Warfarin is the most commonly used oral anticoagulant medicine (blood thinner). Although this medicine works well, it is difficult to know how much medicine a patient needs. Many things affect how much medicine a patient needs and doses can be very different from patient to patient. It is important for patients to get the right dose to prevent clotting or bleeding problems that can happen with this medicine if the dose is too low or too high. These problems can be life-threatening. To help find the right dose, patients on warfarin must have frequent blood tests to measure how well the medicine is working. The investigators know differences in people's genes can affect how much warfarin medicine someone needs, but they don't yet know with certainty how to use this information in making patient care decisions. The hypothesis of this study is that using a patients warfarin related genetic information incorporated into a computer algorithm to be used by a warfarin provider will lead to better warfarin management compared to usual care.

Warfarin is the most commonly used oral anticoagulant medication. Due to the difficulty in determining an individual's proper warfarin dose, therapy is typically initiated with a standard dose followed by INR monitoring with frequent dose adjustment to ensure the medicine is working properly. Unfortunately, therapeutic warfarin doses vary significantly from patient to patient, so that even a standard dose can lead to excessive anticoagulation with its associated risk of causing life-threatening hemorrhaging. Genetic and non-genetic factors both influence an individual's warfarin dose requirement and response characteristics. There has been substantial evidence demonstrating a clear gene-dose relationship. Although this importance of pharmacogenetics to warfarin therapy is understood, clear guidance for how such information should be applied to patient therapy is woefully absent. The Personalized Medicine Interface Tool (PerMIT) is a software utility that supplies this critical guidance by modeling the dose requirements and response characteristics of individual patients based on their genotypic and physical characteristics. Using state-of-the-art multivariate computations, PerMIT calculates a warfarin maintenance dose estimate and also models the influence of repeated dosing on plasma drug concentration.

Both genetic and non-genetic factors (such as age, weight and gender) influence warfarin dose requirement and response characteristics of the individual. Recently, multi-variate mathematical equations, which take into account these genetic and non-genetic factors, such as age, weight and gender, have been developed to calculate an estimate of the warfarin maintenance dose requirement (Linder 2002, Zhu 2007, Sconce 2005, Millican 2007). The temporal response to routine administration of medications is dictated by the clearance rate of the medication and its effective concentration, the blood concentration over the dosing interval that is required to elicit the desired pharmacologic effect. The clearance of S-warfarin is primarily dictated by the patient's Cytochrome P4502C9 (CYP2C9) genotype, whereas the effective S-warfarin concentration is primarily dictated by the patient's vitamin K epoxide reductase complex protein 1 (VKORC1) genotype (Linder 2002, Herman 2005, Zhu 2007).

It is now well-known that genetic variants of CYP2C9 lead to decreased S-warfarin metabolism (clearance) and an increased elimination half-life. The elimination half-life of medications dictates the time required for repeated dosing to result in reproducible drug concentrations over the dosing interval for a given dosage. This situation is referred to as steady-state and is the most reliable time to interpret the dose-response relationship (INR measurements). S-warfarin half-life can be estimated based on the individual's CYP2C9 genotype (Linder 2002, Herman 2005, Loebstein 2001) and the steady-state concentration of S-warfarin under optimal anti-coagulation conditions is closely related to the patient's VKORC1 genotype (Zhu 2007).

PerMIT: Warfarin has clear theoretical benefits and has been demonstrated to be accurate; however, prospective randomized control clinical trials are required to demonstrate the efficacy of the PerMIT: Warfarin software in comparison to standard of care. We have designed this two-arm, prospective randomized control trial to directly assess the efficacy of PerMIT: Warfarin in (a) identifying patients' optimal dose requirements; (b) reducing patients' time to achieve stable therapy; (c) reducing the frequency of out-of-range INR measurements; and (d) reducing the number of dose adjustments. This study will evaluate whether, and to what degree, PerMIT: Warfarin improves these patient care outcomes and, by extension, reduces their risk of adverse drug reactions when compared to patients who receive therapy based on the standard of care.

Inclusion Criteria:

• Prescribed warfarin for any indication, so long as they are naïve to the drug at enrollment and are expected to receive therapy for at least 12 weeks

Exclusion Criteria:

• Recent cardiothoracic surgery as indication for warfarin.