Human Papillomavirus Vaccine Introduction in South Africa: Implementation Lessons From an Evaluation of the National School-Based Vaccination Campaign

The incidence of cervical cancer in southern, central, and east Africa is among the highest in the world. The World Health Organization (WHO) recommends vaccinating girls against the sexually transmitted virus responsible for 99% of cases of cervical cancer, human papillomavirus (HPV), prior to sexual debut (9 to 13 years). Thus, in April 2014, a national school-based HPV vaccination programme was rolled out in South Africa, designed to reach Grade 4 girls aged 9 years and above. At the request of the National Department of Health (NDoH), the researchers conducted an external assessment of the first-dose phase of the vaccination programme to evaluate programme coverage and vaccine safety and identify factors that influenced implementation. The aim of this article is to illustrate what implementation challenges were experienced introducing a new vaccine to a new intended population, outside of the traditional clinic environment, and to offer lessons for HPV vaccine programming - not only in South Africa but also in similar settings elsewhere.

In South Africa, early feasibility and acceptability studies and demonstration projects identified several potential areas of concern for implementing school-based HPV vaccination programmes. For example, in the context of a broader discourse about "sexual risk", the HPV vaccine has become vulnerable to lobbying by diverse anti-vaccination and vaccine-hesitant advocacy groups. To preempt possible opposition to HPV vaccination, policy experts advised a strategy of marketing the vaccine as preventing cervical cancer rather than a sexually transmitted infection (STI). But in South Africa, visibility of cervical cancer is low, and there is little knowledge about the impact of cervical cancer on female morbidity and mortality. The danger, then, was that parents would regard HPV vaccination as "non-essential", leading to poor uptake.

Fortunately, research to discover how receptive the public would be to vaccine messaging, undertaken prior to 2014, found strong support for HPV vaccination of young people among policy and health service representatives, parents, youth, and educators. This may be due to the fact that, in general, vaccines are widely accepted in South Africa.

Prior to campaign initiation, among other activities, the national Ministers of Health and Basic Education jointly convened meetings with school governing bodies, school principal organisations, and teacher unions at national level to explain the HPV vaccination campaign and secure agreement from them to proceed. Social mobilisation efforts involved the development of school-specific informed consent packages that included consent information, education, and communication (IEC) materials, such as posters, fact sheets, frequently asked questions, and a guide for educators. Information about the campaign was placed on government websites and social media networks and relayed through broadcasts on national radio. The Health Minister's official launch of the campaign received wide television exposure on the national broadcaster's "Morning Live" breakfast show.

Training materials - developed by the NDoH with the support of partners - included a field guide and a set of training slides. A 2-day training session was held at national level, and an additional 1-day training session was conducted for provincial, district, subdistrict, and facility-level teams. A bivalent HPV vaccine with a 2-dose (6 months apart) schedule was used. On vaccination days, DoH vaccination teams visited assigned schools and implemented set procedures involving education, eligibility control, vaccination, data recording, and observation of vaccinated girls. "Mop-up" visits were made where necessary to reach eligible girls who had been absent on the day of vaccination.

The researchers based their cross-sectional and mixed-methods approach on a process evaluation framework, which included: a review of key planning and implementation documents and monitoring data; observation at vaccination sites; key informant interviews (N=34); and an assessment of media coverage and content related to the campaign. The main outcomes of interest in the evaluation were programme coverage, vaccine safety, and factors that influence implementation of the programme.

In short, there was overall success in key measures of coverage and safety. Over 350,000 Grade 4 girls were vaccinated in more than 16,000 public schools across South Africa, which translated to 94.6% of schools reached and 86.6% of age-eligible learners vaccinated. No major adverse events following immunisation (AEFI) were detected.

The researchers attributed the campaign's successes to careful planning and coordination and strong leadership from the NDoH. The campaign was introduced in the context of a high-level political mandate, and interview data indicated that strong political commitment to the campaign was an important factor driving results. reportedly, campaign planning and coordination was managed centrally by a team of highly experienced, committed NDoH staff who established strong communication mechanisms at provincial and district levels to monitor progress and address challenges. The primary challenges identified were related to vulnerabilities in cold chain capacity, onsite management of minor adverse events, and obtaining informed consent. With regard to the latter, the wording of the HPV campaign informed consent forms - "I hereby grant/do not grant permission for my child to receive 2 doses of the HPV vaccination" - confused many parents, who believed the form referred to a social grant from the government. Parents were also reported to have been confused by the rollout of the National Contraception and Fertility Planning Policy and Service Delivery Guidelines and, in particular, the launch of contraceptive implants, which took place around the same time as the start of the HPV vaccination campaign. According to key informants, this confusion was responsible for some parents declining consent for vaccination.

Analysis of media coverage found that a total of 373 items on HPV vaccination were published or broadcast in the period March 1 to April 30 2014, the majority (68%) online, with just under a third (28%) in print media and only 4% in broadcast media (radio and television). Over half (55%) of all media items were categorised as neutral, with 38% considered positive and only 7% designated as negative. Of the positive media items, most (70%) were released in March (the first month of the campaign), while 59% of negative coverage was released the following month, suggesting that after the initial time period, a shift in public discourse about the campaign may have occurred. Just over half (51.9%) of the negative media appeared in print, compared with only 18.5% in broadcast media and 29.6% online. In terms of content, of the 27 negative items, a majority (63%) related to parental concerns over vaccine safety, while the remainder either highlighted the high cost of HPV vaccine in the private sector (22%) or were critical of the campaign's exclusion of boys (15%). Although social media - Facebook, email, and short messaging service (SMS) used on cell phones, among others - was not covered in the media assessment, anecdotal information suggests that anti-vaccine messaging disseminated through social media may have posed an important threat to the success of the campaign. While campaign planners anticipated and prepared for some negative media coverage, they did not expect the use of social media for spreading misinformation about HPV vaccination.

Expanding on the role of social media, the researchers observe that, as in many parts of the world, South African parents deciding about HPV vaccination are increasingly likely to search the internet for information about the vaccine; the challenge is that it is often difficult to assess the credibility of these sources. Furthermore, the user-generated content - a key feature of social media - encourages lay persons to engage with medical knowledge, selecting or rejecting information based on what they already believe to be true. This type of content can mobilise those who already have low levels of trust in conventional biomedicine. As the HPV vaccination programme in South Africa matures, the researchers say that it will be important to monitor the influence of Internet-based anti-vaccination groups and social media conversations on local attitudes toward the vaccine.

Six implementation lessons that emerged from the findings of this evaluation are outlined in the report. For example, one lesson relates to the management of adverse events and the role of social mobilisation more broadly. Even a minor AEFI that is not adequately managed has the potential to deter concerned parents from consenting to vaccination, reduce second-dose uptake, or be influenced by anti-vaccination groups. Uptake of the HPV vaccine could be significantly impacted by such a risk, unless it is countered by a sustained, proactive approach to tackle misinformation in a variety of online spaces, including social media. The researchers recommend that a media rapid response plan, prepared prior to implementation, be part of the management of AEFIs. This plan should include public health messaging that conveys complex vaccination information in simple, accessible language and be flexible enough to be activated at any level to respond in real-time to negative media coverage.

"Beyond such targeted responses to an emergency, effective social mobilization should be approached as a long-term investment for building community support....Given the profound changes in the media landscape in the past decade alone, it is time for program designers to explore innovative methods not normally used in public health communication. Possibilities include using the personal narrative format that anti-vaccination groups have appropriated so successfully, and greater use of digital applications that encourage users to interact directly with material, by sharing, commenting on, or uploading content."

In conclusion, the evaluation found that the first phase of the national school-based HPV vaccination campaign was successfully implemented at scale in this setting. Future implementation will require improved partnerships between government ministries, simplified informed consent, and closer monitoring of social media messaging.