Estimated Prevalence of Elder Orphans Using National Health and Aging Trends Study

Abstract

Objectives: By 2050, individuals aged 65 years and older will comprise over 20% of the American population. A portion is at risk for becoming elder orphans: older adults living in the community who lack caregivers or surrogates. Methods: Using the first wave of the National Health and Aging Trends Survey, we estimated the prevalence of older adults who reside in the community, who are socially or physically isolated, and who lack caregivers. Individuals who are “at risk” meet all of these criteria, with the exception that they live with and receive caregiving from their spouse. Results: We estimated the elder orphan prevalence for this population to be 2.62% (2.24–3.00), with an additional 21.29% determined to be at risk. Discussion: As the population of the United States ages, an understanding of the prevalence and demographics of elder orphans will be useful to guide policies and services to assist this population.

Keywords

elder orphans aging alone older adults National Health and Aging Trends Survey

Introduction

The population of adults aged 65 years and older in the United States is rapidly expanding, with some estimating that by 2050, this population will nearly double to 83.7 million individuals, representing over 20% of the population of the United States (Ortman et al., 2014). Of this population, 25.9% are expected to live alone in the community, with this percentage increasing to 39% for those aged 85 years or older (Roberts et al., 2018). A portion of these older adults are isolated and do not have friends or family to act as caregivers. These older adults can be considered elder orphans, a term used by Carney et al. (2016) to describe “aged, community-dwelling individuals who are socially and/or physically isolated, without an available known family member or designated surrogate or caregiver.” Although the literature is limited in its study and documentation of elder orphans, the term elder orphan is becoming accepted in the gray literature and among older adults. When used in the healthcare context, older adults recognize the term elder orphan as a means to highlight their kinless status to healthcare providers who may connect them to appropriate community services (Montayre et al., 2018, 2019; Thaggard & Montayre, 2019).

The characteristics used to outline elder orphan status, including solo community residence, social and/or physical isolation, and a lack of caregiving resources, have been extensively studied as individual variables. Previous research has categorized older adults who live alone as an “at risk” group due to lower self-reported health scores and higher likelihood of comorbidities (Kharicha et al., 2007). Additional research has found that perceived social isolation has a strong relationship with poor mental health and has been associated with increased risk of mortality (Cornwell & Waite, 2009; Holt-Lunstad et al., 2015). Both social isolation and living alone are associated with increased likelihood of institutionalization for older adults (Oh et al., 2019). Caregiver availability can also have an impact on long-term health status for older adults, including increased risk of nursing home placement for those who do not have caregivers in the community (Andel et al., 2007; Blackburn et al., 2018). Although related, these three variables are distinct and can occur in the presence or absence of the others. For example, older adults living alone in the community do not necessarily experience social and/or physical isolation (Djundeva et al., 2019). It is the presence of all three variables combined that distinguishes the categorization of elder orphans from each individual risk factor.

Developing a deeper understanding of the population of elder orphans in the United States will allow healthcare and social support systems to target resources to this vulnerable population. While the primary risk factors associated with becoming an elder orphan are known, including childlessness and single marital status (e.g., widow, divorced, and never married), the determination of an estimated prevalence of this population is still in the early stages (Montayre et al., 2018). Using data from the health and retirement study, Carney et al. (2016) estimated the number of older adults who are at high risk for elder orphan status to be 22.6% by using measures of familial contact. These include marital status and the number of children and siblings, with further distinction for close relatives who remain in contact or live within 10 miles of the respondent. Although this method is useful to define available surrogates, the authors acknowledge the limitations of these measures to define the ability of these individuals to be caregivers and to define the caregiving needs of the older adult respondents. This article expands on the previous framework by including measures of household composition, isolation, and caregiving needs of the older adult respondents and applying it to the National Health and Aging Trends Survey (NHATS) to determine the prevalence of elder orphans in the United States and the potential demographic factors that may be associated with elder orphan status.

Methods

Study Design and Population

The NHATS is a longitudinal survey designed to collect a nationally representative sample of Medicare beneficiaries aged 65 years and older in the United States. The study, conducted by Johns Hopkins University and funded by the U.S. National Institute on Aging, has been repeated yearly since 2011 and is compiled as a publicly available dataset. Initial enrollment included 8245 individuals, with oversampling of the oldest age categories (85+ years) and Black older adults, leading to a 71% response rate (Kasper & Freedman, 2012). Further details of sample design are described elsewhere (Montaquila, Freedman, Edwards et al., 2012). The first wave of the NHATS dataset was used for analysis.

Elder Orphan Status

A stepwise approach to determining the prevalence of elder orphans was used, following the established definition of “aged, community-dwelling individuals who are socially and/or physically isolated, without an available known family member or designated surrogate or caregiver” (Carney et al., 2016). Respondents were separated into one of three categories, including “elder orphan,” “at risk,” and “not an elder orphan.”

Of the 8245 respondents in the original sample, 1048 respondents lived in a residential care home or nursing home and were excluded. Of the remaining 7197 living in the community, 517 respondents were excluded for having a proxy respond to the NHATS due to the concern that if the respondents required proxies to assist with NHATS responses, there may be other factors that would affect their status as elder orphans or as at risk. The population included for analysis comprised older adults living in the community who did not require a proxy to complete the NHATS interview (N = 6680). All older adults who were living in a household that included at least two other individuals (e.g., total of 3+ people in household) were considered not elder orphans (N = 1729). Those living alone in the community (N = 2096) were assessed for elder orphan status, and those living in the community with only a spouse (N = 2855) were assessed for being at risk for becoming elder orphans.

Social and physical isolation were established through variables assessing social network size and the number of times per week the respondent left the house. Respondents living alone with a social network of two or fewer individuals or who left the house less than 5 days per week were deemed socially or physically isolated. Married respondents living with only their spouse fitting these criteria of isolation were considered for at risk status.

Available family members or caregivers were determined through variables assessing if the respondent receives unpaid assistance with activities of daily living (ADLs). Respondents living alone who received only paid assistance with ADLs and respondents who required but did not receive any assistance with ADLs were deemed elder orphans. Those who did not require and did not receive assistance with ADLs were categorized as not elder orphans. Married respondents who received only paid assistance, who required but did not receive assistance, or whose only source of assistance was their spouse were considered at risk (Figure 1).

Figure 1.

Diagram of the determination of elder orphan status for older adults living in the community based on living arrangement, isolation, and availability of caregivers.

Data Analysis

Basic demographic characteristics were determined for the elder orphans, at risk, and not an elder orphan categories. These characteristics included gender, age category, race, ethnicity, self-rated health status, highest level of education, and estimated income. Missing rates across covariates were low, around 1% or less. For race and ethnicity, missing values were included in the “other” category and missing values in highest education were included in “less than high school.” Five respondents (.07%) refused to answer the self-rated health variable and were excluded using listwise deletion. For respondents missing estimated income, almost 44%, an imputed income range variable developed by the NHATS study team was used as replacement (Montaquila, Freedman, & Kasper, 2012). Further analysis was performed to determine any demographic factors associated with elder orphan or at risk status. Multinomial logistic regression was used to estimate risk ratios for both “elder orphan” and “at risk” statuses, using the not an elder orphan group as the reference category. Weighting of data was performed to ensure results were representative of Medicare beneficiaries in the United States who were eligible for the NHATS (Montaquila, Freedman, Spillman, et al., 2012). Statistical analyses were conducted using STATA/IC 15 software (StataCorp, College Station, Texas).

Results

Overview

The prevalence of elder orphans in the United States was estimated using data collected from a nationally representative sample of Medicare recipients in the contiguous United States aged 65 years and older. When defined as older adults living alone, who are socially or physically isolated, with unpaid caregiving resources (N = 6680), the prevalence of elder orphans in the United States is 2.62% (2.24–3.00). In addition to this population of elder orphans, 21.29% of older adults are at risk for becoming elder orphans, as their spouse represents their only resource for caregiving.

Sociodemographic Characteristics

Table 1 outlines basic sociodemographic characteristics of the total population included for analysis, with categorization according to elder orphan status. Of the 6680 older adults included for analysis, more than half (55.8%) were female. When adjusted for elder orphan status, 55.62% of elder orphans were female, while only 31.26% of those at risk were female. Elder orphan status varied by age category, with younger older adults being more likely to be categorized as at risk, and at risk and elder orphan categorization becoming more likely in the older age categories. Elder orphan status also varied by race, with Black, non-Hispanic, and Hispanic older adults likely to be categorized as elder orphans, and white, non-Hispanic older adults being most likely to be categorized as at risk. Low income (US$27,600 or less per year), either directly reported or imputed, was associated with increased likelihood of being categorized as elder orphans. Those with higher income (US$64,000 or more per year) were more likely to be in the at risk category. After exclusion of the five respondents who refused to answer the self-rated health variable, the sample size for self-rated health variable analysis was 6675. Those with excellent or very good self-rated health were more likely to be in the at risk category, while those in fair or poor health were more likely to be categorized as elder orphans. Further analysis using multivariate modeling was performed to determine significance levels of elder orphan and at risk categorization with these demographic characteristics in relationship with each other.

Table 1.

Demographic Characteristics of Population of Elder Orphans, Those at Risk of Becoming Elder Orphans, and Those Who are Not Elder Orphans. Not Elder Orphan was Used as Reference Category for Significance Testing.

Weighted %	Total Population	Not Elder Orphans	At Risk		Elder Orphans
	Unweighted, N = 6680	Unweighted, N = 1729	Unweighted, N = 2855		Unweighted, N = 2096
			%	p-Value	%	p-Value
Gender
Male	44.20	35.78	68.74	Reference	44.38	Reference
Female	55.80	64.22	31.26	<.001	55.62	.063
Age category (years)
65–69	30.13	28.41	35.84	Reference	23.01	Reference
70–74	26.17	25.71	27.93	.070	21.53	.609
75–79	9.40	19.37	19.48	.010	19.49	.779
80–84	4.16	14.85	11.88	<.001	17.12	.586
85–89	7.41	8.36	4.04	<.001	13.85	.191
90+	2.73	3.30	.82	<.001	5.01	.306
Race/ethnicity
White, non-Hispanic	81.11	79.49	86.60	Reference	73.01	Reference
Black, non-Hispanic	8.02	9.20	4.34	<.001	10.52	.866
Hispanic	6.73	7.06	5.30	.001	11.41	.092
Other (Am. Indian, Asian, and Hawaiian)	4.14	4.25	3.76	.080	5.05	.479
Self-rated healtha
Excellent	15.49	14.44	19.08	Reference	9.79	Reference
Very good	30.49	29.10	35.19	.380	23.78	.824
Good	30.61	31.27	28.77	<.001	30.00	.285
Fair	17.55	18.78	13.11	<.001	26.61	.085
Poor	5.85	6.41	3.86	<.001	9.82	.111
Highest education
Less than high school	21.43	22.79	16.97	Reference	26.28	Reference
High school graduate	27.08	27.31	25.85	<.001	33.20	.876
Associate’s or some college	26.61	26.52	27.23	<.001	22.81	.783
Bachelor’s or higher	24.88	23.38	29.95	<.001	17.71	.300
Estimated yearly income
Less than US$27,600	33.07	39.20	17.00	Reference	65.48	Reference
US$27,600–US$41,999	15.82	15.57	16.48	<.001	14.54	.027
US$42,000–US$63,999	16.82	15.99	18.95	<.001	13.01	.051
US$64,000–US$107,999	19.27	16.60	26.49	<.001	1.38	.002
US$108,000 or more	15.03	12.65	21.09	<.001	5.58	.023

Note. Percentages are weighted. Percentages sum to 100% with rounding error.

Unweighted N = 6675 for self-rated health only.

Bold values represent p values that were significant at the 0.05 level.

Relationship with Elder Orphan Status

Multinomial logistic regression was performed to determine the risk ratio of being categorized as an elder orphan or at risk for selected demographic factors (Table 2). Both categories used not an elder orphan as the reference category. Male gender was associated with increased risk for both elder orphan and at risk statuses. Demographics significantly associated with a decreased risk ratio for at risk categorization including age 85 years and older, Black, non-Hispanic race, “fair” or “poor” levels of self-rated health, and associate’s degree level of education and higher. Increasing levels of income (US$27,600 and higher) were associated with increasing risk for at risk status, while income at US$42,000 and higher decreased risk for elder orphan status.

Table 2.

Risk Ratios of Categorization in at Risk or Elder Orphan Status. Not Elder Orphan was Used as Reference Category.

	At Risk			Elder Orphan
	Risk Ratio	95% CI	p Value	Risk Ratio	95% CI	p Value
Gender (female)	.314	.275–.358	<.001	.479	.315–.728	.001
Age category (years)
65–69	1	Reference		1	Reference
70–74	1.012	.861–1.190	.887	1.069	.600–1.906	.820
75–79	.961	.803–1.151	.666	.858	.448–1.645	.645
80–84	.886	.717–1.096	.266	1.218	.631–2.350	.556
85–89	.589	.432–.805	.001	1.825	.895–3.720	.098
90+	.414	.222–.771	.005	1.38	.399–4.788	.609
Race/ethnicity
White, non-Hispanic	1	Reference		1	Reference
Black, non-Hispanic	.569	.426–.761	<.001	.843	.389–1.827	.666
Hispanic	.880	.666–1.164	.371	1.537	.814–2.902	.185
Other (Am. Indian, Asian, and Hawaiian)	.907	.651–1.265	.566	.873	.301–2.532	.802
Self-rated health
Excellent	1	Reference		1	Reference
Very good	1.013	.842–1.219	.893	1.692	.710–4.030	.235
Good	.876	.722–1.064	.181	1.525	.639–3.637	.342
Fair	.768	.608–.970	.026	2.267	.940–5.472	.069
Poor	.635	.451–.893	.009	2.106	.752–5.896	.156
Highest education
Less than high school	1	Reference		1	Reference
High school graduate	.891	.726–1.093	.268	1.466	.827–2.600	.190
Associate’s or some college	.770	.625–.949	.014	.975	.502–1.895	.941
Bachelor’s or higher	.562	.451–.700	<.001	1.783	.897–3.545	.099
Estimated yearly income
Less than US$27,600	1	Reference		1	Reference
US$27,600–US$41,999	2.164	1.758–2.664	<.001	.547	.298–1.005	.052
US$42,000–US$63,999	2.631	2.137–3.239	<.001	.467	.242–.900	.023
US$64,000–US$107,999	3.434	2.799–4.212	<.001	.048	.008–.276	.001
US$108,000 or more	3.425	2.735–4.289	<.001	.238	.090–.629	.004

Bold values represent p values that were significant at the 0.05 level.

Discussion

Summary of Main Findings

The purpose of this study was to identify the prevalence of elder orphans in the United States. Using the NHATS, we estimated that 2.62% of older adults in the United States are elder orphans, with an additional 21.29% of older adults at risk for becoming elder orphans. Demographic factors associated with elder orphan or at risk status were compared to the demographic factors of those who were not categorized as elder orphans. Although women are expected to outlive men and presumably become elder orphans upon the death of their spouse, female gender was associated with reduced risk for both elder orphan and at risk categorization (World Health Organization, 2019). This may be due to women’s increased social network size and connection to children, which can reduce risk for social or physical isolation during older age (Fischer, 1982; McLaughlin et al., 2010). Black race was associated with decreased risk for categorization as at risk. This is in line with previous research that indicates that although Black individuals may have smaller social network size, contact within that social network may be more frequent (Ajrouch et al., 2005). This frequency may decrease levels of social and physical isolation, which may reduce risk for elder orphan status.

Interestingly, several demographic factors have opposite relationships with the at risk and elder orphan categories. Fair health and increasing education both increase risk for elder orphan status, while decreasing risk for at risk categorization. Higher levels of income increase risk for being categorized as at risk and decrease risk for elder orphan categorization. For this characteristic, it is thought that additional resources may allow for married couples to hire outside caregiving resources, without the need for additional unpaid familial support. This balance may change when older adults age alone, where increased financial resources could have a relationship with living arrangements, isolation, or availability of unpaid caregivers.

Limitations

While the results are valuable, there are several limitations to this study. First, the determination of available caregiving resources is based on assistance with ADLs. This does not take into account older adults who may need assistance with other tasks of daily life (e.g., transportation and home maintenance), but who may not need assistance with ADLs. This study also did not take into account available friends and family who do not help with ADLs, but who may be willing to act as surrogate or caregiver during a time of crisis for the older adult. Second, this study does not consider older adults living in nursing homes or assisted living facilities. With the exception of the care received in these institutions, these individuals may otherwise meet the elder orphan criteria of isolation and lack of a surrogate. This can also apply to those who were excluded for having a proxy answer the NHATS. This exclusion assumes that the need for a proxy is indicative of a larger set of factors that may affect the respondent’s elder orphan status and is worth further exploration. Additionally, while having high ADL assistance needs is concerning, some individuals may not report difficulties if they do not engage in activities that cause them to note their own limitations. This framework also does not account for those who were unable to participate in the NHATS process. It is possible that some of the characteristics that define elder orphans, including isolation and a lack of assistance, may have caused some older adults to not be represented in the analysis of the dataset. Further studies to examine characteristics of older adults living in institutions and nonrespondents may be warranted.

Contribution to Literature

The current literature regarding elder orphans is limited. The prevalence of elder orphans in the United States estimated by this study is in line with previous estimates made by Carney et al. This study builds upon previous research by adding a distinction between those suspected to be elder orphans and those who are at risk for becoming elder orphans should their spouse die. This study also adds context of demographic characteristics that may place an older adult at risk for elder orphan status. We used the first wave of the NHATS to set a baseline level of elder orphans and at-risk older adults. Future analytic work can utilize additional waves to examine longitudinal associations between orphan status, risk factors, and outcomes. Suggested directions for further research include additional analysis of the elder orphan and at risk categories to determine variation in health status, mortality risk, and caregiver status.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Drs Clouston and Smith are supported by the National Institute on Aging (NIH/NIA R01 AG058595).

ORCID iD

Regina Roofeh

References

Ajrouch

K. J.

Blandon

A. Y.

Antonucci

T. C.

(2005). Social networks among men and women: The effects of age and socioeconomic status. The Journals of Gerontology: Series B, 60(6), S311-S317. doi:10.1093/geronb/60.6.S311

Andel

Hyer

Slack

(2007). Risk factors for nursing home placement in older adults with and without dementia. Journal of Aging and Health, 19(2), 213-228. doi:10.1177/0898264307299359

Blackburn

Albright

K. C.

Haley

W. E.

Howard

V. J.

Roth

D. L.

Safford

M. M.

Kilgore

M. L.

(2018). Men lacking a caregiver have greater risk of long-term nursing home placement after stroke. Journal of the American Geriatrics Society, 66(1), 133-139. doi:10.1111/jgs.15166

Carney

M. T.

Fujiwara

Emmert

B. E.

Liberman

T. A.

Paris

(2016). Elder orphans hiding in plain sight: A growing vulnerable population. Current Gerontology and Geriatrics Research, 2016, 1-11. doi:10.1155/2016/4723250

Cornwell

E. Y.

Waite

L. J.

(2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50(1), 31-48. doi:10.1177/002214650905000103

Djundeva

Dykstra

P. A.

Fokkema

(2019). Is living alone “aging alone”? Solitary living, network types, and well-being. The Journals of Gerontology: Series B, 74(8), 1406-1415. doi:10.1093/geronb/gby119

Fischer

(1982). To dwell among friends: personal networks in town and city. University of Chicago Press.

Holt-Lunstad

Smith

T. B.

Baker

Harris

Stephenson

(2015). Loneliness and social isolation as risk factors for mortality. Perspectives on Psychological Science, 10(2), 227-237. doi:10.1177/1745691614568352

Kasper

Freedman

(2012). National health and aging trends study round 1 user guide: Final release. Johns Hopkins University School of Public Health. www.NHATS.org

10.

Kharicha

Iliffe

Harari

Swift

Gillmann

Stuck

A. E.

(2007). Health risk appraisal in older people 1: Are older people living alone an “at-risk” group? The British Journal Of General Practice: The Journal of the Royal College of General Practitioners, 57(537), 271-276.

11.

McLaughlin

Vagenas

Pachana

N. A.

Begum

Dobson

(2010). Gender differences in social network size and satisfaction in adults in their 70s. Journal of Health Psychology, 15(5), 671-679. doi:10.1177/1359105310368177

12.

Montaquila

Freedman

Edwards

Kasper

(2012). National health and aging trends study round 1 sample design and selection. Johns Hopkins University School of Public Health, NHATS Technical Paper #1. www.NHATS.org

13.

Montaquila

Freedman

Kasper

(2012). National health and aging trends study round 1 income imputation. Johns Hopkins University School of Public Health, NHATS Technical Paper #3. www.NHATS.org

14.

Montaquila

Freedman

Spillman

Kasper

(2012). National health and aging trends study development of round 1 survey weights. Johns Hopkins University School of Public Health, NHATS Technical Paper #2. www.NHATS.org

15.

Montayre

Thaggard

(2018). The elder orphan in healthcare settings: An integrative review. Journal of Population Ageing, 12, 515-532. doi:10.1007/s12062-018-9222-x

16.

Montayre

Thaggard

Carney

(2019). Views on the use of the term ‘elder orphans’: A qualitative study. Health & Social Care in the Community, 28(2), 341-346. doi:10.1111/hsc.12865

17.

Patel

Boscardin

W. J.

Max

Stephens

Ritchie

C. S.

Smith

A. K.

(2019). Social support and patterns of institutionalization among older adults: A longitudinal study. Journal of the American Geriatrics Society, 67(12), 2622-2627. doi:10.1111/jgs.16184

18.

Ortman

J. M.

Velkoff

V. A.

Hogan

(2014). An aging nation: The older population in the United States. Current population reports: Population estimates and projections.

19.

Roberts

Ogunwole

Blakeslee

Rabe

(2018). The population 65 years and older in the United States 2016.

20.

Thaggard

Montayre

(2019). Elder orphans’ experiences of advance planning and informal support network. SAGE Open, 9(3), 215824401986537.

21.

World Health Organization . (2019). World health statistics 2018: Monitoring health for the sustainable development goals. www.who.int/gho/publications/world_health_statistics/2018/en