Beyond Awareness Response to Nancy Coulson

Nancy Coulson's paper provides a starting point in highlighting the need to understand and reflect upon HIV/AIDS communication campaigns in South Africa. In this response, I highlight the need to develop a more coherent framework for understanding such campaigns, and then discuss more specifically aspects of her points raised in relation to the Beyond Awareness Campaign (BAC). Whilst critical comments are made, I would like to point out that there is a need to understand that research is never a finite process, but rather one of exploration and "truth-seeking" where conclusions are often temporary. In relation to communication campaigns within the context of a rapidly changing epidemic there is a need to foster continued research, reflection and analysis.

Mass media vs communication

As a point of entry, Coulson refers to four campaigns – the Beyond Awareness Campaign, Soul City, loveLife and a Government Communication and Information System (GCIS) campaign as mass media campaigns.

Of the four campaigns mentioned, the GCIS campaign was the only one totally located within mass media as a delivery system – in this case utilising broadcast and print media. The other campaigns all incorporate a complex layering of communication activities that extend beyond conventional use of mass media. Soul City's communication activities include, for example, booklets and other small media, utility items, training manuals, outreach activities, a telephone helpline, public relations and advocacy; loveLife's communication activities include youth magazines, small media items, utility items, youth activities, youth facilities, telephone helplines, and public relations; Beyond Awareness communication activities included leaflets, posters, manuals, research documents, training manuals, utility items (all of which were distributed on demand via a media clearinghouse), developmental activities involving murals and the AIDS Memorial Quilt, tertiary institution HIV/AIDS activities, promotion of HIV/AIDS reporting and photography, various training activities, a living openly project, and the expansion of a telephone helpline service.

These HIV/AIDS campaigns are thus located far more consciously and fundamentally in a contextual communication framework, with mass media components being given greater or lesser emphasis within the context of the overall programme. Although Coulson attempts to locate understanding of the campaigns within the Ottawa Charter, this is not done rigorously enough, and we are left with both a poor general description of the campaigns, as well as inconsistencies in analysis. At times Coulson's analysis also tends to conflate understanding of the part (mass media), with the whole (multilayered communication activities).

In framing mass media Coulson makes brief reference to media reach – suggesting, for example, that 99% of people have access to radio, 75% have access to television, and 7% read newspapers.[1] It is important to explore this aspect more deeply, and to understand socio-economic and rural/urban differentials in terms of media access. For example, a study of six contexts including urban, peri-urban and rural found ranges between sites as follows: Television (45-98%); Radio (80-99%); Daily newspaper (10-79%); Sunday newspaper (2-86%), and Magazines (11-90%).[2] What this demonstrates is severe disparities in media access between urban and rural settings. A more recent study of commuters found media access to be Television (74%); Radio (92%); Newspapers (60%); and Magazines 49%.[3]

Of further relevance is the need to discuss audience segmentation in terms of, for example, age, race, socio-economic status, culture and language. South African society is markedly heterogenous, and there is a need to address the fact that mass media interventions tend towards homogenising messages and audiences. In addition, few campaigns have the capacity or budgets to address audiences in the eleven official South African languages, and key messages are often related only in English, or in a selection of mutually intelligible languages.[4]

The Beyond Awareness Campaign

The concept of communicating beyond HIV/AIDS awareness was a conscious attempt to move past the primacy of mass media as the communication mechanism of choice in national campaigns. The theoretical underpinnings of this approach were grounded in several years of consultative activity amongst a group of HIV/AIDS communicators, government and non-governmental organisation representatives who met regularly within a formally constituted National Communications Forum (NCF). This group, along with a small group of academics and advisors framed the broad terms of reference for the two Beyond Awareness Campaign tenders that were issued by the Department of Health in 1997 and 1998.[5] Year on year budgets for the campaigns were R7-million, R12-million and R14-million respectively.[6]

The first phase of the campaign involved a range of developmental and communication activities that paved the way for the second phase. The first phase included considerable research into the Beyond Awareness concept which was articulated in "Communicating Beyond Awareness: A manual for South Africa."[7] The approach and its underpinnings were widely disseminated via a printed manual and workshops with HIV/AIDS communicators in all provinces in 1998/9.

Beyond Awareness builds on recognising that there are a multiplicity of HIV/AIDS communication activities nationally including those that occur via other national government departments, provincial and local government structures, non-governmental and community-based organisations and workplaces, amongst others. These activities in turn employ a range of communication strategies including mass media, small media items (leaflets, posters, etc) as well as interactive activities. The intention of the BAC approach is to provide a framework for integrating these.

Within the various NCF discussions it was clear that a number of issues had consistently not been addressed. For example, there was a strong demand for a standardised set of leaflets in all South African languages, as well as other small media items. Up to that point there was a disparate set of materials, poor cost efficiency in materials production, an almost total lack of materials in more than one or two languages, and no coherent distribution system. This activity was a core project of BAC I, and an expanded set of materials totaling over 80 separate items was produced during BAC II. These items were all made available via a central clearinghouse at no cost to any individual or organisation requiring them.[8]

I will not go into details of the various aspects of BAC II in detail here, save to say that the Beyond Awareness concept is most simply described as a process of developing and disseminating "tools for action" amongst the wide range of organisations working in the field. Examples of products, as well as reports and overviews of the campaign are available by clicking here. This includes posters presented at the international AIDS Conference in Durban, 2000.

The remainder of this article responds to issues raised by Coulson in relation to mass media elements of BAC II.

BAC mass media components

Beyond Awareness did not document any pre-testing of their messages neither did they specifically evaluate recall of messages in their sentinel site research. The mass media component of the Beyond Awareness campaign that promoted specific messages spent approximately 34% of the total budget over two years. (Coulson, Page 4)

Coulson appears to have misunderstood the nature and purpose of the sentinel site research. In essence, the premise was that most campaigns study their impact via specific reference to themselves, and there was a need to come up with a generic frame of reference to understand what the general communication and behavioural environment was about – media access, differentials between rural and urban contexts, behavioural risk factors, amongst others. It needs to be understood that South African campaigns do not operate in a well-resourced support environment, and an understanding of the enabling context becomes critical to any intervention. For example, one can successfully raise awareness around condoms, but condom use is dependent on a range of factors including whether a condom is available or not as well as critical issues such as condom quality, non-judgemental attitudes in relation to condom distribution, appropriate instructions and a capacity to negotiate and entrench condom use within relationships. We were more interested in these contextual factors than demonstrating impacts of our mass media interventions (which we had other ways of measuring). Rather, the research was critical to the understanding of potential impacts of other national and local campaigns, as well as towards guiding future content. Important findings were, for example, made in relation to proportions and levels of sexual activity amongst youth, sexual debut, age differentials between partners, sexual coercion, secondary abstinence, condom use and the like. Surprisingly, very little attention has subsequently been given to these issues.

It is somewhat disingenuous to frame the budget for mass media as a percentage – the year on year total budget of the campaign was R12-million and R14-million, with only R4-million per annum budgeted for mass media. In year two, only R2-million was committed to mass media activities, because the complex approval process in the Department of Health delayed the second phase. R6-million (approximately US$600 000 at current exchange rates) is a very small amount of money with which to undertake mass media activities and a careful strategy was required.[9]

BAC was a short term campaign, and the limited budget allowed for only a few key messages. These were

• show you care, wear a red ribbon;
• HIV/AIDS information is available from the tollfree AIDS Helpline
• get wise, condomise.

Although additional themes were planned, these could not be implemented because of delays in the Department of Health approval systems.

We knew we were going to be running hard and fast right through the end of the campaign in December 2000, and that it would have been impractical and costly to attempt a complex impact analysis of these messages. Instead, our approach was quite simple – monitor responses via resource delivery systems (AIDS Action Office, AIDS Helpline, DOH condom distribution).

For example, the red ribbon was introduced in South Africa as the primary logo associated with HIV/AIDS communication by the Department of Health in 1995, when the main imperative was to shift away from the "yellow hand" towards an international symbol. The red ribbon was integrated into all BAC media in 1998, as was the tollfree AIDS Helpine number. We also encouraged government departments, NGOs and other organisations to use and promote the combined red ribbon AIDS Helpline logo, and to this end made the logo available via the campaign website.

We had a simple monitoring system in place for both the red ribbon and Helpline. In the case of red ribbon promotion, this was supported by distribution of red ribbon lapel pins via the AIDS Action Office. Over 600 000 red ribbon badges were produced and disseminated during the campaign, with demand exceeding supply throughout the campaign period. The red ribbon was positioned as a symbol of "caring about AIDS", although it was intended that no specific interpretation be associated with the act of wearing a red ribbon. Enabling the simple first step of wearing a red ribbon allows for individual interpretations of the meaning of that act, whilst at the same time, normalising an individual's association with the disease. The manufacture of red ribbons lapel pins became something of a "cottage industry" in South Africa, with local beadworkers producing beaded versions, and many organisations manufacturing their own.

Research has shown that it is relatively easy to generate awareness and basic understanding of HIV/AIDS, but many people have very specific individual and contextual questions and misunderstandings, whilst some may require specialised counselling or referral. A telephone helpline is an ideal system to respond to these needs because it allows for dialogue.

The Helpline component was managed by a combination of monitoring and developmental activities. In its initial form, the AIDS Helpline was hugely under-resourced in terms of technical and personnel capacity, and there was no capacity to respond uniformly or appropriately in terms of information, nor in terms of language.

The impact of advertising bursts on helpline call rates was closely monitored and there was a direct causal relationship. The problem however, lay in the fact that the helpline had insufficient capacity to deal with the calls the campaign was generating. This required a developmental approach that included locating a consultant at the helpline to monitor various aspects and to develop the systems necessary for a more efficient service. Drawing on other partnerships, for example Johns Hopkins University Center for Communications Programs, providing additional resources and expertise via BAC, and working closely with the implementing agency, Life Line, a sophisticated, capable multilingual helpline service was developed.[10]

With regard to condoms: We were in regular contact with the Department of Health logistical system, which reported near constant under-supply of condoms – a solid indication of demand exceeding the capacity of the system. Note that the intention of the "condomwise" slogan and campaign was simply to provide a level of backdrop support to other widespread condom promotion activities, and the impacts on condom demand cannot be reduced to a single campaign.

Overall, monitoring of response in relation to key messages provided a sound and sufficient understanding of mass media impacts – and the need to separately and specifically evaluate recall, as Coulson suggests, is questionable.

In reflection, the results of a November 2001 survey of public transport commuters conducted under the auspices of the Department of Health make for interesting reading. Findings included:

• Reminded of AIDS in the past month - 55% red ribbon, 51% leaflet, 54% poster
• Know a symbol of AIDS - 83%
• Unprompted recall of red ribbon symbol as % of above - 93%
• Heard of AIDS Helpline - 73% (unprompted identification of Helpline - 56% AIDS Helpline)
• Used Helpline - 11%
• Message most recalled (unprompted) – "condomise/use a condom"
• Condom use last - 45%, and 66% in 15-24 years
• Worn a red ribbon or AIDS related clothing - 48%
• Would like to - 83%
• Interested in obtaining leaflets - 93%

It was also interesting to note that a 2002 survey by Markinor of South Africa's top brands included an exercise to test recognition of the red ribbon. Overall recognition and association with HIV/AIDS "was put at 90%, making it one of the most recognised brands in the country". (Sunday Times Top Brands, 29 September 2002. Metropolitan areas, 96% and non metropolitan areas 87%)

These responses are clearly in line with BAC objectives, and at a fundamental level, if one subtracted BAC from the equation, the numbers would be very different. Given that BAC campaign elements were diffused to HIV/AIDS organisations and interventions as a whole, positioning interpretation of such data within a monocausal frame of reference however, is not something one should be inclined to do.

Another issue Coulson raises is that "pre-testing" of messages was not documented. Quite frankly, all of the campaigns described by Coulson would be hard pressed to provide documentation of "pre-testing" in relation to all of their multilayered communication activities. There is a disturbing "monotheism" that seems to exist around pre-testing that runs counter to a number of alternative methodologies for message development – for example participatory development (see Parker W (1997) Action Media: Consultation, Collaboration and Empowerment in Health Promotion, African Media Review, 11(1), 54-63) expert review and reviews of research findings.

Interestingly, most advertising agencies make very little use of pre-tesing, relying on a combination of creative development, expert review and client support. Measuring impact is quite directly related to monitoring sales and brand awareness – ie. Testing impacts on the fly "in the field" so to speak. We followed the exact same approach in terms of monitoring red ribbon, Helpline and condom demand without pre-testing. We did however include in the development team, individuals who had worked in AIDS communication for more than a decade and were able to draw on their experience, members of the NCF who were able to provide comment and expert review, and reviews of qualitative research, which for example, had shown widespread use of the term "condomise", and the integration of experienced advertising "creatives" into message development.

The Helpline, red ribbon, and condoms were promoted via radio messages in all South African languages with advertisements involving two or three sub-messages, as this was the most cost-effective medium to achieve multilingual reach. These were supported by magazine advertisements, outdoor advertising primarily on taxis, and posters and other items disseminated via the AIDS Action Office, with further support being provided via other organisations.

Finally, I must also take issue with the following comment by Coulson:

There is now widespread understanding that formative and evaluative research is essential. Johns Hopkins recommends that 10% of a campaign should be spent on research and evaluation. In the independent appraisal of BAC II, commissioned by the Department of Health, it was concluded that although the campaign was found to be generally competent it was criticised for it's applications of some of these steps. Only 2.5% of the BAC II budget was spent on research and evaluation and it was felt that the campaign had failed to adequately address the following:

• The development of a theoretical model (as required by Step 2).
• To provide a clear definition and segmentation of the target audience. The precise target audience for the Beyond Awareness Campaign changed between various activities and publications.
• To provide evidence of pre-testing of campaign messages, image and branding including the "red ribbon" campaign (as required by Step 3).

(Coulson, page 7)

Research, monitoring and evaluation involve wideranging methodologies and varying costs. At issue is not how much research and evaluation costs, but rather examining the relative sophistication, cost-effectiveness and appropriateness of the approaches used. This cannot simply be reduced to percentages, and it seems somewhat naïve to attempt to critique BAC II on the basis that its research costs didn't hit the right percentage cost. It's worth pointing out that researchers at Johns Hopkins CCP suggest this as a guideline to emphasise the importance of research – rather than a cost proportion that should slavishly be followed.

Similarly, the steps she refers to above, are defined by the Johns Hopkins "P" process, and it is implied that this is the most appropriate and "absolute" framework for understanding BAC II (and I don't think JHU intended it that way either). It is clear that Coulson has a very poor understanding of the campaign – and it is unfortunate that her source of information is referenced only as an "appraisal" that she herself authored. My comments further above have described processes followed, but some direct comments are warranted here.

Firstly, a very closely considered theoretical framework was developed in the form of the manual "Communicating Beyond Awareness", the genesis and application of which is described above.

Secondly, the broad specification of the tender was to target youth. Contrary to Coulson's suggestion of a "changing audience", we "segmented the market" through a wide range of activities, and all activities gave consideration to the primary target audience. In a heterogenous society it is absolutely critical to note that "precise target audiences" do not exist, and audiences are accessed and nuanced through varied interventions, and with different objectives in mind.

Thirdly, I have made some reference to the need to think outside of the box when it comes to pre-testing, and this is probably worthy of a longer debate that does not belong within this response.

Conclusion

With reference to Coulson's conclusions and recommendations:

The ACT consortium were contracted in 2001 to conduct the Department of Health's HIV/AIDS communication campaign. It would be worth reviewing the campaign's activities and achievements. Similarly, there needs to be continued development of understanding of Soul City, loveLife and other national campaigns. However, it is also worth considering the development of understanding of provincial and local level activities that include varied communication approaches. These are seldom described, reviewed or evaluated, yet form the greater mass of HIV/AIDS communication activities in South Africa.

2 (Kelly K (2000) Communicating for Action: A contextual evaluation of youth responses to HIV/AIDS, Beyond Awareness Campaign, Department of Health. Available by clicking here.)

3 (Parker W et al (2002) On the move: the response of public transport commuters to HIV/AIDS in South Africa, Department of Health. Available by clicking here.)

4 (See Moss G and Segal K (1998) Language guidelines for HIV/AIDS communication in South Africa, Beyond Awareness Campaign, Department of Health. Available by clicking here.)

5 (Note: Coulson, Page 1, makes reference to the dates as being 1995 and 1997, suggesting that these were situated to replace the "yellow hand" campaigns. The yellow hand campaigns were commissioned in the early 1990s and were phased out prior to the transition to the 1994 post-apartheid government.)

6 (Note: The one-year Beyond Awareness I Campaign was awarded to a consortium of seven organisations including NGOs, consultancies and commercial companies. The Beyond Awareness II Campaign was awarded to a consortium of four organisations including two commercial companies, a consultancy and an NGO. I was involved in the management of both campaigns, and was director of the BA II Campaign.)

7 (Parker W, Dalrymple L and Durden E (1st Ed 1998, 2nd Ed 2000). The manual is available in Adobe Acrobat format by clicking here).

8 (Note: There is some complexity in developing, producing and managing a system of this nature. It was extremely difficult to balance an almost infinite demand for materials within a finite budget. Over the period of the campaign some 30-million items were distributed on request to over 2 000 organisations. A description and evaluation of this system and other activities is available by clicking here.)

9 (Note: loveLife, for example, expends approximately two and a half times that amount per annum on billboards alone).

10 (Information related to this service, as well as findings from monitoring reports can be found at clicking here)