J Urban Health (2021) 98:394–403
https://doi.org/10.1007/s11524-021-00531-4
Loops and Building Blocks: a Knowledge co-Production
Framework for Equitable Urban Health
Camilla Audia & Frans Berkhout & George Owusu &
Zahidul Quayyum & Samuel Agyei-Mensah
Accepted: 9 February 2021 / Published online: 18 March 2021
# The Author(s) 2021
Abstract This paper sets out a structured process for and engagement practices, which we call building
the co-production of knowledge between researchers blocks. Building blocks are activities and interaction-
and societal partners and illustrates its application in an based methods aimed at bringing together a range of
urban health equity project in Accra, Ghana. The main participants involved in joint knowledge production. In
insight of this approach is that research and knowledge practice, recursive iterations within loops may be limit-
co-production is always partial, both in the sense of ed due of constraints on time, resources, or attention.
being incomplete, as well as being circumscribed by We suggest that co-production loops and building
the interests of participating researchers and societal blocks are deployed flexibly.
partners. A second insight is that project-bound societal
engagement takes place in a broader context of public Keywords Urbanhealth .Co-productionofknowledge .
and policy debate. The approach to co-production de- Policy . Impact . Equity
scribed here is formed of three recursive processes: co-
designing, co-analysing, and co-creating knowledge.
These ‘co-production loops’ are themselves iterative,
Introduction
each representing a stage of knowledge production.
Each loop is operationalized through a series of research
Health equity research highlights how economic, social,
and health policy choices may result in uneven health
C. Audia (*) : F. Berkhout outcomes across populations [1]. This is particularly
Department of Geography, School of Global Affairs, Faculty of
Social Science and Public Policy, King’s College London, Strand, visible in cities, where differential exposure to health
London WC2R 2LS, UK risks is compounded by pre-existing and interconnected
e-mail: camilla.audia@kcl.ac.uk vulnerabilities of urban populations. Despite increasing
understanding about health in cities and its relationship
G. Owusu
Institute of Statistical, Social and Economic Research (ISSER), with risks, such as social exclusion, poverty, housing,
University of Ghana, Accra, Legon, Ghana sanitation, and environmental quality, there continues to
be a gap between available and actionable knowledge,
resulting in decision-makers not acting on best evi-
Z. Quayyum
James P Grant School of Public Health, BRACUniversity, Dhaka, dence, even where policy objectives are agreed. How
Bangladesh can research and policy evolvemore closely, resulting in
options that are useful, usable, and used by decision-
makers? This paper presents a methodological frame-
S. Agyei-Mensah
Department of Geography and Resource Development, University work, drawing on social science literatures, for co-
of Ghana, Accra, Legon, Ghana production of knowledge between researchers and
Loops an d  Building Blocks: a Knowledge co-Production Framework for Equitable Urban Health 395
societal partners in the production of actionable knowl- concern with knowledge production as an active social
edge, aimed at contributing to more equitable urban process set in institutional and political contexts has
health. remained.
Generating information and knowledge that is sa- Approaches to co-production range from normative
lient, timely, and understandable to decision-makers to descriptive [5, 6]. We aim to contribute to the de-
requires transdisciplinary dialogue between researchers scriptive approach [7], which understands co-
from different academic disciplines and with decision- production as an interactive and complex process in
makers—including people who make choices in a vari- which disciplines, practices, and knowledge systems
ety of social contexts, from the household to govern- can confront, shape, and be shaped by each other,
ment [2]. These dialogues start from an assumption that whether by conflict or by cooperation. Diverse perspec-
professional researchers are not the only producers of tives and understandings are reconfigured to generate
useful and legitimate knowledge. Many societal actors new, transdisciplinary knowledge. In this process of
produce knowledge that is valid and useful [3]. Knowl- interaction, new knowledge is intertwined with social,
edge claims generated by scientific research is one form cultural, and political practices [5].
of knowledge that is widely distributed among social Given its mixed intellectual heritage, there is no
actors. The co-production of knowledge recognizes this unifying definition of knowledge co-production in the
distributed nature of knowledge and its production and literature. We use the term co-production to mean a set
organizes interactions between researchers and societal of specific processes and practices that structure and
actors to encourage shared understanding of what is organize the complex, interactive relationships between
known and what can be done. We propose a way of science, society, and policy. The benefits of co-
framing the co-production of knowledge. production are clear and well-described, and include
This paper has four sections. First, we set out core better quality of research, ownership and buy-in, ac-
themes and concepts from the broad literature on co- countability, empowerment, inclusion, and usability of
production of knowledge. Second, we describe a ‘loops knowledge [8–12].
and building blocks’ framework for the practice of Co-production has been explored in the field of pub-
engagement between research, policy, and practice to lic health, intersecting with a literature about ‘knowl-
generate valid and salient knowledge for action. Third, edge translation’ [13]. It has evolved alongside concepts
drawing on research in Accra, Ghana, we describe and of user participation and user input, and is often de-
assess co-production activities carried out in the Path- scribed as a way of working with patients, citizens,
ways for Equitable Healthy Cities project. Finally, we and organizations to design services that are more
consider how co-production may be evaluated. people-centred and user-led [8]. The value of co-
production in health care has been widely explored
[14] and there is broad acceptance of the value of
Why co-Production? Key Themes and Core processes involving multiple actors including patients,
Concepts clinicians, managers, and carers aimed at jointly-
creating knowledge, evaluating practice, or designing
The term ‘co-production of knowledge’ has emerged in strategies [15–17].
a number of fields over the last 30 years, including Our framework is concerned with learning, which
public administration, science and technology studies has emerged as a leading theme in STS co-production
(STS) and sustainability science [1], each developing a literature. Learning is seen as a long-term, evolving
particular set of concerns and agendas. Elinor Ostrom, process, rather than a single event at a discrete mo-
in her work on the provision of urban services like ment. Co-production practices aim to create an envi-
policing, defined co-production as a process by which ronment in which learning occurs through dialogue,
inputs are transformed by individuals not in the same individual and joint reflection, and separate and joint
organization into goods and services [2]. The term was experimentation with solutions [18, 19]. The process
adopted by constructivist STS scholars in the 1980s as of active, structured dialogue points towards knowl-
they sought to analyse the historical interplay between edge that is socially robust [20, 21]. This entails a
science and society [3, 4]. Since then, co-production has shift to a conception of knowledge production as an
been used in a variety of ways, but the underlying open process, with multiple actors agreeing new
396 Audia et al. 
knowledge claims and how they can affect action project team was therefore challenged to develop a
[22]. conceptual model and practical approach to knowledge
Co-production represents a multiple collaboration: co-production across contexts, building on previous
between the sciences, between the different decision- examples.
makers, and between science and societal actors [23, The ‘loops and building blocks’ framework aims to
24]. This collaboration takes effort and, to participate, structure more inclusive forms of engagement, knowl-
actors expect their effort to be rewarded. But the inter- edge production, and governance [28, 30] taking ac-
ests and incentives for will differ between participants. count of practitioner-oriented literature and experience
These differentiated incentives further complicate co- in transdisciplinary projects [27, 31–33]. The frame-
production, with the critique of ‘extractive’ research also work defines co-production as including recursive three
being a concern for the conduct of engagement and loops: co-design, co-analysis, and co-creation (see
dialogue [25]. Fig. 1). Each loop represents a different stage of knowl-
In practice, not all knowledge claims are held to have edge production. A first stage involves identifying part-
equivalent value. A process of co-production often chal- ners and mapping their networks, and joint framing of
lenges existing hierarchies of value and esteem in problems and choices about the focus of research and
knowledge systems. Knowledge is constructed against action and agreement about common research and ac-
a background of social and institutional power relations tion objectives. This co-design loop allows early in-
and cultural factors [3]. Each knowledge claim draws on volvement of relevant societal partners and aims at
experience, assumptions, and expertise [26]. Since trust-building, engagement and buy-in, open collabora-
knowledge is socially embedded, co-production as a tion, and ultimately to project outputs becoming under-
participatory process involving expert, practitioner, stood and influential to starting assumptions, choices,
and lay knowledges may challenge conventional struc- and decisions. The second stage (co-analysis loop) in-
tures of knowledge production. This can entail tensions volves an agreed analytical strategy, the separate or joint
and conflicts. Co-production processes may therefore collection of evidence, and analytical procedures, in-
provide a space in which these tensions surface, are cluding modelling. Researchers and societal partners
acknowledged, and, sometimes, resolved [27, 28]. Deal- may work separately. Validation is at the core of this
ing with tensions is also a means of building trust. The loop: through an exchange of information, researchers
production of knowledge, its appropriation, use, and and partners validate each other’s work to achieve
misuse are processes set in gradients of power between agreed knowledge claims in formats that are fitted to
organizations and people. New forms of information policy and practice design and implementation. Co-
and knowledge governance acknowledging power rela- design and co-analysis are connected as new informa-
tions are needed that mediate such challenges to existing tion or priorities may arise, leading to the identification
systems of knowledge production [29]. of new partners and additional validation steps. The
team will also jointly develop options and policy sce-
narios which identify interventions and actions to be
A Framework for co-Production: Loops tried out and evaluated [34]. The final co-creation loop
and Building Blocks includes implementation of policies, strategies, and in-
terventions, typically in an experimental form, and their
The Pathways to Equitable Healthy Cities project (Path- evaluation. This process may also lead to new knowl-
ways) is a global partnership that aims to improve pop- edge and new, co-produced ideas for further research,
ulation health, health equity, and environmental sustain- leading to a new co-design loop.
ability in cities through knowledge co-production with In a situation of unconstrained time, resources, and
policy and civil society partners in cities in five coun- commitment, it would be possible to move through each
tries (Accra, Tamale, Beijing, Dhaka, London, Vancou- loop and between loops sequentially, recursively, and
ver). The project aims to produce evidence on how carefully. But research and policy projects are typically
urban change and development can be shaped and man- constrained in time and resources. Moreover, knowl-
aged to bring positive impacts to population health and edge production in real-life action-oriented situations is
health equity. The project works in cities of widely messy, fluid, and contingent. Things change, partly in
different economic, social, and health profiles. The response to emerging findings and societal engagement.
Loops an d  Building Blocks: a Knowledge co-Production Framework for Equitable Urban Health 397
A flexible framework, able to adapt to unfolding events Many research projects last three to five years and
and opportunities, is therefore appropriate. Each loop is may be carried out by cross-country consortia entailing
important, but the capacity to work through each one different cultures, contexts, and languages and
recursively may be constrained. Viewing co-design, co- encompassing different partners (NGOs, government
analysis, and co-creation as a fixed sequence therefore bodies, private sector, civil society representatives,
seems too rigid. Instead, we see each loop as creating a etc.). Embedding a co-production framework in such
new entry point for co-produced knowledge, and the contexts has proved to be challenging and may lead to
time and effort placed in any given loop will vary unrealistic goals. However, project reviews suggest that
between projects and action contexts. In this sense, even single workshops, informal interactions, and feed-
knowledge co-production will always be partial: con- back in either direction can produce learning by re-
tingent on specific capacities and contexts, and therefore searchers and societal partners [32]. For this reason,
differing each time. we argue for a realistic approach, matching the resources
Given this freedom to apply comparable but varied and incentives of all participants in the research. We
approaches across the different contexts, projects need believe that a process of organized co-production needs
to recognize their own limits in time and resources to to be adaptive and flexible, picking up dynamically on
develop approaches that are both ambitious and realistic. opportunities as they arise during the flow of a project.
In making choices about the balance of effort, projects The aim should be to maximize inclusion of diverse and
need to develop a strategy for achieving impact. One of weak voices, and to challenge the dominance of specific
the greatest challenges for research is to provide knowl- perspectives, increasing epistemic diversity. This is in
edge claims in a timely way, aligned to the moments of line with most recent literature which places weight on
opportunity when new evidence can have an impact in systemic changes in the governance of knowledge pro-
policy and practice [35]. Such moments of opportunity duction [28–30].
are often outside the gift of researchers to influence and ‘Building blocks’ are the practical expression of
they rarely occur when research has a completed, peer- co-production loops. We define a building block
reviewed result. Moreover, the problem that needs a as any time-bound activity, principle, interaction-
solution may not match the problem which the research based method, or tool aimed at bringing together a
has sought to address. In the translation of co-produced range of participants involved in research to work
knowledge to its use and adoption by decision-makers together on a specific task: framing, analysing, or
and practitioners, there is usually a degree of judgement designing and evaluating interventions. Carter
and improvisation. et al. introduce the idea of building blocks in the
There is a second sense in which co-production of context of co-production of climate and weather
knowledge in a research project will be partial. Every services in Africa [27]. We have developed this
project is embedded in an existing social context. Orga- idea and embedded it in our framework. A build-
nized framing, analysis, and option generation that hap- ing block could be a workshop, a webinar, or
pens within a project (loops and building blocks) is collaboration with a government agency to secure
always against a background of on-going framing, anal- health or population data. We propose a modular
ysis, and experimentation elsewhere. Decision-makers construction of building blocks, activated in the
and practitioners draw on a stream of competing knowl- broader context of the three co-production loops,
edge claims in making choices and judgements [29]. which themselves are conceived of as operating
A recognition of the partial and embedded nature against a wider societal background of knowledge
of engagement organized within a research project production and action. Different building blocks
does not imply a constraint on impact. The loops would be appropriate to different loops, as shown
focus on distinct objectives in a broader process. in Table 1 below. Making loops and building
Achieving these objectives increases the likelihood blocks explicit helps address the question of com-
of impact as it creates buy-in by involving diverse parability across urban contexts and equitable,
actors, aligns outcomes and outputs to concrete health in cities, highlighting the ‘one-size does
needs, translates them into understandable languages, not fit all’ approach, while providing a framework
and ensures that they can be useful, usable, and by for comparative analysis and evaluation across di-
policy professionals and practitioners. vergent contexts.
398 Audia et al. 
Fig. 1 Conceptual framework for knowledge co-production developed for pathways.
Loops and Building Blocks in Practice Who Is co-Producing Knowledge? A Building Block
from Accra, Ghana
Co-production in research has benefits when used to
influence decision-makers. It is a time and resource- In the initial co-design phase of the project, we con-
intensive process that requires commitment from all vened two face-to-face workshops in Ghana in 2019.
involved parties. Because of this, the incentives for This followed more informal consultations with project
participation by researchers and partners need to be partners at the University of Ghana during project de-
clarified: what is in it for each side? [33]. The velopment. The two workshops were convened in Ac-
objective of Pathways is to provide timely, rigorous, cra, Ghana’s capital [36]. The first workshop, held in
and scientific evidence on urban change and devel- May 2019, was a context-setting session where partners
opment to evaluate how urban and health policies from the Accra Metropolitan Assembly, the main polit-
will impact the urban environment and population ical and administrative authority for the city, the large
health, through a lens of health equity. Influencing civil society organization People’s Dialogue on Human
urban planning to improve health equity requires a Sett lements (https:/ /www.pdghana.org/), and
focus on policies. This translates into interactions representatives from Ghana Statistical Service spoke
between researchers from different disciplines with about urban planning and health issues and priorities
policy professionals, government officials, and civil to researchers. Sixteen policy professionals and opinion
society organizations. Such an approach favours leaders were invited, and an open dialogue was
policy professionals and ‘elite’ stakeholders over organized in small groups involving fourteen
direct engagement with citizens and other decision- researchers collaborating in Pathways. Focussing on
makers, to prioritize inputs and influence over attentive listening by researchers, the session aimed to
national- and local-level policies affecting urban reverse the conventional relationship between
health. researchers, and policy makers and activists. Here the
Loops an d  Building Blocks: a Knowledge co-Production Framework for Equitable Urban Health 399
Table 1 Building block examples for each of the loops in pathways. Adapted from Carter et al. (2019).
Loop Building Blocks
Co-design - Spaces for dialogue (online tools; face-to-face interviews; informal talks) and building common ground
- Structured discussion (face-to-face workshops, online engagement meetings)
- Alternative ongoing debates monitoring (Twitter hashtag monitoring; newspapers/blogs reviews; informal 
relations with decision-makers)
- Space and time to recognise diverse knowledge systems as well as gender and cultural differences
- Spaces where listening is prioritised
- Flexibility in budgets, goals and allocations within the project to factor in sufficient time and resources to 
support the steps of co-production
Co-analysis - Separate but simultaneous processes of data collection and analysis / policy making
- Space and time to clearly map out roles and responsibilities in co-production
- Structured spaces for validation of results (face-to-face workshops, online engagement meetings)
- Unstructured, informal conversations (non-project meetings, conferences, “coffee” chats)
- Dialogue around ongoing policy initiatives and suggestions of interventions and strategies to incorporate 
analysis results into policies
- Potential for going “back” to co-design methods as new or different actors may be identified, or new or 
different data may be needed; it is key to maintain an unbiased and open agenda
Co-creation - Two-way, open and honest dialogue about policy options to be modelled in different scenarios
- Spaces for learning and understanding among different actors. These could be formal or informal, face-to-
face or online.
- Engagement to stimulate discussion and gain agreement of the group to ensure sustainability of the proposed 
solutions
- Potential for co-branding and ownership of the “final products” by the whole group
- Activities leading to capacity building of both researchers (for example in terms of engaging and 
communicating outputs) and policy professionals (for example to better understand data and how to integrate 
them in their daily work. This also ensures accessibility of the projects’ outputs.
societal partners were given time to elaborate on their developed at the International Centre for Tropical Agri-
own perspectives on the relationship between urban culture from about 2005. It can be adapted to different
development and health in Accra without any framing contexts and aims, and is designed to help the people
by researchers. involved in a project, program, or organization to make
The informal setting allowed policy makers and ac- explicit how they see themselves achieving their goals
tivists to tell rich stories about the rapid and extensive and impact [37]. The aim of this workshop was to build
growth of the city, the goals of the structure plan for the common ground and create a shared vision of chal-
Greater Accra Metropolitan Area, day-to-day problems lenges and opportunities for urban planning and health
of governance and administrative capacity, and the im- equity in Accra. PIPA engages societal partners in a
portant role played by citizen’s organizations and the structured participatory manner, allowing time for re-
private sector in shaping urban change. This provided flection and promoting learning.
perspectives for researchers of policy priorities, place In Accra, the team focused on how the project would
attachment, ethnic challenges, social dimensions, eco- develop its research outputs and who outside the project
nomic trends, and equity, and how these are discussed in would need to use them to achieve developmental out-
the context of contemporary Accra. This example is comes. In practice, this meant discussing specific fields
useful in showing that, while the act of listening may of urban development (transport, water, sanitation, air
seem like a one-way process, it can create a space for pollution, noise pollution, density, land tenure…), what
reflection and learning that is the basis for co-produc- theymeant for the policy professionals in the room, how
tion. It embodies perspectives, knowledge, and they were prioritized, and what the Pathways project
positionality in real people and places. could do to support policymaking in these areas. Over
For the second more intensive workshop, in October the course of the two days, participants were asked to
2019, we used the Participatory Impact Pathways Anal- design ‘problem trees’ to specify issues that Pathways
ysis (PIPA) approach to discuss issues, priorities, and researchers could, realistically, help to tackle in Accra.
ways forward in urban development and planning. PIPA Drawing from the more specific, easier to tackle issues,
is a project planning and research methodology, participants were then asked to envisage a short- to
400 Audia et al. 
medium-term vision of the city. In this visioning exer- Engaging with the Wider Context: Monitoring Twitter
cise, they were asked to keep the discussions realistic,
guided by a timeline of changes that could be achieved The Pathways project made a conscious choice early on
in the project’s time frame. to target policy professionals. However, co-production
This was followed by a network mapping session aims to be inclusive. In an urban context of equitable
seeking reflection on which actors, authorities, deci- health, this should also include citizens and people
sion-makers, policy professionals, and citizens should living or working in the city and its surroundings. To
be involved in making change happen, as well as address the gap in community-level engagement and to
existing or missing links between them. This discussion assess ongoing public and policy discourse, the project
created awareness among researchers of the complex began to monitor social media commentary on urban
networks of relevant actors and their multiple relation- development and health. Starting in Accra with housing
ships. The network of actors considerably extended the and health issues, we initiated a systematic and quasi-
scope and complexity of change to be envisaged from automated web-scraping system for monitoring Twitter
Pathways research, while also increasing the range of commentary using simple lines of code in Python to
potential partners for the project. Finally, the partici- collect data from Twitter’s application programming
pants were asked to articulate action plans to reflect on interface (API). Tweets are gathered and coded in qual-
the development trajectory of Accra and what identified itative analysis software to allow text, discourse, or
actors would need to do. Deep policy already exists in narrative analysis.
many of the domains that were mentioned (e.g., We assume that Twitter commentary partially re-
upgrading services in slums, land tenure, public trans- flects and could potentially shape how urban and health
port, waste management, and vehicle emissions), with problems are framed. We aim to investigate patterns of
participants agreeing that the real challenge is often to these debates in Accra, including their evolution
implement what is already on paper. This point about through time, summarizing them as an input for research
governance capacity confirmed the need for the project across the project and a background to communication
to be deeply engaged with the existing policy context, and impact-oriented work. This building block offers
including the potential for critical linking interventions the opportunity to identify other partners and more
to unlock implementation of plans. priorities that could feed in the researchers and policy
As a result of this workshop, Pathways priorities and professionals’ agenda and add a social dimension to our
workplans shifted, in particular by reallocating attention agenda, especially at a time when direct interactions
more towards the topic of housing and land tenure. with local people are made harder by the global
Researchers agreed to look into urban planning trade- pandemic.
offs and the provision of health-related services in dif-
ferent scenarios, as well as potential impacts on health
and equity. Researchers were later debriefed in person, Monitoring and Evaluating co-Production
via email, and via an online questionnaire and asked to
reflect on how this workshop had affected the way they Monitoring and evaluation of engagement in knowledge
were carrying out their work. These mini evaluations co-production presents many challenges. There are mul-
showed that the workshop encouraged participants to tiple interactions between project partners with a wide
reflect beyond the scope of the project and work to variety of purposes and encompassing diverse disci-
establish interpersonal and organizational links that plines and societal actors; linking each of these to even-
would be needed for longer-term impact. Researchers tual project outcomes will be difficult, also given the
also highlighted that it gave them an opportunity to existence of multiple alternative factors explaining these
create or strengthen networks and relationships with outcomes. Moreover, the problem of evaluation of
specific actors, government representatives, or NGO small-n, local, and experimental interventions is well-
workers who have since been closely involved with understood [38]. There is likely to be a balance between
working groups in the project. This exemplifies the conventional scientific notions of rigour in evaluation
modular conceptualisation of co-production and shows (randomized control trials, for instance) and more hy-
how different actors can add building blocks, giving the brid, evaluative approaches, including policymakers
process more depth and breadth. and practitioners.
Loops an d  Building Blocks: a Knowledge co-Production Framework for Equitable Urban Health 401
A monitoring and evaluation framework would ide- creating outputs that are relevant, usable, and use-
ally be continuous and in real time, allowing project ful by practitioners and decision-makers.
partners to learn and adjust as a project proceeds. Each Co-production is a process. Having established
loopmay invite an evaluation before moving on. In such that there can be no ‘one size fits all’ approach,
a process, learning is integrated in the project and eval- the framework we offer can be adjusted to the
uation becomes intrinsic to co-production loops and objectives, resources, and interests of research,
building blocks. Previous projects have identified and fitted to the needs of specific contexts and
multi-directional learning as raising the likelihood of social dynamics. By developing a monitoring,
success and sustainability of co-production processes evaluation, and learning framework, we seek to
and outcomes [27, 39]. For co-production to be encourage the application of successful co-
evaluable and effective, projects need built-in feedbacks production tools and approaches in new contexts
to allow for course adjustments [39]. The context of the at different scales. This largely depends on partic-
research should itself guide success criteria, reflecting ipants finding value in the process and integrating
the distributed nature of co-production, as well as the it further into their work but ultimately leads to
diverse interests of the actors. The co-production pro- sustainability.
cess is often overlooked by funders in performance It is important to keep in mind that a co-production
metrics. Similarly, it is often the case that researchers process is always intervening over a background of
lack incentives to carry out co-production and prioritize ongoing relationships and discourses that existed well
academic outputs to engagement. However, some pro- before the project and will persist beyond its comple-
cesses such as behavioural changes in scientists, trust tion. Moreover, it is also key to remember that co-
among actors across different disciplines, or changes in production demands commitment, time, and resources
hierarchies and relationships, can be assessed before the by people who engaged in it. To ensure continued buy-
co-produced product is developed by setting up mech- in and engagement, all participants, including re-
anisms such as key informant interviews over the course searchers, need incentives and rewards for these invest-
of the project. In Pathways, brief online Google Forms ments. If research is usually a collaborative effort, then
are set up for researchers across the project to keep track achieving change through new knowledge production in
of their interactions. This keeps the teams accountable research involves an extended form of collaboration,
for their outreach, but also allows our research to focus with shared benefits. This is effortful and will only be
on comparability across cities. effective if it achieves the right balance between being
planned, flexible, and opportunistic.
The COVID-19 pandemic presents both challenges
and opportunities to co-production processes and think-
Conclusions ing. The absence of face-to-face interactions may sig-
nificantly hinder co-creation processes aiming at creat-
This paper puts forward a practical and flexible ing common ground and building trust among different
‘loops and building blocks’ approach for co- parties. On the other hand, greater familiarity with vir-
production processes which draws on previous re- tual meetings and with more pervasive use of social
search in a diverse range of research fields and media, there are also new and more efficient modes of
aims to support engagement between researchers digital engagement and data collection. An adaptable
and societal partners. Co-production encompasses and modular loops-and-blocks framework can be flexi-
the complex process of formal and informal inter- ble to these new realities. There is an opportunity to
actions between researchers of different back- think beyond face-to-face engagement and explore new
grounds, disciplines, and institutions, and between opportunities for remote outreach.
researchers and policy professionals, decision-
makers, and other social actors. This paper offers Acknowledgments Funding for this research is by the Pathways
some practical ways of creating this awareness, to Equitable Healthy Cities project (Wellcome Trust, 209376/Z/
17/Z). We acknowledge the inputs of Meghan Winters at Simon
encouraging a flexible and adaptive approach in- Fraser University and our project teams and partners who attended
cluding ongoing monitoring and adjustment, the workshops and engaged in the Pathways to Equitable Healthy
supporting through potential obstacles, and Cities Co-Production Working Group.
402 Audia et al. 
Open Access This article is licensed under a Creative Commons 15. Dunston R, Lee A, Boud D, Brodie P, Chiarella M. Co-
Attribution 4.0 International License, which permits use, sharing, production and health system reform: from re-imagining to
adaptation, distribution and reproduction in anymedium or format, re-making. Aust J Public Adm. 2009;68:39–52.
as long as you give appropriate credit to the original author(s) and 16. Hawkins J, Madden K, Fletcher A,Midgley L, Grant A, Cox
the source, provide a link to the Creative Commons licence, and G, et al. Development of a framework for the co-production
indicate if changes were made. The images or other third party and prototyping of public health interventions. BMC Public
material in this article are included in the article's Creative Com- Health. 2017;17:689.
mons licence, unless indicated otherwise in a credit line to the 17. Kirkegaard S, Andersen D. Co-production in community
material. If material is not included in the article's Creative Com- mental health services: blurred boundaries or a game of
mons licence and your intended use is not permitted by statutory pretend? Sociol Health Illn. 2018;40:828–42.
regulation or exceeds the permitted use, you will need to obtain 18. Pelling M, High C, Dearing J, Smith D. Shadow spaces for
permission directly from the copyright holder. To view a copy of social learning: a relational understanding of adaptive capac-
this licence, visit http://creativecommons.org/licenses/by/4.0/. ity to climate change within organisations. Environ Plan A.
2008;40:867–84.
19. Pahl-Wostl C. A conceptual framework for analysing adap-
References tive capacity and multi-level learning processes in resource
governance regimes. Glob Environ Change. 2009;19:354–
1. Miller CA, Wyborn C. Co-production in global sustainabil- 65.
ity: histories and theories. Environ Sci Pol. 2018;113:88–95. 20. Gibbons M. Science’s new social contract with society.
https://doi.org/10.1016/j.envsci.2018.01.016. Nature. 1999;402:C81–4.
2. Ostrom V, Ostrom E. A theory for institutional analysis of 21. Nowotny, H. Re-thinking science: from reliable knowledge
common pool problems. Manag Commons. 1977;157 to socially robust knowledge. Entangled Hist. Negot.
3. Latour B. Postmodern? No, simply amodern! Steps towards Univers. 14–31 (2003).
an anthropology of science. Stud Hist Philos Sci Part A. 22. Salter KL, Kothari A. Knowledge ‘translation’ as social
1990;21:145–71. learning: negotiating the uptake of research-based knowl-
4. Jasanoff S. Beyond epistemology: relativism and engage- edge in practice. BMC Med Educ. 2016;16:76.
ment in the politics of science. Soc Stud Sci. 1996;26:393– 23. Sharp, P. A. et al. The third revolution: the convergence of
418. the life sciences, physical sciences, and engineering. Mass.
5. Jasanoff, S. States of knowledge: the co-production of sci- Inst. Technol. (2011).
ence and the social order. London, UK: Routledge; 2004. 24. Dedeurwaerdere T. Transdisciplinary sustainability science
6. Latour, B. & Woolgar, S. Laboratory life: the construction at higher education institutions: science policy tools for
of scientific facts. Los Angeles, CA: Princeton University incremental institutional change. Sustainability. 2013;5:
Press; 2013. 3783–801.
7. Bremer S, Meisch S. Co-production in climate change re- 25. Chu KM, Jayaraman S, Kyamanywa P, Ntakiyiruta G.
search: reviewing different perspectives: co-production in Building research capacity in Africa: equity and global
climate change research. Wiley Interdiscip Rev Clim health collaborations. PLoS Med. 2014;11:e1001612.
Chang. 2017;8:e482. 26. Stenmark, D. Information vs. knowledge: the role of intra-
8. Filipe A, Renedo A, Marston C. The co-production of what? nets in knowledge management. in Proceedings of the 35th
Knowledge, values, and social relations in health care. PLoS Annual Hawaii International Conference on System
Biol. 2017;15:e2001403. Sciences 928–937 (IEEE Comput. Soc, 2002). doi:
https://doi.org/10.1109/HICSS.2002.994043.
9. Barber R, Beresford P, Boote J, Cooper C, Faulkner A.
27. Carter, S., Steynor, A., Vincent, K., Visman, E., &
Evaluating the impact of service user involvement on re-
Waagsaether, K. (2019). Co‐production of African weather
search: a prospective case study. Int J Consum Stud.
and climate services. Manual. Cape Town: Future Climate
2011;35:609–15.
for Africa and Weather and Climate Information Services
10. Ghate D. Developing theories of change for social for Africa. Retrieved from: https://futureclimateafrica.
programmes: co-producing evidence-supported quality im- org/coproduction-manual
provement. Palgrave Commun. 2018;4:90. 28. West S, van Kerkhoff L, Wagenaar H. Beyond “linking
11. Edwards PN. Knowledge infrastructures for the knowledge and action”: towards a practice-based approach
Anthropocene. Anthr Rev. 2017;4:34–43. to transdisciplinary sustainability interventions. Policy Stud.
12. Shippee ND, Domecq Garces JP, Prutsky Lopez GJ, Wang 2019;40:534–55.
Z, Elraiyah TA, Nabhan M, et al. Patient and service user 29. Decieux JPP. How much evidence is in evidence-based
engagement in research: a systematic review and synthe- policymaking: a case study of an expert group o the
sized framework. Health Expect. 2015;18:1151–66. European Commission. Evid Policy. 2020;16:45–63.
13. Kitto SC, Sargeant J, Reeves S, Silver I. Towards a sociol- 30. NorstromA, et al. Principles for knowledge co-production in
ogy of knowledge translation: the importance of being dis- sustainability research. Nat Sustain. 2020;3:182–90.
interested in knowledge translation. Adv Health Sci Educ. 31. Visman, E. Knowledge is power: unlocking the potential of
2012;17:289–99. science and technology to enhance community resilience
14. Brandsen T, Honingh M. Distinguishing different types of through knowledge exchange. (2014). http://www.odihpn.
coproduction: a conceptual analysis based on the classical org/hpn-resources/network-papers/knowledge-is-power-
definitions. Public Adm Rev. 2016;76:427–35. unlocking-the-potential-for-science-and-technology-to-
Loops an d  Building Blocks: a Knowledge co-Production Framework for Equitable Urban Health 403
enhance-community-resilience-through-knowledge-exchange. 36. Agyei-Mensah S, Aikins A, de-Graft. Epidemiological tran-
Accessed 21 March 2019. sition and the double burden of disease in Accra, Ghana. J
32. Audia, C. et al. Investigating resilience at local, organisational Urban Health. 2010;87:879–97.
and institutional levels: a methodological note. (2016). 37. Douthwaite, B. et al. Participatory impact pathways analysis
http://www.braced.org/resources/i/Investigating-resilience-at- (PIPA) and research priority assessment. Prioritizing Agric.
local-organisational-institutional-levels. Accessed 5 Sept 2019. Res. Dev. 8–24 (2009).
33. Visman, E. et al. Underpinning principles and ways of 38. Sutherland WJ, Shackelford G, Rose DC. Collaborating
working that enable co-production: Reviewing the role of with communities: co-production or co-assessment? Oryx.
research. (2018). http://www.braced.org/resources/i/ways- 2017;51:569–70.
of-working-enable-coproduction/. Accessed 6 Oct 2019. 39. Wall TU,MeadowAM, Horganic A. Developing evaluation
34. Lemos MC, Kirchhoff CJ, Ramprasad V. Narrowing the indicators to improve the process of coproducing usable
climate information usability gap. Nat Clim Chang. climate science. Weather Clim Soc. 2017;9:95–107.
2012;2:789–94.
35. Oliver K, Lorenc T, Innvær S. New directions in evidence- Publisher’s Note Springer Nature remains neutral with regard to
based policy research: a critical analysis of the literature. jurisdictional claims in published maps and institutional
Health Res Policy Syst. 2014;12:34. affiliations.