ENCS
Tuesday, 11 August 2026
Your paper, your pace.

Loneliness: what has actually been tried, and what the surveys found afterwards

In 2022, 11.2% of people in the EU had nobody to ask for help — up from 7.7% four years earlier. Five countries built policy around it. One can point to a measured effect, and it came from a trial of 235 people.

Checked on

All ten problems

In 2022, 11.2% of people in the European Union aged 16 or over said there was nobody they could ask for help if they needed it. In 2018 the figure was 7.7%. It is Eurostat’s dataset ilc_pw06, it is free, and it runs from 2.4% in Norway to 29.3% in Romania. One warning before using it: several countries moved by more than ten points in either direction between the two rounds — Germany from 7.6% to 20.2%, Bulgaria from 18.0% to 4.6% — which is more than a country’s social life plausibly changes in four years. Those are almost certainly the survey moving, not the population, which is why Germany and Bulgaria are not on the board.

The question that dataset asks is about isolation, not loneliness. Isolation is countable: does a person exist who could be asked. Loneliness is a feeling about the gap between the company someone has and the company they want, and the surveys that try to measure it disagree with each other. The OECD, using EU-SILC data for 2022 and publishing on 16 October 2025, found about 6% of people across 25 countries feeling lonely most or all of the time. The EU Loneliness Survey, fielded online in November and December 2022 with 25,646 respondents, found 13%. Both are competent. The instrument changes the answer.

One number in this field travels further than any other, and it is worth getting right. In 2010 Julianne Holt-Lunstad and colleagues pooled 148 studies covering 308,849 people and found that those with stronger social relationships had a 50% higher likelihood of surviving the follow-up period, which averaged 7.5 years. To show what that size of effect meant, they lined it up against other known risks — and the smoking comparison in that chart was for moderate smoking, defined as up to 15 cigarettes a day. That is where “loneliness is like fifteen cigarettes” comes from. Holt-Lunstad has since written that the original comparison was for lacking social connection in aggregate, of which loneliness was one part, and the underlying studies are observational: they show association, not proof of cause. Health is a subject on this page, not counsel; nothing here is a diagnosis or advice, and a doctor is the right address for that.

How this is measured

One number, named and sourced. Everything on this page is argued against it, so you can check us.

Share of people aged 16 and over with no one to rely on for help in case of need — the mirror image of Eurostat's published series, which counts the people who do have someone %

Eurostat, EU-SILC, dataset ilc_pw06, 2022 reference year — the most recent available; dataset last updated 23 January 2026

Who is actually ahead

Ranked on the number above — not on reputation.

Share of people aged 16 and over with no one to rely on for help in case of need — the mirror image of Eurostat's published series, which counts the people who do have someone
Country Result Why
Norway 2.4 % The lowest in Europe, and improving: 3.4% in 2018. Norway is one of the few countries whose answer moved only slightly between the two rounds.
Portugal 3.0 % Down from 4.2% in 2018. Portugal scores far better on this question than on income, which is a reminder that the two are not the same thing.
Finland 3.6 % Up from 2.2% in 2018 — still among the best in Europe, but one of the few Nordic figures that moved the wrong way.
Czechia 4.5 % Up from 3.2% in 2018. Czechia sits in the best quarter of the EU on this measure and well below the EU figure of 11.2%.
Netherlands 4.5 % The largest improvement in western Europe: 9.0% in 2018. The Dutch national loneliness programme ran across exactly this period, which is suggestive and not proof.
Ireland 5.6 % Up from 4.2% in 2018. Ireland scores well here and worst in the EU on self-reported loneliness — the clearest sign that the two questions measure different things.
France 9.4 % Doubled from 4.7% in 2018, the largest deterioration among the big western economies, and still below the EU figure.
Romania 29.3 % The highest in the EU, up from 18.3% in 2018. Almost three in ten Romanian adults say there is nobody they could ask.

What has actually been tried

Real countries, real policies, and what the numbers did afterwards — including where it went wrong.

Finland

Finland tested a group, not a campaign. In the Circle of Friends trial, older people who said they were lonely were put into closed groups of seven or eight with two trained professional leaders, meeting twelve times over three months. Groups did one of three things: writing and group psychotherapy, exercise and discussion, or art. It was a randomised controlled trial — the rarest thing in this field.

What happened. Among 235 people aged over 74, survival at follow-up was 97% in the intervention group (95% CI 91–99) against 90% in the control group (95% CI 85–95), p = 0.047. Health care costs were €943 per person per year lower (95% CI −1955 to −127, p = 0.039), and subjective health improved significantly. By 2022 the programme had run for 12,000 participants in Finland with 1,200 trained facilitators, and a ten-year follow-up reported that nine in ten participants said their loneliness had eased.

The catch. One trial, 235 people, and a p-value of 0.047 is a hair inside the conventional line. It has not been repeated at that scale anywhere else. It also recruited older people who agreed to join a group twelve times, which is not the same population as everyone who is lonely — the least connected are the hardest to enrol, and the trial cannot say anything about them.

Pitkälä and colleagues, 'Effects of psychosocial group rehabilitation on health, use of health care services, and mortality of older persons suffering from loneliness', The Journals of Gerontology Series A, July 2009; Jansson, Circle of Friends, presentation to the European Commission Joint Research Centre

United Kingdom

The United Kingdom went first and went loudest. After the Jo Cox Commission reported in 2017, loneliness was added to a minister's brief and the strategy 'A connected society' was published in October 2018 — the first of its kind. It came with £20 million for organisations working on loneliness, including an £11 million Building Connections Fund, and with social prescribing built into the NHS.

What happened. The Community Life Survey measured 6% of adults in England as often or always lonely in 2018/19. The 2024/25 round, published on 10 December 2025 with more than 175,000 responses, put it at 6.6%, down from 7.1% in 2023/24. The publication describes the series as broadly stable at 5–6% from 2013/14 to 2021/22. Seven years after the strategy, the number is at or above where it started.

The catch. A pandemic sits in the middle of the series, and the two publications round differently — the 2018/19 fact sheet says 6%, the 2024/25 release describes those years as 5–6%. The strategy was also never funded at a scale that could plausibly move a national rate: £20 million across a country of 57 million adults is about 35 pence a head.

UK Department for Culture, Media and Sport, Community Life Survey 2024/25: loneliness and support networks (published 10 December 2025); Community Life Survey 2018/19 loneliness fact sheet; House of Commons Library briefing CBP-8514

Netherlands

The Netherlands built its programme downwards. 'Eén tegen eenzaamheid' — One against loneliness — started in 2018 and works through municipalities: local coalitions of shops, GPs, sports clubs and volunteers, supported and monitored from The Hague rather than run from it. By 2025 about 80% of all Dutch municipalities took part, and of those, more than 80% ran a local coalition.

What happened. The four-yearly Health Monitor of adults and older people found 49.1% of Dutch adults reporting some loneliness in 2022 and 47.2% in 2024. The programme's final progress report of 16 December 2025 concluded that the share feeling lonely 'appears to be declining slightly', while the share who feel very strongly lonely 'barely decreases' — close to one adult in seven.

The catch. The Dutch question counts anyone with any degree of loneliness, so 47.2% is not comparable with the British 6.6%; different instrument, different threshold. A survey run once every four years also cannot separate a national programme from a pandemic that fell inside the same window. The honest summary is that the mild end moved a little and the severe end did not.

Dutch government, final progress report on Eén tegen eenzaamheid, 16 December 2025 (parliamentary document 2025D52385); Health Monitor Adults and Older People, GGD, CBS and RIVM

Japan

Japan legislated. A Minister in charge of measures for loneliness and isolation was appointed in February 2021, the Act on the Advancement of Measures to Address Loneliness and Isolation was enacted in May 2023 and came into force in April 2024 — the first comprehensive national law of its kind. It obliges national and local government to co-ordinate, and asks municipalities to set up regional councils.

What happened. The Cabinet Office's annual national survey recorded 36.4% of respondents reporting loneliness in 2021, 40.3% in 2022 and 39.3% in 2024. In the 2024 round, from 10,876 valid responses, 4.3% said they felt lonely often or always; the highest rate was 7.4% among people aged 20 to 29. Two years after the law came into force, the headline proportion is where it was before it.

The catch. The law arrived with very little money behind it: 38 prefectures set up regional platforms, but the initial national budget for supporting all 1,741 municipalities was ¥210 million in fiscal 2025 — about ¥120,000 per municipality. A statute that requires councils without paying for them measures intent rather than delivery, and two years is a short window in which to expect a national survey to move.

Cabinet Office of Japan, 'Japan's efforts to tackle loneliness and isolation', June 2025; Institute for Social Vision Design, two-year review of the Loneliness and Isolation Act

South Korea

South Korea counts the deaths. A 2021 law on the prevention and management of lonely deaths — godoksa, a person who dies alone at home and is found later — obliged the Ministry of Health and Welfare to survey them and report. In October 2024 the city of Seoul announced a five-year plan worth 451.3 billion won, about $327 million, including a round-the-clock counselling line, home visits, shared meals and community activities.

What happened. The Ministry recorded 3,559 lonely deaths in 2022 and 3,661 in 2023 — 1.04 of every 100 deaths in the country. Men accounted for 84.1% of them, with people in their 50s and 60s more than half. Single-person households rose from 7.16 million in 2021 to 7.82 million in 2023, or 35.5% of all households.

The catch. This is the one page entry where the measured thing is a death rather than a feeling, which makes it hard and also narrow: it counts the end state, not the years before it, and part of the rise is simply more people living alone. The Seoul plan was announced in October 2024 and no outcome has been published yet, so it belongs on the page as a cost, not yet as a result.

Ministry of Health and Welfare of South Korea, lonely deaths report, October 2024, as reported by The Korea Times (17 October 2024) and The Korea Herald (21 October 2024); Seoul plan first reported by CNN, 24 October 2024

What a machine would optimise for

This is arithmetic, not advice and not a prediction. We name the single number being maximised and follow it wherever it goes. The point is to see the shape of the answer a calculator gives.

The number being maximised

The single number is the share of adults with someone to rely on in case of need: 88.8% across the EU in 2022. Push it towards 100 and nothing else enters the calculation.

  1. Answer the question the survey asks. It is a yes-or-no about whether a person exists who could be asked, so assigning every household a named contact — a neighbour, a warden, a case worker on a list — turns a no into a yes without anybody meeting anybody.
  2. Go where the noes are. Romania at 29.3% and France at 9.4% hold the bulk of Europe's total; Norway at 2.4% has almost nothing left to remove. The same intervention moves the European figure ten times further in Iași than in Bergen.
  3. Merge households. The indicator is satisfied by proximity: two people sharing a flat can each name the other, so the cheapest unit of improvement is a lease rather than a friendship, and the calculator cannot see the difference.
  4. Buy the young first. Loneliness among 16 to 24-year-olds across OECD countries rose from 3.3% to 4.3% between 2018 and 2022. They can be reached through schools and universities they already attend, and each one carries sixty more years of the indicator.

The arithmetic

EU27, 2022: 11.2 people in every 100 aged 16 or over had nobody to ask. The EU held about 447 million people; if roughly five in six are 16 or over, that is about 372 million adults and about 42 million with nobody. Romania, with about 19 million people and about 16 million adults, holds about 4.7 million of them — 11% of the European total from 4% of Europe's adults. A calculator reading only this indicator would spend everything in Bucharest, Iași and Craiova and nothing in Lisbon.

What the machine would miss

The most important part of this page. A number that goes up can still be paid for by somebody, and some things never make it into the number at all.

Who pays

The person who becomes the answer. Every yes on this indicator is somebody's daughter, neighbour or volunteer, and none of them appear in the cost column. The Finnish trial saved €943 per person per year in health care, but twelve two-hour sessions were run by two trained professionals for groups of seven or eight, and the Dutch programme reached 80% of municipalities on the backs of local volunteers. The saving lands on the health budget; the hours land on a family.

What the number flattens

Isolation and loneliness are not the same thing and this indicator only sees the first. Ireland is among the better-connected countries on Eurostat's question and the loneliest in the EU on self-reported loneliness. Even the self-reports disagree: the OECD found about 6% of people in 25 countries feeling lonely most or all of the time in 2022, while the EU Loneliness Survey of the same year, run online with 25,646 respondents, found 13%. Same construct, same year, twice the number.

What a spreadsheet cannot see

Whether the company is wanted. A person with five names to call can be lonely and a person with one can be content, and no table separates them. The most-repeated claim in this field — that loneliness equals smoking fifteen cigarettes a day — comes from a 2010 meta-analysis of 148 studies and 308,849 people that found a 50% higher likelihood of survival for people with stronger social relationships, and benchmarked that against smoking defined as up to 15 cigarettes a day. Its author has since written that the comparison was for lacking social connection in the round, not loneliness alone.

Where does your country stand?

Open my country

Where the numbers come from

We are a newspaper, not a government and not an adviser. Nothing on this page is a recommendation to you or to anyone in office. It is what was tried, what it measured, and what a calculator would say if you let it loose on the same problem.

One good piece of thinking a day

The day's most worthwhile story, and the question underneath it. No spam, one click to leave.

One click to leave. We never sell or share your address.