مدار مواد مخدر در میان جوانان تحت درمان: تاثیر مشکلات روانی و بزهکاری

Abstract Previous research has documented associations of addiction with delinquency and psychological problems. However, few studies have evaluated their influence on adolescent's drug use trajectories. The current study aims to examine the influence of these factors on the recovery trajectories of 199 youths aged 15.6 years on average admitted to inpatient and outpatient addiction treatment centers, followed up three and six months later. Results indicate that youth who show higher severity of drug abuse exhibit greater improvement than youth with a lower severity of drug abuse at the onset of treatment. Although psychological problems were associated with baseline drug use, they did not influence drug use trajectory over time. Only delinquency influenced the recovery trajectories of these youth. Results suggest that a high level of delinquency can have a significant effect on the drug recovery process of adolescents and that interventions should attempt to reduce both drug use and delinquency.

Results Drug use problems during treatment and follow-up A description of participants' drug use profiles based on average number of days of use of various substances at the three assessment points is presented for informational purposes (see Table 2). Excluding tobacco, cannabis was the substance most commonly used by youth in the 30 days preceding entry into treatment (T0). For cannabis, the average number of days declined between T0 (14.40) and T1 (4.96), and a slight increase in use was observed at T2 (5.68). The second most frequently used substance was methamphetamine, the use of which diminished between T0 (3.42) and T1 (0.81). The average number of days of methamphetamine use increased between T1 and T2 (1.61) but remained below the initial average (T0). The percentage of total abstinence “in the past 30 days” (excluding tobacco) increased considerably from 10.5% (T0) to 42.8% (T1) and 36.6% (T2). Table 2. Participants' profile of substance use and abstinence in the 30 days preceding their entry into treatment and at the three- and six-month follow-ups. Profile of substance use Entry into treatment (n = 199) Three-month follow-up (n = 152) Six-month follow-up (n = 134) Substance use Average number of days of use (s.d.) (% abstinent; n) Average number of days of use (s.d.) (% abstinent; n) Average number of days of use (s.d.) (% abstinent; n) Alcohol 2.72 (4.01) (37.36; n = 68) 2.05 (4.31) (52.94; n = 81) 2.67 (4.64) (49.25; n = 66) Heroine/opiate/analgesic/narcotic 0.12 (1.49) (98.35; n = 179) 0.21 (1.46) (96.71; n = 147) 0.09 (0.87) (97.76; n = 131) Barbiturate 0.03 (0.38) (98.90; n = 179) 0.06 (0.73) (99.34; n = 151) 0.01 (0.17) (99.25; n = 133) Benzodiazepine 0.60 (3.52) (95.05; n = 173) 0.37 (2.93) (97.37; n = 148) 0.18 (1.37) (97.76; n = 131) Cocaine 0.80 (3.10) (85.16; n = 155) 0.15 (0.89) (94.12; n = 144) 0.19 (0.89) (94.03; n = 126) Methamphetamine or speed 3.42 (6.04) (56.04; n = 102) 0.81 (2.73) (83.01; n = 127) 1.61 (4.83) (79.10; n = 106) Cannabis 14.40 (11.92) (19.78; n = 36) 4.96 (8.79) (56.21; n = 86) 5.68 (9.21) (52.99; n = 71) Hallucinogen 1.58 (4.15) (74.73; n = 136) 0.37 (1.72) (90.20; n = 138) 0.70 (3.05) (85.07; n = 114) Inhalant/solvent/volatile substance 0.12 (1.14) (97.80; n = 178) 0.00 (0.00) (100.00; n = 152) 0.13 (1.32) (98.51; n = 132) GHB 0.02 (0.19) (98.54; n = 135) 0.06 (0.44) (98.43; n = 125) 0.03 (0.16) (97.44; 114) Total proportion of users % % % Abstinence 10.50 (n = 21) 42.76 (n = 65) 36.57 (n = 49) Alcohol only 5.52 (n = 11) 10.53 (n = 16) 11.94 (n = 16) Drugs only 26.52 (n = 53) 9.87 (n = 15) 12.69 (n = 17) Drugs and alcohol 57.46 (n = 114) 36.84 (n = 56) 38.81 (n = 52) Table options Psychological problems at the time of entry into treatment In the 30 days preceding admission into treatment (T0), participants experienced many psychological problems (see Table 3), including issues with self-control (44.4%), depression (42.5%), break rules (42.5%), and concentration (41.4%). Only a small proportion of the participants had attempted suicide during this period (2.2%), but 12.7% indicated that they had experienced suicidal ideations. They generally do not feel preoccupied by their psychological and emotional problems (mode = 0 (not at all); median = 2 (moderately)) and they do not wish to seek help regarding them either (mode = 0 (not at all); median = 1 (not frequent)). However, the counselors consider that the gravity of these problems is significant (mode = 3 (significant problem); median = 2 (average problem)). Table 3. Participants' profile of psychological problems and abuses in the 30 days preceding their entry into treatment. Profile of psychological problems Entry into treatment (n = 199) Psychological symptoms Proportion of self-reported psychological symptoms in the 30 days preceding their entry into treatment (% ; n) Depression 42.46 (n = 76) Anxiety 34.08 (n = 61) Focusing problems 41.44 (n = 75) Low self-esteem 32.60 (n = 59) Extreme agitation 21.67 (n = 39) Low self-control 44.44 (n = 80) Break rules 42.46 (n = 76) Be concern about his weight 19.89 (n = 36) Hallucinations 7.73 (n = 14) Paranoia 19.34 (n = 35) Suicidal thoughts with scenario 12.7 (n = 23) Suicide attempt 2.2 (n = 4) Self-mutilation 6.7 (n = 12) Medication for a psychological or emotional problem 9.4 (n = 17) Verbal abuse 22.8 (n = 41) Physical abuse 2.8 (n = 5) Sexual abuse 0.6 (n = 1) Table options Delinquency at time of entry into treatment A large percentage of the young substance abusers (88.6%) had committed delinquent acts prior to entry into treatment (T0), and 43.2% had previously been arrested for a crime. The number of arrests varied from 0 to 7 (median = 0). Among previously arrested youth, the median number of arrests was 1.5, and the principal offenses reported were drug possession or drug trafficking (67.6%), theft (52.7%), misdemeanors (36.5%), assault (29.7%), and offenses against law enforcement (9.5%), which included disobeying a court order, violating probation, or obstruction of justice. No participant had been arrested for driving under the influence, sexual offenses, homicide, or attempted homicide. Latent growth curve analyses Table 4 presents correlations between the variables in the LGM. Severity of problems with illicit drugs at T0 was significantly and positively correlated with severity of drug problems at T1 and at T2. Severity of problems with illicit drug use at T0 was also correlated with severity of psychological problems at T0; the greater the severity of the drug use problem, the greater the severity of the psychological problems. Finally, number of arrests was correlated significantly and negatively with severity of drug problems. Contrary to expectations, more severe drug use at T0 was correlated with a lower number of lifetime arrests. Table 4. Correlations between variables in the latent growth model. Substance use severity Severity of psychological problems Number of arrests Age Gender Center T0 T1 T2 V1 V2 V3 V4 V5 V6 V7 V8 V1 1 V2 0.189* 1 V3 0.182* 0.411*** 1 V4 0.427*** 0.176* 0.294*** 1 V5 −0.178* 0.107 0.136 −0.117 1 V6 0.186* 0.016 0.026 0.065 −0.020 1 V7 0.208** −0.008 0.066 0.380*** −0.192** −0.059 1 V8 0.312*** −0.020 0.044 0.395*** 0.137 0.186* 0.014 1 *p < 0.05; **p < 0.01; ***p < 0.001. Table options Unconditional model The latent growth curve analyses establish the trajectories of drug use severity, revealing that the statistics are not significant for the linear model (χ2(2, N = 199) = 45.105, p = 0.0001, RMSEA = 0.329, and CFI = 0.000). In the second model, with T2 freely estimated to be 1.200, the statistics are excellent (χ2(2, N = 199) = 2.012, p = 0.3656, RMSEA = 0.006, and CFI = 1.000). The T2 factor shows a decrease in the decline between T1 and T2. Therefore, the freely estimated model is retained. In short, participants were found to have significantly reduced the severity of their drug use by T1, and this improvement was maintained (although to a lesser level) between T1 and T2 (see Table 5). Table 5. Unconditional model parameters. Parameters Sample n = 199 Mean Value (standard error) p Severity of drug use at admission (T0) Intercept γ00 0.213 0.007 0.000 Change over time Slope γ01 −0.090 0.008 0.000 Variance Intercept σ20 0.009 0.001 0.000 Slope σ21 0.007 0.001 0.000 Model type Chi-squared 2.012 DoF 2 p 0.3656 CFI 1.000 RMSEA 0.006 Table options The growth factors exhibited a statistically significant intercept (α = 0.009; SE = 0.001; p < 0.001) and slope (β = 0.007; SE = 0.001; p < 0.001). The specific scores on the “drug” scale will be reported to allow the reader to appreciate the evolution of consumption gravity over time. By reporting the specific scores defining the slope insures a precise description of the participants' degree of change over time. However, it must be kept in mind that these scores, useful in documenting intra-individual and between group changes, do not have a clinical significance in their absolute value at a specific time. The estimators indicate a mean initial drug use severity of 0.213 (SE = 0.007; p < 0.001) and a decrease of 0.090 between T0 and T1. Furthermore, a decrease of 0.018 (0.200*0.090) is observed between T1 and T2. The three mean trajectory values are 0.213, 0.123, and 0.105 for T1, T2, and T3, respectively. The variance of the slope is significant (Var(β) = 0.011; SE = 0.002; p < 0.001), as is the variance of the intercept (Var(α) = 0.009; SE = 0.001; p < 0.001), demonstrating variability in both slope (change in the severity of drug use over time) and intercept (severity of drug use at admission). Conditional model Four of the five covariates were found to be significant with respect to the intercept and/or slope. First, inpatient center (p < 0.05) and IGT-ADO psychological scores (p < 0.001) were positively associated with the intercept. These covariates accounted for 23.5% of the variance in the intercept of the basic model. Second, age (p < 0.05) and addiction treatment center (p < 0.05) were negatively associated with the slope of the drug use trajectory, whereas number of arrests (p < 0.01) was positively associated with the slope of the drug use trajectory. These variables account for 18.2% of the variance in the slope of the basic model. Statistics for the model with the covariates are excellent (χ2(7, N=199) = 4.651, p = 0.7025, RMSEA = 0.000, and CFI = 1.000). Results indicate that older youth had more severe drug use problems at T0 than did younger ones; there was an increase of 0.010 (SE = 0.006; p = 0.064) on the IGT-ADO drug use severity scale for each additional year of age. However, the severity of drug problems of older youth decreased more rapidly than that of younger individuals, with a decrease of 0.014 (SE = 0.007; p < 0.05) on the IGT-ADO drug use severity scale for each additional year of age. Inpatient center youth had more drug use problems at T0 than did outpatient center youth, who exhibited a ratio of 0.031 (SE = 0.014; p < 0.05). The severity of their drug problems decreased more rapidly than did that of outpatient youth, who exhibited a ratio of 0.039 (SE = 0.016; p < 0.05) on the IGT-ADO drug use severity scale. Therefore, older participants and those treated in inpatient centers had more severe drug use problems at T0 but had problems of a similar level of severity at T2 compared to younger and outpatient youth. Participants with more serious psychological problems at T0 also had more severe drug problems at T0, exhibiting an increase of 0.157 (SE = 0.040; p < 0.001) for each unit scored on the IGT-ADO psychological scale. However, the severity of their psychological problems had no significant effect on the slope and did not influence drug use trajectory. Finally, the number of prior arrests at T0 had no significant effect on the intercept of the severity of drug problems. This covariable positively and significantly influenced the slope for drug use severity, with an increase of 0.017 (SE = 0.006; p < 0.01). Number of arrests was associated with poorer addiction treatment outcomes; youth who were arrested more often exhibited more severe drug use problems over time. Table 6 provides a summary of this model, and Fig. 1 describes the effects (using standardized betas) of significant covariates of the latent growth model on the intercept and slope. Only covariates significant at 5% are shown. Among the significant covariates that are predictive with respect to the intercept, psychological problems appear to be the most important contributor (β = 0.313), followed by residential center (β = 0.164). Furthermore, the number of arrests (β = 0.246) has a greater influence on the slope than does attending a residential center (β = −0.224) or age (β = −0.174). Table 6. Conditional model. Parameters Sample n = 199 Mean Standard error p Severity of drug use at admission Intercept γ00 −0.005 0.087 0.951 Age γ10 0.010 0.006 0.064 Sex γ20 0.015 0.014 0.288 Center γ30 0.031 0.014 0.023 Severity of psychological problems γ40 0.157 0.040 0.000 Number of arrests γ50 −0.008 0.005 0.115 Change over time Slope γ01 0.134 0.106 0.204 Age γ11 −0.014 0.007 0.045 Sex γ21 −0.019 0.017 0.254 Center γ31 −0.039 0.016 0.018 Severity of psychological problems γ41 −0.008 0.049 0.876 Number of arrests γ51 0.017 0.006 0.007 Variance Intercept σ20 0.007 0.001 0.000 Slope σ21 0.006 0.001 0.000 Type of model Chi-squared 4.651 DoF 7 p 0.7025 CFI 1.000 RMSEA 0.000 Table options Latent growth model: the effects of the covariates on the intercept and slope. ... Fig. 1. Latent growth model: the effects of the covariates on the intercept and slope. All effects are significant at p < 0.05.