Web First. 2 Health Affairs MAY :5

Transcription

1 By Catherine M. DesRoches, Chantal Worzala, Maulik S. Joshi, Peter D. Kralovec, and Ashish K. Jha Small, Nonteaching, And Rural Hospitals Continue To Be Slow In Adopting Electronic Health Record Systems received $1.4 billion in Medicare incentive payments, and more than 3,000 hospitals had registered for the Medicare or Medicaid EHR incentive program. 1 Additionally, the Affordable Care Act of 2010 provided for many initiatives including the formation of accountable care organizations and bundled payments that will be difficult for health care organizations to implement without well-functioning EHR systems. 2 To date, the pace of adoption of EHR systems in US hospitals has been slow, and the future pace of adoption and distribution of adoption across all hospitals remains uncertain. Since 2008, when we first started systematically mondoi: /hlthaff HEALTH AFFAIRS 31, NO. 5 (2012): 2012 Project HOPE The People-to-People Health Foundation, Inc. ABSTRACT To achieve the goal of comprehensive health information record keeping and exchange among providers and patients, hospitals must have functioning electronic health record systems that contain patient demographics, care histories, lab results, and more. Using national survey data on US hospitals from 2011, the year federal incentives for the meaningful use of electronic health records began, we found that the share of hospitals with any electronic health record system increased from 15.1 percent in 2010 to 26.6 percent in 2011, and the share with a comprehensive system rose from 3.6 percent to 8.7 percent. The proportion able to meet our proxy criteria for meaningful use also rose; in 2011, 18.4 percent of hospitals had these functions in place in at least one unit and 11.2 percent had them across all clinical units. However, gaps in rates of adoption of at least a basic record system have increased substantially over the past four years based on hospital size, teaching status, and location. Small, nonteaching, and rural hospitals continue to adopt electronic health record systems more slowly than other types of hospitals. In sum, this is mixed news for policy makers, who should redouble their efforts among hospitals that appear to be moving slowly and ensure that policies do not further widen gaps in adoption. A more robust infrastructure for information exchange needs to be developed, and possibly a special program for the sizable minority of hospitals that have almost no health information technology at all. Catherine M. DesRoches (cdesroches@ mathematica-mpr.com) is a senior researcher at Mathematica Policy Research, in Princeton, New Jersey. Chantal Worzala is director of policy at the American Hospital Association, in Washington, D.C. Maulik S. Joshi is president ofthehealthresearchand Educational Trust and senior vicepresidentofresearchat theamericanhospital Association. Peter D. Kralovec is the senior director of the Health Care Data Center for the American Hospital Association s Health Forum. Ashish K. Jha is an associate professor of health policy at the Harvard School of Public Health and an associate professor of medicine at Harvard Medical School, in Boston, Massachusetts. Enactment of the 2009 stimulus law, which contained the Health Information Technology for Economic and Clinical Health (HITECH) Act, has resulted in federal investments in an array of programs to increase the adoption and use of electronic health records (EHRs) in US hospitals and physicians offices. To receive incentive payments through the Centers for Medicare and Medicaid Services, these providers must meet the federal standard for meaningful use. As of February 2012, a few more than 800 of the nation s approximately 5,000 hospitals had MAY :5 Health Affairs 1

2 itoring adoption, the percentage of hospitals with any EHR system grew by approximately three percentage points each year, reaching 15 percent in However, these early surveys were conducted before implementation of the meaningful-use incentives. It is possible that with financial incentives, and the specter of future payment cuts for nonadopters, the adoption rate of EHR systems in US hospitals has accelerated, as has been suggested by recent publications from HIMSS (Healthcare Information Management and Systems Society) Analytics, which show increases in hospital adoption across all levels of the HIMSS adoption model. 6 However, nationally representative survey data that is generalizable to all acute care hospitals have not been available. As the federal government prepares to finalize proposed stage 2 meaningful-use requirements, nationally representative data on the pace and spread of adoption across hospitals and barriers to achieving the federal criteria can highlight both successes of the program to date and challenges still to be overcome. Thus, we used the most recent 2011 data from a national longitudinal survey of hospitals to answer the following questions: What proportion of US hospitals had a basic or comprehensive EHR system or could meet our proxy standard of meaningful use in 2011? Are there specific types of hospitals that appear to be making progress more rapidly than others? And finally, which electronic functions appear to be the biggest barriers to hospitals reaching the meaningful-use mark? Study Data And Methods We used data from the American Hospital Association annual survey of health information technology adoption from the period The association sent the survey, as a supplement to its annual survey, to the CEO of each US hospital. The CEO was asked to assign the survey to the most knowledgeable person in the organization, typically the chief information officer or equivalent. All nonrespondents received multiple mailings and follow-up telephone calls to try to achieve a high response rate. (The final response rate was 58 percent.) The survey was fielded during October December All respondents were given the option to complete the survey online or by mail. As with prior surveys, respondents chose online over mail by a wide margin. Survey Content The contents of the survey have been previously described. 3 Briefly, the survey asked each respondent to report on whether the hospital had implemented each of the twenty-four clinical functions that might be part of an EHR system. In the 2011 survey, additional questions were asked about specific functions that are tied to meaningful use. Measures Used We used previously developed definitions of basic and comprehensive EHR systems. 3 A basic system is defined as full implementation of the following technologies in at least one clinical unit of the hospital: computerized systems for patient demographics, physician notes, nursing assessments, patient problem lists, laboratory and radiologic reports, diagnostic test results, and order entry for medications. A comprehensive EHR system must include all of the functions that a basic system can perform and fourteen additional functions. In addition, to meet the comprehensive definition, all functions must be present in all major clinical units in the hospitals. Many of the functions included in the comprehensive definition are necessary for meeting the federal meaningful-use standard. However, there are differences between the two measures. The survey made available data on twelve functions that very closely resembled twelve of the fourteen core criteria of the first stage of meaningful use. These served as our proxy standard for meaningful use. The twelve functions included in the survey were computerized records for patient demographics; problem lists; medication lists; vital signs; smoking status; patient allergies; computerized provider order entry for medications; decision support, including clinical guidelines; drug-allergy and drug-drug alerts; automatic generation of quality metrics; and provision of electronic discharge summaries and health information to patients. This measure using these twelve functions was previously used to assess hospitals readiness for meaningful use in 2009 and ,5 The individual survey items included in this measure were constructed by staff from the Office of the National Coordinator for Health Information Technology in collaboration with the American Hospital Association to assess hospitals progression toward meaningful use. The survey included a measure of whether a hospital participated in health information exchange. However, that measure was a poor proxy for the stage 1 meaningful-use criteria focused on data exchange. The survey item measured a much higher level of exchange than required by the criteria, asking hospitals if they participated in a regional health information exchange with unaffiliated providers. The meaningful-use criteria only require hospitals to attest to performing a test exchange with another entity. However, given the importance of data exchange, we recreated our meaningful-use proxy standard with data exchange as a necessary com- 2 Health Affairs MAY :5

3 ponent in an additional analysis.we present the results, including the exchange variable, in the online Appendix. 7 The survey included an item assessing hospital implementation of electronic systems to review and update privacy and security measures a core objective of meaningful use. However, we did not include this variable in our definition because approximately 80 percent of hospitals did not answer the question. Overall, our proxy standard was far less stringent than the regulatory definition of meaningful use, which requires hospitals to meet all fourteen core objectives plus five of ten menu objectives for using certified EHR technology. Analysis As in prior years, we first began by excluding all nonfederal, non acute care hospitals, and we included only general medical and surgical hospitals that had responded to the survey. We then compared the characteristics of responding to nonresponding hospitals and found small, statistically significant differences between these two groups. We used a standard technique, a regression model that predicts the likelihood of responding to the survey, to develop weights for nonresponse to ensure that our survey was nationally representative. 8 For example, if small hospitals were 20 percent less likely to respond to the survey, we would create a weight that would, in effect, increase the analytic number of smallhospital respondents by 20 percent to ensure that the sample was representative. We calculated the proportion of US hospitals that had adopted basic and comprehensive EHRs in 2008, 2009, 2010, and 2011 by simply calculating rates for each year separately. We plotted the trends in these proportions over time from 2008 the first year that these data were available through We next calculated the proportion of hospitals that could meet our proxy standard of meaningful use those with all core functions implemented in at least one clinical unit. In the Appendix, 7 we also show who meets the somewhat more stringent meaningful-use criteria of having all twelve functions fully implemented in all clinical units. We next determined whether different types of hospitals were moving faster than others from no implementation of an EHR system to a basic or a comprehensive system between 2008 and We also examined whether different types of hospitals had made differential progress toward meaningful use between 2010 the first year that the meaningful-use questions were asked and Finally, we examined rates of adoption for each of the individual criteria required to meet our proxy standard of meaningful use, and we identified which EHR system functions had not yet been adopted by the hospitals that had not met the stage 1 criteria.we stratified these by size and location, urban versus rural, and present those data in the Appendix. 7 We believe these analyses are particularly salient for policy makers as they finalize the rules regarding the types of features hospitals will need to have to meet the next bar (stage 2) for receiving financial incentives. Limitations There are limitations to our work that may have resulted in an overestimation in the proportion of US hospitals that have adopted an EHR system and achieved meaningful use. Although the survey received a 58 percent response rate, as in past years there were differences between responding and nonresponding hospitals. We attempted to account for these differences by adjusting for potential nonresponse bias; however, these adjustments are imperfect. In addition, nonresponding hospitals were more likely than responders to have characteristics associated with low adoption rates for EHR systems. We have also used a relatively generous measure of meeting meaningful-use criteria. Our definition did not map exactly to the core meaningful-use requirements set forth by the government, and we did not include any measure of whether a hospital can meet the menu requirements. Finally, the American Hospital Association survey has changed slightly each year as it has been adapted to measure the evolving meaningful-use criteria. These changes make it difficult to infer that changes in adoption and readiness for meaningful use are the result of external forces such as HITECH. However, we have used the same measures of basic and comprehensive EHR system adoption since 2008, and the measure of meaningful use presented in this article was very similar to the ones presented in prior years. Study Results The American Hospital Association received responses from 3,233 hospitals. Federal hospitals, those located outside of the fifty states and the District of Columbia, and nonmedical or surgical institutions were excluded. Of the total responses, 2, percent of all acute care hospitals in the United States remained for analysis. As noted above, we found small but significant differences between hospitals that responded to the survey and nonresponders. Hospitals that responded to the survey were more likely to be large, teaching hospitals, and located in the MAY :5 Health Affairs 3

4 Exhibit 1 Northeast or Midwest. All analyses were statistically weighted to account for potential nonresponse bias using the approach described above. Changes In Hospitals EHR Adoption We found a substantial increase in the proportion of hospitals with either a basic or comprehensive EHR system between 2010 and 2011 (Exhibit 1). In 2010, 3.6 percent of hospitals had a comprehensive system. By 2011 that number more than doubled, to 8.7 percent. Eighteen percent of hospitals had a basic system in 2011 an increase from 11.5 percent in Therefore, between 2010 and 2011 the proportion of hospitals with at least a basic EHR system increased from 15.1 percent to 26.6 percent. Hospitals with at least a basic system were, compared to those with no system, more likely to be large, major teaching hospitals, and located in the Northeast. During the past four years, the gap in rates of adoption of at least a basic EHR system increased substantially based on hospital size, teaching status, and location. There was a fifteen-percentage-point gap in system adoptions between large and small hospitals in 2010 (25.7 percent compared to 10.7 percent, respectively) a gap that widened to 22.2 percentage points in 2011 (20.8 percent compared to 43.0 percent, respectively). Similar divergences based on hospitals teaching status and urban locations were also found (Exhibit 2). Hospitals in rural areas saw growth in system adoptions over time but had the lowest rate of any group analyzed: 19.4 percent had at least a basic system in Achieving Meaningful Use Using our relaxed proxy standard for determining meaningful use twelve functions implemented in at least one major clinical unit we estimate that after adjustment for nonresponse, just under one in Changes In The Adoption Of Basic And Comprehensive Electronic Health Record (EHR) Systems Among US Hospitals, Percent Any EHR Basic EHR Comprehensive EHR SOURCE Authors calculations of data from the American Hospital Association annual survey information technology supplement. NOTE All analyses were statistically weighted for potential nonresponse bias. five US hospitals 18.4 percent had all twelve core functions implemented in at least one unit a substantial increase from 4.3 percent in 2010 (Exhibit 3). Using the more stringent definition twelve functions implemented across all major clinical units 11.2 percent of hospitals met the criteria. Finally, when we included a measure of data exchange, the percentage of hospitals meeting the proxy measure of meaningful use decreased to 5 percent. The full results are shown in the Appendix. 7 An additional 33.6 percent of US hospitals had between nine and eleven functions implemented in at least one unit in Nearly 22 percent had fewer than five functions, including 7.5 percent of hospitals with no functions a proportion essentially unchanged from 2010 (Exhibit 4). Consistent with our findings on adoption of EHR systems, we found that hospitals achieving a proxy standard of meaningful use were more likely to be large, nongovernment not-for-profit, teaching hospitals, members of a hospital system, and located in an urban area. A similar set of hospital characteristics was also associated with achieving meaningful use between 2010 and The most commonly adopted meaningful-use functions were as follows: a computerized system for recording patient sex, race, ethnicity, and date of birth (81.9 percent of hospitals had this implemented in at least one major clinical unit); patient medication allergy list (79.2 percent); vital signs (75.5 percent); clinical decision support (74.4 percent); and patient medication lists (74.2 percent). The functions that were least likely to be implemented were drug-drug and drug-allergy checks (41.7 percent in at least one major clinical unit); calculation of quality measures (46.8 percent); and providing patients with an electronic copy of their record (49.6 percent). Rural and small hospitals were less likely than their counterparts to have implemented each of the meaningful-use functions. Complete results are available in the Appendix. 7 Missing Functionalities To identify the most challenging EHR features for the 82 percent of US hospitals that could not meet our proxy standard for meaningful use in 2011, we first focused on those that were close to meeting the standard: hospitals with nine to eleven core functions. In this group, 47.1 percent could not generate quality measures; 40 percent had not implemented drug-drug or drug-allergy checks; and 35 percent could not provide patients with a copy of their record upon request (see the Appendix for complete results). 7 Nearly 30 percent of these advanced hospitals did not yet have computerized physician order entry for medications implemented in any unit of the hospital. Hospi- 4 Health Affairs MAY :5

5 Exhibit 2 Percentage Of US Hospitals With Comprehensive Or Basic Electronic Health Record Systems, Hospital characteristic Change, Change, a a Size Small 6.1% 8.3% 10.7% 20.8% Medium Large Location Rural Urban Teaching status Major Minor Nonteaching SOURCE Authors calculations of data from the American Hospital Association annual survey information technology supplement. NOTE All analyses were statistically weighted for potential nonresponse bias. a Percentage points. tals with eight or fewer functions were missing these key functions at a much higher rate. Discussion In the first nationally representative survey of hospital EHR system adoption since the onset of federal financial incentives under the HITECH Act, we found evidence of important progress. After three years of slow gains, hospitals adoption of these systems accelerated dramatically. By late 2011 we estimate that more than one in four hospitals had at least a basic system a far cry from the 9 percent that had achieved such a benchmark in The number of hospitals that could meet our proxy standard for stage 1 meaningful use increased more than fourfold from the prior year. Taken together with the number of hospitals that have registered for the meaningful-use incentive program as of February 2012 more than three thousand hospitals had registered for the Medicare or Medicaid incentive programs this represents very good news for federal policy makers. 1 Despite this good news, our findings raise a series of important concerns that urgently need to be addressed if EHRs are to have a substantial effect on improving the delivery of care across the health care system. First, although the number of hospitals meeting stage 1 meaningful-use criteria, using a very relaxed proxy definition, increased substantially, more than 80 percent of US hospitals still could not do so. Furthermore, data suggest that much of the increase in the proportion of hospitals meeting meaningful-use criteria in 2011 comes from hospitals that were already close to meeting the criteria in Given that the policy discussion is now shifting toward stage 2 meaningful use, these findings should increase the focus on helping the majority of US hospitals move forward on the adoption and use of EHR Exhibit 3 Characteristics Of US Hospitals, By Meaningful-Use Proxy Measure, 2011 Hospital characteristic No meaningful use, %(n = 2,139) Meaningful use, %(n = 507) All hospitals Size Small Medium Large Region Northeast Midwest South West Ownership For-profit Private nonprofit Public Teaching Major Minor Not teaching Location Rural Urban Member of hospital system No Yes SOURCE Authors calculations based on data from the American Hospital Association annual survey information technology supplement. NOTES All analyses were statistically weighted for potential nonresponse bias. All p values for differences were <0.001 except for region (p ¼ 0:011) and system membership (p ¼ 0:01). MAY :5 Health Affairs 5

6 Exhibit 4 Number Of Core Meaningful-Use Functions Implemented By US Hospitals, 2011 Percent SOURCE Authors calculations based on data from the American Hospital Association annual survey information technology supplement. NOTE All analyses were statistically weighted for potential nonresponse bias. systems required in stage 1. Second, the widening of previously described gaps in adoption of EHR systems based on hospital size, teaching status, location, and region of the country grew substantially, at least in absolute terms. However, the relative rates changed very little, which is also troublesome. The gap based on size, for instance, increased from 15 percent to 22 percent, and nonteaching and rural hospitals likewise fell further behind. 4,9 Third, we continue to see a large proportion of hospitals that are making very little progress, including one in thirteen institutions that have none of the twelve core measures included in our proxy standard for meaningful use. We suspect that the increase in adoption is in large part because of the onset of financial incentives. However, it is also possible that the findings may reflect the Rogers traditional S- shaped diffusion curve for the adoption of new innovations. If so, the nation may be at the inflection point where the curve turns sharply upward meaning that the number of electronic health record system adopters could grow quickly. 10 Unfortunately, our data do not allow us to be definitive. The findings of our study echo those of a recent study by HIMSS Analytics. Using the proprietary HIMSS database, the organization estimates that adoption of electronic health record systems and readiness for meaningful use have increased the most among large hospitals, academic medical centers, those located in urban areas, and those that are members of a multihospital setting. 6 The fact that our findings are broadly consistent with their report is reassuring. Policy Implications There are obvious, important policy implications to this work. First, we believe that federal policy makers need to redouble their efforts among hospitals that appear to be moving slowly or starting from a lower base rate of adoption. This slower rate could be because of several factors, including lack of access to capital and market forces. The regional extension center program, designed specifically to help vulnerable providers adopt and meaningfully use EHR systems, needs to demonstrate its effectiveness in this area. Whether the regional program can have a meaningful effect on these practices is unknown. Continued federal efforts are also needed to address the shortage of trained health information technology professionals. 11 In addition, the meaningful-use program has greatly increased demand for high-quality vendors of EHR systems and has strained market capacity. Small hospitals with limited access to capital may have a hard time competing with large, urban facilities in this marketplace. 12 The second important issue is related to the proposed criteria for stage 2 of meaningful use. Specifically, policy makers must decide how high to set the bar. It will be critically important for policy makers now finalizing the stage 2 criteria to take into account that more than 80 percent of hospitals could not meet stage 1 criteria in 2011 as assessed by a relaxed proxy measure. Third, the lack of infrastructure to support the requirements for health information exchange that are key to future stages of meaningful use is a major challenge, particularly for rural hospitals. 2 Even if these hospitals adopt systems capable of data exchange, without the supporting infrastructure they will be unable to meet the meaningful-use criteria. Just as important, the lack of infrastructure will make it more difficult for them to coordinate care and manage population health. Finally, there may need to be a special program designed for the sizable minority of hospitals that have almost no health information technology at all. Getting these institutions on board is likely to require resources and skills that go beyond what the current regional extension center program is designed to achieve. It is also possible that these hospitals have calculated that the cost of implementing an electronic health record system exceeds the potential benefits, or is less than the potential penalties they may incur. Either way, special efforts targeted toward these institutions may need to involve other stakeholders in order to bring these hospitals on board. For example, commercial payers could align their own incentive programs with meaningful-use standards or educate their members about the benefits of EHR technology to build consumer demand Health Affairs MAY :5

7 Conclusion Ensuring that hospitals and the patients they serve are not left behind is critical to achieving what Congress envisioned through the Health Information Technology for Economic and Clinical Health Act: a nationwide electronic health information infrastructure. Without the widespread adoption and meaningful use of health information technology, the broader goal of better care at lower cost, which information technology can enable in the right environment and where incentives are aligned, will be much harder to achieve. Although the uptick in the adoption of EHR systems is a positive sign, keeping focused on improving health system performance as the ultimate goal is critical. We examined changes in adoption rates of electronic health record systems and the ability to meet a proxy measure of meaningful use during the first year of the federal incentive program. Our findings show reasons for optimism, as the pace of adoption accelerated, but there are also areas of concern. Although rates of adoption are increasing rapidly among large hospitals, small, rural, and nonteaching institutions continue to fall behind. Federal policy makers need to take steps to ensure that the stage 2 criteria do not further exacerbate this divide going forward. This work was supported by a grant from the Robert Wood Johnson Foundation. [Published online April 24, 2012.] NOTES 1 Centers for Medicare and Medicaid Services. EHR Incentive Program [Internet]. Baltimore (MD): CMS; 2012 Jan [cited 2012 April 13]. Available from: Legislation/EHRIncentive Programs/downloads//Monthly_ Payment_Registration_Report_ Updated.pdf 2 DesRoches CM, Painter M, Jha AK, editors. Health information technology in the United States: driving towards delivery system reform. Princeton (NJ): Robert Wood Johnson Foundation; Jha AK, DesRoches CM, Campbell EG, Donelan K, Rao SR, Ferris TG, et al. Use of electronic health records in US hospitals. N Engl J Med. 2009;360(16): Jha AK, DesRoches CM, Kralovec PD, Joshi MF. A progress report on electronic health records in US hospitals. Health Aff (Millwood). 2010;29(10): Jha AK, Burke MF, DesRoches CM, Joshi MF, Karlovec PD, Campbell EG, et al. Progress toward meaningful use: hospitals adoption of electronic health records. Am J Manag Care. 2011;17(12 Spec. No.): SP HIMSS Analytics. Annual study [Internet]. Chicago (IL): HIMSS; [cited 2012 Apr 3]. Available from: annualstudy.aspx 7 To access the Appendix, click on the Appendix link in the box to the right of the article online. 8 Woodbridge JM. Inverse probability weighted estimation for general missing data problems. J Econom. 2007;141(2): Jha AK, DesRoches CM, Shields A, Miralles PD, Zheng J, Rosenbaum S, et al. Evidence of an emerging digital divide among hospitals that care for the poor. Health Aff (Millwood). 2009;28(6): Rodgers EM. Diffusion of innovation. New York (NY): Free Press; Office of the National Coordinator for Health Information Technology. American Recovery and Reinvestment Act of 2009; information technology professionals in health care: program of assistance for university-based training. Washington (DC): Department of Health and Human Services; Gale K. Meaningful use leading to improved outcomes. Orem (UT): KLAS; Jain SH, Seidman J, Blumenthal D. How health plans, health systems, and others in the private sector can stimulate meaningful use. Health Aff (Millwood). 2012;29(9): MAY :5 Health Affairs 7

8 ABOUT THE AUTHORS: CATHERINE M. DESROCHES, CHANTAL WORZALA, MAULIK S. JOSHI, PETER D. KRALOVEC & ASHISH K. JHA Catherine M. DesRoches is a senior researcher at Mathematica Policy Research. In this month s Health Affairs, Catherine DesRoches and coauthors report on their analysis of 2011 national survey data, in which they found that the share of hospitals with any electronic health record (EHR) system increased from 15.1 percent in 2010 to 26.6 percent in 2011, and those with a comprehensive system rose from 3.6 percent to 8.7 percent. At the same time, small, nonteaching, and rural hospitals continued to adopt the systems more slowly than others. The authors suggest policy responses, such as a special federal program for the sizable minority of hospitals that have almost no health information technology at all. DesRochesisaseniorscientistat Mathematica Policy Research whose work has included studies of EHR adoption and use among hospitals, physicians, and registered nurses. Prior to joining Mathematica in 2010, she spent ten years as a researcher and faculty member at Harvard Medical SchoolandtheHarvardSchoolof Public Health. DesRoches holds a doctorate in public health from Columbia University. Chantal Worzala is director of policy at the American Hospital Association. Chantal Worzala is director of policy at the American Hospital Association. Her primary area of focusishealthinformation technology use and policy development. Worzala received a doctorate in health services research from the Johns Hopkins University and a master s degreein public administration from Princeton University. Maulik S. Joshi is president of the Health Research and Educational Trust. Maulik Joshi is president of the Health Research and Educational Trust, senior vice president of research at the American Hospital Association, and editor-in-chief of the Journal for Healthcare Quality. He has a doctorate in public health and a master s degree in health services administration from the University of Michigan. Peter Kralovec has served as the senior director of the Health Care Data Center for the American Hospital Association s Health Forum since its inception in He is currently the principal investigator on a grant from the Department of Health and Human Services Office of the National Coordinator for Health Information Technology to measure the 2011 adoption rate of EHRs in US hospitals. Kralovec received a bachelor s degree in sociology from the University of Cincinnati. Ashish K. Jha is an associate professor of health policy at the Harvard School of Public Health. Ashish Jha is an associate professor of health policy at the Harvard School of Public Health and an associate professor of medicine at Harvard Medical School. His research focuses on the quality of care provided by health care systems, health care disparities as a marker of poorquality care, and health information technology as a potential solution for reducing disparities and improving care. Jha has a medical degree and a master s degree in public health from Harvard University. Peter D. Kralovec is the senior director ofthehealthcare Data Center for the American Hospital Association s Health Forum. 8 Health Affairs MAY :5