Funding Informal Social Programmes Without Signing Off on the Risk Blind

A commissioner funding a walking group for isolated older adults and a commissioner funding a clinical befriending service are, on paper, funding the same outcome. They are not taking on the same risk. One puts a paid, supervised, vetted worker into a one-to-one relationship with a vulnerable adult. The other puts strangers who answered a leaflet into a park together, with a volunteer holding a clipboard. Both get called “social prescribing.” Only one of them typically gets a safeguarding line in the grant agreement.

This is the gap this piece is about. The evidence that informal, community-based connection helps is reasonably solid — social prescribing reviews report gains in self-esteem, confidence, and reduced use of GP and emergency services. The evidence on how to fund it so that harm to a vulnerable participant doesn’t happen on your watch is close to nonexistent in the published literature. That absence is itself useful information: it means you cannot commission your way out of this by asking for “an evidence-based safeguarding approach.” No such off-the-shelf approach has been evaluated. You are working from first principles, general safeguarding practice, and judgment about who is actually showing up to these groups.

Step 1: Work out who is in the room before you fund the room

Social prescribing routes people toward community groups precisely because they are isolated, and isolation correlates with other vulnerabilities. The National Academies’ 2020 report puts roughly a quarter of adults 65 and older in the socially isolated category, and frames isolation as something the health care system should be routinely screening for — which tells you that a meaningful share of referred participants will have been flagged by a clinician as at-risk before they ever reach your funded activity. The American Heart Association’s 2022 scientific statement adds that isolated and lonely people carry elevated cardiovascular risk and that socially vulnerable groups are disproportionately represented among them. None of that is a safeguarding finding. It is a population profile, and it tells you the group walking into your funded walking club is not a random sample of the public. Ask the referral pathway — GP, link worker, self-referral — what proportion of participants arrive with an existing care plan, cognitive impairment, or known vulnerability, before you finalise the funding terms, not after an incident.

Step 2: Separate “informal” from “unsupervised” in the funding agreement

Ray Oldenburg’s account of the third place and Eric Klinenberg’s argument about social infrastructure both treat the value of these spaces as coming precisely from their informality — nobody is running an agenda, nobody is being case-managed. That informality is the point, and funders should not try to clinicalise it out of existence. But informal does not have to mean unsupervised. A grant can require a named, accountable coordinator for a group without turning the group itself into a clinical service. The distinction to write into the agreement: activities can be unstructured, but responsibility cannot be undefined. If no named person on the funded programme is responsible for noticing when a participant stops attending, discloses harm, or is being targeted by another attendee, you have funded a space, not a programme.

Step 3: Calibrate vetting to contact type, not to the word “vulnerable”

Blanket background-check requirements for every volunteer in every funded group are expensive, slow recruitment of volunteers you badly need, and don’t map onto actual risk. A person who attends a group as a peer and never has one-to-one unsupervised contact with a vulnerable participant carries a different risk profile than a volunteer befriender who visits someone’s home. Ask three questions of any role before setting the vetting bar:

1. Does this role involve one-to-one contact, or only group contact with others present?
2. Does this role involve entering a participant’s home, vehicle, or another private setting?
3. Does this role involve access to a participant’s personal, financial, or medical information?

A “yes” to any of these moves the role into the tier requiring formal checks. A “no” to all three does not mean no safeguarding requirement — it means a lighter one, focused on code-of-conduct training and a reporting line, not a background-check bureaucracy that will deter the volunteers the programme depends on.

Step 4: Fund the coordination role explicitly, as a line item

Grant applications routinely underfund coordination because “the activity” is what feels fundable — the walking route, the craft materials, the room hire. Coordination — the person who tracks attendance, follows up on someone who stops coming, holds the incident log, and knows what to do if a participant discloses abuse or self-harm risk — gets treated as overhead and cut in negotiation. Refuse that cut. If a funding application has no explicit, paid, named-hours allocation for who does this work, the programme has no safeguarding function, regardless of what the policy document says. The qualitative evidence on social prescribing consistently finds that participants value structured, purposeful activity with someone accountable for it more than unstructured contact alone — which is also, conveniently, the condition under which safeguarding actually functions.

Step 5: Require an incident-reporting minimum as a funding condition, not a hope

Set this before money moves, in the grant agreement itself:

Requirement	What to specify
Reporting threshold	What counts as an incident (disclosure of harm, unexplained non-attendance, conflict between participants, concern about capacity or cognitive decline)
Reporting timeline	Within how many hours a coordinator must escalate
Escalation route	Named contact at the funded organisation and, separately, the statutory safeguarding lead they must notify
Review cadence	How often the funder sees an aggregated (not case-identifying) log

This is not about creating paperwork for its own sake. It is about making sure the first time you learn a programme has a safeguarding problem is not when it appears in a complaint or a news story.

Step 6: Ask what happens when a participant is the source of harm

Programme design conversations focus almost entirely on protecting a vulnerable participant from external risk. They talk far less about the more common scenario: one participant in an informal group behaving in ways that harm another — financial exploitation, harassment, a volunteer overstepping a boundary with someone they visit. Ask any applicant, directly, what their process is for removing a participant or volunteer from a group, and who has the authority to do it. If the answer is vague, that is the answer.

What this means in practice: Before you sign off on funding, get three things in writing from the applicant — a named coordination role with paid hours, a vetting tier matched to actual contact type rather than a blanket policy, and an incident-reporting threshold with a named escalation contact. If any of the three is missing, you are funding a space, not a safeguarded programme, and should say so rather than assume the informality of the activity means the risk is informal too.

The named-coordinator vacuum

Call this the unaccounted volunteer problem: a well-meaning informal group runs for months on the goodwill of one or two volunteers, nobody is formally designated as responsible for noticing when something goes wrong, and when something does go wrong — a participant discloses abuse, a volunteer’s behaviour crosses a line — there is no clear route for it to have been caught earlier or handled once it surfaced. This is not a failure of the volunteers. It is a failure of the funding agreement that never named anyone as accountable in the first place.

Evidence status

Claim	Status
Social prescribing improves self-esteem and reduces some service use	Reasonably supported across systematic reviews, though trial evidence remains thin
Structured, purposeful group activity outperforms unstructured contact	Supported by qualitative synthesis of participant accounts
Specific safeguarding models for informal community programmes reduce harm	Not evaluated in the published literature available here
Isolated older adults referred into community programmes carry elevated background risk	Supported by population-level data on isolation and vulnerability

What this does not solve

None of this eliminates risk, and none of it is validated by an evaluation showing it reduces harm — because that evaluation doesn’t exist yet for informal community programmes. This also does nothing for the participants who never get referred at all: social prescribing and grant-funded groups reach people already in contact with a GP, link worker, or community organisation, which is a narrower slice of the isolated population than commissioners sometimes assume. A funding agreement can make a programme accountable. It cannot make it reach the people who never walked through the door.