Rama,

I agree with you that the work of comparing results between the three largest 
testing vendors was very much needed.  Patrice and Mike are to be commended 
for all the work they put into this, and I hope there will be more extensive 
studies in the future.

You bring up some great points.  And since you invited me to address them, 
here it goes...

First, in spite of the title of the session being "Validation or 
Certification", the specific instructions from Patrice to all three vendors 
were that we were ONLY to discuss testing.  We were not supposed to discuss 
certification, only testing.  I specifically asked Patrice about this because 
it seemed odd, given the title of the session and she specifically forbade me 
to discuss certification or HIPAAmetrics.  My presentation was to be ONLY 
about testing outbound transactions using Velocedi.

Since I was the first presenter, I stuck to the script.  The ONLY thing we 
were supposed to do in our 10 minutes each was to test/analyze/validate the 
one file that Patrice had provided us with, and no other.  No other 
discussion or presentation was to take place, per Patrice's instructions.  
Much to my dismay, the other two presenters chose to present their entire 
product lines rather than sticking to the rules.  Patrice, given the 
situation, passed me a note inviting me to take a second turn of 5 minutes to 
make a "sales pitch", but I declined.

Had I know that I could present not only about outbound testing but also about 
certification and inbound testing and inbound certification, I would have 
used my 10 minutes very differently.

So, given that the only topic was testing/analysis/validation, you can 
probably understand why when you brought the topic of the difference between 
testing and certification from the audience, Patrice diverted your question 
and never addressed the certification part of it.  She was not interested in 
discussing certification in this meeting, only testing.  I hope this email 
addresses your question.

Comparing the offering of these three vendors by only looking at the results 
of testing outbound EDI files is equivalent to comparing automobiles by only 
looking at their tires.  Yes, they do all have four tires on the road.  Yes, 
all tires are made out of rubber, steel, plastic, cotton and other fibers.  
Yet, some tires will separate the thread under certain conditions and other 
tires will work better in the snow or may have studs for ice.  But, would you 
buy a car based on the tires?  There are many other aspects that are much 
more important than the tires.  The Minnesota case study only looked at the 
single aspect of outbound transactions test results, with the explicit 
exclusion of other important aspects, such as certification or inbound 
testing.

So as far as testing is concerned, Patrice showed that the three main 
contenders can test the files with some differences.  However, had we tested 
a different file, a more moderately sized file for example, we would had seen 
a number of significant differences in the quality and completeness of the 
results.  This is left for a future exercise.   For example, I tested a 
medium size file with 3,500 professional claims and it took 133 seconds 
(2m13s) using the Foresight software but only 3 seconds using Claredi's 
Velocedi, using exactly the same hardware.  This sort of differences are not 
visible when the test files are as small as the file Mike and Patrice gave 
us.  I will stop here, as I don't want to start benchmarking discussions.

As for your request for clarification about the disclaimer in the test 
report...  The test report is exactly that, a TEST report.  It is NOT 
certification.  For almost two years now I have been making a big distinction 
between testing and certification.  Some people understand the difference, 
others don't understand the difference.  I think it is clearly expressed in 
the white paper, but obviously it is not as clear as I think it is, or this 
topic would not come up so often.  I will try to explain it in this email, 
but I suspect it is going to need a more detailed description, so I am going 
to start working on a Claredi white paper to make sure people understand what 
WE mean by certification.  I am aware that most of our competitors do NOT 
understand the difference and therefore try to equate the two concepts as if 
they were one and the same.  In their eyes, if you pass a test without 
errors, that equates to being certified HIPAA compliant.  I believe that is 
not the case.  Please keep reading.

When Patrice approached Claredi about the case study, she was very explicit 
about wanting to show only outbound test results.  She did NOT want to show 
inbound testing (with Claredi's Wibbler) and she did not want to show 
certification in either direction.  So we gave her a temporary test account 
of the same kind we offer in our "trial run" accounts.  Those accounts are 
established with only a limited set of capabilities and do NOT have access to 
Claredi's certification, they have access only to testing tools.  In a 
"normal" (that includes certification) account, the report summary that 
contains the disclaimer you are citing also contains a few more sections 
between the "Action" and the "Support" areas of the summary.  These other 
sections are labeled "Certification", "HIPAAmetrics", and "File Specific 
Info".  Again, these sections are only available to Claredi customers, not to 
test accounts.  Also, test accounts are restricted in other ways.  I can give 
you the restriction details off-line.

Since I don't think it would be appropriate to turn this email into a sales 
pitch, I will only describe the outbound "Certification" process.  We also 
have a completely different process for inbound certification, and I could 
describe it another day if there is enough interest.

Once a file has passed the Claredi testing/verification/validation (all 
synonyms) process without HIPAA errors, it becomes eligible for certification 
by Claredi.  The "Certification" section of the summary page then shows a 
button labeled "Submit this file for certification".  If the file has HIPAA 
errors the button does not appear.  Unless the user clicks this button, the 
file is not taken through Claredi's certification process. It is only tested, 
not certified.  This is why the summary report MUST NOT be construed as a 
certification report, even if it shows that a file does not have "HIPAA 
errors".  And thus the disclaimer.

As you can tell, the summary report shows three types of "errors": the "HIPAA 
Errors", with the standard 7 types defined in the SNIP white paper, the 
"Business Errors" section, and the "Warnings" section.  Only the "HIPAA 
Errors" section must be clean for the file to be a candidate for 
certification as HIPAA compliant.  The file could have "Business Errors" and 
"Warnings" and still be certified as HIPAA compliant, since the certification 
of compliance measures against the requirements of the HIPAA implementation 
guides.

In the next few weeks we will be relaxing the cleanliness requirement even 
further.  For a file to qualify as a candidate for certification, it will 
need to have at least one "business unit" (e.g. claim) without "HIPAA Errors" 
rather than the entire file being error free.  The HIPAAmetrics certification 
will give a detail of the contents of the file, including the percentage of 
compliant and non-compliant claims.  This is currently in "beta" with a 
restricted number of accounts and will appear in all the Claredi accounts in 
a few weeks, as well as in the certification details through the directory.

But, back to the certification process... Once the file passes the testing 
clean enough to be "certifiable", and the user clicks on the "Submit for 
certification" button, Claredi takes the file through a process that looks at 
the data content.  Based on the data content of the file we can identify 
patterns, such as (I will use the claim as an example) the presence of 
Billing Provider and/or Pay-To provider, or the presence of both Subscriber 
and Patient or only Subscriber, or the presence of secondary coverage, or the 
presence of a prior adjudication (COB claims), or the presence of a referring 
physician, or the maximum number of service lines submitted by this entity, 
or the presence of a "cluster of data" that indicates this is an 
anesthesiology claim, or a DME claim, or an ambulance claim, etc.

We identify these data clusters through a process we call HIPAAmetrics.  This 
is a unique Claredi pattern recognition process that has never before been 
applied to EDI data, and we have applied for a patent to cover it.  The 
result of the process gives us the "metrics" of this particular file.  I am 
sure you have heard me say many times that the fact that a file does not 
contain errors is not enough.  You have to know what was actually contained 
in the file.

Using HIPAAmetrics we can determine exactly what is in the file, from the 
business perspective.  The testing phase looks at the EDI perspective and 
determines that the file (or the individual claim within the file) does not 
contain errors.  The HIPAAmetrics looks at the healthcare contents and 
"characterizes" it.  At the completion of the HIPAAmetrics phase Claredi can 
say: 
We have seen a file that contained the following data
- Office visits, 234 without errors, 3 with errors
- Primary claims, 180 without errors, 20 with errors
- Primary claims with a second payer, 0 without errors, 20 with errors
- Secondary claims, 0 without errors, 97 with errors
- Consultations, 5 without errors, 40 with errors
- etc.
Claredi is making a statement of what we have seen.  Because there are some 
claims that are "compliant", the provider could claim to be compliant.  
However, the statement from Claredi clearly shows that although this provider 
has the ability to generate HIPAA compliant office visits for "single payer" 
situations, the provider does not have the capability to create correct 
claims when two payers are involved.

The HIPAAmetrics describe the profile that we have seen reflected in the EDI 
data itself.  The HIPAAmetrics certification gives the detailed data profile 
necessary to make a determination of compliance.  Without this sort of 
detail, all you would know is that the provider has generated something that 
did not have errors, but you don't know what that "something" was. Certainly 
none of the other testing entities (neither our competitors nor the 
translator validators) provide these details, and I suspect that they do not 
yet understand the difference between passing a "clean test" and being 
certified as HIPAA compliant.  Both EDIFECS and Foresight show a pass/fail 
approach in their reports. Of course, by the end of this email their eyes 
will be wide open.

So, there is a significant difference between a testing or validation service 
and Claredi's certification.  Big difference.  The problem is that if you 
don't understand the difference and assume that any entity that can produce 
an error free file is automatically "certified" you will certainly find a 
real problem when you look at the quality of the data that entity sends.  And 
thus the bad rap that the "pseudo-certification" provided by the pass/fail 
approach of our competitors is giving to the word "certification".

And this is also the reason why a vendor cannot really be certified unless all 
of its clients are individually certified.  It is just too easy for the 
vendor to produce certain clean transactions, and for the clients to not be 
able to produce the transactions THEY need in a compliant form.  Ditto for 
clearinghouses.

>From the usability side, a clean test that does not have the HIPAAmetrics type 
of data analysis, still needs to be reviewed by an expert to determine 
whether the test is sufficient in volume and quality, and to determine what 
is it that was actually tested.  Did the test include sufficient variety of 
business scenarios to make it relevant?  Were those scenarios complete?  Is 
the volume of the sample for each scenario big enough?  These are some of the 
sort of things the EDI folks at the payer's end have been looking in 
provider's tests.  Using an EDI test tool may tell you that the EDI data does 
not contain errors, but that is only a small part of the answer.  Most 
translators can do that too, so, if that is all the EDI test tool does, it is 
not much better than the translator rules.

The Claredi HIPAAmetrics answer these questions.  The payer can then define 
what are the required parameters.  Things as simple as saying "we need to see 
95% clean claims in a test containing at least 200 claims" can be answered by 
HIPAAmetrics and the Claredi certification, and cannot be answered by any of 
the other test tools in the market today, without human intervention.

I hope this email sheds some more light on these concepts.  I have been 
talking about this for almost two years.  It is (washed out) in the SNIP 
white paper. And yet, the concept is still a mystery to most, even a lot of 
experts.  The reality is that it will take a while to grasp the concept.  
Those of us that have done extensive testing of healthcare EDI before HIPAA 
(NSF, UB92, etc.) understand the concept easily.  For some reason the 
traditional non-healthcare EDI community (X12) has more difficulty 
understanding the concept.

Now let me add another important difference between the "test results" and the 
HIPAAmetrics.  The test results, by their nature, will contain PHI.  If there 
are any errors, the PHI will most likely be part of the error message or 
description.  Claredi's test results also contain PHI.  So we have a bold 
statement on the test results report saying that they must be handled 
according to the appropriate privacy and security procedures.  And this in 
spite or requiring a Business Associate Agreement with all our customers.  
The HIPAAmetrics reports, on the other hand, do NOT contain PHI and thus are 
shareable with the world.  In fact, in a few weeks we will be releasing the 
HIPAAmetrics results as part of the Claredi directory describing the details 
of each certification obtained through Claredi.

There is a concept of "community" testing that is being touted lately.  The 
problem that I see is that if the "community" is comprised by more than one 
payer (e.g. several payers sharing the results of all the providers in the 
community) then all the payers in the "community" are seeing the PHI from all 
the testers.  Unless the provider creates a separate test account for each 
one of the payers in the "community" and then sends the appropriate test 
files to each one of those accounts, so each of the payers will only see the 
test files that corresponds to that payer, there is a problem...  All the 
payers in the "community" will be seeing (through the "community" testing 
center) the PHI that belongs to the other payers.  As far as I can tell this 
is not kosher for healthcare privacy reasons.  Even if the provider has a 
"business associate agreement" with the community test center, and a BAA with 
each one of the payers, and each payer has a BAA with each one of the payers 
in the "community", I still think that the "minimum necessary" rules would 
preclude this sort of bulleting board sharing of PHI within the testing 
"community".

Yes, you guessed right.  If instead of sharing the test results, like these 
"communities" do today, they were to share the HIPAAmetrics certification, 
the PHI would never be shared.

Some people have seen the vision.  We have offered it to other testing vendors 
and translator vendors.  Some have expressed interest, others are not 
interested. Clearly we need to do a lot more education on this topic.  The 
American Hospital Association understood the concept and is now endorsing 
Claredi as the exclusive HIPAA transactions testing and certification service 
to their members.

So, going back to Patrice's case study, you cannot compare testing and 
certification if the only thing you look at is testing.  Testing your own 
transactions is important.  Some people will focus only on testing.  That is 
just fine, as there are many different approaches to putting a trading 
partner into production.  We believe out test tools are better than our 
competitors', but I am sure they believe theirs are better.  But testing is 
only part of the picture.  To establish a business relationship for HIPAA 
transactions, you must go beyond compliance testing.  Knowing that the data 
you receive is clean is important, but you must also know what is contained 
(from the business perspective) in that clean data stream.  Only Claredi 
gives you an automated way to determine that.  We call it HIPAAmetrics 
Certification.  For the last 20 years we have done without it, and we have 
been happy with just test tools.  Some people are just happy with the current 
tools, other people want to explore a more automated way to put a trading 
partner into production.  This country was built on freedom of choice.  No 
need to put someone else down just because you don't like what they have to 
offer.

If you want to see it in action, give me a call and I can show it to you.  It 
is really simple...

Kepa



On Saturday 23 November 2002 12:54 pm, [EMAIL PROTECTED] wrote:
> For the benefit of those who could not make it to the WEDI SNIP HIPAA 
IMPLEMENTATION SUMMIT held in Phoenix from Nov 18 - 21, 2002.
> 
> There was a session on Nov 20th between 9.45 AM - 11.45 AM on the topic 
"Validation or Certification", moderated by Patrice Thaler of Allina Health 
System and Mike Ubl of Blue Cross of Minnesota and the main participants 
being three vendors of testing tools represented by their big chiefs... 
Namely Sunny Singh from EDIFECS, ED Hafner from Foresight Corp, Kepa Zubeldia 
from Claredi, presenting a case study of three vendors.
> 
> First of my sincere thanks to Patrice, Mike and whoever helped in this case 
study, besides the vendors.
> 
> Being the author of the vendor-neutral statement in the Testing and 
Certification White Paper (Hey John, why is my name missing in the 
acknowledgements section? ;-)), and being the first (one of the?) to suggest 
the formation of Vendor (of testing tools AND TRANSLATORS) Consortium in one 
of the teleconference calls in May/June of 2002, I am really glad that this 
case study was done.  It is the first step in that direction of forming the 
vendor consortium to help the CEs be able to pick and choose any one of them 
vendors, for they ALL have to offer the same basic VALIDATION (not 
certification) capability by coming up with the same ERROR messages and 
similar warning messages for the same test files.  This case study proved 
just that point and am truly delighted and am happy to share it with one and 
all.
> 
> One more interesting thing and I would like Kepa Z to clarify for all of 
us...
> 
> On page 3(of the handout we were given there), on a "Claredi - Passed 
Claim", there is a heavy disclaimer in BOLD, which reads "This is not a 
certification or verification of HIPAA compliance  Summary of test report for 
confidential use by Claredi customer only".  I know and agree that it is NOT 
CERTIFICATION.   I can also understand why the disclaimer is there.  But Kepa 
should explain to us why he touts Certification with an evangelical pitch and 
then this disclaimer... saying it is NOT CERTIFICATION!
> 
> Another important point, as suggested by Larry Watkins, as a member of the 
X12N committee (correct me Larry if I am not saying this right), that the 
discrepancies in the standards are not that many and there is a very good 
process to address them.  I can't agree more.
> 
> The bottom line is, it is Testing, Verification and Validation and NOT 
CERTIFICATION.  End of discussion.  I have been saying this all along and one 
more time won't hurt ;-).  That said, I really do believe all the vendors, 
have a lot to offer in the validation part of transaction testing.  I have no 
bias or preference towards one vendor.  The CEs (or their testing people) 
should make the choice.  Don't let anyone fool you that it is soooooooo big 
and only I (my org) can lead you(r org) to salvation ;-).
> 
> Accreditation, creating entities to certify and the process of certification 
is a long one and can be done, but vendor consortium will eliminate the need 
for certification all together, IMHO.
> 
> This message is not intended to be against any individual or a flame and 
rake up any unnecessary things but to share the joy of possible formation of 
vendor consortium and also request the translator vendors to join in the 
vendor consortium.  This will really help in reducing the chaos and eliminate 
the obfuscation and confusion that is there and being created...  We don't 
have time and anybody who can chip in to make things simple to whatever 
extent should be commended... like Patrice and Mike and all those who 
participated in the case study.  In Patrice's words, paraphrased, "It is easy 
and you can do it, and I have done it".
> 
> Thanks for reading up to this point.
> 
> Best Regards,  --Rama.
> 
> ---
> The WEDI SNIP listserv to which you are subscribed is not moderated. The 
discussions on this listserv therefore represent the views of the individual 
participants, and do not necessarily represent the views of the WEDI Board of 
Directors nor WEDI SNIP. If you wish to receive an official opinion, post 
your question to the WEDI SNIP Issues Database at 
http://snip.wedi.org/tracking/.   These listservs should not be used for 
commercial marketing purposes or discussion of specific vendor products and 
services.  They also are not intended to be used as a forum for personal 
disagreements or unprofessional communication at any time.
> 
> You are currently subscribed to wedi-testing as: [EMAIL PROTECTED]
> To unsubscribe from this list, go to the Subscribe/Unsubscribe form at 
http://subscribe.wedi.org or send a blank email to 
[EMAIL PROTECTED]
> If you need to unsubscribe but your current email address is not the same as 
the address subscribed to the list, please use the Subscribe/Unsubscribe form 
at http://subscribe.wedi.org
> 
> 

-- 
This email contains confidential information intended only for the named 
addressee(s). Any use, distribution, copying or disclosure by any other 
person is strictly prohibited.


---
The WEDI SNIP listserv to which you are subscribed is not moderated. The discussions 
on this listserv therefore represent the views of the individual participants, and do 
not necessarily represent the views of the WEDI Board of Directors nor WEDI SNIP. If 
you wish to receive an official opinion, post your question to the WEDI SNIP Issues 
Database at http://snip.wedi.org/tracking/.   These listservs should not be used for 
commercial marketing purposes or discussion of specific vendor products and services.  
They also are not intended to be used as a forum for personal disagreements or 
unprofessional communication at any time.

You are currently subscribed to wedi-testing as: [email protected]
To unsubscribe from this list, go to the Subscribe/Unsubscribe form at 
http://subscribe.wedi.org or send a blank email to 
[EMAIL PROTECTED]
If you need to unsubscribe but your current email address is not the same as the 
address subscribed to the list, please use the Subscribe/Unsubscribe form at 
http://subscribe.wedi.org

Reply via email to