Housing and loneliness among individuals with serious mental illnesses

Abstract

Background:

People with serious mental illnesses (SMI) have higher levels of loneliness than the general population. Furthermore, people with SMI tend to be less satisfied with their housing and tend to move more frequently.

Aim:

This study aims to examine relationships between housing variables (whom they live with, duration of residence, and satisfaction) and loneliness among individuals with SMI.

Methods:

Data were collected from 188 adults with SMI in greater Philadelphia area. Classification and Regression Trees (CART) were used to examine whether whom they live with, duration of residence, and housing satisfaction were associated with loneliness.

Results:

Housing satisfaction was found to be the most prominent predictor of loneliness. Those who were unsatisfied with their overall housing conditions always had the highest level of loneliness, regardless of other factors. Even if they were satisfied with their housing conditions, their loneliness was higher if they had just moved to the new residence. Participants had lower loneliness the longer they lived in a residence and had the lowest loneliness levels after about three years.

Conclusion:

Housing is associated with loneliness among people with SMI. Psychiatric service providers should increase support to factors contributing to housing satisfaction and duration of residence, including active engagement in the community.

Keywords

home environment loneliness community psychiatry mental disorders

Introduction

Loneliness among people with serious mental illnesses (SMI) is not only a primary unmet need (Fortuna et al., 2019), but also a significant risk factor of chronic illness and death (Trémeau et al., 2016). It is evidenced that loneliness is associated with health determinants such as lack of self-efficacy to manage chronic disease (Fortuna et al., 2021) and thoughts of self-harm among people with SMI (Dell et al., 2020). As a result, loneliness leads to compromised psychological well-being, undermined physical functioning, bodily pain, and even high incidence of psychiatric hospitalizations (Fortuna et al., 2020). These findings indicate that loneliness is a serious health concern among people with SMI.

Previous research has focused on social relationships, such as social network size, satisfaction with social relationships, and social support as primary predictors of loneliness among adults with SMI (Chang et al., 2014; Chrostek et al., 2016; Schwartz & Gronemann, 2009; Wang et al., 2020). In addition to social relationship, attitudinal and behavioral factors also predict loneliness. For example, internalized stigma (Chrostek et al., 2016; Prince et al., 2018), depressive symptoms (Lim et al., 2018; Wang et al., 2020), and participation in community activities (Nagata et al., 2021; Schwartz & Gronemann, 2009) are some of the reported predictors of loneliness among people with SMI. Yet, researchers argue that other factors predicting loneliness have largely been unexplored and called for investigating environmental factors (Lim et al., 2018). Housing is such a factor that has a potential relevance to loneliness but has rarely been examined.

Health geography research has identified housing as an important base for the nurturing of social relationships, yet its role in loneliness among adults with SMI is largely unexamined. According to Dunn (2000), housing is conceptualized as “a crucial site in the day-to-day life of most individuals for . . . access to social resources, as well as being an important factor in processes of social identity formation, and the establishment and maintenance of social relationships.” (p. 352). In short, housing is comprehensively understood as a multidimensional concept—not just a physical space to live in, but also a social space to interact with others by inviting guests and interacting with neighbors (Mallett, 2004). In theory, housing is associated with loneliness because people develop closer relationships with others in their homes. While a study among older adults suggested that housing was associated with loneliness (Prieto-Flores et al., 2011), the relationship between housing and loneliness has not been examined among people with SMI.

For people with SMI, housing is a matter of great importance. Place of residence has become increasingly recognized as an important factor in the lives of people with SMI as they face challenges to acquire and maintain a place to live (Forchuk et al., 2006). A previous study identified that the home is one of the most important places identified by adults with SMI (Townley et al., 2009). While the importance of housing among people with SMI is widely acknowledged, the relationship between housing and loneliness is understudied.

The current study focused on three factors related to housing that are potentially related to loneliness: (1) cohabitation status (i.e., whom they live with); (2) duration of residence, and (3) satisfaction with housing. First, one aspect of housing that might be related to loneliness is cohabitation status. Living alone can heighten loneliness as people tend to feel unsafe being alone, and the feeling of loneliness is a sign for seeking companionship (Hawkley & Cacioppo, 2010). Thus, people who live with others would have lower levels of loneliness. Furthermore, whom they live with may also matter for loneliness. For example, those living with close others such as family members or friends would have even lower levels of loneliness than those who live with strangers as can be seen in congregate housing services. Previous studies have reported inconsistent results, as one study showed that living situation (i.e., living alone, with a roommate, with family, or in a group home) was not associated with loneliness (Brown, 1996); another reported that those who live in a group home had lower levels of loneliness than those who live alone (Schwartz & Gronemann, 2009). Thus, cohabitation status should be examined to clarify the relationship with loneliness.

Duration of residence, or the length of time an individual has lived in their current residence, may also relate to loneliness. Individuals with SMI experience greater residential turnover than the general population, and the high rates of poverty and homelessness experienced by this population contributes to residential turnover (Lix et al., 2006). Frequent moving requires people to navigate daily life in an unfamiliar environment. They must begin establishing relationships with neighbors and getting to know the community. New social relationships cannot be established instantly. With a meta-analysis of qualitative studies, Watson et al. (2019) found that some people who move into supported housing experience loneliness and struggle to find people and places to connect with, and activities to get engaged in the community. Considering that it generally takes about 30 contact hours to develop even casual friendship (Hall, 2019), it is likely that those who move to a new place have high loneliness due to the lack of social ties they have in the community.

Finally, satisfaction with housing should also be considered as a potential housing factor related to loneliness. Housing satisfaction is defined as the individuals’ current housing situation as a whole being close to their ideals (Prieto-Flores et al., 2011; Vera-Toscano & Ateca-Amestoy, 2008). While housing satisfaction can be influenced by factors including monthly cost, aesthetics, and the condition of the location, research shows that the quality of social interaction that occur at home is a major determinant of housing satisfaction (Vera-Toscano & Ateca-Amestoy, 2008). Among older adults, those who satisfied with their housing had lower loneliness (Prieto-Flores et al., 2011). Possibly, the same phenomenon may be observed among people with SMI. Furthermore, inclusion of housing satisfaction, a psychological measure, to the above objective measures (i.e., cohabitation status and duration of residence) provides a broader picture of how housing may be associated with loneliness among adults with SMI. Thus, this study aimed to examine whether cohabitation status, duration of residence, and housing satisfaction contributed to greater loneliness among individuals with SMI.

Methods

Sample

A dataset involving 188 individuals was created by appending data from individuals with SMI who participated in two research studies conducted between 2014 and 2016 that aimed to enhance physical activity. The dataset included only the baseline measures. The studies received required approval from two institutional review boards (Temple University IRB Protocol # 21826 and 21827). The individuals included in these studies were recruited from five community mental health centers in the Greater Philadelphia area, were between the ages of 18 and 64, had a desire to increase their community participation, and had a psychotic disorder, major depression, or bipolar disorder based on the Mini International Neuropsychiatric Interview (MINI; Sheehan et al., 1998). For the sake of the current study, we also excluded nine individuals living in temporary housing (e.g., emergency shelter/mission), for a final sample of 179.

Measures

Demographics

Demographic variables including age, sex, education, and history of homelessness were collected.

Cohabitation Status

Study participants were asked to select their current cohabitation status from a list of options. The final cohabitation status variable used in this study had three categories: (1) living alone in own apartment or home; (2) living with family members, roommates, or friends; and (3) living with stranger(s) at a mental health residential housing, which combined the following categories: mental health supervised or unsupervised household with other clients; personal care/boarding care/domestic care; foster or private care home where care is provided; commercial boarding homes, YMCA’s, rooming houses; substance abuse rehabilitation residence).

Duration of Residence

Individuals were asked to indicate how long (number of days, months, or years) they have lived in the current location. Housing duration in days was used in the analyses.

Housing Satisfaction

We asked participants a single-item question to rate their overall satisfaction with their current living situation, taking into consideration numerous factors, including monthly cost, the number of people they live with, the condition of the location, and the distance to work and places they receive their mental health services. The item was rated on a seven-point scale (1 = extremely dissatisfied, 7 = extremely satisfied), but was dichotomized for the analyses, such that individuals who rated their housing satisfaction as 5 (satisfied), 6 (very satisfied), or 7 (extremely satisfied) were in the satisfied group, and everyone else was in the dissatisfied group.

Loneliness

Loneliness was measured with a four-item version of the UCLA Loneliness Scale (Russell et al., 1978). Items were rated on a four-point scale, and asked how often individuals felt left out; isolated from others; that there are people that really understand them; and that there are people they can talk to. The composite score was created as the average of the four items, such that higher scores indicated greater loneliness. The original UCLA Loneliness Scale demonstrates good validity and reliability (Russell, 1996; Russell et al., 1980).

Analysis

In this study, we employed Classification and Regression Trees (CART; Breiman et al., 1984), a set of machine learning algorithms that have been used in mental health research (e.g., Ascher-Svanum et al., 2013; Bettenhausen et al., 2021; Schennach-Wolff et al., 2010), as well as for modeling loneliness and social isolation in geriatric populations (e.g., Bai et al., 2021; Ejlskov et al., 2018; Kotian et al., 2018). Unlike traditional regression models, tree models enable us to find and visualize interesting interactions between predictors, make no assumptions about the data, and are good at handling multicollinearity (Breiman et al., 1984). Regression trees, which are constructed for continuous outcomes, employ an iterative process to partition “the root node,” which refers to the original group containing all the data, into non-overlapping smaller groups named “child nodes”; the terminal nodes at which no additional data splits are performed are known as “leaves.” At each step of the recursive process, an optimal predictor and split value are found that minimize the average squared error, which is the average of the squared differences between the observed and predicted values of the dependent variable, across the child nodes. The more variance in the outcome variable a predictor variable and split level account for, the closer that predictor and split level are to the root of the tree. The CART procedure can, in sum, identify (a) most prominent predictors among the ones that are entered in the model, and (b) the optimum split value within a predictor variable that helps classify groups.

In the current study, PROC HPSPLIT in SAS 9.4 (SAS Institute, 2015) was used to create a regression tree with ten-fold cross-validation and cost-complexity pruning (Breiman, 1984). The outcome variable was loneliness, and predictors included housing satisfaction, duration of residence, cohabitation status, age, sex, and white and black race indicators. To answer the research question, we examined whether the CART procedure determined housing variables as prominent predictors. We also inspected the optimum split values of housing duration variable to identify at what levels it made a meaningful change in loneliness. Additionally, an ANOVA was conducted to examine whether the subgroups in the terminal nodes of the regression tree differed significantly in terms of loneliness.

Results

Sample Description

Seventy-four individuals in the sample (41%) were female, and the rest were male. The average age was 48.3 years (SD = 10.3). In terms of race, 39 (22%) indicated that they were white, whereas 115 (64%) said they were black or African American. More than two-thirds of the sample (n = 122, 68%) were satisfied with their current housing situation, and 57 (32%) were dissatisfied or neutral. On average, individuals stayed at their current residence for 1,888 days, or 5.2 years (SD = 2,579 days, or 7.1 years), but this ranged between 1 day and 40 years. Cohabitation status indicated that 78 (44%) individuals lived alone in their own apartment or home, 54 (30%) shared their residence with family members, roommates, or friends, and 47 (26%) lived in mental health residential housing.

The average score on the UCLA Loneliness Index was 2.37 (SD = 0.72). Loneliness was not significantly associated with age (r = −0.12, p = .125) or housing duration (r = −0.10, p = .166). Table 1 presents additional tests of association between loneliness, demographics, and the housing variables, from which we can see that females were marginally less lonely than males, and those who were satisfied with their housing situation were significantly less lonely than those who were not. Race and cohabitation status were not related to loneliness.

Table 1.

Tests of association between loneliness, demographics, and housing.

	N	Mean	SD	t-value	df	p-value
Sex				−1.93	142	.056
Female	74	2.24	0.78
Male	105	2.45	0.67
Black race				−0.03	116	.972
Yes	115	2.37	0.69
No	64	2.36	0.79
White race				0.36	57	.718
Yes	39	2.33	0.77
No	140	2.38	0.71
Housing satisfaction				4.82	118	<.001
Dissatisfied or neutral	57	2.71	0.64
Satisfied	122	2.2	0.7
	N	Mean	SD	F-value	df	p-value
Cohabitation status				0.66	2; 176	.517
Living alone, in own apartment or home	78	2.32	0.67
Living with family members, roommate, or friends	54	2.35	0.73
Living with stranger(s) in a mental health residential housing	47	2.47	0.81

In addition, tests of association among the housing variables were conducted. There was no significant difference in housing duration between those who were satisfied with their housing and those who were not (t = 0.15, df = 177, p = .878). Moreover, cohabitation status and housing satisfaction were unrelated (χ²(2) = 1.828, p = .401). However, the average housing duration was found to be different depending on cohabitation status, F(2, 176) = 6.85, p = .001. Post-hoc analysis found that those who lived in a mental health residential housing had stayed significantly longer than those who lived in their own homes or those who lived with their family/roommate/friend (p > .05).

Regression Tree Results

Figure 1 presents the final pruned regression tree. When Breiman’s cost-complexity pruning (Breiman, 1984) was specified, the tree had only three terminal nodes (nodes 2-4 in the tree in Figure 1), and the cross-validation average squared error (CV ASE) was 0.47. However, because the pruning algorithm does not always do pruning one leaf at a time, we looked at additional trees with other numbers of leaves as well. The tree with four terminal leaves shown in Figure 1 had the lowest CV ASE (0.43) and was selected as the optimal tree.

Figure 1.

Regression tree diagram obtained by the classification and regression trees (CART) procedure.

We can see that the final model only includes housing satisfaction and duration, and omits cohabitation status and demographics. The first split shows that in our model, housing satisfaction was most predictive of loneliness, and that individuals who were satisfied with their housing (node 2), on average, had lower loneliness levels (n = 122, mean = 2.20, SD = 0.70) than those who were dissatisfied or neutral (node 3; n = 57, mean = 2.71, SD = 0.64; t = 4.82, df = 118, p < .001). While node 3 was a terminal node, those who were satisfied with their housing (node 2) were further split based on housing duration; the optimal split value was determined, by the CART algorithm, to be 1,169 days or approximately 3.2 years. Those who were satisfied and stayed at their current residence for 3.2 years or more were less lonely (node 5; n = 53, mean = 1.90, SD = 0.64) than those who were satisfied and stayed at their residence less than 3.2 years (node 4; n = 69, mean = 2.43, SD = .66; t = 4.50, df = 113, p < .001). Node 4 was further split at the optimal housing duration value which was determined to be 147 days (approximately 4.8 months). We can see from Figure 1 that individuals who stayed at their residence less than 147 days were more lonely (node 6; n = 31, mean = 2.62, SD = 0.65) than those who have stayed there between 147 and 1,168 days (node 7; n = 38, mean = 2.28, SD = 0.63; t = −2.16, df = 63, p = 0.034).

An ANOVA test comparing the four terminal nodes shows groups differ significantly in terms of loneliness (F(df = 3;175) = 16.70; p < .001) and that loneliness was highest in node 3 (dissatisfied with housing situation), followed by node 6 (satisfied with housing situation and housing duration less than 147 days), then node 7 (satisfied with housing situation and housing duration between 147 and 1,168 days) and finally, node 5 (satisfied with housing situation and housing duration 1,169 days or more). Post-hoc tests showed that nodes 3 and 6 were not significantly different from one another (pairwise comparison p-value = .513), but all the other pairwise group comparisons were statistically significant.

Discussion

The current study aimed to examine relationships between housing factors and loneliness among people with serious mental illnesses (SMI). The results identified two major housing factors. First, housing satisfaction was the most prominent predictor of loneliness. Those who were unsatisfied with their overall housing conditions always had high level of loneliness, regardless of other factors. The second strongest predictor was duration of residence. Even if they were satisfied with their housing conditions, the less time they had lived in that home, they reported higher levels of loneliness. While social relationships are understood to be predictors of loneliness (Chang et al., 2014; Chrostek et al., 2016; Schwartz & Gronemann, 2009), other factors have been underexplored. The finding that housing is a predictor of loneliness, therefore, is novel and helps to enhance understanding of loneliness among people with SMI.

The results demonstrate that housing satisfaction was most predictive of loneliness. While housing satisfaction and loneliness might be seen as two unrelated concepts, there are several important connections between these constructs. According to Tsemberis et al. (2003), the experience of living in the residence determines the level of housing satisfaction. Indeed, physical housing quality such as housing size and location matters, but social experiences such as inviting guests in their home and interacting with neighbors are also important determinants of housing satisfaction (Prieto-Flores et al., 2011; Vera-Toscano & Ateca-Amestoy, 2008). Positive assessment of these social experiences might be quite relevant to loneliness, which our data might indicate. Another potential reason is that good housing can facilitate a sense of belonging, which is a correlate of loneliness. Previous studies reported that sense of belonging was a mediator between housing satisfaction and loneliness (Prieto-Flores et al., 2011) and that people with SMI identified home as an important place for sense of belonging (Townley et al., 2009). While social experience at home and sense of belonging could not be examined directly in this study, our findings call for investigation of mechanisms for how housing can facilitate these social factors. Furthermore, future studies should investigate factors that determine housing satisfaction in the context of people with SMI because people with SMI often experience loneliness or live in a residence that is assigned.

While satisfaction with one’s living condition played the most important role in determining loneliness, findings indicate that satisfaction alone is not enough. The duration individuals lived in their home also determined the level of loneliness. Even if they were satisfied with their residence, if they had moved within the last 5 months, the level of loneliness was not different from those who were not satisfied with the residence. A potential reason for this finding is that people typically have few social connections when they move to a new place. Moving often means that one is displaced to an environment with few social connections, and one must start establishing social relationships from scratch (Watson et al., 2019). Especially, they may not know what resources or activities exist in their new neighborhood or community, and even if they do know, they may have no one to go with (Hendriks et al., 2016). Considering overwhelming to-do lists and errands post-move, people may not have time and opportunities to connect with others in the first few months post-move, which might lead to higher loneliness.

However, the longer individuals live in the same home, the more benefits they experience regarding loneliness. According to our results, this benefit reaches a peak at around 3.2 years after a move. After that, individuals enjoy the lowest level of loneliness. One explanation for this phenomenon is that it takes time to nurture a new friendship. A study shows that it takes approximately 140 hours of contact to establish a good friendship (Hall, 2019). Provided that meeting a person 1 hour per week, it would take about 3 years to reach the threshold of good friendship; and close social relationships are a known predictor of less loneliness (Chang et al., 2014; Chrostek et al., 2016; Schwartz & Gronemann, 2009). As people with SMI move residences more often than the general population (Lix et al., 2006), perhaps many people cannot take advantage of the benefit of living in a place for a long time. The body of knowledge in housing stability identified that sense of community is a major factor that made people live in one place for a longer period of time (Leickly & Townley, 2021). It appears consistent that helping newcomers feel included and that they belong in the community is a way to enhance both housing satisfaction and housing duration.

While we found some promising housing factors, whom individuals live with was not a critical factor regarding loneliness. Reports from previous studies have been inconsistent whether living with others resulted in lower loneliness compared to living alone (Brown, 1996; Schwartz & Gronemann, 2009). The current study used Regression Trees, which examined the relative importance among the factors included in the analysis. Perhaps the results were nuanced in a way that whom one lives with has little to do with loneliness compared to satisfaction with the residential conditions and the duration of living. Also, the satisfaction measure included, at least partially, a judgment of cohabitants. Another potential reason that cohabitation status was not related to loneliness is the matter of preference. Some people may like to live alone, rather than live with others. Loneliness is a perceived state of connectedness with others, and it is not necessary to have people physically around them (Hawkley & Cacioppo, 2010).

Implications

The most predictive housing factor for loneliness was satisfaction. To improve housing satisfaction, active engagement in the community is recommended (Watson et al., 2019). Congregate housing such as group homes can enhance social relationships within the housing community through social events such as block parties. The results also support the slogan of the National Alliance to End Homelessness: “housing first, but not only.” It also has policy implications to fund not only the physical housing space, but also the services that support social integration by connecting with neighbors and the new community.

The current study found that satisfaction with housing alone may not be enough. Living in a residence for a longer time may result in reduced loneliness; however, the first few months after a move may be a particularly lonely period. Individuals should be prepared that they might experience higher loneliness after a move and should be encouraged to use strategies to minimize or cope with loneliness as appropriate. In addition, agencies should increase support programs to help clients remain in their current home as long as possible, assuming their living situations are generally satisfactory. If remaining in the same structure is not possible, efforts should be made to support housing within the same neighborhood. Engaging in activities with other residents and neighbors could help enhance sense of belonging and sense of community. The results suggests that stable housing is ideal. Policymakers should continue to advocate for increased funding for permanent supported housing with accompanying services, including programs aimed at increasing social connections and reducing loneliness within housing. Not only would this help to reduce loneliness among residents but could also potentially achieve savings in their medical treatment given findings that loneliness predicts hospital emergency department use (Chamberlain et al., 2022; Geller et al., 1999).

Limitations

While the findings of the current study are informative, some limitations should be acknowledged. First, the results do not demonstrate causality. The data is from a correlational study, and the relationships between loneliness and housing variables should be interpreted carefully. The opposite direction may also be plausible, as feeling lonely can cause dissatisfaction with housing. Second, the housing satisfaction measure was a single-item instrument. The data acquired from a single-item measure may contain unaddressed random measurement error. Third, the results are limited to the variables included in the analyses. Although housing variables were outstanding predictors of loneliness in our study, our results do not take into account other potential predictors not included in the study.

Conclusion

The current study demonstrated a significant relationship between housing and loneliness among people with SMI. Housing satisfaction was most prominent predictor of loneliness in this study. In addition, the length of time living in one residence also matters significantly. As loneliness has gained increasing attention, the findings of the current study suggest paying attention to factors that are beyond interpersonal relationships. Indeed, future studies are needed to advance knowledge about the mechanisms underlying housing satisfaction, duration, and loneliness.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the U.S. Department of Health and Human Services, National Institute on Disability, Independent Living, and Rehabilitation Research (grants #90RTCP0001-01-00 and #90IF0086). The contents do not necessarily represent the policy of the U.S. Department of Health and Human Services, and endorsement by the federal government should not be assumed.

ORCID iDs

Shinichi Nagata

Bryan McCormick

Mark S. Salzer

References

Ascher-Svanum

Novick

Haro

J. M.

Aguado

Cui

(2013). Empirically driven definitions of “good,” “moderate,” and “poor” levels of functioning in the treatment of schizophrenia. Quality of Life Research, 22(8), 2085–2094.

Bai

Wang

Shao

Qin

(2021). Association between social capital and loneliness among older adults: A cross-sectional study in Anhui Province, China. BMC geriatrics, 21(1), 26. https://link-springer-com-443.web.bisu.edu.cn/article/10.1186/s12877-020-01973-2

Bettenhausen

J. L.

Hall

Doupnik

S. K.

Markham

J. L.

Feinstein

J. A.

Berry

J. G.

Gay

J. C.

(2021). Hospitalization outcomes for rural children with mental health conditions. The Journal of Pediatrics, 229, 240–246.

Breiman

Friedman

J. H.

Olshen

R. A.

Stone

C. J.

(1984). Classification and regression trees. Wadsworth.

Brown

(1996). A comparison of living situation and loneliness for people with mental illness. Psychiatric Rehabilitation Journal, 20(2), 59–63.

Chamberlain

S. A.

Savage

Bronskill

S. E.

Griffith

L. E.

Rochon

Batara

Gruneir

(2022). Examining the association between loneliness and emergency department visits using canadian longitudinal study of aging (CLSA) data: A retrospective cross-sectional study. BMC Geriatrics, 22(1), 69.

Chang

C.-W.

Chung

C.-L.

Biegel

D. E.

Pernice-Duca

Min

M. O.

D’Angelo

(2014). Predictors of loneliness of clubhouse members. Psychiatric Rehabilitation Journal, 37(1), 51–54. https://doi.org/10.1037/prj0000052

Chrostek

Grygiel

Anczewska

Wciórka

Świtaj

(2016). The intensity and correlates of the feelings of loneliness in people with psychosis. Comprehensive Psychiatry, 70, 190–199.

Dell

N. A.

Huang

Buttafuoco

K. A.

Vidovic

K. R.

Murphy

A. M.

Farrar

L. A.

(2020). Direct and indirect associations between loneliness and thoughts of self-harm among a clinical sample of older adults with serious mental illness. The Journal of Nervous and Mental Disease, 208(10), 828–832.

10.

Dunn

J. R.

(2000). Housing and health inequalities: Review and prospects for research. Housing Studies, 15(3), 341–366.

11.

Ejlskov

Wulff

Bøggild

Kuh

Stafford

(2018). Assessing the relative importance of correlates of loneliness in later life. Gaining insight using recursive partitioning. Aging & Mental Health, 22(11), 1486–1493. https://www-tandfonline-com-s.web.bisu.edu.cn/doi/full/10.1080/13607863.2017.1370690

12.

Forchuk

Ward-Griffin

Csiernik

Turner

(2006). Surviving the tornado of mental illness: Psychiatric survivors’ experiences of getting, losing, and keeping housing. Psychiatric Services, 57(4), 558–562. https://doi.org/10.1176/ps.2006.57.4.558

13.

Fortuna

K. L.

Brusilovskiy

Snethen

Brooks

J. M.

Townley

Salzer

M. S.

(2020). Loneliness and its association with physical health conditions and psychiatric hospitalizations in people with serious mental illness. Social Work in Mental Health, 18(5), 571–585.

14.

Fortuna

K. L.

Ferron

Bianco

C. L.

Santos

M. M.

Williams

Mois

Pratt

S. I.

(2021). Loneliness and its association with health behaviors in people with a lived experience of a serious mental illness. Psychiatric Quarterly, 92(1), 101–106.

15.

Fortuna

K. L.

Ferron

Pratt

S. I.

Muralidharan

Aschbrenner

K. A.

Williams

A. M.

Deegan

P. E.

Salzer

(2019). Unmet needs of people with serious mental illness: Perspectives from certified peer specialists. Psychiatric Quarterly, 90(3), 579–586.

16.

Geller

Janson

McGovern

Valdini

(1999). Loneliness as a predictor of hospital emergency department use. Journal of Family Practice, 48(10), 801–804.

17.

Hall

J. A.

(2019). How many hours does it take to make a friend? Journal of Social and Personal Relationships, 36(4), 1278–1296.

18.

Hawkley

L. C.

Cacioppo

J. T.

(2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218–227.

19.

Hendriks

Ludwigs

Veenhoven

(2016). Why are locals happier than internal migrants? The role of daily life. Social Indicators Research, 125(2), 481–508.

20.

Kotian

D. B.

Mathews

Parsekar

S. S.

Nair

Binu

V. S.

Subba

S. H.

(2018). Factors associated with social isolation among the older people in India. Journal of Geriatric Psychiatry and Neurology, 31(5), 271–278. https://journals.sagepub.com/doi/full/10.1177/0891988718796338

21.

Leickly

Townley

(2021). Exploring factors related to supportive housing tenure and stability for people with serious mental illness. Journal of Community Psychology, 49(6), 1787–1805.

22.

Lim

M. H.

Gleeson

J. F.

Alvarez-Jimenez

Penn

D. L.

(2018). Loneliness in psychosis: A systematic review. Social Psychiatry and Psychiatric Epidemiology, 53(3), 221–238.

23.

Lix

L. M.

Hinds

DeVerteuil

Renee Robinson

Walker

Roos

L. L.

(2006). Residential mobility and severe mental illness: A population-based analysis. Administration and Policy in Mental Health and Mental Health Services Research, 33(2), 160–171.

24.

Mallett

(2004). Understanding home: A critical review of the literature. Sociological Review, 52(1), 62–89. https://doi.org/10.1111/j.1467-954X.2004.00442.x

25.

Nagata

McCormick

Brusilovskiy

Zisman-Ilani

Wilson

Snethen

Townley

Salzer

M. S.

(2021). Emotional states associated with being in the community and being with others among individuals with serious mental illnesses. American Journal of Orthopsychiatry, 91(1), 1–8.

26.

Prieto-Flores

M. E.

Fernandez-Mayoralas

Forjaz

M. J.

Rojo-Perez

Martinez-Martin

(2011). Residential satisfaction, sense of belonging and loneliness among older adults living in the community and in care facilities. Health & Place, 17(6), 1183–1190.

27.

Prince

J. D.

Oyo

Mora

Wyka

Schonebaum

A. D.

(2018). Loneliness among persons with severe mental illness. The Journal of Nervous and Mental Disease, 206(2), 136–141. https://doi.org/10.1097/NMD.0000000000000768

28.

Russell

Peplau

L. A.

Cutrona

C. E.

(1980). The revised UCLA loneliness scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39(3), 472–480. https://doi.org/10.1037/0022-3514.39.3.472

29.

Russell

Peplau

L. A.

Ferguson

M. L.

(1978). Developing a measure of loneliness. Journal of Personality Assessment, 42(3), 290–294. https://doi.org/10.1207/s15327752jpa4203_11

30.

Russell

D. W.

(1996). UCLA loneliness scale (version 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66(1), 20–40. https://doi.org/10.1207/s15327752jpa6601_2

31.

SAS Institute Inc. 2015. SAS/STAT® 14.1 User’s Guide. Cary, NC: SAS Institute Inc.

32.

Schennach-Wolff

Jäger

Obermeier

Schmauss

Laux

Pfeiffer

Naber

Schmidt

L. G.

Gaebel

Klosterkötter

Heuser

Kühn

K. U.

Lemke

M. R.

Rüther

Klingberg

Gastpar

Seemüller

Möller

H. J.

Riedel

(2010). Quality of life and subjective well-being in schizophrenia and schizophrenia spectrum disorders: Valid predictors of symptomatic response and remission? The World Journal of Biological Psychiatry, 11(5), 729–738.

33.

Schwartz

Gronemann

O. C.

(2009). The contribution of self-efficacy, social support and participation in the community to predicting loneliness among persons with schizophrenia living in supported residences. Israel Journal of Psychiatry and Related Sciences, 46(2), 120–129.

34.

Sheehan

D. V.

Lecrubier

Sheehan

K. H.

Amorim

Janavs

Weiller

Hergueta

Baker

Dunbar

G. C.

(1998). The mini-international neuropsychiatric interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59(20), 22–33.

35.

Townley

Kloos

Wright

P. A.

(2009). Understanding the experience of place: Expanding methods to conceptualize and measure community integration of persons with serious mental illness. Health and Place, 15(2), 520–531. https://doi.org/10.1016/j.healthplace.2008.08.011

36.

Trémeau

Antonius

Malaspina

Goff

D. C.

Javitt

D. C.

(2016). Loneliness in schizophrenia and its possible correlates. An exploratory study. Psychiatry Research, 246, 211–217.

37.

Tsemberis

Rogers

E. S.

Rodis

Dushuttle

Skryha

(2003). Housing satisfaction for persons with psychiatric disabilities. Journal of Community Psychology, 31(6), 581–590.

38.

Vera-Toscano

Ateca-Amestoy

(2008). The relevance of social interactions on housing satisfaction. Social Indicators Research, 86(2), 257–274.

39.

Wang

Lloyd-Evans

Marston

Mann

Solmi

Johnson

(2020). Epidemiology of loneliness in a cohort of UK mental health community crisis service users. Social Psychiatry and Psychiatric Epidemiology, 55(7), 811–822.

40.

Watson

Fossey

Harvey

(2019). A home but how to connect with others? A qualitative meta-synthesis of experiences of people with mental illness living in supported housing. Health & Social Care in the Community, 27(3), 546–564. https://doi.org/10.1111/hsc.12615