The encounter date in question was changed from January 1st to January 5th in the 2014 test bundle to alleviate this issue.

Sam-

Thanks for that. Just the response we were looking for.

Howard,

It is a little unclear from the solution text (which I have just updated), but we plan on updating the patient test deck to move the encounter to another time. We plan on updating the patients by late Spring, though we are trying to move that time frame up.

The interim guidance recommended by the Cypress team to ONC last Friday was to allow both results for the measure in the meantime. Official guidance is pending from ONC.

The response to this issue is plainly absurd.

The fact that the measure is specified in this way is clearly a typo on the part of the measure authors, something that is a by-product of the complexity of the QDM. There is no benefit to anyone, not to the measure authors, to the vendors, to CMS, to publc health, to provide a test patient which exploits this typo. This test patient only makes sense to a mathematician, which regretably is the case for a number of other test patients as well.
It is unreasonable to assert that the proper course of action is to request that the measure developers fix this issue. They cannot do so until June. Instead, the proper course of action is to withdraw the test patient and replace it with another that has an encounter on one of the other 180 days of the measurement period that is not the one affected by this typo.
This ticket should be re-opened and addressed.

Clinigence is having this issue as well. Is there guidance for this issue for the ATL's?

Measure definition error being tracked in CQM-900

Created a ticket in the CQM tracker: CQM-900

The best way to raise this issue with the measure developers is to post an issue on the CQM JIRA (http://oncprojecttracking.org/browse/CQM) with the relevant information.

Looking to follow up on this - what's the best way to raise this issue with the measure developers? Again, we understand that Cypress is calculating it correctly per the measure spec, but we feel this measure warrants a modification to the specifications.

Hi Robert,

We understand your hands are tied. Sounds like this issue belongs in the CQM tracker - is that the correct place to raise issues like this to the measure developers? If so then I'd recommend this ticket be moved to that project since, as you've noted, it's not a problem with Cypress.

Eric,

You need to understand that we cannot modify any of the measures. All modifications must come from the measure developers and released through CMS. I understand your concerns with the way it is coded presently but there is nothing that we can do about that we must use the measures as published from CMS.

I agree with Adam that, while this may not occur frequently with real data, it's definitely a problem with the measure specification and it would be great if this could be addressed in the next set of measure updates.

I wonder if we could also consider some kind of exception for vendors who are failing this measure on this specific corner case. Clearly the intent of the measure is to include patients with an encounter only on January 1 - the USHIK site describes the denominator as "Equals Initial Patient Population and seen for a visit between October 1 and March 31". I understand the need to follow the spec but I think this scenario warrants an exception until the specification is altered.

Hi Robert,
This needs to be changed. What you are saying indicates that we should not be counting patients that were seen on January 1. Clinically this does not make sense as patients will be seen on January 1 and will receive a vaccine.

Thanks,

Adam

There is a slight issue with the temporal constraints in the Denominator that cause the patient to not be included. Specifically these to constraints.
<= 92 day(s) starts before start of "Measurement Period"
<= 91 day(s) starts after start of "Measurement Period"

In order for something to match it must either start after the start of the Measurement period or before the start of the measurement period. The encounter in question on the record starts at the exact same time as the measurement period. As it starts at the exact same time it does not start before or after the start of the measurement period.