https://orcid.org/0000-0003-2676-3551
Figures
Abstract
The faecal contamination of irrigation water threatens public health. Although safe practices can mitigate hygiene and food safety risks along the urban irrigated vegetable value chain, their adoption remains limited. A behaviour framework was combined with a participatory approach to explore how institutions influence stakeholders’ capability, opportunity and motivation to adopt safe practices in Accra, Ghana. After extensive preparation, a dialogue engaged stakeholders and institutions in identifying the actors and interactions influencing stakeholder practices. We found that institutional dynamics and misaligned priorities hinder stakeholders’ opportunity and motivation to adopt safe practices, while their capability (education and skills) is not actively hindered. Knowledge gaps created by top-down approaches and sectoral silos were bridged by engaging participants in conducting the behavioural diagnosis. This shared understanding highlights the need to integrate and harmonise policies, regulations and service provision across water, sanitation, agriculture and health sectors, enabling participants to co-design arrangements that make safe practices easier to adopt.
Citation: Galibourg D, Scott RE, Gough KV, Amoah P (2025) Institutional barriers to food safety: The irrigated vegetable value chain in Accra, Ghana. PLOS Water 4(11): e0000378. https://doi.org/10.1371/journal.pwat.0000378
Editor: Tafadzwanashe Mabhaudhi, London School of Hygiene and Tropical Medicine / University of KwaZulu-Natal School of Agricultural Earth and Environmental Sciences, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND
Received: April 28, 2025; Accepted: November 6, 2025; Published: November 24, 2025
Copyright: © 2025 Galibourg et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data is available in the Loughborough repository on the following link: https://doi.org/10.17028/rd.lboro.28827659.
Funding: This work was supported by the UKRI Engineering and Physical Science Research Council (EPSRC) through a PhD studentship received by DG as part of the EPSRC Centre for Doctoral Training in Water and Waste Infrastructure and Services Engineered for Resilience (Water-WISER). EPSRC Grant No.: EP/S022066/1. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Over 1.7 billion people in cities of Low- and Middle-Income Countries (LMICs) lack access to safely managed sanitation [1]. This leads to the faecal contamination of surface water that urban farmers need to irrigate the vegetables they supply to these cities [2–4]. Perishable vegetables, essential to a healthy diet, are expensive and difficult to transfer from rural areas to cities without adequate cold transport and storage [5,6]. Consequently, urban and peri-urban farmers produce most of the leafy greens consumed in cities and do so year-round thanks to irrigation [7]. Many of these leafy greens are consumed raw as salads or toppings on other dishes, making raw produce and street food the leading exposure pathway to faecal contamination in many LMIC cities [8,9], exposing farmers, vegetable traders, street food vendors and consumers to diarrhoeal and parasitic infections [8–11]. Contaminated vegetables contribute to 420,000 annual deaths and 600 million illnesses from foodborne diseases, costing over USD 100 billion in LMICs [12,13]. Moreover, pathogens can also lead to severe public health crises, such as the 2024 cholera outbreak in Ghana, which led authorities to ban vegetable sales [14]. Food security and nutritious diets must not come at the expense of food safety [15,16].
While achieving universal, safely managed sanitation remains a long-term challenge [17], the World Health Organization (WHO) multiple-barrier approach recommends that stakeholders adopt safe practices to mitigate the risk of faecal contamination along urban irrigated vegetable value chains, ‘from farm to fork’ [18]. Practices include simple on-farm water treatment and non-treatment options, such as growing crops not eaten raw, adopting different irrigation regimes, or washing vegetables safely. Despite its benefits, the uptake of this approach remains low [19]. The Capability+Opportunity+Motivation=Behaviour (COM-B) behavioural model posits that adoption requires stakeholders to have the Capabilities, Opportunities, and Motivations to perform safe Behaviours [20]. Top-down approaches and barriers between water, agriculture and health sectors prevent comprehensive identification of stakeholders’ Capability, Opportunity, and Motivation needs [21]. Additionally, institutional siloes lead to misaligned priorities where some institutions hinder stakeholders’ capabilities and opportunities to adopt safe practices recommended by other institutions [22].
This study aimed to diagnose the institutional factors hindering farmers’ capability, opportunity and motivation to adopt safe irrigation practices. We invited stakeholders and institutional representatives to dialogue, share experiences and develop a shared understanding of the challenges. To our knowledge, this study is the first globally to combine a behavioural framework and a participatory approach for stakeholders and institutions to co-develop a systematic and comprehensive evidence-based behavioural diagnosis and actionable insights to implement a multiple barrier approach. These insights are key to co-designing arrangements that enable farmers and other stakeholders to enhance hygiene and food safety. The process and insights are also relevant for the many cities where irrigated vegetable value chains depend on contaminated water.
The subsequent method section outlines how we integrated WHO’s multiple barrier approach, the COM-B behavioural framework and a participatory method in a multistakeholder dialogue. We then present participants’ perspectives, analyse institutional influences, and elaborate on the shared understanding that emerged from the dialogue. We reflect on the process before concluding.
Methods
Ethics statement
This study was reviewed and approved by Loughborough University’s Ethics Review Sub-Committee (ref 2022-7727-12403) and IWMI’s Institutional Review Board in Ghana (ref 2023_04). Participants’ formal consent was obtained in writing and verbally.
Context
Accra’s urban farmers face misaligned priorities between key institutions, hindering their adoption of recommended safe practices. A notable example, indicated through earlier stakeholder interviews, identified that agricultural extension agents from the Municipal Departments of Agriculture and Public Health motivate farmers to use piped water instead of surface water to reduce faecal contamination [23]. However, the water utility often hesitates to connect farmers willing to use piped water to the piped network, prioritising domestic customers [23]. With over 6% of households in Greater Accra resorting to open defecation, and under 3% being connected to sewers, over 90% use septic tanks, private or public pit latrines, and under 10% of domestic faecal sludge and wastewater reaches a treatment plant [24,25]. As a result, the vast majority of over 1,000 farmers use contaminated water to irrigate leafy greens, including lettuces, spring onions and herbs such as mint or coriander, and about 15,000 street food vendors serve these vegetables raw to over 85% of the city’s population [26–29]. Contaminated vegetables are a primary source of faecal exposure for adults and children in Accra, as well as many LMIC cities, where the burden of foodborne diseases is likely to increase as diets transition to include more raw vegetables [12,29–31].
This study combines a behavioural theoretical framework (COM-B) and a participatory approach (Companion Modelling) to analyse how institutional obstacles hinder hygiene and food safety along Accra’s irrigated vegetable value chain. Fig 1 illustrates how these approaches integrate across four levels: the WHO multi-barrier approach (level 1), the COM-B framework (level 2), Companion Modelling with the ARDI method (level 3), and the multistakeholder dialogue (level 4). As shown in Fig 1, the article focuses on institutions and stakeholders responsible for adopting safe practices that enhance food safety. Consumers of ready-to-eat street foods benefit from these practices but are outside the scope of this article, as they have little direct control over hygiene and currently exert minimal demand for safer food. Notably, all participants are themselves consumers, so the consumer perspective is implicitly reflected. This focus allows the study to concentrate on institutional barriers to the adoption of safe practices.
To protect public health, the WHO recommends a multiple-barrier approach whereby stakeholders adopt a combination of safe practices ‘from farm to fork’ [18]. Using the COM-B behavioural framework ensures all factors influencing stakeholders’ adoption of these safe practices are considered. The Companion Modelling participatory approach helps operationalise the COM-B framework by enabling stakeholders and institutional representatives to participate in a dialogue to collectively develop a shared understanding of the Actors, Resources, Dynamics and Interactions (ARDI) that influence stakeholders’ capability, opportunity and motivation to adopt safe practices. This strategy provides a systematic and comprehensive evidence-based behavioural diagnosis and actionable insights, which are prerequisites to designing an effective and sustainable intervention.
A previous formative study tracked vegetables from farm to fork, interviewing stakeholders along the value chain about their water- and food safety-related practices and the institutions that influence these [23]. Here, “stakeholders” refers to farmers using contaminated water, traders buying and selling farmers’ vegetables, street food vendors, and consumers. “Institutions” are the organisations, formal or informal, that determine the rules influencing stakeholders’ adoption of safe practices. Institutions mentioned by stakeholders include the Departments of Agriculture of the Metropolitan, Municipal and District Assemblies (MMDAs), the water utility Ghana Water Company Ltd (GWCL), the Food and Drug Authority, whose representatives were subsequently interviewed. Analysis of the 75 interviews for this previous study revealed that institutional siloes and resulting sectoral barriers leave intervening institutions unaware that institutions with other priorities may hinder stakeholders’ adoption of the safe practices they recommend.
COM-B framework
Theory-based behaviour models, such as the COM-B model, help practitioners ensure they consider all factors influencing stakeholder behaviour. The COM-B model posits that for a stakeholder to adopt a safe practice or behaviour, they need to have the Capability, Opportunity and Motivation to perform that Behaviour [20]. Capability, physical and psychological, includes the strength, skill, or knowledge needed to perform a behaviour. Opportunity, social and physical, refers to the external factors that make a behaviour possible or easier to perform, such as infrastructure or resource accessibility. Motivation refers to the mental processes, automatic or reflective, that drive behaviour performance. The Capabilities and Opportunities influence the Motivation, all three influencing and being influenced by the Behaviour [20], as illustrated in Fig 1. This model synthesises and expands earlier models, such as the Theory of Planned Behaviour [32] and the Health Belief Model [33] into a simple yet robust framework to understand what influences behaviour [20,34]. Behavioural diagnosis is applied in diverse sectors, including some related to sanitation and hygiene [35], food safety [36], and policy development [37].
Companion modelling approach and the ARDI method
This study draws on the Companion Modelling participatory approach [38,39] to facilitate behavioural diagnosis. It uses dialogue between stakeholders, institutions and researchers to enable them to learn from each other, provide continuous feedback and develop a shared understanding of the factors influencing the adoption of safe practices along the value chain. In particular, the study employs the ARDI method, one of the first steps of the Companion Modelling approach whereby participants identify the Actors, Resources, Dynamics and Interactions that shape a value chain [40]. Systematically analysing these through the lens of the COM-B model addresses the knowledge gaps around mechanisms by which institutions may impede stakeholders’ adoption of safe practices along a value chain.
Multistakeholder dialogue
A multistakeholder dialogue was hosted at the International Water Management Institute (IWMI) in Accra on 13 September 2023. Beginning on 24 August 2023, invitations outlining its purpose and agenda were hand-delivered to urban farmers, vegetable traders and street food vendors involved in prior fieldwork. Invitations were also distributed to the directors of 17 institutions. Follow-up calls confirmed participants’ attendance. The purpose and agenda were verbally clarified, and invitees who could not attend were requested to delegate a substitute. Almost all invitees responded positively. IWMI’s history of action research in collaboration with urban farmers, including on wastewater irrigation, and government bodies, and research institutions in Greater Accra, likely contributed to the high attendance. Previous interactions of the authors with participants also fostered trust and engagement. Unfortunately, most vegetable traders and street food vendors could not attend due to timing conflicts with their income-generating activities. This prevented the direct exchange of perspectives between these actors and institutions. Some institutions delegated representatives from different services, while others assigned representatives at the last minute who lacked a proper briefing.
The dialogue brought together 29 participants (14 females and 15 males) representing 21 stakeholder groups and institutions, including farmers, traders, government officials, and researchers (Table 1). Participants were mostly senior representatives involved in agricultural production, food safety regulation, water and sanitation management, urban planning, environmental protection, and policy implementation. During registration, participants were reminded of the dialogue’s objectives and informed, verbally and in writing, that the session would be audio-recorded for research purposes. They then provided informed consent by signing the attendance sheet, which included a consent statement, and by giving their contact details to record attendance and facilitate follow-up. At the start of the session, participants were verbally reminded that the session would be recorded before the discussion began. No minors participated in the study.
The dialogue lasted five hours, facilitated by two researchers with expertise in agro-hydrology, wastewater, food safety, and microbiology. The lead facilitator is the paper’s first author, who is also trained in COM-B and Companion Modelling. Most participants were previously interviewed by the lead facilitator, offering continuity. Water-related practices affecting hygiene and food safety ‘from farm to fork’ were introduced to participants through a short presentation given by the facilitators, highlighting contamination sources and the economic burden of foodborne diseases. The institutional barriers to the adoption of safe practices, identified during prior interviews, were also introduced by the lead facilitator. These included poor access to safe water for irrigation and safe vegetable washing, lack of infrastructure and land tenure insecurity. Participants were then invited to share their perspectives and respond to each other’s viewpoints, while the facilitators steered the discussion to identify the Actors, Resources, Dynamics, and Interactions at play for each barrier. Participants mentioned underlying factors and proposed and evaluated potential solutions. To ease farmers’ reluctance to challenge authorities, the lead facilitator agreed to present sensitive issues on their behalf.
Participants engaged in a candid and productive dialogue, remaining in a group discussion rather than breaking into activities as planned. The dialogue concluded with the participants summarising key insights, which the facilitators recorded on flip charts. Participants appreciated that such a range of stakeholders and institutions had agreed to discuss often-overlooked food safety concerns. As the Eden Tree representative stated: “Thank you very much for inviting me to this. I’m actually enlightened because, for so long, my perception was that nobody cared … [So much work has been done] but we don’t know”.
The dialogue transcript was manually coded to identify the Actors, Resources, Dynamics, and Interactions mentioned by participants. The coded data were then analysed using the COM-B model to systematically identify institutional barriers that affect stakeholders’ Capability, Opportunity, and Motivation to adopt safe practices, which are discussed in the next section. To protect participants from potential negative consequences, statements on sensitive topics are attributed generically (e.g., ‘a participant stated...’) rather than to specific individuals. In some cases, attribution was not possible due to the dynamic nature of discussions involving multiple participants, where identifying individual speakers was not feasible.
Results and discussion
Stakeholders’ initial perspectives
The Departments of Agriculture and Public Health view urban farmers using piped water for irrigation as an easy solution to reduce faecal contamination and protect public health. Agricultural Extension Agents occasionally condition farmers’ access to training, subsidies, and awards on their adoption of this safe practice. These Departments attribute low compliance to farmers’ limited awareness of risk and lack of incentives, prompting them to prefer the uninterrupted supply of free surface water from canals over the safer yet charged for and intermittent supply of piped water.
GWCL’s perspective is best reflected by a statement one of their representatives made in a formative interview, and that was repeated in the dialogue:
When they [farmers] apply [for a connection], we do install, but we are not encouraged to connect them … because they will be billed at the commercial rate, and they can’t pay. They can also leave their farm. The tendency for them to steal [piped water] is always there because of the rate [charged]. Just recently, in February-March [2023], the tariff has been increased by 167%. Even companies and other commercial users are finding it very difficult [to pay], let alone peasant farmers… We encourage them to sink boreholes.
Farmers using surface water argue that the intermittent supply of piped water disrupts their irrigation schedules. Those relying on piped water confirmed doing so only because they cannot access surface water, raising complaints about issues including pipe bursts and infrastructure failures, leaving them without water for weeks. Their main concern is the rising cost of piped water, which makes bill payment difficult and can lead to disconnection. Reopening the connection involves long delays and extra fees. Utility agents may charge taxi fares and repair costs for interventions in the public domain, a common practice among underfunded public utilities [41].
Developing a shared understanding of the problem
Farmers saw their water bill increase in February 2023, unaware of GWCL’s lower “non-residential” rate, which capped other small businesses at 8.3%. GWCL learned that local agricultural and health officials were promoting piped water for irrigation to enhance food safety. Other officials discovered that GWCL did not systematically connect farmers to the piped network and had received approval from the Public Utility Regulatory Commission (PURC) to charge a substantially higher rate than that for households or small businesses. The actors, resources and interactions discussed during the multistakeholder dialogue are modelled in Fig 2.
Agricultural officers, empathising with farmers, questioned the higher rate. The GWCL representative explained: “As long as you’re using the water to enhance your business, you have to be on the commercial [rate]”, before adding:
Our priority is to supply water for domestic purposes and commercial … Initially, we didn’t supply them [farmers] with water … But we realised that they also need the water … They’re metered, but they don’t pay … Certain areas like Plant Pool, they use a shared meter. So, if a farmer decides not to pay, … we have to cut supply to the area … The intermittent supply, we can’t do anything about it. The raw water source that we have is a lot, but the infrastructure to treat this water and supply it to our customers is a challenge.
GWCL has increased tariffs despite abundant raw water resources to address rising production costs and utility debt. These have been exacerbated by a devalued national currency and a high non-revenue water ratio largely due to ageing infrastructure [42,43]. Local GWCL officers know that urban farmers struggle to pay for water, but the utility prioritises cost recovery. This strategy discourages farmers who use surface water from switching to piped water and incentivises those who use piped water to steal it or revert to using surface water. As a farmer emphasised:
Today [I pay] 2000 [cedi per month, about USD174 at the time], tomorrow 3000, 5000, 7000! All the farmers, it is just like we are stuck.... If you don’t make it affordable for me to work, either I steal from Ghana Water, or I use the stream [i.e., surface water]. If you make it affordable for me, I’ll pay. But if you don’t make it affordable, I’ll use the alternative.
Some farmers lost their motivation to use piped water and reverted to surface water years ago when availability or cost became too challenging. Those without access to surface water lose the opportunity to irrigate altogether and reduce production, losing their livelihood and leaving the market to those using contaminated water. In both cases, hygiene and food safety are jeopardised.
Participants fear decision-makers underestimate the problem, not realising that all consumers may be exposed to contaminated vegetables and that urban farmers’ irrigation practices are key. As a MoFA representative pointed out:
We are all at risk because we don’t know what we are eating. You can go to a hotel [restaurant], you don’t know where they are sourcing their vegetables from, because I have seen people buying at places and their vehicles packed far away. So then you don’t associate them with the source where they are buying these vegetables... Now the chemical [contaminants take time] … but the microbial will kill you.
Through the multistakeholder dialogue, participants developed a shared understanding of the institutional barriers affecting farmers’ access to piped water, bridging knowledge gaps across sectors. This process revealed how pricing policies, infrastructure limitations, and regulatory constraints shaped farmers’ decisions, helping to map the system dynamics. These insights lay the foundation for participants to propose potential solutions while better accounting for their respective constraints.
Co-designing solutions and discussing their feasibility
Participants proposed solutions for farmers to use safer water and discussed their feasibility based on their respective experiences and perspectives. The GWCL representative advised farmers to drill boreholes for groundwater, but concerns arose over salinity risk and high upfront costs. Participants noted landowners are hesitant to allow drilling, fearing installing permanent infrastructure may grant farmers land rights. A farmer claimed that even if his landowner agreed to a borehole, and he managed to secure the investment and treat the salinity, “When, tomorrow, one big man comes for the land, where am I taking the borehole? He’s the one who is going to benefit!”
Land tenure insecurity hinders farmers’ opportunity and motivation to invest in equipment that would make irrigation safer, as a director of MMDA’s Department of Agriculture pointed out:
The land for farming within Accra... on paper, there’s a green belt. On the ground, there’s no green belt; it’s gone. And so, my farmers understand they don’t have the land anymore... We look at ways of intensifying productivity per unit area so that they can still get something out of it for their livelihoods. They don’t invest beyond that. Irrigation, they’ll buy pumping machines, they’ll buy sprinklers [spray tubes] because you can roll it up and take it away. They will not dig a borehole because it is expensive, and tomorrow the landowner will come for their land.
Urban farmers cultivate mostly undeveloped public or customary land, typically under power lines and near railway tracks and surface water [44]. Families with traditional authority own most of the land, while MMDAs regulate its use [45]. A Land Use and Spatial Planning Authority (LUSPA) representative and a University of Ghana researcher noted that a 2020 law supports MMDAs in preventing unauthorised land-use changes. However, outdated plans, land speculation, financial interests, and miscommunication hinder enforcement, leading to rezoning de facto floodable open spaces near streams into residential areas [45].
A participant proposed a piped network could supply raw water to farmers, as in other countries. However, others responded that laying pipes is costly and fixing the existing treated water network should be prioritised. Some suggested rainwater harvesting, but the cost and space for storage between the two rainy seasons could be prohibitive. Representatives of the Departments of Agriculture indicated that they favour backyard and school gardening, where there is better access to land and piped water, to address the demand for safer vegetables from households and schools. MoFA’s Women in Agricultural Development (WIAD) directorate develops educational material for this purpose.
Another participant suggested subsidising piped water for irrigation
If we have to look at some social protection measures, we can redirect it to maybe reducing the tariffs on these farmers who use potable water, and then we can pick it from there [… but if the rates are high, then vices will just creep in, and we won’t win.
Some participants suggested GWCL provide an affordable tariff for farmers, as it does for other businesses:
Ghana Water Company must just change their policy. Ghana Water can go to parliament and PURC [the Public Utility Regulatory Commission] and say it’s about time we provide water for irrigation for our own safety and well-being... And water and sewage must be up and running. It’s about time!
Interestingly, this is a rare instance in which participants acknowledged ineffective sanitation. During formative interviews, most representatives outside of agriculture blamed farmers for using contaminated water rather than questioning why the water had become so polluted.
The role of the Public Utility Regulatory Commission (PURC) is to rule on GWCL’s proposed new tariff structure and customer categories. PURC ratified moving from the commercial to the newly created non-residential rate for any “small and medium scale non-domestic customers that use water for drinking and sanitary purposes only, including churches, mosques, faith-based organisations, registered non-profit organisations, schools, hospitals, barbering shops, welding, carpentry and allied businesses, water sellers, shops and offices including firms and retailers” [46]. PURC then redefined commercial customers as “entities and businesses that use water as a significant input in their operations other than for drinking and sanitary purposes such as hotels, restaurants, chop bars, fuel stations, airports, financial institutions, washing bays, laundries, block factories, hair and beauty salons and public toilets” [46]. Finally, PURC approved a smaller increase for non-residential customers than requested by GWCL, while exceeding GWCL’s request for commercial customers. PURC’s consideration that water is not a significant input for water sellers and that public toilets do not use water for “sanitary purposes”, denotes the flexibility of the categorisation. These arrangements arguably aim to shield domestic customers and small businesses from the major tariff increase, while relying on more affluent customer categories to improve GWCL’s cost recovery.
Keeping urban farmers in the commercial category does not substantially help GWCL’s cost recovery but thwarts the Departments of Agriculture and Health’s efforts to improve hygiene and food safety. While the government subsidises seeds and fertilisers to support the agricultural sector [47], the water tariff is a bottleneck for urban agriculture and food safety, raising public health risks. Moreover, GWCL disincentivising the use of safe water for non-drinking purposes arguably contradicts its mission to meet urban customer demand and protect health and safety [48]. Conversely, applying an affordable rate would reduce the number of unpaid bills by farmers and illegal connections, motivate more farmers to adopt piped water, increase GWCL revenue, and improve hygiene and food safety.
A MoFA representative suggested involving the Ghana Irrigation Development Authority (GIDA) to share their experience with GWCL regarding farmers paying for irrigation water:
The Ghana Irrigation Development Authority... has a way where farmers pay for irrigation service charges... They pay for fuel to draw water from the source... So, I think GIDA and the Ghana Water Company should [have] a high-level discussion so that the two of them will see how GIDA does that for farmers... and replicate the same treatment down to the farmers, as far as Ghana Water Company is concerned.
GIDA was not invited to the dialogue because, during formative interviews, its representative stated that urban irrigated agriculture fell outside its scope. The representative cited the institution’s lack of regulatory power, especially over informal practices. However, GIDA’s experience could help GWCL work with urban farmers. Another MoFA representative noted GIDA’s collaboration with traditional authorities to secure land tenure, which could assist the Municipalities in supporting urban farmers and increasing food sovereignty and safety.
GWCL’s customer categorisation, MMDAs’ lack of enforcement of land use regulation, and GIDA’s disengagement raise the question of institutions’ own capabilities, opportunities and motivations to play their part in enabling the adoption of safe practices that improve food safety and public health. GWCL and PURC may not have the motivation to apply the “non-residential” rate to urban farmers. If this is the case, then MoFA may not have the opportunity to bring the matter to parliament, nor have the motivation to develop a strategy to support safer urban irrigated agriculture, nor the knowledge and skills specific to urban agriculture. Consequently, municipalities may not have the motivation and capabilities to enforce land-use regulations, or allocate adequate resources to allow Departments of Agriculture to support urban agriculture. A Director of a Department of Agriculture explained: “When I went to the Municipal Assembly the first time, the budget officer didn’t want to listen to anything agriculture”. Another added:
The challenge is that the voice of agriculture is not too strong, and it is not too valued or recognised. Because we as a nation talk agriculture [but] we don’t have agriculture in our hearts, we don’t have agriculture in our minds. Even though we say we are an agricultural country, it’s just on paper... Our money doesn’t go into agriculture.
Another participant gave an example:
You go to a farm today, tomorrow … there’s a structure there... Everybody wants to [buy] land, and they are paying hundreds of thousands of dollars for the land. So, the farmers are being pushed out... On paper, there’s a green belt [in Accra] where the land is supposed to be left for farmers to enhance the greenery, but everything is built up, and it’s high-level top politicians and big management people who sit in parliament and make those decisions not implementable.
As a MoFA representative put it:
It means we have to act together. … So I think we’ll have to take this to another level where we can also bring up options if people are using this [water] for irrigating vegetables within the urban space... to safeguard the health of the consumer; it’s very crucial.
Similarly to Fig 3, which was presented at the start of the dialogue, the participant translates the health burden from food safety issues into economic terms and shows the dire situation of Ghana compared to other countries with similar gross national income (GNI). The horizontal axis represents the GNI of selected LMICs, while the vertical axis indicates productivity loss from foodborne diseases. A vertical line compares the economic burden with sanitation coverage in countries with similar GNI. In 2010, productivity loss in Ghana was more than thrice that of Vietnam and ten times that of Uzbekistan; their respective enhanced sanitation coverage was 15%, 69%, and 95%. Translating the health burden from food safety, especially contaminated irrigation water, into economic terms could motivate politicians to develop new policies to align strategies across the agricultural, health and water sectors at local and national levels.
A lack of institutional coordination across levels and sectors limits food safety promoters to targeting potential small improvements by focusing on quick, in-depth changes, such as stakeholder practices. Small gains are essential to securing farmers’ commitment. However, changing practices without changing the environment that generated them is labour- and cost-intensive for farmers, challenging to sustain and monitor, and unlikely to improve food safety and public health at the municipal scale.
This finding aligns with broader governance literature on transformative change, which demonstrates that system-wide improvements require coordination at high levels rather than isolated practice changes [49]. Similar coordination challenges have been documented across diverse contexts in the Global South. In urban food systems across Africa, Asia, and Latin America, fragmented governance approaches that target individual stakeholders without addressing systemic barriers have proven insufficient for achieving food safety at scale, with successful interventions requiring coordination across multiple government levels and sectors [50]. Likewise, health system strengthening efforts in West Africa have shown that focusing on individual practice changes without addressing the broader institutional environment leads to unsustainable improvements, while successful transformations require “vibrant collaborations amongst researchers, decision-makers and civil society” supported by coordinated governance across national, regional, and global levels [51]. The Food and Agriculture Organisation of the United Nations (FAO) similarly emphasises that the effectiveness and sustainability of food safety systems in the Global South depend on coordination among agencies and stakeholders [52].
These converging findings across food safety, urban food systems, and health systems suggest that the coordination challenges observed in Accra reflect a broader pattern in development interventions, where sustainable improvements at scale require high-level institutional coordination rather than fragmented approaches targeting individual stakeholder practices.
Achievements, potential and limitations of the process
The multistakeholder dialogue successfully engaged a broad range of participants in collective learning. The COM-B framework and Companion Modelling approach allowed participants to collectively specify underlying issues not previously addressed. These insights, along with the feedback from participants, enriched everyone’s understanding of the institutional factors hindering farmers’ capability, opportunity, and motivation to adopt safe practices.
Extending this systematic and comprehensive analysis to the entire value chain requires opening the dialogue to the “extended peer community consisting of all those with a stake in the dialogue on the issue” [53]. Participants suggested involving institutions that had not initially been considered or identified. The dialogue occurred during office hours, which was convenient for institutional representatives and manageable for farmers but conflicted with the daily activities of vegetable traders and street food vendors, who were unable to participate. While their concerns were represented through prior interview findings, future multi-stakeholder engagements should prioritise conditions that enable their direct participation to strengthen cross-actor understanding. Moreover, Companion Modelling is an iterative and time-consuming process which may require more time than participants are willing or able to allocate, thus perpetuating a status quo where the most vulnerable groups are not heard [54].
Some institutions were reluctant to engage in the process, delegating representatives who acknowledged being uninformed about the topic or working in a position irrelevant to the matter. None of the participants positioned themselves as decision-makers. Executives are less able to find time to engage in collective learning and may be more interested in preserving their position, resulting in a disconnect between knowledge and decision-making powers [55]. While this limited the immediate impact of the first iteration, over time, further multi-stakeholder dialogue could bridge this gap. Participants would need to comprise a balanced mix of executives, managers and experts (including stakeholder representatives), with interactions regular enough to foster trust and reciprocity to enable brokering knowledge and policy-making [55].
The multistakeholder dialogue illustrated how stakeholders’ adoption of safe practices depends on institutions’ own capabilities, opportunities and, particularly, their motivations. Participants call for the urgent prioritisation of hygiene and food safety in local bylaws and national policymaking, raising the question of the political will (i.e., institutions’ motivations) and its influence on institutional capabilities and opportunities. Developing narratives that connect the dialogue with institutions’ value systems could motivate executives to engage or delegate representatives with decision-making powers [56,57]. For example, while food safety along the vegetable value chain may not be at the core of GWCL priorities, and they may push against farmers’ use of piped water for irrigation, they might engage in a dialogue that gives them the opportunity to increase their cost recovery from farmers. Neoliberal shifts often lead to a marginalisation of urban agriculture, which is then perceived as a practice that does not belong in the modern city [58,59]. Since so many urban dwellers depend on urban agriculture for their livelihoods and access to a nutritious diet, it is paramount to secure strong political will at the highest level to integrate urban agriculture into urban planning for more sustainable and safer outcomes.
Further steps include simulating current and prospective practices and arrangements, comparing outcomes, and assessing necessary adaptations from stakeholders and institutions. Participants could then select the combination of safe practices most suited to their context and co-design arrangements that enable adoption by accounting for both stakeholders’ and institutions’ capabilities, opportunities and motivations.
Conclusion
The multistakeholder dialogue brought actors around the table who rarely meet. It clarified for all participants the misalignment between institutions’ strategies and farmers’ realities. To enhance hygiene and food safety, Departments of Agriculture and Health seek to motivate farmers to use piped water for irrigation. However, to improve cost recovery, the water utility implements tariffs that restrict farmers’ opportunity to use piped water. Moreover, the municipalities’ failure to enforce land use regulations limits farmers’ opportunities and motivation to secure alternative safe water resources. These institutional barriers threaten farmers’ livelihoods, undermine food safety and perpetuate public health risks. Participants suggest collaboration between the water utility, the regulatory authority, the irrigation authority, and their respective parent ministries to offer farmers an affordable rate.
Such open discussion allows participants to diagnose stakeholders’ capability, opportunity, and motivation needs more extensively than interviews. Combining the COM-B framework with the Companion Modelling participatory approach enables participants to bridge knowledge gaps caused by sectoral silos and top-down approaches, broadening discussion beyond generic needs for education and training, incentives and sanctions. Together, stakeholders can identify actors, resources and interactions previously overlooked, leading to participants developing a shared understanding of each other’s perspectives. The dialogue makes explicit how institutions’ objectives and constraints may hinder stakeholders’ opportunity and motivation to adopt safe practices. Such open discussion lays the foundation for co-designing arrangements that better account for stakeholders’ and institutions’ capabilities, opportunities and motivations.
Acknowledgments
We are grateful to all the participants for their time, expertise, and trust. We would also like to express our thanks to Dr. Pay Drechsel and Dr. Olufunke Cofie from the International Water Management Institute (IWMI), and the CGIAR initiative on Resilient Cities for their support.
References
- 1.
WHO, UNICEF. Joint Monitoring Programme. 2022 [cited 2025 Apr 19]. JMP - Household data - Income - Urban - 2022 - Service levels. Available from: https://washdata.org/data/household#!/dashboard/new
- 2. Mills F, Foster T, Moe C, Amin N, Liu P, Rahman M, et al. Unsafe containment: Public health risks of septic tanks discharging to drains in Dhaka Bangladesh. PLOS Water. 2024;3(12):e0000325.
- 3. Steele M, Odumeru J. Irrigation water as source of foodborne pathogens on fruit and vegetables. J Food Prot. 2004;67(12):2839–49. pmid:15633699
- 4.
UN-Habitat. Global Report on Sanitation and Wastewater Management in Cities and Human Settlements [Internet]. Nairobi, Kenya: United Nations Human Settlements Programme; 2023 [cited 2024 Nov 21]. 192 p. Available from: https://www.un-ilibrary.org/content/books/9789211329001
- 5. Andersson Djurfeldt A. Urbanization and linkages to smallholder farming in sub-Saharan Africa: Implications for food security. Global Food Security. 2015;4:1–7.
- 6.
World Bank Group. www.databank.worldbank.org. 2017 [cited 2024 Dec 6]. Food Prices for Nutrition. Available from: https://databank.worldbank.org/source/food-prices-for-nutrition#advancedDownloadOptions
- 7. Orsini F, Kahane R, Nono-Womdim R, Gianquinto G. Urban agriculture in the developing world: a review. Agron Sustain Dev. 2013;33(4):695–720.
- 8. Coulibaly PZZ, Dongo K, Christoph L. Multi-pathway assessment of fecal contamination in urban areas of Abidjan: The case of Abobo municipality. PLOS Water. 2023;2(6):e0000074.
- 9.
SaniPath. SaniPath. 2025 [cited 2025 Apr 19]. Results Dashboard - Multi-City Exposure Results. Available from: https://www.sanipath.net/results-dashboard
- 10. Adegoke AA, Amoah ID, Stenström TA, Verbyla ME, Mihelcic JR. Epidemiological Evidence and Health Risks Associated With Agricultural Reuse of Partially Treated and Untreated Wastewater: A Review. Front Public Health. 2018;6:337. pmid:30574474
- 11. Beuchat LR. Vectors and conditions for preharvest contamination of fruits and vegetables with pathogens capable of causing enteric diseases. British Food Journal. 2006;108(1):38–53.
- 12.
Jaffee S, Henson S, Unnevehr L, Grace D, Cassou E. The Safe Food Imperative: Accelerating Progress in Low- and Middle-Income Countries [Internet]. Washington, DC, USA: World Bank; 2019 [cited 2023 June 23]. Available from: http://hdl.handle.net/10986/30568
- 13.
WHO. WHO estimates of the global burden of foodborne diseases: foodborne disease burden epidemiology reference group 2007-2015 [Internet]. Geneva, Switzerland: World Health Organization; 2015 [cited 2021 June 26]. Available from: https://apps.who.int/iris/handle/10665/199350
- 14.
Osei L. Western Region bans vegetable sales amid Cholera outbreak [Internet]. Citi Newsroom. 2024 [cited 2025 Feb 5]. Available from: hhttps://citinewsroom.com/2024/12/western-region-bans-vegetable-sales-amid-cholera-outbreak/
- 15. Unnevehr L. Food safety in developing countries: Moving beyond exports. Global Food Security. 2015;4:24–9.
- 16. Vipham JL, Amenu K, Alonso S, Ndahetuye JB, Zereyesus Y, Nishimwe K. No food security without food safety: Lessons from livestock related research. Global Food Security. 2020;26:100382.
- 17. Mara D, Evans B. The sanitation and hygiene targets of the sustainable development goals: scope and challenges. Journal of Water, Sanitation and Hygiene for Development. 2017;8(1):1–16.
- 18.
WHO. Guidelines for the safe use of wastewater excreta and greywater. 3rd edn. Vols 1–4. Geneva, Switzerland: World Health Organization; 2006. 115 p.
- 19. Drechsel P, Qadir M, Galibourg D. The WHO Guidelines for Safe Wastewater Use in Agriculture: A Review of Implementation Challenges and Possible Solutions in the Global South. Water. 2022;14(6):864.
- 20. Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6:42. pmid:21513547
- 21. Galibourg D, Scott RE, Gough KV, Drechsel P, Evans BE. Effectiveness of behaviour change interventions to reduce the risk of faecal contamination in urban irrigated vegetable value chains – applying the COM-B behavioural framework. Journal of Water, Sanitation and Hygiene for Development. 2024;14(8):654–69.
- 22. Dewi D, Aytekin D, Schneider KR, Covic N, Fanzo J, Nordhagen S. Defining and measuring policy coherence for food system transformation: A scoping review. Global Food Security. 2024;43:100803.
- 23. Galibourg D, Amankwaa EF, Gough KV, Scott R. Informal irrigated vegetable value chains in urban Ghana: potential to improve food safety through changing stakeholder practices. International Development Planning Review. 2024;46(4):391–414.
- 24.
GSS. Ghana 2021 Population and housing census - volume 3M - Water and Sanitation. Accra, Ghana: Ghana Statistical Service; 2022 Feb p. 86.
- 25. Sagoe G, Danquah FS, Amofa-Sarkodie ES, Appiah-Effah E, Ekumah E, Mensah EK, et al. GIS-aided optimisation of faecal sludge management in developing countries: the case of the Greater Accra Metropolitan Area, Ghana. Heliyon. 2019;5(9):e02505.
- 26. Fianko JR, Korankye MB. Quality characteristics of water used for irrigation in urban and peri-urban agriculture in greater accra region of Ghana: health and environmental risk. West African Journal of Applied Ecology. 2020;28(1):131–43.
- 27.
Marras S, Bendech MA, Laar A. Street Food Vending in Accra, Ghana: Field Survey Report 2016 [Internet]. Rome, Italy: FAO; 2016 [cited 2024 Feb 9] p. 57. Available from: https://www.fao.org/documents/card/en/c/b9668d2d-e1cf-4546-a64b-c9bda8dfb6d2/
- 28. Puppim de Oliveira JA, Ahmed A. Governance of urban agriculture in African cities: Gaps and opportunities for innovation in Accra, Ghana. Journal of Cleaner Production. 2021;312:127730.
- 29. Robb K, Null C, Teunis P, Yakubu H, Armah G, Moe CL. Assessment of Fecal Exposure Pathways in Low-Income Urban Neighborhoods in Accra, Ghana: Rationale, Design, Methods, and Key Findings of the SaniPath Study. Am J Trop Med Hyg. 2017;97(4):1020–32. pmid:28722599
- 30. Raj SJ, Wang Y, Yakubu H, Robb K, Siesel C, Green J, et al. The SaniPath Exposure Assessment Tool: A quantitative approach for assessing exposure to fecal contamination through multiple pathways in low resource urban settlements. PLoS One. 2020;15(6):e0234364. pmid:32530933
- 31.
SaniPath. SaniPath. 2020 [cited 2024 Jan 22]. SaniPath Results Dashboard - Distribution of Behavior by City (combined neighborhoods) for Selected Pathways (Street food). Available from: https://www.sanipath.net/results-dashboard
- 32. Ajzen I. The theory of planned behavior. Organizational Behavior and Human Decision Processes. 1991;50(2):179–211.
- 33. Rosenstock IM. Why people use health services. The Milbank Memorial Fund Quarterly. 1966;44(3):94–127.
- 34.
Cane J, O’Connor D, Michie S. Validation of the theoretical domains framework for use in behaviour change and implementation research. Implement Sci. 2012;7:37. https://doi.org/10.1186/1748-5908-7-37 22530986
- 35.
MacLeod C, Davies K, Mwenge MM, Chipungu J, Cumming O, Dreibelbis R. Household behaviour change interventions to improve sanitation and hygiene practices in urban settings: a scoping review [Internet]. 2024 [cited 2024 Oct 19]. Available from: http://medrxiv.org/lookup/doi/10.1101/2024.08.20.24312313
- 36. Mosimann S, Ouk K, Bello NM, Chhoeun M, Vipham J, Hok L. Describing capability, opportunity, and motivation for food safety practices among actors in the Cambodian informal vegetable market. Front Sustain Food Syst. 2023;7:1060876.
- 37.
West R, Michie S, Chadwick P, Atkins L, Lorencatto F. Achieving behaviour change: A guide for national government [Internet]. London, UK: Public Health England; 2020 Nov p. 75. Available from: https://assets.publishing.service.gov.uk/media/5fa537c7d3bf7f03b249aa12/UFG_National_Guide_v04.00__1___1_.pdf
- 38. Barreteau O, Antona M, d’Aquino P, Aubert S, Boissau S, Bousquet F. JASS. 2003;6(1):n.p.
- 39.
Etienne M, Abrami G, Allier F, Arrault S, Barnaud C, Barreteau O, et al. La modélisation d’accompagnement: partager des représentations, simuler des dynamiques [Internet]. Étienne M, editor. Paris, France: INRA; 2015. 299 p. Available from: https://collaboratif.cirad.fr/alfresco/s/d/workspace/SpacesStore/0e6fecd7-f07e-4864-8a5c-46157b146793?attach=true
- 40. Etienne M, Du Toit DR, Pollard S. ARDI: A Co-construction Method for Participatory Modeling in Natural Resources Management. E&S. 2011;16(1).
- 41. Amankwaa EF, Gough KV. Everyday contours and politics of infrastructure: Informal governance of electricity access in urban Ghana. Urban Studies. 2022;59(12):2468–88.
- 42.
GWCL. Proposals for review of aggregate revenue requirement and tariff. Accra, Ghana: GWCL; 2022 Apr p. 70.
- 43.
USAID. Improving water quality management, water equity, and non-revenue water in Ghana - Component 3 Report: Non-Revenue Water [Internet]. Washington, D.C., U.S.A: USAID-URBAN WASH; 2024 Apr [cited 2024 Nov 15] p. 52. (Improving water quality management, water equity, and non-revenue water in Ghana). Available from: https://www.globalwaters.org/sites/default/files/ghana_buy-in_component_3_report_non-revenue_water_final_508.pdf
- 44.
Drechsel P, Keraita B, editors. Irrigated urban vegetable production in Ghana: characteristics, benefits and risk mitigation [Internet]. 2nd edn. Colombo, Sri Lanka: International Water Management Institute (IWMI); 2014 [cited 2022 Jan 31]. 247 p. Available from: http://csirspace.csirgh.com:80/handle/123456789/217
- 45. Yeboah E, Shaw DP. Customary land tenure practices in Ghana: examining the relationship with land-use planning delivery. IDPR. 2013;35(1):21–39.
- 46.
PURC. PURC. n.d. [cited 2024 Nov 19]. Definition of Water Customer Categories. Available from: https://www.purc.com.gh/categ/resources/subcategories/definition-of-water-customer-categories
- 47.
MoFA. Planting for Food and Jobs Phase II (PFJ 2.0) [Internet]. 2023 [cited 2025 Jan 10]. Available from: https://mofa.gov.gh/site/pfj-2
- 48.
GWCL. GWCL.com.gh. 2025 [cited 2025 Jan 8]. Company Profile, Vision, Mission, Core Values and Core Business. Available from: michie 2014
- 49. Termeer K, Dewulf A, Biesbroek R. Three archetypical governance pathways for transformative change toward sustainability. Current Opinion in Environmental Sustainability. 2024;71:101479.
- 50. Moustier P, Holdsworth M, Anh DT, Seck PA, Renting H, Caron P, et al. The diverse and complementary components of urban food systems in the global South: Characterization and policy implications. Global Food Security. 2023;36:100663.
- 51. Godt S, Mhatre S, Schryer-Roy A-M. The change-makers of West Africa. Health Res Policy Syst. 2017;15(Suppl 1):52. pmid:28722548
- 52.
Hopper M, Boutrif E. Strengthening national food control systems: a quick guide to assess capacity building needs [Internet]. Rome: Food and Agriculture Organization of the United Nations; 2007. 45 p. Available from: https://openknowledge.fao.org/server/api/core/bitstreams/43e6a7b0-5dd2-4afc-b7ba-b8d4d7878214/content
- 53. Funtowicz SO, Ravetz JR. Science for the post-normal age. Futures. 1993;25(7):739–55.
- 54. Barnaud C, Van Paassen A. Equity, power games, and legitimacy: dilemmas of participatory natural resource management. Ecol Soc. 2013;18(2):21.
- 55. Rittelmeyer P, Lubell M, Hovis M, Heikkila T, Gerlak A, Pozzi T. Knowledge is not power: Learning in polycentric governance systems. Review of Policy Research. 2024;42(3):444–65.
- 56. Nordhagen S, Lambertini E, DeWaal CS, McClafferty B, Neufeld LM. Integrating nutrition and food safety in food systems policy and programming. Global Food Security. 2022;32:100593.
- 57. Wainaina GK, Truffer B, Lüthi C. The role of institutional logics during participation in urban processes and projects: Insights from a comparative analysis of upgrading fifteen informal settlements in Kenya. Cities. 2022;128:103799.
- 58. Reyes Tejada N, Kooy M, Zwarteveen M. Farming and the city: the changing imaginary of the city and Maputo’s irrigated urban agriculture from 1960 to 2020. Water International. 2024;49(2):185–200.
- 59. Veldwisch GJ, Amerasinghe P, Letema S, Wessels MT. The practices and politics of irrigated urban agriculture. Water International. 2024;49(2):129–43.