Research · for befriending services, day centres & older people's charities

Housebound in Later Life: 5.6% of Older Americans Rarely or Never Left Home

Published 2026-11-15

In the United States in 2011, 5.6% of people aged 65 and over living in the community, about 2 million people, never or rarely left their homes[1]. Housebound older people have more long-term conditions, fewer chances to see others and report more social isolation and, in several studies, more loneliness than other older people. Yet we found only a handful of trials of the services that reach them at the door.

Being housebound is rarely a choice. It usually follows illness, pain, frailty, a fall, failing sight or breathlessness, and it is held in place by steps without a rail, flats without a lift, buses that do not stop nearby and streets that do not feel safe. Housebound people are among the hardest people for a club or day centre to reach, and their contact often comes from a meal driver, a care worker, a befriending volunteer or a phone call. This study sets out how many older people are housebound, who they are, how isolated and lonely they are, and what trials of home-delivered meals and home-based support have found.

17.7–19.5%of people in two representative Israeli cohorts with an average age of about 83 were homebound, and in one of the cohorts they were significantly lonelier than those who were not homebound[2]
626older people on Meals on Wheels waiting lists at eight US programmes took part in a randomised trial; those who received meals had lower adjusted loneliness scores after 15 weeks than those still waiting[3]

Between about 1 in 20 and 1 in 5 older people are housebound, depending on age, place and definition

There is no single definition of being housebound, which is why estimates vary so much. A 2022 systematic review of 19 studies found that, depending on the definition each used, the share of older people who were homebound ranged from 3.5% to 39.8%[4]. A scoping review of 47 studies likewise found that various definitions and measures have been used[5].

The deepest national data come from the US National Health and Aging Trends Study, an in-person survey of Medicare beneficiaries aged 65 and over who live in the community, so we use it as a worked case. In 2011, 5.6% were homebound: an estimated 395,422 people never left home in the previous month and 1,578,984 rarely did. A further 3.3% never left home without someone to help them, and 11.7% needed help or had difficulty going out[1]. In the same survey in 2022, 4.9% never or rarely went outside[6]. The two analyses worded the definition slightly differently, so the small difference is not evidence of a trend.

Studies of older and more rural groups find higher shares. In Israel, two representative cohorts with an average age of about 83 found that 17.7% to 19.5% were homebound[2]. In rural Saitama, Japan, in 2021, during the pandemic, 10.4% of people aged 65 and over who managed their daily activities independently went out once a week or less: 6.6% of men and 13.8% of women[7].

Share of older people found to be homebound, by study, country and definitionShare of older people classed as homebound in four studies that used different definitions and age groups: Israel, two representative cohorts with an average age of about 83, 17.7% to 19.5%; rural Saitama, Japan, 2021, people aged 65 and over who were independent in daily activities and went out once a week or less, 10.4%; United States, National Health and Aging Trends Study 2011, Medicare beneficiaries aged 65 and over who never or rarely left home in the last month, 5.6%; same survey 2022, never or rarely went outside in the past month, 4.9%.0%5%10%15%20%Israel, average age ~8317.7–19.5%Rural Japan, 65+, 202110.4%US, 65+, 20115.6%US, 65+, 20224.9%
Being homebound, not loneliness. Definitions, ages and years differ (US: never or rarely left home in the past month; rural Japan: out once a week or less, during the pandemic), so the figures are not directly comparable. Israeli bar shows the top of the range. Sources: Cohen-Mansfield et al. 2012; Koizumi et al. 2024; Ornstein et al. 2015; Choi et al. 2026.

All of these surveys cover people in their own homes, not care-home residents.

Housebound older people are older, poorer and in worse health, and their surroundings matter too

In the 2011 US data, people who were completely homebound were on average 83.2 years old, against 74.3 for those who were not. They were more likely to be women (67.9% against 53.4%) and of nonwhite race (34.1% against 17.6%), and had less education and income. They had on average 4.9 chronic conditions, against 2.5, and 52.1% had been in hospital in the past year, against 16.2%. Only 11.9% of completely homebound people received primary care at home[1].

Completely homebound and non-homebound older Americans compared, 2011Characteristics of completely homebound and non-homebound Medicare beneficiaries aged 65 and over in the US National Health and Aging Trends Study, 2011: women 67.9% vs 53.4%; nonwhite race 34.1% vs 17.6%; hospitalised in the last 12 months 52.1% vs 16.2%.Completely homeboundNot homebound0%20%40%60%80%Women67.9%53.4%Nonwhite race34.1%17.6%In hospital in the pastyear52.1%16.2%
Completely homebound means never leaving home in the last month. Nationally representative survey of people living in the community, so care-home residents are not included. “Nonwhite race” is the source’s own category. Source: Ornstein et al., JAMA Internal Medicine 2015.

Health explains much of the pattern, but not all of it. In the Israeli cohorts, homebound people were more likely to have stairs and no lift at home, and having stairs without a lift predicted becoming homebound later in both cohorts, along with difficulty with everyday tasks such as shopping and cooking[2]. In the US in 2022, older people who rated their neighbourhood’s social cohesion more highly had lower odds of being homebound (adjusted odds ratio 0.79), after allowing for the size of their support network, the physical state of the neighbourhood, and their health and circumstances. Being homebound was not fixed: between 2022 and 2023, 2.2% of the sample moved out of it, and improving neighbourhood cohesion went with higher odds of doing so[6].

In Japan, the Japan Gerontological Evaluation Study, which followed 89,914 people aged 65 and over who were not homebound in 2019, 2.8% had become homebound by 2022, from 1.3% to 6.4% across 49 municipalities. People living in municipalities that had made less progress in building local daily-living support systems had higher odds of becoming homebound (odds ratio 1.22) than those in municipalities with high scores, after allowing for individual and municipal characteristics; a second score, for long-term care prevention, showed no difference[8]. These findings are observational, but they point to causes outside the person. For long-term disability itself, see our study of aging with disability.

Housebound older people are more isolated, and in several studies lonelier, than others their age

Social isolation and loneliness are different things, and the studies measure them differently, so we keep them apart here. On isolation, a US analysis of nine rounds of the National Health and Aging Trends Study (6,464 people) scored isolation across four areas: marital status, religious attendance, club participation, and contact with family and friends. At the start, homebound and semi-homebound people were more isolated than others; over the following years, though, their isolation did not change at a different rate from everyone else’s[9].

On loneliness, the clearest figures come from the pandemic. In a US survey in the summer and autumn of 2020, older people asked directly how often they felt lonely, 13.2% of those who had been homebound in 2019 (here including people who left home only with help or difficulty) said every day or most days, against 7.7% of others. Before adjustment, they were also more likely to have limited social contact, meaning in-person, phone, video or email contact less than once a week (13.2% against 6.5%), and only 16.3% had learned a new technology during the pandemic, against 30.4% of others[10]. Both groups reported similar rates of loneliness getting worse during the pandemic[10]. Our digital inclusion study covers their phone and computer ownership.

Contact and loneliness among homebound and other older Americans during the COVID-19 pandemic, 2020Among older Americans in the National Health and Aging Trends Study COVID-19 survey, summer and autumn 2020, by homebound status in 2019: limited social contact (in-person, phone, video or email contact less than once a week) 13.2% of homebound vs 6.5% of non-homebound; felt lonely every day or most days 13.2% vs 7.7%; learned a new technology during the pandemic 16.3% vs 30.4%.Homebound in 2019Not homebound0%10%20%30%40%Limited social contact13.2%6.5%Felt lonely every or mostdays13.2%7.7%Learned a new technology16.3%30.4%
Homebound includes people who left home only with help or difficulty. Loneliness: direct question, “every day or most days” against “some days, rarely or never”. Pandemic-era figures. Source: Ankuda et al., Journal of General Internal Medicine 2022.

Outside the US, the Israeli cohorts found homebound people significantly lonelier in one of the two cohorts[2]. In Germany, the German Ageing Survey followed adults aged 40 and over across four waves and defined homeboundness as spending six or more days a week at home. Becoming homebound was robustly associated with a rise in loneliness within the same people over time; but among people aged 65 and over taken separately, changes in being homebound were not significantly associated with changes in loneliness or wellbeing[11]. Being at home a lot is not the same as feeling lonely.

Being housebound is linked with depression and earlier death, and loneliness appears to carry part of the link to low mood

In Israel, homebound people in a national cohort with an average age of 83 had a higher risk of death during follow-up than others (risk ratio 1.33), after allowing for background characteristics, health and everyday function. They were also more likely to report depressed mood, again after allowing for health and function[12]. In an analysis of the 2023 US National Health and Aging Trends Study (7,547 Medicare beneficiaries aged 65 and over), being homebound was associated with more depressive and anxiety symptoms, and loneliness and low psychological wellbeing both statistically mediated part of that association[13]. In a study of 1,301 older people in Shandong Province, China, homebound people had worse mental health than others after allowing for demographics and physical health and loneliness widened the gap[14].

These are associations from observational data. Illness often causes both being housebound and low mood, so being housebound may be a marker of poor health as much as a cause, and a mediation analysis from one survey wave cannot show which came first. What the studies show is a heavy load of illness, low mood and isolation together.

A meal at the door came with lower loneliness scores in a US trial, but the evidence base is small

Home-delivered meals reach housebound older people every week, often every weekday. The strongest test of their effect on loneliness is a US randomised trial with Meals on Wheels programmes, published in 2016. Researchers took 626 older people from the waiting lists of eight programmes and randomly assigned them to daily meal delivery, a once-weekly delivery, or staying on the waiting list. They were surveyed at the start and again after 15 weeks[3].

People who received meals had lower loneliness scores at follow-up than those still waiting, after adjustment. People who had daily deliveries were more likely than those with weekly deliveries to say that the meals had improved their loneliness[3]. We could see only the published abstract, which does not give the size of the difference or the loneliness scale used, so we report the direction of the finding and no more. The trial was short and a meal delivery cannot be hidden from those who receive it, so expectations may have played a part; still, randomisation from a waiting list is a fairer comparison than most loneliness research offers.

Deliver-EE, a larger trial under way, plans to enrol 2,300 people from home-delivered meals waiting lists and compare, over six months, weekday lunchtime meals brought by a volunteer or driver who also offers a chat and a wellness check, against a delivery of 10 frozen meals every two weeks; loneliness is among its secondary outcomes[15]. In interviews with 54 of its participants, recipients described positive contact with drivers, though interviews do not measure loneliness[16]. Shared meals at a lunch club are a different service with different evidence; see our study of day centres and lunch clubs.

Meal services can open the door to other support, and one small trial found lasting gains

Meal programmes already assess the people they serve. In a two-site US trial, 89 people with an average age of 74 were recruited by home-delivered meals programmes during their routine assessments; all reported loneliness. They received five weekly video sessions with a lay coach, either tailored behavioural activation or “friendly visits” by video as the comparison. Loneliness was measured with the PROMIS Social Isolation Scale. The behavioural activation group had a greater fall in loneliness and greater gains in social interaction and satisfaction with support, along with less depression and disability, than the friendly-visit group[17].

When 64 of the 89 were interviewed again after a year, the gains in loneliness, social interaction and satisfaction with support had held, though to a lesser degree[18]. The trial was small, compared two active approaches with no group receiving nothing, and lost more than a quarter of its participants by one year, and the authors themselves called for more research before the approach is treated as proven. Our study of befriending sets this trial alongside telephone and volunteer befriending schemes.

What the evidence asks of us

The WHO Commission on Social Connection, reporting in 2025, named people with disabilities among the groups who “may face discrimination or additional barriers that make social connection harder”, and set out a roadmap built on five areas: policy, research, interventions, better measurement and public engagement[19]. For housebound older people, the evidence in this study points to several practical implications.

  • Count housebound people. Prevalence ranges from 3.5% to 39.8% across studies largely because definitions differ[4]. Outside the US survey used here, our search found no regular national figure. A shared definition, used in national ageing surveys and local needs assessments, would let services see who they are missing.
  • Treat the door as a point of contact. Meal drivers, home-care workers and home-visiting clinicians may be the only people some housebound older people see. In 2011, only 11.9% of completely homebound older Americans received primary care at home[1]. The video-support trial above recruited through meal programmes’ routine assessments, which shows that services already at the door can link people to further support[17].
  • Protect services with evidence behind them, and keep testing them. Home-delivered meals are one of the few services for this group that we found tested in a randomised trial with loneliness as an outcome[3]. Funders can build evaluation into such services rather than cut the contact that comes with the meal before its value is measured.
  • Fix the steps as well as the person. Stairs without a lift, weak neighbourhood ties and thin local support systems were all linked with being or becoming homebound[2][6][8]. Housing adaptations, transport and local support are part of the response to isolation.

For befriending services, older people’s charities and day centres, the message is that the people they cannot see at their doors are often the people with the fewest other contacts. More evidence across the Loneliness in Later Life library.

Language: this piece says "older people" and treats loneliness as a common human experience that anyone can have, not a personal failing or an illness. It keeps apart three things that are often mixed up: loneliness (how people feel about their relationships), social isolation (how few social contacts they have) and living alone (a household arrangement). Where a study uses another term, such as "seniors" or "older adults", we keep it when quoting that study.

Why we're publishing this

Housebound older people rarely appear in the membership lists, attendance figures and member surveys that most community groups use to show their value, because they cannot come along. That makes them easy to overlook in funding and local plans. We have gathered the evidence specific to this group so that befriending services, meal programmes, charities and day centres can make the case for the people they serve at home, with the figures and their limits in plain view.

How to read this data

Housebound has no single definition. The US figures count people who never or rarely left home in the past month (some analyses add those who left only with help or difficulty); the rural Japanese study counts going out once a week or less; the German study counts six or more days a week at home. Prevalence figures from different definitions, ages and years are not directly comparable.

Isolation and loneliness are measured separately, as each section says. For the Meals on Wheels trial and the German and Israeli studies we could read only abstracts, which do not name the loneliness measure, so we report the direction of their findings only.

Study types. Most findings come from cross-sectional or longitudinal surveys, which show associations, not causes; illness can lead to being housebound, loneliness and low mood all at once. The two trials were small or short. Most studies are from the US; an integrative review of loneliness among homebound older people found only 14 studies published between 1999 and 2020[20]. The surveys cover people living in their own homes and leave out care-home residents.

Use this data

Free to cite with attribution to Understanding Aging. u3a groups, men's sheds, senior centres, lunch clubs, befriending services, charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Suggested citation: Understanding Aging (2026). Housebound in Later Life: 5.6% of Older Americans Rarely or Never Left Home. https://understandingaging.org/research/loneliness-housebound-older-people

Important: This article summarises published research and statistics for general information. It is not medical or care advice and cannot be used to diagnose, treat or predict anything about an individual. If loneliness, low mood or isolation is affecting your health or someone else’s, a doctor or another health professional can help, and local services for older people can often put you in touch with groups and support nearby.

Sources

  1. Ornstein KA, Leff B, Covinsky KE, et al. Epidemiology of the homebound population in the United States. JAMA Internal Medicine, 2015. europepmc.org/article/MED/26010119
  2. Cohen-Mansfield J, Shmotkin D, Hazan H. Homebound older persons: prevalence, characteristics, and longitudinal predictors. Archives of Gerontology and Geriatrics, 2012. europepmc.org/article/PMC/PMC3174334
  3. Thomas KS, Akobundu U, Dosa D. More than a meal? A randomized control trial comparing the effects of home-delivered meals programs on participants’ feelings of loneliness. Journals of Gerontology, Series B, 2016. europepmc.org/article/MED/26613620
  4. Lee J, Suh Y, Kim Y. Multidimensional factors affecting homebound older adults: a systematic review. Journal of Nursing Scholarship, 2022. europepmc.org/article/PMC/PMC9299136
  5. Ko Y, Noh W. A scoping review of homebound older people: definition, measurement and determinants. International Journal of Environmental Research and Public Health, 2021. europepmc.org/article/PMC/PMC8069440
  6. Choi NG, Fons B, Gutierrez A, Vences K. Associations of homebound state in older adults with neighborhood social cohesion. Journal of Applied Gerontology, 2026. europepmc.org/article/PMC/PMC12318369
  7. Koizumi S, Ohta A, Kamei M. Homebound older adults who live independently in rural Japan: prevalence and contributing factors during the COVID-19 pandemic. Preventive Medicine Reports, 2024. europepmc.org/article/PMC/PMC10873719
  8. Ho DSM, Yamaguchi K, Ide K, et al. Municipal governments’ long-term care prevention efforts and homebound status of older adults: a three-year cohort study in Japan. Geriatrics & Gerontology International, 2025. europepmc.org/article/PMC/PMC12584943
  9. Cho J. The association between homebound status and social isolation in the United States: a longitudinal study with 9 years of follow-up. International Journal of Aging & Human Development, 2026. europepmc.org/article/MED/42095590
  10. Ankuda CK, Kotwal A, Reckrey J, Harrison KL, Ornstein KA. The experience of homebound older adults during the COVID-19 pandemic. Journal of General Internal Medicine, 2022. europepmc.org/article/PMC/PMC8853401
  11. Hajek A, Soysal P, Gyasi RM, et al. Behind closed doors: homeboundness and psychosocial outcomes. Evidence from a longitudinal study of middle-aged and older adults. Archives of Gerontology and Geriatrics, 2025. europepmc.org/article/MED/39862581
  12. Cohen-Mansfield J, Shmotkin D, Hazan H. The effect of homebound status on older persons. Journal of the American Geriatrics Society, 2010. europepmc.org/article/PMC/PMC3119862
  13. Choi NG, Gutierrez A, Fons B, Vences K, Marti CN. Depressive/anxiety symptoms in homebound older adults: mediation effects of loneliness and psychological well-being. Clinical Gerontologist, 2026. europepmc.org/article/MED/40247655
  14. Chen T, Zhou R, Yao NA, Wang S. Mental health of homebound older adults in China. Geriatric Nursing, 2022. europepmc.org/article/MED/34864541
  15. Thomas KS, Bernard KP, Clark M, et al. Evaluating effects of meal delivery on the ability of homebound older adults to remain in the community via a pragmatic, two-arm, randomized comparative effectiveness trial: study protocol for the Deliver-EE trial. Trials, 2024. europepmc.org/article/PMC/PMC11583665
  16. Balkan E, Gadbois EA, Tucher EL, Bernard KP, Thomas KS. Home-delivered nutrition services for older adults under the Older Americans Act. JAMA Network Open, 2025. europepmc.org/article/PMC/PMC12485633
  17. Choi NG, Pepin R, Marti CN, Stevens CJ, Bruce ML. Improving social connectedness for homebound older adults: randomized controlled trial of tele-delivered behavioral activation versus tele-delivered friendly visits. American Journal of Geriatric Psychiatry, 2020. europepmc.org/article/PMC/PMC8767809
  18. Bruce ML, Pepin R, Marti CN, Stevens CJ, Choi NG. One year impact on social connectedness for homebound older adults: randomized controlled trial of tele-delivered behavioral activation versus tele-delivered friendly visits. American Journal of Geriatric Psychiatry, 2021. europepmc.org/article/PMC/PMC8286332
  19. World Health Organization. Social connection linked to improved health and reduced risk of early death (news release on the report of the WHO Commission on Social Connection). WHO, 2025. www.who.int/news/item/30-06-2025-social-connection-linked-to-improved-heath-and-reduced-risk-of-early-death
  20. Ezeokonkwo FC, Sekula KL, Theeke LA. Loneliness in homebound older adults: integrative literature review. Journal of Gerontological Nursing, 2021. europepmc.org/article/MED/34309447