Characteristics of Non-coerced and Coerced Clients at Admission
Texas treatment programs which are funded by DSHS are public community mental health centers or non-profit private entities and represent approximately 72% of treatment services in Texas. They are licensed to meet health and safety standards and are contracted to provide specific treatments which are reimbursed based on units of service. Eligibility is based on clinical and financial need. Thus, individuals with the means to enter private treatment are not included in this dataset.
Some 69% of cannabis admissions were involved with the criminal justice system, including those who had a legal status (awaiting trial, diverted to treatment, on probation, parole, or in jail) and those referred to treatment from a criminal justice source (probation, parole, police, or courts). Referral sources for non-coerced admissions included self (20%), social services or protective services (26%), community mental health centers (8%), family or friends (6%), or local councils on alcohol and drugs (6%).
Some 47% of the clients entered programs which reported DSM-IV diagnoses, and of those diagnosed, 84% of the coerced and 69% of the voluntary clients had no condition on Axis I or Axis II. However, some 7% of the coerced and 14% of the voluntary clients had a mood depressive disorder (χ2 284.5, p < .0001) and another 4% of coerced and 9% of voluntary clients were diagnosed with bipolar disorders (χ2 283.7, p < .0001). At admission, 6% of all-coerced and 13% of all non-coerced patients were prescribed anti-depressant or anti-anxiety medications (χ2 400.70, p < .0001) with less than 2% of clients being prescribed any other medication at admission.
Of those patients who received a substance-related DSM-IV diagnosis, 24% were diagnosed as cannabis abusers, 55% were cannabis dependent, and 9% were polysubstance dependent. Fifty-five percent of the coerced clients and 54% of the voluntary clients met the criteria for cannabis dependence (p = 0.1098), 26% of the coerced and 18% of the non-coerced met the criteria for cannabis abuse (p < .0001), and 8% of the coerced and 12% of the non-coerced met the criteria for polysubstance dependence (p < .0001),
These voluntary or non-coerced referrals, as compared to coerced clients, entered treatment with more days of problems in the past month as measured on the six domains of the ASI, they were more likely to be homeless, to have sought care for themselves at hospitals or emergency rooms at least once in the past year, to have used cannabis daily in the 6 months prior to admission, and to have been placed on medication for depression or anxiety problems at admission (Table ).
Characteristics of Clients at Admission to Treatment with a Primary Problem with Cannabis. Based on Their Legal Status and Odds Ratios Predicting Legally Coerced Status at Admission: 2000–2005
To determine which demographic and psych-social functioning characteristics best predict criminal justice status, binominal and multinominal logistic regression models were constructed using referral status (1 = coerced and 0 = non-coerced admission). As shown in Table , being male, employed at admission, having health problems not related to substance use, and being younger predicted being coerced into treatment, while being homeless, seeking care in the emergency room or hospital at least once in the past year, using daily, and having more days of family, psychological, or alcohol or drug problems in the month prior to admission as measured on the ASI predicted being a voluntary or non-coerced client.
Some 75% of the clients who entered treatment with a primary problem with cannabis received outpatient services, with 20% receiving residential services and 3% receiving detoxification services. Voluntary clients were more likely to receive detoxification (5% versus 2%, χ2 166.7, p < .0001) and residential services (36% versus 18%, χ2 979.9, p < .0001), and less likely to be in outpatient care (64% versus 80%, χ2 862.8, p < .0001).
Characteristics of Non-coerced and Coerced Clients Who Completed Treatment
At discharge, 34% of non-coerced and 42% of coerced patients successfully completed the treatment service for which they were enrolled and they were either discharged or referred to an additional level of treatment at another location (χ2 136.56, p < .0001). Another 17% of non-coerced and 21% of coerced were discharged because of violation of program rules (χ2 48.50, p < .0001) and 12% of non-coerced and 6% of coerced left against program advice (χ2 302.9, p < .0001). Seventy-five percent of the non-coerced and 71% of the coerced clients were abstinent in the last month of their treatment before discharge (χ2 31.05, p < .0001).
The average length of stay from admission to date of the last face-to-face treatment contact was 70 days for coerced clients, as compared to 57 days for non-coerced clients (p < .0001). Coerced clients stayed in residential longer (31 days versus 28 days, p < .0001) and in outpatient longer (78 days versus 73 days, p < .0001).
Since coerced clients were more likely to complete treatment, data were analyzed to determine which variables predicted completing treatment. A binominal model was constructed using the six ASI variables, length of stay in treatment, number of family and friends involved in the treatment process, abstinence in the last month of treatment, being in residential treatment, and number of 12-Step meetings attended in the last month of treatment. Those variables which were significant were then included in multinominal models for coerced and non-coerced clients (1 = completed treatment and 0 = non-completer). As Table shows, for both groups, being abstinent in the month prior to discharge, having more family and friends involved during treatment, attending more 12-step meetings during the last month of treatment, and a longer length of stay were significant factors in predicting treatment completion, while having more days of psychological problems in the month before admission predicted non-completion.
Multivariate Prediction of Treatment Completion for Legally Coerced and Voluntary Clients with a Primary Problem with Cannabis: 2000–2005
Status of Clients at 90 day Follow-up
Information on status of clients 90 days after last treatment encounter was obtained on 68% of the clients. Contact was more likely to be made with coerced clients (75% versus 66%, χ2 179.6, p < .0001). Of the clients, 84% of coerced and 77% of the non-coerced clients had not used cannabis in the month prior to follow-up (χ2 77.2, p < .0001).
Binominal models were constructed to predict past-month abstinence at follow-up (1 = no use and 0 = use) using number of months employed since discharge, whether or not the client was living in a household where he or she was exposed to alcohol abuse or drug use, number of arrests since discharge, number of emergency room or hospital visits since discharge, all the ASI problem indices except days of alcohol or drug problems, number of 12-Step meetings attended in the month prior to follow-up, and whether or not the client had been in residential treatment. Those variables which were significant were then included in multinominal models for coerced and non-coerced clients (Table ).
Multivariate Prediction of Past Month Abstinence from Cannabis at 90 Day Follow-Up for Legally Coerced and Voluntary Clients: 2000–2005
For both groups, living in situations where they were exposed to alcohol abuse or drug use and having more days of employment, family, or psychological problems in the month prior to follow-up predicted not being abstinent at follow-up, while attending more 12-Step meetings in the previous month predicted past-month abstinence for both groups. For coerced clients, cannabis use at follow-up was also predicted by more arrests since discharge, while for voluntary clients, having been in residential treatment and having more days of social problems in the month before follow-up predicted cannabis use.
Overall, clients admitted to residential services were more impaired than those admitted to outpatient services, yet they were more likely to be abstinent in their last month of treatment (87% versus 67%, χ2 870.2, p < .0001) and to complete treatment (70% versus 53%, χ2 578.4, p < .0001). But they were less likely to be abstinent at follow-up (34% versus 44%, χ2 207.8, p < .0001).