At-a-glance – Living arrangements and health status of seniors in the 2018 Canadian Community Health Survey

Abstract

Currently, 1 in 3 Canadian seniors meet the criteria for successful aging, which include low probability of disease and disability, high cognitive and physical ability and active engagement in life. The sociodemographic characteristic of living alone can identify high-risk seniors, due to its association with lower social support and interactions, thus increasing susceptibility to negative health outcomes in older age. However, limited data exists on the living arrangements of Canadian seniors. In this analysis, we present sociodemographic characteristics and measures of health and social well-being of seniors by living arrangement. This information should be used to identify and support vulnerable seniors and increase the prevalence of healthy aging among Canadians.

Highlights

An understanding of living arrangements may help those who develop intervention programs better target seniors at higher risk for negative health outcomes in older age.

We found that seniors who were female, older, lower-income, divorced or separated, living in a population centre, renters and less educated were most likely to live alone.

Seniors who were living alone were also more likely to report poor perceived health and social well-being.

These results may be useful in targeting policies and programs aimed to improve health outcomes among seniors.

Introduction

Due to increasing life expectancy¹ and decreasing fertility rates,² the proportion of seniors in Canada is growing faster than ever. In 2011, this trend began to accelerate as the first baby boomers turned 65 years old. Seniors now total over 6 million (1 in 6) persons in Canada,³ and outnumber children aged 0 to 14 years for the first time.⁴ Moreover, current trends suggest that this age group will continue to grow, with the proportion of seniors set to rise to 1 in 5 by 2024 and 1 in 4 by 2055.³ Implementing policies and programs to promote health in older age will be of increasing importance, as only about 1 in 3 seniors currently meet the criteria for successful aging,⁵ defined by Rowe and Kahn⁶ as low probability of disease and disability, high cognitive and physical capacity and active engagement in life.

Recent international studies have shown that living arrangements of seniors are an important determinant of healthy aging, as they predict social support and interactions. Seniors living with a spouse or partner were more likely to have lower incidence rates of dementia,⁷ better mental health, and fewer limitations due to multimorbidity on their involvement in all aspects of life (including social life, housework and leisure-time activities)⁸; those living with family demonstrated lower rates of chronic and acute diseases⁹; and those living with others reported better mental health, social support and engagement in more physical activities, compared to those living alone.¹⁰ However, there are only a few studies that have assessed the living arrangements of Canadian seniors,^{11, 12, 13, 14} and none aimed to identify the subpopulations of seniors who are more likely to live alone, putting them at higher risk for negative health outcomes in older age. As well, half of those studies focussed on Asian-Canadian seniors alone.^{13, 14} Recent and complete data on this topic are necessary to identify and address gaps in the promotion of healthy aging among seniors.

The purpose of this brief analysis was to examine the living arrangements of Canadian seniors in the most recent (2018) Canadian Community Health Survey (CCHS) by sociodemographics and health and social well-being, stratified by sex.

Methods

The CCHS is an annual, cross-sectional survey that collects representative data on the health status and determinants of the noninstitutionalized Canadian population in all provinces.¹⁵ Those living in the territories were excluded from the annual component due to small samples and non-representativeness.¹⁵ We employed data from the 2018 CCHS cycle on individuals aged 65 and over living in private dwellings who responded to the living arrangement question. Information on living arrangements, sociodemographics (age group, race, health region–level household income ratio [quintiles], marital group, region of residence, classification of region [population centre vs. rural area], dwelling ownership, participant education and household size) and self-reported well-being (perceived health, perceived mental health, life satisfaction and sense of community belonging) were used for this analysis. The household income ratio measures a participant’s household income relative to other residents in their health region, adjusting for household and community size.¹⁵ A population centre is defined as an area with a population of at least 1000 and a density of at least 400 persons per km²; all other areas are considered rural.¹⁵ Descriptive statistics were stratified by sex and weighted using bootstrap methods to produce data representative of the Canadian senior population living in the provinces. Please note that in 2016, the CCHS asked respondents whether they were male or female; we recognize that perceptions and behaviours are influenced by a person’s gender and should be taken into consideration when interpreting our results. Data are shown as row percent with bootstrapped 95% confidence intervals (95% CI) and compared using the Rao-Scott ??² test. All analyses were run on SAS Enterprise Guide version 5.1 (SAS Institute Inc., Cary, NC, USA).

Results

In total, 8261 female and 6532 male seniors in the 2018 CCHS were included in the analyses. Data on the living arrangements of females and males were weighted and stratified by selected characteristics in Table 1 and Table 2, respectively. Females were almost twice as likely to live alone (35.7% vs. 19.1%) and 1.5-fold less likely to live with a partner (48.9% vs. 71.3%) compared to males. Among both sexes, the prevalence of living alone was highest for participants who were older, lower-income, divorced or separated, living in a population centre, renters, and less educated. The opposite set of characteristics were associated with living with a partner (younger, higher-income, married or common-law, living in a rural area, owners and more educated). Asians were most likely to live with children, relatives or nonrelatives and least likely to live alone (for both sexes). For seniors of both sexes, living arrangement did not differ by region. However, some sex differences were noted. Among females, White participants were more likely to live with a partner than Asians and those of “other” ethnicity (e.g. Black, Latin American and Arab), though no racial differences were found among males living with a partner. Female renters were much more likely to live alone than with a partner (62.4% vs. 25.0%), though no such tendency was displayed among male renters (45.5% vs. 47.9%). Finally, the prevalence of living with a partner increased with higher education for both sexes, though the largest increase was between females with less than a high school education and females who graduated from high school (females: 38.6% vs. 52.7; males: 64.9% vs. 74.1%).

Table 1. Living arrangements of 8261 female seniors in the 2018 CCHS, weighted and stratified by selected characteristics

Characteristics	Living alone (n = 4313)	Living with partnera (n = 3214)	Otherb (n = 734)	p-value
Weighted N (%)	1 170 194 (35.7)	1 604 305 (48.9)	507 312 (15.5)	N/A
Age group				< .001
65–74	28.4 (26.3–30.4)	57.1 (54.8–59.5)	14.5 (12.3–16.7)
75–84	42.6 (39.5–45.7)	43.0 (39.7–46.3)	14.5 (11.5–17.4)
85 or over	58.4 (53.1–63.6)	18.3 (13.7–22.9)	23.3 (18.3–28.4)
Race				< .001
White	36.8 (35.0–38.6)	50.9 (49.1–52.7)	12.3 (10.8–13.9)
Indigenous	34.1 (25.6–42.6)	51.3 (41.0–61.7)	14.6 (7.6–21.6)
Asianc	16.0 (10.6–21.4)	38.7 (30.1–47.2)	45.3 (35.9–54.8)
Otherd	38.6 (28.2–49.0)	34.3 (24.3–44.3)	27.1 (16.3–37.8)
Health region–level household income ratio (quintiles)e				< .001
1	54.2 (50.9–57.5)	32.1 (29.0–35.2)	13.7 (11.0–16.4)
2	35.7 (32.5–38.9)	48.0 (44.6–51.4)	16.3 (13.1–19.4)
3	29.1 (25.8–32.5)	55.9 (51.9–60.0)	14.9 (10.9–18.9)
4	22.6 (19.3–26.0)	58.9 (53.7–64.1)	18.5 (13.1–23.9)
5	20.6 (16.8–24.5)	64.9 (59.8–70.1)	14.4 (10.6–18.3)
Marital group				N/Af
Married or common-law	1.6 (1.2–1.9)	92.6 (91.2–94.0)	5.8 (4.4–7.2)
Widowed	73.7 (70.3–77.2)	0	26.3 (22.8–29.7)
Divorced or separated	76.9 (71.1–82.8)	0	23.1 (17.2–28.9)
Single	66.9 (58.2–75.6)	0	33.1 (24.4–41.8)
Region of residenceg				0.2
Atlantic	34.4 (31.0–37.9)	53.3 (49.6–57.0)	12.3 (9.5–15.1)
Central	35.8 (33.5–38.0)	48.0 (45.5–50.5)	16.2 (14.0–18.5)
Prairies	37.6 (33.9–41.3)	49.5 (45.7–53.2)	12.9 (9.0–16.8)
West	33.8 (30.3–37.3)	50.0 (45.5–54.5)	16.2 (12.5–19.9)
Classification of regionh				< .001
Population centre	37.7 (35.7–39.7)	46.2 (44.0–48.4)	16.1 (14.1–18.1)
Rural area	26.7 (24.5–28.8)	60.6 (57.8–63.3)	12.7 (10.4–15.0)
Dwelling ownership				< .001
Owned	26.9 (25.3–28.6)	56.7 (54.7–58.7)	16.4 (14.5–18.2)
Rented	62.4 (58.4–66.4)	25.0 (21.7–28.3)	12.6 (9.1–16.2)
Personal education				< .001
Less than HS	41.0 (37.5–44.4)	38.6 (35.1–42.1)	20.4 (16.9–23.9)
HS grad	35.3 (32.0–38.5)	52.7 (49.0–56.4)	12.0 (9.4–14.6)
Postsecondary grad	33.3 (30.9–35.7)	52.5 (49.7–55.2)	14.2 (11.5–16.9)
Abbreviations: CCHS, Canadian Community Health Survey; CI, confidence interval; HS, high school; N/A, not applicable. Note: Data are row % (95% CI), unless otherwise stated. Includes living with or without children. a Includes living with children, relatives and nonrelatives. b Includes South Asian, West Asian, Southeast Asian, Chinese, Korean, Japanese and Filipino. c Includes Black, Latin American, Arab, other racial background and multiple ethnicities. d Distribution of participants in each health region based on the adjusted ratio of total household income over the low-income cut-off corresponding to household and community size. e No tests can be computed for the table since at least one cell has 0 frequency. f Atlantic includes Prince Edward Island, New Brunswick, Nova Scotia and Newfoundland and Labrador; Central includes Quebec and Ontario; Prairies includes Alberta, Manitoba and Saskatchewan; and West includes British Columbia. g A population centre is defined as an area with a population of at least 1000 and a density of at least 400 persons per km2; all other areas are considered rural. h

Table 2. Living arrangements of 6532 male seniors in the 2018 CCHS, weighted and stratified by selected characteristics

Characteristics	Living alone (n = 2105)	Living with partnera (n = 4065)	Otherb (n = 362)	p-value
Weighted N (%)	540 770 (19.1)	2 022 160 (71.3)	273 833 (9.7)	N/A
Age group				< .001
65–74	17.5 (15.9–19.0)	73.4 (71.2–75.6)	9.1 (7.3–10.9)
75–84	20.0 (17.6–22.4)	71.1 (68.1–74.1)	8.9 (6.6–11.1)
85 or over	28.3 (23.3–33.4)	54.5 (48.0–61.0)	17.1 (10.9–23.4)
Race				< .001
White	20.6 (19.1–22.0)	72.0 (70.3–73.8)	7.4 (6.1–8.7)
Indigenous	20.0 (11.4–28.6)	63.8 (52.8–74.8)	16.2 (7.0–25.4)
Asianc	8.5 (3.8–13.2)	67.8 (58.7–76.9)	23.7 (15.2–32.2)
Otherd	13.7 (7.4–20.1)	69.6 (60.0–79.2)	16.7 (8.6–24.8)
Health region–level household income ratio (quintiles)e				< .001
1	30.7 (27.4–34.0)	57.8 (54.0–61.7)	11.5 (8.6–14.3)
2	18.0 (15.5–20.4)	71.0 (67.5–74.6)	11.0 (7.8–14.2)
3	15.8 (13.6–18.0)	74.4 (70.8–77.9)	9.9 (6.5–13.2)
4	12.5 (10.0–15.0)	80.9 (77.6–84.3)	6.6 (4.1–9.1)
5	14.6 (11.6–17.5)	78.2 (74.2–82.1)	7.3 (4.4–10.2)
Marital group				N/Af
Married or common-law	1.3 (0.9–1.6)	92.1 (90.7–93.5)	6.6 (5.2–8.0)
Widowed	78.4 (73.7–83.2)	0	21.6 (16.8–26.3)
Divorced or separated	83.0 (77.3–88.6)	0	17.0 (11.4–22.7)
Single	78.0 (69.7–86.2)	0	22.0 (13.8–30.3)
Region of residenceg				.07
Atlantic	18.5 (15.5–21.6)	75.1 (71.8–78.5)	6.3 (4.2–8.4)
Central	19.6 (17.9–21.3)	71.2 (68.9–73.5)	9.2 (7.3–11.1)
Prairies	16.1 (13.6–18.5)	73.2 (69.5–76.9)	10.7 (7.3–14.0)
West	20.0 (16.7–23.2)	67.8 (63.3–72.3)	12.2 (8.3–16.2)
Classification of regionh				.02
Population centre	19.6 (18.0–21.2)	70.2 (68.1–72.2)	10.2 (8.5–11.9)
Rural area	17.2 (15.3–19.0)	75.1 (72.6–77.6)	7.7 (5.7–9.8)
Dwelling ownership				< .001
Owned	13.6 (12.4–14.7)	76.3 (74.6–78.1)	10.1 (8.5–11.7)
Rented	45.5 (40.8–50.1)	47.9 (42.9–52.9)	6.7 (4.0–9.3)
Personal education				< .001
Less than HS	23.2 (20.5–25.9)	64.9 (61.2–68.5)	12.0 (8.8–15.1)
HS grad	22.8 (19.2–26.3)	69.2 (65.2–73.1)	8.1 (5.3–10.8)
Postsecondary grad	17.0 (15.3–18.6)	74.1 (71.8–76.3)	9.0 (7.1–10.8)
Abbreviations: CCHS, Canadian Community Health Survey; CI, confidence interval; HS, high school; N/A, not applicable. Note: Data are row % (95% CI), unless otherwise stated. Includes living with or without children. a Includes living with children, relatives and nonrelatives. b Includes South Asian, West Asian, Southeast Asian, Chinese, Korean, Japanese and Filipino. c Includes Black, Latin American, Arab, other racial background and multiple ethnicities. d Distribution of participants in each health region based on the adjusted ratio of total household income over the low-income cut-off corresponding to household and community size. e No tests can be computed for the table since at least one cell has 0 frequency. f Atlantic includes Prince Edward Island, New Brunswick, Nova Scotia and Newfoundland and Labrador; Central includes Quebec and Ontario; Prairies includes Alberta, Manitoba and Saskatchewan; and West includes British Columbia. g A population centre is defined as an area with a population of at least 1000 and a density of at least 400 persons per km2; all other areas are considered rural. h

Measures of perceived health and social well-being were also stratified by sex and living arrangement (Table 3). Across all four measures, seniors living with a partner were less likely to report poor health and social well-being. Compared to those living with a partner, females living alone or living with children, relatives or nonrelatives reported poorer general health and mental health; further, females living alone reported lower life satisfaction, while those living with children, relatives or nonrelatives had a weaker sense of community belonging. Among males, those living alone reported poorer mental health, life satisfaction and sense of community belonging compared to those living with a partner. Perceived general health did not differ between males in the three living arrangements. Further, males living with children, relatives or nonrelatives were no more likely than those living with partners to report poor health and social well-being across the four measures.

Table 3. Measures of perceived health and social well-being among participants of the 2018 CCHS, weighted and stratified by sex and living arrangement

Characteristics	Perceived general health		Perceived mental health		Life satisfaction		Sense of community belonging
	Less than very good	Very good or excellent	Less than very good	Very good or excellent	Less than satisfied	Satisfied or very satisfied	Somewhat or very weak	Somewhat or very strong
Females	p < .001		p = .001		p < .001			p < .001
Living arrangement
Living alone	53.4 (51.1–55.7)	46.6 (44.3–48.9)	31.2 (28.9–33.6)	68.8 (66.4–71.1)	11.9 (10.3–13.6)	88.1 (86.4–89.7)	24.8 (22.7–26.9)	75.2 (73.1–77.3)
Living with partnera	46.9 (44.3–49.6)	53.1 (50.4–55.7)	25.9 (23.6–28.1)	74.1 (71.9–76.4)	7.1 (5.7–8.4)	92.9 (91.6–94.3)	21.7 (19.4–24.1)	78.3 (75.9–80.6)
Otherb	61.7 (55.2–68.3)	38.3 (31.7–44.8)	35.2 (28.8–41.7)	64.8 (58.3–71.2)	11.7 (8.1–15.2)	88.3 (84.8–91.9)	35.7 (29.1–42.2)	64.3 (57.8–70.8)
Males	p = .01		p = .007		p < .001			p = .003
Living arrangement
Living alone	56.5 (53.2–59.8)	43.5 (40.2–46.8)	31.7 (28.9–34.6)	68.3 (65.4–71.1)	13.0 (11.1–15.0)	87.0 (85.0–88.9)	29.7 (26.7–32.7)	70.3 (67.3–73.3)
Living with partnera	51.6 (49.2–54.0)	48.4 (46.0–50.8)	25.0 (22.8–27.2)	75.0 (72.8–77.2)	6.2 (4.9–7.4)	93.8 (92.6–95.1)	22.0 (19.9–24.1)	78.0 (75.9–80.1)
Otherb	61.1 (53.1–69.1)	38.9 (30.9–46.9)	30.2 (22.3–38.1)	69.8 (61.9–77.7)	8.7 (4.9–12.5)	91.3 (87.5–95.1)	27.6 (19.1–36.1)	72.4 (63.9–80.9)
Abbreviations: CCHS, Canadian Community Health Survey; CI, confidence interval. Note: Data are row % (95% CI), unless otherwise stated. Includes living with or without children. a Includes living with children, relatives and nonrelatives. b

Discussion

Using the 2018 Canadian Community Health Survey data, we found that nearly half of seniors were living with a partner (49.2% in total). Other studies have identified that living with others may be especially beneficial for healthy aging, and living alone may be detrimental.^{7, 8, 9, 10} Still, the prevalence of living alone may be rising among this age group. In an analysis of the 2011 census, other researchers found that 31.5% of females and 16.0% of males aged 65 or over lived alone¹¹; those numbers increased to 33.0% and 17.4% in the 2016 census,¹² further rising to 35.7% and 19.1% in our 2018 analysis of the CCHS.

Seniors who were female, older, lower-income, divorced or separated, living in a population centre, renters, and less educated were most likely to live alone. Similar results have been demonstrated elsewhere. Female seniors have consistently been found to have a greater likelihood of living alone, both in Canada^{11, 12} and in other countries,^{16, 17} in part because of their greater life expectancy compared to opposite-sex partners.¹¹ Seniors with low income^{17, 18} or low education¹⁶ have been reported in other studies to be more likely to live alone, which may be a consequence of their inability to afford the high cost of assisted living facilities.¹⁹ Moreover, seniors living in urban areas were more likely to live alone, which may be due to the disparity in supports and services available to seniors in rural areas²⁰; in fact, the most-used services for seniors are senior centres, homemaker services and transportation services,²¹ all of which are more likely to be available in urban areas. Finally, we replicated the result that living alone was associated with poorer perceived health and social well-being among seniors, though the temporality of this association and the others mentioned cannot be determined from this cross-sectional analysis. Still, identifying this vulnerable senior subpopulation will help us more effectively develop policies and programs to promote healthy aging. Of course, other factors, such as desire to live alone, loneliness (vs. solitude) and social capital, should all be taken into consideration to focus more specifically on those at highest risk.

Strengths and limitations

This analysis is based on cross-sectional questionnaire data, making causal inferences problematic. Moreover, data on seniors residing in the territories or in institutions were not included, meaning that these data cannot be used to infer living arrangements of seniors in those areas or circumstances. Nevertheless, the sample size used is large and has been weighted in an attempt to be representative of the Canadian, provincial, noninstitutionalized population. Further, this analysis uses the most recent data available on seniors in Canada.

Conclusion

Our analysis found that seniors who were female, older, lower-income, divorced or separated, living in a population centre, renters, less educated and who demonstrated poor perceived health and social well-being were most likely to live alone and potentially most vulnerable to negative health outcomes in aging. This information could help programs and policies identify and target older adults at higher risk of negative health outcomes due to their living arrangements, with the aim of increasing the prevalence of healthy aging among Canadian seniors. As an example, health policy makers could promote the development of community programs that increase the social participation and inclusion of older females who live alone, for the purpose of increasing their sense of community belonging.

Acknowledgements

We would like to acknowledge the contributions of Kerry Anderson, Patti Gorr and Rachel Milliken from the Aging and Seniors Unit of the Centre for Health Promotion at PHAC. Their thorough understanding of the issues facing seniors in Canada helped guide the analysis and the interpretation of the findings. As well, we would like to acknowledge Dr. Howard Morrison for his professional advice.

Conflicts of interest

The authors have no conflicts of interest to disclose. This research did not receive any funding.

Authors’ contributions and statement

All authors conceived and designed the methods. SAS conducted the analyses and wrote the initial draft. All authors interpreted the results and critically reviewed the paper. All authors read and approved the final version of the manuscript.

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.

References

1 Shumanty R. Report on the demographic situation in Canada. Mortality: overview, 2014 to 2016. Ottawa (ON): Statistics Canada [Catalogue No.: 91-209-X]; 2018. 14 p.

2 Provencher C, Milan A, Hallman S, et al. Report on the demographic situation in Canada. Fertility: overview, 2012 to 2016. Ottawa (ON): Statistics Canada [Catalogue No.: 91-209-X]; 2018. 19 p.

3 Statistics Canada. Annual demographic estimates: Canada, provinces and territories, 2018. Analysis: population by age and sex [Internet]. Ottawa (ON): Statistics Canada; 2018 [cited 2019 May 29]. Available from: https://www150.statcan.gc.ca/n1/pub/91-215-x/2018002/sec2-eng.htm

4 Statistics Canada. The Daily—Canada’s population estimates: age and sex, July 1, 2018 [Internet]. Ottawa (ON): Statistics Canada; 2019 Jan 25 [cited 2019 May 29]. Available from: https://www150.statcan.gc.ca/n1/daily-quotidien/190125/dq190125a-eng.htm

5 Meng X, D’Arcy C. Successful aging in Canada: prevalence and predictors from a population-based sample of older adults. Gerontology. 2014;60(1):65-72.

6 Rowe JW, Kahn RL. Successful aging. Gerontologist. 1997;37(4):433-40.

7 Arai A, Katsumata Y, Konno K, et al. Sociodemographic factors associated with incidence of dementia among senior citizens of a small town in Japan. Care Manag J. 2004;5(3):159-65.

8 Müller F, Hagedoorn M, Tuinman MA. Chronic multimorbidity impairs role functioning in middle-aged and older individuals mostly when non-partnered or living alone. PLOS ONE [Internet]. 2017 [cited 2019 May 29];12(2):e0170525. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0170525

9 Agrawal S. Effect of living arrangement on the health status of elderly in India: findings from a national cross sectional survey. Asian Popul Stud. 2012;8(1):87-101.

10 Yu C-Y, Hou S-I, Miller J. Health for older adults: the role of social capital and leisure-time physical activity by living arrangements. J Phys Act Health. 2018;15(2):150-8.

11 Statistics Canada. Living arrangements of seniors [Internet]. Ottawa (ON): Government of Canada; 2012 [cited 2019 May 21]. Available from: https://www12.statcan.gc.ca/census-recensement/2011/as-sa/98-312-x/98-312-x2011003_4-eng.cfm

12 Tang J, Galbraith N, Truong J. Living alone in Canada. Ottawa (ON): Statistics Canada [Catalogue No.: 75-006-X]; 2019.

13 Gee EM. Living arrangements and quality of life among Chinese Canadian elders. Soc Indic Res. 2000;51(3):309-29.

14 Ng CF, Northcott HC, Abu-Laban SM. Housing and living arrangements of South Asian Immigrant Seniors in Edmonton, Alberta. Can J Aging. 2007;26(3):185-94.

15 Statistics Canada. Canadian Community Health Survey (CCHS) – Annual Component user guide. 2018 Microdata File. Ottawa (ON): Government of Canada; 2018.

16 You KS, Lee H. The physical, mental, and emotional health of older people who are living alone or with relatives. Arch Psychiatr Nurs. 2006;20(4):193-201.

17 Beckett MK, Elliott MN, Haviland AM, Burkhart Q, Gaillot S, Montfort D, et al. Living alone and patient care experiences: the role of gender in a national sample of Medicare beneficiaries. J Gerontol A. 2015;70(10):1242-7.

18 Administration on Aging. A profile of older Americans: 2011. Washington (DC): US Department of Health and Human Services; 2011. Available from: https://acl.gov/sites/default/files/Aging%20and%20Disability%20in%20America/2011profile.pdf

19 White G, Singh T, Caine K, Connelly K. Limited but satisfied: low SES older adults’ experiences of aging in place. In: Proceedings of the 9th International Conference on Pervasive Computing Technologies for Healthcare. 2015 May 20-23; Istanbul, Turkey:121-8.

20 Siconolfi D, Shih RA, Friedman EM, et al. Rural-urban disparities in access to home- and community-based services and supports: stakeholder perspectives from 14 states. J Am Med Dir Assoc. 2019;20(4):503-508.e1.

21 Ewen HH, Washington TR, Emerson KG, et al. Variation in older adult characteristics by residence type and use of home- and community-based services. Int J Environ Res Public Health [Internet]. 2017 [cited 2019 May 28];14(3):330. Available from: https://www.mdpi.com/1660-4601/14/3/330