Adherence: Adherence will be defined including missing doses and treatment discontinuation using the following approaches:

"Drug at disposal" approach: Based on information on prescribed dosing (mg/day) and information on filled prescriptions (mg expensed from the pharmacy); it is possible to calculate for how long a period the drug will last for that specific patient. Adherence is then defined using the Medication Possession Ratio (MPR) calculated as the ratio of number of days of treatment supply divided by the total length of follow-up time, and converted to a percentage (in the graph below illustrated as the sum of the grey horizontal bars divided by the length of follow-up, ie, time between index date and end of follow-up). The MPR will be computed as a continuous and binary variable (MPR>80%, ie, high vs low adherence)

Persistence: Persistence will be defined as treatment discontinuation and will be measured by estimating the following:

• Time to discontinuation: Date of therapy cessation is not possible to determine with certainty due to the nature of registry data. In order to be able to estimate the time of discontinuation, patients will be considered on treatment for the full time between two purchases as long as they have at least one refill in a six month period. If there is no new purchase within six months the patient is considered to have terminated treatment at a date corresponding to how long the last prescription will last (in the graph below illustrated as the time between index date and the end of the "last" horizontal bar).
• Cumulative incidence of persistent patients still on index treatment at the every 3- month interval after index date (treatment initiation)

Describe and compare patient characteristics (eg, demographics and clinical) among AF patients who:

• Initiate apixaban without previous anticoagulant treatment (warfarin or NOAC use)
• Initiate apixaban with previous war

Demographic parameters : Age-groups, Sex, Socioeconomic variables (e.g. education, disposable income)