Predictors of Substance Abuse Treatment Completion or Transfer to Further Treatment, by Service Type

In Brief
In 2005, clients discharged from short-term residential treatment were more likely to complete treatment than clients discharged from long-term residential, outpatient, or intensive outpatient treatment settings Significant predictors of treatment completion or transfer among clients who were discharged from outpatient, intensive outpatient, long-term residential, or short-term residential treatment included: alcohol as the primary substance of abuse, less than daily use at admission, being over age 40, having 12 or more years of education, being White, referral to treatment by the criminal justice system, and being employed Among clients who were discharged from intensive outpatient treatment, men were more likely than women to complete treatment or transfer to another program or facility; however, among clients who were discharged from outpatient or long-term residential treatment, women were more likely than men to complete treatment or transfer to another facility

Among adults who seek treatment for an alcohol or drug abuse problem, many do not complete an entire course of treatment. This finding is a concern given the research showing that length of stay in treatment is one of the strongest predictors of positive treatment outcomes.^1,² Identifying factors that predict treatment completion is an important step towards understanding what leads to successful treatment.

The Treatment Episode Data Set (TEDS) is an annual compilation of data on the demographic characteristics and substance abuse problems of those admitted to and discharged from substance abuse treatment. TEDS also collects information on reasons for leaving substance abuse treatment. These include treatment completion, transfer to another substance abuse program or facility within a single episode of treatment, left against professional advice (i.e., dropped out), terminated by the facility (i.e., discharge was not because the client dropped out, was incarcerated, or any other client reason), and other reasons, such as death.³ Clients’ treatment may be terminated by a facility for a variety of reasons, such as not following facility rules or exhibiting violent behavior.

This report focuses on the 973,000 clients who were discharged from outpatient, intensive outpatient, long-term residential (more than 30 days), and short-term residential (30 days or fewer) treatment in 2005.⁴ Specifically, this report examines the proportion of clients discharged who completed treatment or transferred to further treatment and the demographic and substance use characteristics that predict treatment completion and transfer. Because treatment completion and transfer to further treatment represent positive conclusions to a treatment episode, understanding the characteristics of clients who are completing treatment or transferring to further treatment may assist providers to tailor programs that will yield more successful outcomes for their clients.

In 2005, clients discharged from short-term residential treatment were more likely to complete treatment than clients discharged from long-term residential, outpatient, or intensive outpatient treatment settings (57 vs. 38 percent or less) (Figure 1). Clients discharged from intensive outpatient and short-term residential treatment were more likely to transfer to another program or facility than clients discharged from long-term residential and outpatient treatment settings (19 and 17 vs. 13 and 12 percent).

Figure 1. Reason for Discharge from Outpatient, Intensive Outpatient, Long-Term Residential, and Short-Term Residential Treatment: TEDS 2005
Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

Figure 1 Table. Reason for Discharge from Outpatient, Intensive Outpatient, Long-Term Residential, and Short-Term Residential Treatment: TEDS 2005
Reason for Discharge	Outpatient	Intensive Outpatient	Short-term Residential	Long-term Residential
Completed	36%	36%	57%	38%
Transferred	12%	19%	17%	13%
Dropped Out	29%	24%	15%	31%
Terminated	11%	13%	7%	9%
Other	12%	8%	4%	9%
Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

To examine the client characteristics associated with treatment completion or transfer to further treatment, a statistical analysis was conducted.⁵ This analysis identifies the likelihood of one group completing treatment or transferring to further treatment compared with another group. The remainder of this report presents the client characteristics that significantly predicted treatment completion or transfer to further treatment for each of the four major service types.

As shown in Tables 1 through 4, across the four treatment types (outpatient, intensive outpatient, short-term residential, and long-term residential), the strongest predictors of treatment completion or transfer to further treatment were use of alcohol as the primary substance, being referred to treatment from the criminal justice system, and being employed. However, the strength of each predictor varied across service types. For example, clients referred from the criminal justice system (versus any other referral source) were 58 percent more likely to complete outpatient treatment or be transferred to further treatment, 37 percent more likely to complete short-term residential treatment or be transferred to further treatment, 34 percent more likely to complete intensive outpatient treatment or be transferred to further treatment, and 28 percent more likely to complete long-term residential treatment or be transferred to further treatment.

Table 1. Significant Predictors of Treatment Completion or Transfer to Further Treatment among Clients Discharged from Outpatient Treatment: 2005

Outpatient Treatment
Clients who […]	were […] more likely to complete treatment or be transferred to further treatment	than clients who […]
were referred to treatment by the criminal justice system	58 percent	were referred to treatment by some other source
reported alcohol as their primary substance of abuse	48 percent	reported a drug other than alcohol as their primary substance of abuse
were employed full-time or part-time at the time of admission	27 percent	were unemployed or not in the labor force at the time of admission
were non-Hispanic White	26 percent	were of another race or ethnicity
used substances less than daily at admission	14 percent	used substances daily at admission
were female	10 percent	were male
were over age 40 at admission	9 percent	were age 40 or younger at admission
had 12 or more years of education	5 percent	had fewer than 12 years of education

Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

Table 2. Significant Predictors of Treatment Completion or Transfer to Further Treatment among Clients Discharged from Intensive Outpatient Treatment: 2005

Intensive Outpatient Treatment

Clients who […] were […] more likely to complete treatment or be transferred to further treatment than clients who […]

reported alcohol as their primary substance of abuse 36 percent reported a drug other than alcohol as their primary substance of abuse

were referred to treatment by the criminal justice system 34 percent were referred to treatment by some other source

were employed full-time or part-time at the time of admission 29 percent were unemployed or not in the labor force at the time of admission

were non-Hispanic White 28 percent were of another race or ethnicity

used substances less than daily at admission 15 percent used substances daily at admission

had 12 or more years of education 14 percent had fewer than 12 years of education

were over age 40 at admission 10 percent were age 40 or younger at admission

were male 9 percent were female

Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

Table 3. Significant Predictors of Treatment Completion or Transfer to Further Treatment among Clients Discharged from Short-term Residential Treatment: 2005

Short-term Residential Treatment

Clients who […] were […] more likely to complete treatment or be transferred to further treatment than clients who […]

reported alcohol as their primary substance of abuse 50 percent reported a drug other than alcohol as their primary substance of abuse

were employed full-time or part-time at the time of admission 45 percent were unemployed or not in the labor force at the time of admission

were over age 40 at admission 45 percent were age 40 or younger at admission

were referred to treatment by the criminal justice system 37 percent were referred to treatment by some other source

had 12 or more years of education 25 percent had fewer than 12 years of education

used substances less than daily at admission 23 percent used substances daily at admission

were non-Hispanic White 11 percent were of another race or ethnicity

Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

Table 4. Significant Predictors of Treatment Completion or Transfer to Further Treatment among Clients Discharged from Long-term Residential Treatment: 2005

Long-term Residential Treatment

Clients who […] were […] more likely to complete treatment or be transferred to further treatment than clients who […]

were employed full-time or part-time at the time of admission 85 percent were unemployed or not in the labor force at the time of admission

reported alcohol as their primary substance of abuse 33 percent reported a drug other than alcohol as their primary substance of abuse

were over age 40 at admission 31 percent were age 40 or younger at admission

were referred to treatment by the criminal justice system 28 percent were referred to treatment by some other source

were non-Hispanic White 27 percent were of another race or ethnicity

had 12 or more years of education 15 percent had fewer than 12 years of education

were female 15 percent were male

used substances less than daily at admission 11 percent used substances daily at admission

Source: 2005 SAMHSA Treatment Episode Data Set (TEDS).

Identifying factors that predict treatment completion is a critical part of understanding what leads to successful treatment. The TEDS analyses presented here look at the relationship between treatment completion or transfer to further treatment and gender, age at admission, race/ethnicity, primary substance of abuse, frequency of use, referral source, employment status, and education. The results show that the client characteristics that most strongly predicted treatment completion varied depending on the type of treatment (outpatient, intensive outpatient, short-term residential, or long-term residential). The findings underscore the strong positive influence of being employed on the likelihood of completing treatment or transferring to further treatment. These findings provide important insights to treatment providers and policymakers as they work to ensure that treatment is successful for all clients in all service settings.

¹ McKay, J. R., & Weiss, R. V. (2001). A review of temporal effects and outcome predictors in substance abuse treatment studies with long-term follow-ups: Preliminary results and methodological issues. Evaluation Review, 25, 113-161.
² Gossop, M., Marsden, J., Stewart, D., & Rolfe, A. (1999). Treatment retention and 1 year outcomes for residential programmes in England. Drug and Alcohol Dependence, 57(2), 89-98.
³ Other reasons for discharge include incarceration, death, or other factors external to the treatment regimen (i.e., client moved, was hospitalized, or some other reason out of the client’s control). Clients may be terminated from a treatment program by a facility for a variety of reasons, such as refusing to follow the prescribed treatment program, failing to follow facility rules and procedures, and exhibiting violent behavior.
⁴ In 2005, 34 States submitted discharge data, including AR, AZ, CA, CO, CT, FL, GA, HI, IA, ID, IL, KS, MA, MD, ME, MI, MN, MO, MS, MT, NE, NH, NJ, NY, OH, OK, OR, RI, SC, SD, TN, TX, UT, and VA.
⁵ Statistical analysis of odds ratios was conducted using multiple conditional logistic regression. This method involves simultaneous entry of a set of predictors into a logistic odds model. The results from the model give the statistical significance of each predictor variable relative to a reference value (e.g., the estimate for males provides a comparison to females).

Substance Abuse and Mental Health Services Administration, Office of Applied Studies. (February 26, 2009). The TEDS Report - - Predictors of Substance Abuse Treatment Completion or Transfer to Further Treatment, by Service Type. Rockville, MD