How to make noncoherent problems more productive: Towards an AMR management plan for low resource livestock sectors

The study investigates how global antimicrobial resistance (AMR) policies to reduce antibiotic use (ABU) intersect with on-the-ground realities of livestock breeding in low-resource settings, focusing on Benin and Burkina Faso. Motivated by the mismatch between standardized guidelines and local constraints, the research asks how antibiotics function within complex socio-material ecologies of production and how reduction can occur without jeopardizing livelihoods, food security, or animal welfare. The authors frame antibiotics as infrastructural “quick fixes” that manage uncertainty and precarity where diagnostic capacity, veterinary extension, biosecurity, and financial safety nets are limited. Using the concept of noncoherence, they explore how heterogeneous practices and institutional logics come together productively yet imperfectly, and identify context-specific opportunities to safely reduce ABU.

The paper reviews the WHO’s 2015 Global Action Plan (GAP) for AMR and the 2017 guidelines for reducing antibiotic use in food animals, which emphasize awareness, surveillance, infection control, use optimization, and investment in new treatments and diagnostics. Livestock sectors, especially in LMICs, are key targets due to high ABU and weak regulatory and monitoring infrastructures. A dominant ‘knowledge deficit’ framing posits that education will correct misuse. Social science critiques argue this is insufficient, highlighting antibiotics’ embedding in organizational cultures, expectations of care, and food production processes. Particularly in low-income contexts, antibiotics remain integral to poverty reduction and productivity; removing them without replacing their functions risks harm. Theoretical grounding draws on STS concepts of antibiotics as infrastructure (Denyer Willis and Chandler) and Law et al.’s ‘noncoherence’—syncretic modes where differing practices and logics co-exist (denial, domestication, separation, care, conflict, collapse). The authors argue the GAP’s push for harmonization ‘domesticates’ heterogeneity, obscuring conflicts and implementation gaps. They build on literature indicating that northern management models often do not suit more fragmented, less structured breeding networks in the Global South, necessitating context-responsive strategies.

The study is part of the AMRIWA consortium (interdisciplinary team of microbiologists and social scientists in Europe and West Africa). Fieldwork in 2019–2020 applied qualitative and ethnographic methods: nine focus groups with veterinarians and breeders; interviews with seven breeders, three non-governmental training agencies, one veterinary scholar, three vet pharmacists, one ministry official, three government para-veterinarians, and 12 ministry officials/local representatives of international organizations implementing the GAP. The team also conducted three farm observations (including a 3-week stay on a Beninese layer poultry farm and 5-day visits to others). Data (recorded or handwritten) were transcribed in French, digitally translated and checked, coded and analyzed in Atlas.ti. Analysis proceeded inductively to identify themes around antibiotic practices and AMR, then mapped relationships among organizational logics and practices to understand noncoherence within production ecologies. The authors ultimately focused on three overlapping noncoherence styles—implementation, risk, and production. Participants varied by country: Beninese breeders tended to be young men running small peri-urban farms with little formal training; Burkinabe participants were more often entrepreneurs running larger agribusinesses with training in management/marketing and hired labor. This variance enabled comparison of perspectives but limited detailed cross-country sectoral comparison.

- Three overlapping noncoherence styles structure antibiotic decision-making: implementation, risk, and production. - Implementation noncoherence: Ministries lack resources to register and support numerous small and mid-sized farms, hindering GAP objectives (awareness and surveillance). In Benin, a modest registration fee (~10,000 CFA ≈ €15) was offset by high loan interest, required use of expensive state-approved services, and informal ‘taxes’; some applications went unanswered and many producers were unaware of requirements. In Burkina Faso, too few staff could monitor too many farms. Farmers thus rely on private training (often HACCP-aligned) and digital/vernacular networks (WhatsApp, Facebook, phone) for advice. Laboratory capacity is minimal (Benin: two veterinary reference labs; Burkina Faso: one), stalling national AMR surveillance programs. - Risk noncoherence: Evidence-based diagnostics are impractical: antibiograms are costly and slow; breeders fear losses if they delay treatment (e.g., losing a 200 kg bull). Calling vets carries biosecurity risks (possible cross-farm contamination), and official limits of two farm visits/day are financially unfeasible for some vets. Breeders often avoid vet callout fees (≈2000 CFA per visit or ~€20/month) and self-diagnose/treat, or opt for remote mobile consultations. Vets who avoid entering farms during suspected viral outbreaks to reduce contamination risk may be criticized by clients. These competing strategies create tensions between diagnostic ideals and biosecurity/economic realities, potentially increasing unauthorized ABU. - Production noncoherence: Intensive systems using ‘exotic’ breeds require robust biosecurity and inputs that many peri-urban/backyard farms cannot afford. Even professionalized farms face thin margins and high exposure to disease and market fluctuations (e.g., unvaccinated imported chicks causing cohort losses; feed and input costs tied to exchange rates). Example: a 20,000-layer farm in Benin, producing ~12,000 eggs/day (400 trays of 30 eggs), priced 1500–1800 CFA/tray (~€1000/day at 1800 CFA/tray). Despite good practices, constrained finances limited biosecurity upgrades (no showers/handwashing at gates; only shared sandals; footbaths changed every 2–3 days). Laborers worked in unsanitary, low-wage conditions, sometimes self-medicating with animal antiparasitics or purchasing antibiotics (e.g., Trimethoprim) from informal markets. Such conditions undermine sanitary barriers and can create AMR pathways even when farm-level ABU is controlled. - Perils of withdrawal (interlinked risk/production noncoherence): Ledger data showed near-monthly bacterial infections in early 2019; diagnosis via litter observation led to metaphylactic treatment of the entire flock. With 5 days of treatment and 7 days withdrawal (12 days total), observing withdrawal would mean losing 144,000 eggs and about €13,000—potentially catastrophic for the farm and dependent value chains (supermarkets, restaurants, market sellers). Thus, decisions to prioritize continuity of supply can conflict with antibiotic withdrawal rules, shaped by absent financial protections and tight margins. - Noncoherent possibilities: Given these constraints, the authors identify context-appropriate avenues: (1) strengthen digital extension/mobile veterinary consultations and social media-based awareness to help rule out bacterial infections and reduce unnecessary ABU where lab access and in-person visits are limited; (2) develop alternative indicators and low-cost, vernacular biosecurity and health solutions by leveraging local research and practices (e.g., traditional remedies, effective microorganisms for hygiene and gut health; analyzing local pharmacopeia via units like URMAPha; potential co-benefits from using brewers’ spent yeasts/grain to improve gut microbiota). Collaborative experiments can substitute for large-scale surveillance infrastructures in the short-to-medium term.

Findings show that antibiotics in West African livestock sectors act as infrastructural supports compensating for gaps in biosecurity, diagnostics, veterinary extension, and financial safety nets. Implementation noncoherences (weak state support and surveillance) push breeders toward private and digital networks for knowledge; risk noncoherences (diagnostic delays, biosecurity threats from vet visits, thin margins) incentivize rapid, often unauthorized ABU; production noncoherences (intensive breeds in low-resource settings, labor precarity) generate vulnerabilities that antibiotics help manage. Addressing AMR thus requires accommodating heterogeneity and enabling noncoherent practices to become productive ‘possibilities’ rather than enforcing uniform standards. The authors argue for adapting the GAP to local realities by: embracing digital extension to improve awareness and triage; investing in low-cost, context-specific biosecurity and health interventions; and fostering collaborations that turn vernacular practices into tested, scalable solutions. This approach can reduce reliance on antibiotics without undermining livelihoods, food security, or animal welfare.

The paper contributes a noncoherence framework to understand and manage antibiotic use in low-resource livestock sectors. By mapping implementation, risk, and production noncoherences, it shows why standardized, knowledge-deficit-based policies fall short and how antibiotics function as infrastructural ‘quick fixes.’ The authors propose transforming noncoherent problems into noncoherent possibilities through context-responsive strategies: digital veterinary extension and AMR awareness; alternative indicators and low-cost biosecurity/health technologies; and collaborations to validate vernacular practices. They call for global AMR policy to prioritize flexibility over uniformity, stitching together diverse assets and forms of labor to adapt guidelines to local conditions. Future directions include medium-to-long term investments in laboratory capacity and surveillance, compensation/insurance mechanisms to buffer withdrawal-related losses, vaccination programs, and improvements in water, sanitation, and hygiene infrastructures, alongside citizen-generated data and adaptive governance that recognizes contextual drivers of ABU.

- Sampling and comparability: Participant composition differed between Benin (mostly small-scale peri-urban breeders with limited formal training) and Burkina Faso (larger, more professionalized agribusiness owners), reflecting partner networks rather than a designed sampling frame, limiting detailed cross-country sectoral comparisons. - Scope and setting: Focus on peri-urban, semi-modern agribusinesses and backyard farms; pastoral systems and rural contexts were not included. - Fieldwork constraints: The COVID-19 pandemic halted fieldwork in March 2020, potentially limiting data breadth. - Generalizability: Qualitative, context-specific findings from a limited number of farm observations (including one 3-week stay and two shorter visits) may not generalize across the region. - Data access: Laboratory and surveillance data were limited in the study settings, and datasets are not publicly available due to ethical constraints.