Uploaded image for project: 'eCQM Issue Tracker'
  1. eCQM Issue Tracker
  2. CQM-4756

"Overlaps" condition - Start date/time missing

XMLWordPrintable

    • Icon: EC eCQMs - Eligible Clinicians EC eCQMs - Eligible Clinicians
    • Resolution: Answered
    • Icon: Moderate Moderate
    • None
    • None
    • Hide
      Thank you for your inquiry for the hospice exclusion in CMS122v10 "Diabetes: Hemoglobin A1c (HbA1c) Poor Control (> 9%)". As you noted, the logic does consider periods with missing start or end date-times, and therefore both are acceptable per the current specification. As explained in BONNIEMAT-630, if the beginning (or end) of an interval is null inclusive, calculations should use the minimum (or maximum) for that datatype, e.g., DateTime(1,1,1,0,0,0,0), or DateTime(9999,12,31,23,59,59,999). This allows your example (null-7/30/2013) to meet the overlaps criteria (1/1/2012-12/31/2012), as you have observed in Bonnie.

      CQL Author's Guide: https://cql.hl7.org/02-authorsguide.html#operating-on-intervals

      Appendix B-CQL Reference: https://cql.hl7.org/09-b-cqlreference.html#minimum

      We cannot provide guidance on whether implementers should manipulate incomplete data in their records as that is beyond the measure requirements and logic specification.
      Show
      Thank you for your inquiry for the hospice exclusion in CMS122v10 "Diabetes: Hemoglobin A1c (HbA1c) Poor Control (> 9%)". As you noted, the logic does consider periods with missing start or end date-times, and therefore both are acceptable per the current specification. As explained in BONNIEMAT-630 , if the beginning (or end) of an interval is null inclusive, calculations should use the minimum (or maximum) for that datatype, e.g., DateTime(1,1,1,0,0,0,0), or DateTime(9999,12,31,23,59,59,999). This allows your example (null-7/30/2013) to meet the overlaps criteria (1/1/2012-12/31/2012), as you have observed in Bonnie. CQL Author's Guide: https://cql.hl7.org/02-authorsguide.html#operating-on-intervals Appendix B-CQL Reference: https://cql.hl7.org/09-b-cqlreference.html#minimum We cannot provide guidance on whether implementers should manipulate incomplete data in their records as that is beyond the measure requirements and logic specification.
    • CMS0122v10

      With "Overlaps" conditions, I understand that cases where an event lacks a recorded end date/time, the missing end date/time will interpreted as ongoing (i.e., positive infinity). 

       

      1. Is the same rule applied to start date/time if it is missing?

      2. Is it acceptable to mandate start date/time for "overlaps" criteria to avoid misreporting patients for these criteria or should we accommodate incomplete data?

       

      I have noticed the following scenario is passing:

      Requirement: "Overlaps the measurement period"

      Start date/time: Null

      End date/time: After the measurement period

       

       

      Logically, with start date/time null in this scenario, there no way to confirm the intervention actually overlapped the measurement period. 

        1. CMS122v10-HospiceExclusion startdate-NULL enddate-AfterMP.PNG
          CMS122v10-HospiceExclusion startdate-NULL enddate-AfterMP.PNG
          46 kB

            edave Mathematica EC eCQM Team
            crintoul Cory Rintoul
            Votes:
            1 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved:
              Solution Posted On: