8) The lack of significant differences for symptoms might have been due to the low sample size.
The paper does mention that patients preferred ivabradine, but it looks like this mostly came from a qualitative interview with a subgroup (n = 11/37) of participants.
