Factors associated with retention of drug abusing women in long-term residential treatment

Abstract

This study examines factors associated with retention at 50 projects funded by the Center for Substance Abuse Treatment, in the Substance Abuse and Mental Health Services Administration, under its Residential Women and Children and Pregnant and Postpartum Women (RWC/PPW) Demonstration Program. These programs provided long-term, intensive residential treatment for pregnant and parenting women and their children. Data for this study were collected from 3265 clients from 24 six-month and 26 twelve-month RWC/PPW projects, admitted to and discharged from treatment between January 1, 1995 and March 31, 2001. Results from an analysis of covariance (ANCOVA) model indicate that, for both 6- and 12-month projects, significant predictors of retention include: bringing children into treatment, age, and coercion (either through CJS or CPS actions). In six-month projects, longer LOS was also associated with frequency of client–counselor contact, as well as with pregnancy status. Taken together, these findings suggest that enabling parenting women to remain together with their children during residential treatment is an important key to achieving the extended period of stay needed to accomplish treatment objectives.

Introduction

Retention—keeping clients in treatment long enough for them to achieve their treatment goals—is a major challenge for substance abuse treatment programs (Hser, Anglin, & Liu, 1990). Substance abuse treatment research has consistently found strong associations between client length of stay (LOS) in treatment and posttreatment outcomes, including decreased drug use, decreased involvement in criminal activities, and increased employment (Anglin and Hser, 1991, Gerstein and Harwood, 1990). However, the reality of drug treatment is that retention and completion rates are often quite low; completion rates of 50% or less are common across various treatment modalities (Copeland and Hall, 1992, Mammo and Weinbaum, 1993).

A good deal of past research has examined relationships between client characteristics and treatment LOS. A number of studies have found LOS and treatment completion to be positively correlated with age in programs serving women (Bell et al., 1997, Lewis et al., 1997, Scully, 1997). Other client factors consistently associated with longer LOS are higher education level (Bell et al., 1994, Knight et al., 2001) and higher motivation for treatment (Anglin and Hser, 1991, SAS Institute, 2001). Higher education level has also been linked to treatment completion (Brindis, Clayson, & Berkowitz, 1997). A recent study by the Center for Substance Abuse Treatment (CSAT) found that age, education, employment, and legal supervision were all related to longer LOS, while severe drug use consistently predicted shorter LOS (Hser, Maglione, Joshi, & Chao, 1998).

Recent literature suggests that psychological dysfunction is related to shorter LOS in treatment. Simpson et al. (1997), in analyzing data from the Drug Abuse Treatment Outcome Study (DATOS), found that individuals with psychological dysfunction paired with heavy cocaine and alcohol use had comparatively short LOS. Depression and anti-social personality disorder have also been shown to be associated with comparatively short LOS in drug treatment among women (Haller et al., 1997, Williams and Roberts, 1991). Depression may also reduce the likelihood of completing treatment (Williams & Roberts, 1991).

Factors predictive of retention in treatment that are specific to female drug abusers have not been clearly identified. Several studies have reported that pregnancy status and trimester of pregnancy at treatment admission are associated with retention (Grella, Joshi, & Hser, 2000); however, findings on these subjects have been inconsistent. Haller, Elswick, Dawson, Knisely, and Schnoll (1992) reported that women who entered treatment in their third trimester of pregnancy stayed longer than women who entered during the second trimester or those who entered during the postpartum period. Conversely, Schnoll (1994) reported that women who entered treatment in the first trimester stayed in treatment longer than women who entered during the second or third trimester. Wald (1992) reported a 20% completion rate overall among pregnant clients.

Some client characteristics have not been consistently associated with treatment retention or completion. For example, African–American and Hispanic women have been found in some studies to have comparatively short average LOS in long-term treatment (Center for Substance Abuse Treatment, 1997, McCaul et al., 2001); others have found race/ethnicity unrelated to retention (Hser et al., 1998). Among women, it is also unclear whether a legal mandate to attend substance abuse treatment programs leads to longer stays in such programs, although the majority of studies have found that women who are legally mandated to attend treatment tend to stay longer than women who enter treatment voluntarily (Brindis et al., 1997, Knight et al., 2001, Lewis et al., 1997, Wald, 1992, Williams and Roberts, 1991). Finally, research examining primary drug of choice, severity of drug use, and prior treatment history have produced inconsistent findings as to these variables' associations with retention (Brindis et al., 1997, Scully, 1997, Simpson et al., 1997).

Although program characteristics have not been studied as extensively as client characteristics, several studies have examined program factors that may affect retention of women clients. Several studies have found that residential programs that allowed women to bring children into treatment had significantly higher retention rates than programs that did not allow children (Coletti et al., 1992, Hughes et al., 1995, Szuster et al., 1996). Similar findings were reported by Metsch et al. (2001), who compared retention rates between women who entered treatment with children and those who did not.

Hser et al. (1998) found several program-level factors statistically associated with retention, including provision of case management services, availability of vocational training, and engagement of clients in the development of treatment plans. Other variables that have been linked with longer LOS are a short waiting list for admission (Lewis et al., 1997), greater intensity of treatment as measured by the number of hours in treatment related activities (Hoffman et al., 1994, Strantz and Welch, 1995), family involvement in the treatment process (Stanton, 1997), and high clinical staff/client ratios and small caseloads (Stark, 1990).

The present study extends the body of research described above by examining factors associated with retention at a distinctive set of residential substance abuse treatment facilities. The data are derived from projects funded by CSAT, in the Substance Abuse and Mental Health Services Administration, under its Residential Women and Children (RWC) and Pregnant and Postpartum Women (PPW) demonstration programs (CSAT, 2001). In fiscal years 1993–1995, CSAT awarded 50 5–6 year grants under these programs to support development of facilities providing long-term residential substance abuse treatment for pregnant and parenting women (Clark, 2001). Each RWC/PPW project, while having the latitude to develop its own unique treatment model and philosophy, was required to meet specified service requirements, to develop programming for either a six-month or 12-month planned length of stay (Ns=24 and 26 projects, respectively), and to focus either on pregnant and postpartum women (postpartum meaning having given birth within the past 12 months) or on women with children over one year of age stay (Ns=24 and 26 projects, respectively). This paper examines retention (LOS) patterns in this program and factors affecting LOS.