Statement To

DEPARTMENT OF HEALTH AND HUMAN SERVICES

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS

SUBCOMMITTEE ON STANDARDS AND SECURITY

Health Insurance Portability and Accountability Act of 1996 (Public Law 104-191)

Early Implementation Experiences with HIPAA Transactions

February 1, 2001

By: Donald L. Bechtel
Strategic Advisor for
Healthcare Data Exchange Corporation (HDX), a Siemens Company

Representing
HDX and Siemens Health Services Corporation
and
The Workgroup for Electronic Data Interchange
Strategic National Implementation Process (WEDI SNIP)

Mr. Chairman and members of the committee, I am Don Bechtel, an Advisory Analyst for HDX, a wholly owned subsidiary of Siemens Health Services Corporation, where I am currently focused on strategic issues related to HIPAA compliance and the implementation of new requirements for Administrative Simplification. In addition to my HDX responsibilities, I co-chair the Eligibility Work Group within the Insurance Subcommittee of the ANSI Accredited Standards Committee X12, I co-chair the Transactions Work Group within WEDI SNIP, and I am a board member of AFEHCT (Association For Electronic Health Care Transactions). On behalf of the WEDI SNIP co-chairs Christine Stahlecker, Rob Tennant, and Larry Watkins I want to thank you for the opportunity to testify today on issues related to our analysis and early implementation experiences with using the HIPAA X12 transaction standards.

WEDI SNIP began its work to evaluate the transactions in June 2000, starting from a number of issues that were brought forward at the first WEDI SNIP Conference by the WEDI members and several AFEHCT white papers. During this meeting we divided our work effort into the five sub work groups which were described to you by Lee Barrett during his recent testimony in November. Today I’d like to share with you some of the issues that we have begun to identify.

First, I would like to point out that the goal of SNIP is to resolve implementation issues related to the implementation requirements set forth in the law, regulations, and implementation guides. We are not attempting to discuss or recommend changes to the HIPAA regulations. Rather, we will try to identify work around solutions or best practices that might mitigate a problem and allow implementation to proceed. That being said, we are hearing issues that may require changes to the regulations, which are being noted and passed to the appropriate DSMO’s, the DHHS, and the WEDI HIPAA Success Task Group.

Sequencing

One of the objectives of WEDI SNIP is to promote a deployment schedule for an orderly sequenced implementation of the transactions. Our concern is that without such a national plan, each covered entity will work to their own schedule or a local/regional schedule that does not align with neighboring regions or States. This approach would put most entities in a position of having to be prepared to implement all of the transactions at the same time, but not for all trading partners, because each trading partner could potentially take a different approach. From this we would begin to experience implementation difficulties resulting from a non-orderly and confusing implementation schedule that could be dictated to some organizations differently by each trading partner.

This scenario would put many organizations at risk of not having enough resources to focus on the implementation tasks that will be created. We know that some transactions will need more time to develop than others, and some will need more time to implement than others. Therefore, we’re trying to find a sequence approach that addresses these concerns and is logical and effective for most organizations. We also considered return on investment, trying to put those transactions that offer the biggest return on investment up front to help fund the development of the other transactions.

We recently submitted a sequencing proposal to the industry via our website and email announcements asking organizations to review the proposal and comment on their ability to follow the schedule. We also asked for other information, such as expected costs to implement, expected resource commitments, implementation strategies, and implementation barriers that we might be able to address within WEDI SNIP to keep the implementation process moving forward.

The results of our survey are still being collected and evaluated. But, here is what we’ve heard so far through some very consistent messages:

• The three most critical implementation barriers are cost, limited time, and elimination of local codes. Other issues that ranked very high were NDC codes and new transaction content.
• In terms of sequencing, our fears that many organizations may have different plans proved well-founded. Of those organizations responding, 15% said they would follow the WEDI SNIP proposal, 40% said they were still evaluating their plans and couldn’t comment on a sequence, and the remaining 45% offered different alternatives, of which very few matched anyone else. From this we only can hope that WEDI SNIP will convince the 40% that are undecided to follow our plans. As we will evaluate the comments we will try to determine if there are modifications that we can make to the plan that will make it more appealing to the majority of organizations. We expect to publish our findings this month.
• Many organizations are still doing their gap analysis. Many respondents indicated that their implementation plans were not firm, that they were still doing their analysis, and that when it was complete they might need to make changes. Many other respondents indicated that they couldn’t give us information, because they were just beginning to perform their gap analysis.
• There is concern that the staggered release of final rules will impact work already completed for HIPAA compliance, forcing them to go back into their applications to make additional changes, thus increasing the cost and stability of their enhancements. Some organizations felt they would rather wait until all final rules were published before they began to make their enhancements.
• A number of organizations commented that the time to complete the development, testing, and implementation was too short. They felt that the scope and complexity of the changes are very large, and they are concerned that, if rushed to complete this work, they could introduce potential problems to the health community related to providing good and timely service to the transactions being processed.
• A number of organizations reported that they will need to acquire new systems to implement the HIPAA transactions and associated changes for code sets and data content. They felt that the time to go through the acquisition and implementation process will take more time than is available to complete.
• There were many other issues identified, and many of them are not issues that can be addressed by WEDI SNIP. These are being compiled into a report and will be given to the WEDI HIPAA Success Task Group to review and consider for their input to the Department of Health and Human Services and to this committee.

Data Review – transaction inconsistencies and ambiguities

To date, this group has not identified many issues. I believe there are two primary reasons for this. First, we are very early in the implementation, and many, if not most, organizations are still in the gap analysis phase of their own evaluations. This can be deduced from the survey results of our sequencing proposal. WEDI SNIP relies on volunteers to come forward with issues they’ve uncovered during their own analysis. We are not sponsoring this analysis as a standalone effort. Therefore, we are not hearing much at this time.

The second reason few issues have been brought forward is that those who are doing this analysis are the same people who are deeply involved with X12. Many of the implementation problems that are being found are in the implementation guides and are currently being addressed by X12N via their errata process. These issues are not coming to WEDI SNIP for work around or best practice solutions – rather, they are seeking fixes to the implementation guides. Those issues that are not fixed, I believe will eventually find their way to this work group.

Following are issues that have been identified in the Data Review sub work group:

• The X12 837 claims standards do not crosswalk well to the UB92 or NSF formats. The issues are related to details like data lengths and new identifiers. We are finding that many organizations will need to significantly modify their systems to collect and handle new data requirements. For example, currently the industry uses one facility code for place of service, as assigned by the health plan. Under HIPAA, a health plan is required to support 4 identifiers: health plan, provider, patient (as assigned by provider or a national identifier), and claim.
• We are just beginning to identify transaction inconsistencies. The first issue identified is the use of the X12 837 dental claim transaction to get predetermination of benefits. For other lines of health business, this is done using the X12 270/271 health care eligibility transaction or the X12 278 Authorization and Referral transaction, if the provider is looking for an authorization. This becomes confusing for the implementers since it is unclear which transactions to use when. Further, this could lead to other issues of consistency and begin to create unnecessary transaction maintenance resulting in two transaction types supporting the same functions. This kind of issue can be resolved with a best practice recommendation, and suggestions to the DSMO (X12N) to correct the problem in future versions.
• We have also identified a problem with the X12 278 Authorization and Referral transaction that could be a “show stopper” for use with Medicaid programs. In this situation, the transaction does not support the clinical information that is required by many States’ Medicaid programs to provide authorization for services. This will probably prevent those Medicaid programs from implementing the 278 transaction, and force them to continue to use their existing methods of processing. There appears to be no remedy for this problem without changing the actual transaction. This is not just an implementation guide issue, it is a transaction limitation that must be corrected. Such a change will likely take a year to move through the X12 process.

Data gap analysis is also being done by AFEHCT in the Administrative Simplification Print Image Research Effort (ASPIRE) project. I believe you’ll be hearing a report from them today as well.

Translations issues

Here are issues that have been identified in the WEDI SNIP Translations sub work group:

• NDC codes do not cross walk well to the HCPCS J-Codes. While it is possible to translate from the NDC code to the J-Code, this is not possible from the J-Code to the NDC, because there are many possible NDC codes that would be described by a J-Code. This is sometimes referred to as a “many to one” translation problem.

This particular issue becomes a problem for those organizations that cannot convert their systems internally to support the NDC codes in place of the J-Codes. Some organizations are considering the notion of converting the NDC codes to the J-Code equivalents, so they can perform adjudication processing. However, when they must report back on the remittance transaction information about a drug on the claim, it will not be possible to convert the J-Code back to the NDC. The translation engine won’t be able to determine which NDC to use.

An alternative to the J-Code to NDC translation issue is to store the inbound transaction and re-associate the NDC to the J-Code based on some location reference. However, I am skeptical that this re-association can always be done due to the complexities. This is still being evaluated.

• Mapping Taxonomy codes to the HCFA specialty codes causes a similar problem. The new Taxonomy code list is much more granular than the HCFA codes, and, if translation were required for interim processing during the implementation transition period of HIPAA, there could be similar results. One could translate from the Taxonomy codes to the older specialty codes, but it is not possible to translate from the specialty to the Taxonomy codes.

An important option to avoid these transitional problems is to implement such code sets at the end of the implementation phase. In other words, the X12 transactions would be implemented using the J-Codes and Specialty codes we have today, which are currently supported in the X12 transactions. Then, following the conversion efforts for the format and other data content, the new code sets would be implemented. The resistance to this solution is that it adds another step to the implementation process, and therefore to the development efforts.

Testing issues

The WEDI SNIP Testing work group is looking at ways to improve our implementation validation process. A major concern in the industry is centered on the fact that it takes a lot of time and testing to implement a new transaction between trading partners. With the short window of time available for the industry to complete the system development and implementation activities for the 9 transactions (multiplied by the number of trading partners for each), this becomes a time critical task. WEDI SNIP has concluded that the mandated timeframes will not be met unless we can find a more efficient way to conduct these implementations. This may be a topic for another time.

Business issues

The number one concern that the WEDI SNIP Business Issues work group has identified is in the interpretation of the regulations. Although there needs to be a place to take interpretation issues for quick resolution, such a formal mechanism does not exist today. Many times we simply ask someone from HCFA or HHS who is working in the work group for an answer. Official answers take more time. We believe the number of issues will begin to expand rapidly as organizations begin to analyze their requirements and determine their data gaps and questions. Specific issues where clarification is needed include the role of clearinghouses, coordination of benefits, permissible implementations of direct data entry solutions, and more.

Code set deficiencies This sub work group has written a draft white paper entitled “Data and Code Set Compliance”, which outlines a number of code issues that are being raised. I have included an electronic copy of this draft white paper with my testimony. Some examples of identified issues:

• Implementation issues related to the use of NDC codes and the impact these have on provider and payer systems when processing with these codes. This problem is reviewed in terms of the business issues created by this change. WEDI SNIP proposes some suggestions on how organizations might proceed with their implementation. More will be heard on this issue in later presentations today, but much of the background material can be found in this white paper.
• Business impacts of eliminating local procedure, diagnosis, modifier, and other trading partner specific codes. This paper reviews how providers and payers might mitigate those problems, and other issues. Some examples are anesthesia services, dental codes, DME, home infusion, and others.
• Other codes, which are not governed by the DSMO’s, may require mapping to enable payers to utilize them for processing purposes. This paper considers how these mappings might be developed into best practices for greater consistency across the industry. Examples are taxonomy and adjudication reason codes.
• Version control of data standards and HIPAA implementation guides need to be managed. The implementation guide requires that the claim be filed with the medical code structure in effect at the time services were rendered, while other code sets are required to follow the standard in effect at the time the transaction is generated.

Another hot topic within this sub work group is the use of alternative sources of transaction data for the named business transactions -- specifically for direct data entry devices and browser-based applications that might be operated by health plans and clearinghouses. This white paper is finding a number of issues related to data content, including explicit vs. derived data, stringent vs. reasonable field length requirements. (For example, phone numbers within X12 standards are 80 characters in length, but in most case 20 characters is more than enough. I have included an electronic copy of this draft white paper with my testimony for your convenience.

Batch vs. Real Time. The eligibility X12 270/271 transactions will require many health plans to perform a significant development project to support the functions -- in particular to support batch processing. In today’s environment, many health plans process eligibility either by phone or voice response systems, which is a real-time process. However, the implementation guide requires the support of batch transactions, which in these situations is not typically supported. (This is especially true for State Medicaid programs.)

Another issue raised concerns when an X12 835 transaction must be sent electronically. This is not made clear in the adopted implementation guide. For example, must a provider who is only receiving paper remittance advice today only take the electronic form under HIPAA? Many believe that paper remittances will still be appropriate, but WEDI SNIP believes clarification, such as is discussed above, is necessary on this issue.

Concluding Remarks:

Mr. Chairman and members of the committee, these are the primary issues WEDI SNIP has uncovered to date. While we believe we have only just begun the process of identifying the implementation issues, we are also committed to the consensus process to bring resolution to these issues whenever possible.

This concludes my statement and as always, I am available for follow up questions at any time. Again, thank you for this opportunity to share our experiences.