Week 3 - Discussion 1,
Progress in information technology implementation in health care has
lagged behind most other information-intensive service sectors. Is this
good or bad? What should the federal government do about it?
Week 3 - Discussion 2,
In the course text, read case 8-1, pages 219-232. What is your response to discussion question # 4 on page 232?
Case 8–1 REGIONAL HEALTH INFORMATION ORGANIZATIONS (RHIOs): DISRUPTIVE TECHNOLOGY OR BUSINESS AS USUAL?
HIT provides an opportunity for insights into why and how the health industry remains so far behind others in the adoption of new information technologies, despite leadership and motivation for change. Entrepreneurial vendors abound. Influential politicians and professionals call for change, and the public wants it; however, this potentially disruptive set of innovations continues to sputter and lurch forward, often threatening to stall. This case contrasts two approaches to community-wide health information and communication networks.
• The first approach is the Whatcom Health Information Network (HINET), a grassroots, community-based, community-initiated, distributed model that links physician practices, the local hospital, and other providers of health-related services. This network dates to the early 1990s.
• The other approach is the NHIN—the federal government’s recent attempt to exert market-oriented technological leadership, including the encouragement of demonstration projects of RHIOs. This national strategy seems to favor a more top-down, centralized model and has tended to emphasize extending the reach of complex, large-scale, dedicated systems used by large hospitals and integrated health care systems.
These two approaches are not mutually exclusive and, indeed, have significant overlap. Much of the value of the Whatcom Health Information Network stems from the fact that it gives providers online access to EMRs stored at the local hospital. HINET is an example of a RHIO, and the federal NHIN strategy is designed to support RHIOs. Community networks similar to HINET participate in the four teams the Office of the National Coordinator for Health Information Technology (ONCHIT) assembled to develop regional models that might serves as models for a national system.
The approaches are, nonetheless, different in several ways. One attempts to deploy emerging products from a wide set of vendors in a manner that is responsive to the capacity and cost requirements of smaller practices and the local community. The other represents an attempt to speed up the agonizingly slow and often disappointing development process for large-scale technologies first attempted with time-shared systems in the late 1960s.
DEFINITION OF TERMS
Narrowly defined, a HIN is a web of hardware (desktop computers, imagers, wires, fiber optic cables, servers, etc.) and basic communications software that allows the exchange of documents (e-mails, image files, database records, and other items) between providers, but the term can also encompass the software (databases and other programs) that allow for maintenance of EMRs, also called EHRs.
The highest form of HIN is achieved when EMR programs are compatible, can contribute to personal health records that are accessible to or controlled by consumers, allow for CPOE, and are able to exchange patient information across a network or through other media.
The Stakes Are High
The impetus for promoting EMRs that can be exchanged across HINs is fourfold—cost containment, quality, patient empowerment, and data.
• Cost containment: Interoperable EMR systems have the potential to eliminate waste and duplication of effort, streamline workflow, increase productivity, and slow health care inflation.
• Quality: The 2001 Institute of Medicine Report Crossing the Quality Chasm identified five activities in which the use of health IT has been demonstrated to improve the quality of patient care: (1) researching treatment alternatives or recommended guidelines; (2) sharing clinical data and images; (3) reviewing patient notes, lists of medications, and lists of problems; (4) creating reminders for preventive care; and (5) writing legible prescriptions.
• Patient empowerment: Medical practice is moving toward patient-centered care. When knowledgeable patients are actively involved in self-management and decision making, outcomes can improve along with patient satisfaction. Many cost-containment strategies also rely on patient involvement, which is enhanced by access to one’s own EHR.
• Data: The fourth argument for EMRs is simply to improve the quality and quantity of data that can then be used for a variety of purposes. “Trying to create an accountable system or a well-functioning health care marketplace without accurate, accessible, meaningful, and timely date is an exercise in futility” (Halvorson, 2005, p. 1266).
Forecasting the Impact
The RAND HIT team began a study in 2003 to better understand the impacts of HIT (Hillestad et al., 2005). The researchers, after reviewing the literature on HIT and developing an economic model, made the following findings:
• If the adoption rate of HIT reached 90% between 2002 and 2016 and the resulting productivity gains were on the order of 4% per year, efficiency savings for inpatient and outpatient care could reach $77 billion per year.
• After concluding that “improved patient safety from EMR use in hospitals and ambulatory care largely focus on alerts, reminders, and other components of CPOE,” the researchers estimated that EMRs could prevent 2 million adverse drug events annually, at a savings of $3.5 billion.
• Health IT could result in improved care and better outcomes by increasing the delivery of proven preventive services, improving chronic diseases management in the near term, and preventing and managing chronic diseases in the long-term. Some $40 billion in net savings could be realized there each year.
• The cost of implementing HIT is significant, but would not outstrip savings. Reaching a 90% adoption rate by 2016 would require an average annual investment for hospitals of $6.5 billion and for physicians of $1.1 billion.
• Net cumulative savings over the 15-year implementation period from 2002 through 2015 from greater efficiency and safety savings would be $513 billion—$371 billion from hospital systems and $142 million from provider systems.
The RAND team concluded, “Given our analysis, we believe there is substantial rationale for government policy to facilitate widespread diffusion of interoperable HIT…. It is not known what changes should or will take place after widespread EMR system adoption—for example, increased consumer-directed care, new methods of organizing care delivery, and new approaches to financing. It is increasingly clear, however, that a lengthy uneven adoption of nonstandardized, noninteroperable EMR systems will only delay the chance to move closer to a transformed health care system” (p. 1115). Such fragmented adoption will also cost significantly more than a strategic, standards-based approach.
Most observers would agree that the state of affairs the RAND researchers warn against—an attenuated period of adoption with fragmented networks, no clear standards, and the proliferation of systems with limited or no interoperability—is where the United States finds itself. Roughly 1 in 4 or 1 in 5 hospitals have EMR systems, whereas the number for ambulatory care practices in probably closer to 1 in 10 (Taylor et al., 2005). The rate of IT investment per employee per year is around 5%—relatively low compared with other successful industries (Bower, 2005).
The situation is changing, however slowly. The Center for Health Systems Change compared data from the 2000 to 2001 and 2004 to 2005 HSC Community Tracking Study Physician Survey (Reed & Grossman, 2006). It found that significant improvements (5% or more) had been made in the number of physicians using information technology to obtain guidelines, exchange clinical data, access patient notes, generate reminders, and write prescriptions (see Figure 8-3).
FIGURE 8-3 Percent of Physicians in Practice with IT for Specific Clinical Activities, 2000–01 and 2004–05
Changes from 2000–2001 to 2004–2005 are statistically significant to P < .0001 for all activities.
Source: Community Tracking Study 2004-2005 Physician Survey, ©Center for Studying Health System Change, from Grossman and Reed (2006).
Barriers to Adoption
In an April 24, 2006, column in the online version of Time, pediatrician Donald Berwick, president of the Institute for Healthcare Improvement, lamented, “My pizza parlor is more thoroughly computerized than most of health care.”
There are several explanations for slow adoption of HIT. One is the initial cost of purchasing EMR systems, which are expensive, complex, and disruptive. Initial costs for a system to serve a provider office are in range of $25,000 to $50,000 per full-time equivalent (FTE). Miller et al. (2005) put initial costs to solo or primary care small-group practices at $44,000 per FTE and estimated ongoing costs of about $8,500 per year.
A major barrier to health IT adoption is that costs are generally paid by hospitals and providers, but providers only see about 11% of the benefits; the bulk of the benefits accrue to payers and consumers (Bates, 2005). Miller et al. (2005), however, reported that small practices generally recouped expenses in 2.5 years and profited thereafter, largely because they increased their reimbursements through more aggressive coding (a finding that will be of concern to many). Despite some evidence health IT will pay for itself, Hackbarth and Milgate (2005) are among many observers who argue that government should make a bigger investment. This is in contrast to ONCHIT’s market-based approach. Former national coordinator David Brailer said he supports pay-for-performance but not pay-for-health IT use. “I think that it [EMR adoption] becomes incredibly hard for a provider to resist because the evidence is so overwhelming that care is better whenever it is done” (Cunningham, 2005, p. 1153).
Middleton (2005) argued that health IT in the Unites States provides a classic example of market failure. He urges that it be seen as a public good, saying that “a Third Hand is needed to address the failure of the Invisible Hand.” Kleinke (2005) bluntly explained why the market has not produced a viable health information system and articulated the case for more direct government intervention. He maintained that rather than searching out the business case for health IT, we should recognize that many players in the health care industry have a strong business case for not implementing IT (and many of those players, interestingly enough, are in leadership positions on the national organizations charged with rationalizing health IT and promoting a NHIN). Insurers, he pointed out, are judged by their “medical loss ratio” and are rewarded for slowing down or preventing transactions. Faster, better, more transparent claims-processing is not in their interest. Similarly, hospitals get reimbursed according to the number of services they provide, even if they are necessary because of medical errors. Reducing duplication, eliminating unnecessary procedures, and avoiding errors are not outcomes from which they benefit. It is not in the interest of any provider or facility to have a sick patient empowered to transfer to another facility or practice with his or her medical records in hand. EMRs and e-mail could reduce the number of billable office visits per patient, which would be economically detrimental to many practices. Few players would actually benefit from transparency in pricing, detailed billings, and more consumer choice. “It is noteworthy not that so few practices computerize their clinical activities,” he wrote, “but that so many of them actually do” (p. 1252).
WHATCOM HEALTH INFORMATION NETWORK
In 1990, St. Joseph Hospital in Bellingham, Washington convened a 2-day retreat with 84 area physicians to develop a vision for seamless care in Whatcom County. A group convened later included technology experts. As an outgrowth of that discussion, the hospital knew when it began looking into EMRs in 1994 that what it wanted to create was not merely a hospital records system, but what it called a “community health record”—a common electronic record that would promote an integrated delivery system and ensure area providers had access to the right information in the right place at the right time (Nichols, 2006).
St. Joseph is the only hospital in Whatcom County, which has about 175,000 residents. Of those residents who had health insurance in 1994, about two thirds were covered through the local Blue Shield licensee, the Whatcom Medical Bureau. While the hospital was looking to computerize its records, the bureau was looking to share eligibility and benefits information electronically. It already processed 68% of claims electronically. The two organizations agreed to put $1.7 million each into startup costs for the Whatcom Health Information Network—a way for providers to communicate with one another, link up with the hospital records system, and submit claims. The system would be free to providers for the first few years and for at least 3 years after that the monthly fees for physician offices would not exceed $70 per month per physician.
HINET launched in 1995. Two technicians working with the hospital technical staff designed and built the system, supplying each provider office with three computer lines per physician. They supplied one computer and one printer per practice. The network itself began as a frame relay system. Users could access a bulletin board system and send and receive files using FTP (file transfer protocol) software.
HINET had to come up with a sustainable business model for when the $3.4 million in seed money ran out. It did not want to create disincentives for using the system, and that meant no transaction costs. It decided to apportion costs to each sector of the industry based on the share of the health care dollars it received. Using federal data, it decided the hospital should carry 40% of the costs, and 22% should be distributed across the doctors. Under the first year of this scheme, medical practices paid $45.88 per doctor per month—far less than the pledged $70 ceiling. Costs were similarly distributed among other users—long-term care facilities, pharmacies, and “others.” In 2006, HINET spun off from the hospital. A limited liability corporation with an annual budget of about $1 million, it maintains the network and the servers and provides a help desk, Internet, e-mail, security, and spam protection. It links 2,400 personal computers in health care providers’ offices, as well as the county jail, mental health service organizations, skilled nursing facilities, and even the fire department. The network itself is a private intranet built with a mix of fiber and cable. In the spring of 2006, the cost per physician was $74.95 a month. For that fee, providers can not only communicate better and access online information sources, but they can also call up hospital emergency room reports on their patients, receive laboratory results, and view x-rays and other digital medical images. A coalition of independent community organizations, funded through a Robert Wood Johnson Foundation Pursuing Perfection grant, developed a personal health information Web application called Shared Care Plan. The network also participates in the county’s e-Prescribing Project, which will provide electronic prescription software.
NATIONAL HEALTH INFORMATION NETWORK
As national coordinator for HIT, David Brailer saw the need for a government policy to promote HIN development on a national scale. He argued for using a market-based approach to achieve that end. Government’s role would be to facilitate the development of leadership structures. ONCHIT laid out a three-pronged strategy: (1) support and promote RHIOs, (2) encourage development of a NHIN “to exchange patient health information accurately and in a timely manner under stringent security, privacy, and other protections,” and (3) drive EMR deployment by reducing the costs and risks to physicians. To implement this voluntary, nonregulatory strategy:
• ONCHIT issued a request for information on November 15, 2004, asking what an NHIN might look like. Two months later, it received nearly 5,000 pages of responses from 512 organizations. It issued a summary report in June 2005.
• It chartered the American Health Information Community, a federal advisory commission, on September 13, 2005. Health and Human Services Secretary Michael O. Leavitt chairs “the community” and the rest of its membership comprises 16 health care, government, consumer, and business leaders.
• It established the Healthcare IT Standards Panel (HITSP) under a $3.3 million contract awarded to the American National Standards Institute on October 6, 2005. The panel’s charge was to harmonize standards. On June 29, 2006, HITSP recommended selected standards for EHRs, biosurveillance, and consumer empowerment.
• At the same time, ONCHIT awarded $11.5 million to RTI International to oversee the Health Information Security and Privacy Collaboration (HISPC), a multidisciplinary team that was to work with state and territorial governments to develop baseline requirements for security and privacy protections. State and territorial groups that received HISPC contracts were to report their assessments, proposals, and plans to RTI by June 9, 2006.
• The third contract awarded that same day was for $2.7 million and went to the nonprofit Certification Commission for Health Information Technology (CCHIT). The commission was charged with figuring out criteria for evaluating and certifying HIT products and solutions that comply with recommendations of HITSP and HISPC. In phase 1, CCHIT looked at inspection and certification of ambulatory EMRs, issuing commercial criteria on May 3, 2006. CCHIT issued its first certifications in July 2006. They seemed to be tailored for a variety of systems and vendors; however, at CCHIT’s May 2006 conference in Baltimore, a “town hall” meeting elicited serious concerns from small vendors. Later phases will examine inpatient EMRs and HIT networks.
• On November 10, 2005, ONCHIT announced that it was providing $18.6 million to four teams that would develop regional HINs (see Table 8-4). Although they would serve specific markets, they would be prototypes for a NHIN. Existing community RHIOS were key players on each team. “We called for regional organizations as business conveners,” said Brailer (Cunningham, 2005, p. 1154). Eventually, these regional systems would have to speak seamlessly to one another to begin to create a nationwide network.
• The four regional consortia reported on their efforts in Fall 2006. In June 2007 ONCHIT issued a request for proposals to fund an additional ten state, regional and non-geographic consortia.
TABLE 8-4 The Four RHIO Partnerships Selected
The four regional teams supported by the Office of the National Coordinator for Health Information Technology are:
• Accenture, with partners Apelon, Cisco, CGI-AMS, Creative Computing Solutions, eTech Security Pro, Intellithought, Lucent Glow, Oakland Consulting Group, Oracle, and Quovadx, which is working regionally with the Eastern Kentucky Regional Health Community, CareSpark (Tennessee), and the West Virginia eHealth Initiative
• CSC, with partners Browsersoft, Business Networks International, Center for Information Technology Leadership, Connecting for Health, DB Consulting Group, eHealth Initiative, Electronic Health Record Vendors Association, Microsoft, Regenstrief Institute, SiloSmashers, and Sun Microsystems, which is working regionally with the Indiana Health Information Exchange, MA-SHARE (Massachusetts), and Mendocino HRE (California)
• IBM, with partners Argosy Omnimedia, Business Innovation, Cisco, HMS Technologies, IDL Solutions, Ingenium, and VICCS, which is working regionally with Taconic Health Information Network and Community (New York); North Carolina Healthcare Information and Communications Alliance (Research Triangle and Rockingham County)
• Northrop Grumman, with partners Air Commander, Axolotl, Client/Server Software Solutions, First Consulting Group, SphereCom Enterprises, and WebMD, which is working regionally with Santa Cruz RHIO (California); and HealthBridge (Cincinnati, Ohio) and University Hospitals Health System (Cleveland, Ohio).
Source: Office of the National Coordinator for Health Information Technology.
A community-based RHIO such as HINET that is developed with participation from the entire local medical community can employ a variety of solutions geared to the needs of small practices—an important consideration since nearly a third of all physicians work in solo or two-physician practices and more than half work in practices with fewer than 10 physicians (Grossman & Reed, 2006) and small practices lag behind large ones in their use of health IT (see Figure 8-4). A HINET-type network could easily incorporate an ASP (application service provider) Internet system and open standards technology, which would reduce the investment required by a practice. Groups such the American Academy of Family Physicians have launched demonstration projects aimed at developing Internet-based EMR systems costing about $100 per physician per month to maintain.
FIGURE 8-4 Physicians in practices with IT for at least three of five clinical activities in 2000–2001 and 2004–2005 by practice size
Changes between 2000–2001 and 2004–2005 in the gaps between group practices with 1–9 physicians and group practices with 10–50 physicians and 51+ physicians were statistically significant at P < .05.
Source: Community Tracking Study 2004-2005 Physician Survey, ©Center for Studying Health System Change, from Grossman and Reed (2006).
Such approaches, however, also reduce the market for proprietary systems and standalone hardware and software, setting up a conflict with hospitals and large practices that use systems built around the HL7 language developed by Health Level Seven, Inc., a nonprofit standards development organization, and with technology companies intent on developing and marketing systems they control. Low-costs systems also reduce the competitive advantage of larger practices that can afford high-cost systems.
The federal NHIN approach, by comparison, has the potential to favor large-scale systems that would benefit the technology companies and large health care companies that sit on the four regional development teams and other leadership groups that are part of the NHIN effort. The tendency will be to try to extend the reach of HL7-based, hospital-type systems into the offices of smaller provides. This could help promote interoperability by discouraging the proliferation of many potentially incompatible systems, but it also has the potential to drive up costs and complexity and to delay uptake by solo providers and small group practices.
While delivering the keynote address at the Health Information and Management Systems Society (HIMSS) meetings in February 2005, Brailer spoke openly of the inherent conflicts. Health Data Management reported him saying the following:
The electronic records investment by large group practices is strategic, intended to shift the market toward them. They should be recognized for their leadership and not faulted for their inventiveness. However, if we believe electronic records improve health status, then we have an obligation to level the playing field. We can’t allow one part of the physician community to have the technology and another unable to afford it.
Brailer also took aim at the mindset in the health care industry that patient data is a proprietary strategic asset, a concept that could undermine regional networks. “Interoperability is not just an IT issue. We need guidelines on how to share and access data between competitive organizations,” he said. “Data sharing is a precursor to widespread use of electronic records. Proprietary boundaries are growing around health information at the same time that talk about interoperability has become commonplace.”
He also noted that in comments ONCHIT received in response to the Request for Information on RHIOs, “Many respondents recommended that the NHIN be a virtual network that leverages the public Internet.”
According to a New York Times report on the HIMSS meeting, Brailer asked technology executives in small group meetings to put aside their narrow corporate interests and give up some of their proprietary lock on customers in order to open up a larger market opportunity to everyone. “Each company cannot get all it wants,” he said. “The elephant in the room in what we are trying to do is the small physician practices…. That’s the hardest problem, and it will bring this effort to its knees if we fail” (Lohr, 2005).
4. Where does the government’s NHIN strategy fall on the political and economic spectrum of alternative approaches to health care policy?