Tuesday, March 23, 2010

Meaningful Use and Health Care Reform (Commentary updated)

Since my last commentary on "meaningful use" of an electronic health record (EHR) on June 22, 2009, progress has been made in elaborating relevant definitions as well as specific metrics for evaluation and measurement. Progress has also been made in developing certification criteria for EHR software products available on the market. The Proposed Rule on the CMS Electronic Health Record Incentive Program, published on January 13 in the Federal Register, builds on previous consultations and hearings concerning definition and measurement of "meaningful use" of EHRs. "This proposed rule would implement the provisions of the American Recovery and Reinvestment Act of 2009 (ARRA) (Pub. L. 111-5) that provide incentive payments to eligible professionals (EPs) and eligible hospitals participating in Medicare and Medicaid programs that adopt and meaningfully use certified electronic health record (EHR) technology." (Page 1844, Federal Register, Vol. 75, No. 8) My comment and recommendations address the context of the U.S. health care system through the lens of organizational behavior and theory.
On December 30, 2009, CMS proposed a definition of meaningful use of EHR technology. In summary this definition considers three phases of EHR adoption and meaningful use. In stage I (2011): "... criteria for meaningful use focus on electronically capturing health information in a coded format, using that information to track key clinical conditions, communicating that information for care coordination purposes, and initiating the reporting of clinical quality measures and public health information." Definitions to be applied for stages 2 and 3 remain to be finalized in time for 2013 and 2015 incentive payment years, as do corresponding dimensions of EHR certification. (In the proposed rule published on January 13, CMS refers to ONC definitions of qualified and certified EHR technology.)

This linear design of EHR adoption omits critical dimensions of organizational context and the reticular, multilevel process of change and development in social systems. Furthermore the logical structure of the process has been reversed in two important ways. First, narrow, short-term criteria have been defined for 2011 in stage 1, leaving adopters at a loss to understand how early adoption prepares them for future returns on their investment. Second, infrastructures necessary for meaningful use seem to be viewed as an outcome of the process rather than a prerequisite. The absence of health information exchange (HIE) organizations forming a Nationwide Health Information Network (NHIN) invalidates the most important intrinsic benefit of meaningful use, the communication of health information across institutional and geographical boundaries.

Discussion of the EHR reporting period for determination of ARRA incentive payments (page 1849) reveals concern for a tradeoff between "robust verification and time available to achieve compliance." To resolve this tradeoff, the EHR reporting period in the first incentive payment year is defined as any continuous 90 day period, while subsequent reporting periods should be extended over the entire payment year for more robust measurement of compliance rates. This provision recognizes neither the dynamic, nonlinear character of this complex process nor the critical importance of the initial determination of meaningful use in the process of awarding incentive payments. It might be more reasonable to establish continuous monitoring, for example over a period of the previous 3 to 6 months of EHR use. Measurement of health information exchange (HIE) would further require identification of the organizational configuration of associated Regional Health Information Organizations (RHIOs) or the Nationwide Health Information Network.

The policy priorities for meaningful use presented in the proposed rule are generally consistent with the recommendations of the HIT Policy Committee: 1- To improve quality, safety, efficiency and to reduce health disparities; 2- To engage patients and families in their health care; 3- To improve care coordination; 4- To improve population and public health; and 5- to ensure adequate privacy and security protections for personal health information. Stage 1 criteria for meaningful use include a set of objectives for each policy priority, along with the requirement that all objectives be satisfied for provider qualification. This requirement may be too rigid to accommodate the increasing diversity of health care providers in the U.S.. For example, physician and emerging mid-level provider organizations may satisfy different objectives.

Careful review of the proposed rule reveals that issues related to development of a national health information infrastructure have not been addressed since June, 2009. The criteria for meaningful use exclude functions requiring effective health information exchange - in recognition of the absence of HIT infrastructure and the current low rate of EHR adoption in the U.S.: "Given the anticipated maturity of HIT infrastructure inherent in the strengthening criteria (through 3 stages of meaningful use) and the increased adoption of certified EHR technology predicted in section V. of this proposed rule, these barriers to meaningful use will be removed." (page 1853) Unfortunately, there is no evidence-based argument to support these apparent assumptions at the foundation of EHR meaningful use.

In the proposed rule, there are several examples of criteria adjusted to the lack of infrastructure for HIE. For example, the use of computer provider order entry (CPOE) is defined as "the provider's use of computer assistance to directly enter medical orders (from a computer or mobile device) captured in a digital, structured and computable format for use in improving safety and organization. It does not include the transmittal of that order to the pharmacy, laboratory, or diagnostic imaging center in 2011 or 2012."(page 1856) The formulation of this meaningful use criterion illustrates how the intrinsic motivations for CPOE - the ability to communicate such orders to other health care providers - may be compromised where no infrastructure exists for HIE. Another example is the exclusion of the objective to provide access to patient-specific education resources upon request. An important reason for this exclusion is the lack of infrastructure for open access to such resources.

On the other hand, on page 1856, the goals associated with improved care coordination require health information exchange, and the proposed rule apparently assumes the presence of necessary infrastructure. For example, qualified providers and eligible hospitals must demonstrate capability to exchange key clinical information (such as problem list, medication list, allergies and diagnostic test results) among providers of care and patient authorized entities electronically. Furthermore, concerning meaningful use objectives related to administrative simplification, it is proposed that the phrase "where possible" be deleted from the requirement that insurance eligibility be checked and that claims be submitted electronically since these are already standard HIPAA transactions. However, there is no indication that these electronic transactions are "standard" in practice. An AMA interpretation of HIPAA regulations (AMA Practice Management Center, Understanding the HIPAA Standard Transactions: The HIPAA Transactions and Code Set Rule, published in 2009 -page 7) states that CMS will focus on "...voluntary and complaint-driven enforcement." The responsibility for securing insurer compliance appears to belong to the physician and his or her practice - a substantial additional burden associated with "meaningful use."

The above discussion shows that the assumption of available NHIN infrastructure varies from one section of the proposed rule to another. This is especially inappropriate in the formulation of Stage 1 criteria for meaningful use.

As stated on page 1858, it is surely important to measure conformity with the objectives of meaningful use. However, the measures presented in the proposed rule focus mainly on percentages based on frequency, for example in use of CPOE. This measure ignores the fact that without appropriate HIE, such orders cannot be effectively communicated. The data generated for these measures (the numerator and denominator for calculation of conformity to the required percentage of use) are further difficult to verify, and would be subject to legal challenge especially for qualified providers practicing in multiple settings and through multiple payers (page 1859). EPs practicing in multiple settings are required to conduct 50% of their patient encounters in locations or practices equipped with certified EHR technology. Is it really feasible that "…in evaluating the 50 percent threshold, our proposal is to review all locations-organizations at which an EP practices."? (page 1859) Among EPs whose pattern of practice is changing over time, this suggested review cannot be valide. The measurement further excludes HIE :

"As this objective (CPOE) relies solely on a capability included as part of certified EHR technology and is not, for purposes of Stage 1 criteria, reliant on the electronic exchange of information, we believe it would be appropriate to set a high percentage threshold....For other objectives that are reliant on the electronic exchange of information, we are cognizant that in most areas of the country, the infrastructure necessary to support such exchange is not yet currently available..."(page 1859) The effect of these stipulations will be conflation of "meaningful use" with implementation of certified EHR. The entry of data in a certified EHR is not by itself evidence of meaningful use.

The most persistent problem in creation of infrastructure for HIE is the business model for these enterprises to be created across the U.S.. Until this problem is solved, no meaningful use of EHR will be possible. In a study reported by Wright et al. (Physician Attitudes Toward Health Information Exchange: Results of a Statewide Survey, JAMIA 2010 17: 66-70), 45% of respondents reported no usage of EHR while 28% reported usage of a simple EHR defined as "an integrated clinical information system that tracks patient health data and may include such functions as visit notes, prescriptions, lab orders, etc." (page 67). It is difficult to conclude from this survey that physicians would be willing to pay for HIE, as so few have any meaningful experience with this function. Further, in a study by Adler-Milstein et al. (Characteristics Associated with Regional Health Information Organization Viability, JAMIA 2010 17: 61-65), it is concluded that "Exchanging a narrow set of data and invoking a broad group of stakeholders were independently associated with a higher likelihood of being operational." (Abstract page 61)

This conclusion omits reference to the extremely low rate of viability observed among RHIOs under study. In another very recent study by Ross et al. (Health Information Exchange in Small-to-Medium Sized Family Medicine Practices: Motivators, Barriers, and Potential Facilitators of Adoption, IJMI 2010 79: 123-129), electronic prescribing was ranked favorably among health information exchange functions, but no surveyed practice identified available government incentives as a significant motivator for adoption of this function in HIE, particularly in practices where eprescribing was already included in EHR functions.

In conclusion to this commentary, I would agree with some other observers: physicians and other health care providers should not wait for U.S. government incentives for adoption of certified EHR or "meaningful use". Rather they should assess the intrinsic benefits to their workflow as well as ROI resulting from prospective software adoption, and they should act in accordance with internal scenario analysis. System level efficiencies at the regional or national levels of analysis will only become accessible with sustained public investment in necessary infrastructures. Unfortunately in the U.S. health care sector such investment remains unlikely.
More promising than the NHIN configured among local and regional RHIOs is the National Information Exchange Model (NIEM) originated in 2005 by the Department of Justice (DOJ) and the Department of Homeland Security (DHS) to address information exchange among government agencies in the context of national security. Even though this development of the NIEM has suffered some of the same difficulties as HIE in defining an effective and sustainable business model, it has benefited from more consistent and longer term public funding.
While the NHIN is designed as a many to many mapping of communication among participating entities, the NIEM proposes a canonical mapping through the common infrastructures of the model shared among communities of interest. Such a national – and eventually global - infrastructure offers services as well as a system of governance to assure economies of scale and scope in information exchange across enterprise domains served. [1] [2] [3] [4]
Websites: Justice Information Sharing ; National Information Exchange Model

1-Emphasize intrinsic benefits available through meaningful use rather than unsustainable extrinsic incentives. NIEM implementation model values stakeholder participation in system development and governance. HIE is a fundamental intrinsic motivation for meaningful use of EHRs as well as a central criterion for its assessment. For example, where CPOE does not include the functionality for communication across health care providers and pharmacies, it seems meaningless to require that 80% of such orders be entered electronically. Modest financial incentives for individual providers will not be adequate to motivate behavior perceived to be meaningless.
2-Recognize the benefits of meaningful use of certified EHR software to integrate clinical research and practice. According to Chris Thorman of Software Advice, a website that reviews electronic health records this important motivation for EHR adoption surpasses the HITECH incentives in terms of return on investment in health information technology. According to Synergyst Research only 10% of more than 720,000 licensed American physicians participate in clinical trials, mostly because of the significant burdens associated with data collection, extra paperwork, compliance with regulations, and staff training. EHRs contribute to solve these problems. Chris Thorman has summarized the usefulness of EHRs in clinical trial participation in the following table appearing in his article, Electronic Health Records and Clinical Trials, An Incentive to Integrate:
Identify potential opportunities
EHR vendors whose software integrates with clinical trial providers will have access to trials, studies, and registries that your practice is eligible to participate in.
Identify number of potential trial subjects
The search function in an EHR database allows a user to quickly identify how many of a practice's patients are potentially eligible for a clinical trial. From there, the clinical trial provider can determine if a practice would be a good partner.
Patient enrollment
The EHR has the capability to implement trial-specific screening requirements into new patient records to determine their eligibility for a study. The EHR will also have the ability to identify patients who meet the exact requirements of a study.
Study execution
During the trial, the EHR can create trial-specific data fields that can be populated during routine patient encounters. Conflict alerts can also be created to notify providers of actions that violate a study's protocol.
Data submission
The EHR will be able to submit information to EDC software without having to convert the data. This eliminates redundant data entry and increases accuracy of the data.

­3-Recognize the reticular, nonlinear process of EHR adoption and meaningful use by considering the configuration of communities of interest in the process of use assessment. Greater emphasis, particularly in definition of the time frame, should be placed on evaluation of meaningful use at the initial and most important phase. Failure to adequately assess the initial phase will result in serious problems in subsequent phases of evaluation for incentives. The time frame should be a moving period of 3 to 6 months to better reflect the dynamic character of the process.
4-Examine assumptions related to HIE infrastructures as they shape the formulation of the rules throughout the document. These assumptions appear to vary, affecting whether the exchange of information is required or not. They should be realistic and consistent throughout. Furthermore, qualified providers should not be charged with obtaining compliance on the part of insurers or other partners, as appears to be the case in checking insurance eligibility.
5-The U.S. could develop a more effective long-term strategy through collaboration with the Brazilian BIREME and the Canadian Infoway. This approach would contribute as well to creation of an effective regional and hemispheric health information system.
[1] Allen, C. Information sharing and the federal state and local levels. Testimony before the Senate Committee on Homeland Security and Governmental Affairs. July 23, 2008. Washington, DC. http://www.dhs.gov/xnews/testimony/testimony_1216992676837.shtm
[2] Carter DL, Carter JG. The Intelligence Fusion Process for State, Local, and Tribal Law Enforcement. Criminal Justice and Behavior 2009 December 1;36(12):1323-1339.
[3] Garson GD. Securing the Virtual State: Recent Developments in Privacy and Security. Social Science Computer Review 2006 November 1;24(4):489-496.
[4] Rollins J. Fusion Centers: Issues and Options for Congress. 2008; RL34070. http://fas.org/sgp/crs/intel/RL34070.pdf