MAT version 5.5 Training
MAT version 5.5 Training

We’re going to start today with an overview
of what has changed between the MAT version 5.4 and version 5.5. We’ll give a
live demonstration of those changes. We will review the timeline for the MAT
version 5.5 release, and then we will open it up for a question and answer
session, and then we will turn the presentation over to the Bonnie team for
them to show you their new interface. So we have made a number of changes with
the MAT version 5.5, both to the user interface and to the behind-the-scenes
operations. The major change we made to the behind-the-scenes operations was
that we changed the way the MAT was filtering out unused items upon
packaging. The previous code was missing some of the items that were used and
omitting them upon packaging, which then made the CQL invalid and the package
failed. This change to the code now ensures that what is used, no matter how
deeply nested or even if it’s only in a return clause, will be included upon
packaging. Every instance that was brought to our attention was tested with
the new filtering and is now working as expected. We also updated the version of
the CQL to ELM parser. The MAT is now using version 1.2.20 of the CQL to ELM
parser which includes a validation for UCUM units and an updated formatter. More
user facing items include a way to replace an included library with a newer
version, a way to retrieve a value set with a program and release, a way to
enter value sets and/or codes that have the same name, changes to the applied
value sets and applied code tables, a way to choose whether or not to include the
code system version in the exports for your direct reference codes, a way to
copy value sets or codes from one measure to another, the ability to edit
direct reference codes, some changes to the shortcut keys and insert icon
options, a new CQL error report, loading bars on all the screens, changes to the
exports, and, to top it all off, a complete redesign of the population workspace. So
with that, let’s get into the system and we’ll take a look at all of the changes.
Starting on the measure library tab, we have added a search feature to the
measures sharing. When you click on the share icon you can now see a search box
at the top of the screen. This can help you narrow down the list of users for a
quicker selection. We’re going to move on to the includes section in the CQL
workspace. And here we determine that as we move into the second year of annual
updates using the MAT with CQL, that there may be a need for people to update
libraries they included in their measures with a newer version of that
library. The only way to do this currently is to remove all references to
the library in your logic. Then delete out the included library and include a new
library with the latest version and then redo the references in the logic. This
was a lot of work just to get a new version of the same library. To address
this, we have now added an edit feature. When you open the included library,
clicking on this edit icon will open up a box with a list of all of the
available versions of that library. To change versions, you simply click on the
check box next to the version you would like to include and click apply. Looking
at the CQL file view of the included library you can now see that the version
has changed. As the alias you originally gave the library is the same, you should
not have to redo the logic in your measure, unless of course, one of the
expressions names has changed. In the value set section, we added a number of
features. The first one is the ability to retrieve an OID based on a program and
release. These are optional fields at this point, but if you want to retrieve a
value set based on a specific program and release you simply make those
selections before you click on the retrieve OID
button. If that value set exists within the release, it will be retrieved. If it
is not included in that release, you will receive an error message stating that it
cannot be retrieved from VSAC. This message does not necessarily mean that
the value set does not exist in VSAC, but that it does not exist within that
particular program and release combination. We have adjusted the applied
value sets table to accommodate the listing of the release if one is chosen.
So, you can see here in the release column. Again the program and release are
optional at this point. You can still retrieve a value set without them the
same way you do today; however, if you do choose a program and release, you will not
be able to choose a version. Additionally, we have changed the modify label in the
applied codes table and separated it out into edit, delete, and copy. We felt it was
confusing for all three of these icons to be under the same heading and some
users thought they had to check the checkbox in order to delete the value
set. It is now clear that the checkbox is for the copy feature and not related
to the Delete icon. This brings us to the copy and paste
feature that has been added. This is functionality that used to exist in the
MAT prior to the introduction of CQL. Per user request, we are bringing it back in
the MAT version 5.5. You once again have the capability to copy value sets out of
one measure and paste them into another measure. To do this, you would click on
the checkbox in-line with the value set or sets you would like to copy and
then click on the copy icon. Then you would open another measure, go to the value set section, and click
paste. The system will paste in any value sets that you have copied that do not
already exist within that measure. If you have copied one that is already in the
second measure, the system will not add a second instance. It will simply skip that
one. You can also copy value sets from a measure and paste them into a stand-alone CQL library or copy from the stand-alone CQL library and paste them into a
measure. We have one more feature in the value set section and that is the
ability to add a suffix onto the value set. This is a user requested feature
that allows you to include two different value sets into the same measure that
have the exact same name. In the current system, you’re not able to do this as the
names of the value sets are used as the identifiers in the CQL logic. If you have
two with the same name, neither the system nor the implementers will be able
to tell which one you meant to reference when it is used in the logic. Therefore,
we needed to find a way to make the identifiers unique to each value set.
So, to do this, you would apply the first value set as you normally would and then
retrieve the second OID. And then you need to add a suffix. So, I’m going to go to
the suffix field here and I’m going to enter 02 and then click apply. You can
see that both value sets are now in the applied codes table and you can
differentiate them by the suffix added. Moving into the code section, you will
see some of the same updates here as we made to the value set section such as,
the ability to add a suffix and the ability to copy codes from one measure
and paste them into another. I’m actually going to switch my measures back because I have a better example in this one. These features work the same way in the
Codes section as they did in the Value Sets section. In addition, we have added a
way for you to choose whether or not you would like the code system version
included with your direct referenced codes. For those codes with which you
want the latest version of the code, you would retrieve the code and apply as
usual. However, if you have a legacy code that is not available in the most recent
code system version and therefore, you would need to state which code system
version the code would be found in, you can check the checkbox that states
Include Code System Version and then click apply. You can see in the table
below that there will be a checkmark in- line with the codes that will have the
code system version included. And those codes will have the code system version
stated within the exports. We have also included a way to edit a direct
referenced code. In the past, you either added the code or you didn’t. There were
not any fields to change and therefore, there was no edit feature. If you no
longer wanted the code, you deleted it. However, with this release, you can now
add suffixes and choose whether or not to add the code system version. As people
may change their minds on these items, we felt an added option was necessary to
edit a direct referenced code. You would click on the Edit icon. This will put the
code information back into the top-half of the screen. Here, you can either add or
change a suffix or you can check or uncheck the Include Code System Version
box and then click apply. You should then see the changes reflected in the table
below. With the MAT version 5.5, we have made a
couple of minor changes to the insert icon and the shortcut keys. Under
functions, ctrl-alt-f, we have now added the age-in-weeks functions. You can see
we have age-in-weeks, age-in-weeks-at and we have calculate-age-in-weeks and
calculate-age-in-weeks-at. Under units, ctrl-alt-u, we have changed what
used to be {H.B./min} to {beats/min} And now when you use a code as an attribute in the attribute builder, it’s
going to insert for you into the CQL editor using a tilde rather than an
equal sign. It was determined during the measure reviews that the equivalent was
more correct in the logic than a hard equals, so we made the change for the
insert icon. However, if you need an equals sign rather than a tilde, you can change
it once it has been entered into the CQL editor. Another user requested
enhancement was to have a way to get a hard copy of the errors in their CQL
file. To accommodate this, we have added an Export Error File button on the view
CQL section of the CQL workspace. Clicking this button will produce a text
file that shows the CQL library in its entirety, identified with line numbers.
Below the CQL file, you will find the errors and which line number they
correspond to. If there are no errors in the CQL file, you will just get a text
file with your CQL file in it. And now we come to the biggest change in the MAT
version 5.5, and that is a complete redesign of the Population Workspace.
This change stems from feedback we received, both through
user satisfaction surveys and through user acceptance testing, that the tool
was not very intuitive and that the Population Workspace was one of the
bigger culprits. Redesigning this space to include a left-hand navigation like
the rest of the tool and eliminating the right-click menus, which were difficult
for our users utilizing assistive technology to navigate, has made the
process of adding, removing, and changing expressions in the populations easier to
do for everyone. When you first enter the Population Workspace, you will be taken
to the View Population section. This section is still laid out in the tree
view that you are used to seeing from the old Population Workspace; however,
there is no longer any functionality on this screen. Just as the View CQL section
in the CQL Workspace is read-only, so is the View Population section. The purpose
of this view is to give you an overall look as to which expressions are
currently paired to which populations. The default view for this section is to
have all populations expanded so that you can see everything at once.
However, the collapse and expand buttons (the plus and the minus signs) still
function so you can adjust the view however you would like. If the measure
you are in is in a read-only state, meaning it is versioned or it is a
measure that belongs to someone else and you do not have editing rights, the View
Population section is the only section that will appear in the Population
Workspace. Again this view allows you to see what expression is included in each
population but it does not allow you to make any changes. If you are currently
editing a draft measure that you do have editing rights for, you will see in the
left-hand navigation all of the populations that are allowed for the
measure based on the measure scoring type and the selection made for the
patient based indicator, which were chosen when the measure was created. To
add expressions into your populations you would click on the appropriate
population heading in the left-hand navigation, pulldown the
drop-down menu, and select the expression you would like to stand in for that
population, and click the Save icon. To add more instances of that population,
click on the add new link next to the Save icon. This will add the next
sequential population with another drop- down box to allow you to choose an
expression. To delete an added population, just click the Delete icon in-line with
the population you wish to remove. It’s important to note here that the
numbering of the population still works the way it did in the old Population
Workspace. Meaning, if you have two populations for example Initial
Population 1 and Initial Population 2, and you delete Initial Population
1; the remaining population will still be named Initial Population 2. Adding
another population will create Initial Population 3 and so forth. It’s also
important to note that the system will not allow you to remove the last
remaining population. For example, if you have a Denominator Exclusion and then
you determine that you do not need it any longer, the system will not allow you
to delete Denominator Exclusion 1 as it is the only one in the section. To remove
the expression from your population, simply reset the drop-down to “Select
Definition” and click Save. We have also retained the ability to view the logic
as it will appear in the human readable export. And you do that by clicking on
the binoculars that are in-line with the population. This will open up a new
window, so, you may need to allow pop-ups for the application before it will
generate. The human readable will also not generate if you have an error within
your CQL file. So, you will need to go back to the CQL Workspace and correct
your errors, and then try again. So, I did have an error. I’m going to come
back here, delete out my error, go back to the Population Workspace, Initial Population
and now you can see the view human readable generate. All of the populations
function the same way with the exception of two. The first one that is slightly
different is the Stratifications. On the Stratification section you have two add
new links. The one at the top will add additional stratifications, while the one
in-line with the stratification heading will add additional stratum to that
stratification. You pair your populations to the stratum level. Remember that each
stratification can have more than one stratum and you can have multiple
stratifications in the same measure, but you can only have one stratification in
each measure grouping. In this section you can add, you can delete, and you can
view the human readable just like you can in the other populations. So, I’m
going to switch my measures again, just so we can take a look at the Measure
Observations. In the Measure Observation section of
the Population Workspace, you will be pairing functions instead of definitions.
A measure observation requires both an aggregate function and a valid user-defined function in order to be complete and to be included in a measure package.
Again here, you can add additional, you can delete, and you can view the human
readable of that population. It’s important to note that if, while in the
Population Workspace, you make one of the populations that you have included in
your measure package grouping invalid, that measure grouping will be removed
and will need to be recreated on the Measure Packager tab of the MAT. You
may have noticed as we were navigating the tool that there is a tab missing. The
Measure Notes tab, which used to be right here, has been removed from the
MAT. We did some research and we determined that it was not being
utilized; therefore, we just removed the entire tab. And next, we did make a couple
of changes to the measure exports. And most of those changes affect the human
readable. I’m going to open a human readable document for you here. In the human readable
export, we have removed the word using in the terminology section. We have also
removed the code system declaration from any direct referenced codes in the
terminology section of the human readable and from the HQMF. These
changes are mainly a cosmetic change and were requested by the implementers.
Additionally, as we stated when we looked at the code section of the CQL Workspace,
we are only showing the code system version for a direct referenced code if
the measure developer has indicated that there is a need to do so. This also
applies to the human readable export and the HQMF and it was requested by
implementers and corroborated by NLM. So, this ends our live demonstration portion
the MAT version 5.5. Let’s go back to the PowerPoint and we’ll take a look at
the schedule. The MAT version 5.5 is scheduled for release into the
production environment the evening of Tuesday, May 29th, 2018. Release Notes will
be posted to the MAT public website after the release is complete. And an
updated user guide will be available within 10 days following the release. please stay tuned to hear a disclaimer
regarding this presentation. This presentation was prepared as a tool to
assist providers and is not intended to grant rights or impose obligations.
Although every reasonable effort has been made to assure the accuracy of the
information within these pages, the ultimate responsibility for the correct
submission of claims and response to any remittance advice lies with the provider
of services. This publication is a general summary that explains certain
aspects of the Medicare Program, but is not a legal document. The official
Medicare Program provisions are contained in the relevant laws,
regulations, and rulings. Medicare policy changes frequently, and links to the
source documents have been provided within this document for your reference.
the Centers for Medicare and Medicaid Services (CMS) employees, agents, and staff
make no representation, warranty, or guarantee that this compilation of
Medicare information is error-free and will bear no responsibility or
liability for the results or consequences of the use of this guide.

Leave a Reply

Your email address will not be published. Required fields are marked *