What’s Faith Got to Do with It? Religiosity Among Women Who Use Methamphetamine

Abstract

Religiosity is not found to be consistently protective in mental health and substance use outcomes among illicit drug users. This study examines the association between religiosity, mental health and drug use among a community-recruited sample of women who use methamphetamine. The majority of the sample (74%) had high scores of religious faith. In multivariate analysis, those with high scores had higher odds of self-reporting a mental health diagnosis and of being psychologically dependent upon methamphetamine, and were less likely to report injection risk. Further examination of the role of religiosity in the lives of women who use methamphetamine is advised.

The protective effects of religiosity on mental health have been well documented. Studies of mental well-being routinely find that extrinsic religiosity (i.e. church attendance; people who use their religion) is protective against psychological distress (Levin & Chatter, 1998) and internalizing disorders such as major depression and generalized anxiety disorders (Kendler et al., 2003). For example, older adults who were highly religious had lower levels of depression (Roff et al., 2004). In a meta-analysis of 35 studies, religion was positively associated with psychological adjustment (Hackney & Sanders, 2003). Levels of intrinsic religiosity (i.e. internal beliefs like faith; people who live their religion) are also associated with mental well being (Payne, Bergin, Bielma, & Jenkins, 1991). Plante and Boccaccini’s (1997a) sample of undergraduate students found an association between strength of religious faith and both self-esteem and adaptive coping. Among homeless mothers, scores of spiritual well being among African-American women were inversely associated with higher levels of anxiety and rates of trauma symptoms (Douglas, 2008). Similarly, Koenig and Larson’s (2001) systematic review of 850 studies revealed that those studies where religiosity was associated with mental health outcomes, approximately two-thirds found rates of depression and anxiety to be lower among those who were more religious.

Among religiosity and addiction research it is a common finding that there is an inverse relationship between religiosity and substance abuse (Geppert, Bogenschutz, & Miller, 2007; Kendler et al., 2003; Koenig, McCullough & Larson, 2001). A study of female-female twin pairs found that personal devotion and personal and institutional conservativism were inversely associated with levels of drinking and lifetime risk for alcoholism (Kendler, Gardner, & Prescott, 1997). However, the aspects of religiosity that are measured have the potential to alter this association. It has been suggested that the likelihood of developing a substance use problem may be associated with an individual’s perception of God, where those who maintain a vision of a punitive God are thought to be at an increased risk of developing alcohol or drug related problems (Gorsuch, 1995). Furthermore, some researchers have posited that a depleted sense of faith may motivate people towards drug or alcohol use to fulfill unmet needs, or, what Grof refers to as the “thirst for wholeness” (Duvall, Staton-Tindall, & Leukefeld, 2008; Grof, 1994).

Most studies that examine religiosity do so among non-drug using populations. Most studies that examine religiosity do so among non-drug using populations. The limited literature on religiosity among current substance users show that they often differ from non-substance users. Fallot and Heckmans’ (2005) study of women with co-occurring mental health and substance use disorders found that negative spiritual coping, when someone feels that events in their lives are representative of God’s punishment or abandonment and then choose to address future situations without God’s assistance, was associated with mental health symptoms and severity of post-traumatic symptoms. A study of Drug Court clients found that self reports of higher levels of persistent faith across time was associated with less frequent substance use (Duvall et al., 2008), yet a study of rural stimulant users found that for women religiosity was positively correlated with crack cocaine use. This same study however found that religiosity was negatively associated with frequency of methamphetamine use (Staton-Tindall, 2008). Others have also found an inverse association between religiosity and using certain drugs, and a positive association with other drugs. Chitwood’s (2008) review of the literature on religiosity among current drug users, found that religiosity was protective for marijuana use in 84% of the studies, but protective for powder and rock cocaine in only 43% of the studies.

Findings about the association between religiosity and injection drug risk behaviors are also inconsistent. A study of cocaine using, and opiate dependent individuals enrolled in a methadone maintenance program found that strength of religious faith was not an independent predictor of injection risk behaviors (Avants, Marcotte, Arnold, & Margolin, 2003). Although Des Jarlais et al. (1997) found an association among injection drug users between faith and the avoidance of HIV risk behaviors, other studies have found that religiosity is associated with reuse of injection equipment (Hasnain, Sinacore, Mensah, & Levy, 2005; Weiss, Chitwood, & Sanchez, 2008).

Few studies have looked at the association between religiosity and substance use among community-recruited populations of drug users (Chitwood, Weiss, & Leukefeld, 2008). The present study examines the association between strength of religious faith, drug use, mental health and HIV-related injection risk behaviors among a cross-section of women in San Francisco, CA who are active methamphetamine users. In this study, our two primary research questions were: (1) what demographic factors are associated with strength of religious faith among women who are active methamphetamine users? And (2) is strength of religious faith associated with mental health, substance use, HIV-related injection risk behaviors, or involvement and/or interest in drug treatment?

To our knowledge, there have been no studies that explore strength of religious faith among urban women who use methamphetamine. Since some researchers have hypothesized that a depleted sense of faith may motivate people towards problematic patterns of drug use (Duvall et al., 2008; Grof, 1994), we would expect to find low rates of religious faith among this sample of women who use methamphetamine. However, taking into account the inconsistent findings from religiosity research among current substance users, we did not make any predictions about the scores of religious faith. Likewise, with some researchers finding a positive association between religiosity, mental health and substance use behaviors (Fallot & Heckman, 2005; Staton-Tindall, 2008), and others an inverse association (Chitwood, 2008; Payne et al., 1991), we predicted that religiosity would be associated with mental health and substance use behaviors, but did not specify the directionality of association. With drug treatment experiences and desires, we predicted a positive association with strength of religious faith as we felt that those with a high strength of religious faith would be motivated to decrease or stop their methamphetamine use so that their behaviors would be more in line with their religious faith.

METHODS

We enrolled 322 women who use methamphetamine from 2007 to 2009 in San Francisco, CA. Participants were recruited using respondent driven sampling (RDS), a methodology often utilized for hard to reach populations (Heckathorn, 1997; Iguchi et al., 2009). Briefly, RDS is a form of chain-referral sampling where a group of initial recruits (or “seeds”) are identified by the research team. These initial seeds are enrolled in the study, interviewed, and then given coupons to recruit other eligible participants. Eligibility requirements were (1) biological female, (2) methamphetamine use in the past 30 days, (3) 18 years of age or older, (4) had sex with at least one man in the past 6 months, and (5) referred by another participant with an RDS recruitment coupon (except initial recruits, or ‘seeds’). Eligibility was determined through a screening process that masked criteria by including several questions unrelated to eligibility. Data were collected at a community-based field site, where all participants gave informed consent and completed a computer assisted personal interview (CAPI) with a trained interviewer in a confidential space.

The study was approved by the Institutional Review Board at RTI International, and a Certificate of Confidentiality was obtained from the National Institutes of Health. To compensate the participants for their time and expertise, they received $40 for the interview, and $10–$20 (this incentive was increased midway through the study to improve recruitment) for each eligible referral.

MEASURES

Strength of Religious Faith Variable

Religious faith was measured by the Santa Clara Strength of Religious Faith Questionnaire (SCSORFQ). This 10-item measure is scored on a 4-point scale (range 10–40), with the convention being that a score of 26 or greater is considered high (Plante & Boccaccini, 1997b). The questionnaire assesses strength of religious faith and does not assume that a person is religious or of a specific affiliation. The questionnaire’s internal consistency has been shown to be high (Cronbach’s alpha = 0.94 – 0.97), and it has demonstrated both high reliability and validity (Plante & Boccaccini, 1997a, 1997b; Sherman et al., 1999).

The SCSORFQ has been positively correlated with intrinsic religiosity (a personal sense of the importance of religion), perceived coping and hope. The questionnaire is inversely associated with anxiety and depression, and inversely correlated with God control (Plante, Saucedo, & Rice, 2001; Plante, Yancey, Sherman, Guertin, & Pardini, 1999). Using the Belief in Personal Control Scale (BPCS; Berrenberg, 1987), Plante and Boccaccini (1997b) found that those with a high score of religious faith also scored lower on the God-mediated control dimension of the BPCS. This means that those who have a high strength of faith have a greater belief in personal control as opposed to God being an absolute arbiter of life events. Similarly, among undergraduate students high scores on the SCSORFQ were negatively correlated with God control (Plante & Boccaccini, 1997a) as assessed by the Religious Life Inventory (RLI; Batson & Ventis, 1982), a scale that measures three religious orientations: means (extrinsic religiosity), end (intrinsic religiosity), and quest (interactional religiosity). This association between strength of religious faith and God control, as evidenced by the lower scores on the BPCS and RLI, indicates that the person does not believe that God is the arbiter of control, but instead that God can be enlisted in the achievement of outcomes (Berrenberg, 1987).

For descriptive variables (age, race, homeless, and knowledge of positive HIV status), the SCSORFQ score was the outcome variable. For all other variables, the SCSORFQ score served as the independent variable.

Independent and Potential Confounding Variables

We report age as a categorical variable: 18–29; 30–39; 40–49; ≥50. For race, we asked “which one of these racial/ethnic groups do you consider yourself?” and read a list of eight options that included “other” and “mixed.” Homelessness was defined as an affirmative response to the question, “Do you consider yourself homeless?” Known HIV status was determined by the question, “Have you ever been given a positive HIV test result?”

Explanatory and Outcome variables

Several questions measured mental health. To determine whether a participant had recently received outpatient mental health care, we asked, “In the past 6 months, did you receive any outpatient treatment or counseling for any problem you were having with your emotions, nerves, or mental health? Please do not include treatment for alcohol or drug use.” Answers to these two questions were dichotomized yes or no.

Current levels of depression symptomatology were assessed with the CES-D scale (Radloff, 1977). Each item is scored on a 4-point scale, with a score range of 0 to 60. Higher scores are indicative of more depressive symptoms. Scores of 16 or greater on the scale are interpreted as suggestive of clinically significant depression (Radloff, 1977).

Symptomatology of trauma from sexual abuse was assessed with the Sexual Abuse Trauma Index subscale of the Trauma Symptom Checklist-40 (TSC-40). This subscale is comprised of 7 items and is scored on a 4-point scale, with the range of scores being from 0 to12 (Briere & Runtz, 1998). Scores of 4 or greater on the scale are labeled as suggestive of trauma from sexual abuse.

Utilizing the Severity of Dependence scale (Gossop et al., 1995), we assessed the women’s psychological dependence on methamphetamine. This five item scale measures a person’s psychological dependence upon different types of illicit drugs through inquiry about the past 12 months. In this study we only used the scale for one drug, methamphetamine. Each item is scored on a 4-point scale, and the total score is a sum of the items. Higher total scores are indicative of higher levels of dependence. We grouped the scores into three categories: a high score was 11–15, medium was 5–10 and low was 0–4.

Several questions were asked to explore the frequency of methamphetamine use and the participant’s desire to decrease use. We also assessed desire to cut back on their methamphetamine use with the following two questions: “Are you seriously thinking of cutting down or quitting meth?” and “Have you cut down or quit meth for any time in the past 30 days?”

We also queried about other drug use and injection related risk behaviors. Two sets of questions were asked, one for non-injection drug use and the other for injection drug use, with both assessing use in the past 30 days. Answers were dichotomized into yes or no. Participants that reported injection drug use in the past 6 months were asked “In the past 30 days, have you been injected by another person?”, “How many people did you give or loan your used needles to, including a close friend or lover, in the past 30 days?”, and “How many different people did you inject behind or follow, that is inject with a needle someone else has already used, including that of a close friend or lover, in the past 30 days?” The first item was dichotomized as yes or no; the last two items were dichotomized into none and one or more.

All participants were asked about experiences with drug treatment. If they answered yes to, “Have you ever participated in any type of drug treatment program including outpatient, inpatient, or residential treatment; methadone; acupuncture, or 12-Step (e.g., NA, AA)?” they then were asked about involvement in the past six months as well as current involvement. We also explored the likelihood of participants wanting to participate in outpatient treatment programs for their methamphetamine use. We dichotomized the following question into unlikely and likely: “How likely is it that you would participate in an outpatient program designed to help you stop using meth?”

Analyses

We analyzed data from 321 women, as religiosity data were missing for one participant, and conducted a series of bivariate and multivariate logistic regression analyses to examine associations between strength of religious faith and mental health, substance use, and HIV-related drug risk behaviors. Variables that were significantly associated with religious faith in bivariate analysis at p<0.10 level were entered into multivariate logistic regression models. Only significant variables at p<0.05 level were retained in the final model. Second, we conducted a series of analyses assessing whether strength of religious faith was associated with several outcomes. For each of these outcomes, we first assessed the bivariate relationship and then followed up with multivariate models that forced strength of religious faith into the model, while also controlling for potential confounding variables including demographic variables. The p value for strength of religious faith needed to be below 0.05 for it to be considered as statistically significantly associated with the outcome.

RESULTS

Sample Characteristics

Almost half of the sample was African American, and slightly more than half of the women were forty years of age or older (see Table 1). Fifty-seven percent of the sample considered themselves homeless. Nearly half of the women injected methamphetamine in the past 30 days, while 85% reported non-injection methamphetamine use.

Table 1.

Sample Characteristics of Women Who Use Methamphetamine in San Francisco (n=321)

	Crude %
Race/ethnicity    African American    White    Native American    Latina    Asian or Pacific Islander    Mixed race    Other/refused	45.9 32.8 4.7 4.4 2.2 9.4 0.6
Age    18–29     30–39     40–49     50 or older	20.9 22.7 32.7 23.7
Homeless	56.9
Methamphetamine use past 30 days Non-injection use Injection use	85.1 47.0
Other Drug Use past 30 days Injected heroin, speedball or goofball Used crack cocaine Smoked marijuana	27.6 61.8 58.3

The mean strength of religious faith score assessed among individuals in this sample was 29.7, and the median score was 31 (quartile range: 25, 37). Nearly three-quarters (73.8%) had a high score (26 or higher). The Cronbach alpha score was 0.95, indicating that the SCSORFQ has a high level of internal consistency when used with women who are current methamphetamine users. Table 2 reports the crude frequencies for each item in the scale.

Table 2.

Prevalence of Responses to Strength of Religious Faith Questionnaire Scale Items among Women who Use Methamphetamine in San Francisco (N=321)

SCSORFQ item

% Strongly Agree

My religious faith is extremely important to me. I pray daily. I look to my faith as a source of inspiration. I look to my faith as providing meaning and purpose in my life. I consider myself active in my faith or church. My faith is an important part of who I am as a person. My relationship with God is extremely important to me. I enjoy being around others who share my faith. I look to my faith as a source of comfort. My faith impacts many of my decisions.

77.5 67.8 75.4 73.8 36.9 75.3 83.4 71.7 82.2 61.7

Identifying as African-American and being older were significantly associated with strength of religious faith in bivariate analysis (see Table 3). Because of this association, and since others have found these two variables to be associated with religiosity (Miller, 1998), we controlled for both race and age in all multivariate analyses.

Table 3.

Bivariate Analysis of Demographic Variables Associated With Strength of Religious Faith among Women who Use Methamphetamine (n = 321)

Demographics	Adjusted Odds Ratio	95% Confidence Interval
Homeless	1.22	(0.71, 2.09)
Race: African-American	3.73	(2.12, 6.55)
Age: 18–29 vs. 50+	0.31	(0.14, 0.66)
Know HIV-positive result	2.80	(0.78, 10.08)

^*bolded items indicate significance at the p<.05 level.

Mental Health, Drug Use, Drug Treatment, and Injection Risk Behaviors

Table 4 displays the results from bivariate and multivariate analysis. Strength of religious faith was positively associated with severity of psychological dependence on methamphetamine and with self-reported mental health diagnosis in bivariate and multivariate analysis where race and age were controlled for. Strength of religious faith was not associated with current levels of depression symptomatology or with symptomatology of sexual trauma.

Table 4.

Characteristics and Behaviors Associated With Strength of Religious Faith in Bivariate Analysis and Multivariate Analysis (n = 321)

Characteristics or Behaviors	Bivariate OR (90% CI)*	Multivariate AOR (95% CI)
Mental Health
Self-report mental health diagnosis	1.55 (0.93, 2.56)	1.95 (1.12, 3.40)
Outpatient mental health care ≤ 6 months	1.07 (0.64, 1.78)	--
Depression Symptomatology	1.28 (0.65, 2.54)	--
Sexual Trauma Symptomatology	1.06 (0.56, 1.99)	--
Psychological dependence on MA	3.57 (1.41, 9.08)	3.02 (1.15, 7.97)
MA Use past 30 days
Injection	0.62 (0.37, 1.02)	--
Binge use	1.12 (0.69, 1.85)	--
Desire to Decrease use	1.60 (0.85, 3.03)	--
Self reported decreased use	1.41 (0.83, 2.39)
Non-IDU Drug Use past 30 days
Marijuana	1.80 (1.09, 2.97)	1.94 (1.13, 3.32)
Heroin	1.38 (0.60, 3.13)	--
Crack Cocaine	1.45 (0.88, 2.40)	--
IDU Drug Use
Ever inject	0.83 (.48, 1.43)	--
IDU Drug Use past 30 days
Heroin	0.45 (0.26, 0.77)	--
Speedball	0.64 (0.33, 1.22)	--
Goofball	0.71 (0.37, 1.37)	--
Crack cocaine	0.68 (0.30, 1.59)	--
Injection Risk Behaviors
Injected by someone (30 days)	0.91 (0.47, 1.77)	--
Loaned needles (30 days)	0.25 (0.09, 0.71)	0.24 (0.07, 0.81)
Injected behind another person (30 days)	0.42 (0.18, 0.95)	--
Drug Treatment
Ever any drug treatment	0.73 (1.01, 2.94)	--
Likelihood of participating in outpatient program for MA use	0.76 (0.91, 3.39)	--
Current Drug Treatment
12-step	4.99 (1.50, 16.66)	6.96 (2.03, 23.81)
Methadone Maintenance	0.90 (0.44, 1.85)	--
Outpatient	1.24 (0.51, 2.99)	--

^*bolded items indicate significance at the p<.10 level.

In bivariate analysis, strength of religious faith was protective against injection methamphetamine use in the past 30 days. Controlling for race and age eliminated this association. Binge methamphetamine use in the past 30 days and desire to reduce methamphetamine use were not associated with strength of religious faith.

In multivariate analysis, the association between marijuana use in the past 30 days and strength of religious faith remained. Strength of religious faith was not statistically associated with all other non-injection and injection drug use in the past 30 days, and it was not associated with having ever injected a drug.

Among the 163 women who reported injection drug use in the past 6 months, lending used needles and injecting behind another person were associated with lower strength of religious faith in both bivariate. Only the association between lending used needles and strength of religious faith remained significant in multivariate analyses. We saw no statistically significant association between strength of faith and being injected by another person in the past 30 days.

In multivariate analysis an association was found between strength of religious faith and current 12-step program involvement. Strength of religious faith was not statistically associated with past or current involvement in outpatient, inpatient, residential, methadone, or acupuncture treatment programs.

DISCUSSION

This study provides one of the few sources of data on the strength of religious faith among a community-recruited sample of women who use illicit substances, specifically methamphetamine. The fact that almost half of the sample is comprised of African-American women provides a unique perspective since studies of methamphetamine users are largely comprised of white populations. Our findings demonstrate that there is a high degree of religious faith among urban women, especially African-American and older women (≥ 50 years old), who use methamphetamine. We found that the SCSORFQ maintains its high internal reliability among this population, and that although the mean score was higher than those reported in studies with non-drug using populations (Plante & Boccaccini, 1997a, 1997b; Plante et al., 2001; Plante, Vallaeys, Sherman, & Wallston, 2002; Plante et al., 1999), it was slightly lower than some found in studies of those in recovery from drugs or alcohol (Pardini, 2000; Plante et al., 1999). Our finding highlights that among this sample, strength of religious faith is not incompatible with substance use.

This study did not find an association between strength of religious faith and lower levels of depression or trauma symptoms. Rather, we found that women with a high strength of religious faith had higher odds of self-reporting a mental health diagnosis, and were more likely to be psychologically dependent upon methamphetamine. This begs the question of why strength of religious faith is not associated with mental well being among this sample of women who use methamphetamine. Since the SCSORFQ is inversely associated with God control, it is likely that this group of women feel that they are able to enlist God in achieving their hoped for outcomes. However, with suggestions that individuals who perceive God to be punitive are more likely to develop a substance use problem (Gorsuch, 1995), and that religiosity may exacerbate the effects of certain life stressors for those with a vengeful or wrathful view of God (Pargament, 2002), work needs to be done with women who use methamphetamine to explore their perceptions of God. Although this group of women has a high strength of religious faith, with increased reports of mental illness and a heightened dependency on methamphetamine, we cannot assume that their religiosity is serving them well. This suggests that before a program uses religious language, an exploration with individuals about the role religiosity plays in their lives and their view of God may offer important insights into how to assist individuals in using their religiosity as a helpful resource.

In multivariate analysis, strength of religious faith was not associated with most of the drug use variables. Women with a high strength of religious faith were neither more nor less likely to inject drugs, to use methamphetamine in specific ways (mode and frequency), or to use certain drugs (except for marijuana). This does not reflect findings from other studies where drug users were more or less likely to use certain drugs depending on their religiosity. Among the women who injected drugs, we found an inverse association between strength of religious faith and the injection risk behavior of lending their used needles for another person’s use. This was a somewhat unanticipated finding as we would have assumed that since there was no association between religiosity and types of drugs used and the mode of use (injection vs. non-injection), that there would not be an association with injection risk variables. However, all the other variables measure individual level behaviors, but injection risk measures group behaviors. Furthermore, this was the injection risk behavior that potentially puts someone else at risk, as opposed to the women using someone else’s used needle which would put them at risk. It is possible that this finding is driven by gender, and that women are less likely to engage in injection behaviors that put someone else at risk. It is equally plausible that those women with a high strength of religious faith are motivated to protect others. Future work will want to examine what it is about having a high strength of religious faith that is associated with women who inject drugs being less likely to share their used needles with someone else.

Our prediction that there would be a positive association between a high strength of religious faith and past, current and future involvement in drug treatment programs was largely unsupported. Although we found that women with a high strength of religious faith are more likely to report current involvement in 12-step meetings, we did not find an association with any of the other drug treatment variables. The association with 12-step meetings is not surprising, since programs like Alcoholics Anonymous and Narcotics Anonymous are low-threshold and based upon a belief in a higher power. For women whose lives are often characterized by instability and uncertainty, and who have a high strength of religious faith, the 12-step model allows them to access meetings at their convenience. That, in addition to the use of a language of a higher power, may make these programs particularly attractive to them.

Examining the other drug treatment variables, we were surprised to find that they were not associated with strength of religious faith. A further look at the methadone maintenance variable however, reveals a plausible explanation. In bivariate analysis, we found an inverse association between strength of religious faith and heroin use. When put in the multivariate model, we discovered that African-American and older women were less likely to report heroin use. Once we controlled for those two variables, the association did not remain. If the two groups of women with high strengths of religious faith are less likely to have used heroin, then it is not surprising that there was not an association between religious faith and involvement in a methadone maintenance program. This explanation, however, does not carry over to experiences with outpatient treatment or an interest in participating in an outpatient program specifically for methamphetamine use. For these two variables future work that examines the potential role that religious faith plays in women’s decisions to enter more structured types of treatment programs may offer useful insight that can be used to make these types of programs more enticing to women who use methamphetamine and have a high strength of religious faith.

Study Limitations

There are several potential limitations that need to be considered when interpreting the data from this study. Measures of strength of religious faith, like that of the SCSORFQ, are limited in that they only measure levels of intrinsic religiosity. They do not capture extrinsic religiosity, views of God, or individuals’ experiences with religiosity over time. Potentially, it is the combination of these aspects of religiosity that produce favorable health outcomes, as opposed to simply having a high strength of religious faith. Using only the SCSORFQ, as opposed to incorporating other measures of religiosity, prevented us from assessing whether this group of women may be experiencing a spiritual struggle between their virtues and behaviors, and the ways in which that may be associated with their drug use behaviors and mental health outcomes. We relied on self-reported data, which is potentially subject to biases associated with misclassification due to poor recall or social desirability. Since our study was cross-sectional we cannot infer causality from our statistical associations. Lastly, our sample was comprised primarily of urban poor women who use methamphetamine. Our findings may not be representative of other groups of women who use methamphetamine.

CONCLUSION

Findings from our study indicate that among women who use methamphetamine, especially African-American and older women, there is a high strength of religious faith. The role of service providers is to draw upon all the personal and structural resources that their clients have, and for this group of women their strength of religious faith may be one such important resource. Therefore, interventions and treatment programs that target women who use methamphetamine will want to consider the role religiosity can play in service delivery. Our finding that women with a high strength of religious faith were more likely to be psychologically dependent upon meth suggests that programs working with women who use methamphetamine need to consider the ways in which their faith is both a positive and negative force in their lives. By working with women who use methamphetamine to determine how religiosity influences their behaviors, service providers may be able to help women draw from their religious faith in a way that enhances their well being.