Factors Associated With Initiation With Atomoxetine Versus Stimulants in the Treatment of Adults With ADHD: Retrospective Analysis of Administrative Claims Data

OBJECTIVES: To determine which factors are associated with use of atomoxetine (ATX) relative to stimulant medications (STIMs) for treatment initiation in adults with attention-deficit/hyperactivity disorder (ADHD). A similar exploratory analysis of the use of ATX versus STIMs in children has been published previously. METHODS: This was an exploratory analysis using a retrospective observational cohort design applied to administrative pharmacy and medical claims from an integrated managed care database. Patients were identified if they had at least 1 administrative claim with a diagnosis for ADHD. Treatment initiation was defined as a new prescription for an ADHD medication preceded by 3 months without similar therapy. Two separate analyses were done, one comparing medication starts for ATX with those of any STIM, the other comparing starts of ATX with long-acting stimulants (LA-STIMs). Logistic regression analyses of prior-year administrative claims were used to compare the frequencies of differential predictors of the use of medication. RESULTS: There were 10,359 patients aged greater than18 years who initiated ATX or a STIM between April and December of 2003 and had at least 1 claim with a diagnosis for ADHD (International Classification of Diseases, Ninth Revision, Clinical Modification codes 314.0x). Approximately one third (28 of 82) of the comparisons related to patient demographics, diagnostic history, and previous treatment history was found to be related to the use of ATX versus STIMs and/or LA-STIMs. Patients were more likely to have received ATX than a STIM if they had prior diagnoses of bipolar disorder (odds ratio [OR] 1.47; 95% confidence interval [CI], 1.16-1.87), alcohol dependence (OR 1.80; 95% CI, 1.26-2.58), anxiety (OR 1.21; 95% CI, 1.05-1.40), previous use of antipsychotic medication (OR 1.55; 95% CI, 1.22-1.96), or previous antidepressant use (OR 1.14; 95% CI, 1.01-1.28). Prior use of behavioral services greater than 12 visits was associated with the use of ATX relative to STIMs (OR 1.46; 95% CI, 1.20-1.77) but not for ATX relative to LA-STIMs. Conversely, ATX was used less often than STIMs for initiation in younger adults aged 18 to 24 years (OR 0.66; 95% CI, 0.58-0.74), female patients (OR 0.89, 95% CI, 0.80-0.99), patients with personality disorders (OR 0.53; 95% CI, 0.34-0.82), and those with prior use of STIMs (OR 0.62; 95% CI, 0.56-0.69). The majority of comparisons (54 of 82) related to demographics, diagnostic history, and previous treatment history did not show statistically significant associations. CONCLUSIONS: During the first year of ATX's market introduction, some differences in the frequency of various clinical factors were found in adults treated with ATX compared with those patients who received STIMs. This association may suggest that STIMs and ATX are used to address different treatment needs in adults with ADHD. Future studies will need to determine the significance of the practice pattern differences inferred here and if they persist after ATX has been on the market longer.