How To Write A Loneliness Programme Bid A Commissioner Will Fund

The most common reason a good social connection programme fails to get funded is that the bid was written in the language of the programme rather than the language of the commissioner. You describe a weekly walking group with a volunteer host and a shared lunch. The panel is scoring against reduced non-elective admissions, service demand, or a wellbeing metric they already report upward. Those are not the same document.

This note is about closing that gap without overstating what the evidence supports. It assumes you already know your programme works, or at least that you have a defensible reason to believe it might. The problem is translation.

Work out what the commissioner is buying

Before you write a word, establish which of four things the money is for. Commissioners rarely say this explicitly, and the specification often blurs it, but every pot behaves as one of these:

1. Demand reduction. They want fewer contacts with an expensive service — GP appointments, emergency attendances, social care assessments.
2. Population risk. They have a strategy or an outcomes framework with loneliness, isolation, or wellbeing in it, and they need something to point at.
3. Statutory or contractual cover. Something obliges them to have a provision. They need it to exist, be safe, and reach the stated group.
4. Innovation or test-and-learn. They are buying evidence, not outcomes. The deliverable is a defensible answer to “did this work”.

These require materially different bids. A demand reduction bid that leads with wellbeing scores will lose. An innovation bid that promises admission reductions will win the money and then fail its own evaluation. If the specification does not tell you which one it is, ask at the market engagement event, and ask what the commissioner is required to report and to whom.

The three claims you can make, and what backs each one

Most loneliness bids rest on a chain: our programme reduces loneliness, loneliness causes poor health, therefore our programme improves health and saves money. Every link in that chain has a different evidential strength, and commissioners with any analytical support will find the weak one.

Claim	Evidence status
Loneliness and isolation predict earlier death	Strong. Holt-Lunstad’s 2015 meta-analysis found odds ratios of 1.26 for loneliness, 1.29 for social isolation, and 1.32 for living alone, holding after adjustment for health status. The 2010 review across 148 studies and 308,849 people found a 50% increased likelihood of survival with stronger relationships.
Loneliness is prevalent enough to justify population-level attention	Strong. Harvard’s Making Caring Common survey put serious loneliness at 36% of Americans in early 2021, including 61% of 18-to-25-year-olds. AARP’s 2018 survey of 3,020 adults aged 45 and over, using the 20-item UCLA scale, found one in three lonely. The National Academies concluded in 2020 that roughly a quarter of adults over 65 are socially isolated.
Social connection is a modifiable risk factor, not just a correlate	Moderate, and improving. Holt-Lunstad’s 2021 review in the American Journal of Lifestyle Medicine argues connection belongs alongside diet, exercise and smoking in preventive frameworks. This is a well-argued position, not a demonstrated causal chain.
Group-based social interventions reduce loneliness measurably	Weak to moderate. The 2021 systematic review in Perspectives in Public Health found all nine included social prescribing studies reported positive individual impacts, but the studies are small and heterogeneous.
Such programmes reduce health service use	Weak. Three of those nine studies reported reductions in GP, emergency, social worker or inpatient use. Three studies is not a business case.
Such programmes improve self-esteem and confidence	Moderate for direction, weak for magnitude. The 2021 systematic review in IJERPH identifies these as the key reported outcomes, while flagging limited trial evidence and heterogeneity.

Read that table again as a commissioner would. The risk-factor evidence is excellent. The intervention evidence is thin. Any bid that presents the strength of the first as though it applies to the second is being read by someone who knows the difference.

What this means in practice: cite the epidemiology to establish that the problem is worth money. Cite the intervention literature to establish that your approach is reasonable and not novel. Do not let the first do the work of the second. Write one sentence that says so explicitly — something like “the association between social isolation and mortality is well established; the evidence that group programmes reduce that risk is early, which is why we are proposing the measurement plan below.” Commissioners fund people who tell them where the uncertainty sits.

Translate outcomes into the units they report in

A commissioner’s outcomes framework already exists. Your job is to map onto it, not to introduce a new one.

Find out what they measure. In the UK, the four ONS loneliness questions became a standard instrument after the 2018 national strategy embedded loneliness measurement into official statistics, and most local authority and integrated care frameworks now use them. If your commissioner reports on those four items, use those four items. Adding the UCLA-3 alongside is fine and often useful for comparability with the academic literature — that is exactly why the AARP survey used the 20-item version. But your headline number should be the one they already report.

Then do the arithmetic they will do. If you claim reduced GP contacts, state the baseline contact rate for your cohort, the reduction you expect, the number of people, and the unit cost. If you cannot source a baseline, say so and propose collecting it in the first quarter. A bid that shows its working with an honest gap beats a bid with a confident unsourced figure.

Name your cohort narrowly enough to be checkable

“Socially isolated adults” is not a cohort. It cannot be counted, recruited to, or evidenced against.

Narrow it until the commissioner can picture the referral. Adults aged 45 to 64 living alone in the three most deprived wards. Men who have had a bereavement in the last eighteen months. Employees returning from long-term sick leave. Mothers of children under five — Harvard’s survey found 51% of that group reporting serious loneliness.

Specificity also lets you use the demographic evidence properly. If you are targeting men, the Survey Center on American Life’s 2021 friendship survey is directly useful: the share of men reporting at least six close friends fell from 55% in 1990 to 27%, and 15% now report no close friends, a fivefold rise. If you are targeting under-25s in workplaces, Cigna’s 2020 report found 73% of workers aged 18 to 22 reporting loneliness, and lonely workers missing work twice as often through illness and five times as often through stress. That last figure is the one an employer-funded bid should lead with.

The fourth-quarter problem

Here is the failure mode that kills renewal more often than poor delivery.

Most annual contracts require an evaluation report in the final quarter. Most programmes do not collect a baseline properly in the first month, because the first month is spent on recruitment, safeguarding, venue contracts and getting the referral route working. By the time anyone thinks about measurement, the cohort has already started, and the baseline is reconstructed from memory or skipped.

Come the fourth quarter, you have attendance data, some warm quotes, and no before-and-after on the instrument the commissioner reports. The programme may have worked. You cannot show it. The renewal decision is then made on relationship and impression, which means it is made on budget.

Fix this in the bid, not in delivery. Put baseline collection in the mobilisation milestones as a named deliverable with a date. Specify who administers the instrument, in what setting, and what the minimum completion rate is before you would consider the data usable. Ask for the first quarter to be a mobilisation quarter with reduced delivery volume, and justify it on measurement grounds. Commissioners grant this more often than people ask.

What to do with the venue

Bids frequently treat the physical space as a line item. It is often the intervention. Klinenberg’s Palaces for the People makes the case that libraries, parks and other shared spaces measurably shape rates of social contact, drawing on survival differences in the 1995 Chicago heat wave.

If you are delivering in a library, a leisure centre, or a community hall the commissioner already owns or funds, say so and quantify it. You are increasing utilisation of an asset already on their books. That argument lands with estates and finance colleagues who never read the clinical section.

What this does not solve

None of this helps with reach, and reach is where the honest weakness in this whole field sits.

Referral-based and self-referred programmes overwhelmingly reach people who have already identified themselves as looking for something. The AARP data shows the pattern plainly: 33% of people who had spoken to their neighbours were lonely, against 61% of those who never had. The second group is who you most want and least often get. The National Academies’ recommendation that health systems routinely assess isolation — argued through for clinicians in the 2020 American Journal of Geriatric Psychiatry commentary — is the most serious proposal on the table for changing that, and it requires clinical time nobody has costed.

Write your bid knowing this. If you promise to reach the most isolated quartile, you will be held to it and you will miss. If you state that your reach is limited to people who present, and propose one specific outreach mechanism to test alongside, you have given the commissioner something they can defend when someone asks the same question upward.