Safeguarding concerns, practices, and resources in COVID-19 mutual aid groups

ABSTRACT Community mutual aid groups that sprang up during the COVID-19 pandemic were characterized by a lack of formal regulation and therefore potential issues around safeguarding. In this study, eight organizers of COVID mutual aid groups in the UK describe their group’s concerns, and existing safeguarding resources, and resources they would find useful in the future. Groups often created their own policies, training, reference materials, and safeguarding roles as a way of implementing safeguarding measures. Interviewees expressed three types of concerns around safeguarding: protection from harm (of both volunteers and members of the public), retaining the character of mutual aid, and making appropriate referrals. Solutions suggested by interviewees included educational materials, training, funding, consultation, a signposting database, and a forum for COVID mutual aid groups. Some of these needs could be provided by developing relationships with formal community groups. We suggest some of the mechanisms whereby informal community support groups can retain their grassroots identity and yet protect the health, wellbeing, and rights of volunteers and those they work with.


Introduction
When the COVID-19 pandemic began, over 4,000 "mutual aid" groups sprang up in the UK to provide support in local communities (Booth, 2020;Mao et al., 2021).Similar groups emerged in many other countries around the world (Sitrin & Sembrar, 2020).Mutual aid groups are defined as "self-organizing groups where people come together to address a shared health or social issue through mutual support" (Seebohm et al., 2013).Mutual aid is often distinguished from charity by its ethic of mutualism and is typified by the slogan "solidarity, not charity," erasing the traditional "hierarchical" divide between helper and helped in favor of "horizontalist" principles of democratic participation and reciprocity.Mutual aid groups often seek not only to provide goods and services to those in need, but to change the underlying social conditions which produce those needs: in this sense, it can constitute a form of political participation (Spade, 2020).However, conceptualizations of mutual aid vary even amongst participants in such groups, from those who see mutual aid as similar to traditional aid and adjunctive to government support, to those who see mutual aid as an ongoing support mechanism which should be separated from government structures (Littman et al., 2023).Here we use "mutual aid" as a broad term referring to groups which are characterized by their spontaneous and informal organization, local or neighborhood identity, and/or adherence to principles of mutual aid or solidarity (Mao, Drury, et al., 2021).
Safeguarding is defined as protecting the health, wellbeing and human rights of people at risk, enabling them to live safely, free from abuse and neglect (Care Quality Commission, 2022).People who are understood to be "at risk" include children and young people.It also includes adults who, because of issues such as dementia, mental illness, or learning disability, have support needs that render them more vulnerable to abuse or neglect (National Health Service England, 2024).Safeguarding practices include identification of abuse or neglect, assessment of mental capacity and whether a relevant individual is "at-risk" and escalating the case (or "referring") to the relevant organization (generally the local governmental authority) (Boland et al., 2014).
Participating in COVID mutual aid groups can expose volunteers to vulnerable individuals and to risky or ambiguous situations, including those involving potential abuse or neglect (Mao, Drury, et al., 2021).Without proper management of safeguarding issues, mutual aid groups run the risk of allowing abuse or neglect to continue.In cases where personal information is shared inappropriately or incorrect advice is provided, mutual aid groups may even end up causing harm to those they seek to support (Southern Brooks, n.d..).Therefore, if such groups intend to continue in the long term, safeguarding arrangements are necessary (Power & Benton, 2021).
There has been little research on COVID mutual aid groups' safeguarding practices (Benton & Power, 2021).Chevéé's (2021) study of a North London mutual aid group found that the group gradually developed a more formal structure and tried a range of documents on safeguarding.Tiratelli and Kaye (2020) describe the case of a mutual aid group who were helped by the local authority with logistical and other support, and who developed their own system for effective record-keeping, safeguarding, and data protection.There is a need to further understand COVID mutual aid groups' safeguarding concerns, and any potential gaps in resources which are preventing them from fulfilling this function optimally.The present paper describes an interview study in which these issues are explored with a sample of COVID mutual aid group organizers from across the UK, to understand the support needed for these groups' safeguarding requirements.

COVID mutual aid groups in the UK and safeguarding
Little is currently known about the extent to which mutual aid groups have prioritized safeguarding or utilized such measures in their work.By contrast, voluntary sector organizations with a formal constitution are required by UK law to have a safeguarding policy (The Charity Commission, 2022).But whilst some UK-based COVID mutual aid groups applied for charitable status (and hence were required to develop safeguarding policies), most did not (Power & Benton, 2021).Professionalization -salary, charity status, formal policieswas often understood as threatening the independence and the "grassroots" nature of such organizations, which were key to maintaining trust with local communities (Rendall et al., 2024).
An alternative to professionalization is to have relationships with more established community groups (Fernandes-Jesus et al., 2021).Indeed, these groups, as well as national umbrella organizations, have many resources such as training and guides that can help volunteers with safeguarding.However, the quality of COVID mutual aid groups' relationships with such groups is somewhat mixed.On the one hand, mutual aid organizers often see relationships with other organizations as crucial to obtaining various resources (Fernandes-Jesus et al., 2021).On the other hand, many groups have poor relationships with important local structures (such as local government) (Rendall et al., 2024).Another type of organization which mutual aid groups could collaborate with is Local Infrastructure Organizations: these are defined as formal organizations that provide advice and support to voluntary and community groups (National Association for Voluntary and Community Action, 2023).The extent to which mutual aid groups have relationships with such organizations is currently unclear.Prior research in UK social care contexts has shown that multi-agency relationships and information-sharing is a critical component of effective safeguarding and is associated with higher levels of referrals (Institute of Public Care, 2013).Therefore, from the perspective of organizations seeking to enhance their relationships with mutual aid groups, there is a need to gauge both mutual aid groups' knowledge of other community groups, as well as the quality of any existing relationships.This knowledge can then be used to inform better models of outreach and collaboration.

The present study
We interviewed eight organizers from different COVID mutual aid groups around the UK.Our first aim was to examine their safeguarding practices and any resources available to them.Second, we aimed to explore their concerns and needs.Third, we sought to examine any safeguarding resources they would find useful in the future.We use this analysis to suggests ways in which a fruitful collaboration between the mutual aid groups and established community groups could take place going forwards.

Materials and methods
This study was co-produced by researchers from the University of Sussex and the National Association for Voluntary and Community Action's (NAVCA) safeguarding lead.The University of Sussex team had previously researched factors that help sustain COVID mutual aid groups over time (Fernandes-Jesus et al., 2021;Mao, Drury, et al., 2021).NAVCA had already developed a website and resources for safeguarding.What was missing was an evidence-base for developing resources aimed at groups who were not part of NAVCA's network.While previous interviews conducted by the University of Sussex team contained some examples of safeguarding issues (Fernandes-Jesus et al., 2021), these studies did not look at the topic systematically.Therefore, the University of Sussex researchers and NAVCA team jointly designed the present study.

Participants
We conducted eight semi-structured interviews with coordinators of COVID mutual aid groups across the UK.We use "coordinator" and "organizer" interchangeably throughout, to mean individuals who took on greater responsibilities of decision-making, organizing, and leadership in the groups.We use "clients" to mean the recipients of services offered by the mutual aid group: in some cases, clients were also volunteers or organizers for the group.We specifically looked to recruit people with an organizing role, as such individuals were more likely than other volunteers to have been actively involved with the safeguarding process.We sought participants across a range of geographical locations to capture a variety of experiences and practices.
Participants were recruited through the existing networks of the researchers.We approached 26 individuals for interview: 18 either refused or did not respond to our solicitations, whilst eight agreed to be interviewed.Recruited interviewees belonged to groups from around the UK: Glasgow (2), Northumberland (1), Hampshire (1), West Sussex (1), London (1) and Derbyshire (1), and one group which requested that its location not be disclosed.Gender balance of participants was equal (four men and four women).We did not gather information on age.
We offered each participant a £20 voucher as compensation for their time, and interviews were conducted and recorded on the video-conferencing platform Zoom.

Interview schedule
The interview schedule covered questions about the group's activities (e.g., "What are the current aims of the group?"),concerns around safeguarding (e.g., "Has your group encountered any safeguarding concerns?"),existing resources and practices around safeguarding (e.g., "What resources does your group have?"),relationships with local infrastructure organizations (e.g., "Has your group been in contact with local infrastructure organizations?"), and areas where they would welcome support (e.g., "Would you welcome any guidance or advice from local infrastructure organizations on safeguarding?").See the project OSF site for the full interview schedule: https://osf.io/er7p4/?view_only= 4918659360c8487f98a892db17186911 We conducted the interviews between January and February 2022.Interviews ranged in length from approximately 40 minutes to an hour, with a mean of 45 minutes.The interviews were transcribed verbatim by a single professional transcriber.

Analytic approach
The transcribed interview material was analyzed using thematic analysis (Braun & Clarke, 2021).The analysis was guided by the five research aims described above (resources, practices, concerns, needs, areas where support would be welcome).
The analytic procedure was as follows: the first author read through the interview material, highlighting statements relevant to the research aims.Extracts were initially grouped together under five superordinate themes corresponding to each research aim, with extracts under each superordinate theme subdivided into subordinate themes.As discussions of "resources" and "practices" often overlapped with one another, these two themes were combined into one superordinate theme.A distinction continued to be drawn between "concerns" understood as immediate issues faced by mutual aid groups, and "needs" understood as more general deficits in resources.This led to four final superordinate themes, each with corresponding subordinate categories.

Results
The Results section is organized according to the four themes identified: "Existing Practices and Resources," "Concerns," "Needs," and "Resources which would be useful in future."

Existing practices and resources
Most of the participants we interviewed described some form of existing safeguarding process within their groups.All the groups demonstrated an awareness of the concept of safeguarding and the need to escalate potential safeguarding concerns to the relevant governmental authorities.However, despite this ostensible similarity at one level, there was variation in the measures implemented to address safeguarding priorities and, in the resources deployed to support this.Below, we outline a number of measures and resources which groups utilized to support safeguarding, from more common to less common approaches.

Policies
The most common measure to address safeguarding amongst our interviewees was the development and adoption of a written safeguarding policy which detailed correct practice surrounding safeguarding scenarios.These policies then served as internal resources for volunteers to refer to or were used as part of volunteer induction.Five of the interviewees reported having some form of written policy: three of these were explicit safeguarding policies which were badged as such; one was a policy for adults at risk of harm; and one was a handbook of procedures.In some cases, such policies were developed as a response to requests from third parties such as insurance providers.

Training
A second common measure was safeguarding training for group members: four coordinators reported having some form of training in their groups.Training covered a variety of topics and could be completed as part of an induction process.

Reference materials
Groups disseminated information on safeguarding through a variety of reference materials.This took a number of formats including handbooks, graphics, videos, scripts, and flowcharts.One organizer perceived well-made reference materials as superior to face-to-face training: We did a couple of, like, Zoom trainings for people and stuff, although by that point the, like, engagement with the more administrative side of things was, like, extremely low.But one really nice thing that my team-mate put together is she managed to, like, brief this graphic designer to create.like.a safeguarding graphic for us, which we then, like, put on WhatsApp a few times [. ..]Like a flow-chart.(Caitlin, London)

Specific safeguarding roles
Four groups dedicated specific team members to safeguarding management.Of these four: two groups created a "committee" or "case management team" of coordinators to discuss safeguarding cases; one allocated potential safeguarding cases to skilled group members such as social workers; and one nominated a safeguarding contact within the group to provide specific safeguarding support.This last group also installed a provision for anonymous reporting of concerns.

Skilled group members
Three groups mentioned having team members with transferable skills and experience such as teachers, charity workers, or community leaders.

Collaborating with other organizations
All the groups reported collaborating with other organizations.Three of our interviewees reported relationships with Local Infrastructure Organizations (LIOs), whilst the remaining five reported relationships with churches, charities, or local authorities.All the interviewees reported benefits from collaborating, including access to funding, physical facilities, or advice.Despite this, however, two of our interviewees also reported some concerns around larger organizations "taking over" or "pushing out" mutual aid groups.

Concerns
Coordinators' concerns related to safeguarding could be grouped into three sub-themes: potential harms (to either volunteers or clients); losing group identity; and complexities around referrals.

Potential harm being caused to clients
Many of the interviewees expressed a concern around volunteers potentially causing harm to their clients.The most common concern in this category, raised by three interviewees, was that clients could worsen a complicated situation by overextending themselves and substituting their own judgment for that of professionals: My aim with the policy was rather more about making sure that . . .people weren't just doing their own things, and sitting in judgements or taking things on themselves and saying "oh I'll sort that out for you and I'll have a word with your grandson about this, or I'll do this or that".(Rita, Derbyshire) Three interviewees were also concerned about the potential for the reverse to happen.Two gave specific examples of clients behaving inappropriately toward their volunteers: So we got one of our admin team to be one-to-one with [the client] because it was becoming a bit of a, a hassle and she started sending her abusive texts and all that.(Matt, Glasgow) Whilst safeguarding literature often focuses on situations where professional service providers react to potential neglect or abuse toward vulnerable service users, the informal nature of mutual aid means that many of the service providers themselves may be inexperienced or even vulnerable and therefore require measures to protect them.

Losing group identity
The group's identity being based on mutual aid principles (as distinct from conventional volunteering, charity, or professional services) was important to several interviewees.This commitment toward mutual aid principles led to a suspicion within some groups toward measures which could be perceived as "professionalizing" the group or imposing a more formal structure or regulations.However, some interviewees acknowledged that this commitment to informality could conflict with the implementation of safeguarding policies: the more we, we put that framework around it, the less it became mutual aid, you know it more became like structured volunteering and you know, almost like a staffmanagement relationship.(Dominic, Glasgow) This last quote illustrates the difficulty for mutual groups in balancing the two conflicting priorities.A robust safeguarding process may require a more formal or even hierarchical structure where coordinators possess oversight over the actions of volunteers; but this is seen as in conflict with mutual aid's egalitarian approach.Commitment to mutual aid principles also circumscribed the types of partnerships which such groups were prepared to enter: I think we hesitate to become more entangled with the council [local government] because of their data sharing stuff which we are not really comfortable with.A lot of the people we support are, like, NRPF [no recourse to public funds] or you know they've had bad experiences with the council and so I think the fact that we are not the council and are not too closely intertwined with the council makes people much more comfortable seeking support from us.(Violet, Oxford) Many of the individuals this group supports had "No Recourse to Public Funds" -a term for those in the UK who are unable to access state welfare due to their immigration status.Here we again see a principle of mutual aid (unconditional, universal provision of support) leading to conflict with the usual safeguarding practices of information sharing with authorities.Even those groups who did not prioritize mutual aid principles were wary of having their protocols and systems "taken over" by larger organizations, enough to render them wary of resources being offered which may have strings attached.

Complexities around referrals
Several interviewees noted complexities around the safeguarding referral process.Difficulties around privacy and protection of information were mentioned.One interviewee pointed out that making referrals to local governmental authorities often required a great deal of information, information their group could not handle properly without robust infrastructure: And it opens up a can of worms when you, like, take more information from people because then you have to protect it which means you need much more robust, like, infrastructure . . .and we don't have the capacity to put that in place, so we would rather not have the information.(Violet, Oxford) A second interviewee applied this requirement to protect privacy even to information which was shared internally: we would say to the volunteer we sent, "if you have a concern, obviously I mean we've got to be careful about what information we can share" . . .it's not fair for them to turn around and tell us all the, you know, what is essentially private and personal circumstances of an individual.(Bradley, West Sussex) These difficulties could have led to worsened safeguarding outcomes: in the former case, it may have led to fewer referrals or referrals not being accepted; in the latter, it may have led to volunteers not feeling like they could discuss their concerns or worries properly with other group members.
Relatedly, two interviewees raised concerns around the difficulty of maintaining a positive relationship with clients whilst also informing them of the need to share information, and potentially overriding their consent if it was deemed necessary.Volunteers in COVID mutual aid groups often had a close, occasionally preexisting, relationship with those they were helping.This could serve to steer volunteers away from safeguarding actions which could damage that relationship: Because we sort of toe the line as mutual aid people, as neighbour and friend and volunteer, like, I get the sense that with other organizations we've spoken to where they have like a befriending service, while those relationships are personal, there's sometimes, like, a line that they don't cross.Whereas in our situation there's, like, this tension where you're actually, like, mates with the person because they live down the road from you.(Caitlin, London) Finally, one interviewee pointed out that the safeguarding process was seen as a "drag" and considered to be burdensome "admin" by many volunteers.

Needs
Interviewees' safeguarding needs could be categorized in three ways: "Needs for knowledge," "Needs for funding/personnel," and "Needs for connections."

Needs for knowledge
Interviewees described several knowledge needs, both on an organizational level (i.e.how internal safeguarding processes could be optimized) and on an individual level (i.e.how given safeguarding cases should be managed).In terms of organizational knowledge, three interviewees felt that they required assurance around their policies being appropriate.One interviewee felt unsure about whether the policy they developed in consultation with a large charity was appropriate for a smaller organization: As a fairly informal tiny, tiny grassroots organization, like how much policy do we actually need or, like, how much is actually like common sense?So, obviously [charity name] could only give us the advice that is kind of, you know, they can tell us what they do, which is fairly formal, what we didn't quite know is, like, how do we, we did translate it for ourselves, but it's not technically clear on whether it was, like, overkill or not.(Caitlin, London) In terms of individual knowledge, three interviewees identified a need for more training.One interviewee further identified a need for a flowchart resource to address potential knowledge gaps amongst team members.Although many groups had already implemented training themselves, they felt that this existing training was not sufficiently in-depth enough to guarantee confidence in volunteers' decision-making.

Needs for funding/personnel
The need for funding to support safeguarding was raised by two interviewees.One noted that it could go toward quality, in-person safeguarding training for volunteers.

Needs for connections
A need for more connections was raised by three interviewees, including links to organizations such as the local authority, charities, or other COVID mutual aid groups.Two potential benefits from connections were identified.Firstly, connections could enable sharing of best practice.Secondly, connection with other groups would enable a more strategic approach, both to planning services and to increasing the leverage and influence of mutual aid groups in the wider area.

Resources which would be useful in future
Our interviewees suggested a plethora of potential solutions to the needs and concerns outlined above.Here we briefly explore these solutions, from the least resource-intensive (money, time, labor) to the most.

Educational materials
Educational materials around safeguarding were identified by six interviewees as a simple, cost-effective, and scalable way of addressing knowledge needs at both an organizational and individual level.At the organizational level, three interviewees brought up the idea of a template or "model" safeguarding policy which could be adapted according to the organization.At the individual level, information which took more of a visually appealing format was especially desirable.One participant pointed out that visual material helped in communities with low levels of literacy or where several different languages were spoken.Thus, two interviewees felt that a well-designed flowchart detailing the safeguarding process would be helpful.

Training
Training was suggested by four interviewees as a way to address lack of knowledge amongst volunteers.Three of the four were agnostic as to what format this should take; the fourth advocated strongly for in-person training.One participant also indicated a preference for training to be scenario-based with a test, arguing that this increased engagement.

Funding
Funding could be used to employ extra individuals for the explicit purpose of volunteer management, supervision, and training: this could address concerns around potential harms, as well as needs around volunteer knowledge.

Consultation
Four interviewees suggested an ongoing consultation service offered by a professional third party to discuss issues around safeguarding.This service could be used both for advice on resolving complex cases and to assure and validate policy.Ongoing collaboration would also serve to generate a more upto-date picture of the group's needs, allowing them to be met in a timely fashion.

Signposting database
A signposting database or directory with up-to-date information on who to report concerns to was suggested by three interviewees.One interviewee pointed out that this could be especially useful as a service provided by larger organizations, as they are equipped with the resources to check information in a way which smaller ones are not.

A forum for mutual aid groups
One interviewee suggested facilitating a forum for mutual aid groups, charities and other organizations to convene, share best practice, and plan together.

Discussion
Our findings on the practices adopted by COVID mutual aid groups to meet safeguarding needs are echoed in some previous case studies (Chevée, 2021;Tiratelli & Kaye, 2020).However, our findings go beyond previous work through identification of novel strategies to assist such groups, notably the creation and maintenance of an up-to-date signposting database and the creation of a forum for local COVID mutual aid groups to assist one another.Future research could seek to evaluate such initiatives, both in terms of promoting safeguarding practices and in their ability to broker stronger working relationships between formal organizations and informal community groups.
Our findings on groups' needs and suggestions also points to the importance of relationships with other organizations, including more established community groups.All our participants reported beneficial relationships with organizations such as churches, charities, or LIOs, who were able to provide resources such as funding, knowledge or consultation.However, these were in some cases coupled with concerns about a loss of autonomy or identity, especially as the partner organization became more formal or associated with local governmental authorities.
These examples of the concern around losing group autonomy or distinctiveness are like those noted in previous studies of COVID mutual aid groups (e.g., Fernandes-Jesus et al., 2021).Based on the feedback of the groups who reported positive experiences, established organizations should properly respect the role that grassroots groups play and seek to use their resources to support and learn from them rather than supplant them.Some of the strategies identified through our research, such as the creation of an up-to-date signposting database, would enable established groups to leverage their greater resources, whilst also empowering smaller groups to make more informed decisions.
CONTACT Guanlan Mao g.mao@sussex.ac.ukOxford Institute of Clinical Psychology Training and Research, Isis Education Centre, Oxford, UK My present affiliation for Guanlan Mao: Oxford Institute of Clinical Psychology Training and Research, Isis Education Centre, Oxford, UK