Stream: cql
Topic: Artifacts
rekha (Oct 02 2019 at 10:13):
Hi All,
We are having different artifacts in CDS Connect,How can we know that these Artifacts related to Measure or Cds cards??
Thanks in advance.
Chris Moesel (Oct 02 2019 at 12:58):
Hi Rekha -- the mission of CDS Connect is to advance Clinical Decision Support (CDS). So all artifacts are CDS. None of the artifacts are electronic Clinical Quality Measures (eCQM). Perhaps some of the confusion is because some artifacts might be derived from an eCQM -- but they are not an eCQM anymore; they are CDS that uses similar logic and value sets as an eCQM -- but they're not a measure.
rekha (Oct 02 2019 at 13:20):
Okay.Thank you @Chris Moesel .Your answer is very helpful.
I want to know that ,If I had a CQL File ,How can I know which type of hook is generated based on the CQL Logic.
Chris Moesel (Oct 02 2019 at 13:54):
CQL only defines logic -- and doesn't have any formal way of indicating how it relates to CDS Hooks. In fact, some CQL artifacts may not be intended for CDS Hooks at all. Instead, they might be intended to be used with FHIR Clinical Reasoning resources, or the CDS Knowledge Artifact Specification, or perhaps require the implementor to read human documentation to understand how they integrate w/ systems.
Chris Moesel (Oct 02 2019 at 13:58):
Your best bet is to read whatever documentation accompanies them and see what other files are provided (e.g., FHIR PlanDefinition, etc). One other thing I'll note: There are a few CQL artifacts that use conventions to encode CDS Hooks information into the CQL. One potential approach is described a bit here: http://build.fhir.org/ig/cqframework/opioid-cds/implementation-documentation.html
rekha (Oct 03 2019 at 06:40):
Thank you @Chris Moesel .Apart from all these ,If we are having a CQL file ,based on which parameters we will categorize them into Measure or CDS cards?
Chris Moesel (Oct 03 2019 at 13:00):
Hi @rekha. Technically speaking, the CQL specification does not specify any way of indicating if a CQL library is a measure, CDS, or something else. CQL libraries are just logic -- so you need to see how the CQL is used to determine if it's for a measure or CDS. That's why I suggested you'd need to look at surrounding documentation or accompanying standards-based files.
There might be some hints in the CQL file itself, but they are only hints. For example, if you see an expression called "Measure Population" or "Numerator", then it's _probably_ a measure -- because the CQL Formatting and Usage (for eCQMs) Wiki suggests these names for all CMS-based measures. You might run into CDS that defines a "Numerator" expression, but it doesn't seem likely (but "Initial Population" and "Denominator" very well _might_ show up in CDS). Similarly, if you run into an expression called "Recommendation" or "Suggestion" then there's a good chance it is CDS and not a measure.
But as I said, these are only hints -- and they will not work 100% of the time.
Surya (Oct 09 2019 at 12:00):
Hi @Chris Moesel Among these AHRQ artifact cql logic's, can anyone get cds cards and measure report??
please share that information if any one get those...
Thanks in advance.
Chris Moesel (Oct 09 2019 at 13:05):
@Surya - as far as I know there are no ready-to-go CDS Hooks services or executable eCQMs in CDS Connect. We do not plan to host eCQMs in CDS Connect, so you'll likely never get CQL w/ MeasureReports from CDS Connect. And while we do have CQL that could be integrated into a CDS Hooks service, there is work required to do so (involving either the CQL Services tool I've already pointed to or the CQF Ruler that Bryn has already pointed to).
Jerry Goodnough (Oct 10 2019 at 19:16):
So are there any published constraints or recommendations on syntax elements such as maximum identifier length?
Bryn Rhodes (Oct 10 2019 at 19:24):
There are informative formatting conventions in the CQL specification itself: https://cql.hl7.org/14-g-formattingconventions.html
Bryn Rhodes (Oct 10 2019 at 19:25):
There are also conventions in the Quality Measure IG: http://hl7.org/fhir/us/cqfmeasures/2019May/using-cql.html
Bryn Rhodes (Oct 10 2019 at 19:25):
However, none of those actually put a recommended constraint on maximum identifier length, though that would be a good one to recommend.
Jerry Goodnough (Oct 11 2019 at 02:44):
Thanks Bryn
Surya (Oct 15 2019 at 06:50):
Hi all, Are there any artifacts that which will generate cds cards using cql logic??
Thanks in Advance...
Lloyd McKenzie (Oct 15 2019 at 12:29):
@Bryn Rhodes
Bryn Rhodes (Oct 15 2019 at 14:24):
Hi @Surya , the recommendation logic here will produce the elements required for a CDS Hooks card: http://build.fhir.org/ig/cqframework/opioid-cds-r4/
Surya (Oct 16 2019 at 05:18):
Thanks @Bryn Rhodes i already gone through that opioid recommendations logic and i got cds hooks cards too, but i want to know" is there any other artifacts which will generate the CDS Hooks cards??" .
Last updated: Apr 12 2022 at 19:14 UTC
