Health Affairs, 24, no. 5 (2005): 1281-1289 doi: 10.1377/hlthaff.24.5.1281 © 2005 by Project HOPE	New Online   Senate Health Reform Bill   Rewarding Providers   Public Option Policy Brief   Health Reform & Abortion   Delivery System Reform

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Kaushal, R. Search for Related Content PubMed PubMed Citation Articles by Kaushal, R. Related Collections Health Reform Quality Of Care Research And Technology Health Spending Politics Health Information Technology

Health Tracking

TRENDS

Functional Gaps In Attaining A National Health Information Network

Abstract

 
We describe the health information technology (HIT) required for a model of a national health information network (NHIN). Specifically, we define the functional HIT capabilities of an attainable NHIN and determine the gap between the current state of HIT and its likely state in five years, given current trends of adoption. Administrative functionalities will be more prevalent than clinical functionalities in five years. Smaller stakeholders, such as home health care agencies, will lag in adoption. Policy changes, such as financial incentives to HIT end users or regulatory measures, may help accelerate the adoption of HIT for a model NHIN.

For this paper, we aimed to determine the HIT functionalities required for an attainable model NHIN in five years. We first sought to define the gap between the HIT functionalities of this model NHIN and the estimated HIT functionalities in five years if current national adoption trends continue. Unfortunately, resource constraints precluded our defining all aspects of a model NHIN—most importantly, interoperability. Nevertheless, the insights gained in defining needed functionalities, current adoption rates, and future rates and in achieving consensus on a best-attainable model should be useful in other modeling exercises.

Study Methods

Top Study Methods Study Results Discussion NOTES

 
We convened an expert panel to develop a model of the HIT functionalities required for an NHIN. After presenting data from interviews of HIT stakeholders in two markets, we then asked the panel to estimate the current national state of HIT functionality and its likely state in five years, given current rates of adoption. We then determined the gap between the two estimates.

NHIN model. The expert panel (acknowledged at the end of this paper) developed a model of an NHIN that would be attainable in five years, with a focus on required HIT functionalities. This model incorporated temporal as well as reasonable financial, personnel, and technical constraints. At the time of our study, the U.S. Department of Health and Human Services (HHS) had not yet developed a national vision for an NHIN.

The panel then identified the most important stakeholders of an NHIN, defined as those who were most likely to need or benefit from electronic health information as well as most likely to directly bear primary financial responsibility for HIT. Although consumers are perhaps the most critical stakeholder in an NHIN, they were excluded from the analysis because they do not make primary financial decisions for HIT acquisitions. For each stakeholder, the panel defined the critical functional HIT capabilities that were achievable in five years. The panel focused on capabilities that were essential to providing high-quality care or improving its efficiency. They then described each element within the functional HIT capabilities.

Interviews and data sharing. The expert panel achieved consensus through a modified Delphi approach, which consisted of iterative rounds of interviewing followed by a sharing of the group’s opinions with participants.⁵ The group achieved a high level of consensus about HIT functionalities for a model NHIN.

Using these functionalities as a framework, we conducted a study of stakeholders in Boston and Denver to determine present levels of HIT functionality in these two markets; those results are presented elsewhere.⁶

The NHIN expert panel was then reconvened to present data from Denver and Boston as well as its estimates of present levels of HIT functionality adoption nationwide. We also presented available external data regarding HIT adoption costs and rates.⁷ After the presentations, the expert panel reached consensus, using the modified Delphi approach, about the current level of adoption for each functional HIT element of the NHIN nationally and the expected level of HIT adoption in five years, assuming that the United States remains on its current trajectory of HIT spending. There was a wider range of opinions during this process than during the discussion of a model NHIN. After discussion, each member independently provided a final estimate for each stakeholder’s HIT functionality. The reported estimates are the mean of the responses by the expert panel.

Estimates of HIT gaps. We then determined the functional HIT gaps nationwide between a model NHIN and the expected state in five years on the current trajectory of U.S. IT adoption. We assumed that there would be 100 percent adoption of each functional HIT element in the model NHIN. To determine the gap, we subtracted the expected percentage of HIT functionality adoption by each stakeholder in five years from the model’s 100 percent level of adoption. An institutional review board approved this study.

Study Results

Top Study Methods Study Results Discussion NOTES

 
Critical stakeholders included physicians’ offices, hospitals, skilled nursing facilities (SNFs), home health agencies, clinical laboratories, and pharmacies. Payers were initially included and then eventually excluded, because most already have the critical functionalities assigned by the expert panel, namely advanced electronic capabilities for claim and eligibility checks. Patients also were excluded because they do not bear the primary financial responsibility for the development of HIT capabilities. However, the expert panel incorporated the needs and perspectives of patients in the development of key functionalities by focusing on those functionalities that improve the safety, quality, and efficiency of care.

We segmented physicians’ offices into small practices (1–4 providers), medium-size practices (5–20 providers), and large practices (more than 20 providers), because their expected rates of HIT adoption were different. Smaller physician offices were expected to adopt at slower rates than larger practices because of greater financial pressures. Similarly, we divided hospitals into those with fewer or more than 300 beds, with the similar expectation that larger hospitals would have greater financial resources, allowing them to adopt HIT more rapidly.

HIT functionalities of a model NHIN. The panel then identified critical functional capabilities for a model NHIN in five years: electronic result viewing (of ordered tests), electronic health records (EHRs), computerized physician order entry (CPOE), electronic claims submission, electronic eligibility verification, secure electronic patient communication, and electronic prescription acceptance by pharmacies (Exhibit 1).

For EHR functionality, the expert panel determined that requirements for hospitals and SNFs would be the same in five years. Although most institutions are purchasing vendor products that combine result viewing and EHR capabilities in one package, some institutions have implemented the two separately, so the panel considered them separately.

CPOE systems automate the ordering process, thereby ensuring standardized, legible, and complete orders. The systems used by SNFs would be simpler than those used by hospitals, allowing only medication order entry with simpler drug alerts. CPOE systems used by office practices would be similar to those in hospitals but would include some added functionalities to optimize benefits (Exhibit 1).

Another important functionality of a model NHIN would be automated claims submission and member-specific eligibility verification for all stakeholders. Automating the claims submission process allows for more rapid and accurate billing, while automating the eligibility verification process allows for more rapid pre-approval. Functionality would be the same for all stakeholders examined (Exhibit 1).

The final two areas of functionality, patient communication and e-prescribing, are not shown in Exhibit 1. Secure Web messaging, electronic health questionnaires, and electronic home monitoring would facilitate patient communication with physicians’ offices. Patients and caregivers could access customized patient education materials to tailor and monitor their therapy, as well as electronically request refills, referrals, and appointments. They would also receive electronic visit and test reminders. Finally, pharmacies would have the capability of electronically accepting and processing prescriptions.

Estimates of current functionality. After a presentation of the data from interviews in Boston and Denver, available literature, and their own initial prevalence estimates, the expert panel discussed and reached consensus about the national prevalence of each functional capability. By far, the most developed capability at present is electronic claim submission and eligibility checking. Result viewing is the most common clinical functionality, followed by EHRs and then CPOE. This is presented in more detail elsewhere.⁸

Functionality estimates at five years. Exhibits 2–6 show our expert panel’s estimation of the prevalence of critical functionalities after five years, assuming no significant changes in the current trajectory of HIT adoption. The penetration of automated claims submission in five years is projected to be greatest, with availability in approximately 86 percent of home health agencies and 99 percent of large hospitals (Exhibit 2). The penetration of automated eligibility verification is also expected to be high, ranging from 53 percent in small physician offices to 92 percent in pharmacies (Exhibit 3).

View larger version (24K): [in this window] [in a new window]   EXHIBIT 2 Estimate Of Electronic Claims Processing Functionality For Stakeholders At Present And In Five Years

View larger version (21K): [in this window] [in a new window]   EXHIBIT 3 Estimate Of Electronic Eligibility Determination Functionality For Stakeholders At Present And In Five Years

View larger version (19K): [in this window] [in a new window]   EXHIBIT 4 Estimate Of Electronic Result-Viewing Functionality For Stakeholders At Present And In Five Years

View larger version (18K): [in this window] [in a new window]   EXHIBIT 5 Estimate Of Electronic Health Record (EHR) Functionality For Stakeholders At Present And In Five Years

View larger version (17K): [in this window] [in a new window]   EXHIBIT 6 Estimate Of Computerized Physician Order Entry (CPOE) Functionality For Stakeholders At Present And In Five Years

For office practices, the next most prevalent functionality would be secure patient communication (33–46 percent; data not shown) followed by EHRs (25–38 percent, Exhibit 5) and then CPOE (21–32 percent, Exhibit 6). In contrast, CPOE was estimated to be more prevalent than EHRs in the inpatient setting. On the other hand, only 21 percent of home health agencies and 14 percent of skilled nursing facilities were expected to have an EHR system, and 14 percent of skilled nursing facilities were expected to have a CPOE system. Finally, the expert panel estimated that 58 percent of pharmacies would be able to accept electronic prescriptions in five years.

In general, large hospitals were expected to have the most clinical functionalities, followed by smaller hospitals, large office practices, medium-size office practices, and small office practices. Home health agencies and SNFs lagged further behind.

Discussion

Top Study Methods Study Results Discussion NOTES

 
Home health agencies and SNFs could have the largest gap in five years between projected levels of functionality and a model NHIN. Administrative functionalities such as claims submission and eligibility checks appear to be the furthest advanced, while clinical functionalities could lag further behind in five years. Of the clinical functionalities, result viewing could be the most advanced, followed by EHRs and CPOE in the ambulatory setting. In contrast, in the hospital setting, CPOE could be adopted more rapidly than EHRs.

Administrative functionalities could be further advanced than clinical ones because incentives for adoption of administrative capabilities are tangible and well aligned. For example, automated claims submission and eligibility verification enable institutions to eliminate costs and decrease turnaround time for bills. Also, legislative pressure from the Health Insurance Portability and Accountability Act (HIPAA) of 1996 pushed institutions to comply with HIPAA standards and to submit claims electronically.

Of the clinical functionalities, result viewing is now the most prevalent. It is easily integrated into existing work flow and provides great benefit to users. In contrast, EHRs and CPOE restructure physicians’ work flow, which generates greater resistance and thus requires careful integration into new HIT systems.⁹ EHR and CPOE systems also tend to be more costly than result-viewing systems.

Inpatient settings are more advanced in HIT than ambulatory settings. This is likely attributable to a variety of resources: financial, technological, and personnel. Also, probably because of their resource limitations, home health agencies and short-term nursing facilities lag behind office practices. For home health agencies, there are also issues of geographical dispersion, as visiting nurses travel from home to home. The remote and mobile nature of home health will require portable computers and maybe even wireless technology, making implementation incrementally more difficult. In addition, the workforce of visiting nurses may require a great deal of technological training and support.

Support for model NHIN. Approaches for speeding up the process to achieve a model NHIN include overcoming heterogeneity in rates of adoption of HIT functionalities, addressing financial concerns, developing standards to ensure interoperability of systems, and addressing issues of privacy and security.

Widespread adoption of HIT is an initial step in developing an NHIN. If large stakeholders such as hospitals continue to rapidly advance in HIT adoption and smaller stakeholders such as office practices and home health care agencies lag behind, large inequities will result. Then the quality benefits of HIT systems will be limited as patients move from highly computerized inpatient settings to paper-based outpatient settings.¹⁰ These handoff points are particularly susceptible to medical errors. Policy interventions should preferentially target adoption by smaller stakeholders to prevent the development of likely inequities. For example, smaller office practices often must bear the financial responsibility of implementing a CPOE system, although the quality and financial benefits accrue to many parties, including patients and payers.¹¹ Rural stakeholders may need focused attention to accelerate HIT adoption because of their even more limited resources.

Financial constraints. Among all stakeholders studied, financial concerns remain a powerful force in determining the degree of HIT adoption. In today’s climate of declining revenues, monetary incentives tied to quality improvement initiatives represent an important facilitator for HIT diffusion.¹² Third-party payers and purchasers who stand to benefit the most from overall savings as a result of improved health care outcomes may best provide these pay-for-performance incentives. Stakeholders with smaller operating budgets may require direct financial assistance rather than just incentives.

Need for standards. Development and adoption of standards will be another important component of efforts to create an NHIN, in conjunction with regulatory measures, to ensure eventual interoperability. For example, HIPAA has been an important catalyst for full conversion to an electronic claim submission system. Had a standard for electronic transaction not been mandated through HIPAA, paper-based claims submission would undoubtedly be far more prevalent today. Much work has recently been done to define electronic standards for communicating and interpreting health care data.¹³ Discussion is also under way to define a private-sector certification process for vendor-based EHRs.¹⁴

Finally, the development of an NHIN will require careful attention to privacy and security issues. The health care industry may be able to learn from the successes and problems of the banking industry. Discussions will need to address whether patients have single or multiple national personal health identifiers and whether they are voluntary or required.

Study limitations. This study has several limitations. The data consist of expert panel estimates rather than primary data. Furthermore, the study is incomplete, because it was restricted to the most important stakeholders and functionalities as defined by the expert panel, rather than all stakeholders and functionalities. It was beyond the scope of this study to address interoperability in addition to functionality, although interoperability will clearly confer considerable benefits.

President Bush has called for nationwide adoption of EHRs by 2014, and several HIT bills are pending in Congress. If the nation decides to move forward to adopt an NHIN, private and public efforts can accelerate adoption through ensuring widespread equitable adoption of HIT functionalities, better aligned financial incentives, adoption of standards, and protection of privacy and security.

Editor's Notes

 
Rainu Kaushal (rkaushal{at}partners.org) is a staff physician at Brigham and Women’s Hospital in Boston and assistant professor, Department of Public Health, at Cornell Medical School in New York City. David Bates is chief, General Medicine Division, at Brigham and Women’s Hospital; medical director, Clinical and Quality Analysis, at Partners HealthCare System; and a professor of medicine at Harvard Medical School. Eric Poon is an instructor in medicine, Division of General Medicine and Primary Care, at Brigham and Women’s. Ashish Jha is an assistant professor of Health Policy at Harvard School of Public Health. David Blumenthal is director of the Harvard University Program for Health Systems Improvement and the Samuel O. Thier Professor of Medicine and professor of health care at Harvard Medical School.

This work was supported by the Commonwealth Fund and the Harvard Interfaculty Program for Health Systems Improvement. The authors thank the members of the Harvard Interfaculty Program for Health Systems Improvement NHIN Working Group: Melissa Christino, Gary Fanjiang, Rushika Fernandopoulle, Calvin Franz, John Glaser, Melissa M. Honour, Gilad Kuperman, Blackford Middleton, Joseph Newhouse, and Seth Woolf. They also thank the members of the expert panel: David Brailer, Health Technology Center; Janet Corrigan, Institute of Medicine; Mark Frisse, First Consulting Group; Lucian Leape, Harvard School of Public Health; Janet Marchibroda, eHealth Initiative; Eduardo Ortiz, Agency for Healthcare Research and Quality; Joseph Scherger, University of California, San Diego; the late Elliot Stone, Massachusetts Health Data Consortium; William Yasnoff, U.S. Department of Health and Human Services; and Barry Zallen, Blue Cross Blue Shield of Massachusetts.

NOTES

Top Study Methods Study Results Discussion NOTES

 

1. See, for example, A.A. Gawande and D.W. Bates, "The Use of Information Technology in Improving Medical Performance, Part II: Physician-Support Tools," Medscape General Medicine 2, no. 1 (2000), www.medscape.com/viewarticle/408033 (27 June 2005; registration required); and A.A. Gawande and D.W. Bates, "The Use of Information Technology in Improving Medical Performance, Part III: Patient-Support Tools," Medscape General Medicine 2, no. 1 (2000), www.medscape.com/viewarticle/408035 (27 June 2005; registration required).
2. Regarding slow adoption, see J.S. Ash et al., "Computerized Physician Order Entry in U.S. Hospitals: Results of a 2002 Survey," Journal of the American Medical Informatics Association 11, no. 2 (2004): 95–99.[Abstract/Free Full Text]
3. C.J. McDonald et al., "The Regenstrief Medical Record System: A Quarter Century Experience," International Journal of Medical Informatics 54, no. 3 (1999): 225–253.[CrossRef][Web of Science][Medline]
4. D.J. Brailer et al., Moving toward Electronic Health Information Exchange: Interim Report on the Santa Barbara County Data Exchange (Oakland, Calif.: California HealthCare Foundation, 2003).
5. See, for example, R. Meijer et al., "The Use of a Modified Delphi Procedure for the Determination of Twenty-six Prognostic Factors in the Sub-Acute Stage of Stroke," International Journal of Rehabilitation Research 26, no. 4 (2003): 265–270.
6. E.G. Poon et al., "Assessing the Level of Healthcare Information Technology Adoption in the United States: A 2003 Snapshot of the Boston and Denver Markets," Proceedings from Medinfo 2004 (San Francisco: Medinfo, 2004), 1815.
7. See, for example, S.J. Wang et al., "A Cost-Benefit Analysis of Electronic Medical Records in Primary Care," American Journal of Medicine 114, no. 5 (2003): 397–403[CrossRef][Web of Science][Medline]; J. Janas III and D. Morrison, "Capital Region Healthcare: Cost and Quality Benefits," 26 August 2003, www.medicalogic.com/emr/user_experience/capregion.html (10 March 2005); and T. Swanson et al., "Recent Implementations of Electronic Medical Records in Four Family Practice Residency Programs," Academic Medicine 72, no. 7 (1997): 607–612.[Web of Science][Medline]
8. Poon, "Assessing the Level of Healthcare Information Technology."
9. C.H. Cheng et al., "The Effects of CPOE on ICU Workflow: An Observational Study," Annual Symposium Proceedings/AMIA Symposium (2003): 150–154; and E.G. Poon et al., "Overcoming Barriers to Adopting and Implementing Computerized Physician Order Entry Systems in U.S. Hospitals," Health Affairs 23, no. 4 (2004): 184–190.[Abstract/Free Full Text]
10. J. Walker et al., "The Value of Health Care Information Exchange and Interoperability," Health Affairs, 19 January 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.10 (3 June 2005).
11. R. Kaushal, C.P. Landrigan, and D.W. Bates, "Improving Patient Safety in Massachusetts," in Memos to the Governor: Management Advice from the Commonwealth’s Experts in Public Administration and Policy, ed. R.T. Moore (Bloomington, Ind.: Lightning Source Inc., 2003).
12. A.M. Epstein, T.H. Lee, and M.B. Hamel, "Paying Physicians for High-Quality Care," New England Journal of Medicine 350, no. 4 (2004): 406–410.[Free Full Text]
13. D.J. Brailer, "Health IT Strategic Framework" (letter to Sec. Tommy Thompson), 21 July 2004, www.os.dhhs.gov/healthit/transmittalletter.html (27 June 2005).
14. See the Certification Commission for Healthcare Information Technology home page, www.cchit.org/about.htm.

                      What's this?

This article has been cited by other articles:

				H. E. Resnick, B. B. Manard, R. I. Stone, and M. Alwan Use of Electronic Information Systems in Nursing Homes: United States, 2004 J. Am. Med. Inform. Assoc., March 1, 2009; 16(2): 179 - 186. [Abstract] [Full Text] [PDF]

				A. H. Goroll, S. R. Simon, M. Tripathi, C. Ascenzo, and D. W. Bates Community-wide Implementation of Health Information Technology: The Massachusetts eHealth Collaborative Experience J. Am. Med. Inform. Assoc., January 1, 2009; 16(1): 132 - 139. [Abstract] [Full Text] [PDF]

				E. A. Miller, M. Booth, and V. Mor Assessing Experts' Views of the Future of Long-Term Care Research on Aging, July 1, 2008; 30(4): 450 - 473. [Abstract] [PDF]

				R. G. Locke and M. Srinivasan Attitudes Toward Pay-for-Performance Initiatives Among Primary Care Osteopathic Physicians in Small Group Practices J Am Osteopath Assoc, January 1, 2008; 108(1): 21 - 24. [Abstract] [Full Text] [PDF]

				A. G. Fiks, R. W. Grundmeier, L. M. Biggs, A. R. Localio, and E. A. Alessandrini Impact of Clinical Alerts Within an Electronic Health Record on Routine Childhood Immunization in an Urban Pediatric Population Pediatrics, October 1, 2007; 120(4): 707 - 714. [Abstract] [Full Text] [PDF]

				S. M. Handler, R. L. Altman, S. Perera, J. T. Hanlon, S. A. Studenski, J. E. Bost, M. I. Saul, and D. B. Fridsma A Systematic Review of the Performance Characteristics of Clinical Event Monitor Signals Used to Detect Adverse Drug Events in the Hospital Setting J. Am. Med. Inform. Assoc., July 1, 2007; 14(4): 451 - 458. [Abstract] [Full Text] [PDF]

				M. Srinivasan, C. R. Keenan, and J. Yager Visualizing the Future: Technology Competency Development in Clinical Medicine, and Implications for Medical Education Acad Psychiatry, December 1, 2006; 30(6): 480 - 490. [Abstract] [Full Text] [PDF]

				J. Z. Ayanian Rising Rates of Cardiac Procedures in the United States and Canada: Too Much of a Good Thing? Circulation, January 24, 2006; 113(3): 333 - 335. [Full Text] [PDF]

				R. Kaushal, D. W. Bates, and D. Blumenthal National Health Information Network Cost and Structure Ann Intern Med, January 17, 2006; 144(2): 147 - 147. [Full Text] [PDF]