The purpose of the study is to assess the occurrence of venous thromboembolism (blood clots), stroke, and heart attack in current users of ORTHO EVRA compared to current users of norgestimate-containing oral contraceptives with 35 mcg ethinyl estradiol with special attention to duration of use. The study uses data from the PharMetrics Patient-Centric Database and MarketScan database, which are US medical claims databases.

ORTHO EVRA is a newly developed transdermal (absorbed through the skin) contraceptive (birth control) system (a "patch") available in the U.S. since April 2002. The study uses data from the PHARMetrics and MarketScan databases to assess the occurrence of venous thromboembolism (blood clots), ischemic stroke (blockage of an artery that supplies blood to the brain), and acute myocardial infarction (heart attack) in current first time users of ORTHO EVRA compared to current first time users of norgestimate-containing oral (by mouth) contraceptives with 35 mcg ethinyl estradiol. The PHARMetrics database is an ongoing longitudinal database with information on about 17 million people. It is made up of data contributed by managed care plans in the United States and contains information on paid claims for pharmaceuticals, medical diagnoses and procedures, and demographic (age, race, etc) information. There will be 3 sets of cases reflecting women who have a first-time recorded claim for an ICD-9 (International Classification of Diseases , ninth revision) diagnosis during the study period (a) venous thrombosis or pulmonary embolism (blood clot in the lung) with hospitalization during the study period and who had subsequent claims for anticoagulant treatment, (b) ischemic stroke and who were hospitalized, (c) acute myocardial infarction or acute coronary revascularization (restoration of blood supply to the heart) and who were hospitalized. Three separate sets of controls (patients) will be identified for each outcome. One analysis will estimate the relative risk of idiopathic (of unknown cause) ischemic stroke or myocardial infarction, and another analysis will estimate the relative risk of idiopathic venous thromboembolism (deep vein thrombosis or pulmonary embolism). A specific analysis called a "Conditional logistic regression" (conditional on the matching factors) will be used in this study. Analyses will be stratified by calendar year. The analyses will be repeated including non-idiopathic (ie, of known cause) cases of venous thromboembolism, stroke, and heart attacks. The original study runs through March 2005, with updates to through August 2006, and October 2007. A transdermal patch containing 6 mg norelgestromin (NGM) and 0.75 mg ethinyl estradiol (EE) is worn for 1 week and replaced weekly for 3 consecutive weeks; the fourth week is patch-free. Monophasic (all pills have the same hormones at the same concentration) or triphasic (3 different pills are taken during the cycle) oral contraceptive pill is taken for 21 consecutive days followed by no pill or a drug-free pill for 7 days.

Inclusion Criteria:

• Users of Ortho Evra or first time users of norgestimate-containing oral contraceptives with 35 mcg ethinyl estradiol between April 1, 2002 and December 31, 2004, who are identified in the PHARMetrics database using the National Drug Code (NDC) assigned by the FDA and modified by Pharmetrics
• 6 months of enrollment in a health plan prior to the event date of their matched case
• Start of study contraceptive use after January 1, 2002
• Updates to the original study included users of Ortho Evra or first time users of norgestimate-containing oral contraceptives identified in the PHARMetrics database and a US healthcare claims database through August 2006 and October 2007

Exclusion Criteria:

• Patients with any ICD-9 code for cancer (except for non-melanoma skin cancer), renal failure, or chronic inflammatory disease

Data collected from PharMetrics, a United States based, ongoing longitudinal database of data contributed by managed care plans that contains information on unpaid claims for pharmaceuticals, medical diagnoses and procedures and from MarketScan database, a US health care claims database.