Spirituality and Mental Health among Homeless ... - Oxford Journals

210 downloads 0 Views 146KB Size Report
Mothers are one of the fastest growing segments of the homeless population in the United. States. Although mental health problems often contribute to ...
Spirituality and Mental Health among Homeless Mothers David R. Hodge, Stephanie E. Moser, and Michael S. Shafer Mothers are one of the fastest growing segments of the homeless population in the United States. Although mental health problems often contribute to homelessness, little is known about the factors that affect mothers’ mental health. To help identify protective factors, this longitudinal study examined the relationship between spirituality and mental health among a sample of homeless women with children (N = 222). A growth curve analysis was conducted to examine relationships over a 15-month time span. Forgiveness, congregational problems, negative religious coping, and spiritual meaning all variously predicted mental health outcomes. The implications of these findings are discussed as they intersect practice with homeless mothers. KEY WORDS:

homelessness; mental health; mothers; religion; spirituality

O

ver the course of the past few decades, homelessness has emerged as a significant social issue (Fertig & Reingold, 2008). Obtaining accurate estimates of the number of homeless individuals is difficult because of the use of different definitions of homelessness and problems counting the homeless (Rollins, Saris, & Johnston-Robledo, 2001). For instance, according to the U.S. Department of Housing and Urban Development’s latest Annual Homeless Assessment Report (2010), an estimated 643,067 people lived on the streets or in shelters on a single night, and 1.56 million people used an emergency shelter or a transitional housing program over the course of a 12-month period. Although estimates vary, general agreement exists that the number of homeless people in the United States has increased dramatically since the late 1970s (Wachholz, 2005). In addition to growing in size, the composition of the homeless population has changed substantially over the past few decades (Averitt, 2003; T. N. Richards, Garland, Bumphus, & Thompson, 2010). Families with children have emerged as a major component of the homeless population (MeadowsOliver, 2003; Paquette & Bassuk, 2009; Weinreb, Nicholson, Anthes, & Williams, 2007). The majority of homeless families are headed by single mothers (Goldberg, 1999). Indeed, according to some commentators, women are the fastest growing segment of the homeless population in the United States (Arangua, Andersen, & Gelberg, 2006).

doi: 10.1093/swr/svs034

© 2013 National Association of Social Workers

The causes underlying the changing composition of the homeless population are not fully understood (Lehmann, Kass, Drake, & Nichols, 2007). The increase in the number of homeless families is not limited to the United States, but extends to Canada (Schiff, 2007), the United Kingdom (Tischler & Vostanis, 2007), and perhaps other countries as well (Daiski, 2007). Although a small but growing body of research on homeless mothers exists, this area of inquiry is still in its infancy (Cosgrove & Flynn, 2005; Gelberg, Browner, Lejano, & Arangua, 2004; Stainbrook & Hornik, 2006). One relatively established contributor to homelessness among mothers is mental health status (Arangua et al., 2006; Fertig & Reingold, 2008; Lee & Oyserman, 2009; Tischler, Rademeyer, & Vostanis, 2007; Williams & Hall, 2009; Zlotnick, Tam, & Bradley, 2007). The relationship between mental health and homelessness is complex. Poor psychological health can be both an antecedent to, and a consequence of, homelessness (Philippot, Lecocq, Sempoux, Nachtergael, & Galand, 2007). In terms of the former, the onset of a mental disorder can, for example, lead to deteriorating social and economic conditions that eventually result in homelessness (Weinreb et al., 2007). Many events unrelated to the onset of a mental disorder can result in women becoming homeless, including domestic violence, unaffordable rents, divorce or separation, condemned housing, loss of employment, and so on (Meadows-Oliver, 2003;

245

Rollins et al., 2001; Tischler et al., 2007). On becoming homeless, mothers often report experiencing deep senses of loss, stress, or depression (Meadows-Oliver, 2003). Negative life events, cumulating with the loss of their homes and their struggle to adapt to a homeless lifestyle while parenting children, can overwhelm mothers, resulting in increased depression, anxiety, and other mental health problems (Banyard & Graham-Bermann, 1998; Tischler et al., 2007). In turn, the onset of various forms of psychological distress caused by becoming homeless can hinder women’s ability to exit homelessness, causing a downward spiral (Daiski, 2007). Although it is generally accepted that homelessness is stressful and the mental health of homeless mothers is often poor, relatively little is known about the factors that affect the mental health of these women (Tischler et al., 2007). Given the stigma homeless mothers often face, it is particularly important to focus on the strengths or protective factors that help mothers deal with mental health problems (Cosgrove & Flynn, 2005; Wachholz, 2005). Protective factors can be understood as variables that facilitate positive outcomes by buffering individuals from constructs that place them at risk (Fraser, Richman, & Galinsky, 1999; Smith, 2006). One such factor that may help engender positive mental health among homeless mothers is spirituality (Larkin, Beckos, & Martin, in press). SPIRITUALITY AND MENTAL HEALTH

mothers (Cosgrove & Flynn, 2005; MeadowsOliver, 2003) use spirituality to cope with the stress of being homeless. For instance, a meta-synthesis of qualitative research on homeless mothers revealed that praying was among the most common strategies used to deal with the difficulties resulting from homelessness (Meadows-Oliver, 2003). At least two longitudinal studies have explored the relationship between spirituality and mental health among homeless mothers, with, at best, mixed results. In the United Kingdom, the relationship between five different coping strategies and mental health outcomes was examined (Tischler & Vostanis, 2007). Seeking spiritual support was assessed using the Family Crisis Oriented Personal Evaluation Scales. Mental health was assessed using the General Health Questionnaire at baseline (N = 72) and at four-month follow-up (n = 44). Seeking spiritual support was unrelated to mental health, both at baseline and at follow-up. In the United States, the relationship between spirituality and mental health was explored among a sample of African American (n = 88) and non– African American mothers (n = 101) in the Connecticut area (Douglas, Jimenez, Lin, & Frisman, 2008). Spirituality was measured with the Spiritual Well-being Scale, and multiple standardized batteries were used to measure 10 dimensions of mental health. Outcomes were assessed at baseline and three follow-up points over a 15-month period. Although no significant main effects emerged, a moderator analysis was significant. Among African American mothers, higher levels of spiritual wellbeing predicted lower levels of anxiety and posttraumatic stress over time. Among the possible explanations for the largely nonsignificant findings recorded in these two studies is the operationalization of spirituality. In other words, how spirituality was operationalized may account for the failure of spirituality to predict mental health (Hackney & Sanders, 2003).

A growing body of evidence suggests that spirituality is positively associated with women’s mental health (Dailey & Stewart, 2007). Similarly, reviews of the extant research on spirituality and mental health have found generally positive associations (Ano & Vasconcelles, 2005; Hackney & Sanders, 2003; H. G. Koenig, 2007; H. G. Koenig, McCullough, & Larson, 2001; Shreve-Neiger & Edelstein, 2004). Although it is important to note that the results are not uniformly positive, in aggregate, higher levels of spirituality tend to be linked to greater psychological well-being in hundreds of studies (H. G. Koenig, 2008). This emerging body of evidence is consistent with studies exploring coping strategies among homeless women. A number of qualitative studies have found that both women (Bhui, Shanahan, & Harding, 2006; Montgomery, 1994; Washington, Moxley, Garriott, & Weinberger, 2009) and

Spirituality and religion are increasingly defined as distinct but overlapping constructs (Derezotes, 2006; Miller & Thoresen, 2003). Spirituality is commonly defined in individual, existential, or relational terms, typically incorporating some reference to the sacred or the transcendent (Hill & Pargament, 2003; Hodge, 2005). Conversely,

246

Social Work Research Volume 36, Number 4 December 2012

SPIRITUALITY AND RELIGION AS MULTIDIMENSIONAL CONSTRUCTS

religion tends to be conceptualized in communal, organizational, or structural terms (Canda & Furman, 2010). As part of the process of distinguishing spirituality from religion, observers have increasingly recognized that these constructs are multidimensional (Miller & Thoresen, 2003). In other words, spirituality and religion, regardless of which is considered to be the more encompassing construct, consist of multiple dimensions in much the same way that mental health consists of multiple dimensions (for example, depression, anxiety) (Berry, 2005). Examples of various dimensions include forgiveness, spiritual meaning, and positive and negative approaches to spiritual coping (Idler et al., 2003). Thus, outcomes can vary depending on how spirituality is operationalized (Hackney & Sanders, 2003). In some cases, different dimensions of spirituality can even be inversely related to mental health (Ano & Vasconcelles, 2005). Positive spiritual coping, for example, tends to be positively associated with mental health, whereas negative spiritual coping tends to be inversely associated (Ano & Vasconcelles, 2005; H. G. Koenig et al., 2001; Pargament, 2002). Recognition of the multifaceted nature of spirituality and religion has sparked calls for the use of multidimensional measures in research exploring the relationship between spirituality and health outcomes (Berry, 2005; H. G. Koenig et al., 2001; Miller & Thoresen, 2003). Using psychometrically sound multidimensional instruments can help clarify which aspects of spirituality and religion are linked to mental health (Shreve-Neiger & Edelstein, 2004). Accordingly, this study used a multidimensional measure to examine the relationship between spirituality and religion and mental health among a sample of homeless mothers in the United States. As discussed in greater detail in the following sections, a prospective longitudinal design was used to conduct this examination. Specifically, this study explored the degree to which 11 dimensions of spirituality and religion were associated with nine dimensions of mental health over the course of a 15-month time span. METHOD

Participants

Participants were women (N = 222) enrolled in an emergency homeless shelter program in a large

Southwestern city. The program provided up to 120 days of emergency shelter and supportive services (for example, child care; employment education; social skills training; substance abuse treatment, material support in form of transportation assistance, clothing, food boxes, and toiletries) and 12-months of aftercare services to homeless families. To be enrolled in the program, participants had to meet the following criteria: be 18 years of age or older, be homeless (that is, lack a fixed, regular, or adequate nighttime residence), be a female head of household, have at least one child between two and 16 years of age living with the mother at program entry, and have a Diagnostic and Statistical Manual of Mental Disorders (4th ed.) (American Psychiatric Asssociation, 1994) Axis I diagnosis for either a mental health or substance use disorder in the past year. Over the course of 20 months, 423 women were screened, of which 262 met the eligibility criteria for program enrollment. Of these, 84.73% (N = 222) agreed to participate in the study. The mean age of the participants was 31.7 years (SD = 7.9), and their average number of children was 3.3 (SD = 1.9). Twenty-seven percent (n = 60) reported living with a partner at baseline. In terms of race/ethnicity, 47.7% (n = 106) identified as non-Hispanic white or Caucasian, 14.4% identified as Hispanic/Latina (n = 32), and the remaining participants selected alternative descriptors or declined to self-identify. Twenty-five participants (11.26%) could not be located for the final 15-month interview. An attrition analysis was conducted to see if baseline participants differed from those who completed the study. No significant differences emerged between the two groups on demographic, baseline mental health, and spirituality measures. Measures

Spirituality and religion were assessed using a slightly modified version of the NIA/Fetzer Short Form for the Measurement of Religion and Spirituality (Fetzer Institute, 1999; Idler et al., 2003). This measure is designed to assess 11 different dimensions of spirituality and religion that are theoretically related to health and well-being among adults of all ages. The instrument was validated with a nationally representative sample of adults (Idler et al., 2003) and has been widely used

Hodge, Moser, and Shafer / Spirituality and Mental Health among Homeless Mothers

247

(Maselko & Kubzansky, 2006; Neff, 2006; ShreveNeiger & Edelstein, 2004). The 11 domains, the number of items in each domain, a sample item from the subscale, means and standard deviations for each scale averaged across the four time points, and Cronbach’s alpha reliability coefficients for each subscale are listed in Table 1. Reliability coefficients were generally acceptable (Kline, 2000), particularly given the low number of items making up many of the domains (Cortina, 1993), and comparable to the coefficients recorded among the general population (Idler et al., 2003). The major exception was the beliefs and values domain, which recorded an unacceptable level of error (α = .46 in the present study compared with α = .64 in the validation study). As is the case with the following dependent measures, all subscales were scored so that higher values represent greater degrees of the individual construct assessed. Mental health was measured using the Brief Symptom Inventory (BSI) (Derogatis & Melisaratos, 1983). The BSI is a widely used measure of mental health status (Ryan, 2007) and has been used previously to assess psychological symptoms among other homeless populations (Solorio, Milburn, Andersen, Trifskin, & Rodgriquez, 2006), including homeless mothers (Douglas et al., 2008). It assesses emotional distress in the past seven days using a five-point scale (“not at all” = 0 to “extremely” = 4) in nine psychological dimensions: (1) somatization, (2) obsessive–compulsive behavior, (3) interpersonal sensitivity, (4) depression, (5) anxiety,

(6) hostility, (7) phobic anxiety, (8) paranoid ideation, and (9) psychoticism. Further information about these nine dimensions can be found in Derogatis and Melisaratos (1983). The BSI was previously validated with a primarily female sample (Derogatis & Melisaratos, 1983). Reliability coefficients for the nine subscales in the validation study ranged from .71 to .85, with similar values obtained in subsequent research with samples of homeless adolescents (Solorio et al., 2006) and mothers (Douglas et al., 2008). In this study, the coefficients for the nine subscales ranged from .73 to .90. Procedures

Under the supervision of a university institutional review board, participants were interviewed within one month of entering the program (baseline measurement). Subsequent interviews were conducted at three months, nine months, and 15 months post baseline. The interviews were conducted by trained interviewers. Participants were paid for their time. The amount of compensation was increased with each subsequent interview to maximize participant retention. Initial interviews were compensated at $30, three-month follow-up interviews were compensated at $40, and 15-month interviews were compensated at $60. Data Analytic Approach

To examine the relationship between spirituality and mental health outcomes over the 15-month

Table 1: Psychometric Properties of Spirituality and Religion Measures across All Time Points Number of Items

Measure (Sample Item)

Private religious practices (How often do you pray privately in places other than at church or synagogue?) Public religious practices (How often do you go to religious services?) Congregational problems (How often do people in your spiritual group make too many demands on you?) Negative spiritual coping (I wonder whether God has abandoned me.) Forgiveness (I know that God forgives me.) Congregational benefits (If you were ill, how much would the people in your congregation help you out?) Intensity (To what extent do you consider yourself to be a spiritual person?) Positive spiritual coping (I think about how my life is part of a larger spiritual force.) Daily spiritual experiences (I feel God’s presence.) Beliefs and values (I believe in a God who watches over me.) Spiritual meaning (The events in my life unfold according to a divine or greater Plan.)

248

M

SD

α

3 2

4.19 1.70 .77 2.56 1.16 .60

2 2 3

1.43 0.48 .63 1.55 0.62 .77 3.23 0.59 .68

2 2 3 6 3 2

2.43 2.53 2.59 3.77 3.05 3.77

0.93 0.65 0.74 1.21 0.45 1.21

.91 .63 .77 .89 .46 .72

Social Work Research Volume 36, Number 4 December 2012

period, separate multilevel longitudinal growth curve analyses were constructed to predict each of the nine mental health outcomes using the mixed regression procedure in SPSS version 17 (Raudenbush & Bryk, 2002). Within this framework, each mother received an intercept and a slope that defined her personal regression line. The intercept was an indicator of the initial level of mental health for a given mother at the beginning of the study period. The slope represented the individual’s rate of change. Additional explanatory variables, such as the 11 spirituality and religion measures, can be added to the model to ascertain their effects on the intercept. Interaction terms, comprising the explanatory variables × time, were used to assess the variables’ effect on the slope. This approach allowed us to examine the effect of spirituality on both initial mental health outcomes and the trajectory of mental health outcomes over time. Analysis proceeded in a hierarchical manner. As implied in the introduction, mental health is malleable and may be affected affected by homelessness. Examining the linear trend of the nine mental health outcomes provides an understanding of changes in mental health status over time. Toward this end, in level 1 we predicted each mental health outcome for an individual as a function of her intercept (for example, mental health score at baseline), her growth rate (slope), and the residual mental health score at each time point. In these models, the time variable was coded 0 = baseline interview, 1 = three-month interview, 2 = nine-month interview, and 3 = 15-month interview. Following the examination of the linear trend in mental health outcomes, the following demographic characteristics were included as timeinvariant covariates of the intercept and slope in the level 2 model: woman’s age at program admission (mean centered at age 31), whether she was living with a partner (0 = no, 1 = yes), whether she identified as Hispanic/Latina (0 = not Hispanic/Latina, 1 = Hispanic/Latina), and whether she identified as white/Caucasian (0 = white/Caucasian, 1 = not white/Caucasian). In addition, the mean centered scores of the 11 spirituality measures were calculated across the four time points and included as timeinvariant predictors of intercept and slope. This approach was selected because adult female spirituality tends to be relatively stable over time (Dalby, 2006; Koenig, McGue, & Iacono, 2008).

Thus, in level 2 we specified the intercept for an individual as a function of the average mental health outcome at baseline for 31-year-old, nonHispanic, Caucasian women, not living with a partner, with average scores on all 11 spirituality measures, and the average effects of each individual covariate/predictor on mental health outcomes at baseline. As implied earlier, the effects of the spirituality measures on the intercepts provides a cross-sectional understanding of the relationship between spirituality and mental health at the beginning of the study. We further specified the growth rate of an individual as a function of the average growth rate in mental health outcomes for 31-year-old, nonHispanic, Caucasian women not living with a partner, with average scores on all 11 spirituality measures, and the average individual effects of each individual covariate/predictor on the growth rate of mental health outcomes, controlling for every other predictor/covariate. The effect of the spirituality measures on the slopes provides an understanding of the effects of spirituality on mental health over the 15-month study period. Model fit was assessed using likelihood ratio tests. Full information maximum likelihood estimation was used to account for missing data (Enders & Bandalos, 2001). RESULTS

Correlations among Spirituality Measures

The correlations among the 11 spirituality measures, averaged across the four time points, are presented in Table 2. The correlations varied substantially across the measures, reflecting the multidimensional nature of the NIA/Fetzer instrument. Coefficients ranged from .88 to −.29. Linear Trend of Mental Health Outcomes

Mental health outcome scores over time are depicted in Table 3. With the exception of hostility, the coefficients were all negative, suggesting that mental health decreased over time. This apparent decrease was only significant for interpersonal sensitivity and depression. Inclusion of Demographic Covariates and Spirituality Measures

After examination of the linear trend of the mental health outcomes over time, the demographic

Hodge, Moser, and Shafer / Spirituality and Mental Health among Homeless Mothers

249

Table 2: Correlation Matrix of Spirituality and Religion Measures Measure

1. Private religious practices 2. Public religious practices 3. Congregational problems 4. Negative spiritual coping 5. Forgiveness 6. Congregational benefits 7. Intensity 8. Positive spiritual coping 9. Daily spiritual experiences 10. Beliefs and values 11. Spiritual meaning

1

2

3

4

5

6

7

8

9

10

11

— .68 .21 –.18 .50 .53 .58 .71 .74 .57 .55

— .26 –.16 .42 .62 .47 .60 .57 .48 .50

— .11 .11 .22 .20 .24 .16 .23 .23

— –.28 –.20 –.23 –.25 –.29 –.21 –.28

— .43 .53 .61 .65 .52 .52

— .43 .54 .55 .41 .42

— .64 .71 .62 .58

— .88 .73 .70

— .73 .74

— .66



Note: All correlations significant at p < .001.

Table 3: Linear Trend of Mental Health Outcomes Model 1

SOM

OCD

IS

DEP

ANX

HOS

PHOB

PAR

PSY

Intercept Time Deviance

0.72 –0.03 1,488.49

1.19 –0.04 1,885.86

0.98 –0.05* 1,982.43

1.04 –0.06* 1,977.68

0.84 –0.03 1,803.87

0.80 0.02 1,645.10

0.59 –0.03 1,641.45

1.04 –0.04 1,846.67

0.75 –0.03 1,632.38

Note: SOM = somatization; OCD = obsessive–compulsive behavior; IS = interpersonal sensitivity; DEP = depression; ANX = anxiety; HOS=hostility; PHOB = phobic anxiety; PAR = paranoid ideation; PSY = psychoticism. *p< .05.

covariates and spirituality measures were incorporated into the model (see Table 4). The inclusion of these predictors significantly improved model fit for each mental health outcome, as evidenced by log likelihood tests [each χ2(26 > 54.05, p < .001]. The effects of the spirituality measures on the intercepts are depicted in the top half of Table 4, and the effects on the slopes are depicted in the bottom half of the table (for example, Private Religious Practices × Time). Effects on Intercepts. Four spirituality measures were significantly related to mental health outcomes at baseline: (1) forgiveness, (2) negative spiritual coping, (3) beliefs and values, and (4) spiritual meaning. Of these, forgiveness exhibited the strongest and most pervasive relationship. This construct was associated with significantly lower scores on each of the nine mental health outcomes. This indicates that greater overall scores on forgiveness were associated with better mental health outcomes at baseline, controlling for all other spirituality measures and demographics. Two other spirituality measures also exhibited relatively consistent associations with mental health: negative spiritual coping and spiritual meaning. Greater overall scores on negative spiritual coping were associated with significantly

higher levels of interpersonal sensitivity, depression, anxiety, paranoid ideation, and psychoticism at baseline. In other words, with all other spirituality measures and demographics controlled for, negative spiritual coping was linked to worse mental health at baseline. Spiritual meaning was also linked to five mental health outcomes. More specifically, higher spiritual meaning scores were associated with significantly higher levels of interpersonal sensitivity, hostility, phobic anxiety, paranoid ideation, and psychoticism at baseline. In contrast to the other significant relationships that emerged at baseline, these relationships are inconsistent with previous research. In other words, the existing literature suggests that higher levels of spiritual meaning will be associated with better, rather than worse, mental health (Koenig, 2007; Koenig et al., 2001). Finally, the beliefs and values measure was associated with one mental health outcome. In keeping with extant theory and research, higher scores on beliefs and values were associated with significantly lower levels of paranoid ideation at baseline. Interpretation of this result is complicated, however, by the unacceptable level of error associated with the beliefs and values measure (that is, a Cronbach’s alpha of .46).

250

Social Work Research Volume 36, Number 4 December 2012

Hodge, Moser, and Shafer / Spirituality and Mental Health among Homeless Mothers

Table 4: Effects of Spirituality and Religion on Mental Health at Baseline and Over Time Model 2

Intercept Time Private religious practices Public religious practices Congregational problems Negative spiritual coping Forgiveness Congregational benefits Intensity Positive spiritual coping Daily spiritual experiences Beliefs and values Spiritual meaning Private Religious Practices × Time Public Religious Practices × Time Congregational Problems × Time Negative Spiritual Coping × Time Forgiveness × Time Congregational Benefits × Time Intensity × Time Positive Spiritual Coping × Time Daily Spiritual Experiences × Time Beliefs and Values × Time Spiritual Meaning × Time Mother’s age Living with partner at baseline Hispanic/Latina White Deviance

SOM

OCD

IS

DEP

ANX

HOS

PHOB

PAR

PSY

0.79 –0.02 –0.02 0.10 0.12 0.14 –0.23* –0.06 0.06 –0.02 0.003 0.10 0.18 0.02 –0.001 0.10* –0.03 0.03 –0.02 0.01 –0.02 –0.01 0.03 –0.07 0.01 –0.12 0.14 –0.22* 1,321.19

1.26 –0.03 –0.04 0.14 0.12 0.09 –0.34** –0.04 0.07 –0.12 0.02 –0.02 0.31 0.03 –0.02 0.17***† 0.03 0.06 –0.03 0.000 –0.06 –0.01 0.07 –0.04 0.00 –0.12 0.17 –0.29* 1,717.55

1.06 –0.04 0.02 0.13 0.11 0.26** –0.44***† –0.04 0.002 –0.11 0.05 –0.22 0.52***† 0.01 –0.05 0.20***† –0.03 0.08 –0.05 0.03 –0.04 0.02 0.20* –0.21*** 0.00 –0.14 0.19 –0.32***† 1,771.66

1.12 –0.04 0.02 0.09 0.04 0.23** –0.43** –0.07 0.04 –0.13 0.14 0.10 –0.02 0.01 –0.01 0.18***† 0.01 0.09 –0.05 –0.01 –0.08 0.02 0.07 –0.04 0.003 –0.23* 0.19 –0.24* 1,768.53

0.97 –0.01 0.04 0.12 0.18 0.20* –0.35** –0.01 0.04 –0.16 0.05 0.04 0.20 –0.004 –0.002 0.09 0.0001 0.07 –0.03 0.03 0.01 0.000 0.04 –0.11 0.002 –0.16 0.02 –0.32***† 1,604.91

0.78 0.03 0.01 –0.01 0.14 0.14 –0.27** 0.04 –0.02 –0.23 0.08 0.16 0.31* 0.01 –0.01 0.20*** 0.01 0.07 –0.04 –0.03 –0.06 0.07 0.10 –0.17** –0.01 0.04 0.12 –0.15 1,432.43

0.68 –0.02 0.05 0.07 0.02 0.15 –0.27* –0.03 0.11 –0.14 –0.02 –0.05 0.30* 0.01 –0.001 0.12* –0.04 –0.01 –0.02 0.03 0.01 –0.02 0.12 –0.13* –0.01 –0.21* 0.04 –0.22* 1,445.88

1.11 –0.03 0.04 0.07 0.15 0.22* –0.51***† –0.07 0.11 –0.08 0.13 –0.40* 0.37* –0.02 –0.01 0.14** –0.01 0.09 –0.04 –0.03 –0.03 0.01 0.16* –0.11 0.00 –0.19* 0.19 –0.24* 1,659.13

0.80 –0.02 0.00 0.07 –0.07 0.22**† –0.46***† –0.04 –0.01 –0.09 0.07 0.12 0.29* 0.01 –0.003 0.16*** 0.02 0.08 –0.04 0.001 –0.04 0.02 0.01 –0.09 0.00 –0.23** 0.16 –0.16* 1,407.38

Notes: Superscript daggers denote significance at the corrected alpha level. SOM = somatization; OCD = obsessive–compulsive behavior; IS = interpersonal sensitivity; DEP = depression; ANX = anxiety; HOS = hostility; PHOB = phobic anxiety; PAR = paranoid ideation; PSY = psychoticism. *p < .05. **p < .01. ***p < .001. †p .0045.

251

Effects on Slopes. Three spirituality measures demonstrated significant interaction effects with time. The measure with the most pervasive effect on mental health was congregational problems. Higher levels of congregational problems predicted significant increases in somatization, obsessive– compulsive behavior, interpersonal sensitivity, depression, hostility, phobic anxiety, paranoid ideation, and psychoticism over the 15 months. Spiritual meaning was associated with four mental health outcomes. In contrast to the crosssectional findings, higher levels of spiritual meaning predicted significant decreases in interpersonal sensitivity, hostility, and phobic anxiety over time. As noted, this finding is consistent with prior theory and research (Idler et al., 2003; Koenig et al., 2001). The beliefs and values measure predicted two outcomes: Higher levels were associated with significantly higher levels of interpersonal sensitivity and paranoid ideation over time. As is the case with the previous cross-sectional finding, the high level of error associated with the beliefs and values measure implies that this result should be treated cautiously.

It is widely recognized that mental health challenges contribute to the onset of homelessness and that the state of being homeless itself can also engender psychological problems (Philippot et al., 2007). As mothers cope with the stress of homelessness, it is all too easy for them to become ensnared in a downward spiral as the increasing severity of psychological distress compromises their ability to escape homelessness. Identifying strengths or protective factors can play a crucial role in helping mothers exit homelessness or avoid its onset. Toward this end, the present study used a longitudinal design to examine the relationship between spirituality and mental health among homeless mothers over the course of 15 months. Of the 11 dimensions of spirituality examined, the strongest, most pervasive relationship emerged for forgiveness. At baseline, forgiveness was positively associated with all nine dimensions of mental health. No interaction with time occurred, indicating a stable relationship between forgiveness and mental health over time (that is, the slope did not change over time). In other words, mental health outcomes did

not improve over time for those with higher levels of forgiveness beyond the relationship identified at baseline. The second most pervasive relationship occurred with congregational problems. Although this variable was not associated with mental health at baseline, it was associated with eight of the nine mental health dimensions over time. Congregational problems predicted increasingly worse mental health over the course of the 15-month study. Negative religious coping was positively related to five dimensions of mental distress at baseline. As was the case with forgiveness, the slope did not change over time. Higher levels of negative coping did not predict increasingly worse levels of mental health over time. Spiritual meaning was positively associated with five dimensions of mental health at baseline. Unlike the other significant relationships that emerged in this study, this finding runs counter to existing theory and research on spirituality and religion (Ano & Vasconcelles, 2005; Idler et al., 2003; Koenig et al., 2001; Pargament, 2002). It is interesting to note that the longitude findings were consistent with prior theory and research. In other words, over time, higher levels of spiritual meaning predicted better mental health. Although more research is needed to understand this finding, the results may be partially explained by dynamics related to homelessness. At baseline, mothers may have been wrestling with the onset of homelessness and induction into the program. Women with a strong sense of spiritual meaning, as exemplified by the sample item “The events in my life unfold according to a divine or greater Plan,” may have felt abandoned by God (Pargament, 1997; Pargament, 2002). However, over time, as they adjusted to their status and continued to understand life events through a spiritual lens, mental health may have improved. The results obtained in this study are also consistent with the existing qualitative research on spirituality and homelessness. As noted in the introduction, a number of studies indicate that spirituality is a strength that helps mothers deal with the stress of homelessness (for example, Cosgrove & Flynn, 2005; Meadows-Oliver, 2003). The present research corroborates this prior body of work and expands it by adding a longitudinal perspective. Conversely, the results differ from those obtained in previous longitudinal studies of spirituality and

252

Social Work Research Volume 36, Number 4 December 2012

DISCUSSION

mental health among homeless mothers, in which largely nonsignificant relationships have emerged (Douglas et al., 2008; Tischler & Vostanis, 2007). Although various explanations may account for the differences, it is notable that previous research used unidimensional spirituality measures, whereas this study used a multidimensional instrument (Idler et al., 2003). The differential results may be attributed to the fact that multiple dimensions of spirituality and religion were assessed in the present study, allowing for a more nuanced understanding of the relationship between spirituality and mental health among homeless mothers (Berry, 2005; Miller & Thoresen, 2003; Shreve-Neiger & Edelstein, 2004). The results have important implications for practitioners working with homeless mothers. Some qualitative research suggests that homeless mothers recognize the barriers that mental health problems present in terms of escaping homelessness but find it difficult to imagine pathways that might lead to improvement (Lee & Oyserman, 2009). Thus, interventions designed to foster improvement in this area may be particularly important. Toward this end, practitioners may wish to explore mothers’ understanding of forgiveness, perhaps while conducting a spiritual assessment (Hodge, 2004). The relationship between forgiveness and mental health is relatively well established (McCullough, Paragament & Thoresen, 2000). Interventions designed to foster forgiveness have appeared in the literature (DiBlasio, 1998). Practitioners may assist homeless mothers by tapping clients’ spiritual strengths to enhance forgiveness. In a similar manner, exploring clients’ sense of spiritual meaning may also help foster mental health. However, as is the case in all work that engages client spirituality, it is important to consider one’s degree of cultural competence with the client’s spiritual tradition and use collaboration or referral strategies as necessary (Richards & Bergin, 2000). Practitioners may also wish to ascertain the existence of congregational problems during a spiritual assessment. The collaborative exploration of strategies that alleviate these problems may foster mental health. In addition, resolving tensions between mothers and their congregations, with clients’ consent, may have the added, secondary benefit of connecting mothers with social support resources that can help them in their efforts to escape homelessness (Irwin, LaGory, Ritchey, & Fitzptrick, 2008; Zugazaga, 2008).

These implications must be considered tentatively in light of the study’s limitations. For example, the low reliabilities obtained with some of the spirituality and religion measures should be noted. One measure had an alpha coefficient of just .46, and four others had coefficients between .60 and .70. Although the small number of items comprised by these subscales helps account for the low reliabilities (Cortina, 1993), those coefficients all fall below the widely accepted .70 value (Kline, 2000). The lack of generalizability should also be noted. The sample was drawn from one homeless program with a unique set of services. It is possible that the set of services provided by the program affected the results. Results might vary with different programs or samples, particularly in geographic regions in which the background demographics differ. For instance, outcomes might differ with samples of African American mothers, given the salience of spirituality in the black community (Douglas et al., 2008; Stahler, Kirby, & Kerwin, 2007). Further research is needed with other samples to confirm the findings obtained in this study. Future research should use multidimensional measures to assess both spirituality and mental health (Berry, 2005; Miller & Thoresen, 2003; ShreveNeiger & Edelstein, 2004). Given the growing number of homeless mothers in the United States, such research should be prioritized (Goldberg, 1999; Paquette & Bassuk, 2009). CONCLUSION

Female-headed familie may constitute the fastest growing segment of the homeless population (Arangua et al., 2006; Goldberg, 1999; Paquette & Bassuk, 2009; Richards et al., 2010). Many of these women wrestle with mental health issues that inhibit their escape from homelessness (Lee & Oyserman, 2009). Using a longitudinal design with multidimensional measures of spirituality and mental health, this study identified a number of spiritual and religious variables that may help mitigate mental health problems. In short, the present results clarify some of the pathways that may help mothers exit homelessness or avoid it entirely. REFERENCES Ano, G. G., & Vasconcelles, E. B. (2005). Religious coping and psychological adjustment to stress: A metaanalysis. Journal of Clinical Psychology, 61, 461–480.

Hodge, Moser, and Shafer / Spirituality and Mental Health among Homeless Mothers

253

Arangua, L., Andersen, R., & Gelberg, L. (2005). The health circumstances of homelessness women in the United States. International Journal of Mental Health, 34, 62–92. Averitt, S. S. (2003). “Homelessness is not a choice!” The plight of homeless women with preschool children living in temporary shelters. Journal of Family Nursing, 9(1), 79–100. Banyard, V. L., & Graham-Bermann, S. A. (1998). Surviving poverty: Stress and coping in the lives of housed and homeless mothers. American Journal of Orthopsychiatry, 68, 479–489. Berry, D. (2005). Methodological pitfalls in the study of religiosity and spirituality. Western Journal of Nursing Research, 27, 628–647. Bhui, K., Shanahan, L., & Harding, G. (2006). Homelessness and mental illness: A literature review and a qualitative study of perceptions of the adequacy of care. International Journal of Social Psychiatry, 52, 152–165. Canda, E. R., & Furman, L. D. (2010). Spiritual diversity in social work practice: The heart of helping (2nd ed.). New York: Oxford University Press. Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98–104. Cosgrove, L., & Flynn, C. (2005). Marginalized mothers: Parenting without a home. Analysis of Social Issues and Public Policy, 5, 127–143. Dailey, D. E., & Stewart, A. L. (2007). Psychometric characteristics of the spiritual perspectives scale in pregnant African-American women. Research in Nursing & Health, 30, 61–71. Daiski, I. (2007). Perspectives of homeless people on their health and health needs priorities. Journal of Advanced Nursing, 58, 273–281. Dalby, P. (2006). Is there a process of spiritual change or development associated with aging? A critical review of research. Aging and Mental Health, 10(1), 4–12. Derezotes, D. S. (2006). Spiritually oriented social work practice. Boston: Pearson. Derogatis, L. R., & Melisaratos, N. (1983). The Brief Symptom Inventory: An introductory report. Psychological Medicine, 13, 595–605. DiBlasio, F. A. (1998). The use of a decision-based forgiveness intervention within intergenerational family therapy. Journal of Family Therapy, 20, 77–94. Douglas, A. N., Jimenez, S., Lin, H.-J., & Frisman, L. K. (2008). Ethnic differences in the effects of spiritual well-being on long-term psychological and behavioral outcomes within a sample of homeless women. Cultural Diversity and Ethnic Minority Psychology, 14, 344–352. Enders, C. K., & Bandalos, D. L. (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation modeling. Structural Equation Modeling: An Interdisciplinary Journal, 8, 430–457. Fertig, A. R., & Reingold, D. A. (2008). Homelessness among at-risk families with children in twenty American cities. Social Service Review, 82, 485–510. Fetzer Institute. (1999). Multidimensional measurement of religiousness/spirituality for use in health research. Retrieved from http://www.fetzer.org/component/content/ article/18-main/248-dses Fraser, M. W., Richman, J. M., & Galinsky, M. J. (1999). Risk, protection, and resilience: Toward a conceptual framework for social work practice. Social Work Research, 23, 131–143. Gelberg, L., Browner, C. H., Lejano, E., & Arangua, L. (2004). Access to women’s health care: A qualitative

study of barriers perceived by homeless women. Women & Health, 40, 87–100. Goldberg, J. E. (1999). A short term approach to intervention with homeless mothers: A role for clinicians in homeless shelters. Families in Society, 80, 161–168. Hackney, C. H., & Sanders, G. S. (2003). Religiosity and mental health: A meta-analysis. Journal for the Scientific Study of Religion, 42, 43–55. Hill, P. C., & Pargament, K. I. (2003). Advances in the conceptualization and measurement of religion and spirituality. American Psychologist, 58, 64–74. Hodge, D. R. (2004). Spirituality and people with mental illness: Developing spiritual competency in assessment and intervention. Families in Society, 85, 36–44. Hodge, D. R. (2005). Spiritual ecograms: A new assessment instrument for identifying clients’ spiritual strengths in space and across time. Families in Society, 86, 287–296. Idler, E. L., Musick, M. A., Ellison, C. G., George, L. K., Krause, N., Ory, M. G., et al. (2003). Measuring multiple dimensions of religion and spirituality for health research. Research on Aging, 25, 327–365. Irwin, J., LaGory, M., Ritchey, F., & Fitzptrick, K. (2008). Social assets and mental distress among the homeless: Exploring the roles of social support and other forms of social capital on depression. Social Science & Medicine, 67, 1935–1943. Kline, P. (2000). The handbook of psychology testing (2nd ed.). New York: Routledge. Koenig, H. G. (2007). Spirituality in patient care (2nd ed.). Philadelphia: Templeton Foundation Press. Koenig, H. G. (2008). Religion and mental health: What should psychiatrists do? Psychiatric Bulletin, 32, 201–203. Koenig, H. G., McCullough, M. E., & Larson, D. B. (2001). Handbook of religion and health. New York: Oxford University Press. Koenig, L. B., McGue, M., & Iacono, W. G. (2008). Stability and change in religiousness during emerging adulthood. Developmental Psychology, 44, 532–543. Larkin, H., Beckos, B., & Martin, E. (in press). Applied integral methodological pluralism: Designing comprehensive social service program evaluation. In S. Esbjorn-Hargens (Ed.), Enacting an integral future: New horizons for integral theory. SUNY Press. Lee, S. J., & Oyserman, D. (2009). Expecting to work, fearing homelessness: The possible selves of lowincome mothers. Journal of Applied Social Psychology, 39, 1334–1355. Lehmann, E. R., Kass, P. H., Drake, C. M., & Nichols, S. B. (2007). Risk factors for first-time homelessness in low-income women. American Journal of Orthopsychiatry, 77, 20–28. Maselko, J., & Kubzansky, L. D. (2006). Gender differences in religious practices, spiritual experiences and health: Results from the US General Social Survey. Social Science & Medicine, 62, 2848–2860. McCullough, M. E., Paragament, K. I., & Thoresen, C. E. (Eds). (2000). Forgiveness. New York: Guilford Press. Meadows-Oliver, M. (2003). Mothering in public: A meta-synthesis of homeless women with children living in shelters. Journal for Specialists in Pediatric Nursing, 8, 130–136. Miller, W. R., & Thoresen, C. E. (2003). Spirituality, religion, and health: An emerging research field. American Psychologist, 58, 24–35. Montgomery, C. (1994). Swimming upstream: The strengths of women who survive homelessness. Advances in Nursing Science, 16, 34–45. Neff, J. A. (2006). Exploring the dimensionality of “religiosity” and “spirituality” in the Fetzer

254

Social Work Research Volume 36, Number 4 December 2012

multidimensional measure. Journal for the Scientific Study of Religion, 45, 449–459. Paquette, K., & Bassuk, E. L. (2009). Parenting and homelessness: Overview and introduction to the special issue. American Journal of Orthopsychiatry, 79, 292–298. Pargament, K. I. (1997). The psychology of religion and coping. New York: Guilford Press. Pargament, K. I. (2002). The bitter and the sweet: An evaluation of the costs and benefits of religiousness. Psychological Inquiry, 13, 168–181. Philippot, P., Lecocq, C., Sempoux, F., Nachtergael, H., & Galand, B. (2007). Psychological research on homelessness in Western Europe: A review from 1970–2001. Journal of Social Issues, 63, 483–504. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods. Thousand Oaks, CA: Sage Publications Richards, P. S., & Bergin, A. E. (Ed.). (2000). Handbook of psychotherapy and religious diversity. Washington, DC: American Psychological Association. Richards, T. N., Garland, T. S., Bumphus, V. W., & Thompson, R. (2010). Personal and political? Exploring the feminization of the American homeless population. Journal of Poverty, 14, 97–115. Rollins, J. H., Saris, R. N., & Johnston-Robledo, I. (2001). Low-income women speak out about housing: A high-stakes game of musical chairs. Journal of Social Issues, 57, 277–298. Ryan, C. (2007). British outpatient norms for the Brief Symptom Inventory. Psychology and Psychotherapy: Theory, Research and Practice, 80, 183–191. Schiff, J. W. (2007). Homeless families in Canada: Discovering total families. Families in Society, 88, 131–140. Shreve-Neiger, A. K., & Edelstein, B. A. (2004). Religion and anxiety: A critical review of the literature. Clinical Psychology Review, 24, 379–397. Smith, E. J. (2006). The strength-based counseling model. The Counseling Psychologist, 34, 13–79. Solorio, M. R., Milburn, N. G., Andersen, R. M., Trifskin, S., & Rodgriquez, M. A. (2006). Emotional distress and mental health service use among urban homeless adolescents. Journal of Behavioral Health Services & Research, 33, 381–393. Stahler, G. J., Kirby, K. C., & Kerwin, M. E. (2007). A faith-based intervention for cocaine-dependent black women. Journal of Psychoactive Drugs, 39, 183–190. Stainbrook, K. A., & Hornik, J. (2006). Similarities in characteristics and needs of women with children in homeless family and domestic violence shelters. Families in Society, 87, 53–62. Tischler, V., Rademeyer, A., & Vostanis, P. (2007). Mothers experiencing homelessness: Mental health, support and social care needs. Health and Social Care in the Community, 15, 246–253. Tischler, V. A., & Vostanis, P. (2007). Homeless mothers: Is there a relationship between coping strategies, mental health and goal achievement? Journal of Community & Applied Social Psychology, 17, 85–102. U.S. Department of Housing and Urban Development. (2010). The 2009 annual homeless assessment report to Congress. Retrieved from http://www.hudhre.info/ documents/5thHomelessAssessmentReport.pdf Wachholz, S. (2005). Hate crimes against the homeless: Warning-out New England style. Journal of Sociology and Social Welfare, 32, 141–163. Washington, O.G.M., Moxley, D. P., Garriott, L., & Weinberger, J. P. (2009). Five dimensions of faith and spirituality of older African American women transitioning out of homelessness. Journal of Religion & Health, 48, 431–444.

Weinreb, L., Nicholson, J., Anthes, F., & Williams, V. (2007). Integrating behavioral health services for homeless mothers and children in primary care. American Journal of Orthopsychiatry, 77, 142–152. Williams, J. K., & Hall, J. A. (2009). Stress and traumatic stress: How do past events influence current traumatic stress among mothers experiencing homelessness? Social Work Research, 33, 199–207. Zlotnick, C., Tam, T., & Bradley, K. (2007). Impact of adulthood trauma on homeless mothers. Community Mental Health Journal, 43, 13–32. Zugazaga, C. B. (2008). Understanding social support of the homeless: A comparison of single men, single women, and women with children. Families in Society, 89, 447–455.

David R. Hodge, PhD, is associate professor, School of Social Work, Arizona State University, Phoenix, and senior nonresident fellow, Program for Research on Religion and Urban Civil Society, University of Pennsylvania, Philadelphia. Stephanie E. Moser, MA, is a doctoral student, Department of Psychology, Arizona State University, Tempe. Michael S. Shafer, PhD, is professor, School of Social Work, and director, Center for Applied Behavioral Health Policy, Arizona State University, Phoenix. The research reported in this article was supported by funding from the Substance Abuse and Mental Health Services Administration (SAMHSA) (Cooperative Agreement 5 UD1 SM53054). The present study analyses site-specific data collected as part of this federally funded investigation. The analysis and interpretation of the results reported in the article are strictly those of the authors, and no endorsement by SAMHSA is to be inferred. Address correspondence to David R. Hodge, School of Social Work, Arizona State University, Mail Code 3920, 411 North Central Avenue, Suite 800, Phoenix, AZ 85004-0689. Original manuscript received April 1, 2010 Final revision received October 15, 2010 Accepted November 17, 2010 Advance Access Publication January 9, 2013

Hodge, Moser, and Shafer / Spirituality and Mental Health among Homeless Mothers

255