Single-Gender Treatment of Substance Abuse: Effect on Treatment Retention and Completion

Article excerpt

Increasing emphasis has been placed on the development of gender-specific treatment programs to address the unique needs of women substance abusers. Some authors have suggested that this task can be accomplished by simply providing single-gender programs; yet others have argued that simply segregating substance abuse clients by gender has no effect on outcomes. The latter have suggested that in addition to providing women-only programs, different treatment approaches must be adopted. The purpose of the study described in this article was to investigate the effect of one agency's change from mixed-gender to single-gender treatment on client retention and treatment completion. Data were collected on 305 men and 102 women who were treated in either mixed-gender or single-gender settings. Results indicate that substance abuse treatment provided in single-gender settings does not significantly increase treatment retention and completion. These findings lend support to the contention that to improve treatment outcomes for women substance abusers, gender-specific treatment must do more than provide traditional treatment in a single-gender environment.

Key words: gender; retention; substance abuse treatment

**********

Although women have struggled with chemical dependency for centuries, society has generally viewed substance abuse as a male problem (Blumenthal, 1998; Goldberg, 1995). Not until feminist theory and the women's liberation movement questioned male-dominated societal norms did addicted women begin to receive attention in their own right (Kandall, 1998). Because women have been underrepresented in research studies and treatment groups, the effects of and treatment for drug abuse among women are less understood than for men (Mondanaro, 1989; Wilke, 1994). Recognition of this fact has led to an increasing emphasis on the development of treatment programs that address the unique needs of women substance abusers. Despite a number of articles describing gender-specific treatment programs (Burman, 1992; Coletti et al., 1995; Finkelstein, 1993; Finkelstein, Kennedy, Thomas, & Kearns, 1997; LaFave & Echols, 1999; Schliebner, 1994; Wald, Harvey, & Hibbard, 1995; Yaffe, Jenson, & Howard, 1995; Zankowski, 1987), few studies have examined the outcomes of these programs (Beckman, 1994; Hodgins, el-Guebaly, & Addington, 1997; Wallen, 1998).

LITERATURE REVIEW

It has become accepted generally that the patterns, consequences, and reasons for substance abuse are different among women and men (Lex, 1994). Studies comparing men and women in chemical dependency treatment have found that women report more psychiatric symptoms than men, more depression and anxiety, lower self-esteem, and higher rates of childhood sexual abuse (Wallen, 1992). Also, chemically dependent women differ from men in patterns, psychosocial characteristics, and physiological consequences of drug use (Nelson-Zlupko, Kauffman, & Dore, 1995). Rates of treatment entry, retention, and completion are significantly lower for women than for men, and current substance abuse treatment models are less effective for women than for men (Beckman & Amaro, 1984; Blume, 1990; Reed, 1985; Stevens, Arbiter, & Glider, 1989). Because of these differences, it is not surprising that treatment strategies designed for men fail to adequately address the needs of female substance abusers (Schliebner, 1994), thus preventing treatment programs from attracting and retaining female substance abusers (Wells & Jackson, 1992).

Gender-Specific Treatment

Blumenthal (1998) noted that to increase retention of women in substance abuse treatment, treatment approaches must address gender differences in the etiology and treatment of addictive disorders and respond to the unique needs of women. Gender-specific programming gives women an opportunity to concentrate on their needs and desires away from their traditional concerns of social approval and the welfare of others (Copeland, Hall, Didcott, & Biggs, 1993). According to Copeland and colleagues, the basic requirements of specialized treatment for women are a female therapist, availability of individual counseling, and women-only groups. Other components may include sexual and physical abuse counseling, child care services, family counseling, and vocational training and job-seeking support (Burman, 1992). Although these services can be provided in a mixed-gender setting (Beckman, 1994), George (1990) recommended that women have treatment programs separate from men. Women-only programs allow women to discuss issues that they will not discuss in mixed-gender groups and display a wider range of behaviors, and they may prevent the experience of sexual harassment (Hodgins et al., 1997). Also, women-only programs are more likely to provide services to meet the specific needs of women (Pendergast, Wellisch, & Falkin, 1995) and may be more attractive and effective (Grella, Polinsky, Hser, & Perry, 1999).

However, some authors have argued that to be effective, gender-specific treatment must do more than segregate female clients and employ only female staff. For example, treatment for women should be more supportive and less confrontational, grounded in women's experiences, and focus on empowerment and women's strengths (Finkelstein et al., 1997; Grella et al., 1999; Nelson-Zlupko et al., 1995). Despite the cogency of these arguments, there is a paucity of supporting empirical data. Only one study (Copeland et al., 1993) has provided empirical support for the hypothesis that providing gender-specific treatment without substantially altering the treatment approach fails to improve outcomes for female clients. As noted by Grella and colleagues, it is still not clear whether gender-specific programs must adopt treatment approaches that are different from those used in traditional mixed-gender programs to improve treatment outcomes, or whether providing services in a women-only environment is sufficient to do so.

Studies on Gender-Specific Treatment

Despite Wilke's (1994) call for research comparing mixed-gender and single-gender treatment groups, few such studies have appeared in the literature. Several evaluations of gender-specific programs have been published (Smith, 1985), but a review of the literature found only three studies involving control (Dahlgreen & Willander, 1989) or comparison groups (Copeland et al., 1993; Dodge & Potocky-Tripodi, 2001). In the first study, Dahlgreen and Willander randomly assigned 200 women in the early phases of alcohol problems to a specialized outpatient women's program or to one of two traditional mixed-gender treatment settings. A two-year follow-up revealed decreased alcohol consumption and improved social adjustment in both groups, with the women in the specialized female unit demonstrating greater improvement. However, there was a significant correlation between length of treatment and outcome in both groups. Because the women in the experimental group were in treatment longer, the apparently better outcomes of this group may have been a result of differences in treatment intensity rather than the gender-specific programming (Hodgins et al., 1997).

In the second study, Copeland and colleagues (1993) compared changes in alcohol and other drug-associated problems among 80 women from a specialist women's service and 80 women from two traditional mixed-gender treatment services. Six months after treatment there were no significant differences in any measure of outcome between the groups. The authors concluded that simply providing women with an all-female environment without changing the program content did not substantially improve treatment outcome.

In the third study, Dodge and Potocky-Tripodi (2001) compared dropout rates and changes in levels of self esteem, social support, depression, and severity of addiction among 89 women treated in three inpatient programs: gender-specific, mixed-gender, and a combination mixed-gender/gender-specific treatment. Results indicated that none of the three groups demonstrated improvement on any of the outcomes examined or in dropout rates.

A limitation of each of the three studies was that independent treatment providers served each group, introducing the possibility that differences among the groups in treatment approaches or philosophy may have contributed to the failure of each study to find significant differences between mixed-gender and gender-specific groups. For example, in the Copeland et al. (1993) study, the three groups included a women-only residential program based on a feminist model; a mixed-gender, residential program based on a medical model; and a mixed-gender detoxification program. Furthermore, the treatment groups differed in the recommended length of stay, exposure to 12-step groups, and the amount, if any, of individual counseling, health care, and group topics such as parenting and self-esteem.

The present study sought to address this limitation by a retrospective comparison of retention and completion outcomes between single- and mixed-gender residential treatment in one agency that changed from mixed- to single-gender programming. Because the agency did not alter its treatment approach or philosophy with the change to single-gender treatment, the discrepancies in treatment approaches noted in the previous studies were minimized. However, it should be noted that the longitudinal design introduced the possibility that issues such as staff turnover and program drift may have resulted in undocumented differences between the groups.

Importance of Treatment Retention in Substance Abuse Treatment

Treatment retention is one of the most important intermediate outcome measures of successful substance abuse treatment (Chou, Hser, & Anglin, 1998). Treatments for substance abuse are considered effective to the extent that they demonstrate the ability to retain clients (Carroll, 1997). Simpson and Sells (1982) demonstrated that better retention is associated with reductions in substance abuse. More recently, Messina, Wish, and Nemes (2000) found that both women and men who completed treatment demonstrated reductions in drug use and arrest and had increased employment. A literature review by Condelli and Hubbard (1994) found that the longer clients were in therapeutic communities, the less likely they were to use drugs or commit predatory crimes and the more likely they were to be employed at a one-year follow-up. Furthermore, Higgins and Budney's (1997) analysis of four published studies revealed that treatment duration of three months or more predicted improved outcome proportionate to the length of time spent in treatment. In sum, length of time in drug treatment is related to positive treatment outcome and failure to complete treatment is associated with drug use relapse (Jones, Haug, Stitzer, & Svikis, 2000). Therefore, the evaluation of methods to increase substance abuse treatment retention and completion is a legitimate aim of social work research.

Using a retrospective, quasi-experimental cohort design (Cook & Campbell, 1979), the study described in this article examined whether the provision of single-gender substance abuse treatment, without a significant change in treatment philosophy, would increase rates of treatment retention and completion. Three primary research questions guided the study: (1) Does single-gender treatment have an effect on the rate of completion of an intensive day program treatment phase? (2) Does single-gender treatment have an effect on the rate of completion of ninety days of treatment? (3) Does single-gender treatment have an effect on client length of stay? Because the implications of single-gender groups for men have not been studied, we explored the research questions for men as well as women. Follow-up data on outcomes such as drug and alcohol use were not available.

METHOD

Program Description

The programs studied were in a private nonprofit agency that provides comprehensive substance abuse services to address the emotional, physical, recreational, employment, spiritual, and social needs of the client, with the goal of long-term abstinence from drugs and alcohol. The agency serves a population that is primarily low-income or homeless in a large city in the southeastern United States.

After admission to the program, clients entered an intensive day treatment phase lasting 28 days, during which clients attended five, hour-long groups daily, Monday through Friday. Group topics included group therapy, spirituality, meditation, stress and anger management, psychoeducation, relationship issues, social skills, family issues, addiction issues, 12-step principles, and employment issues. Residents also attended evening activities such as group therapy, community meetings, a spirituality group, and a "Big Book" group and had individual sessions with their assigned case manager at least biweekly. Clients were required to document attendance at up to six 12-step meetings each week and obtain a 12-step sponsor. Evening activities and involvement in a 12-step network, including meetings, were required for the duration of treatment. After completing the intensive day treatment phase, clients were required to obtain and maintain full-time employment. An employment specialist worked with clients to conduct a job search, prepare a resume, and develop interview skills. A treatment goal for all clients was completion of a minimum of 90 days of treatment, but they were encouraged to remain in residential treatment for one year. In January 1998 the agency announced a change from mixed-gender to single-gender programming to increase client retention. There was no evidence to suggest that factors associated with this change affected staff effectiveness.

Mixed-Gender Group (MG). Men and women were housed in a single facility with 35 beds for men and 12 beds for women. Men and women were integrated in all group services and activities, which were facilitated by both male and female staff members. Although assignment to case managers was based on client needs and known strengths of particular staff members, gender congruence between client and case manager was rarely given priority.

Single-Gender Group (SG). With the introduction of single-gender programming, the men and women were housed in separate facilities (35 beds for men and 12 beds for women), forming two programs: the Women's Residential Program (WRP) and the Men's Residential Program (MRP). In the WRP, female staff members facilitated all group activities, and all case managers were women. In the MRP, however, such gender congruence between client and staff was not possible because there were not enough male staff members. The treatment structure used for the mixed-gender group remained relatively unchanged for the single-gender group, except that the specific content of groups changed depending on the gender of the clients. For instance, a group to address relationship issues was conducted with both MG and SG groups. However, the MG group had focused primarily on men's issues, because the group often was composed of more men than women. The women in the MG group focused on relationship issues of importance to women in their weekly women-only meeting. The primary differences between the MG and SG groups were the segregation of clients by gender and the introduction of women-only staff in the women's group, rather than any clinically significant differences in treatment structure or process.

Data Collection

I collected data retrospectively from agency records. Agency policy requires that each chart be audited following discharge for accuracy and completion of records. Review of audit records indicated that this policy was consistently complied with, and agency records were judged to be sufficiently complete and accurate. In fact, there were no missing values associated with any variables examined in this study. Demographic information was collected regarding client age, ethnicity, education, marital status, drug of choice, polysubstance abuse, and number of previous substance abuse treatment episodes.

Completion of the day treatment phase was chosen as a dependent variable, because studies consistently report that most attrition occurs early, with the majority of dropouts usually occurring in the first month of treatment (Baekeland & Lundwall, 1975; Carroll, 1997; DeLeon, 1991; Silberfeld & Glaser, 1978; Swett & Noones, 1989). Completion of 90 days of treatment was included as a second dependent variable, because research has demonstrated that treatment duration of three months or longer predicts improved outcome (Higgins & Budney, 1997). Both day program and 90-day completion were recorded as dichotomous variables. Length of stay (LOS) was chosen as the third dependent variable, because a number of studies have reported that longer treatment duration predicts improved outcome, with the degree of improvement proportional to the length of time spent in treatment (Higgins & Budney). LOS was calculated as the total number of days in treatment. Because of rule violations or relapse, some clients were suspended temporarily from treatment for seven or 14 days. Such suspensions were considered to be part of the therapeutic process, and thus were included in the calculation of total LOS up to a maximum of 21 days. In the event that a client was suspended but did not return, the date of suspension was considered to be the final day of treatment.

Sampling Procedure

All clients discharged during the 18 months before announcement of the change to single-gender programming (July 1, 1996-December 31, 1997) were eligible for inclusion in the MG group (n = 174). Clients discharged during the month between announcement and establishment of single-gender programming were excluded to minimize the possibility that they chose to end treatment in response to the announcement. All clients admitted during the 18 months following the implementation of single-gender programming (March 1, 1998 through August 31, 1999) were eligible for inclusion in the SG group (n = 230).

During the first four months of single-gender programming, the women occupied vacant space in a state psychiatric hospital. In July 1998 the women were relocated permanently to a renovated two-story house in a residential neighborhood. The temporary and institutional nature of the first setting and the disruption caused by moving to a new facility had the potential to affect retention negatively. However, analysis failed to find significant differences on dependent variables between clients who were admitted to the temporary facility and those who were admitted to the permanent site, because the clients admitted to the temporary facility were retained for the analysis. Any clients discharged for major medical or psychiatric reasons (n = 7), or for arrest resulting from past offenses (n = 1) were excluded because they were precluded from return to the program.

Data Analysis

Demographic Variables. Preliminary analyses were conducted to examine differences among the four treatment groups on demographic variables. One-way analyses of variance (ANOVAs) were conducted on continuous variables (age, education, and prior treatments), and chi-square tests of independence were conducted on categorical variables (ethnicity, marital status, drug of choice, and polysubstance abuse). Because seven separate analyses were conducted, increasing the probability of a Type I error, Bonferroni's procedure was used and resulted in an operational alpha level of .007 (.05/ 7). In conducting chi-square analyses, numerous expected cell counts less than 5 will negatively affect the validity of results (George & Mallery, 1999). One approach to this problem is to redefine the response categories such that two or more of the original categories can be collapsed together (Huck & Cormier, 1996). Consequently, ethnicity was converted to two categories (African American and other), and drug of choice was converted to two categories (cocaine and other).

Dependent Variables. Because of the nominal level of the data, chi-square tests of independence were used to determine if any differences existed by treatment setting (mixed-gender or single-gender) and gender on the dependent variables of day program completion and 90-day completion. To determine differences in LOS, a 2 (treatment setting) x 2 (gender) analysis of covariance (ANCOVA) was planned initially with the demographic variables found to vary across groups considered for inclusion as covariates (years of education, prior treatments, cocaine addiction, and polysubstance abuse). However, when the assumptions for ANCOVA were tested with each potential covariate, the assumptions of linearity and homogeneity of regression slopes were untenable (Huck & Cormier, 1996). Thus, it was decided that ANCOVA would be inappropriate, and a 2 (treatment setting) x 2 (gender) ANOVA would be used instead. Levene's test for equality of variances indicated that homoscedasticity could be assumed, but Kolmogorov-Smirnov tests indicated that the data distribution was not normal. However, ANOVA is very robust to violation of assumptions, especially the normality assumption (Howell, 1992). In general, if the populations can be assumed to be similar in shape and not severely skewed (that is, [+ or -] 2.0) then the analysis is likely to be valid (George & Mallery, 1999; Howell). Because the skewness ranged from 1.16 to 1.31, it was decided that analysis could proceed. Because the use of three separate analyses (two chi-squares, one ANOVA) increases the Type I error rate, Bonferroni's procedure was used to maintain the experiment-wise alpha level at .05, resulting in an operational alpha level of .017.

RESULTS

Sample Characteristics

The ethnicity of the study sample (N = 404), was primarily African American (64 percent) or white (34 percent), with the remainder (2 percent) identifying as Asian, (n = 5), Hispanic (n = 4), or biracial (n = 1). The large majority of participants were separated, divorced, or widowed (44 percent) or single (42 percent), with only 14 percent married at the time of treatment. Ages ranged from 18 to 67, with a median age of 37.4. Participants had completed an average of 13.6 years of education, with a range between six and 27 years. The majority of participants identified cocaine (72 percent), alcohol (15 percent), or opiates (9 percent) as their drug of choice, with the remainder (4 percent) identifying marijuana (n = 5), amphetamines (n = 4), or sedatives (n = 6). Three-quarters (75 percent) of participants reported abuse of more than one substance. The mean number of prior treatments for substance abuse was 2.1 (SD = 2.2).

No significant differences were found between groups on age, ethnicity, and marital status. A significant difference was found in years of education [F(3, 391) = 5.18, p [less than or equal to] .005] (Table 1). Post-hoc comparisons using the Tukey HSD test revealed that SG men had completed more years of education than MG women (p [less than or equal to] .01) and SG women (p [less than or equal to] .05). Similarly, a significant difference was found in the number of previous substance abuse treatment episodes [F(3, 397) = 9.24, p [less than or equal to] .001]. A Tukey HSD post-hoc multiple comparison indicated that MG men had significantly fewer prior treatment attempts than SG men (p [less than or equal to] .001) and SG women (p [less than or equal to] .01). A significant difference in the proportion of cocaine addicts was noted across groups [[chi square] (3, N = 404) = 17.50, p < .001]. Post-hoc pair-wise comparisons indicated that MG men were less likely to indicate cocaine as their drug of choice than SG men (p [less than or equal to] .001). Comparison of polysubstance abuse also revealed a significant difference across groups [[chi square] (3, N= 403) = 16.83, p [less than or equal to] .001]. Follow-up using pair-wise comparisons revealed that MG men were more likely to report polysubstance abuse than S G women (p [less than or equal to] .001).

Treatment Retention and Completion Outcomes

Among the total sample, 80.9 percent completed the intensive day program, and 41.5 percent completed 90 days of treatment (Table 2). However, no significant differences were found in the rate of day program completion [[chi square] (3, N = 404) = 1.80, p = .616] or in the rate of 90-day completion [[chi square] (3, N = 404) = .535, p = .911]. These results indicate that regardless of gender, people treated in the mixed-gender setting were as likely to complete the intensive day program and the initial 90 days of treatment as people treated in the single-gender setting. Among the total sample, the mean LOS was 109.8 (SD = 103.9). Among women the mean LOS was 115.0 (SD = 112.5) and among men, the mean LOS was 108.1 (SD = 101.1). Results of the two-way analysis of variance failed to reveal a significant effect for treatment

setting [F(1,400) =. 109, p = .741] or gender [F(1,400) = .332, p = .565] and failed to reveal a significant interaction effect [F(1,400) = .032, p = .859]. Thus, regardless of treatment setting or client gender, no differences were found on the total time spent in treatment.

DISCUSSION

Consistent with the only other study addressing this issue (Copeland et al., 1993), the findings of this study suggest that providing a women-only environment for the treatment of chemically dependent women does not increase treatment retention and completion. The results further suggest that providing a men-only setting neither increases nor decreases treatment retention and completion. The latter finding is in contrast to Hodgins et al.'s (1997) suggestion that greater retention of male clients would be found in mixed-gender groups compared with single-gender groups. This is the first study to examine empirically the effect of single-gender programming on male substance abusers. These findings lend support to the argument that to retain female substance abusers, gender-specific treatment must go beyond simply providing a women-only environment by adopting treatment approaches different from those used in traditional mixed-gender programs (Grella et al., 1999; Nelson-Zlupko et al., 1995). Furthermore, a program's change from mixed-gender to single-gender programming is unlikely to affect negatively retention of male clients.

A notable aspect of this study is how the sample differs from that of earlier studies. First, nearly two-thirds of the sample in this study identified as African American. Dodge and Potocky-Tripodi's (2001) sample was primarily white, and neither Copeland et al. (1993) nor Dahlgreen and Willander (1989) reported on ethnicity. Given the countries in which the latter two studies were undertaken (Australia and Sweden, respectively), it is reasonable to assume that these samples were largely white as well. Second, nearly three-quarters of this study's sample identified cocaine as their drug of choice. In each of the previously cited studies, the samples primarily identified alcohol as their drug of choice, and only Dodge and Potocky-Tripodi reported on any cocaine users.

However, the quasi-experimental, cohort design of this study requires that results be viewed with caution. A previously mentioned limitation of this design is the passage of time between the two cohorts (mixed-gender and single-gender treatment groups), which introduced the possibility that some unmeasured differences existed in the treatment program variables between cohorts. For example, researchers have suggested that program characteristics such as size, staff, organizational structure, staff turnover, and major program changes may have an effect on retention (Lewis & Ross, 1994).

Another limitation of the study, lack of randomization, increases the possibility that the cohort groups were not equivalent on important variables. Indeed, there were several differences among groups on the demographic variables: prior substance abuse treatments, cocaine addiction, polysubstance abuse, and education. Although studies have not found education or drug of choice to be related to retention (Lewis & Ross, 1994; Roberts & Nishimoto, 1996), a history of previous treatments and polysubstance abuse are positively related to retention (Gainey, Wells, Hawkins, & Catalano, 1993; Lewis & Ross). Unfortunately, because of assumption violations, these variables could not be controlled through the use of ANCOVA. However, it must also be noted that the differences found on previous treatments and polysubstance abuse were not found between MG women and SG women, reducing the probability that demographic differences explain the failure to find differences on the outcome variables.

This study is further limited in that it examines the effect of only one agency's change from mixed-gender to single-gender programming. The agency under study may not be representative of other agencies providing similar services, limiting the generalizability of findings. For instance, it should be noted that an alternative explanation for failure to find a difference in treatment retention and completion between MG women and SG women is the possibility that the agency was sufficiently responsive to female clients even in a mixed-gender setting. Some credence to this explanation is provided by the fact that gender differences were not found in the MG group, although previous research suggests that women have poorer completion and retention rates than men (Messina et al., 2000).

Despite these limitations, this study is an important contribution to the limited empirical knowledge of gender-specific treatment of substance abuse. This study adds to earlier research by its inclusion of a large number of African Americans and people who abuse cocaine in the study sample. Along with Copeland et al. (1993), this study lends empirical support to the contention that gender-specific treatment does more than provide traditional substance abuse treatment. As noted by Nelson-Zlupko and colleagues (1995), "treatment programs that simply layer specialized women's programs over male-normed philosophies fail to provide a treatment atmosphere sufficiently responsive to female clients" (p. 58). These findings are particularly germaine in light of the major role social workers play in the development and provision of substance abuse treatment services. Indeed, social workers such as Finkelstein et al. (1997) and Nelson-Zlupko et al. have been among the most vocal proponents of gender-specific treatment. Obviously, these findings do not provide evidence that the changes prescribed by Finkelstein and Nelson-Zlupko will result in increased retention. Further study is needed to determine the viability of such a conclusion.

TABLE 1--Demographic Characteristics, by Substance Abuse Treatment
Setting and Gender (N = 404)

                                    Mixed-Gender

                                 Female        Male
                                (n = 47)    (n = 127)
                                 M (SD)       M (SD)

Age (years)                    36.4 (8.5)   37.6 (8.8)
Education (years)              12.5 (2.9)   13.8 (3.3)
Substance abuse treatments      1.5 (1.5)    1.4 (2.1)

                                 n (%)        n (%)
Ethnicity
  African American             34 (72.3)    73 (57.5)
  Other                        13 (27.7)    54 (42.5)
Marital status
  Married                       6 (12.8)    18 (14.5)
  Single                       19 (40.4)    48 (38.7)
  Single/widowed/divorced      22 (46.8)    58 (46.8)
Drug of choice
  Cocaine                      32 (68.1)    76 (59.8)
  Other                        15 (31.9)    51 (40.2)
Polysubstance abuse            37 (78.7)   105 (82.7)

                                  Single-Gender

                               Female        Male
                              (n = 52)    (n = 178)
                               M (SD)       M (SD)            F

Age (years)                  35.7 (7.3)   37.8 (8.0)       1.07 NS
Education (years)            12.7 (2.4)   14.2 (3.5)       5.18 *
Substance abuse treatments    2.5 (2.4)    2.6 (2.3)       9.24 **

                               n (%)        n (%)      [chi square], df

Ethnicity                                                 3.99, 3 NS
  African American           35 (67.3)    115 (64.6)
  Other                      17 (32.7)     63 (35.4)
Marital status                                            4.33, 6 NS
  Married                     4  (7.8)     28 (15.7)
  Single                     27 (52.9)     75 (42.1)
  Single/widowed/divorced    20 (39.2)     75 (42.1)
Drug of choice                                            17.50, 3 **
  Cocaine                    37 (71.2)    145 (81.5)
  Other                      15 (28.8)     33 (18.6)
Polysubstance abuse          28 (53.8)    133 (75.1)      16.83,3 **

NOTE: NS = not significant.

* p < .005. ** p [less than or equal to] .001.
TABLE 2--Summary of Findings on Day Program Completion, 90-Day
Completion, and LOS, by Treatment Setting and Gender (N = 404)

                            Treatment Setting

            Mixed-Gender (n = 174)     Gender-Specific (n = 230)

Gender          n         %                n           %

             Day Program Completion

Female          41       87.2              40         76.9
Male           103       81.1             143         80.3
  Total        144       82.8             186         80.9

              90-day Completion

Female          20       42.6              20         38.5
Male            50       39.4              76         42.7
  Total         70       40.2              96         41.7

               Length of Stay

                M         SD               M           SD

Female        118.3     115.9            112.1       110.4
Male          109.1     102.2            107.3       100.5
  Total       111.6     105.8            108.4       102.6

REFERENCES

Baekeland, F., & Lundwall, L. (1975). Dropping out of treatment: A critical review. Psychological Bulletin, 82, 738-783.

Beckman, L. J. (1994). Treatment needs of women with alcohol problems. Alcohol Health & Research World, 18, 206-211.

Beckman, L. J., & Amaro, H. (1984). Patterns of women's use of alcohol treatment agencies. In S. C. Wilsnack & L. J. Beckman (Eds.), Alcohol problems in women (pp. 319-348). New York: Guilford Press.

Blume, S. B. (1990). Chemical dependency in women: Important issues. American Journal of Drug and Alcohol Abuse, 16(3/4), 297-307.

Blumenthal, S. J. (1998). Women and substance abuse: A new national focus. In C. L. Wetherington & A. B. Roman (Eds.), Drug addiction research and the health of women (NIH Pub. No. 98-4290, pp. 229-236). Washington, DC: U.S. Department of Health and Human Services.

Burman, S. (1992). A model for women's alcohol/drug treatment. Alcoholism Treatment Quarterly, 9(2), 87-99.

Carroll, K. M. (1997). Enhancing retention in clinical trials of psychosocial treatments: Practical strategies. In L. S. Onken, J. D. Blaine, & J. J. Boren (Eds.), Beyond the therapeutic alliance: Keeping the drug-dependent individual in treatment (NIH Pub. No. 97-4142, pp. 4-24). Washington, DC: U.S. Department of Health and Human Services.

Chou, C., Hser, Y., & Anglin, M. D. (1998). Interaction effects of client and treatment program characteristics on retention: An exploratory analysis using hierarchical linear models. Substance Use & Misuse, 33, 2281-2301.

Coletti, S. D., Schinka, J. A., Hughes, P. H., Hamilton, N. L., Renard, C. G., Sicilian, D. M., Urmann, C. F., & Neri, R. L. (1995). PAR Village for chemically dependent women: Philosophy and program elements. Journal of Substance Abuse Treatment, 12, 289-296.

Condelli, W. S., & Hubbard, R. L. (1994). Relationship between time spent in treatment and client outcomes from therapeutic communities. Journal of Substance Abuse Treatment, II, 25-33.

Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis for field settings. Chicago: Rand McNally.

Copeland, J., Hall, W., Didcott, P., & Biggs, V. (1993). A comparison of a specialist women's alcohol and other drug treatment service with two traditional mixed-sex services: Client characteristics and treatment outcome. Drug and Alcohol Dependence, 32, 81-92.

Dahlgreen, L., & Willander, A. (1989). Are specialized treatment facilities for female alcoholics needed? A controlled 2-year follow-up study from a specialized female unit (EWA) versus a mixed male/female treatment facility. Alcoholism: Clinical and Experimental Research, 13, 499-504.

DeLeon, G. (1991). Retention in drug-free therapeutic communities. In R. W. Pickens, C. G. Leukenfeld, & C. R. Schuster (Eds.), Improving drug abuse treatment (DHHS Pub. No. 91-1754, pp. 218-244). Washington, DC: U.S. Department of Health and Human Services.

Dodge, K., & Potocky-Tripodi, M. (2001). The effectiveness of three inpatient intervention strategies for chemically dependent women. Research on Social Work Practice, 11, 24-39.

Finkelstein, N. (1993). Treatment programming for alcohol and dependent pregnant women. International Journal of the Addictions, 28, 1275-1309.

Finkelstein, N., Kennedy, C., Thomas, K., & Kearns, M. (1997). Gender-specific substance abuse treatment. Alexandria, VA: National Women's Resource Center for the Prevention and Treatment of Alcohol, Tobacco, and Other Drug Abuse and Mental Illness.

Gainey, R. R., Wells, E. A., Hawkins, J. D., & Catalano, R. F. (1993). Predicting treatment retention among cocaine users. International Journal of the Addictions, 28, 487-505.

George, D., & Mallery, P. (1999). SPSS for Windows: A simple guide and reference. Boston: Allyn & Bacon.

George, R. L. (1990). Counseling the chemically dependent: Theory and practice. Boston: Allyn & Bacon.

Goldberg, M. E. (1995). Substance-abusing women: False stereotypes and real needs. Social Work, 40, 789-798.

Grella, C. E., Polinsky, M. L., Hser, Y., & Perry, S. M. (1999). Characteristics of women-only and mixed-gender drug abuse treatment programs. Journal of Substance Abuse Treatment, 17(1/2), 37-44.

Higgins, S. T, & Budney, A. J. (1997). From the initial clinical contact to aftercare: A brief review of effective strategies for retaining cocaine abusers in treatment. In L. S. Onken, J. D. Blaine, & J. J. Boren (Eds.), Beyond the therapeutic alliance: Keeping the drug-dependent individual in treatment (NIH Pub. No. 97-4142, pp. 25-43). Washington, DC: U.S. Department of Health and Human Services.

Hodgins, D. C., el-Guebaly, N., & Addington, J. (1997). Treatment of substance abusers: Single or mixed gender programs? Addiction, 92, 805-812.

Howell, D. C. (1992). Statistical methods for psychology (3rd ed.). Boston: PWS-Kent.

Huck, S. W., & Cormier, W. H. (1996). Reading statistics and research (2nd ed.). New York: HarperCollins.

Jones, H. E., Haug, N. A., Stitzer, M. L., & Svikis, D. S. (2000). Improving treatment outcomes for pregnant drug-dependent women using low-magnitude voucher incentives. Addictive Behaviors, 25, 263-267.

Kandall, S. R. (1998). Women and addiction in the United States--1920 to the present. In C. L. Wetherington & A. B. Roman (Eds.), Drug addiction research and the health of women (NIH Pub. No. 98-4290, pp. 53-79). Washington, DC: U.S. Department of Health and Human Services.

LaFave, L. M., & Echols, L. D. (1999). An argument for choice: An alternative women's treatment program. Journal of Substance Abuse Treatment, 16, 345-352.

Lewis, B. F., & Ross, R. (1994). Retention in therapeutic communities: Challenges for the nineties. In F. M. Tims, G. DeLeon, & N. Jainchill (Eds.), Therapeutic community: Advances in research and application (NIH Pub. No. 94-3633, pp. 99-116). Washington, DC: U.S. Department of Health and Human Services.

Lex, B. W. (1994). Alcohol and other drug use among women. Alcohol Health and Research World (Special Focus: Women and Alcohol), 18, 212-220.

Messina, N., Wish, E., & Nemes, S. (2000). Predictors of treatment outcomes in men and women admitted to a therapeutic community. American Journal of Drug and Alcohol Abuse, 26, 207-227.

Mondanaro, J. (1989). Chemically dependent women. Lexington, MA: Lexington Books.

Nelson-Zlupko, L., Kauffman, E., & Dore, M. M. (1995). Gender differences in drug addiction and treatment: Implications for social work intervention with substance abusing women. Social Work, 40, 45-53.

Pendergast, M. L., Wellisch, J., & Falkin, G. P. (1995). Assessment of and services for substance abusing women offenders in community and correctional settings. Prison Journal, 75, 240-256.

Reed, B. G. (1985). Drug misuse and dependency in women: The meaning and implications of being considered a special population or minority group. International Journal of the Addictions, 20, 13-62.

Roberts, A. C., & Nishimoto, R. H. (1996). Predicting treatment retention of women dependent on cocaine. American Journal of Drug and Alcohol Abuse, 22, 313-333.

Schliebner, C. T. (1994). Gender sensitive therapy: An alternative for women in substance abuse treatment. Journal of Substance Abuse Treatment, Il, 511-515.

Silberfeld, M., & Glaser, E B. (1978). Use of the life table method in determining attrition from treatment. Journal of Studies on Alcohol, 39, 1582-1590.

Simpson, D. D., & Sells, S. D. (1982). Effectiveness of treatment for drug abuse: An overview of the DARP research program. Advances in Alcohol and Substance Abuse, 2, 7-29.

Smith, D. I. (1985). Evaluation of a residential AA programme for women. Alcohol and Alcoholism, 20, 315-327.

Stevens, J., Arbiter, M., & Glider, P. (1989). Women residents: Expanding their role to increase treatment effectiveness in substance abuse programs. International Journal of the Addictions, 24, 425-434.

Swett, C., & Noones, J. (1989). Factors associated with premature termination from outpatient treatment. Hospital and Community Psychiatry, 40, 47-51.

Wald, R. W., Harvey, S. M., & Hibbard, J. H. (1995). A treatment model for women substance abusers. International Journal of Addictions, 30, 881-888.

Wallen, J. (1992). A comparison of male and female clients in substance abuse treatment. Journal of Substance Abuse Treatment, 9, 243-248.

Wallen, J. (1998). Need for services research on treatment for drug abuse in women. In C. L. Wetherington & A. B. Roman (Eds.), Drug addiction research and the health of women (NIH Pub. No. 98-4290, pp. 229-236). Washington, DC: U.S. Department of Health and Human Services.

Wells, D.V.B., & Jackson, J. F. (1992). Client perceptions of the ideal addictions counselor. International Journal of the Addictions, 27, 571-585.

Wilke, D. (1994). Women and alcoholism: How a male-as-norm bias affects research, assessment, and treatment. Health & Social Work, 19, 29-35.

Yaffe, J., Jenson, J. M., & Howard, M. O. (1995). Women and substance abuse: Implications for treatment. Alcoholism Treatment Quarterly, 13(2), 1-15.

Zankowski, G. L. (1987). Responsive programming: Meeting the needs of chemically dependent women. Alcoholism Treatment Quarterly, 4(4), 53-66.

Original manuscript received July 5, 2000

Final revision received April 25, 2001

Accepted August 6, 2001

An earlier version of this article was presented at the Fifth Annual Conference of the Society for Social Work and Research, January 21, 2001, Atlanta. This study was conducted while the author was a doctoral candidate at the School of Social Work, University of Georgia.

Brian E. Bride, LCSW, PhD, is assistant professor, College of Social Work, University of Tennessee, 193 East Polk Avenue, Nashville, TN 37210; e-mail: bbride@gw.utk.edu.

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.