Funding the Handoff, Not Just the Referral

Most social prescribing budgets are built around the first fifteen minutes: the GP conversation, the referral form, the link worker’s initial call. Almost none are built around the fortnight after that, when a person is meant to walk into a walking group, a men’s shed, or a choir and actually stay. That second stretch is where most programmes lose people, and it is the part commissioners fund least.

The UK’s 2018 national loneliness strategy set the template most health systems have since copied: fund link workers, embed them in primary care, connect them to a directory of community activities. It was the first national strategy of its kind and it did the hard political work of getting loneliness onto a government agenda. But a strategy that starts and ends at the referral treats the community activity on the other end as a fixed resource that will simply be there, staffed, and welcoming, whenever a link worker sends someone its way. That assumption is where the pathway breaks.

What the evidence actually says works

Two systematic reviews published in 2021 looked at social prescribing’s effect on loneliness specifically, rather than on wellbeing in general. Both found positive individual-level outcomes were commonly reported — but a 2021 review in Perspectives in Public Health also found that only three of nine included studies reported any reduction in service use (GP visits, emergency attendances, inpatient care), and the trial evidence underneath the whole field remains thin and heterogeneous. A parallel review the same year, in the International Journal of Environmental Research and Public Health, reported gains in self-esteem and self-confidence but flagged the same weakness: programmes vary so much in design that “social prescribing” is closer to a category than a single intervention.

A 2022 qualitative meta-synthesis in BMC Health Services Research gets closer to the mechanism. Participants who benefited did not describe social contact alone as the value. They described restored participation and purpose — being useful, having a role, being expected somewhere. The synthesis notes that structured, purposeful group activity appears to do more than unstructured contact. That is a claim about programme design, not about referral volume, and it points funding attention at the receiving end of the pathway rather than the sending end.

The named failure mode: the warm handoff that goes cold

Link workers are trained to make a “warm handoff” — a personal introduction rather than a phone number on a leaflet. The National Academies’ 2020 consensus report on isolation in older adults, and the clinical commentary that followed it, both call for health systems to routinely screen for isolation and connect people onward. Neither addresses what happens if the group on the other end of that handoff has no capacity, no consistent facilitator, or folds after eight weeks because its own funding ran out. The handoff was warm. It arrived cold, because nobody funded the destination to still exist.

This is the design failure worth naming: the warm handoff that goes cold. It happens when the referral relationship is funded on an annual commissioning cycle and the community activity it depends on is funded, if at all, on a grant cycle that does not line up with it — or not funded at all, running on volunteer goodwill that this quarter happens to hold and next quarter might not.

What this means in practice: if you are commissioning a social prescribing pathway, do not fund the link worker role and stop there. Fund, or at minimum map and monitor, the capacity of the activities the link worker refers into. A pathway is only as reliable as its weakest funded link, and in most designs the weakest link is not the referral — it is the thing being referred to.

Where to put commissioning attention

Pathway stage	Typically funded?	What the evidence says matters
Screening / identification (clinician flags isolation)	Usually, where screening is mandated	Recommended by the National Academies, but screening without a functioning destination is a dead end
Referral / link worker contact	Yes, almost always the core commissioned role	Warm, personal introduction; evidence base is mostly about this stage
Community activity capacity	Rarely, or via separate uncoordinated grants	The 2022 meta-synthesis ties benefit to structured, purposeful activity — which requires stable facilitation, not just a venue
Follow-up / re-engagement after first visit	Almost never commissioned explicitly	Not directly evidenced in the sources here, but structurally this is where attrition is least visible and least measured

What to ask a provider before you fund a pathway

1. What happens to a referral if the intended activity has no space that week — is there a fallback, or does the person just not hear back?
2. Is the community activity’s funding term matched to the referral pathway’s funding term, or shorter?
3. Who is accountable for knowing whether a referred person actually attended a second time, not just a first?
4. If a facilitator leaves, is there a named backup, or does the group quietly stop?

None of this is answered by the current literature with any precision — the field’s own reviewers say so. Holt-Lunstad’s 2021 review argues connection belongs in preventive frameworks alongside diet and exercise, which is a case for funding it at all. It is not a case for funding it as currently structured.

What this does not solve

Pathway design fixes leakage between services. It does not reach the people who never get referred in the first place — those without a GP relationship, without English as a first language, or without a link worker in their area. And better handoffs cannot compensate for a community activity sector that is, in most places, running on time-limited grants regardless of how well any pathway feeding it is designed. Fixing the join is necessary. It is not sufficient if the thing on the other side of the join is itself unfunded.