How adolescents with substance use disorder spend research payments

The outcomes of this research were published in Drug and Alcohol Dependence in October 2010.

Study title: How adolescents with substance use disorder spend research payments

Authors: Christian Thurstone, Stacy Salomensen-Sautel, and Paula D. Riggs

Citation: Drug Alcohol Depend. 2010 Oct 1; 111(3): 262–264

Abstract

There is concern that research reimbursements to adolescents may increase substance use. However, these concerns have not been examined empirically. Participants were 70 adolescents (13-19 years) with at least one non-nicotine substance use disorder (SUD) enrolled in a 12-week clinical trial of atomoxetine/placebo for attention/deficit-hyperactivity disorder (ADHD). Adolescent participants received cash reimbursement after each study visit (maximum possible = $425 over 12 weeks). Participants reported each week how they spent the previous reimbursement. Results were tallied, and correlates of spending a payment on substances were examined. Results showed that 26 of 70 subjects reported spending at least one research payment on alcohol or drugs, and 25 of 70 subjects reported spending at least one payment on tobacco. Comparing those who did and did not spend a research payment on alcohol/drugs, those who did had more frequent baseline alcohol/drug use but did not differ in demographics (age, gender) or other clinical characteristics (ADHD severity, diagnosis of conduct disorder, number of SUD diagnoses, number of treatment sessions attended, or pre/post change in number of days used substances in the past 28 days). Comparing those who did and did not spend a payment on tobacco, those who did were slightly older and had more frequent baseline tobacco use. In conclusion, a significant proportion of subjects used at least a portion of one research payment to buy alcohol, drugs or tobacco. However, there was little indication that research payments increased substance use.

Results

Sixty-five (92.9%) of the subjects completed the study. In general, the sample was predominantly male (78.6%) and the average age was 16.1 years (SD = 1.6). Table 1 indicates the current baseline substance use disorder diagnoses for the sample. The average amount of money received during the study was $313.64 (SD = 57.79, minimum = $90, maximum = $375, median = $330). Overall, the pre/post change in the number of days used non-nicotine substances in the past 28 days for the sample of study completers (n = 65) was −4.63 days (SD = 9.76; Z= -3.59, p = 0.0005). The pre/post change in the number of days used tobacco in the past 28 days for the sample of completers (n = 65) was −2.03 days (SD = 7.22; Z = -2.17; p = 0.030).

A total of 26 of 70 subjects (37.1%) reported spending research payments on alcohol or non-tobacco drugs at least one time during the 12-week study. For those reporting at least one purchase on alcohol or drugs, the mean number of times research money was spent on alcohol/drugs was 3.46 (SD = 2.70). Nine participants reported purchasing alcohol, and 8 of these 9 also purchased drugs. All 25 subjects who purchased illicit drugs reported purchasing marijuana and no other drugs. Those who reported spending at least one research payment on alcohol or drugs had a greater number of baseline days of alcohol or non-tobacco drug use in the past 28 days (22.9 days, SD = 7.0), compared to those who did not report spending any research payments on alcohol or drugs (14.7 days, SD = 10.3; U = 308.50, p = 0.001). There was no difference between those who did and did not use research payments to purchase alcohol or drugs in baseline demographic (age, gender, or court mandated status) or clinical characteristics (ADHD severity, conduct disorder, nicotine dependence, number of SUD diagnoses, or past 28 day nicotine use). There was also no difference between groups in the mean number of MI/CBT sessions attended or in the mean pre/post change in the number of days used at least one non-tobacco substance in the past 28 days (-2.42 days, SD = 9.85, for those who bought alcohol/drugs compared to -5.93 days, SD = 9.60, for those who did not buy alcohol/drugs; t63 = 1.41, p = 0.164).

A total of 25 subjects (35.7%) reported spending research money on tobacco. Of these, 11 also spent at least one research payment on alcohol or drugs. For those reporting at least one purchase on tobacco, the mean number of times research money was spent on tobacco was 2.56 (SD = 2.52). Those who reported spending at least one research payment on tobacco had a greater number of baseline days of tobacco use in the past 28 days (23.48 days, SD = 7.97) compared to those who did not report spending any research payments on tobacco (9.91 days, SD = 11.36; U = 201.50, p = 0.0005). They were also more likely to have DSM-IV nicotine dependence (88% versus 40%; χ21 = 15.12, p = 0.0005) and to be slightly older (mean age = 16.56 years, SD = 1.58 compared to 15.82 years, SD = 1.51, t68 = -1.92, p = 0.059). There were no other differences between those who did and did not use research payments to purchase tobacco in baseline demographic (gender or court mandated status) or clinical characteristics (ADHD severity, conduct disorder, number of SUD diagnoses, or past 28 day non-tobacco substance use). There was no difference between groups in the mean number of MI/CBT sessions attended or in the mean pre/post change in the number of days used tobacco in the past 28 days (-2.0 days, SD = 7.55, for those who purchased tobacco compared to -2.1 days, SD = 7.12, for those who did not; U = 475.50, p =0.915). A multivariate logistic regression revealed that one additional day of tobacco use at baseline increased the odds of spending at least one research payment on tobacco by 12%, controlling for participant’s age and baseline diagnosis of nicotine dependence (B = 0.11, SE = 0.06; OR = 1.12, p = 0.049; 95% CI: 1.00 to 1.25). Age and diagnosis of nicotine dependence were not significant in the multiple logistic regression.

Discussion

Data from this preliminary report indicate that a significant proportion of adolescent research participants with SUD used at least a portion of one research payment to buy alcohol, drugs, or tobacco. Those who spent their payments on alcohol/drugs had more severe baseline alcohol/drug use. Those who spent their payments on tobacco were more likely to have baseline nicotine dependence, more frequent baseline tobacco use, and were slightly older.

Although a significant proportion of adolescent research participants in this study used at least a portion of one research payment to buy alcohol, drugs, or tobacco, these findings do not necessarily mean that adolescents with SUD should not receive cash reimbursement. First, all participants had a current SUD at baseline and, therefore, were obtaining and using illegal substances before the study began. Moreover, because tobacco and non-tobacco substance use decreased for those who did and did not use a research payment to buy substances, there is little indication that research payments increased substance use.

Limitations of this study include the reliance on self-report and the lack of different payment schedules. Therefore, trials are needed that use different payment schedules and compare how money is spent. A baseline assessment of the amount of money spent on substances would also allow for a pre/post comparison of how much money was spent on obtaining substances. Finally, because all participants in this study attended a research-based substance treatment (MI/CBT), the results may not generalize to research that does not include treatment.

For much more information about this study, including methods and references, please see: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2950251/.