CQM for MU2 - Discussion

tmccormi wrote on Thursday, April 30, 2015:

We have a need for a work on CQM compliance for MU2. Several of our contributors have offered to do the work or contribute work they have already done. This is a case of too many options for me to decide.

So. I am opening this topic for any one that wishes to submit a proposal for providing all or part of the CQM requirements so that the community can consider the options and help chose a good path.

The vendors I have heard from, directly and indirectly are Sunset Systems, Visolve and Ensoftek.

Dr Michael Brody has also proposed a solution.

This topic is open to anyone that wishs to submit a proposal. Please be aware that you are expected to have researched the actual requirements and understand the gaps in MU1 as it currently is.

Payment will be based on passing the certification tests with InfoGard for the CQM related measures.

Thanks
Tony
Pres OEMR

sunsetsystems wrote on Friday, May 01, 2015:

The problem I ran into is that proposing a solution (and its cost) requires a lot of study and analysis and overlaps greatly with the task of actually doing it. Has any analysis been done that can be shared here? And/or has any been done that is worth paying for?

To get things started I will share (as I did with Brady) the following quote from http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Downloads/eMeasures_GuidetoReading.pdf :

HQMF is a component of a larger quality end-to-end framework in which providers will ideally be able to push a button and import these eCQMs into their EHRs. The eCQMs can be turned into queries that automatically retrieve the necessary information from the EHR’s data repositories and generate quality data reports. From there, individual and/or aggregate patient quality data can be transmitted to the appropriate agency using Quality Reporting Document Architecture (QRDA) Category I (individual patient level) or Category III (aggregate patient data) reports.

So I think we should very seriously explore the possibility of parsing these things. If that can be done sensibly, then I think the “hard part” mostly boils down to writing functions that do the right things in OpenEMR to pick out data corresponding to the different “value sets”.

And obviously, such an approach will greatly simplify the task of supporting additional eMeasures in the future.

Rod
http://www.sunsetsystems.com/

fsgl wrote on Friday, May 01, 2015:

To ensure good value for money, I would suggest competitive bidding based on cost & delivery date.

If delivery is late, a reduction of 10% per week (or whatever the community thinks to be reasonable) can be levied.

sunsetsystems wrote on Wednesday, May 06, 2015:

… aaaand crickets.

One way or another, we need to move forward. Not good to squander the time reprieve that was granted. No comments on my suggestion? Nobody wants to share their insights?

Regarding competitive bidding, the sad fact is that most software projects fail. Would be very bad to go with the lowest bidder and then have them unable or unwilling to complete with no time left for alternatives.

Rod
http://www.sunsetsystems.com/

bradymiller wrote on Thursday, May 07, 2015:

Yep,

All of MU2, including this item, need to be finalized and tested within 3 months. A loud, high-pitched, alarm should be ringing in all of our ears right now.

For CQM, agree the button needs to be pressed very soon.

My input is that whomever does this needs to have already demonstrated that:

  1. They have successfully produced code for OpenEMR with minimal community resources (ie. do not need umpteen reviews and extensive code refactoring by the community).
  2. They have demonstrated ability to understand and fulfill technical guidelines.

-brady
OpenEMR

tmccormi wrote on Thursday, May 07, 2015:

For the sake of listing the options, we could choose to skip CQM, do modular only and require the end users to adopt a certified CQM tool independently. Not my idea of a good option, but it is an option.

Ensoftek has contributed code, been driving the MU2 testing for us so far (with ZHH for their parts), has just done tons of work with Peace Corp on OpenEMR and is about to be certified MU2 with their fork (DrCloudEMR). They offered to port their code back to us. That has been a conversation with me at this point. So unless they submit a proposal to the community soon is will be hard to consider them.

Same is true of Dr Brody and Visolve … Need a proposal/offer/response…

Gut feeling (For me) is if you are not paying reasonable attention to the forum enough to see this and respond then that points against … :slight_smile:

MI-Squared is trying to stay out of this as a developer. We are swamped and it might be considered insider influence in any case.

I don’t think this is going to be one resource, it could be several including paying for code review, integration of contributions.

sunsetsystems wrote on Thursday, May 07, 2015:

Are Ensoftek, Dr. Brody and Visolve all aware of this thread? You might send them a note to make sure.

Rod
http://www.sunsetsystems.com/

tmccormi wrote on Thursday, May 07, 2015:

I told them all personally I would be posting a request for proposal on the forum.

bradymiller wrote on Thursday, May 07, 2015:

Hi,

We should definitely shoot for full certification for several reasons, one of them being only having modular will also force need to import for CQM.

What about partially funding a developer to see what comes from it and then go from there?

-brady
OpenEMR

tmccormi wrote on Thursday, May 07, 2015:

Current funds are $27,600

Due on completion of full cert to InfoGard: $10,000 +/- depending on any extra time we need to buy for retesting when things fail.
So… reserve $2500 just in case.

That leaves $15,100 for CQM and AMC and anything else that needs funded development…

we are getting about $600/m in average contributions… April was no exception, thanks y’all!

mbrody wrote on Thursday, May 07, 2015:

Hello Everybody,

I think that the approach needs to change and be more modular. Each CQM, which happens to be related to PQRS which is a VERY complex task. The first step is to identify the clinical quality measures and PQRS measures that we wish to support in OpenEMR. Once we have identified those measures we then need to look at the workflow to allow users to capture the data points that apply to each of these measures. For example there is a measure related to BMI. The data points that need to be collected for that are height and weight, then the software needs to calculate the BMI and provide an alert if the BMI is above or below the ‘target’ BMI. The doctor then needs to act on that alert.

There are a limited set of actions that allow a doctor to report that he successfully performed for the BMI measure. Successful performance is JUST as important as successful reporting under the new upcoming payment paradigms.

We want to identify measures that meet the following criteria:

  • can be incorporated into the workflow of both primary care providers and specialists with a minimum disruption of the workflow in the practice
  • when possible measures where the data is already being collected in OpenEMR
  • Select measures that are reported once per reporting period rather than on every visit to reduce the workload on the provider

Staging this out :
First identify the measures that we want to support in OpenEMR
Second identify the modules withing OpenEMR that need appropriate modification to collect the necessary data points
Third develop alerts to remind providers to perform the actions necessary to report positive performance on the measures and develop a method of capturing the data necessary to document positive performance
Finally a tool that can ‘scrape’ the information from the OpenEMR database to allow for reporting of these measures.

This should not be a BIG project it should be a series of incremental changes to the program. I believe that with the right quarterbacking it should not require a significant cash disbursement from OpenEMR.

This is not a bid, nor is it a prelude to a bid. It is an offer to work with the community to break this up into manageable components that can be done by the members of the community. If the community wishes to take this approach I will be more than happy to become involved in ‘quarterbacking’ the implementation of CQM / PQRS for MU2 and PQRS 2015.

For those of you who do not know who I am, I have not been involved in this community since the MU1 certification, but I was an active member then. Since that time I have become heavily involved in HIMSS, HL7, and the Standards and Interoperability Framework. I have a very good working knowledge of the requirements for Meaningful Use, and the process needed to certify software for Meaningful Use, and the direction that the project needs to head.

Michael L. Brody

bradymiller wrote on Thursday, May 07, 2015:

Hi Michael,

I think the selection and creation of the MU2 CQM rules is the more straightforward part. Albeit, it appears you’d be selecting rules that will also fulfill PQRS, which brings me to the question of would satisfying PQRS further complicate things or can they both be dealt with without additional resources?

The bottleneck (why it seems to be a big project) appears to be the creation of the QRDA Category I and QRDA Caegory III files. Is your group able to provide guidance on that?

thanks,
-brady
OpenEMR

tmccormi wrote on Thursday, May 07, 2015:

Dr Brody,
We welcome your expertise in the matter.

–Tony

mbrody wrote on Thursday, May 07, 2015:

Good Morning Brady,

QRDA is the format of files that need to be sent to CMS if a doctor is doing REGISTRY reporting. I am the chief compliance officer of a PQRS registry. We work with a number of EHR vendors and accept csv files from the EHR. We then convert the CSV file to the QRDA format and submit those files to CMS on behalf of the provider. Therefore, should OpenEMR choose to produce a csv file that can be uploaded to the registry I work with we can eliminate the need for OpenEMR to produce the QRDA files.

FYI if you want to take a look at the website for the registry visit https://pqrsregistry.com

I think I have just removed a bottleneck if OpenEMR users elect to use this registry. All that needs to be done is for OpenEMR to produce a CSV file with all of the required data elements.

rnagul wrote on Thursday, May 07, 2015:

Hello,

EnSoftek is currently working on getting the AMC/CQM criteria certified for DrCloudEMR and I have spoken to Tony about contributing the successful modules to the community. There will be merging and testing required.

Will know by end of the day tomorrow if our approaches are acceptable to InfoGard.

Ramesh

sunsetsystems wrote on Thursday, May 07, 2015:

Hi Dr. Brody,

Thank you for offering your expertise! It will surely be a very valuable contribution to the OpenEMR project.

If OpenEMR requires use of a registry to create QRDA reports, does that not imply modular and not full certification?

Also what do you think of my suggestion to import eMeasure specifications (in their XML format) into OpenEMR, parse and use them to determine what values to collect and what calculations to perform? Of course this would imply additional projects to collect and extract the data for the various required value sets, resulting in callable functions to supply the data to the calculation engine.

EDIT: I guess the underlying question is, are the eMeasure specs sufficiently complete, unambiguous and reliable enough for this to be possible?

Rod
http://www.sunsetsystems.com/

sunsetsystems wrote on Thursday, May 07, 2015:

Hi Ramesh,

My apologies for being blunt, but why would it be in EnSoftek’s best interest to donate that work?

Rod
http://www.sunsetsystems.com/

tmccormi wrote on Thursday, May 07, 2015:

Why is it in your best interest Rod to contribute what you do or any of us for that matter? I think that is a strange question. In any case that is is divergence from the thread. If you want to discuss the philosophy of an open source ecosystem, I’d be thrilled to do that in another thread.

sunsetsystems wrote on Thursday, May 07, 2015:

Tony, just trying to understand what the proposal is and how it fits with the business model. I think that’s very much on topic. Don’t mean to insinuate anything at all - it’s a very sincere question and certainly not strange. I don’t know EnSoftek as you do.

As for me, while I contribute time to the project for administration and some things like code review, I do very little development that’s not paid for.

Rod
http://www.sunsetsystems.com/

tmccormi wrote on Thursday, May 07, 2015:

But you do contribute the code that was paid for back to the community, they are doing the same thing. Really don’t think this is the thread for this discussion.