COMPARATIVE SAFETY OF STIMULANT AND ATOMOXETINE ASSOCIATED WITH THE RISK OF SUBSTANCE USE DISORDER AMONG ADOLESCENTS WITH ATTENTION-DEFICIT/HYPERACTIVITY DISORDER
Author(s)
Bhattacharjee S1, Chen H2, Bhatara V3, Aparasu RR21West Virginia University, Morgantown, WV, USA, 2University of Houston, Houston, TX, USA, 3University of South Dakota Sanford School of Medicine, Sioux Falls, SD, USA
OBJECTIVES: This study compared the risk of developing substance use disorder in children with Attention-Deficit/Hyperactivity Disorder (ADHD) utilizing stimulant and atomoxetine. METHODS: This study involved retrospective, propensity score matched cohort assessing the risk of developing substance use disorder among stimulant and atomoxetine users with ADHD using the IMS LifeLink Health Plan Claims Database. Adolescents between 12 and 18 years of age starting on stimulant or atomoxetine therapy from July 1, 2004 to December 31, 2005 were included in the study sample. Substance use disorder was classified into (i) tobacco use (ICD-9-CM code- 305.1); (ii) drug abuse (ICD-9-CM codes- 305.2-305.9, 304.x, 292.x, 304.00-304.03, 304.70-304.73); and (iii) alcohol abuse (ICD-9-CM codes- 265.2, 303.x, 305.0, 357.5, 425.5, 291.0-291.5, 291.9, 571.0, 571.2, 571.3, 535.3, and 790.3). Patients with stimulant and atomoxetine were matched on propensity scores calculated on the basis of baseline characteristics. Conditional logistic model was developed using the STRATA option to account for the matched pair design to assess the risk of substance use disorder development. Persistency measured as duration of therapy was included as a covariate in the final model along with other covariates which were significant after matching. RESULTS: The propensity score matched cohort consisted of 2,030 adolescents with ADHD in both the stimulant and atomoxetine user groups (total of 4,060 adolescents). Conditional logistic regression analysis did not show any statistically significant difference between stimulant or atomoxetine use and the risk of substance use development (Odds Ratio [OR]- 0.86; 95% CI- 0.29-2.50). Age was the only covariate that was significantly associated with the substance use disorder (OR-3.55; 95% CI-1.25-10.13). CONCLUSIONS: Utilization of stimulant was not significantly associated with higher risk of substance use disorder compared to atomoxetine in adolescents. More research is needed to evaluate the long-term effects of use of medications in ADHD.
Conference/Value in Health Info
2011-05, ISPOR 2011, Baltimore, MD, USA
Value in Health, Vol. 14, No. 3 (May 2011)
Code
DU1
Topic
Epidemiology & Public Health
Topic Subcategory
Safety & Pharmacoepidemiology
Disease
Mental Health