 |
 |
|
Rapid Responses to:
|
|
Electronic letters published:
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif){kind=icon}
Important Considerations for a National Health Information Network
Speeding the Adoption of a National Health Information Network
Interoperability is a prerequisite to functionality
A bit too simple
|
|
|||
|
Rainu Kaushal, MD MPH Brigham and Women's Hospital, Harvard Medical School, David W. Bates, David Blumenthal
Send rapid response to journal:
rkaushal{at}partners.org Rainu Kaushal, et al.
|
Each letter in response to our article, “The Costs of a National Health Information Network” (NHIN) raises interesting and thoughtful points. Dolin and Wiesenthal raise concerns that we separated functionality from interoperability costs while Kretz raises concerns that we applied the Santa Barbara County Data Exchange (SBCDE) as a national model of interoperability. We agree that functionality and interoperability go hand in hand, and did not mean to imply that one could be implemented without the other. However, to model costs, it was conceptually and methodologically easier to approach each set of costs separately. The nation as a whole has not reached consensus about how to approach interoperability. SBCDE is one of the most successful early examples of broad clinical data exchange, and we were able to access costs for it, making it a reasonable choice. This is, however, only one of the possible architectures that can be used. We agree with Kretz that it may not be the best approach ultimately, especially as the data exchanged were not coded thereby limiting benefits.(1) Kretz further states that the VISTA system should be an significant consideration in developing a model NHIN. The VISTA system has provided tremendous benefits to its population (2), but we do not believe that its interoperability approach will be the most useful outside this closed system. Kretz questioned our assertion that an important financial barrier to wider adoption of HIT is that benefits accrue to parties other than those bearing costs. From the policy perspective, who gets the benefits versus who must pay is critical, as addressed by several recent studies.(1,3) It is not simply whether the party bearing the costs accrues benefits greater than the costs but over what period of time and whether there is equitable distribution of costs to those who benefit financially. This is particularly essential for small physician practices as highlighted by Doherty. We agree that financial savings from HIT will be substantial.(1,4) Finally, we completely agree with Rothkopf and Jackson about the importance of appropriate user interfaces. For successful implementation of HIT, key success factors include addressing cultural issues, speed of the system, appropriate training, technical support, and seamless connectivity.(5) All these factors much be addressed to realize the dream of widespread, interconnected HIT and the attendant benefits in safety, quality, and efficiency. The recent experience of tremendous health data problems for Hurricane Katrina victims highlights the urgency of implementing HIT in the U.S. Rainu Kaushal, MD MPH David W. Bates, MD MSc David Blumenthal, MD MPP References 1. Walker J, Pan E, Johnston D, Adler-Milstein J, Bates DW, Middleton B. The Value of Health Care Information Exchange and Interoperability. Health Affairs. Web Exclusive Jan 19, 2005. 2. Jha AK,Perlin JB, Kizer KW, Dudley RA. Effect of the Transformation of the Veterans Affairs Health Care System on the Quality of Care. N Engl J Med 2003;348(22):2218-27. 3. Kaushal R, Bates DW, Poon EG, Jha AK, Blumenthal D. Functional Gaps In Attaining A National Health Information Network. Health Affairs. 2005;24(5):1281-9. 4. Hillestad R, Bigelow J, Bower A, Girosi F, Meili R, Scoville R, Taylor R. Can Electronic Medical Record Systems Transform Health Care? Potential Health Benefits, Savings, And Costs. Health Affairs. 2005;24(5):1103-17. 5. Bates, DW. Physicians And Ambulatory Electronic Health Records. Health Affairs. 2005;24(5):1180-9. Conflict of Interest:None declared |
|||
|
|
|||
|
Robert B. Doherty SVP, Department of Governmental Affairs and Public Policy, The American College of Physicians
Send rapid response to journal:
bdoherty{at}acponline.org Robert B. Doherty
|
We read with interest the study by Kaushal(1) and colleagues and the editorial by Basch(2) regarding the costs of rolling out an interoperable National Health Information Network (NHIN). As Congress considers financial incentives to speed the adoption of health information technology, they must pay attention to the findings and conclusions of the study by Kaushal and colleagues. The most striking of which is the $48 billion annual operating costs of maintaining a NHIN. These on-going costs and the lack of financial return on investment to providers is perhaps the single greatest barrier to attract small physician practices into the information technology age. As Basch correctly points out, it’s this “misalignment of incentives” that translates into vast savings for payers and leaves small physician practices holding the bag. Sensing the need for federal intervention, Kaushal and colleagues suggest various policy options to encourage the widespread adoption of a NHIN. They single-out legislation by Senators Frist and Clinton and Representatives Murphy and Kennedy as a remedy. Unfortunately, these bills will do little to attract those small physician practices trying to justify the business case for electronic health record (EHR) adoption into their practices. The American College of Physicians has worked closely with Congressional leaders to develop incentives aimed at small physician practices by proposing building into the Medicare physician payment system an add-on code for office visits and other E/M services when using an EHR. Senators Stabenow and Snowe and Representatives McHugh and Gonzalez have introduced similarly bipartisan bills that would do just that.(3) By recognizing the substantial financial savings likely to result from a reduction in medical errors and duplicative tests, policymakers should strategically target on-going incentives toward small physician practices to help offset the daily operating costs of an EHR system. This return on investment will be well worth it. (1) Kaushal R, Bluemenathal D, Poon EG, Jha AK, Franz C, Middleton B, et al. The Costs of a National Health Information Network. Ann Intern Med. 2005; 143:165-173. (2) Basch P, Electronic Health Records and the National Health Information Network: Affordable, Adoptable, and Ready for Prime Time? Ann Intern Med. 2005; 143:227-28. (3) S. 1227, “the Health Information Technology Act of 2005,” and H.R. 747, the National Health Information Incentive Act of 2005.” Conflict of Interest:None declared |
|||
|
|
|||
|
Robert H Dolin, M.D. Kaiser Permanente, Andrew M. Wiesenthal, MD
Send rapid response to journal:
Robert.H.Dolin{at}kp.org Robert H Dolin, et al.
|
To the Editor, We agree with Kaushal, et al's conclusion, that the cost of building a NHIN is underestimated, but believe that the separation of costs into functionality and interoperability may perpetuate the lack of recognition that the latter is critical in achieving the former. Interoperability is an essential and foundational component of all the functions (results viewing, EHR, CPOE, claims submission, eligibility verification, patient communication, electronic prescriptions) that are described. This misunderstanding also exists in the HL7 Electronic Health Record Minimal Function Set specifications [1], where for instance, the management of allergy and medication lists are deemed to be essential, while adoption of interchange and terminology standards are deemed to be future functions. Further, we believe that interoperability standards are best broken down into technical and semantic components (such as mapping from a local terminology to a standard terminology), as these require different skill sets to implement, and the cost of the latter is the component most often underestimated. Recognizing that functional requirements have a dependency on interoperability standards will enable the United States to focus on building the necessary foundation. Good standards take time to build. Just as three women can't pool their resources to deliver one baby in three months, there are constraints on the rate at which a standard can be developed, regardless of expenditures. Good standards do exist today (such as SNOMED CT [2] and HL7 V3 including the HL7 Clinical Document Architecture [3]), as do large scale successful implementations [4]. A focus on foundational components, along with a push towards adoption of these components, will move the United States towards a NHIN with a higher likelihood of success and a reduced probability of wasting resources. REFERENCES [1] Health Level 7. http://www.hl7.org [2] SNOMED Clinical Terms. http://www.snomed.org [3] HL7 Clinical Document Architecture, Release 2.0. ANSI-approved HL7 Standard; May 2005. Ann Arbor, Mich.: Health Level Seven, Inc., 2005. [4] Dolin RH, Mattison JE, Cohn S, Campbell KE, Wiesenthal AM, Hochhalter B, et al. Kaiser Permanente's Convergent Medical Terminology. Medinfo. 2004;11(Pt 1):346-50. (PMID: 15360832). Conflict of Interest:None declared |
|||
|
|
|||
|
James M Kretz, B.A.,MA,ABD Reproductive Endocrinology Software
Send rapid response to journal:
jimkretz{at}speakeasy.net James M Kretz
|
I was dismayed by the superficiality underpinning much of “The Costs of a National Health Information Network.” Adopting wholesale the Santa Barbara County Data Exchange (SBCDE) as a working model for the nation because the SBCDE could provide good cost data seems something like the drunk looking for his keys not near where he dropped them but beneath the street light because that is where light is. Using the SBCDE as a template lead the authors directly to the assertion that “Each data supplier requires a dedicated server to store aggregated data for exchange…” (p167) With that server comes a number associated expenses including, but not limited to, staff and equipment to perform routine data backups, staff or contracted services to keep the server software updated, creation of secure physical conditions to house the server, etc. While that was a design decision made in Santa Barbara County it is not a requirement to exchange data. If the design had included a secure Web based approach, the individual provider locations would not be saddled with those costs and economies of scale achievable, for example, by having data maintained and backup managed at a central site. By itself, this would drive estimated startup and ongoing maintenance costs down. Nation wide training costs were virtually ignored. Again relying upon the SBCDE experience, the authors conclude “Training costs were relatively small since the system relies on user-friendly browsers.” (p167). Ignored is the training costs associated with how users are to interact with the material and the design of the software presented within the browser, not the browser itself. Browser based or not, application user interfaces always require training: well written ones take less and more poorly designed ones more. Appendix Table 2 indicates that one to four-physician offices would require $600 of training. Unless the interoperability software were exquisitely designed, training costs seem to be underestimated. Throughout the article, various discursive assertions are offered as justification or explanation. The authors claim that “A major barrier to widespread IT adoption is that costs are generally incurred by a few entities, while benefits accrue to many.” (p171). The first question in this context might more appropriately be couched in terms of whether sufficient benefit accrues to those who bear the cost. The authors claim “high rate of automated claims submissions are largely due to HIPAA.” (p171). HIPAA certainly promoted standards, but well before its passage in 1996, health insurance claims clearing houses were already $1 billion per year businesses, the impetus for which was simply that third party payers would adjudicate claims within days instead of weeks. Finally, the paper suggests adoption of the VA’s VISTA system, while in the public domain is written in an older programming language and lacks billing and claims functionalities. “Older” in this context is hardly a four letter word and it is instructive to note that MUMPS or its variants known as M-Technology, Magic, and MIIS is the language used by the Department of Defense, the Indian Health Service, Partners Healthcare, IDX, and RBOC in health care and is still widely used in the banking industry. Some VA installations do produce electronic claims that are processed and disturbed by claims clearing houses and in any event it could be made part of VISTA for something less than $156 billion. Conflict of Interest:None declared |
|||
