Safety Study of Atomoxetine and Cerebrovascular Outcomes

• Cerebrovascular Accident (CVA) [ Time Frame: January 2003 - June 2007 ] [ Designated as safety issue: Yes ]
• Transient Ischemic Attack (TIA) [ Time Frame: January 2003 - June 2007 ] [ Designated as safety issue: Yes ]

Using a proprietary insurance health claims database, Eli Lilly and Company has contracted with an external party to conduct a retrospective cohort study of health claims for the time period from 1 January 2003 through 31 December 2006 (with follow-up of patients through 30 June 2007). This study will evaluate the potential association between atomoxetine and cerebrovascular events. In this study, the incidence of selected cerebrovascular outcomes as represented in health claims data among adult patients who initiate therapy with atomoxetine will be estimated. In particular the study will focus on cerebrovascular accident (CVA) and transient ischemic attack (TIA) as the outcomes of interest. The incidence for each outcome among atomoxetine initiators will then be compared to the incidence in a cohort of similar patients who initiate stimulant treatment and an age and gender-matched general population cohort. The atomoxetine and stimulant-initiating cohorts will be matched on a broad variety of variables, including age, gender, diagnoses, medication use, and healthcare utilization through the use of propensity score matching in order to minimize the influence of confounding by indication. The analysis will include the cohorts (atomoxetine and stimulant initiators) from a previous completed study with increased follow-up time (1 January 2003 through 30 June 2007) and accrue new atomoxetine and stimulant ADHD medication initiators over a 2 year period, so that the study will represent initiators between January 1, 2003 and December 31, 2006 with follow-up through June 30, 2007.

The source population is derived from the proprietary Ingenix RDM, with a cross-sectional population of approximately 12 million current health plan members across the US at the beginning of 2007 who have both medical and prescription benefit coverage.

• Cerebrovascular Accident
• Transient Ischemic Attack
• ADHD

• Drug: atomoxetine
  Other Names:

  • LY139603
  • Strattera
• Drug: Stimulants
  Methylphenidate, amphetamines (including Adderall and mixed salts)
• Other: No intervention (general population)

• A
  All patients 18 years or older who received a first dispensing of atomoxetine during the time period of the study (January 1, 2003 through December 31, 2006) and had at least 6 months of continuous enrollment prior to first dispensing are included in the study cohort. Patients are excluded for presence of pre-existing arrhythmia and heart failure during the baseline period. The study entry date for this cohort is the date of first atomoxetine dispensing.
  Intervention: Drug: atomoxetine
• B
  All patients 18 years or older who received a first dispensing of a stimulant medication (methylphenidate or mixed salts of amphetamine) during the time period of the study with no dispensing of the same drug in the prior 6 months and had at least 6 months of continuous enrollment prior to the first dispensing are identified. Patients are excluded for presence of pre-existing arrhythmia and heart failure during the baseline period. Patients who are matched to atomoxetine initiators using this propensity score method are retained and followed as one comparator cohort. The study entry date is the date of the first dispensing of a comparator ADHD medication.
  Intervention: Drug: Stimulants
• C
  Patients with at least 6 months of continuous enrollment in the database, and without a history of arrhythmia or heart failure during the baseline period are sampled and frequency matched on age and gender to the atomoxetine cohort in a 2:1 ratio. Study entry dates are assigned so as to be similar to the distribution of study entry dates in the atomoxetine cohort. Patients identified and matched as initiators of atomoxetine or stimulant ADHD medications are not eligible for inclusion in this cohort
  Intervention: Other: No intervention (general population)

Inclusion Criteria:

• patients 18 years or older
• received dispensing of one of the study medications during the time period of the study (January 1, 2003 - December 31, 2006)
• 6 months of continuous enrollment prior to first dispensing

Exclusion Criteria:

• presence of pre-existing arrhythmia and heart failure during the baseline period