|
Article Excerpt
CONFIDENCE IN THE ABILITY TO CARRY out a specific set of behaviors, or self-efficacy (Bandura, 1982), is hypothesized to predict future success in the same behavioral domain. Others have expanded on Bandura's concept of self-efficacy and applied it to the field of addictions research (DiClemente, 1986; DiClemente et al., 1995; Marlatt and Gordon, 1985). As described by Witkiewitz and Marlatt (2004), self-efficacy is an important predictor of lapse or relapse to substance use after treatment, and higher self-efficacy to remain abstinent in high-risk situations should be associated with a greater likelihood of abstinence outcomes.
Research on the treatment of alcohol-use disorders and, to a lesser extent, other substance-use disorders (SUDs), indicates that higher self-efficacy to abstain from all substance use at the beginning (Miller and Longabaugh, 2003; Rychtarik et al., 1992; Stephens et al., 1993) and end (Allsop et al., 2000; Goldbeck et al., 1997; Ilgen et al., 2005) of treatment is associated with more positive treatment outcomes. Distal posttreatment measures of abstinence self-efficacy also predict less subsequent substance use (McKay et al., 2005). Thus, a fairly consistent picture emerges whereby increases in self-efficacy precede and coincide with reductions in substance use. Given the general consistency of these findings, it is important to identify specific aspects of treatment that predict self-efficacy.
The relationship between self-efficacy and treatment outcomes does not appear to be strongly influenced by the type of treatment (DiClemente et al., 2001; Project MATCH Research Group, 1997; Stephens et al., 1993). In fact, independent of the type of treatment received, self-efficacy tends to increase during treatment (Ilgen et al., 2005; Stephens et al., 1995). Additionally, Connors and colleagues (2001) found that, irrespective of assignment to one of three psychosocial treatments for alcohol use disorders, more Alcoholics Anonymous (AA) involvement during the first 6 months of treatment was associated with higher self-efficacy at 6 months. But, beyond these general explorations of overall treatment influences on self-efficacy, little is known about the specific determinants of posttreatment self-efficacy. In one of the few studies that examined predictors of self-efficacy, Stephens and colleagues (1995) found that several pretreatment factors, including lower frequency of marijuana use, lower temptation to use, higher likelihood of coping effectively with temptation, less perceived stress, and less contact with other substance users, predicted higher self-efficacy following treatment for marijuana dependence. However, to the best of our knowledge, no one has specifically examined the comparative contribution of different patient and treatment-related factors to posttreatment self-efficacy.
The present study examines predictors of self-efficacy 12 months after discharge from SUD treatment in community residential facilities (CRFs). After verifying that higher 1-year self-efficacy is associated with less 1-year substance use, we focus on the associations between patient-related factors (such as demographic characteristics and baseline general confidence in abstinence) and treatment-related factors (such as treatment orientation, social relationships with others, and participation in skills training) and 1-year self-efficacy. Then, we test whether any of these treatment factors are associated with self-efficacy beyond baseline patient factors.
Method
Participants
Participants in this study were patients with SUDs who were treated in a CRY affiliated with the Department of Veterans Affairs. These patients were predominantly unmarried men who had few social and economic resources, had a history of substance-abuse problems, and were in need of additional support in the transition between hospital-based treatment and independent life in the community. A total of 2,822 patients entered 88 CRY programs and completed intake assessments. Of these, 2,350 patients (83%) were assessed at both discharge and 1-year followup. To attain high follow-up rates, we adhered to established methods for locating and following individuals in longitudinal studies, including informing them at the baseline interview that follow-up evaluations would be conducted and when the evaluations would take place. Additionally, we obtained the names, addresses, and telephone numbers of at least two family members and/or friends with whom the participant did not live, we told participants to inform research assistants about changes in residence, and we kept detailed records of every effort to reach participants.
The sample was predominantly men (99%) and, on average (SD), patients were 41.9 (8.3) years old. Slightly more than half (52.3%) of the sample reported that their ethnic background was white, 36.6% were black, 4.7% were Hispanic, and the remaining 6.4% of the sample reported that their ethnic background was Asian American, Native American, or other. The majority (84%) of patients were unemployed at treatment entry and, on average, patients had 12.9 years of education. Additionally, 83.9%...
|
