaviors

The percentage of listeners who reported adopting an HIV/AIDS prevention measure as a result of listening to Twende na Wakati increased in the intervention area between 1994 and 1995 (82 % in 1995 compared to 73 % in 1994), with listeners largely reporting a reduction in their number of sexual partners (77 % in 1995), rather than an increase in their use of condoms (15 %) or their ceasing to share razors (6 % in 1995).

Study data indicated that two of the behavioral variables (monogamy and condom use) did correlate with better HIV/AIDS knowledge and higher HIV/AIDS self-efficacy. The number of reported sexual partners of sexually active men declined in both the intervention (0.7 partners) and comparison areas (0.3 partners) from 1993 to 1995, and the difference in the decreases was statistically significant (p = 0.01). In addition, when the radio show aired in 1995 to 1997 in the comparison area, the number of partners further declined by 0.6 in the that area, while in the intervention area it declined further by only 0.2 partners.

The number of reported sexual partners of sexually active women declined in both the intervention and comparison areas from 1993 to 1995. The decline was greater in the intervention area (0.7) than in the comparison area (0.5), and this difference was statistically significant. After airing the radio show in the comparison area in 1995, rates declined by 0.1, but the rates in the intervention area did not further decline.

The percentage of respondents who reported currently using condoms in the intervention area increased from 1993 to 1995 (6 % vs. 13 %). While during that time in the comparison area, the percentage declined (15 % vs. 2 %). The logit, loglinear, and logistic regression tests indicated that those differences were statistically significant, but the MLR test did not. Notably, after introducing the program to the comparison area from 1995 to 1997, condom use increased by only 3 percentage points to 16 % in the intervention area, whereas in the comparison area, it increased 11 percentage points to 13 %.

During Twende na Wakati, both the intervention and comparison areas showed equal declines in razor sharing and needle sharing, demonstrating no apparent intervention effect on these behaviors.

Conclusions

The researchers found that the entertainment-educational radio soap opera, Twende na Wakati, had measurable effects on HIV/AIDS knowledge, attitudes, and prevention behaviors. They concluded that the intervention led to (1) decreases in the number of sexual partners for both men and women, (2) increases in the number of people with more than one sex partner who reported currently using condoms, and (3) increases in other HIV/AIDS prevention behaviors. Twende na Wakati appeared to foster adoption of HIV/AIDS prevention behaviors wherever it aired.

Implications and Lessons Learned

Much of Twende Na Wakati’s success may be attributed to its ability to incorporate educational material about HIV prevention into a highly entertaining drama with good stories and great characters. With nine story lines running at any given time, and 18 main characters, the soap opera appeared to gain widespread popularity among listeners in a short amount of time. According to the program developers, controversial subjects were introduced slowly into the drama, so that listeners had the opportunity to engage with the drama and understand characters before educational topics were introduced.

Supplementary Materials Available

Additional References

Rogers, E.M. & Singhal, A. (1999). Entertainment education: A communication strategy for social change. Mahwah, NJ: Lawrence Earlbaum Associates Inc.

Singhal, A., Cody, M., Rogers, E., & Sabido, M. (Eds.). (2004). Entertainment-education and social change: History, research, and practice. Mahwah, NJ: Lawrence Earlbaum Associates Inc.

Case Study 10: A Safer Sex Media Campaign in Lexington, Kentucky: Using Public Service Announcements to Promote Condom Use

Original Program Developers and Evaluators

Rick Zimmerman

Philip Palmgreen

Seth Noar

Mia Liza Lustria

Hung-Yi Lu

Mary Lee Horosewski

Case Study Contents

• Abstract

• Program at a Glance

• Program Information and Implementation

• Original Program Evaluation

• Implications and Lessons Learned

• Supplementary Materials Available

Bibliography

Zimmerman, R. S., Palmgreen, P. M., Noar, S. M., Lustria, M. L., Lu, H., & Horosewski, M. L. (2009). Effects of a televised two-city safer sex media campaign targeting high-sensation-seeking and impulsive-decision-making young adults. Health Education & Behavior, 34(5), 810–826.

Abstract

The researchers designed televised condom promotion public service announcements to target young adults who were high sensation seekers and impulsive decision-makers. Television stations in Lexington, Kentucky, showed the announcements from January through April 2003. The researchers evaluated the program by conducting interviews before and after the intervention in Lexington and in Knoxville, Tennessee, a similar city where the intervention was not aired. The surveys gathered information about sexual risk behaviors, condom use, condom self-efficacy, and condom use intentions.

The media campaign increased condom use, condom use intentions, and condom self-efficacy among the target population in the intervention city. The effects were somewhat short-lived, however, and there was a wearing-out effect after the announcements were taken off the air. The researchers estimated that the levels of the effects were higher at the final evaluation than they would have been had the intervention never occurred. The effect size of the intervention was slightly greater than small (Cohen’s d = 0.26); nevertheless, an estimated 181,224 unprotected sexual intercourse acts were prevented over the course of the study as a result of the intervention.

Program at a Glance

Goal: To increase condom use, condom self-efficacy, and condom-use intentions among high sensation-seeking and impulsive-decision-making young adults through the use of safer sex televised public service announcements

Target Populations: High sensation-seeking and impulsive-decision-making young adults aged 18–23 in Lexington, Kentucky. The individuals in the geographical region were mainly Caucasian (80.7 % in the study), but the media programs were balanced to target both Caucasians and African Americans.

Geographic Location and Region: Lexington, Kentucky, served as the intervention site and Knoxville, Tennessee, served as the comparison site. The researchers chose those two moderate-sized southeastern cities in the United States because they were demographically similar.

Establishment and Duration: The researchers initiated the project in August 2001. Television stations in Lexington ran the public service announcements from January through April 2003.

Resources Required and Goods and Services Provided: Scripts, actors, and funds to create safer sex themed public service announcements; funds to pay for television time; funds to support the services of a professional media buyer to coordinate the placement of announcements; and funds and staff to support focus-group activities to evaluate the announcements

Strategies and Components

• Designed public service announcements about safer sex and tested their appeal with focus groups

• Placed announcements in television shows popular with sensation-seeking and impulsive-decision-making young adults

• Utilized an interrupted time-series design with a control community to evaluate the campaign

Key Partners: A grant from the National Institute of Mental Health supported the development and evaluation of the media campaign.

Key Evaluation Findings

Statistically Significant

• Reached the target audience successfully

• Increased condom self-efficacy, condom-use intentions, and condom use in the high-risk group

No Effect

• No change in condom use, condom self-efficacy, or condom-use intentions in the low-risk group

Program Information and Implementation

Background, History, and Public Health Relevance

Mass media messages have the potential to reach a large number of individuals and may be an effective means to support safer behaviors across an entire population. Many interventions have used mass media campaigns to promote safer sex behaviors, such as decreasing risky behaviors that lead to the transmission of HIV and other STIs. Media campaigns have contributed to an increase in HIV testing services in clinics and in an increase in AIDS information discussions between family and friends. Safer sex media messages may be able to target populations at greater risk for acquiring HIV and other STIs over other, lower-risk populations. The media campaign in Lexington, Kentucky, aimed to change risky sexual behaviors among high sensation-seeking and impulsive-decision-making young adults by tailoring media messages to them.

Theoretical Basis

The researchers relied on a multi-theoretical approach to guide the content of public service announcements (PSAs). They used elements of Social Cognitive Theory, such as modeling and self-efficacy related to condom use. They also employed the Theory of Planned Behavior and depicted the negative consequences of not using condoms in some announcements. The researchers also took into account the Trans-Theoretical Model and the Stages of Change Model when designing the announcements. They aimed to create PSAs that covered the range of theoretical concepts—from those that focused on STIs and HIV for individuals who might not even be thinking of condom use to those that targeted self-efficacy and skills (R. Zimmerman, personal communication, March 9, 2011).

Objectives

The program aimed to increase condom use, condom self-efficacy, and condom-use intentions among high sensation-seeking and impulsive-decision-making young adults.

Class and Type of Outcome or Behavior Change Targeted

• ☐ Decrease IDU risk

• ☑ Decrease noncommercial sex risk

• ☐ Decrease commercial sex risk

• ☐ Increase health services utilization (exams, testing, and treatment)

Target Population and Venue for HIV Prevention

The program aimed to influence young adults in Lexington, Kentucky, who were high sensation seeking and impulsive decision making.

Pathways for Structural Change

• ☐ Changes in programs

• ☑ Changes in practices

• ☐ Changes in policies and laws

The program focused on changing people’s sex and condom practices by changing the acceptability of safer sex and condom usage.

Strategies and Tactics for Structural Change

The program designed televised public service announcements about safer sex, tested their appeal through focus groups, and placed the PSAs in television shows popular with the target audience.

Core Components

The researchers designed PSA messages to target sensation-seeking and impulsive-decision-making young adults in order to persuade them to make safer sex choices. They created a set of PSAs and then used focus groups to evaluate their relevance to the intended audience. Televisions stations showed the PSAs during television programs popular with the targeted audience over the course of 4 months. The researchers collected information about what types of television shows the target audience watched, and a professional media buyer helped to choose the types of shows and networks that should carry the PSAs.

The PSAs addressed four different themes in order to promote safer sex behaviors: (1) three videos addressed threat and presented facts about the HIV infection, (2) four videos addressed personal risk and focused on HIV transmission, (3) three videos presented information about the costs and benefits of engaging in safer sex, and (4) six videos demonstrated skills such as self-efficacy in negotiating condom use in challenging situations and modeling condom-use skills and abilities.

Resources Required

The campaign required scripts, actors, and funds to create safer sex themed public service announcements, funds to pay for PSA presentation time, and funds to support the services of a professional media buyer who coordinated PSA placement. The media campaign relied on extensive market research, and funds were required to support focus-group activities.

Management Structure

The campaign had a principal investigator, a co-principal investigator, and an investigator. Graduate students worked on the project, a media buyer placed PSAs, and survey research centers collected data (R. Zimmerman, personal communication, March 9, 2011).

Implementation Themes

The campaign relied on televised public service messages that were targeted to the intended audience and were broadcast at times and in conjunction with shows that the audience watched.

Main Challenges Faced

During the 9 months before the PSAs began to air, the intervention and control communities showed different trends in condom use, condom-use intentions, and condom self-efficacy. During the campaign, the intervention community population reported steady declines in each of the measures over time until the PSAs aired. The control sample did not follow the same trend and essentially remained stable on all of the measures throughout the duration of the study. Because there was a great amount of variation in the measures in the control community, the researchers suspected that there might have been some sampling problems in the control city even though the survey staff were identically trained in the two cities. In addition, the researchers also noted that the large proportion of Caucasian individuals in the two cities might limit the generalizability of the study’s findings.

Program Continuity and Present-Day Status

The media campaign lasted from January 2003 to April 2003.

Other Locations and Regions that Have Implemented Similar Programs

No similar media campaigns, which specifically target high sensation-seeking and impulsive-decision-making young adults to reduce STI and HIV risk, had been implemented before the current program.

From August 2006 to January 2008, six cities in the United States aired an HIV prevention and condom promotion media campaign. This campaign specifically targeted African American youth, a demographic group shown to be at greater risk of acquiring STIs including HIV. The mass media campaign included culturally appropriate TV and radio ads that were designed to appeal to and actively engage African American youth by featuring hip-hop music and African American actors. These ads stressed the importance of using condoms during 30-s TV ads and 60-s radio ads. The media campaign improved outcome expectancies regarding condoms and reduced unprotected sexual contacts among youth infected with STIs.

Original Program Evaluation

Study Design

Timeline and Duration

The researchers initiated the project in August 2001, and television stations in Lexington ran the PSAs from January through April 2003. The researchers conducted surveys in Lexington and the control city, Knoxville, every month from May 2002 until January 2004.

Cohorts

• ☑ Cross-sectional (snap shots in time)

• ☐ Longitudinal (same people followed over time)

Temporal Direction of Data Collection Relative to Intervention

• ☑ Prospective

• ☐ Retrospective

Assessment Time Points (Temporal Comparison)

• ☐ Before and after intervention (baseline and follow-up measures)

• ☐ After only

• ☑ Serial (more than two measures taken over time)

Implementation Level (Geographic Comparison)

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☑ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☐ Individuals

The intervention was conducted in Lexington, Kentucky, and Knoxville, Tennessee, served as a control. The two cities were demographically similar.

Sampling Unit

• ☐ Countries

• ☐ Regions

• ☐ Counties

• ☐ Cities

• ☐ Towns

• ☐ Villages

• ☐ Households

• ☐ Couples, pairs, and dyads

• ☑ Individuals

Recruitment Techniques

The researchers used random digit dialing and called random samples of registered students at the University of Kentucky and University of Tennessee. The researchers included students in the study if they were eligible according to a brief screener interview and were heterosexually active in the past 3 months, a US citizen, and in the appropriate age range.

Randomization

• ☐ No

• ☑ Yes

  • ☐ Random assignment

  • ☑ Random sampling

The researchers used a 21-month controlled time-series design and surveyed independent cross-sectional samples of 100 randomly selected young adults in each city each month.

Study Type

Quasi-experimental

Methods

Data Collection

Data Sources

• ☑ Questionnaire or survey

• ☐ Chart information or surveillance

• ☐ Record of biological specimen (e.g., urine sample)

Interview

• ☐ Interviewer administered

• ☑ Self-administered

Instruments

• ☐ Paper and pencil (data entry after fieldwork)

• ☑ Computer (ACASI or direct data entry in the field)

Modality

• ☑ In-person

• ☐ Mail

• ☐ Phone

• ☐ Internet

Individuals completed the survey with a self-administered interview loaded on a laptop computer either at a research center or in their own homes. The interviews were private and anonymous, and participants provided informed consent before participating.

Data Analysis

Exposure Variables Measured

Survey respondents self-reported whether or not they had seen at least one of the public service announcements.

Outcome Variables Measured

• ☑ Knowledge, attitudes, and beliefs

• ☑ Behaviors and practices

• ☐ Biomarker and clinical data

The surveys queried participants about sexual descriptors, condom self-efficacy, condom-use intentions, condom use, and how often they had seen each PSA.

Other Variables Measured

• ☑ Demographics

• ☐ Risk groups

• ☑ Behaviors

The researchers also measured demographics and sensation seeking and impulsive decision making with two scales.

Statistical Methods

The researchers separated the samples into high and low sensation seekers and compared the two groups across a number of sexual risk variables. They also analyzed monthly means for all dependent variables with a regression-based ITS procedure, by modeling slopes and also intercept and slope changes as a result of the campaign. The researchers also extrapolated the pre-campaign regression lines in the Lexington group to estimate dependent variable values assuming the intervention had never taken place.

Strengths and Weaknesses of the Study Design and Methodology

• ☐ Cross contamination between intervention and comparison groups

• ☐ Concurrent interventions occurring in experimental and comparison areas

• ☑ Historical bias or trend due to historical factors

Before the campaign began, condom use was decreasing in Lexington and was essentially flat in Knoxville. Each community may have been experiencing different historical trends, making comparisons between the communities more difficult.

Results

Sample Size

Out of the 199,940 telephone numbers the researchers called, 94 % did not yield participants. The researchers spoke to 8,315 individuals, and 60 % agreed to complete a brief screener interview. Of those screened and determined to be eligible (4,989), a total of 4,032 individuals completed an interview during the study. The researchers dropped a small number of individuals from the analysis after their data were collected because they were not sexually active.

Retention and Loss to Follow-Up (Cohort Studies Only)

Not applicable to this study

Sample Demographics

Age

Ages ranged from 18 to 26 years. The mean age was 21.9 at baseline in Lexington, 21.7 in Knoxville, and 21.8 overall.

Race or Ethnicity

	Lexington %	Knoxville %	Total %
White or Caucasian	80.7	86.9	83.8
Black or African American	15.9	10.0	12.9
Other/multiracial	3.4	3.1	3.2

Gender

	Lexington %	Knoxville %	Total %
Male	42.6	44.3	43.4
Female	57.4	55.7	56.6

Sexual Orientation

Not reported

Outcome Measures

High sensation seekers scored significantly higher on all sexual risk measurements than low sensation seekers. Individuals scoring low on the sensation-seeking and decision-making measure did not exhibit any developmental or campaign trends in either city for any of the dependent variables explored, including condom use, condom self-efficacy, and condom-use intentions.

In Lexington, the intervention city, high sensation seekers showed immediate and significant upward trends in condom self-efficacy, condom-use intentions, and condom use after the campaign started. For all three variables, there were significant downward trends after the campaign ended. The researchers determined that the final scores on the measures were higher than the levels projected by a regression estimate if the intervention had never occurred, although the intervention group actually scored lower on these variables at the end of the study than the comparison group. There were no significant changes across any of these variables in the Knoxville comparison group over the length of the study.

The researchers calculated the effect size for the Lexington condom-use campaign as Cohen’s d = 0.26 (slightly greater than a small effect), by comparing condom use 4 months before the intervention (M = 2.8) to condom use during the 3 months after the intervention (M = 3.03). Knoxville condom use decreased from pre-intervention to post-intervention.

The researchers also calculated the campaign impact by comparing the amount of unprotected intercourse in the past 30 days in the Lexington group 12 months after the campaign began to the estimated number had the intervention never occurred. They found that high sensation seekers engaged in 10.49 fewer occasions of unprotected intercourse during the 12 months after the campaign began than would have been expected if the pre-campaign pattern had continued. They estimated that 181,224 fewer unprotected sexual intercourse acts occurred among high sensation seekers between January 2003 and December 2003 because of the intervention.

Conclusions

As a likely result of the intervention, high sensation seeking young adults increased their condom use by an estimated 13 %, and 181,224 unprotected sexual intercourse acts were averted. Although the intervention appeared to be effective in increasing safer sex behaviors, the effects diminished soon after the public service announcements were taken off the air. The developers speculated that reinforcement messages in the form of booster campaigns would be necessary to maintain any benefits that the program conferred on the high-risk population. The developers also noted that combining a similar media campaign with other behavioral interventions would help sustain the effects of the intervention.

Implications and Lessons Learned

The intervention community was primarily Caucasian, and the researchers created and selected PSAs with predominately Caucasian actors, although some racial diversity was built into scenes with more characters. Racial diversity may be common in other areas interested in implementing a similar media campaign, and race may play a large role in determining whether an audience will relate to the characters portrayed in the public service announcements. In racially diverse communities, PSAs should be tailored to the specific racial profile of the community. The developers noted that especially for African Americans, it is critical that the actors in the PSAs are also African American, and the more the PSAs can be tailored to a specific population, the better (R. Zimmerman, personal communication, March 9, 2011).

Although the campaign showed efficacy in the short term, as soon as the PSAs were taken off the air, the target population steadily declined in all condom-use measures. The researchers noted two ways in which an intervention could be more effective for longer-term use. First, implementers could use a two-pronged approach: air cover and ground cover. The air cover would be provided by the PSAs, but the ground cover might involve face-to-face behavioral interventions or changes in the environment such as condom distribution or making condoms readily available and convenient. The second approach would be to keep presenting the PSAs on a consistent, regular basis with updated messages to make them new and relevant (R. Zimmerman, personal communication, March 9, 2011).

The researchers were aware that they were dealing with potentially sensitive subjects and encountered some resistance to the messages and content of the PSAs. They had to work with the television stations to figure out what PSAs were appropriate for audiences. They stressed, however, that they were able to show the PSAs successfully in two conservative cities, so others interested in the intervention should not hesitate to try a similar campaign (R. Zimmerman, personal communication, March 9, 2011).

Supplementary Materials Available

Additional References

Palmgreen, P., Noar, S. M., & Zimmerman, R. S. (in preparation). A mass media campaign to increase condom use among high sensation-seeking and impulsive decision-making young adults. In R. E. Rice & C. K. Atkin (Eds.), Public communication campaigns (4th Ed). Thousand Oaks, CA: Sage.

Romer, D., Sznitman, S., DiClemente, R., Salazar, L. F., Vanable, P. A., Carey, M. P., et al. (2009). Mass media as an HIV prevention strategy: Using culturally sensitive messages to reduce HIV-associated sexual behavior of at-risk African American youth. American Journal of Public Health, 99(12), 2150–2159.