Overlap of multiple observation periods - Error in validateGeneratedCohortSet()

[ZorzArg]

Hi,

This is more of a question rather than a description of a bug.

While trying to run incidence analysis, the code stopped while trying to run function generateTargetDenominatorCohortSet.
The error (part of it) I got is the following:

Error in `validateGeneratedCohortSet()`:                                     
! There is overlap between entries in the cohort, 18 overlaps detected first 5:

I also got the same error message (but with more overlaps) while running function cdm_from_con but I could get past that by setting argument .soft_validation to FALSE.
From the CDMConnector documentation of the function above we read:

Other analytic packages may break or produce incorrect results if soft_validation is TRUE and the observation period table contains overlapping observation periods.

I assume that this is the case for the IncidencePrevalence package but I would like to confirm with you.
My questions are the following:

1. Is function generateTargetDenominatorCohortSet errors because ignoring overlapping observation periods would overestimate the time in the denominator?
2. Is there a potential workaround to prevent this error from popping up? Or do we have to resolve this during ETL?

Thanks,
George