Meaningful Use of Health Information Technology by Rural Hospitals

ORIGINAL ARTICLE Meaningful Use of Health Information Technology by Rural Hospitals Jeffrey McCullough, PhD; Michelle Casey, MS; Ira Moscovice, PhD; & Michele Burlew, MS Division of Health Policy and Management, School of Public Health, University of Minnesota, Minneapolis, Minnesota Abstract This research was supported by the federal Office of Rural Health Policy, Health Services and Resources Administration, PHS Grant No. U1CRH03717. For further information, contact: Jeffrey McCullough, PhD, Division of Health Policy and Management, School of Public Health, University of Minnesota, 420 Delaware St. SE, MMC 729, Minneapolis, MN 55455; e-mail mccu0056@umn.edu. doi: 10.1111/j.1748-0361.2010.00359.x Purpose: This study examines the current status of meaningful use of health information technology (IT) in Critical Access Hospitals (CAHs), other rural, and urban US hospitals, and it discusses the potential role of Medicare payment incentives and disincentives in encouraging CAHs and other rural hospitals to achieve meaningful use. Methods: Data from the American Hospital Association (AHA) Annual Survey IT Supplement were analyzed, using t tests and probit regressions to assess whether implementation rates in CAHs and other rural hospitals are significantly different from rates in urban hospitals. Findings: Of the many measures we examined, only 4 have been met by a majority of rural hospitals: electronic recording of patient demographics and electronic access to lab reports, radiology reports, and radiology images. Meaningful use is even less prevalent among CAHs. We also find that rural hospitals lag behind urban institutions in nearly every measure of meaningful use. These differences are particularly large and significant for CAHs. Conclusion: The meaningful use incentive system creates many challenges for CAHs. First, investments are evaluated and subsidies determined after adoption. Thus, CAHs must accept financial risk when adopting health IT; this may be particularly important for large expenditures. Second, the subsidies may be low for relatively small expenditures. Third, since the subsidies are based on observable costs, CAHs will receive no support for their intangible costs (eg, workflow disruption). A variety of policies may be used to address these problems of financial risk, uncertain returns in a rural setting, and limited resources. Key words Health information technology, Medicare, rural hospitals. Health information technology (IT) is a potential vehicle for improving the quality and safety of health care; however, rural communities face many challenges in health IT adoption including financial constraints, limited access to capital, inadequate infrastructure, and limited health IT workforce. 1 Reflecting these challenges, small rural hospitals in the United States have lagged behind larger urban facilities in implementation of health IT, including adoption of electronic medical records (EMRs) and computerized provider order entry (CPOE) systems. 2-9 Additional organizational factors, including private nonprofit ownership, teaching hospital status, system affiliation, and accreditation, have also been found to be associated with higher health IT adoption rates. 4,5,7 The American Recovery and Reinvestment Act of 2009 (ARRA) established financial incentives for hospitals, including Critical Access Hospitals (CAHs), rural hospitals with 25 or fewer beds, and other rural hospitals, to become meaningful users of health IT. The legislation requires use of certified electronic health record (EHR) technology that provides for the exchange of health information to improve the quality of health care and The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association 329

Meaningful Use of Health IT by Rural Hospitals McCullough et al. submission of information on clinical quality measures and other measures selected by the Secretary of Health and Human Services. The EHR must include patient demographic and clinical health information and have the capacity to provide clinical decision support, support physician order entry, capture and query information relevant to health care quality, and exchange and integrate electronic health information with other sources. 10 The Centers for Medicare and Medicaid Services (CMS) issued a final rule in July 2010 establishing criteria for eligible professionals and hospitals that participate in the Medicare and Medicaid programs to qualify for ARRA incentive payments based on their meaningful use of certified EHR technology. 11 The Office of the National Coordinator for Health IT also issued a final rule setting forth an initial set of standards, implementation specifications, and certification criteria for EHR technology. 12 The CMS rule outlines a 3-stage approach to implementing meaningful use. 11 The Stage 1 criteria focus on electronically capturing health information in a coded format; using the information to track key clinical conditions and for care coordination purposes; implementing clinical decision support tools to facilitate disease and medication management; and reporting clinical quality measures and public health information. CMS anticipates updating the criteria on a biennial basis. Stage 2 criteria (beginning in 2013) will focus on using health IT for continuous quality improvement at the point of care and the exchange of information. Stage 3 criteria (beginning in 2015) will focus on promoting improvements in quality, safety and efficiency, decision support for national highpriority conditions, patient access to self-management tools, access to comprehensive patient data, and improving population health. This study examines the current status of meaningful use of health IT in CAHs, other rural, and urban US hospitals, and it discusses the potential role of Medicare payment incentives and disincentives in encouraging CAHs and other rural hospitals to achieve meaningful use. The CMS rule includes 24 objectives, including 14 core objectives and a menu set of 10 other objectives. Hospitals must meet the 14 core objectives and 5 objectives from the menu set to achieve Stage 1 meaningful use. Specific measures are aligned with each objective. The focus of our analysis is on the clinically related objectives in the rule (eg, recording of patient data in EHRs, use of drug alerts and clinical guidelines, implementation of CPOE, use of data for quality improvement and measurement, and clinical data exchange and coordination). The remaining objectives address data privacy and security, and population health. While these objectives are important, they are of less concern for this analysis, as the issues for hospitals are similar regardless of rurality. In addition, previous research has found that CAHs and rural hospitals have much higher technology use rates for administrative applications than for clinical applications. 2 Methods Data for this study came from the American Hospital Association (AHA) Annual Survey IT Supplement for Fiscal Year 2007. These data were linked to the AHA Annual Survey (Fiscal Year 2007), which provides detailed information on hospital characteristics. The sample comprised acute care, general, US hospitals. We further excluded non-cah federal hospitals, as they face a very different set of IT-related regulations and incentives. Of these 4,646 hospitals, 2,829 (61%) responded to the AHA IT Supplement. Hospitals were defined as CAH, other rural, and urban. Response rates were similar across these categories; however, responding hospitals were slightly larger and more likely to be nonprofit institutions. Thus, we explore the role of survey response bias below. The AHA IT Supplement provides extensive data regarding the adoption and utilization of health IT. We matched individual survey measures to meaningful use criteria in the CMS rule. While the survey questions do not perfectly reflect regulatory rules, they do provide a close approximation for many core measures. We also included some criteria that do not directly link to a single core measure but which may be important steps in achieving meaningful use. These distinctions are noted in Table 1 and discussed in the text below. These data describe not only the presence of a technical capability (eg, medication order entry or decision support) but also the extent of implementation. In the following sections (Tables 1-4 in particular), we focus on capabilities fully implemented across all units. While we do not report these results for brevity, we also examined other stages of implementation and found similar differences. For each type of technology, we first test whether prevalence rates in CAH and other rural hospitals are significantly different from urban hospital prevalence rates, using t tests. To test for differences across hospital types in the 5 ordered categories of the percentage of patients for whom medication orders, lab orders and clinical documentation are written electronically, we employed a set of 3 ordered probit regressions. We also modeled the count of criteria achieved by CAHs, other rural, and urban hospitals, while controlling for individual hospital characteristics. This analysis used a count data (negative binomial) regression technique. Finally, we consider a variety of alternative tests and specifications described in the robustness section. 330 The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association

McCullough et al. Meaningful Use of Health IT by Rural Hospitals Table 1 Implementation of Health Information Technologies Related to Meaningful Use Objectives by Hospital Type and Location CAHs Other Rural Urban Recording of Patient Data in EHRs: Electronic Clinical Documentation Patient demographics 61% 78% 88% Physician notes, 9% 13% 12% Nursing assessments, 20% 37% 44% Problem lists 14% 27% 32% Medication lists 24% 42% 56% Discharge summaries 28% 44% 55% Use of Computerized System for Drug Alerts and Clinical Guidelines Drug-drug interaction alerts 26% 46% 55% Drug-allergy alerts 27% 47% 56% Drug dosing support 15% 32% 38% Clinical guidelines 7% 16% 21% Clinical reminders 10% 24% 29% Electronic Access to Test Results and Images Lab reports 52% 78% 90% Radiology reports 53% 78% 91% Radiology images 46% 68% 81% Diagnostic test results 23% 46% 67% Diagnostic test images 16% 31% 48% Consultant reports 22% 47% 61% Implementation of CPOE Laboratory tests 17% 16% 22% Radiology tests 17% 17% 22% Medications 11% 13% 20% Consultation requests 6% 10% 18% Nursing orders 13% 13% 23% Patient Access to Online Information from Hospital Electronic System Discharge summary 8% 10% 14% Test results 11% 14% 19% Operative notes 5% 9% 13% Medication lists 8% 11% 15% Denotes significant differences between (1) CAHs and urban hospitals and (2) other rural and urban hospitals at: P <.05; P <.01; and P <.001. Indicates technology that is aligned with core meaningful use objectives. The related core meaningful use objective is to record and chart changes in vital signs, which would generally be a nursing assessment function. Results Use of Electronic Health Records Meaningful use will require a hospital to use an electronic health record system that is certified according to standards adopted by the Office of the National Coordinator for Health IT (ONC). Overall, CAHs (P <.001) and other rural hospitals (P <.01) are significantly less likely than urban hospitals to fully use an EHR system, with only 3% of CAHs and 5% of rural hospitals reporting full use, compared to 9% of urban hospitals. In addition, some of the EHR systems currently being used may require upgrading or other modifications to meet the ONC certification standards. Recording of Patient Data Elements in EHRs Meaningful use core objectives will require a hospital to electronically record patient demographic information, vital signs, and smoking status (for patients 13 years old or older). In addition, the hospital must maintain: (1) an up-to-date problem list of current and active diagnoses; (2) an active medication list; and (3) an active medication allergy list. The measures for the diagnosis and medication objectives require that at least 80% of patients have at least one entry recorded as structured data or an indication that the patient had none (eg, no medications); at least 50% of patients must have demographic information, vital signs, and smoking status recorded. While nursing assessment and physician notes are not explicit meaningful use criteria, they are potentially related to the electronic documentation of patient data. Table 1 describes the differences between CAHs, other rural, and urban hospitals. In each case, CAHs lag urban hospitals in the automated capture of patient data. For example, 61% of CAHs automated the recording of patient demographic information, while this task was fully Table 2 Percent of Patients for Whom Orders Are Written Electronically Percent of Patients for Whom Orders Are Lab Orders Medication Orders Clinical Documentation Written Electronically CAHs Other Rural Urban CAHs Other Rural Urban CAHs Other Rural Urban 0% 63% 56% 46% 75% 67% 54% 70% 50% 42% 1%-25% 5% 10% 12% 8% 13% 15% 9% 20% 24% 26%-50% 3% 5% 4% 3% 3% 5% 4% 7% 9% 51%-90% 6% 8% 8% 5% 6% 7% 7% 11% 13% 91%-100% 23% 21% 29% 10% 10% 20% 10% 12% 12% Ordered probits were used to test whether CAHs and other rural hospitals were significantly different from urban hospitals. CAHs and other rural hospitals had significantly lower electronic order rates for each category. The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association 331

Meaningful Use of Health IT by Rural Hospitals McCullough et al. Table 3 Electronic Clinical Data Exchange by Hospitals CAHs Other Urban Hospital Electronically Exchanges With Hospitals in Its System: Clinical care record 75% 61% 82% Lab results 68% 55% 78% Radiology reports 78% 62% 85% Medication lists 62% 54% 78% Discharge summaries 74% 62% 81% Hospital Electronically Exchanges With Hospitals Outside Its System: Clinical care record 19% 19% 12% Lab results 22% 22% 14% Radiology reports 17% 16% 10% Medication lists 32% 27% 15% Discharge summaries 20% 19% 11% Hospital Electronically Exchanges With Ambulatory Providers Outside Its System: Clinical care record 39% 49% 43% Lab results 44% 58% 54% Radiology reports 25% 34% 32% Medication lists 45% 58% 52% Discharge summaries 38% 51% 47% This table has smaller hospital sample sizes primarily due to nonresponse from hospitals not in systems. Denotes significant differences between (1) CAHs and urban hospitals and (2) other rural and urban hospitals at: P <.05; P <.01; and P <.001. automated in 78% and 88% of other rural and urban hospitals, respectively. Differences were large and statistically significant for each case except for physician notes, which were not directly addressed by the meaningful use rule. Table 4 Effect of Hospital Characteristics on Count of Fully Implemented Meaningful Use Criteria Hospital Characteristics Relative Rates (Standard Error) Critical access hospital 0.62 (0.03) Other rural hospitals 0.92 (0.04) Staffed beds (in 100 seconds) 1.10 (0.02) Staffed beds squared (in 100 seconds) 0.996 (0.00) Government-owned 1.14 (0.06) Not-for-profit 1.22 (0.06) Multihospital system member 1.21 (0.04) Denotes significant differences between (1) CAHs and urban hospitals and (2) other rural and urban hospitals at: P <.05; P <.01; and P <.001. Use of Drug Alerts and Clinical Guidelines/Reminders Meaningful use core objectives will require a hospital to implement drug-drug and drug-allergy checks; drugformulary checks are part of the menu set. Another core objective for hospitals is to implement one clinical decision support rule related to a high-priority condition and track compliance with those rules. Just over one-fourth of CAHs report having a computerized system for drug-drug interaction alerts and drugallergy alerts fully implemented across all hospital units, compared to about 46% of other rural hospitals and 55% of urban hospitals (Table 1). Clinical guidelines are much less commonly implemented, with fewer than 7% of CAHs and 16% of other rural hospitals reporting full implementation. While drug-dosing support is not explicitly required, it is an important and related decision support capability. Electronic Access to Test Results and Images The meaningful use menu set includes an objective requiring a hospital to incorporate lab test results as structured data into the EHR. Of the clinical lab tests ordered whose results are in a positive/negative or numerical format, more than 40% must be incorporated as structured data. Over half of CAHs and three-fourths of other rural hospitals have electronic access to lab test results fully implemented across all hospital units (Table 1). Electronic access to radiology reports is similar, while electronic access to diagnostic test results lags behind. Use of CPOE A meaningful use core objective will require a hospital to use CPOE for medication orders directly entered by any licensed health care professional who can enter orders into the medical record. The Stage 1 measure for the CPOE objective is that 30% of all patients with at least one medication must have a medication order entered by CPOE. CPOE adoption is relatively infrequent; less than onefourth of urban hospitals have implemented CPOE capabilities for each service type. Once again, critical access and other rural hospitals lag behind their urban counterparts. CAHs and other rural hospitals are most likely to have implemented CPOE for lab and radiology tests across all hospital units (Table 1). 332 The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association

McCullough et al. Meaningful Use of Health IT by Rural Hospitals Patient Access to Online Information A meaningful use core objective will require hospitals to provide patients, upon request, with an electronic copy of their health information (including diagnostic test results, problem list, medication lists, and allergies). Providers must also provide to patients an electronic copy of their discharge instructions and procedures at the time of discharge. Very few rural or urban hospitals have online systems that allow patients to view discharge summaries, test results, operative notes, or medication lists online (Table 1). However, these numbers may underestimate electronic data dissemination capabilities as the rule permits alternative formats (eg, via CD, USB, etc). Use of Electronic Orders for Patients Several meaningful use objectives require electronic orders for patients. Specific thresholds vary across clinical functions. For each order category, CAHs and other rural hospitals had significantly fewer electronically written orders than urban hospitals. More than half of CAHs and other rural hospitals report that lab orders are entered electronically for 0% of their patients (Table 2). Similarly, three-fourths of CAHs and two-thirds of other rural hospitals report that medication orders are entered electronically for 0% of their patients, and over two-thirds of CAHs and half of other rural hospitals report that clinical documentation is done electronically for 0% of their patients. Quality Measurement and Reporting Another meaningful use core objective will require all hospitals, including CAHs, to report data on hospital quality measures to CMS (and to states, for hospitals that are eligible for Medicaid EHR incentives). The CMS rule requires hospitals to report summary data to CMS on a set of clinical quality measures starting in the 2011 payment year, with electronic submission of the measures using certified EHR technology to CMS for the 2012 payment year. This requirement is the first time that CMS has specifically required CAHs to submit quality measures. Since 2004, acute care hospitals paid under the Medicare Prospective Payment System (PPS) have been required to submit quality measure data to CMS and have the data publicly reported on the CMS Hospital Compare website. Data submission is a requirement for hospitals to receive the market basket update to their PPS reimbursement rate. CAHs are exempt from this requirement, as they are reimbursed by Medicare on a reasonable cost basis rather than through the PPS system. Approximately 70% of CAHs voluntarily reported data on at least one quality measure to Hospital Compare for 2008 discharges. 13 However, in the AHA IT Supplement, only 15% of CAHs, 24% of other rural, and 27% of urban hospitals report that their electronic systems allow them to automatically generate the CMS quality measures, suggesting that electronic submission of the measures required by the meaningful use rule may pose an additional challenge for many hospitals. Electronic Clinical Data Exchange A meaningful use core objective will require hospitals to have the capability to electronically exchange key clinical information (eg, discharge summary, procedures, problem list, medication list, allergies, diagnostic test results) among providers and patient-authorized entities. Electronic exchange of clinical information is a key component of care coordination across providers and is an especially important issue for rural patients. For all types of hospitals, electronic exchange of clinical data is most common with hospitals inside their own system (Table 3). Radiology reports and discharge summaries are exchanged most frequently. While CAHs are less likely than urban hospitals to share data with other hospitals in their system, they are more likely to share data with hospitals outside of their system. CAHs are less likely than urban hospitals to exchange data with ambulatory providers outside their system, while differences between other rural and urban hospitals are not significant. Achieving Meaningful Use Ultimately, the benefits of health IT may be derived only from the adoption and implementation of complementary clinical applications. For example, drug interaction alerts depend on up-to-date medication lists, while drugdosing support systems are more effective with accurate patient demographic data. Consequently, meaningful use incentives are focused on the adoption and implementation of a wide range of capabilities. This implies that the number of meaningful use measures met is an important measure of meaningful use objectives. Figure 1 describes the numbers of meaningful use measures met by CAHs, other rural, and urban hospitals. The 21 criteria used include those reported in Table 1 except for physician notes and online access measures. The electronic order entry measures (Table 2) were not included as they capture information similar to the CPOE measures. On average, CAH and other rural hospitals implemented 5.1 and 8.1 criteria, while urban institutions had met 10.2 criteria. The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association 333

Meaningful Use of Health IT by Rural Hospitals McCullough et al. Figure 1 Count of Meaningful Use Objectives Met by Hospital Type While these differences are relatively large, IT adoption and implementation decisions depend on many factors other than critical access status and geographic location. We attempt to disentangle the relative roles of hospital characteristics on meaningful use. We regress the count of fully implemented criteria on type and location of hospital, hospital size, hospital governance, and system membership. This model is estimated by negative binomial regression. The parameters, reported in Table 4, are expressed in relative rates. For example, CAHs implement criteria at 62% of the urban hospital implementation rate, while the difference between urban and other rural hospitals is not statically significant. Meaningful use also increases with hospital size; the number of implemented criteria increased by 10% for each additional 100 staffed inpatient beds. However, this size effect is slightly diminished (0.4% per 100 beds) for larger institutions. Hospital governance also played an important role in IT implementation. Not-for-profit and government-owned hospitals implemented 22% and 14% more criteria than their for-profit counterparts. Similarly, multihospital system members implemented 21% more criteria than stand-alone hospitals. These findings demonstrate a marked difference between CAHs and other institutions. While smaller, forprofit, and stand-alone institutions meet fewer meaningful use criteria, no other characteristic has as large an effect as critical access status. To illustrate this point, holding all else equal, the difference between a CAH and urban hospital is equivalent to a 515-bed size difference. Robustness and Limitations We applied a series of specification and robustness tests to our empirical analysis. In particular, our regression results could be sensitive to our criteria definitions, unobserved hospitals heterogeneity, and model specification. Furthermore, all of our empirical findings could be subject to selection bias in the AHA IT Supplement response rates. We explored each of these issues in our analysis. Since we measure meaningful use imperfectly, our criteria definitions could drive our results. Consequently, we estimated a series of models using different alternative count measures. For example, we estimated one set of models including only the core measures (noted in Table 1) that most closely corresponded with the core meaningful use definitions. An alternative model excluded patient access to online information measures. Other specifications examined counts based on partial, rather than full, implementation. Alternative measures of hospital characteristics were also considered. Additional characteristics included more detailed hospital governance measures (eg, nonprofit religious vs nonprofit secular) and payer mix (eg, percentage Medicare and Medicaid). In each case, the results were similar to those in Table 4. 334 The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association

McCullough et al. Meaningful Use of Health IT by Rural Hospitals Count data models are often sensitive to estimation technique. We chose the negative binomial regression model, as the data were overdispersed. Alternative models using generalized negative binomial and gamma distributions were also estimated. Similarly, we considered specifications with separate shape parameters for CAH, other rural, and urban institutions. While we cannot reject these alternative specifications, their parameter estimates were nearly identical to those reported in Table 4. The AHA IT Supplement response rate was 61%. We found that respondents were, on average, larger and more likely to be nonprofit institutions. Response bias could have affected all of our empirical findings. We explored this issue by predicting meaningful use for nonresponding hospitals essentially using the full AHA sample to measure the bias caused by observable differences between responders and nonresponders. Ultimately, the magnitude of our estimated selection bias was extremely small. For example, the reported parameter estimates might overstate meaningful use by one-tenth of an application and had almost no effect on the difference between CAH and urban hospitals. While this is reassuring, this approach faces important limitations. Our selection correction strategy is based only on observable characteristics from the general AHA survey unobserved factors could also affect IT Supplement responses. Similarly, unobserved differences across hospitals such as their financial status might influence meaningful use (ie, technology adoption might be endogenous). Incentives to Achieve Meaningful Use The CMS rule includes both incentives and penalties for meaningful use (Table 5). Incentive payments begin in 2013 and gradually fall through 2016. Payments are proportional to a hospital s Medicare share and increase in volume. Hospitals are eligible for similar incentives under Medicaid. Beginning in 2015 hospitals that do not achieve meaningful use will face penalties decreases in Medicare payment updates. Our findings suggest that the meaningful use compliance costs will be higher for small, critical access, and forprofit hospitals. The final rule has several features that may mitigate the financial burden for small and CAHs. First, much of the subsidy is independent of hospital size; it is a function of the percentage, rather than number, of Medicare patients. This likely will provide some additional support to relatively small hospitals CAHs face several important differences. Their Medicare reimbursement is based on 101% of reasonable Medicare-related costs rather than prospective payment. Consequently, CAHs will receive adoption subsidies based on allowable IT adoption costs. The subsidy will be based on the hospital s Medicare share plus 20%, with a maximum subsidy of 100% of allowable costs. This payment system may be risky for CAHs since reasonable costs are determined after adoption. Thus, hospitals may not be certain of their subsidy until after IT adoption decisions are made. This uncertainty may also result in higher capital costs for CAHs. Discussion We examined the prevalence of meaningful use measures for CAH, other rural, and urban hospitals. Of the many measures we examined, only 4 have been met by a majority of rural hospitals: electronic recording of patient demographics and electronic access to lab reports, radiology reports, and radiology images. Meaningful use is even less prevalent among CAHs. We also find that rural hospitals lag behind urban institutions in nearly every measure of meaningful use. These differences are particularly large and significant for CAHs. These findings suggest that the costs of meaningful use compliance may be particularly high for CAHs and other rural hospitals. This conclusion is supported by on-theground assessments from experts with experience implementing health IT in small rural hospitals and from rural hospitals themselves. 14 For example, the AHA reported in 2010 that 66% of CAHs and 56% of rural hospitals did not expect to be able to demonstrate meaningful use in 2015. 15 Small rural hospitals may face challenges beyond their limited initial technology stock. The benefits of health IT have largely been studied in academic institutions. Health IT may, for example, exhibit returns to scale or be more valuable for patients with relatively severe illnesses. 6 Rural hospitals may also have limited access to employees with IT-related skills. The final CMS rule allows CAHs to expense a wide range of health IT expenditures and they may do so shortly after investments are made. However, these expenditures must be reviewed prior to reimbursement; thus, CAH subsidies are uncertain and risky. This risk may also affect financing costs for CAHs health IT investments. The final rule has addressed many rural hospital concerns. Meaningful use requirements are now more streamlined and flexible. These changes will make it easier for rural hospitals to catch up to their urban counterparts. CAHs are now eligible for Medicaid subsidies. Similarly, CAHs can receive reimbursement for a wider range of IT expenditures. A variety of policies may be used to address small rural hospitals problems of financial risk, uncertain returns The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association 335

Meaningful Use of Health IT by Rural Hospitals McCullough et al. Table 5 Incentives for Achieving Meaningful Use of EHRs and Penalties for Failure CAHs Rural and urban PPS hospitals Medicare Incentives For cost reporting periods beginning after FY 2010 but before FY 2016, the CAH may receive an incentive payment equal to the product of (1) the reasonable costs incurred for the purchase of certified EHR technology (acquisition costs minus depreciation and interest) in that cost reporting period and previous periods to the extent they have not been fully depreciated and (2) the CAH s Medicare share plus 20 percentage points, not to exceed 100%. Penalties Reasonable cost reimbursement is reduced from 101% to 100.66% in 2015, 100.33% in 2016, and 100% in 2017 for CAHs that fail to achieve meaningful use. Annual hardship exemptions are available on a case-by-case basis not to exceed 5 years. Incentives For each eligible hospital, the incentive payment for each payment year is calculated as the product of: (1) an initial amount; (2) the Medicare share; and (3) a transition factor applicable to that payment year. The initial amount is a base amount of $2,000,000 plus $200 per discharge for the 1,150th through the 23,000th discharge. The Medicare share is a fraction based on estimated Medicare inpatient bed days, divided by estimated total inpatient bed-days, modified by charges for charity care. The applicable transition factor equals 1 for the first, 3 / 4 for the second, 1 / 2 for the third, and 1 / 4 for the fourth payment year for an eligible hospital for which the first payment year is 2013. Penalties Eligible hospitals that are not meaningful users will receive 1 / 4, 1 / 2,and 3 / 4 reductions of their market basket updates in FY 2015, FY 2016, FY 2017, and subsequent years. Medicaid Incentives CAHs are considered eligible hospitals for the purpose of qualifying for Medicaid incentive payments, using the same methodology as PPS hospitals. Penalties None Incentives To be eligible for Medicaid incentive payments, Medicaid patients must constitute a minimum of 10% of a hospital s patients. The payments are largely based on the methodology applied to Medicare incentive payments. States may pay eligible hospitals up to 100% of an aggregate EHR hospital incentive amount provided over a minimum of a 3-year period and a maximum of a 6-year period. The aggregate EHR hospital incentive amount is calculated using an overall EHR amount (base amount and discharge related amount calculated similarly as Medicare) multiplied by the Medicaid share. The Medicaid share is a fraction expressed as estimated Medicaid inpatient bed-days plus estimated Medicaid managed care inpatient bed-days divided by estimated total inpatient bed-days multiplied by ([estimated total charges minus charity care charges] divided by estimated total charges). Penalties None Source: Medicare and Medicaid Programs; Electronic Health Record Incentive Program; Final Rule, July 13, 2010. in a rural setting, and limited resources. First, the government could reduce the risk faced by CAHs under the current system. Risk may be reduced by providing clear rules on what health IT costs will be allowable or through providing ex ante expenditure review. Second, if the expected benefits of health IT are smaller in rural settings, we might explore alternative policy options. The 2015 penalties for not achieving meaningful use could be delayed. This would allow more time for IT costs to fall and for the development of information systems that are targeted to the needs of rural providers. Third, adoption barriers may remain even if IT value is high for rural hospitals. This may be addressed through technical assistance. While CMS has awarded supplemental funding to Regional Extension Centers to work with CAHs, the magnitude of the funds available to rural hospitals remains small. References 1. Institute of Medicine. Committee on the future of rural health care quality, board on health care services. Quality Through Collaboration: The Future of Rural Health Care. Washington, DC: National Academy Press; 2005. 2. Casey M, Klingner J, Gregg W, et al. The Current Status of Health Information Technology Use in CAHs. Flex Monitoring Team Briefing Paper No. 11. Minneapolis, MN: University of Minnesota; 2006. 336 The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association

McCullough et al. Meaningful Use of Health IT by Rural Hospitals 3. Culler S, Atherly A, Walczak S, et al. Urban-rural differences in the availability of hospital information technology applications: a survey of Georgia hospitals. J Rural Health. 2006;22(3):242-247. 4. American Hospital Association. Continued Progress: Hospital Use of Health Information Technology, 2007. Available at: http://www.aha.org/aha/content/ 2007/pdf/070227-continuedprogress.pdf. Accessed November 28, 2010. 5. Jha AK, DesRoches CM, Campbell EG, et al. Use of electronic health records in U.S. hospitals. N Engl J Med. 2009;360(16):1628-1638. 6. McCullough J, Casey M, Prasad S, Moscovice I. The effect of computerized physician order entry and electronic medical record adoption on quality in US hospitals. Health Affairs. 2010;29(4):647-654. 7. Menachemi N, Burke D, Clawson A, Brooks R. Information technologies in Florida s rural hospitals: does system affiliation matter? J Rural Health. 2005;21(3):263-268. 8. Ward MM, Jaana M, Bahensky J, Vartak S, Wakefield DS. Clinical information system availability and use in urban and rural hospitals. JMedSyst. 2006;30(6):429-438. 9. Jha AK, DesRoches CM, Kralovec PD, Joshi MS. A progress report on electronic health records in US hospitals. Health Affairs. 2010;29(11):1951-1957. 10. American Recovery and Reinvestment Act of 2009. Available at: http://frwebgate.access.gpo.gov/cgi-bin/ getdoc.cgi?dbname=111 cong bills&docid=f:h1enr.pdf. Accessed December 8, 2009. 11. Department of Health and Human Services. Centers for Medicare & Medicaid Services. 42 CFR Parts 412, 413, 422, and 495. CMS-0033-F RIN 0938-AP78. Medicare and Medicaid Programs; Electronic Health Record Incentive Program. Federal Register. 2010;75(144): 44314-44588. 12. Department of Health and Human Services, Office of the Secretary. 45 CFR Part 170 RIN 0991-AB58. Health information Technology: Initial Set of Standards, Implementation Specifications, and Certification Criteria for Electronic Health Record Technology. Federal Register. 2010;75(144):44589-44654. 13. Casey M, Burlew M, Moscovice I. Critical Access Hospital Year 5 Hospital Compare Participation and Quality Measure Results. Flex Monitoring Team Briefing Paper No. 26. 2010. Available at: http://flexmonitoring.org/ documents/briefingpaper26-hospitalcompare5.pdf. Accessed November 28, 2010. 14. Wenzlow L. Rural Wisconsin Health Cooperative Comments on CMS Proposed HIT Incentive Rule, 2010. Available at: http://www.worh.org/hit/2010/02/ recommendation-to-cms-on-proposed-hit-incentive-rulea-rural-perspective/. Accessed November 28, 2010. 15. American Hospital Association. The road to meaningful use: what it takes to implement electronic health record systems in hospitals. AHA Trendwatch. 2010. The Journal of Rural Health 27 (2011) 329 337 c 2011 National Rural Health Association 337