The following data, provided by the Johns Hopkins Bloomberg School of Public Health Center for Communication Programs, describe the context of HIV/AIDS and sexual behaviour among youth in Zambia:

• Approximately 1 in 6 urban youth aged 15 to 19 is HIV-positive (Zambian Sentinel Surveillance, 1999)
• By the age of 15, 37% of boys and 27% of girls have had sex (Zambia Sexual Behaviour Survey, 1998)
• Among 15-19 year olds, 62% of the boys and 59% of the girls say they had had sex (Zambia Sexual Behaviour Survey, 1998)
• Of those who have had sex, 84% DID NOT use a condom the last time they had sex (Zambia Sexual Behaviour Survey, 1998)
• By the age of 19, only 16% of Zambian youth report that they have never had sex. (Zambia Sexual Behaviour Survey, 1998)
• 64% of girls and 70% of boys think they are at no risk of contracting HIV (Zambia Demographic and Health Survey, 1996)
• Among sexually active, never-married youth, 24% of boys and 13% of girls reported that they had more than one partner in the past year (Zambia Sexual Behaviour Survey, 1998)

In light of this situation, the Government of the Republic of Zambia (GRZ) asked USAID and its implementing partner, the Zambia Integrated Health Programme (ZIHP), to work with the Central Board of Health, the AIDS Council and Secretariat, the Johns Hopkins Bloomberg School of Public Health Center for Communication Programs, and young people themselves, to implement a mass media campaign to address youth and HIV/AIDS. As part of HEART - Helping Each Other Act Responsibly Together - a design team made up of 7 young Zambians designed 5 TV spots that included messages (e.g., how a girl can say "no" to sex) and that promoted regular condom use. Producers of radio spots and songs adapted the same messages and concepts to the rural context and translated them into local languages. Other materials such as posters, book stickers, exercise books, messages on buses, and music videos complemented the mass media.

The findings summarised below highlight two sets of data:

1. A 2001 evaluation of Phase One of the HEART Campaign, intended to help shape Phase Two, which began in the fall of 2000, and Phase Three, which started in the fall of 2001.
2. A 2003 evaluation of Phases Two and Three

With regard to the 2001 evaluation...

A quasi-experimental, separate sample pre- and post-test design was used. The pre-test survey was conducted from July to November 1999; it was designed to collect data on indicators related to households, health facilities (private and public) and district health management teams. The total sampling frame for the household-based component was the aggregate of all households in the 4 clusters, which was a total population of 2,683,982 living in 430,670 households.

The impact survey (post-test) sample comprised all young people ages 13-19, inclusive, living in households located within the 12 demonstration districts. The post-test questionnaire was modelled after the structured questionnaire used in the baseline survey. Sections about peer pressure, access to health facilities, and viewership of the HEART Campaign were added. The post-test questionnaire was field tested among 50 young men and 50 young women; adjustments to the survey instrument were made accordingly.

Data were field edited, cleaned, and entered using SPPS 6.0. Bivariate and multivariate analyses were conducted. All correlations were tested for significance. To examine responses on attitudinal statements, a Likert-type summative scale was used.

From the 2001 evaluation: Phase One of HEART reached over 50% of the intended audience; 71% of urban and 37% of rural youth saw one of more of the televised health communication spots. Young women (both urban and rural) were as likely as were young men to have seen some or all of the TV adverts. Between 60 and 90% of viewers spontaneously identified the correct message with any given advert.

From the 2003 evaluation: About 64% of urban 2003 respondents reported that they had seen all or some of the television spots, while approximately 14% of rural respondents saw one or more.

From the 2001 evaluation: Approximately 74% of male viewers and 68% of female viewers said the health communication spots prompted them to talk with others about the advert, decide to abstain from sex until more mature, or use a condom. On average, male as well as female viewers discussed abstinence or safer sex with a significantly wider range of people than did their counterparts.

From the 2001 evaluation: Overall, viewers were more knowledgeable about HIV/AIDS than were nonviewers. Nearly 86% of viewers of two of the adverts (Mutale & Ing'utu) compared with 72% of nonviewers recognised that a person who looks healthy can be HIV-positive.

From the 2003 evaluation: Viewers were significantly more likely than non-viewers to mention abstinence as a way to avoid transmission in both survey years. In 2003, viewers compared with non-viewers were more likely to mention condom use and having only one partner as ways to avoid HIV.

From the 2001 evaluation: Among women who are sexually experienced, 82% of viewers contrasted with 69% of baseline and 64% of impact nonviewers reported they feel confident that they have "the ability to say no to unwanted sex". Among both men and women, the perceived efficacy to use condoms was positively and significantly correlated with viewership.

From the 2001 evaluation: Viewers were 1.68 times more likely to report primary or secondary abstinence then were non-viewers (using logistic regression and holding sex, age, residence and education constant). There was a dose effect: The more health communication spots recalled, the greater the likelihood that the respondent is abstinent. Viewers were 1.91 times more likely to have ever used a condom and 1.63 times more likely to report condom use during last sex when contrasted with nonviewers (using logistic regression and holding sex, age, residence and education constant). Older, better-educated respondents were more likely to use condoms. Women were more likely to report condom use than were men when background characteristics were held constant.

From the 2003 evaluation:

• Approximately 53% of viewers in the 2003 survey reported that they took at least one action as a result of having seen the campaign.
• Overall, respondents were more likely to say they chose "to abstain" than to report that they decided to use a condom as a result of seeing the spots in both survey years.
• Viewers were significantly more likely than non-viewers to report primary or secondary abstinence. Holding background variables constant, the odds that respondents were abstinent was 1.4 times greater for viewers compared with non-viewers in 2000 and 1.7 times greater in 2003.