[Z-Note: This message is being resent to the two
listservs, since it was initially rejected as a bit longish. The length
limit has been temporarily reset.]
Kumar,
Nicely put questions. And, as we have
seen, they raise many other issues, some of which have been getting
kicked around the industry for some time now. In fact, I believe that
the American Hospital Association (AHA) wrote to the Secretary of DHHS just a
month or two ago urging that future code set changes be
coordinated. And I remember making some related suggestions as part of
some TCS NPRM comments I wrote back in 1997. I also
agree with many of the other responses that you have received thus
far.
The externally maintained code values can change at any
time, and don't require a 180-day notice. A move to a different
code set, or a major version change to an existing code set [like ICD-9-CM
=> ICD-10-CM], however, probably would require a new rule and at least
a 180-day implementation period. I have also seen external
coding changes that were described as "retroactive". In fact, I think
that one of the most recently approved NUBC Manual changes is supposed to
work that way, and another one is supposed to be "effective
immediately".
I have also heard that X12 is considering changing their
code-level maintenance processes so that the current internal code sets can be
changed at any time and can also be used in older versions of the
transactions. I don't recall whether the proposal was that their
maintenance be externalized, or if it was just that the X12 maintenance
rules be changed to provide the necessary flexibility. Either
approach would certainly help with HIPAA.
One consideration I haven't seen noted yet is the need to
retain previously valid codes for several years in order to support data
warehouses and other multi-year data processing applications, as well as
long-term research activities.
And I would clarify that the "Code Maintenance Committee"
that maintains the Claim Adjustment Reason Codes and the Claim Status Codes is
not part of the BCBSA, but is simply sponsored or hosted by the BCBSA in order
to establish that these are, indeed, externally maintained code
sets.
As for possible solutions, I think we will have to have
multiple strategies. Several years ago HHS was trying to standardize the
use of data within their own department. After doing many months of
research, they reported back that they had found out that data is a four
letter word, and that it is spelled "T-U-R-F"! The point is that many
different entities and constituencies are involved in the development and
maintenance of the code sets used in the HIPAA transactions. Some of
these entities would probably be very happy to coordinate their maintenance
schedules. But I wouldn't expect maintainers of general code sets, such
as Zip Codes, Area Codes, etc., to be interested in rescheduling their
activities just to suit the needs of a single industry. Similarly, I
wouldn't expect international bodies such as the WHO (which maintains the ICD
code set) or the ISO to be interested in aligning their activities to suit the
needs of one country. But some alignment is probably possible, and
should be encouraged.
Another approach is to automate the update notification
process. About a year ago I tried to describe a scheme that would
automate this process so that, at the beginning of any given day's processing,
a computer could inquire of a national code set update tracking system
whether any updates had been received since their last update query, and
then automatically download any such changes and update their local code
validation data bases as needed. A certification organization such
as Claredi would be an obvious choice for hosting such an update service,
since they would need to track these changes anyway in order to carry out some
of their certification functions. And this service could also use a
positive notification, or "push" model, if that would work better. Some
major health care industry publishers might also be interested in this
business.
I think it would also be helpful to urge all code set
maintainers to standardize their code maintenance data. I.e., all of
this works better, with or without HIPAA, if certain information is included
as part of any code list. This would include the code value, the code
description, the date the code was added, the date the code is effective, and
the date the code was retired, plus some others, no doubt. Changes to
both the code values and the code descriptions would need to be tracked.
Some code sets probably already have this data, but many don't. For many
small code sets it might even be practical to reconstruct the historical
record. At any rate, we can always start at some point and just note
that any codes that are effective at that point in time were effective as of
that date or earlier and maintain the more complete historical data going
forward.
WEDI SNIP has established two Special Interest Groups (SIGS)
related to code sets at various points within the Business Issues Subworkgroup
(BI SWG). One of these groups, led by Liza Moran and others, produced a
Data & Code Sets Compliance white paper (WP), while a second one, led by
Julie Thompson, researched how the code set update process worked.
Neither of these SIGS, which were recently merged, attacked the problem
you described directly, although the Code Set Resolution SIG was probably
headed that way. But, with so many different entities involved in code
maintenance, this is tough issue to attack successfully. You pretty much
have to accept that you are going to have very little influence, let alone
control, over the activities of many of these entities. Simply
developing and maintaining a contact list becomes a major chore.
As a result of the current discussion, the BI SWG is
revisiting this issue. I would be surprised if there would be any
problem with revising the scope of the current code sets SIG as needed to
include these issues, or we could simply start a separate SIG, or a separate
WP within the existing code sets SIG. But I would hope that we could
build on some of the prior work of the separate SIGS, and also draw on other
reasonably current information resources, such as that on the Claredi
site.
So, to sum up:
1) I think that we could synchronize code set updates
to a significant extent if we can have an open discussion of the
considerations and give the code maintainers an opportunity to voluntarily
synchronize and coordinate their processes.
2) I think that we also need to identify a "minimum
data set" that would be needed to track compliance with a code set,, given the
expected compliance criteria, and encourage all code maintainers to provide
that data as part of their official code lists.
3) I think that we might have a valid business case
for one or more national code set update services that could relieve several
million covered entities of the need to track those update processes
independently.
4) I suspect that SNIP is a reasonable place to do
some of this, given a sufficient quantity and quality of volunteer
resources. This could be done under the BI SWG, or it could be elsewhere
in SNIP, or even a separate Transactions WG SWG. The problem is getting
a long-term commitment of the resources, since it could take several
years to even start to show a difference, unless such cooperation were
explicitly mandated.
Any other thoughts?
----- Original Message -----
Sent: Wednesday, May 22, 2002 8:01 AM
Subject: HIPAA code sets update synchronization
Hi All,
On the subject of HIPAA code sets I wish to request your
comments on the following.
Problem statement
-----------------
