In general AWS API rate throttling log entries are not an indication of a problem and are expected based on how AWS implements their APIs.
AWS API rate throttling rates are set per account, per api, per minute, and per region. In most cases a large list of throttling events works out to one or two events per hour on a given API, region and account.
AWS has set the throttling limits on RI description APIs very low which is why there are often many throttling log entries for DescribeReservedInstances, DescribeReservedInstancesListings and DescribeReservedInstancesModifications.
It's important to note a throttling on one API will have no affect on another api/account/region. CloudHealth waits a few minutes and tries the collection again, and the call is eventually successful. Again this is working as designed by AWS to guarantee performance across all the users of the service. The only way this would impact another customer application is if that application was making calls against the same APIs in the same time window and region.
The following document has some useful details on how throttling functions:
https://docs.aws.amazon.com/AWSEC2/latest/APIReference/throttling.html