Housing as a Resource for Justice-involved Women

Hobfoll’s (1989) Conservation of Resources (COR) theory of stress posits that resource loss is central to stress reactions (Hobfoll, 1989; Hobfoll, Johnson, Ennis, & Jackson, 2003). Hobfoll’s model is dependent upon the notion that people have a basic need to acquire and protect resources, and stress occurs when these resources are threatened, perceived to be threatened, or actually lost (Hobfoll, 1989). Naturally, resources vary from person to person, as they are defined as objects, conditions, or energies that are valued by the individual (Hobfoll, 1989). For example, there are resources that fall under the title of personal characteristics, such as feeling successful, that may not apply to everyone. Other resources can be thought of as social support, for example, companionship, involvement in the community, or religious affiliation. Resources, and therefore stress, are both cultural and context-specific (Hobfoll, 2001; Hobfoll, 2002). Building on a previous study examining resource loss in underserved populations (Siegel, Ram, Pope, Landreth, & Jason, 2015), we examined the relationship between resource loss and an individual’s housing context in a population of women in recovery from substance use who have had criminal justice involvement.

Of interest to this current study is the notion of housing as a resource, which is understudied in the COR literature. However there is some evidence that housing as a resource may be essential for particular populations. For instance, in a group of survivors of Hurricane Katrina (Fussell & Lowe, 2014) those who were relocated to a new a community or experienced unstable housing (e.g. multiple moves) post-disaster, had significantly higher psychological distress and perceived stress, respectively.

Previous work in COR theory also suggests that the context of resources can vary widely among different populations (Halbesleben, Neveu, Paustian-Underdahl, & Westman, 2014). In a different sample of hurricane survivors, for example, psychological distress was most closely related to the loss of personal characteristic and social support resources, whereas object and energy resources were least closely related (Sattler et al., 2002). This differed from the “cascade of losses” experienced in a sample of individuals in New York City assessed after the 9/11 attacks, where material, personal, and interpersonal resources were all similarly affected (Hobfoll, Tracy, & Galea, 2006).

Loss is further complicated by a study that examined resource loss following an earthquake (Sattler et al., 2006). Sattler and colleagues found personal characteristic (e.g., feeling life has purpose) and object resource loss (e.g., food, possessions) were related to psychological distress in both a college-aged and community sample, but the college-aged sample was also affected by condition (e.g., family stability, stable employment) resource loss (Sattler et al., 2006). This suggests that loss behaves differently depending on one’s situational context within a sample.

There is evidence that this need for context vis-à-vis housing may be especially salient for the sample of the current study: underresourced women with substance use histories and criminal justice involvement (Iguchi, London, Forge, Hickman, & Riehman, 2002; Hunter, Jason, & Keys, 2013). Iguchi and colleagues described multiple adverse effects on health and stress in a group of justice-involved drug users, evidenced through decreased access to housing and employment. Therefore housing as a resource may be especially revealing if one has experience living in jail, prison, or substance abuse treatment settings, as these settings allow for varying degrees of agency. If one lives in a controlled environment such as jail compared to a slightly less restricted environment such as a treatment center or recovery home, will there be differences reported in resource loss? These differences in loss may further argue the context-specific nature of resources and how housing may be an understudied, yet critical variable in COR literature.

In the current study, we examined resource loss in a group of women with substance abuse histories who have also been involved in the criminal justice system. In order to obtain a more detailed picture of the context for these women, COR loss was examined as a function of housing. Though this study was exploratory in nature, we sought to understand whether resource loss was dependent on the type of setting. The following settings were analyzed: Controlled (prison or jail), Independent (living on one’s own), Precarious/Homeless (couch surfing or homeless), Recovery/Treatment (in an Oxford House or in residential treatment). We believed that individuals who were in controlled or recovery settings, who had few resources to lose, would have the lowest amount of resource loss, while those who were in independent housing, suggesting that they had many resources to lose, would have the highest resource loss.

Method

Participants

The present study analyzed baseline data from a National Institute of Health funded project conducted over the course of two years. Two hundred women between the ages of 18 and 59 (M = 39.94, SD = 8.58) who reported having an alcohol and/or other substance use disorder were recruited from multiple sites in metropolitan Chicago and its suburbs from 2008 to 2011 (see Jason, Salina, & Ram, 2015, for more details). The recruitment sites included the Cook County Sheriff’s Women’s Justice Program at Cook County Jail and various substance abuse treatment sites throughout Chicago and Northern Illinois. In addition, recruitment flyers were posted and/or distributed in multiple community-based organizations (CBOs) which provide services to formerly incarcerated women or substance users. Participants were also recruited using snowball sampling techniques. All participants were enrolled in the treatment study with IRB approved informed consent procedures that included outlining the study. Participants received a $40 grocery store gift certificate for participating in the initial interview. Interviewers tracked and interviewed participants during the 24-month span of their involvement in the study. A complete list of demographics is listed in Table 1.

Table 1.

Univariate analysis of time most spent and total core loss scores

	Total core loss score
Source	N	Mean	Std. Error
Controlled	37	91.622	6.242
Independent	40	106.275	6.003
Precarious/Homelessness	43	87.07	5.79
Recovery/Treatment	80	79.325	4.245

Procedure

Participants completed a two-hour interview. We collected data on the types of criminal charges for which participants had been arrested, their previous criminal histories, drug use and misuse histories, and the number of arrests and months incarcerated. In addition to demographic questions, participants completed several psychological measures.

Measures

Resource Loss

Resource loss was measured by the Conservation of Resources-Evaluation (COR-E; Hobfoll, 1989; Hobfoll & Lilly, 1993). The version of COR-E utilized in this study was a 45-item measure assessing perceived loss of various types of resources (e.g., transportation, health of family members, adequate food, and good relationship with my children). Participants were asked to rate how much they felt they lost each item over the previous three months (e.g., “housing that suits my needs”) on a 5-point scale (1= not at all to 5=greatly). COR-E is widely used and has been shown to be both reliable and valid (Ennis, Hobfoll, & Schroder, 2000; Hobfoll & Lilly, 1993). Cronbach’s alpha for our sample on this measure was 0.97.

Housing

Housing was measured by the Housing Timeline Followback (HTLFB). The HTLFB is Adapted Housing Timeline Follow-Back Inventory (New Hampshire Dartmouth Psychiatric Research Center, 1995) (HTLFB). Housing transitions were collected through the use of an adapted version of the Housing Timeline Follow-Back Inventory. During interviews participants reported all housing settings and corresponding living companions from the past six months.

Analyses

We ran a general linear model to examine the relationship between the setting where participants spent the most time and CORE-L. Mean CORE-L scores per group are presented in table 1. Because of small sample size and concerns about normality, a 1000 bootstrap resampling procedure was used. Standard errors from the bootstrap estimates are reported.

Women who had spent the majority of their time in independent settings had significantly higher (t(81)=2.19, p<.05, d=.481) CORE-L scores (M=106.28, S.E.=6.00) compared to those who spent the majority of their time in precarious settings (M=87.07, S.E.=5.79). Women who spent the majority of their time in independent settings also had significantly higher (t(118)= 3.24) CORE-L scores (M=106.28, S.E.=6.00) compared to those who spent the majority of their time in treatment/recovery settings (M=79.33, p<.01, d=.627). There were no other significant differences among groups.

Discussion

This study examined the association of housing and resource loss in a population of women in recovery from substance use. We found that individuals who spent the majority of their time in independent housing had higher feelings of resource loss than individuals who were in recovery settings, controlled environments or who experienced homelessness. However there was no difference in resource loss between individuals who were in independent settings and those who were in controlled settings. These findings suggest that individuals in independent housing report more loss because they have more to lose. This is perhaps best supported by the finding that individuals spending the majority of their time in precarious or homeless situations had the lowest feelings of COR loss.

Perhaps counter-intuitive to one’s notion of being in a homeless setting, there is evidence to support that those who experience homelessness find ways to minimize resource losses (Ennis, Hobfoll, & Schroeder, 2000). In a sample of inner city women, low education, income, and employment were not significantly associated with depressive mood. Our results bolster the notion that in certain niches, such as homelessness, there are indeed ways to minimize loss by remaining resilient to potential adverse effects of lacking material resources.

To characterize housing as a mere material resource may result in missing out on several nuances regarding the effect settings have on individuals. For instance, one experimental design found improvements in subjective quality of life among individuals who have experienced homelessness after receiving housing (Patterson et al., 2013). Participants who were randomized into a Housing First model (HF) reported significantly greater quality of life than those who were in the usual care group, regardless of where they ended up. Specifically, the areas most significantly greater in the HF group were the domains of safety and living situation. Safety is not a material resource but rather an emotional resource. Housing thus provides more comfort than simple shelter.

In the present study the settings in which people lived varied in terms of individual agency. One would assume those living in a controlled setting such as jail would report the highest losses among the other conditions we analyzed, which were independent, homeless, and recovery-oriented settings. Yet those living in independent and recovery settings reported the highest loss. It’s possible that where one has agency one feels responsible for his or her resources, which is to say one would be more likely to feel loss if what they’re losing is indeed theirs to lose. This finding suggests those living independently or in a recovery-setting are feeling the stress of living on their own for perhaps the first time in a long while. Research has shown that stress plays a key role in perpetuating drug use and relapse (Sinha, 2001).

This paper has multiple limitations. Firstly, in regard to the study as a whole, participants were not randomized into these settings. There could therefore be a bias of participants self-selecting into the various settings we analyzed. With respect to the current analysis, we collapsed people into the four settings based on where they spent the most time. This may not be an accurate indicator of where they were living when resource loss was captured. We are also missing key historical details, such as those who reported being homeless, are they chronically homeless? Or those living in recovery settings or jail, have they been there before? This information could be useful for future research on COR and housing and research.

In summary, our findings support previous work that highlights how different aspects of loss may be more salient to different populations (e.g., Hobfoll et al., 2006; Sattler et al., 2002). These findings also bring up the importance of housing, and how it is representative of more than just shelter, but can affect an individual’s state, both mentally and physically. Lastly, the notion of housing as a salient resource could be extended to individuals beyond the sample of the present study. Future COR research should consider housing as a resource that functions beyond material well-being and examine it through a lens of emotional well-being.

Our findings, coupled with the adverse of effects of stress, show that people with criminal histories or substance use disorders who live in independent settings are vulnerable to resource loss, as they are new to these environments. Based on these findings, we suggest that those who are incarcerated or leaving treatment require continued support, such as jobs training and counseling, to ease their transition to independent living settings. Entering independent settings too early after release from these institutions may precipitate stress and consequently the use of substances, which may lead to new crimes and potential rearrests. By having continued support available for people who leave these institutions as they reenter the community, they will be better able to succeed in their reintegration.