Results
Figure 1 illustrates the sample selection process in detail, resulting in 783 final eligible patients. Table I presents the characteristics of 229 patients in
the rivastigmine group and 554 patients in the donepezil group. About three-quarters of the pa- tients were >74 years of age, with similar average ages for the two groups (78.5 vs 79.5 years for the rivastigmine and donepezil groups, respectively). More than half of rivastigmine and donepezil users were females and were enrolled in the comprehen- sive health plan.
Approximately one-quarter of the patients in each group used antidepressant medications and 10% of donepezil and 15% of rivastigmine users had taken antianxiety medications prior to initiation of the study drug, but the difference between the two groups was not statistically significant. The most common co-morbidities were hypertension (24%), followed by cardiovascular disease (19%) and stroke (16%). While patients continued to take their index drug, 63% of patients visited their physician office at least once specifically for an AD-related reason (on average 1.52 times), whereas 5% of patients were hospitalised at least once because of AD-related problems (0.11 times on average).
The duration of continuous use and daily doses of rivastigmine and donepezil are compared in table II. During the study period, 47% of patients in both groups continued to take the index drug. However, a smaller proportion of patients discontinued rivas- tigmine (37%) than donepezil (43%).
Overall, rivastigmine users continued to take their medication for 234 days on average (median = 312) while donepezil users continued to take their medication for 235 days (median = 315); this differ- ence was not statistically significant. Rivastigmine and donepezil discontinuers used their medications for an average of 134 days and 122 days, respective- ly. When patients switched their index drug, patients using rivastigmine as their initial cholinesterase in- hibitor continued to use rivastigmine for 84 days while patients using donepezil continued to use donepezil for 104 days. This trend may reflect the switching of rivastigmine users during the initial dose titration period.
The daily doses of the study drugs were calculat- ed by dividing the total dose of the index medication dispensed by the persistency days. The average dai- ly dose for rivastigmine was 7.17 mg/day and the average daily dose for donepezil was 7.81 mg/day. Among rivastigmine users, switchers took the larg-