This study was conducted between May and August 2025, more than one year after the Belgium ToT, to allow sufficient time for trainees to implement local activities. Semi-structured key informant interviews were used to explore participants’ experiences and perceptions. The study design was exploratory and interpretivist, seeking to capture participants’ experiences and perceptions to answer three interrelated questions: a.

Implementation – How have ToT alumni implemented and adapted the training in their home countries? What audiences were reached and through what formats? How were hands-on diagnostics integrated with theoretical instruction?

The study was designed and conducted by an independent research team from Laterite, a third-party research firm. The core team consisted of two research analysts (one female, one male), two research associates (two female), and a research manager (female) based in Laterite’s East African offices in Ethiopia, Kenya, and Rwanda. All team members held graduate-level training in social science research/economics, and had prior experience conducting key-informant interviews and analyzing qualitative data. Interviews were conducted by the research analysts and the research associates, who were trained and supervised by the research manager. Technical oversight was provided by a Principal Investigator (female, PhD in Education) who is trained in qualitative methods and based in Kenya.

No member of the research team had a prior personal or professional relationship with any participant. At the start of each interview, interviewers introduced themselves, explained the purpose of the study, clarified their role as independent evaluators rather than representatives of Zoetis’ commercial operations, and emphasized that participation was voluntary. This approach helped establish rapport and minimized perceived power dynamics or concerns about consequences for speaking openly.

Throughout data collection and analysis, the team remained attentive to how their role as external evaluators and their professional backgrounds might shape the research process. Internal debriefs were used to discuss emerging impressions, reflect on possible biases, and consider how factors such as outsider status or social desirability dynamics could influence participant responses. Interviewers probed for challenges, failures, and divergent views to counterbalance the tendency of some participants to emphasize positive outcomes. The team also reflected on how their training in social science research and lack of clinical veterinary backgrounds might affect how they interpreted technical descriptions of mastitis management and udder-health practices. To mitigate this, the team shared several analytical drafts with Zoetis staff involved in the program who have specialized expertise in veterinary medicine and dairy production. Their technical review helped clarify veterinary concepts and ensured accurate interpretation of udder-health practices. To enhance rigor and minimize interpretive bias, two researchers independently coded transcripts, cross-checked a subset for consistency, and resolved discrepancies through discussion before applying the shared codebook to the full dataset. While data saturation was not used as a formal stopping criterion, the research team noted that the nine ToT interviews yielded recurring patterns with no substantially new themes emerging, indicating that thematic saturation had been reached for the purposes of this study.

The initial ToT cohort comprised thirteen professionals representing eight countries; however, four were excluded from the follow-up due to relocation, organizational transitions, or voluntary withdrawal from the program. We purposively invited the remaining nine alumni: three from Uganda, two from Zambia, and one each from Ethiopia, Kenya, Nigeria, and Tanzania. Participants were identified by country rather than by name to maintain confidentiality. In Zambia, we conducted two supplementary stakeholder interviews: one with a dairy veterinarian working with the trainer’s previous employer and another with a partnering laboratory manager. Insights from these interviews were incorporated into the overall thematic analysis.

Semi-structured key-informant interviews (KIIs) were conducted virtually via Google Meet or WhatsApp and lasted 60–90 minutes, with one interview extending to 120 minutes. Interview duration varied depending on the depth of detail provided, particularly when participants described their training activities. The interview guide was organized into modules covering: (i) experiences during the Belgium ToT and perceived learning; (ii) post-ToT implementation, including audiences reached, delivery formats, frequency, curriculum adaptation and use of diagnostics; (iii) impacts on veterinary professionals and farmers; (iv) challenges and enabling factors such as supply chains, partnerships and behavior change; and (v) sustainability considerations, resource needs and future plans. Each module included prompts encouraging participants to provide concrete examples, discuss negative or contrasting cases, and reflect on contextual influences.

Interview guides were not shared with participants in advance. The guides were also not pilot-tested due to the small sample size and the need to avoid exhausting limited respondents; instead, iterative refinements were made during early interviews to improve flow and clarity while maintaining consistency across all participants. Interviews with Ethiopian participants were conducted in Amharic by a bilingual researcher and later translated into English by a professional translator. All other interviews were conducted in English and recorded. They were automatically transcribed in Google Meet, and the research team reviewed and corrected the verbatim transcripts.

Data were analyzed using a structured, iterative thematic approach to organize and interpret qualitative data (adapted from Gale et al., 2013 ¹² ). After familiarizing themselves with the transcripts, two researchers independently coded the data in MAXQDA. An initial set of codes was developed based on the research questions and interview guide, while allowing space for new themes to emerge from the data. The codebook covered categories such as ToT experiences, respondent roles, training implementation, enabling environment (diagnostic access, supply chains, partnerships, logistics, market incentives), mechanisms of practice change, perceived impacts on professionals and farmers, success stories, challenges and coping strategies, resource needs, sustainability, and anticipated obstacles. Deductive codes were complemented by inductive ones that captured unanticipated insights, such as farmers improvising diagnostic methods or difficulties forming public–private partnerships. Transcripts from the two supplementary Zambian interviews were coded alongside those of the nine trainers.

To enhance analytical rigor, two researchers independently coded an initial subset of transcripts and reconciled discrepancies through discussion before finalizing a shared codebook that was applied to the full dataset. We examined co-occurrence patterns (e.g., diagnostic access × practice change), actively sought negative and contradictory cases, and maintained an audit trail in MAXQDA. Quotations were selected for their distinct analytic functions (mechanisms, contrasts, and disconfirming examples) and are introduced with brief analytic bridges, attributed by country and role.

To ensure reliability, both researchers cross-checked a subset of transcripts and resolved discrepancies through discussion. The finalized codebook was applied to the full dataset, after which coded segments were reviewed, summarized, and synthesized into overarching themes. Patterns of co-occurrence (for example, between diagnostic access and reported challenges) were examined and compared across countries and participant roles. While frequency of mention suggested salience, analysis emphasized the depth and diversity of perspectives rather than simple counts. Negative or contradictory cases were intentionally explored to avoid confirmation bias. Quotations were selected to illustrate themes following guidance on quote selection: each quote serves a specific function (e.g., evidencing a mechanism, illustrating a contrast), is succinct and relevant, and is attributed by country and role. ¹³

Ethical approval was obtained from the Institutional Review Board of the Ethiopian Society of Sociologists, Social Workers and Anthropologists (ESSSWA, Reference No. ESSSWA/L/AA/05913/2025), which classified the study as minimal risk. All participants provided verbal informed consent prior to data collection. Interview recordings were stored securely with access restricted to the research team, and transcripts were anonymized to protect participant confidentiality. No incentives were offered to respondents.

3.1.1 Respondent landscape and roles in the ToT cascade

The nine participants interviewed were selected as ToT trainers because of their strategic positions within the dairy ecosystem and their ability to bridge multiple domains of practice. The participant group consisted of three private veterinarians who also held university teaching positions, two government extension officers embedded within cooperative networks, two technical managers associated with milk collection centers, one Zoetis distributor coordinating product supply and training, and one academic serving on a national veterinary board. Collectively, these roles placed them at critical junctions where clinical expertise, teaching responsibilities, institutional authority, and market functions converged. Their credibility stemmed not only from technical expertise but also from established social capital, reflected in long-standing relationships with cooperatives, local governments, and producer networks. These attributes enabled them to convene diverse audiences—including veterinarians, paraveterinarians, veterinary shop owners, and farmers—and positioned them as effective multipliers within the ToT cascade, which defined the different stakeholders impacted (as shown in Figure 1).

Participants emphasized that their roles went beyond delivering workshops. They actively brokered supplies and reagents, organized laboratory practicums, supported milk collection centers to enforce quality standards, and mentored junior staff. One Ugandan participant explained that local government veterinarians were pivotal:

This account illustrates how institutional linkages between cooperatives and government extension officers enabled continuous support. The wide reach of these individuals was also evident in the number of platforms through which they engaged farmers. In Uganda, trainers worked with a union of cooperatives covering 6,523 farmers and integrated the ToT curriculum into existing cooperative meetings. In Zambia, the trainer leveraged an integrated laboratory-cooperative hub; farmers brought milk samples for CMT screening at the laborary, and technicians were trained to run tests and provide advice. Respondents’ ability to operate across these diverse spaces enabled the ToT cascade to extend beyond single events and to integrate diagnostic routines into everyday workflows.

3.1.2 Post-ToT training implementation

All participants, except those from Ethiopia, who cited product unavailability as a challenge, delivered multiple training sessions upon returning from Belgium. The intensity and format varied by country, audience, and resource availability. The trainings successfully demonstrated the cascade of the udder health management content offered in Belgium, covering infection prevention, good milking practices, and mastitis treatment.

On average, trainers who reported organizing activity conducted three to five formal workshops within a twelve-month period, supplemented by informal demonstrations. Professional trainings were typically longer (one to three days), with the participant from Kenya noting the incorporation of Continuing Professional Development (CPD) sessions. These sessions blended lectures on udder health management – specifically mastitis epidemiology, anatomy, and antimicrobial stewardship – with hands-on practicums in CMT and culture testing. Participants emphasized the crucial link between theory and practice, with one Kenyan trainer reporting that colleagues often:

Trainers also targeted veterinary students to improve their understanding of udder health. The Tanzanian participant delivered sessions to over 40 final-year students and integrated CMT practices into laboratory classes. One Zambian participant trained 10 recent veterinary graduates and five veterinarians from the country’s largest agricultural cooperative, extending the training to 16 commercial dairy farms (targeting dairymen, supervisors, milkers, and farm owners) using small groups of 10 to 16 people for hands-on learning. The focus was mainly on mastitis management, diagnosis, treatment, prevention, and dairy farm auditing. They also trained staff at milk processing companies on milk hygiene, somatic cell counts, and fair grading practices.

Farmer-facing trainings were shorter and more frequent. Trainers piggybacked on existing cooperative meetings and milk collection center visits. Udder health sessions lasted two to four hours, focused on simple, repeatable routines: CMT screening for subclinical mastitis; cow segregation; cleaning and disinfection of udders before and after milking; milking affected cows last; and discarding the first few squirts of milk. Demonstrations were conducted with live animals at demonstration farms or watched online videos recorded during the ToT.

Participants introduced several innovative delivery methods. A Ugandan trainer described running two-day camps where farmers rotated through supervised stations (milk sampling, CMT performance, results interpretation and teat dipping). Another trainer turned a milk collection center into a mini-lab, providing farmers with immediate CMT and antibiotic residue test results on 100 milliliters of milk samples.

Trainers tailored their training content to specific audiences. Veterinarians and paravets were taught sample collection, aseptic technique and interpretation of culture and sensitivity results, along with guidance on prudent antimicrobial use. Vet-shop owners were instructed on hygiene, proper storage of intramammary tubes and referral protocols. Farmers were provided with simple checklists and encouraged to incorporate screening and hygiene practices into their daily routines. Trainers reported that audiences appreciated the practical focus and the demonstration of immediate, tangible benefits. For example, one farmer who had previously hand-milked switched back to using his abandoned milking parlor after realizing that poor hygiene practices were contributing to repeated infections.

Attendance varied from small professional sessions to large gatherings of farmers. One Ugandan trainer estimated reaching over 6,800 farmers through the union’s cooperative meetings within six months. In Zambia, one trainer conducted at least 15 large farm visits and several group trainings, which collectively reached more than 700 farmers. The small-group approach for commercial farms (10–16 people) emphasized active participation and on-farm learning.

3.2.1 The enabling environment: diagnostics, supplies, partnerships and incentives

Participants repeatedly emphasized that the success of their udder health training, particularly in terms of mastitis testing, hinged on the availability of diagnostic kits and consumables. The CMT emerged as the cornerstone of the cascade; trainers viewed it as an accessible, low-cost test that empowers farmers and veterinarians to detect subclinical mastitis early. However, supply chains were fragile. Stock-outs of CMT kits and reagents were frequent, particularly in remote areas. A Zambian participant recounted hearing farmers complain that:

Others noted that delays in customs clearance, distributor bottlenecks, and high import costs hampered consistent availability. In Ethiopia, participants emphasized that diagnostic capacity gaps were the dominant issue; without basic lab infrastructure, farmers could not confirm infections or test antibiotic sensitivity.

Participants also identified reliable laboratory networks as critical to sustaining a diagnostics-led approach. Where laboratories were far away or poorly equipped, trainers reserved culture testing for complicated cases and emphasized field tests instead. In Tanzania and Kenya, slow turnaround times for culture results undermined farmers’ willingness to pay for testing; many opted for empirical treatment (based on clinical signs and past experience, without confirming the diagnosis through laboratory tests). To mitigate this, trainers advocated establishing small diagnostic hubs at cooperatives or milk collection centers. For example, a trainer in Zambia suggested selecting “five of the biggest collection centers” and equipping each with a master CMT kit, along with training for technicians on antibiotic residue and mastitis detection. This decentralized model will reduce travel barriers and normalize routine screening.

Partnerships with cooperatives, universities, laboratories, and government agencies played a decisive role. Cooperatives and milk collection centers provided venues, mobilized farmers, and enforced quality penalties that reinforced training messages. Government extension officers lent legitimacy and offered follow-up support. Universities and professional associations offer accredited training for CPD, attracting veterinarians and para-veterinarians. Laboratories, whether public or private, served as reference points for complex cases. Participants highlighted that aligning incentives across these actors was essential; when milk collection centers penalized milk with mastitis or offered premiums for high-quality milk, farmers were more motivated to adopt hygiene and screening routines. Some trainers were exploring public-private partnerships (PPPs) to co-finance herd-health programs: cooperatives would pool farmer contributions to pay for routine diagnostics and veterinary visits, while governments and private sector would subsidize start-up costs.

Logistical resources were also essential, as trainers required funds for travel to remote farms, participant meals, and technician reimbursements. Several emphasized that without modest, predictable budgets, they could not deliver on-site coaching or follow-up visits. While some trainers used personal resources upfront and were later reimbursed, others relied on in-kind support from cooperatives or local governments. Two participants identified limited human capacity as a constraint; many farms lacked skilled staff to maintain hygiene routines, and staff turnover meant that training benefits were often dissipated unless refresher sessions were conducted.

3.2.2 Mechanisms that supported the adoption of practices

Beyond material resources, several mechanisms emerged that enabled the ToT cascade to translate into changed practices. First, hands-on learning was critical. Participants noted that demonstrating the CMT and allowing trainees to perform the test themselves helped build confidence and demystify diagnostics. A Nigerian trainer explained:

This experiential approach extended to hygiene and milking routines; farmers practiced washing hands, cleaning teats, and dipping teats in disinfectant, with trainers correcting technique.

Second, trainers framed diagnostics within the context of economic incentives. Farmers were shown that milk with high somatic cell counts or antibiotic residues could be rejected at collection points or fetch lower prices. Linking CMT results to penalties and premiums helped make the value of preventive practices more concrete and visible. One Zambian participant observed that cost was a barrier, but argued that:

By integrating training with cooperative quality systems, trainers leveraged existing economic drivers.

Third, the cascade relied on social capital. Trusted veterinarians served as gatekeepers, introducing trainers to farmer groups and contextualizing the messages. In Zambia, a trainer emphasized the importance of working through local veterinarians who knew the farmers personally and could vouch for the training. In Uganda, the union’s manager and cooperative leaders championed the program, encouraging farmers to attend and adhere to recommendations. Social networks also facilitated diffusion; farmers who adopted the CMT and saw improvements often shared their experiences with neighbors, creating peer-to-peer momentum.

Fourth, the cascade incorporated continuous support rather than one-off events. Trainers provided follow-up visits to troubleshoot implementation, observe milking routines, and reinforce messages. In some cases, they established WhatsApp groups where farmers could ask questions, share pictures of CMT results, and receive advice. This ongoing engagement addressed uncertainties and helped embed new habits. Participants noted that when follow-up was not possible due to logistics, adoption was weaker and backsliding more common.

3.2.3 Challenges in delivering training

Despite enthusiasm and success stories, the cascade faced significant obstacles. The most pervasive challenge was access to diagnostics and supplies. CMT kits, reagents, and intramammary tubes were often out of stock or delayed in shipment. When supplies were available, high costs deterred farmers, especially those with small herds. Trainers recounted that farmers mixed dishwashing liquid with milk as a substitute for the CMT reagent—a practice that yielded unreliable results and undermined confidence in diagnostics. Others described having to cancel trainings or demonstrations because kits did not arrive on time. The fragility of supply chains threatened the fidelity of the cascade; udder management training risked becoming theoretical when tools were unavailable.

Laboratory infrastructure was another constraint. In Tanzania and Kenya, trainers noted that government veterinary laboratories were underfunded, lacked reagents, and had slow turnaround times. Farmers had to travel long distances to access culture testing, and results sometimes came too late to inform treatment. In Zambia, diagnostic labs are concentrated in the capital, Lusaka, and for farms located hundreds of kilometers away, the turnaround time for culture and sensitivity testing can be 7 to 14 days, making results impractical for immediate treatment decisions. Without reliable laboratory results, trainers were reluctant to promote culture and sensitivity testing, focusing instead on CMT and hygiene. This forces a reliance on broad-spectrum antibiotics, contributing to concerns about antimicrobial resistance. Similarly, stock-outs of antibiotic residue test kits hindered efforts to promote withdrawal periods and milk safety.

Logistics and resources posed barriers. Trainers needed funds for transport, venue hire, accommodation, and meals for participants. In Uganda, one trainer lamented that:

In Zambia, there is a limited availability of specialized dairy veterinarians and accessible, timely diagnostic services. As one veterinarian stated, there are

This scarcity of expertise means that knowledge transfer and extension services are critical, yet difficult to scale.

Another noted that the cost of providing full CMT kits to all targeted farmers exceeded their budget. Trainers also faced time constraints; delivering hands-on practicums required extended interaction, and scaling such sessions across many cooperatives was labor-intensive. With limited staff and competing duties, trainers sometimes had to prioritize fewer, higher-impact sessions over broad coverage.

Behavioral and cultural factors also influenced adoption. Some farmers questioned the need for diagnostics, perceiving mastitis as an inevitable part of dairy farming. Others believed that the cost of CMT kits outweighed the benefits, particularly when milk prices were low. Trainers observed that adoption was stronger when farmers saw the link between quality penalties or premiums and their incomes. Additionally, some paravets and veterinary shop owners reported to trainers that focusing on testing and prevention measures led to a reduction in sales of broad-spectrum antibiotics. Trainers, therefore, had to navigate conflicting commercial incentives and encourage responsible stewardship, placing greater importance on preventative and continued care measures rather than treatment alone.

Environmental factors and high exposure to mastitis-causing pathogens continue to pose constant threats. The Zambian veterinarian noted that mastitis cases spike during the hot and rainy seasons, when bacterial contamination in milking environments increases, and that coliform mastitis is a significant concern, leading to culling.

As the cascade aimed to scale, institutional and partnership challenges became evident, with participants noting that forming public-private partnerships (PPPs) with cooperatives, laboratories, and government agencies was often slow and hindered by bureaucratic obstacles. The time needed to formalize agreements, align regulations, and secure approvals delayed implementation. A Ugandan participant noted that:

Trainers also worried about long-term sustainability: would cooperatives continue to invest in udder management, specifically diagnostics and training, beyond the project period? Would government support endure? These uncertainties influenced planning and caution.

3.2.4 Strategies to address challenges

Participants developed creative strategies to overcome barriers, including recommending the decentralization of testing to improve diagnostic access. By placing CMT kits and simple culture tests at milk collection centers and cooperatives, and training local technicians, they shortened the distance between farmers and diagnostics. Mobile laboratories were proposed for remote areas. Trainers also suggested working with distributors to extend supply networks into rural zones and to stock reagents alongside other veterinary products. Some proposed developing local manufacturing or packaging of CMT reagents to reduce import costs.

To address resource constraints, trainers advocated integrating udder-health interventions into existing cooperative budgets and extension programs. Several participants emphasized that farmers were willing to pay small fees when benefits were clear. Trainers also sought sponsorship for meals, venues, and transport from local government offices and dairy boards. Aligning training with CPD calendars and cooperative meetings reduced additional logistical burdens.

Behavioral change strategies focused on tailoring messages to farmer motivations. Trainers framed mastitis management as an investment in milk quality and income. In Zambia, the trainer linked screening to milk collection penalties: farmers who delivered mastitis-positive milk faced rejection or lower prices, motivating them to adopt more effective practices. In Nigeria, a trainer recommended embedding udder health into broader herd-health programs so that diagnostics were seen as part of general animal care rather than an added cost. Storytelling and peer testimonials were used to highlight benefits; farmers who had successfully reduced infections shared their experiences at meetings, reinforcing social norms.

To navigate institutional challenges, trainers proposed engaging regulators early and aligning programs with national veterinary strategies. By working with veterinary councils to recognize mastitis training as CPD, they could formalize the cascade and secure accreditation. Some participants called for national dairy boards to adopt CMT screening as part of quality standards, creating an institutional incentive. Others suggested that PPP formation be simplified through the use of standard templates and guidelines. Engaging both regulators and farmer leaders simultaneously helped align expectations and reduce delays.

3.3.1 Impacts on veterinary professionals

Interviewees unanimously reported that the ToT increased their knowledge and confidence. They appreciated the combination of theory and practice and the exposure to advanced diagnostics at the Belgium laboratory. Several had since become “resource persons” for colleagues seeking advice on udder health. Trainers described shifts in professional practice: colleagues who previously treated mastitis empirically now purchased their own CMT kits and performed tests during farm visits. Diagnostic results informed treatment decisions, with intramammary tubes and systemic antibiotics being reserved for confirmed infections. In some cases, veterinarians used culture and sensitivity results to select narrow-spectrum drugs and avoid unnecessary antibiotic use. Trainers also noted improved record-keeping; veterinarians began logging CMT results and treatment outcomes, creating a data trail that could inform herd-health plans.

The ToT program also enhanced trainers’ teaching skills. Participants reported increased confidence in facilitating workshops, structuring content for different audiences, and using participatory methods. One participant reported being invited to speak at national veterinary conferences about mastitis control. The training thus expanded professional roles from purely clinical work to encompass education and leadership in herd health management.

3.3.2 Impacts on farmers

Farmers exposed to the ToT cascade reportedly adopted several practices. The most widely mentioned change was routine CMT screening. Many farmers learned to test quarter milk samples regularly (weekly or monthly) and to segregate cows that tested positive. Trainers recounted farmers exclaiming that they could “see the problem before the cow stops producing milk,” giving them urgency to act early. Other changes included improved milking hygiene: washing hands, cleaning udders with warm water and disinfectant, discarding the first streams of milk, using separate towels for each cow, dipping teats in iodine solution after milking, and milking infected cows last. Some farmers invested in new equipment, such as CMT paddles, disinfectant sprays, and milking machines.

Participants provided anecdotal evidence of improved outcomes. In Uganda, trainers described farmers who had struggled with recurrent mastitis but, after adopting screening and hygiene routines, saw infections decline. Milk yields reportedly increased, and milk rejection at collection centers decreased. Farmers expressed gratitude, sometimes bringing milk as a token of appreciation. In Zambia, a trainer reported that culture tests identified specific pathogens, enabling targeted treatment and reducing the likelihood of relapse. These accounts suggest that the combination of diagnostics and behavior change resonated with farmers and delivered tangible benefits.

Nevertheless, adoption was uneven. Some farmers initially used CMT kits but discontinued them when the reagents ran out or when they faced other pressing financial priorities. Trainers observed that behavioral change required persistence; farmers might revert to old habits if not reminded. They also noted that improved practices were more common in cooperatives with strong management and clear quality incentives. Remote farmers, those outside cooperative structures or those with low literacy, were harder to reach and less likely to adopt routine screening.

Several narratives illustrate how the ToT cascade produced notable successes when conditions aligned. In Uganda, trainers organized a two-day “dairy camp” at a model farm owned by the cooperative union. Farmers camped on site and rotated through stations covering CMT, culture testing, milking hygiene, milking machine operation, and record keeping. The trainers included cooperative leaders, government veterinarians, and private practitioners. Feedback from farmers indicated that the immersive environment deepened their understanding and motivation. Six months later, trainers observed that participants continued to perform CMT, had invested in teat dip solutions, and were sharing knowledge with neighbors. Milk rejection rates at the union’s collection centers reportedly decreased, and the union attributed this to improved overall milk quality.

3.3.3 Sustainability prospects and resource needs

Across countries, participants agreed that sustaining and scaling the cascade would require material, technical, and institutional support. Key resource needs identified include: (i) continuous availability of diagnostic kits (CMT reagents, paddles, culture media, antibiotic residue tests) and intramammary tubes; (ii) simple laboratory infrastructure at regional hubs (including incubators, refrigerators and basic reagents); (iii) essential training aids (such as projectors, demonstration paddles, protective gear and printed materials); (iv) small, reliable operating budgets to cover local logistics (transport, accommodation, refreshments and allowances); and (v) ongoing mentorship, knowledge updates, and refresher training.

Participants envisioned a sustainability model that combined cooperative engagement, government support, and private investment from local stakeholders. This approach aims to transition the initiative from a project-based activity to a self-sustaining system.

In Uganda, trainers were piloting a PPP where farmers paid for herd-health services through their cooperatives, with government extension teams subsidizing the initial costs. This cost-sharing effort is intended to build long-term demand for udder health products and services, with trainers hoping cooperatives will eventually assume responsibility for procuring diagnostic kits and financing training. The existing accessibility gap in diagnostic kits and intramammary tubes presents a clear market opportunity for local stakeholders to target products that can prevent udder health issues. Scaling and distributing affordable, repeatable interventions (such as CMT reagents and paddles) through vet shops will create consistent local demand. Veterinary shops are seen as pivotal links in the value chain, capable of shifting their focus from solely selling mastitis treatment antibiotics to stocking and incentivizing the sale of preventive udder health solutions.

In Kenya and Zambia, participants planned to embed udder-health modules into CPD frameworks and universities through seminars, thereby ensuring preventative mastitis management becomes a standard practice for future veterinarians. Trainers also called for integrating mastitis management into national dairy policies, including quality standards and incentive mechanisms. These concerns underscore the importance of institutionalizing udder health practices within cooperative governance and regulatory frameworks. Some participants worried about what would happen beyond the current project timeline: would partnerships with cooperatives continue? Would trainers remain motivated without external support? These concerns underscore the importance of institutionalizing udder health practices within cooperative governance and regulatory frameworks.

Routine monitoring and evaluation using simple local tracking, including CMT logs, milk rejection rates, and kit use, will provide actionable and timely evidence. This evidence offers value and strengthens the commercial case for broader adoption and improved milk quality. Effective implementation of this tracking will require the close cooperation of all stakeholders ( Figure 1) engaged by the ToT participants during the cascading training sessions.

The successful execution of post-ToT activities was fundamentally contingent upon the trainers’ ability to leverage existing institutional networks and social capital. Rather than merely transferring information, participants embedded their efforts within recognized social and professional structures—such as cooperatives, local governments, and milk collection centers—that function as critical gatekeepers and reinforcing networks. This approach provided legitimacy and ensured the training was perceived not as an external imposition but as an integrated component of local development, a finding that aligns with literature on effective ToT approaches for udder health. ^{14 , 15}

This operational strategy highlights two key concepts. First, integrating the modules into existing platforms, such as cooperative meetings, aligns with the tenets of implementation science by transforming a one-off event into a continuous, organizationally supported process. ¹⁶ Second, the trainers’ personal credibility and relationships facilitated organization, demonstrating the power of relational embeddedness, where trust lowers the barrier to entry for new practices. ^{17 – 19} This reliance on local partnerships reinforces broader calls for strengthening African veterinary services through collaborative models, particularly at the operational level. ^{20 – 23} Furthermore, efforts to accredit the training for Continuing Professional Development (CPD) by Kenyan and Zambian participants represent a strategic move to institutionalize this knowledge, a key factor in enhancing the long-term quality of udder health services. ²⁴

This evaluation demonstrates that the Training-of-Trainers (ToT) program generated meaningful and measurable shifts in veterinary practices and farmer behaviors across the six African dairy systems. At the professional level, veterinarians and paraveterinarians moved away from empirical treatment of mastitis toward evidence-based decision-making. Many began purchasing their own California Mastitis Test (CMT) kits, logging results, and reserving antibiotics for confirmed infections rather than using broad-spectrum drugs by default. Several trainers reported that colleagues increasingly sought them out as “resource persons” for udder health issues, extending the program’s influence beyond direct training events. These findings suggest that the ToT model succeeded in seeding a diagnostics-led culture among veterinary professionals, with potential long-term benefits for antimicrobial stewardship.

Among farmers, the program catalyzed early adoption of preventive routines. Trainers described how farmers learned to conduct routine CMT screening, segregate positive cows, and integrate milking hygiene such as hand-washing, teat disinfection, and milking order protocols. In several instances, these practices translated into improved milk yields and reduced rejection rates at collection centers.

The study’s findings highlight a potent synergy that successfully drove this behavior change: hands-on training combined with the economic framing of diagnostics. In the adoption of agricultural technology, skill acquisition alone is often insufficient. ²⁵ Our results indicate that practical training, through mini-labs and dairy camps, was vital for building farmer self-efficacy, ²⁶ allowing them to confidently use tools like the California Mastitis Test (CMT). This hands-on approach acts as a critical de-risking mechanism for new ‘experience goods’, ²⁷ as it allows farmers to observe direct benefits before committing their own resources.

Crucially, this technical skill was consistently framed within a behavioral economics perspective, ²⁸ where diagnostic use was directly linked to market incentives, such as milk price penalties and premiums, offered by farmer cooperatives and processors. This shifted the farmer’s motivation from mere compliance to an active cost-benefit analysis, transforming the CMT kit from an abstract tool into a direct investment in their income. The subsequent decline in milk rejections and increase in yield provided tangible reinforcement, consistent with evidence supporting incentives and improving farmer knowledge on the economic impacts of mastitis to overcome adoption hurdles in Sub-Saharan Africa. ^{29 , 5 , 30}

However, the data also revealed that adoption was uneven and often discontinued when farmers faced financial challenges or ran out of supplies. This aligns with the broader literature, which suggests that financial capacity remains a key determinant of sustained practice, even when strong market incentives are present. ^{31 , 32} Sustaining adoption, therefore, requires not only a clear economic signal but also a viable local supply chain for diagnostic tools and a means to increase farmers’ wealth. Additionally, adoption also varied between cooperatives. Well-organized cooperatives with strong management and quality-linked payment structures experienced high uptake, whereas isolated or resource-limited farmers were more challenging to reach and less likely to adopt new practices.

The long-term sustainability of diagnostic practices introduced by trainers is threatened by systemic barriers to sustained diagnostic access, including frequent stockouts, high input costs, and heavy reliance on distant, centralized laboratory infrastructure. These challenges are not unique to participants’ experiences but reflect a well-documented, pervasive failure within the last-mile logistics of Africa’s animal health supply chain, which remains highly fragmented, unreliable and inadequately regulated. ^{33 , 34} For instance, the geographic distance of laboratories is often cited as the principal constraint to their use, ³⁵ which renders culture and sensitivity testing impractical (as reported by trainers in Kenya and Tanzania) due to slow turnaround times. This systematic fragility undermines behavior change, as the lack of reliable and timely access to reagents and consumables when needed makes it impossible for farmers to sustain new quality assurance practices.

In response, the alumni’s initiative to establish local diagnostic hubs with co-financing mechanisms represents a practical, bottom-up approach to decentralizing service delivery. This approach mirrors successful market systems development (MSD) strategies, such as those used by the Mercy Corp RIPA-North program and Boehringer Ingelheim’s LatMile initiative, which shift from centralized “push” systems to market-driven models that empower local actors. ³⁶ Such a model leverages the respective strengths of public and private sectors. There is emerging evidence, for example, from Ethiopia, showing that while farmers rely on the public sector for services like vaccination campaigns, they prefer private providers, despite higher costs, for clinical services, especially mastitis management, due to greater accessibility and responsiveness. ³⁷ By fostering these public-private complementarities, decentralized hubs can build a more resilient and sustained udder health system, an approach strongly supported in animal health literature. ³⁸

Finally, the study’s findings also reveal that vet-shops emerged as an important actor in the enabling environment. Trainers observed that these outlets remain primary suppliers of intramammary tubes and systemic antibiotics. Their sales incentives often ran counter to preventive messages, reinforcing a curative over a diagnostic-led approach. This dynamic presents both a challenge and an opportunity: unless engaged, vet-shops risk undermining stewardship goals, but if repositioned, they could become effective allies in expanding access to preventive products such as teat dips, disinfectants, and CMT kits.

Ultimately, these findings suggest that the primary constraint on long-term impact is not a lack of technical training or farmer willingness, but the failure to integrate diagnostic tools into a resilient, profitable, and locally managed commercial pipeline. Therefore, integrated logistics and supply chain reform appear to be a prerequisite for the sustainability of technological adoption in this context.

Only nine of the thirteen alumni were interviewed; perspectives from South Africa and Zimbabwe are absent, and our findings may not capture the full diversity of experiences. Farmer voices were indirect, filtered through trainer perceptions; direct interviews with farmers could provide deeper insight into adoption and challenges. Outcome measures were self-reported; we did not verify the number of farm visits or the number of trainings conducted, for example. Therefore, the study cannot quantify the extent of adoption or its effect on production or income. Additionally, the evaluation occurred within a specific time frame; longer-term follow-up is needed to assess sustainability.

This evaluation demonstrates that the ToT cascade effectively transferred practical mastitis control skills across six African dairy systems, generating measurable improvements in veterinary practices and farmer behavior. Success was driven by trainers’ institutional embeddedness, social capital, hands-on training, and economic framing of udder-health interventions. However, long-term sustainability is constrained by systemic barriers, including unreliable diagnostic supply chains and market pressures. Strengthening public-private partnerships, local diagnostic hubs, and veterinary shop engagement is critical to sustaining adoption, reinforcing the importance of integrating technical training within resilient, locally managed service systems.

We make the following recommendations for policy and practice ▪

Frame udder health as an economic imperative: Position mastitis control as essential for safeguarding farmer livelihoods, driving profitability, and ensuring cooperative viability, as mastitis is a costly disease directly threatening market access.

Our evaluation opens several avenues for further study. First, longitudinal research could assess the durability of practice change and measure farm-level outcomes such as somatic cell counts, milk yield, milk rejection rates and antibiotic usage. Mixed-methods designs combining surveys, observational visits, and laboratory tests would provide a more comprehensive picture. Second, comparative studies could test different support models, such as providing diagnostic kits versus grants for logistics, and examine their effects on adoption and cost-effectiveness. Third, research should amplify farmer voices through focus groups, ethnographic observation, or participatory action research to understand drivers and barriers from their perspectives. Fourth, evaluations could explore the gender dynamics of udder-health practice: how do women and men farmers adopt diagnostics and hygiene routines, and what additional support might be needed for equitable participation? Finally, meta-analyses of ToT programs across sectors could refine conceptual frameworks and identify cross-cutting determinants of cascade success.