[CYPRESS-218] Cypress expects more data than we should provide Created: 09/10/13  Updated: 06/20/18  Resolved: 09/18/13

Status: Closed
Project: CYPRESS Issue Tracker
Component/s: None

Type: Question Priority: Major
Reporter: Ben (Inactive) Assignee: Robert Dingwell (Inactive)
Resolution: Delivered Votes: 2
Labels: None

Attachments: HTML File 2_Denise_Craig.html    
Issue Links:
Duplicate
is duplicated by CQM-822 Smoking Gun in Cypress Closed
Solution: Resolution: Cypress should require just enough data to provide confirmatory proof that a patient meets a criteria. Fixed in Cypress 2.4.1
Previous Issue Type: Logic affecting more than 1 eCQM

 Description   

Patient Denise Craig has two encounters with the same code. With the new functionality introduced in v2.4.0, Cypress is now requiring that both of these encounters be sent.

However, section 3.2.3 "How Many Data Should Be Sent?" of the QRDA-I specification states that we should include at least "For each data element in each referenced eMeasur [sic], smoking gun data that offers confirmatory proof, where a patient has met the criterion." Since the criterion only requires one encounter for the measure, "confirmatory evidence" consists of just one encounter and therefore two entries are not expected.

The verbatim error is: Expected to find 2 entries with templateId 2.16.840.1.113883.10.20.24.3.23

Note that this is similar to, but separate from, CQM-620, CQM-822, and other "smoking gun" issues as this doesn't address the concern with disjunctive criteria.



 Comments   
Comment by Eric Gunther (Inactive) [ 12/04/13 ]

In case anyone stumbles on this ticket, guidance was posted on p. 28 of the test procedure http://www.healthit.gov/sites/default/files/cypress_test_procedure_11272013.pdf

Comment by Eric Gunther (Inactive) [ 11/25/13 ]

Still haven't heard any update on this ticket. We are just looking to understand a timeline for when guidance will be posted, or even if it's going to be posted, so we can figure out next steps for our certification process. Please help.

Comment by Eric Gunther (Inactive) [ 11/19/13 ]

This issue was resolved during the Cypress tech talk on 9/17 and the guidance still hasn't been posted. This is preventing us from certifying and appears to be affecting other vendors as well. When will this be posted?

Comment by Judi Dorsey [ 11/11/13 ]

We have also encountered this issue in Cypress v2.4 for the following measures:
CMS 22
CMS 65
CMS 135
CMS 144
CMS 145
CMS 146
CMS 148
CMS 160
CMS 161

We are also looking for ONC to publish guidance on this issue. Thank you.

Comment by Eric Gunther (Inactive) [ 11/11/13 ]

Any update on this?

Comment by Eric Gunther (Inactive) [ 10/29/13 ]

Just looking to follow up on this ticket - any update?

Comment by Eric Gunther (Inactive) [ 09/26/13 ]

Since this issue is resolved will the ONC publish official guidance on this or should we just reference this ticket?

Comment by Jean Colbert (Inactive) [ 09/18/13 ]

Dr. Kevin Larsen from ONC helped clarify the issue raised in Cypress 218 on the Cypress Tech Talk today (9/17/2013).

How many data elements do you need to send along in the QRDA I XML? You really have two options –

1.Currently, Cypress v2.3 and Cypress v2.4 require all the “smoking gun” data elements for a patient that meet a specific criteria, or
2.Send just enough data to provide confirmatory proof that a patient meets a criteria.

Resolution: Cypress should require just enough data to provide confirmatory proof that a patient meets a criteria.

Comment by Eric Gunther (Inactive) [ 09/13/13 ]

We are very interested in this case as well. It seems to me that it would be logical to only send data that serves as confirmatory proof, so in the example above that would be just one encounter. Think about how this would play out with real data - we could be returning dozens of encounters for a single patient. When you consider the hundreds of data points related to these measures, and the hundreds of thousands of patients that exist in some large system, we're talking about huge data files with largely unnecessary data.

We've been operating under the assumption that don't submit all the "smoking gun" data and it's starting to sound like we were wrong on that. Was this different in Cypress 2.3?

Comment by Ben (Inactive) [ 09/10/13 ]

Hey Rob,

I understand your concern with disjunctive criteria, and will continue any conversation about that on CQM-822.

This is a different issue however. In this case, there are not two separate disjunctive criteria which are independently being satisfied. Here we have the exact same criterion being satisfied twice.

I believe the specification is fairly clear in this circumstance, but please let me know if there is guidance that I am missing.

Comment by Robert Dingwell (Inactive) [ 09/10/13 ]

Ben,

From the QRDA Category I standard:

QDM-based QRDA adheres to a "smoking gun" philosophy where, at a minimum, the conclusive evidence needed to confirm that a criterion was met shall be included in the instance.
the very least, the QRDA document should include:

  • For each data element in each referenced eMeasur, smoking gun data that offers confirmatory proof, where a patient has met the criterion. (For disjunctive criteria, e.g., where a criterion can be satisfied by either of two data elements, include smoking gun data for both data elements).

CQM-626 states that disjunctive data criteria must be submitted for all items. That is the resolution/guidance offered by CMS/ONC.

I know you believe that there is a change in the QRDA Category I standard that is coming, as it stands now though the standard is clear as is the guidance from CQM-626. Disjunctive data criteria must be submitted for all items meeting the criteria.

Comment by Ben (Inactive) [ 09/10/13 ]

By the way, this patient is template "B ,Diabetes_Peds". This problem seems to be pretty widespread - we've only tested a couple measures with the new cypress but already found that A ,Cancer_Adult_Female has the same problem in CMS 61 in addition to this problem with 74.

Generated at Fri Apr 26 08:32:31 EDT 2024 using Jira 9.12.4#9120004-sha1:625303b708afdb767e17cb2838290c41888e9ff0.