Research Spotlight: Understanding Anthrax Risk Where Farming, Wildlife, and Livelihoods Meet

That is the human setting behind a recent qualitative study in PLOS Neglected Tropical Diseases, where Chisoni Mumba and colleagues explored how farming communities living near game management areas understand anthrax, experience its impacts, and navigate the risks created by wildlife contact and the game meat trade.

We spoke with one of the authors, Chisoni Mumba of the Department of Disease Control, School of Veterinary Medicine, University of Zambia, to learn more about anthrax risk, community knowledge, and responsible AI assisted qualitative analysis.

Anthrax sits at the intersection of public health, veterinary medicine, conservation, livelihoods, and food security. In the study settings, communities depend on livestock and nearby ecosystems, yet these same connections can create exposure pathways when infected animals, contaminated meat, or unsafe carcass handling enter everyday life.

The research matters because it shows why disease control cannot rely only on biomedical messaging. Effective prevention must also account for trust, cultural knowledge, poverty, food access, conservation governance, and the practical realities that shape decisions about meat and livestock.

Professor Mumba reflected on the broader significance of the study: “What this study shows is that you cannot separate anthrax control from how people actually live around these parks. Community knowledge, cultural belief, poverty, and the game meat trade are all part of the transmission picture, not side issues. If veterinary services and public health teams keep treating this as a purely biomedical problem, they will keep missing the reasons outbreaks persist. The real implication is that response has to be multi-sectoral from the start, combining surveillance, veterinary care, and education with an honest look at the economic pressures driving risky behaviour.”

Across the focus group discussions, the authors show that anthrax is experienced as a social, economic, ecological, and health problem at once. Three contributions stand out.

Community knowledge was uneven. Some participants recognized anthrax as dangerous and linked it to infected animals, while others said they did not know the signs, as one Mumbwa participant explained, “This disease, we do not know about it, not even the sign.” The evidence points to the need for targeted and locally grounded education.

Wildlife contact and the game meat trade were perceived as important risk pathways. Participants described concerns about buffaloes, hippos, uncertified meat, and meat moved through informal networks, which helped the authors connect disease risk with conservation spaces, market incentives, and livelihood pressures.

The impacts went beyond infection. Participants described livestock losses, fear around meat consumption, household strain, and mistrust of formal campaigns, including one account of losing 14 adult cattle during the outbreak peak. These accounts show why anthrax prevention needs to address both health risk and economic vulnerability.

Asked what surprised him most, Professor Mumba noted: “What struck us most was how uneven awareness was, even within the same district. Some participants could describe anthrax symptoms in detail, while others, particularly in Mumbwa, told us plainly that they did not know the signs at all and were only just hearing about the disease through phones and radio. We also noticed a clear pattern by gender: men tended to focus on how the disease spreads through wildlife and livestock, while women spoke more about its severity and the toll it takes on the household. That split wasn’t something we expected going in, and it has real implications for how education campaigns should be designed.”

The authors used a qualitative research design grounded in an interpretivist paradigm. They conducted six focus group discussions in Simalaha Community Conservancy, Lochinvar, and Blue Lagoon National Parks, with participants selected through community gatekeepers to include people whose livelihoods, beliefs, or practices intersected with anthrax risk, including livestock farmers, game meat traders, and community leaders.

The data were complex because discussions covered sensitive illegal trade, local languages, cultural beliefs, gendered experiences, disease knowledge, and livelihood pressure. Audio recordings were transcribed and translated into English, then analyzed through reflexive thematic analysis, AI assisted analysis, manual integration, theme validation, and final synthesis, with the authors grounding conclusions in participant quotations.

The paper states that Evidano was used to conduct sentiment analysis on the focus group transcripts, helping the team capture emotional tone and subjective experience in participant narratives.

According to the article, Evidano identified anxiety and fear in discussions about consuming infected meat and economic losses from livestock deaths. It also detected frustration and mistrust toward formal health campaigns and authorities, which added depth to the theme on the impact of anthrax on communities.

The article is careful to show that AI tools supported, rather than replaced, scholarly analysis. The authors manually integrated AI generated outputs, compared themes with direct quotations, reviewed themes across researchers, and relied on human judgment for final synthesis. Professor Mumba also noted how AI tools can support careful research workflows when used responsibly: “Evidano gave us a layer of insight we would have struggled to capture through manual coding alone. It picked up the emotional tone running through the transcripts, the anxiety around eating meat that might be infected, the fear tied to livestock losses, and the frustration and mistrust participants expressed toward formal health campaigns. That helped deepen our theme on the impact of anthrax on communities. But I would caution other researchers against treating AI output as a finished result. Every theme Evidano surfaced was checked against the original transcripts and verbatim quotes, and reviewed independently by the research team. The AI is a tool for pattern recognition, not a replacement for the researcher’s own judgement about what the data actually means.”

For public health and veterinary teams, the article suggests that anthrax response should be grounded in community experience. Messages about safe meat handling and carcass disposal are more likely to resonate when they acknowledge local beliefs, economic realities, and the reasons people may continue risky practices despite knowing some of the dangers.

For researchers, the study offers a practical example of methodological pluralism in qualitative work. AI can help identify patterns, emotional tone, and relationships across transcripts, but the article shows that credibility depends on researcher reflexivity, careful validation, and close attention to participant voices.

For policy and conservation leaders, the takeaway is that disease control, livelihood resilience, and wildlife governance are deeply connected. Professor Mumba offered this final reflection for readers: “The takeaway is that you cannot enforce your way out of this problem. Communities turn to the illegal game meat trade because of poverty and limited access to legal alternatives, not because they don’t understand the risk. If policymakers want to reduce anthrax transmission, they need to invest in surveillance and veterinary services, yes, but also in the livelihoods that make people reliant on unsafe practices in the first place. Community partners and researchers both have a role in making sure interventions are grounded in what people are actually experiencing on the ground, not just what looks good on paper.”

Together, the article argues for One Health interventions that combine education, surveillance, veterinary services, policy enforcement, and sustainable livelihood support.