Assignment: Research Design and Sampling - SOCW 6301 Week 7

profileballs269
Week5approvedtopic.pdf
Home>Social Science homework help>Sociology homework help>Assignment: Research Design and Sampling - SOCW 6301 Week 7

Substance Use & Misuse, 50:653–663, 2015 Copyright C© 2015 Foundations Recovery Network ISSN: 1082-6084 print / 1532-2491 online DOI: 10.3109/10826084.2014.997828

ORIGINAL ARTICLE

Gender Differences in Treatment Retention Among Individuals with Co-Occurring Substance Abuse and Mental Health Disorders

Sam Choi1, Susie M. Adams2, Siobhan A. Morse3 and Sam MacMaster1

1School of Social Work, University of Tennessee—Knoxville, Nashville, Tennessee, USA; 2Vanderbilt University, Nashville, Tennessee, USA; 3Research and Fidelity, Foundations Recovery Network, Brentwood, California, USA

Background: A significant number of individuals with co-occurring substance abuse and mental health dis- orders do not engage, stay, and/or complete residen- tial treatment. Although prior research indicates that women and men differ in their substance abuse treat- ment experiences, our knowledge of individuals with co-occurring substance abuse and mental health disor- ders as well as those attending private residential treat- ment is limited. Objectives: The purpose of this study is to examine gender differences on treatment reten- tion for individuals with co-occurring substance abuse and mental health disorders who participate in private residential treatment. Methods: The participants were 1,317 individuals (539 women and 778 men) with co- occurring substance abuse and mental health disorders receiving treatment at three private residential treat- ment centers. Bivariate analyses, life tables, and Cox regression (survival analyses) were utilized to exam- ine gender effects on treatment retention, and iden- tify factors that predict treatment retention for men and women. Results: This study found that women with co-occurring disorders were more likely to stay longer in treatment when compared to men. The find- ings indicate the factors influencing length of stay dif- fer for each gender, and include: type of substance used prior to admission; Addiction Severity Index Compos- ite scores; and Readiness to Change/URICA scores. Age at admission was a factor for men only. Conclu- sions/Importance: These findings can be incorporated to develop and initiate program interventions to min- imize early attrition and increase overall retention in private residential treatment for individuals with co- occurring substance use and mental health disorders.

Keywords gender differences, co-occurring disorders, dual diagnosis, substance abuse and mental health disorder, retention, predictors, residential treatment

Address correspondence to Siobhan A Morse, Foundations Recovery Network, Research and Fidelity, 5409 Maryland Way, Brentwood, CA 37027 USA; E-mail: [email protected]

Length of stay (LOS) in substance abuse treatment is a strong predictor of treatment outcomes with longer lengths of stay in treatment associated with lower post- treatment substance use rates (DeLeon & Schwartz, 1984; Greenfield et al. 2003; Simpson, Joe, & Rowan-Szal, 1997). Longer periods of treatment engagement are also associated with lower readmission rates (Moos & Moos, 1995). Although the significance of remaining in treat- ment is well established, leaving treatment prior to com- pletion or against clinical advice remains a treatment con- cern (Ball, Carroll, Canning-Ball, & Rounsaville, 2006) and is associated with poor treatment outcomes (Deane, Wootton, Hsu, & Kelly, 2012).

Treatment retention is a widely used proxy for treat- ment outcomes such as substance use relapse, recidivism to crime, and sustained recovery. Factors related to both treatment program characteristics and individual patient characteristics have been investigated for their impact on retention in treatment. Program factors such as therapeu- tic alliance, the use of motivational interviewing, gender- specific programming, gender-specific interventions, pay- ment source, and inclusion of treatment for co-morbid mental health issues have all been studied for their impact on retention.

Therapeutic alliance is strongly associated with re- maining in treatment (Marsh, Angell, Andrews, & Curry, 2012; Simpson, Joe, Rowan-Szal, & Greener, 1997) and treatment attendance (Mullins, Suarez, Ondersma, & Page, 2004) but appears to impact outcomes somewhat less significantly than retention or attendance (Simpson, Joe, & Rowan-Szal, 1997). Motivation for treatment is considered a robust predictor of retention (Adamson, Sellman, & Frampton, 2009). The impact of using mo- tivational interviewing or enhancement on retention has mixed findings with positive results for the early stages of treatment retention (Carroll et al., 2006). However, Mullins and colleagues (2004) found that the use of

653

654 S. CHOI ET AL.

motivational interviewing did not demonstrate significant improvement over education counseling. Gender-specific programming has been found to be a significant factor in both treatment retention and outcomes for women (Ashley, Marsden, & Brady, 2003; Greenfield et al., 2008; Greenfield, Cummings, Kuper, Wigderson, & Koro-Ljungberg, 2013). Further studies have found that attention to issues, such as employment, trauma, and mental health in a gender-specific manner resulted in positive retention and treatment outcomes (Adams et al., 2011; Brady & Ashley, 2005; Grella, 2008; Green, Polen, Dickinson, Lynch, & Bennett, 2002; Greenfield, Back, Lawson, & Brady, 2010). Payment source has also been associated with LOS in treatment with private insurance patients having shorter lengths of stay (Brady & Ashley, 2005). The recognition of the co-occurring nature of psychiatric disorders and substance use disorder and the need to simultaneously treat both has resulted in the development of specialty treatment programs based on an integrated model (SAMHSA, DASIS Report, 2003), positively impacting both outcomes and retention.

Personal and substance use characteristics have been found to be inconsistent predictors of treatment reten- tion and outcomes (Hiller, Knight, Leukefeld, & Simpson, 2002; Hser, Joshi, Maglione, Chou, & Anglin, 2001; Joe, Simpson, & Broome, 1998, 1999). Age is the only socio- demographic characteristic that consistently predicts re- tention in substance abuse treatment regardless of gen- der, with older age associated with longer lengths of stay (Adams et al., 2011; Hall, Prendergast, Wellisch, Pat- ten, & Cao, 2004: Pelissier, Motivans, & Rounds-Bryant, 2005). Primary drug of choice has been investigated as a predictor of retention in 90-day residential treatment with alcohol use associated with longer retention when compared to other substance use (Deane, et al., 2012). Drug Abuse Treatment Outcomes Survey data analysis re- vealed that programs serving patients with greater psycho- logical severity and higher cocaine and alcohol use had shorter lengths of stay (Simpson, Joe, & Brown, 1997). Patients with a higher motivation at intake are also more likely to remain in treatment than those with lower moti- vation at intake (Simpson et al., 1997). A history of trauma (Tull, Gratz, Coffey, Weiss, & McDermott, 2013; Logan, Walker, Jordan, & Leukefeld, 2006), stress response (Tull et al., 2013), multiple life stressors (Kelly, Blacksin, & Mason, 2001; Comfort, Sockloff, Loverro, & Kaltenbach, 2003), and self-efficacy (Cummings, Gallop, & Green- field, 2010) have also been found to impact retention.

A study of insured outpatient treatment attendees re- vealed that having fewer and less severe drug problems improved retention and that there were gender differences in the factors that impacted retention (Mertens & Weis- ner, 2000). In women, higher retention was associated with being married, having a higher income, not being African American, having lower psychiatric severity, and being unemployed; and in men, age, employer involve- ment, and abstinence goals were predictors of higher re- tention (Mertens & Weisner, 2000). By contrast, a study of 268 patients at a publicly funded outpatient center, patient

substance use did not predict retention; however, being male, Caucasian and having higher severity in employ- ment composite score as measured by the ASI were as- sociated with longer retention and greater attendance in outpatient treatment (McCaul, Svikis, & Moore, 2001).

Overall, women appear to be less likely to use sub- stance use treatment services (Wu, Ringwalt, & William, 2003; Kim et al. 2011), and are among those at most risk for not accessing mental health treatment when needed (Roll, Kennedy, Tran, & Howell, 2013). Although, in gen- eral, women use substances for shorter periods of time prior to entering treatment than men, they appear to enter treatment with greater severity of issues (Greenfield et al., 2010; Piazza, Vrbka, & Yeager, 1989; Hernandez-Avila, Rounsaville, & Kranzler, 2004; Arfken, Klein, di Menza, & Schuster, 2001). Some research has shown significant differences in retention by gender (Arfkin et al., 2001; Kim et al., 2011; Choi, Adams, MacMaster, & Seiters, 2013) and others have stipulated that while gender itself has not been predictive of retention, issues traditionally associated with gender, such as child-care, employment and trauma, are related to variations found in retention by gender (Mertens & Weisner, 2000; Green et al., 2002; Tull et al., 2013; Greenfield et al., 2007). Despite having more severe and complex problems, studies found that incar- cerated women with substance abuse histories were less likely to relapse than their male counterparts (Fiorentine, Anglin, Gil-Rivas, & Taylor, 1997; Gil-Rivas, Fiorentine, & Anglin, 1996; Grella, Stein, & Greenwell, 2005). This “gender paradox” in has been explained by the hypothesis that women engage more readily and actively in counsel- ing and other treatment services (Fiorentine et al., 1997; Kim et al., 2011).

In studies focusing on factors impacting the retention of women in treatment, higher income and education and lower psychiatric severity are often predictors of higher retention and completion rates (Mertens & Weisner, 2000; Kelly et al., 2001). The impact of psychiatric severity on employability and thus the level of distress and anx- iety on substance abusing women has been suggested (Hernandez-Avila et al., 2004) which could lead to shorter stays in treatment for women. Women also appear to ben- efit most from single-gendered treatment groups (Cum- mings et al., 2010; Greenfield et al., 2013; Greenfield et al., 2008; Green et al., 2002).

In summary, research indicates that there are signif- icant differences by gender in the factors that influence treatment retention. Although women have less access to services overall, once they enter treatment; they do so with more serious substance dependencies, and with more health and social problems than do men (Kim et al., 2011; Wu et al., 2003). Women are also more likely than men to use, and benefit from, services available in comprehen- sive programs (Greenfield et al., 2013; Green et al., 2002). Through prior investigation of this dataset, Choi and col- leagues (2013) found that in private residential treatment, age, gender, types of drug used, ASI medical and psy- chiatric severity scores and URICA readiness to change scores predicted treatment retention at 30 days of their

GENDER EFFECTS ON TREATMENT RETENTION 655

initial treatment. The prior study used bivariate analysis to identify independent variables significantly correlated with 30-day treatment retention, then used logistic regres- sion to determine predictors of treatment outcomes (Choi et al., 2013). Based on the prior study (Choi et al., 2013) and what we know from the available literature on the im- pact of gender on treatment retention, specifically Merten and Weisner (2000), this study was designed to further examine the effect of gender on treatment retention us- ing survival analyses, and to identify factors that predict treatment retention in each gender while enrolled specifi- cally in private, residential treatment for individuals with co-occurring mental health and substance use disorders.

METHODS

Setting Data were collected at three private residential facilities that provide integrated substance abuse and mental health treatment services in Memphis, Tennessee, Malibu, Cali- fornia, and Palm Springs, California. Foundations Recov- ery Network (FRN), a private for-profit substance abuse treatment provider offering residential and outpatient sub- stance abuse treatment services, operates all three pro- grams. Service recipients’ at all three facilities are drawn from across the United States and Canada. Treatment ser- vices are based on an integrated model of mental health and substance abuse services consisting of both individual and group evidence-based interventions (Foundations Re- covery Network, 2010). In most cases, the expected LOS is between 28 and 40 days. Recommended length of time in treatment is individualized on the basis of clinical as- sessment and medical necessity; however, other factors may contribute to the actual LOS.

Participants All program participants who enter residential services are offered an opportunity to participate in an ongoing evalua- tion during the initial phase of treatment. A trained intake person located at each facility describes the evaluation, re- views and obtains informed consent, and collects the loca- tor information for post discharge interviews. The Addic- tion Severity Index (ASI) (McClellan, 1983; McClellan et al., 1992) from the initial clinical assessment is used as the baseline ASI assessment if informed consent is ob- tained. Masters’ level clinicians complete the initial clini- cal assessment within the first 4 days following admission to FRN’s residential programs. Data are collected at three additional time points: 1, 6, and 12 months post discharge. A community-based Institutional Review Board reviewed study protocols to assure the protection of Human Subjects.

Data for this study are drawn only from the baseline in- terview and LOS in treatment days. The participants were 1,317 individuals who voluntarily sought residential treat- ment at one of the three treatment centers. All participants received an intake assessment by a multidisciplinary team which provides the basis for an individual treatment plan to address substance use, psychiatric disorder, and medi-

cal and social service needs. At this time, the evaluation data was collected for those individuals who were will- ing to consent to participating in the evaluation process. Compared to all patients who attended treatment during the same time period, the population who agreed to par- ticipate in the research project was not appreciably differ- ent in terms of demographic characteristics. The average age for the overall population was 36.6 versus 36.0 for the study population. The overall population was 60.8% male and the study population was 59.1%. Caucasians represented the largest percent by far in both the general population and the study group. The average LOS was 29.90 for the overall population compared to 32.08 for our sample. Co-occurring disorders were assessed over the course of treatment starting with initial screening, assess- ment, and psychiatric evaluation. A master’s level clini- cian conducted a complete psychiatric evaluation with one of the programs psychiatrists within 72 hours after arriv- ing to the residential treatment facility. Each patient is as- signed to one of the program’s licensed clinicians who utilize the information gathered through initial screen- ing and assessment to develop the initial treatment plan with the patient during an initial individual session with the first week of treatment. Ongoing psychiatric and in- dividual therapy sessions are utilized along with weekly treatment team meetings to update each patient’s treat- ment plan, including updates/confirmation of specific co- occurring substance use and psychiatric disorders. This process provides input from a multidisciplinary team of clinicians in order to thoroughly assess co-occurring dis- orders throughout treatment as symptoms may change or become clearer during the course of treatment. Retention was measured in treatment days through a retrospective review of discharge records, which also included the base- line interview data.

Instruments Addiction Severity: The scalable questions that make up the composite scores of the Addiction Severity Index (ASI) (McClellan et al., 1992) were utilized to measure addiction severity. The ASI was developed to measure problem severity in each of seven potential problem areas that include: medical, employment, alcohol, drug, legal, family/social, and psychiatric problems. In order to ensure that each question within a given problem area is given the same weight in calculation of the composite score each item in a subscale is divided by its maximum value and by the total number of questions in a composite. This scoring yields a score from 0 to 1 in each composite.

Readiness for Change: The University of Rhode Island Change Assessment (URICA) (DiClemente & Hughes, 1990) is a measure of readiness to change that has been studied with a range of different populations. The instru- ment consists of 32 statements that subjects endorse on a 5-point scale from strongly agree to strongly disagree. The URICA yields scores on each of four scales; Precon- templation, Contemplation, Action, and Maintenance, or each of the stages of change described by Prochaska, Di- Clemente, & Norcross (1992). In addition, the scores from

656 S. CHOI ET AL.

these scales are used to create a Readiness to Change com- posite score. The Readiness to Change score was derived for this study in the same manner used in Project MATCH Research Group (1997). The average Contemplation, Ac- tion, and Maintenance scores were added and the Precon- templation score was subtracted from the sum. The Readi- ness to Change score was used as a predictor variable in subsequent analysis.

Treatment Retention: This study focuses on treatment retention. Retention status was observed and calculated by days between a program start date and a discharge date. Data were collected on all admissions between February 1, 2008 and through July 31, 2010 with final observation of discharge data on August 31, 2010.

Data Analysis Initial analyses consisted of basic descriptive statistics and bivariate analyses to identify and examine any gen- der difference on any pre-treatment demographic and use- related factors as well as components of treatment reten- tion. Next, a life table was developed to investigate the trajectory of treatment retention by gender. Finally, Cox regression was employed to investigate the impact of var- ious predictors of treatment retention by gender. One of the essential features of Cox regression is that the tech- nique allows for the unbiased analysis of time to event data controlling for covariates. The event of interest in the current study is a discharge from treatment. Like logis- tic regression, the exponential of the coefficients from the Cox model gives the relative risk of the odds for the co- variate. Cox regression also proves superior to ordinary least squares regression (OLS), in that the Cox regression algorithm allows for censoring of persons who discontin- ued or did not experience the event (treatment retention in the current study) during the study period. In this study, Cox regression was developed in three steps. Model 1 con- tains demographic characteristics, types of substance use disorders, types of mental health disorders, and treatment location. Model 2 also includes the various scalable ASI subscale measures. Finally, in Model 3, the score of readi- ness to change was added in the final model.

RESULTS

Sample Description Demographic, substance use, and mental health disorder and treatment characteristics are provided in Table 1. The mean age in this study sample was 36 years (SD = 12.1) with 40.9% of the sample being female. The majority of study participants (90.2%) were Caucasian, 8.1% were African American, and 1.7% were Latino. Nearly 56% were employed in last 30 days. Approximately 67% had alcohol-related disorders (alcohol abuse or dependence), 18.8% had opioid related disorders (opioids abuse or dependence), and 18.1% had cocaine-related disorders (cocaine abuse or dependence). In terms of identified mental health disorders, the majority of study partici- pants (82.9%) had a diagnosis of an anxiety disorder, followed by major depression (74.9%), and mood dis-

order (26.6%). The majority of participants (80%) were diagnosed with more than one mental health disorder. Of the 1,317 study participants, 43.7% stayed in treatment for at least 30 days. The average LOS in treatment for the study sample was 32.08 days (SD = 19.29).

Bivariate Analysis Bivariate analyses were conducted to determine the relationship between various independent variables and gender using chi-square and t tests. The results of bivariate analyses are displayed in the second and third column of Table 1. Statistically significant differences were found in participants’ age, employment in last 30 days, and race/ethnicity. Men in this study were younger than women. Men also had a higher rate of employment in the last 30 days (60.0%) compared to women (46.6%). In terms of race and ethnicity, women were more predominately Caucasians (94.1%) compared to men (87.5%). Women stayed in treatment longer than men by an average of approximately four and a half days. In addition, women had a higher rate of treatment retention (47.7%) at 30 days compared to men (40.9%).

Statistically significant gender differences were found in six of the seven domains of the ASI composite score measurement. Women had higher mean ASI composite scores in the areas of medical, employment/support, fam- ily/social relationships, and psychiatric issues indicating greater severity in these areas than men. Men had higher mean ASI scores in areas of drug, and legal measures than women. ASI composite score for alcohol severity were slightly higher in men. Men entered treatment with higher URICA scores in the precontemplative stage. Women were less likely to be in the precontemplative stage and therefore more likely to be in a stage reflecting greater readiness for change; however, overall URICA and other URICA subscale scores did not reflect significant differ- ences between men and women. Gender differences were also found in type of substance use and mental health dis- order. Men had higher rates of cocaine and cannabis use disorders as compared to women. Women had higher rates of major depression, anxiety disorders, mood disorders, and eating disorders compared to men.

Life Table The life table was developed to investigate the trajectory of treatment retention by gender. Figure 1 illustrates the survival lines indicating time to discharge for both men and women. Observation of Figure 1 revealed that the sur- vival lines are similar for the first 20 days of treatment. These lines split into different trajectories after approxi- mately 20 days which continues to broaden as time pro- gresses. The largest gap appears from 30 days forward. At 30 days, approximately 30% of men remained in treat- ment compared to 40% of women. Comparison of the sur- vival lines was performed using the Wilcoxon (Gehan) statistic (13.415, df = 1. p < .000). The result further high- lights the statistical significance of the differences in the trajectories of these lines.

GENDER EFFECTS ON TREATMENT RETENTION 657

TABLE 1. Sample description

Total Sample N = 1,317 Men N = 778 (59.1%) Women N = 539 (40.9%) Mean (SD) Mean (SD) Mean (SD)

Age∗∗ 36 (12.11) 36.08 (11.94) 38.26 (12.26) ASI: Medical∗∗∗ .26 (.36) .21 (.33) .34 (.38) ASI: Employment/Support∗∗ .40 (.27) .38 (.28) .43 (.26) ASI: Alcohol .38 (.34) .40 (.33) .38 (.35) ASI: Drug∗ .17 (.16) .18 (.16) .16 (.16) ASI: Legal∗ .11 (.21) .12 (.21) .10 (.20) ASI: Family/Social

Relationships∗∗∗ .30 (.26) .28 (.25) .34 (.26)

ASI: Psychiatric∗∗∗ .49 (.20) .45 (.21) .55 (.18) Readiness for Change 10.77 (1.56) 10.66 (1.61) 10.94 (1.47) Precontemplation∗∗∗ 1.67 (.52) 1.73 (.54) 1.57 (.46) Contemplation 4.41 (.44) 4.38 (.46) 4.45 (.41) Action 4.27 (.47) 4.25 (.45) 4.30 (.49) Maintenance 3.75 (.64) 3.75 (.63) 3.76 (.64) Days in Treatment∗∗∗ 32.08 (19.29) 30.17 (16.4) 34.85 (22.5)

N (%) N (%) N (%)

Treatment Retention at 30 days∗∗ 575 (43.7) 318 (40.9) 257 (47.7) Race/Ethnicity∗∗∗

African American 107 (8.1) 83 (10.7) 24 (4.5) Caucasian 1,188 (90.2) 681 (87.5) 507 (94.1) Latino 22 (1.7) 14 (1.8) 8 (1.5) Employment in last 30 days-Yes∗∗∗

733 (55.7) 467 (60.0) 250 (46.4)

No 584 (44.3) 311 (40.0) 289 (53.6) Type of Substance Use Disorders

Alcohol 885 (67.2) 515 (66.2) 370 (68.6) Cocaine∗∗∗ 238 (18.1) 170 (21.9) 68 (12.6) Cannabis∗ 88 (6.7) 60 (7.7) 28 (5.2) Opioid 248 (18.8) 147 (18.9) 101 (18.7) Poly Substance 145 (11.0) 87 (11.2) 58 (10.8) Others 29 (2.2) 4 (1.3) 6 (2.3)

Type of Mental Health Disorders Major Depression∗∗∗ 986 (74.9) 540 (69.4) 446 (82.7) Anxiety Disorder∗∗∗ 1090 (82.8) 617 (79.3) 473 (87.8) Mood Disorder∗∗ 350 (26.6) 186 (23.9) 164 (30.4) Bi-Polar Disorder 16 (1.2) 11 (1.4) 5 (0.9) Eating Disorder∗ 9 (0.7) 2 (0.3) 7 (1.3) ADHD 10 (0.8) 5 (0.6) 5 (1.9) Dementia 19 (1.4) 13 (1.7) 6 (1.1) Missing∗∗∗ 89 (6.8) 70 (9.0) 17 (3.5)

Locations A 291 (22.1) 176 (22.6) 115 (21.3) B 64 (4.9) 29 (3.7) 35 (6.5) C 962 (73.0) 573 (73.7) 389 (72.7)

∗p < .05. ∗∗p < .01. ∗∗∗p < .00.

Cox Regression Table 2 provides the results of models constructed to as- sess relative effects on the likelihood of retention for in- dividuals with co-occurring substance abuse and mental disorders. The results for men suggest that age, an ADHD diagnosis, location, and ASI employment subscale com- posite score were associated with treatment retention. In interpreting these results, it is important to note that a hazard ratio greater than 1 indicates a higher likelihood

of treatment retention. Age was significantly and posi- tively related to retention. The Exp (b) of 1.009 indicates that for each increase in age by a year, the likelihood of treatment retention increased by 0.9%. Men diagnosed with ADHD were 41% more likely to stay in the treat- ment longer than men diagnosed with mood disorder. The ASI employment score was significantly and negatively associated with the likelihood of retention for men. Men with higher ASI employment scores, reflecting greater

658 S. CHOI ET AL.

FIGURE 1. Life Table—Treatment Retention by Gender.

severity of employment issues, were less likely to stay in treatment. There were significant differences in male re- tention rates predicted by treatment location. Readiness to change was not predictive of treatment retention for men.

The Cox regression model for women suggests that co- caine use, depression, location, ASI alcohol subscale com- posite score, and readiness to change were significantly correlated with treatment retention. Women who were co- caine dependent were approximately 41% less likely to re- main in treatment compared to women who were alcohol dependent. However, having a higher ASI drug subscale composite score (indicating greater severity) was not pre- dictive of retention; but greater severity on the ASI alco- hol subscale composite was predictive of treatment reten- tion for women. The likelihood of remaining in treatment improved by 56% for every point increase on the ASI al- cohol composite subscale. In addition, women diagnosed with depression were 92% more likely to remain in treat- ment longer than women diagnosed with a mood disor- der. Location was also predictive of the decision to remain in treatment in women. Women who were scored in the precontemplative or contemplative stages on the URICA readiness to change scale were significantly less likely to remain in treatment than those who scored in the action or maintenance stages of change.

DISCUSSION

The purpose of the current study was to (1) examine gen- der effects on treatment retention, and (2) identify fac- tors that predict treatment retention for men and women in private residential treatment for individuals with co- occurring substance abuse and mental health disorders. This study identifies significant differences by gender in treatment retention for individuals with co-occurring substance abuse and mental health disorders. The find-

ings indicate that women are more likely to stay in treat- ment compared to men. Men with co-occurring sub- stance abuse and mental health disorders in this study had more difficulty staying in treatment than women. For example, men stayed an average of 30.17 days in treat- ment, 4.5 days fewer than women (34.85 days). Similarly, women were significantly more likely to remain in treat- ment for 30 days. Typically, more intensive treatment has been associated with lower retention for women. For ex- ample, Arfken and colleagues (2001), in their study of publicly funded residential and intensive outpatient treat- ment in a major metropolitan area, found that women had lower retention and completion rates than men. In con- trast, in a study of substance abuse care provided in pri- mary care settings women remained in treatment on av- erage longer than men (Kim et al., 2011). However, both studies, as well as much of the research evaluating treat- ment retention have been conducted in publicly funded programs. This research provides an important evidence that funding source may be an over-riding factor in patient decisions to remain in private, residential treatment.

The study also finds that different factors appear to con- tribute to the likelihood of remaining in treatment for each gender. Women appear to enter treatment with greater severity in the areas of medical, employment/support, family/social relationships, and psychiatric measures ev- idenced by higher mean ASI composite scores. This re- flects findings in studies by Greenfield and colleagues (2010) and others noted in the literature review.

Consistent with the literature, age is associated with treatment retention in men with older men more likely to remain in treatment longer (Adams et al., 2011; Hall, Prendergast, Wellisch, Patten, & Cao, 2004: Pelissier, Mo- tivans, & Rounds-Bryant, 2005); however, this was not the case with women. Age was not found to be predictive of treatment retention in females.

Drug use characteristics impacted treatment retention differently for males and females. Similar to the Drug Abuse Treatment Outcomes Survey data analysis results, cocaine use in women was associated with shorter stays in treatment (Simpson, Joe, & Brown, 1997) when compared to alcohol as the control group. In contrast to Simpson, Joe and Brown (1997), women whose ASI alcohol subscale composite score indicated greater severity than their coun- terparts were more likely to stay in treatment. This was similar to results found by Deane and colleagues (2012) who found that women who used alcohol were likely to re- main longer in treatment than those who used other drugs.

Women with depression were also more likely to re- main in treatment longer than their counterparts with mood disorders. Men diagnosed with ADHD were signif- icantly more likely to remain in treatment for longer com- pared to those diagnosed with a mood disorder. Mertens & Weisner, 2000 found that lower psychiatric severity was related to improved retention; however we did not find that psychiatric severity measured by the ASI psychiatric sub- scale was influential in retention for either gender.

Similar to findings in the literature (Grella, Stein, & Greenwell, 2005) women did enter treatment with greater

GENDER EFFECTS ON TREATMENT RETENTION 659

TABLE 2. Cox regression models for treatment retention

Model for Men Model for Women

Variables B SE Exp (B) p-Value 95% CI B SE Exp (B) P-Value 95% CI

Age .009 .005 1.009 .048 (1.000–1.018) −.003 .005 .997 .575 (.986–1.008) Caucasian1 −.271 .147 .762 .065 (.571–1.017) −.262 .275 .760 .339 (.449–1.318) Employed in last

30days–yes −.029 .127 .971 .819 (.756–1.247) −.220 .148 .803 .137 (.600–1.073)

Opiate Abuse/Dependence2

.195 .142 1.216 .168 (.921–1.605) .107 .143 1.113 .452 (.842–1.472)

Cocaine Abuse/Dependence2

−.155 .133 .857 .243 (.660–1.111) −.541 .169 .582 .001 (.418–.811)

Cannabis Abuse/Dependence2

.043 .127 1.044 .735 (.814–1.339) .136 .163 1.146 .403 (.833–1.575)

Poly Substance Abuse/Dependence2

.020 .148 1.020 .892 (.763–1.365) −.255 .166 .775 .123 (.560–1.072)

Depression3 .025 .144 1.025 .862 (.773–1.360) .656 .207 1.928 .002 (1.285–2.892) Anxiety Disorder3 −.149 .167 .861 .371 (.621–1.195) −.006 .217 .994 .977 (.649–1.521) Bipolar3 −.132 .621 .876 .829 (.264–2.907) −.594 1.206 .552 .622 (.052–5.565) Eating Disorder3 −1.385 1.031 .250 .179 (.033–1.889) −.011 .436 .989 .980 (.421–2.324) ADHD3 1.528 .594 4.611 .010 (1.438–14.782) .267 .527 1.306 .612 (.465–3.666) Dimentia3 .282 .332 1.326 .396 (.691–2.544) −.121 .531 .886 .819 (.313–2.508) Missing Mental

Disorders–Yes .048 .246 1.049 .847 (.647–1.699) .138 .560 1.148 .805 (.383–3.443)

Location A4 −.313 286 .731 .272 (.418–1.279 −1.182 .286 .307 .000 (.175–.537) Location B4 .320 .123 1.377 .009 (1.083–1.750) .049 .151 1.051 .744 (.781–1.413) ASI: Medical −.085 .149 .918 .566 (.686–1.229) −.052 .157 .950 .742 (.698–1.292) ASI: Employment −.549 .218 .577 .012 (.376–.886) −.244 .288 .784 .397 (.446–1.378) ASI: Alcohol −.020 .172 .817 .240 (.584–1.144) .446 .191 1.562 .019 (1.075–2.269) ASI: Drug −.077 .544 .926 .888 (.319–2.689) 1.061 .570 2.890 .062 (.946–8.830) ASI: Legal −.188 .244 .828 .441 (.513–1.337) −.468 .287 .626 .103 (.356–1.100) ASI: Family/Support −.011 .208 .989 .958 (.658–1.487) −.236 .233 .790 .311 (.500–1.247) ASI: Psychiatric −.463 .351 .629 .187 (.316–1.253) −1.612 .570 2.890 .062 (.946–8.830) Readiness to Change Precontemplation .181 .121 1.199 .133 (.946–1.519) −.393 .181 .675 .029 (.474–.962) Contemplation −.153 .170 .858 .366 (.615–1.196) −.549 .241 .577 .023 (.360–.926) Action .040 .142 1.041 .778 (.788–1.374) .144 .158 1.155 .362 (.847–1.575) Maintenance .053 .089 1.055 .549 (.886–1.255) −.195 .102 .823 .054 (.674–1.004) -2 Log Likelihood 4868.477 3326.112 χ 2, df, p-value 58.795,27,000∗∗∗ 82.221, 27, .000∗∗∗

1African American and Latino were the reference group. 2Alcohol abuse/dependences and others were the reference group. 3Mood disorders was the reference group. 4Location C was the reference group.

severity of 4 of the 7 subscale composite scores in the ASI. However, despite entering treatment with less severity in ASI employment composite score than females, employ- ment issues were a factor in predicting retention for males but not females.

Females in the early stages of readiness were less likely to remain in treatment; however, readiness for change as measured by the URICA was not a factor in predicting re- tention for men. This is in contrast to the results of Adam- son, Sellman, & Frampton (2009) whom results report that motivation is of the most robust predictors of retention.

This study also found differences in retention for men by location. Further investigation is required to determine the factors causing these differences; however,

an interesting point is raised. Although those three different facilities have similar therapeutic philosophies and operation systems, the findings of this study indicate the needs of recognizing the uniqueness of each program. According to Simpson, Joe and Brown (1997), each treatment facility differ in staff skills, resources, service intensity, environmental setting, and client demands which may impact their retention and effectiveness. Ac- cordingly, treatment evaluation studies must recognize the multi-level factors—client level and program level—on treatment retention (Simpson et al., 1997; Simpson, Joe, & Brown, 1997). Organizational program planners may want to consider a variety of factors that can influence retention differences across sites, including state and

660 S. CHOI ET AL.

local legislative requirements, payer mix, population dif- ferences, and other factors across both the organizations and the patients that could prove to be significant.

These findings show patterns in treatment retention that are different from prior studies. There may be several reasons as to why these findings are different. Further research is needed to investigate the impact of funding source; e.g., private versus public treatment, on retention patterns in both genders. The data for this study was drawn only from private treatment which serves as both a limita- tion to the findings and a strength of the article due to the paucity of research drawn from this source.

These results further demonstrate what was found in the earlier study by Choi and colleagues (2013) using bivariate analysis and logistic regression to determine predictors of retention. These researchers found that gender was a significant predictor of treatment retention at 30 days in private, residential, dual diagnosis treat- ment. This study adds to the current body of literature investigating factors impacting retention in treatment and differences by gender in dual diagnosis treatment and identifies several key predictors which may be addressed through programming interventions. It is also important to note that historically individuals with co-occurring substance abuse and mental health disorders have low retention rates. Prior research indicates that longer lengths of stay in treatment generally predicts better treatment outcomes at follow-up (Simpson et al., 1997). Accord- ingly, this finding highlights the importance of early engagement efforts for both men and women developing gender-specific strategies to improve treatment retention for individuals with co-occurring substance abuse and mental health disorders.

The current study makes a unique contribution to the literature for individuals with substance abuse and men- tal disorders, despite some limitations. First, the results of this study may be unique to private residential treat- ment programs. The predictors of treatment retention in publicly funded residential treatment programs may be different. In addition, this study assumes that three treat- ment programs operate identically, although there is some variability in staff skills, resources, service intensity, envi- ronmental setting, and client demands which may impact their effectiveness (Simpson, Joe, & Brown, 1997). This study was limited to client-level variables and did not ex- amine the impact of program level variables on treatment retention. Although the findings and implications are im- portant, one important limitation may exist in the sample of research participants. While individuals who partici- pated in the research component appear to be demograph- ically similar to all other treatment participants, there was a slight difference in the overall LOS between individuals who participated in the research component and those did not.

In summary, the current study investigated gender- specific factors that may explain treatment retention vari- ations for individuals with co-occurring substance abuse and mental health disorders in private residential treat-

ment settings. The findings suggest the importance of fur- ther research investigating the factors predicting retention in privately funded treatment.

Declaration of Interest

Siobhan A. Morse is employed by Foundations Recovery Network, the operator of the sites supplying the data. She received compensation in the form of salary as Director of Research. Samuel A. MacMaster was under contract with Foundations Recovery Network, the operator of the sites supplying the data. He received compensation from Foundations Recovery Network. The authors alone are re- sponsible for the content and writing of the article.

THE AUTHORS Siobhan A. Morse, MHSA, CRC, CAI, MAC is the Director of Research and Fidelity at Foundations Recovery Network. Her current research interests focus on providing high-quality care and outcomes research in private residential dual diagnosis treatment.

Samuel MacMaster, Ph.D. is an Associate Professor at the University of Tennessee within the College of Social Work. Dr. MacMaster’s research interests center on the intersection of substance use and HIV/AIDS; and have focused specifically on the development of culturally appropriate interventions to overcome barriers to service access for underserved and incarcerated populations.

Sam Choi, Ph.D is a director at Tennessee Korean American Social Service Center and research fellow at Children and Family Research Center. Dr. Choi’s research revolves around two main areas: the relations of service delivery to child welfare and treatment outcomes for parents with co-occurring problems and the relations of service delivery to treatment outcomes for individuals with

co-occurring substance abuse and mental health problems.

GENDER EFFECTS ON TREATMENT RETENTION 661

Susan M. Adams, PhD, RN, is Faculty Scholar for Community Engaged Behavioral Health at Vanderbilt University School of Nursing in Nashville, TN. Dr. Adams’ current research concerns are efficacy of trauma- informed interventions and sustained recovery for women with co-occurring substance use and mental health disorders in community based programs.

GLOSSARY

Co-occurring Disorders [COD] (previously called Dual Diagnosis): refers to individuals who have one or more disorders relating to the use of alcohol and/or other sub- stances of abuse as well as one or more mental health disorders. The diagnosis of co-occurring disorders is used when at least one disorder of each type occurs in- dependent of the other and is not a cluster of symp- toms resulting from one disorder alone. COD replaces the term Dual Diagnosis which can be confusing since it has been used to identify other co-morbid disorders such as a primary medical disorder and a mental health disorder.

Cox regression (or proportional hazards regression): is a method for investigating the effect of several variables upon the time a specified event takes to happen (such as in treatment). The method does not assume a specific “survival model,” although it is not truly nonparametric because it does make the assumptions that the effects of the predictor variables on survival are constant over time and that they are additive in one scale.

Life table: is a statistical calculation of survival analysis that deals with “time to an event” such as death, relapse, time in treatment, or other health events. It can answer the question of the chance of survival after diagnosis or entry to treatment. It can address the variable of entry and withdrawal from treatment. The life table generates a survival curve.

Predictors, sometimes called independent variables, are factors or variables that can be used to “predict” or forecast the value of another variable, called the depen- dent or outcome variable, based on observations and measurements. Within the addictions field, predictor variables can include characteristics of an individual or population (such as age, gender, education, severity of disorder, involvement in criminal justice system, readi- ness to change, motivation for treatment, etc.), char- acteristics of the treatment environment, theoretical approach to treatment, models of service delivery, char- acteristics of the therapist/counselor and therapeutic alliance.

Private residential treatment: is a 24 hour/7 days a week treatment program for co-occurring substance abuse and mental health disorders provided in a residential

setting for extended time periods (up to 6–12 months) beyond an acute detoxification or psychiatric inpatient hospitalization stay. Residential treatment may be pub- licly funded (Medicaid/Medicare, state/federal block grants, or nonprofit agencies without fees) or privately funded (private insurance or direct out-of-pocket pay- ment).

Treatment retention: refers to the quantity or amount of time in treatment. Most commonly treatment reten- tions refer to the length of stay in treatment measured by days, months, or specific time period. Historically, longer treatment retention is a consistent predictor of better post-treatment outcomes.

REFERENCES

Adams, S. M., Peden, A. R., Hall, L. A., Rayens, M. K., Staten, R., & Leukefeld, C. G. (2011). Predictors of retention of women offenders in a community-based residential substance abuse treatment program. Journal of Addictions Nursing, 23(3), 103–116.

Adamson, S. J., Sellman, J. D., & Frampton, C. M. A. (2009). Pa- tient predictors of alcohol treatment outcome: A systematic re- view. Journal of Substance Abuse Treatment, 36(1), 75–86.

Arfken, C., L., Klein, C., di Menza, S., & Schuster, C. (2001). Gen- der differences in problem severity at assessment and treatment retention. Journal of Substance Abuse Treatment, 20, 53–57.

Ashley, O. S., Marsden, M. E. and Brady, T. M. (2003). Effec- tiveness of substance abuse treatment programming for women: A review. The American Journal of Drug and Alcohol Abuse, 29(1), 19–53.

Ball, S. A., Carroll, K. M., Canning-Ball, M., & Rounsaville, B. J. (2006). Reasons for dropout from drug abuse treatment: Symp- toms, personality, and motivation. Addictive Behaviors, 31(2), 320–330. doi:http://dx.doi.org/10.1016/j.addbeh.2005.05.013

Brady, K. T. & Ashley, O. S. (Eds). (2005). Women in substance abuse treatment: Results from the alcohol and drug services study (ADSS) (DHHS Publication No. SMA 04-3968, Analytic Series A-26). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Carroll, K. M., Ball, S. A., Nich, C., Martino, S., Frankforter, T. L., Farentinos, C., . . . Woody, G. E. (2006). Motivational interview- ing to improve treatment engagement and outcome in individu- als seeking treatment for substance abuse: A multisite effective- ness study. Drug and Alcohol Dependence, 81, 301–312.

Choi, S., Adams, S., MacMaster, S. A., & Seiters, J. (2013). Pre- dictors of residential treatment retention among individuals with co-occurring substance abuse and mental health disorders. Jour- nal of Psychoactive Drugs, 45, 122–131.

Comfort, M., Sockloff, A., Loverro, J., & Kaltenbach, K. (2003). Multiple predictors of substance-abusing women’s treat- ment and life outcomes. A prospective longitudinal study. Addictive Behaviors, 28(2), 199–224. doi:http://dx.doi.org/ 10.1016/S0306-4603(01)00227-1

Cummings, A. M., Gallop, R. J., & Greenfield, S. F. (2010). Self-efficacy and substance use outcomes for women in single-gender versus mixed-gender group treatment. Journal of Groups in Addiction & Recovery, 5(1), 4–16. doi:http://dx.doi.org/10.1080/15560350903543915

Deane, F. P., Wootton, D. J., Hsu, C., & Kelly, P. J. (2012). Predicting dropout in the first 3 months of 12-step residential drug and alcohol treatment in an australian sample. Journal of

662 S. CHOI ET AL.

Studies on Alcohol and Drugs, 73(2), 216–225. Retrieved from http://search.proquest.com/docview/993100437?accountid= 14766

DeLeon, G. & Schwartz, S. (1984). Therapeutic communities: What are the retention rates? American Journal of Drug and alcohol Abuse, 10, 267–284.

DiClemente, C. C. & Hughes, S. O. (1990). Stages of change pro- files in outpatient alcoholism treatment. Journal of Substance Abuse, 2, 217–235.

Foundations Recovery Network. (2010). Baseline intake and 30-day treatment retention de-identified data between February 1, 2008 through August 31, 2010 following community IRB review. Available at: http://www.foundationsrecoverynetwork.com/

Fiorentine, R., Anglin, M. D., Gil-Rivas, V., & Taylor, E. (1997). Drug treatment: Explaining the gender paradox. Substance Use and Misuse, 32, 653–678.

Gil-Rivas, V., Fiorentine, R., & Anglin, M. D. (1996). Sexual abuse, physical abuse, and post-traumatic stress syndrome among women in outpatient drug treatment. Journal of Psychoactive Drugs, 28, 95–102.

Green, C. A., Polen, M. R., Dickinson, D. M., Lynch, F. L., & Bennett, M. D. (2002). Gender differences in predictors of ini- tiation, retention, and completion in an HMO-based substance abuse treatment program. Journal of Substance Abuse Treat- ment, 23(4), 285–295. doi:http://dx.doi.org/10.1016/S0740- 5472(02)00278-7

Greenfield, L., Burgdorf, K., Chen, X., Porowski, A., Roberts, T., & Herrell, J. (2003). Effectiveness of long-term residential sub- stance abuse treatment for women: Findings from three na- tional studies. American Journal of Drug and Alcohol Abuse, 30, 537–550.

Greenfield, S. F., Potter, J. S., Lincoln, M. F., Popuch, R. E., Kuper, L., & Gallop, R. J. (2008). High psychiatric symptom severity is a moderator of substance abuse treatment outcomes among women in single vs. mixed gender group treatment. The American Journal of Drug and Alcohol Abuse, 34(5), 594–602. doi:http://dx.doi.org/10.1080/00952990802304980

Greenfield, S. F., Brooks, A. J., Gordon, S. M., Green, C. A., Kropp, F., McHugh, R. K., Lincoln, M., Hien, D. & Miele, G. M. (2007). Substance abuse treatment entry, retention and outcomes in women: A review of the literature. Drug and Alcohol Depen- dence, 86, 1–21.

Greenfield, S. F., Back, S. E., Lawson, K., & Brady, K. T. (2010). Substance abuse in women. Psychiatric Clin- ics of North America, 33(2), 339–355. doi:http://dx.doi.org/ 10.1016/j.psc.2010.01.004

Greenfield, S. F., Cummings, A. M., Kuper, L. E., Wigderson, S. B., & Koro-Ljungberg, M. (2013). A qualitative analysis of women’s experiences in single-gender versus mixed-gender sub- stance abuse group therapy. Substance use & Misuse, 48(9), 772–782. doi:http://dx.doi.org/10.3109/10826084.2013.787100

‘Grella, C. E. (2008). From generic to gender-responsive treatment: Changes in social policies, treatment services, and outcomes of women in substance abuse treatment. Journal of Psychoactive Drugs, Sarc Suppl 5, 327–343. doi:http://dx.doi.org/10.1080/02791072.2008.10400661

Grella, C. E., Stein, J. A., & Greenwell, L. (2005). Associations among childhood trauma, adolescent problem behaviors, and adverse adult outcomes in substance-abusing women offenders. Psychology of Addictive Behaviors, 19, 43–53.

Hall, E. A., Prendergast, M. L., Wellisch, J., Patten, M., & Cao, Y. (2004). Treating drug-abusing women prisoners: An outcomes evaluation of the Forever Free program. The Prison Journal, 84(1), 81–105.

Hernandez-Avila, C., Rounsaville, B. J., & Kranzler, H. R. (2004). Opioid-, cannabis- and alcohol-dependent women show more rapid progression to substance abuse treat- ment. Drug and Alcohol Dependence, 74(3), 265–272. doi:http://dx.doi.org/10.1016/j.drugalcdep.2004.02.001

Hiller, M. L., Knight, K., Leukefeld, C., & Simpson, D. D. (2002). Motivation as a predictor of therapeutic engagement in man- dated residential substance abuse treatment. Criminal Justice and Behavior, 29, 56–75.

Hser, Y. I., Joshi, V., Maglione, M., Chou, C. P., & Anglin, M. D. (2001). Effects of program and patient characteristics on reten- tion of drug treatment patients. Evaluation Program Planning, 24, 331–341.

Joe, G. W., Simpson, D. D., & Broome, K. M. (1998). Effects of readiness for drug abuse treatment on client retention and as- sessment of process. Addiction, 93, 177–1190.

Joe, G. W., Simpson, D. D., & Broome, K. M. (1999). Retention and patient engagement models for different treatment modalities in DATOS. Drug and Alcohol Dependence, 57, 113–125.

Kelly, P. J., Blacksin, B., & Mason, E. (2001). Factors affecting substance abuse treatment completion for women. Issues in Mental Health Nursing, 22(3), 287–304. doi:http://dx.doi.org/10.1080/01612840152053110

Kim, T. W., Saitz, R., Cheng, D. M., Winter, M. R., Witas, J., & Samet, J. H. (2011). Initiation and engagement in chronic disease management care for substance depen- dence. Drug and Alcohol Dependence, 115(1–2), 80–86, http://dx.doi.org/10.1016/j.drugalcdep.2010.10.013

Logan, T. K., Walker, R., Jordan, C. E., & Leukefeld, C. G. (2006). Victimization manifestations and vulnerability factors. (pp. 17–49) Washington, DC: American Psychological Associ- ation. doi:http://dx.doi.org/10.1037/11364-002

Marsh, J. C, Angell, B., Andrews, C.A., & Curry, A. (2012). Client- provider relationship and treatment outcome: A systematic re- view of substance abuse, child welfare, and mental health ser- vices research. Journal of the Society for Social Work and Re- search, 3(4):233–267. DOI:DOI:10.5243/jsswr.2012.15

McCaul, M. E., Svikis, D. S., & Moore, R. D. (2001). Predictors of outpatient treatment retention: Patient versus substance use characteristics. Drug and Alcohol Dependence, 62, 9–17.

McClellan, A. T. (1983). Patient characteristics associated with out- come. In J. R. Cooper, F. Altman, B. S. Brown & D Czechowicz (Eds.). Research in the treatment of narcotic addiction. NIDA treatment research monograph series. Rockville, MD: National Institute of Drug Abuse.

McClellan, A. T., Kushner, H., Metzger, D., Peters, R., Smith, I., Grissom, G., & Argeriou, M. (1992). The fifth edition of the Ad- diction Severity Index. Journal of Substance Abuse Treatment, 9, 199–213.

Mertens, J. R., & Weisner, C. M. (2000). Predictors of sub- stance abuse treatment retention among women and men in an HMO. Alcoholism: Clinical and Experimental Research, 24(10), 1525–1533. Doi: 10.1111/j.1530-0277.2000tb04571.x

Mullins, S. M., Suarez, M., Ondersma, S. J., & Page, M. C. (2004). The impact of motivational interviewing on substance abuse treatment retention: A randomized control trial of women in- volved with child welfare. Journal of Substance Abuse Treat- ment, 27, 51–58.

Moos, R. H., & Moos, B. S. (1995). Stay in residential facili- ties and mental health care as predictors of readmission for pa- tients with substance use disorders. Psychiatric Services, 46(1): 66–72.

Pelissier, B., Motivans, M., & Rounds-Bryant, J. L. (2005). Sub- stance abuse treatment outcomes: A multi-site study of male and

GENDER EFFECTS ON TREATMENT RETENTION 663

female prison programs. Journal of Offender Rehabilitation, 41, 57–80.

Piazza, N. J., Vrbka, J. L., & Yeager, R. D. (1989). Telescoping of alcoholism in women alcoholics. International Journal of the Addictions, 24(1), 19–28. Retrieved November 20, 2013 from http://search.proquest.com/docview/617696002?accountid= 14766

Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to addictive behav- iors. American Psychologist, 47, 1102–1114.

Project MATCH Research Group (1997). Matching alcoholism treatment to client heterogeneity: Project MATCH post treat- ment drinking outcomes. Journal of Studies on Alcohol, 58(1), 7–29.

Roll, J. M., Kennedy, J., Tran, M., & Howell, D. (2013). Dis- parities in unmet need for mental health services in the United States, 1997-2010. Psychiatric Services, 64(1), 80–82. doi:http://dx.doi.org/10.1176/appi.ps.201200071

SAMHSA, Drug and Alcohol Services Information Sys- tem [DASIS]. 2003. Admissions of persons with co- occurring disorders: 2000. Retrieved May 31, 2011 from http://www.oas.samhsa.gov/2k3/dualTX/dualTX.pdf/

Simpson, D. D., Joe, G. W., Broome, K. M., Hiller, M. L., Knight, K., Rowan-Szal, G. A. (1997). Program diversity and treatment

retention rates in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11(4), 279–293. doi: 10.1037/0893-164X.11.4.279

Simpson, D. D., Joe, G. W., & Brown, B. S. (1997). Treatment re- tention and follow-up outcomes in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11(4), 294-307. doi: 10.1037/0893-164X.11.4.294

Simpson, D. D., Joe, G. W., & Rowan-Szal, R. G. (1997). Drug abuse treatment retention and process effects on follow-up out- comes. Drug and Alcohol Dependence, 47(3), 227–235. doi: 10.1016/S0376-8716(97)00099-9

Simpson, D. D., Joe, G. W., Rowan-Szal, G. A., & Greener, J. M. (1997). Drug abuse treatment process components that im- prove retention. Journal of Substance Abuse Treatment, 14(6): 565–72.

Tull, M. T., Gratz, K. L., Coffey, S. F., Weiss, N. H., & McDermott, M. J. (2013). Examining the interactive effect of posttraumatic stress disorder, distress tolerance, and gen- der on residential substance use disorder treatment reten- tion. Psychology of Addictive Behaviors, 27(3), 763–773. doi:http://dx.doi.org/10.1037/a0029911

Wu, L. T., Ringwalt, C. L., & William C. E. (2003). Use of sub- stance abuse treatment services by persons with mental health and substance use problems. Psychiatric Services, 54, 363–369.

Notice of Correction: The order of authors and their corresponding affiliations has changed since the article’s original online publication date of January 14, 2015.

Copyright of Substance Use & Misuse is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.