Fig. 1

Three key elements in using pharmacy refill databases to determine a cross-sectional drug count. (1) Allowable index gap: The first element is the allowable index gap, which is the length of time before the index date in which a drug must be filled to be counted. The investigator assigns a fixed or flexible allowable index gap. A fixed allowable index gap does not vary by fill (e.g. 90 days for all fills). A flexible allowable index gap varies between fills because it is based on data specific to a fill (e.g. days supply of the last fill prior to index). The observed index gap is the time between the last fill prior to index and the index date. If the observed index gap is less than the allowable index gap, then the drug is included in the count. (2) Post-index fills: The second element is whether to consider fills that occur after the index date. This involves calculating the observed pre-post fill gap, which is the time between the last fill prior to index and the first fill after index. If the observed pre-post fill gap is less than the fixed (e.g. 90 days) or flexible (2 times the days supply of the last fill prior to index) allowable pre-post fill gap assigned by the investigator, then the drug is included in the count. (3) Prior fill history: The third element is whether to consider fills that occurred further back in time than the last fill prior to index. The rationale is to estimate the drug supply the patient had on-hand at the time of the last prior fill. For example, if the patient’s last fill prior to index occurred 27 days following a prior fill of 30 days supply, they would be estimated to have 3 remaining days on hand, which can be incorporated into the allowable index gap for this drug