| Background. Since the Food and Drug Administration (FDA) released a warning with regards to potential cardiovascular risks of stimulant medications, few studies have evaluated trends in the utilization of ADHD related medications. Currently, no research has analyzed the effect of surgery to treat sleep-disordered breathing on ADHD medication discontinuation and the effect of immortal time bias on surgical exposures.;Methods. We analyzed trends of stimulant use in United States outpatient physician office visits by utilizing data from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) from 1998-2007, and evaluated the effect of the FDA warning using logistic regression. To evaluate the impact of surgery on medication discontinuation, we utilized Vermont Medicaid data and developed a time-varying Cox proportional-hazards regression model to compare the hazard rate of medication discontinuation in patients that were exposed to surgery compared to those that were not. Finally, we performed a systematic literature review to assess the prevalence of immortal time bias in recent epidemiologic literature and provide a practical example of the biases impact.;Results. While United States office visits related to ADHD nearly doubled from 1998-2007, the odds of treatment with a stimulant medication in 2007 declined by nearly 60% when compared to 2005 (adjusted odds ratio: 0.38; 95% confidence interval (CI): (0.22-0.67). Adenotonsillectomy and related surgeries were shown to be unrelated to ADHD medication discontinuation (adjusted hazard ratio: 0.87; 95% CI: 0.58-1.31). Immortal time bias is prevalent in the surgical literature, and only 3 of 17 abstracts identified as potentially susceptible to immortal time bias explicitly corrected for this bias.;Conclusions. Although ADHD office visits are on the rise, treatment with stimulant medications is decreasing. Surgical intervention did not appear to impact use of medications, however, limitations in the study design may have biased our estimated association. Future studies should focus on the adequacy of treatment for ADHD due to a shift away from use of stimulant medications. |
