Skip to main content
NCBI home page
Journal of the American Medical Informatics Association : JAMIA logoLink to Journal of the American Medical Informatics Association : JAMIA
. 2018 May 30;25(9):1189–1196. doi: 10.1093/jamia/ocy058

Hospitals’ adoption of intra-system information exchange is negatively associated with inter-system information exchange

Joshua R Vest 1,2,✉, Kosali Simon 3
PMCID: PMC7646893  PMID: 29860502

Abstract

Introduction

U.S. policy on interoperable HIT has focused on increasing inter-system (ie, between different organizations) health information exchange. However, interoperable HIT also supports the movement of information within the same organization (ie, intra-system exchange).

Methods

We examined the relationship between hospitals’ intra- and inter-system information exchange capabilities among health system hospitals included in the 2010-2014 American Hospital Association’s Annual Health Information Technology Survey. We described the factors associated with hospitals that adopted more intra-system than inter-system exchange capability, and explored the extent of new capability adoption among hospitals that reported neither intra- or inter-system information capabilities at baseline.

Results

The prevalence of exchange increased over time, but the adoption of inter-system information exchange was slower; when hospitals adopt information exchange, adoption of intra-system exchange was more common. On average during our study period, hospitals could share 4.6 types of information by intra-system exchange, but only 2.7 types of information by inter-system exchange. Controlling for other factors, hospitals exchanged more types of information in an intra-system manner than inter-system when the number of different inpatient EHR vendors in use in health system is larger.

Conclusion

Consistent with the U.S. goals for more widely accessible patient information, hospitals’ ability to share information has increased over time. However, hospitals are prioritizing within-organizational information exchange over exchange between different organizations. If increasing inter-system exchanges is a desired goal, current market incentives and government policies may be insufficient to overcome hospitals’ motivations for pursuing an intra-system-information-exchange-first strategy.

Keywords: medical informatics, hospitals, policy, health information exchange

Introduction

A long-standing and repeated goal of U.S. health policy has been to increase patient information exchange in hopes of improving efficiency and performance.1,2 Information exchange is fundamentally the movement of relevant content, such as data and information, between different actors’ information systems.3 Historically, health care’s isolated information systems, inconsistent methods of storage and representation, and incompatible transport formats prohibited the effective movement of patient information. As a result, health care professionals routinely did not have access to prior patient information from other institutions,4 could not effectively share information with other providers, or even effectively obtain information from other facilities within their own organization. Interoperable health information technology (HIT) attempts to address these barriers through standards for data, transport, security, and terminologies, so that information may be consistently and accurately exchanged.5–7 Evidence indicates that interoperable HIT adoption is growing8,9 and that patient information is increasingly accessible for care delivery, quality improvement, and public health purposes.10

Much of the policy focus on interoperable HIT has been on the exchange of patient information between actors from different organizations, also known as inter-system health information exchange.2,11,12 Nevertheless, interoperable HIT can also be used to support the movement of information within the same organization (ie, intra-system information exchange).12 This is a noteworthy fact because many health care organizations, especially health systems, have not yet mastered their intra-system information exchange challenges. As a case in point, the practice of adopting multiple non-interoperable specialty-focused information systems (eg, pharmacy or radiology systems) can create inefficiencies in information sharing between departments within organizations.13–16 Likewise, patient information could still be difficult to access within a single health system due to the use of different electronic health record systems (EHR) in ambulatory, inpatient, and emergency settings.17,18 In similar fashion, through mergers and acquisitions of hospitals and practices, a single health system may have multiple HIT systems in use at the same time.19 While policy and incentives encourage inter-system exchange, health care organizations also face the choice to invest in interoperable HIT in support of intra-system information exchange needs.

Given that interoperable HIT supports both inter- and intra-system information exchange, and both are needed to effectively support patient care, is there evidence that health care organizations pursue one form of investment more than the other? That is, do health care organizations prioritize the adoption of interoperable HIT to intra-system information exchange? The mere fact that information from intra-system and inter-system exchange can be combined to create comprehensive patient information does not imply that the decision to adopt and pursue each approach functions in a complementary manner. Nevertheless, an organization choosing to make intra-system information exchange the larger focus of its interoperable HIT strategy makes intuitive sense. Furthermore, some evidence indirectly suggests that organizations may prefer to adopt intra-organizational capabilities. For example, qualitative research suggests that hospital leaders see the need to assure that intra-system information sharing capabilities are in place before inter-system information exchange is adopted.19 Similarly, the lack of a clear return on investment20 and the obvious externalities that arise from sharing information with other providers may motivate health care organizations in competitive markets to pursue inter-system information exchange less vigorously.21,22 Additionally, the rise of enterprise EHRs, which integrate all information systems into one single vendor’s product within an organization, suggests organizations’ emphasis may be on addressing intra-system information exchange needs first.23 Moreover, surveillance data suggests that even though interoperable HIT adoption of some kind is becoming nearly ubiquitous, intra-system information sharing is more common with inter-system information exchange activity still lagging behind.8,24–26

This investigation seeks to quantify the relationship between health care organizations’ intra- and inter-system information exchange adoption and to examine sources of heterogeneity. Specifically, we test the hypothesis suggested by qualitative research on hospital strategy and quantitative research on the effects of competition, that health system member hospitals adopt intra-system exchange more than inter-system exchange. Quantifying the relationship between intra- and inter-system information exchange adoption provides insights into organizational strategy and informs the effect of policies intended to increase interoperable HIT usage.

Methods

We examined the relationship between hospitals’ intra- and inter-system information exchange adoption in a sample of health system hospitals included in 5 years of health IT tracking surveys. First, we described the overall trends in intra- and inter-system information exchange adoption among U.S. hospitals that belong to health systems. Second, we described the factors associated with hospitals that adopted more intra-system exchange capability than inter-system exchange capability. Third, we explored the extent of new information exchange capability adoption among a subset of these hospitals that reported neither intra- or inter-system information capabilities at baseline.

Data

Our data come from the 2010 to 2014 American Hospital Association’s (AHA) Annual Surveys and Annual Health Information Technology (HIT) Surveys. The AHA annual survey records key facility characteristics of nearly all hospitals in the United States. The HIT Survey, administered independently, includes key questions about technology adoption. The 2010 to 2014 time period contained the questions necessary to construct our measures.

Sample

The study sample was limited to hospitals that are health system members and responded to both the AHA and HIT surveys. The AHA HIT survey is considered a reliable and valid instrument for assessing hospitals’ HIT usage,27 however, response rates are lower than in the annual overall survey.28 Additionally, respondents do differ from non-respondents, with the HIT survey under-representing small and for-profit hospitals.29,30 The sample was limited to non-federal hospitals in the lower 48 states. Membership in a health system was required in order to measure intra-system exchange as this concept specifically refers to information exchange among different hospitals and organizations within the system. Hospitals had to respond to at least two years of the AHA and HIT survey to be included in the panel. The sample included a total of 8540 hospital-years, representing 2432 unique hospitals.

Measures of Information Exchange Capability

To assess inter-system information exchange adoption, respondents to the AHA HIT survey indicated their hospital’s ability to electronically exchange/share patient demographics, laboratory results, medication histories, radiology reports, and/or clinical care records with hospitals or ambulatory care providers outside their system. Intra-system information exchange capability was assessed by parallel questions about the hospital’s ability to electronically exchange/share patient information with hospitals or ambulatory care providers “inside” their health system. We used these variables to create indicators of any ability to share information via intra- or inter-system exchange, counts of the number of aforementioned information types that could be shared via intra- and inter-system exchange, and a binary indicator of more intra-system exchange capability (ie, a greater count of information types) than inter-system information exchange capability in a given year. Because this article’s focus is on interoperable HIT policy, hospitals had to report having at least a basic EHR29 in order to report information exchange capabilities.

Other Measures

Hospital factors included those structural characteristics known to be associated with technology and HIE adoption such as service type (general acute care, specialty, children’s, long-term, psychiatric and rehabilitation), control (nonprofit, public, or for-profit), bed size, and rural/urban status.21,31,32 In order to explore the role of potential constraints on technology adoption decisions for the hospitals in the sample, we relied on health system cluster types. Cluster types (centralized, moderately centralized, decentralized, independent hospital system, and unclassified) describe the various dominant combinations of service diversification and centralization of decision-making observed across health systems in the U.S.33 Cluster types have been associated with EHR34 and HIE adoption.26 For example, centralizing decision-making to the system level, instead of decentralized at the member hospital level, may delay technology adoption.35 Annual health system size was measured as the total number of hospitals within the system. Increasing system size could indicate a business and clinical need for intra-system exchange. We also measured annual market concentration using the Herfindahl– Hirschman Index (HHI) at the hospital referral region level. The HHI was based on the shares of admissions (as reported in the AHA survey) among general acute care hospitals only and categorized as low concentration, moderate concentration, and high concentration. Previous research suggests competition is a barrier to hospitals’ participation in inter-system information exchange.31 We also included the state in which each hospital was located because there may be differences in policies that may or may not support HIE.36

Analysis

Following previous analyses of AHA survey data,32,37 individual responses were weighted to adjust for non-response bias using inverse-probability weights.38 Frequencies and percents described the study sample at baseline (defined as the first year included in our study panel, 2010). Changes in the annual prevalence of intra- and inter-system information exchange were graphed and assessed using the Cochran-Armitage trend test. Annual prevalence and assessment of changes over time, were stratified into general acute care hospitals and all other hospitals. Specialty, children’s, long-term, psychiatric, and rehabilitation hospitals were largely ineligible for EHR incentive payments under the 2009 HITECH Act,39 and as a result their usage of interoperable HIT and health information exchange adoption is significantly different than that of general acute care hospitals.32 Hospital HIT researchers generally do not analyze general acute care hospitals and these specialty hospitals together (eg,29,30,40).

We fit random-intercept probit regression models to describe the factors associated with a greater extent of intra-system information exchange adoption compared to an equal or greater number of information types exchanged by inter-system exchange. We included random intercepts for each hospital and each health system to account for the repeated and nested nature of the data, a series of state dummies to account for any differences in state policy environments, and year dummies to account for temporal trends in adoption. We presented both unadjusted and fully-adjusted results. Models were limited to general, acute care hospitals due to the small size of specialty hospitals (but results for specialty hospitals are provided as an Appendix). To facilitate interpretation, we presented coefficients as marginal effects at the means.

Lastly, we explored in detail the subsample of hospitals that advanced from having no health information exchange capability at baseline observation (neither intra- or inter-system) to having at least some exchange capability at their next observation period (n = 1147). We report the weighted prevalence of the different types of information shared by intra- and inter-system exchange and the mean number of information types shared. Again, we stratified by general acute care and all other specialty hospital types.

Results

Reflective of the overall structure of the U.S. health care system, hospitals in the sample were mostly general acute care (74%), nonprofit (65%), in urban areas (74%), and commonly had more than 100 beds (Table 1). Decentralized systems, where decision making and services are not centralized at the system-level, were the most frequently represented health system type (44%). At baseline, information exchange capability was not common, with 3 out of 4 hospitals reporting neither intra- nor inter-system information sharing. Approximately 15% of hospitals could share information through both intra- and inter-system exchange. More hospitals could share only intra-system (9%) than could share only inter-system (<1%).

Table 1.

Characteristics of health systems member hospitals by inter-system information exchange status at baseline

Characteristic n (%)a
Service type
 General acute care 2032 (74.1)
 Specialty 31 (1.9)
 Children’s 50 (2.1)
 Psychiatric 145 (8.3)
 Long term 85 (7.8)
 Rehabilitation/ other 89 (5.9)
Control
 Public 286 (10.7)
 Nonprofit 1580 (64.8)
 For-profit 566 (35.2)
Bed size
 <50 598 (30.0)
 50-99 445 (21.1)
 100-299 869 (33.6)
 ≥300 520 (16.4)
Geography
 Urban 1764 (73.6)
 Rural 302 (11.4)
 Suburban 366 (15.0)
System cluster type
 Centralized 477 (16.3)
 Moderately centralized 548 (19.6)
 Decentralized 1023 (43.6)
 Independent hospital system 344 (18.0)
 Unclassified 40 (2.6)
System size (mean, sd) 18.9 (28.8)
Number of inpatient EHR vendors in system (mean, sd) 1.2 (1.7)
Market concentration
 Low 1485 (61.4)
 Moderate 531 (21.6)
 High 416 (17.1)
Health information exchange
 Intra-system only 229 (9.3)
 Inter-system only 11 (0.5)
 Both 444 (15.6)
 Neither 1748 (75.6)
Open in a new tab
a

Frequencies unweighted, percentages weighted.

Trends in Health Information Exchange Adoption

Over time, exchange capability has increased significantly. The percent of general acute care hospitals annually reporting inter-system information exchange capability increased from 12.5% to 71.4% between 2010 and 2014 (p < .0001), intra-system exchange capability increased from 20.0% to 83.0% (p < .0001) (Figure 1). Both inter-system (p < .0001) and intra-system (p < .0001) capabilities among specialty hospitals also increased significantly over time. However, each year the intra-system and inter-system information exchange adoption rates were much lower than among general acute care hospitals (p < .0001 each year). For all hospital types, inter-system information exchange lagged behind intra-system exchange capability each year.

Figure 1.

Figure 1.

Open in a new tab

Trends in general acute care and specialty hospitals’ intra-system information sharing and inter-system information exchange capability, 2010 to 2014.

Sample sizes for each year: 2010 general = 1376, specialty = 242; 2011 general = 1338, specialty = 227; 2012 general = 1531, specialty = 270; 2013 general = 1544, specialty = 284; and 2014 general = 1474; specialty = 254.

Factors Associated with Hospitals Adopting More Intra-system Information Exchange Capability

Over the complete study sample of hospitals that had adopted information exchange: 50.7% had adopted more intra-system exchange; 46.7% had adopted both intra- and inter-system exchange to the same extent; and 2.6% had adopted more inter-system exchange. Table 2 (column 1) displays the probit coefficients describing the association between hospital characteristics and more intra-system information exchange without controlling for repeated measures, health system, and hospital state. The marginal effects at the means for each of these unadjusted coefficients is presented in column 2. Without controlling for other factors, compared to small hospitals, hospitals with more than 300 beds were 6.6 percentage points less likely to report more intra-system exchange. That is, larger hospitals had as many or more types of inter-system information exchanged. In addition, reflective of the overall increasing trends in health information exchange adoption, in later survey years, hospitals were less likely to have more intra-system exchange than inter-system exchange.

Table 2.

Association between hospital characteristics and greater adoption of intra-system information exchange than inter-system information exchange

[1] Unadjusted [2] Unadjusted [3] Adjusted [4] Adjusted
β (95% CI) p Marginal effect p β (95% CI) p Marginal effect p
Control
 Nonprofit Reference Reference Reference Reference
 Public −0.035 (−0.237, 0.167) .735 −0.014 .735 0.024 (−0.222, 0.269) .851 0.009 .851
 For-profit 0.023 (−0.212, 0.258) .847 0.009 .847 −0.005 (−0.292, 0.282) .972 −0.002 .972
Bed size
 <50 Reference Reference Reference Reference
 50-99 −0.091 (−0.254, 0.071) .272 −0.036 .272 −0.089 (−0.67, 0.089) .327 −0.035 .326
 100-299 −0.145 (−0.293, 0.004) .056 −0.057 .055 −0.175 (−0.363, 0.089) .069 −0.069 .066
 ≥300 −0.66 (−0.324, -0.007) .041 −0.066 .040 −0.233 (−0.432, -0.035) .021 −0.092 .020
Geography
 Suburban Reference Reference Reference Reference
 Urban −0.147 (−0.304, 0.010) .067 −0.059 .067 −0.142 (−0.334, 0.050) .147 −0.056 .148
 Rural −0.051 (−0.229, 0.126) .571 −0.020 .571 −0.49 (−0.246, 0.148) .627 −0.019 .626
System cluster type
 Centralized Reference Reference Reference Reference
 Moderately centralized −0.226 (−0.528, 0.076) .143 −0.090 .141 −0.270 (−0.586, 0.045) .093 −0.105 .092
 Decentralized −0.147 (−0.424, 0.130) .298 −0.058 .297 −0.314 (−0.624, -0.003) .048 −0.122 .046
 Independent hospital system −0.367 (−0.745, 0.011) .057 −0.146 .054 −0.407 (−0.802, -0.012) .043 −0.160 .043
 Unclassified −0.377 (−1.092, 0.338) .302 −0.149 .296 −0.480 (−1.362, 0.402) .386 −0.189 .288
System size −0.012 (−0.016, -0.008) <.001 −0.005 <.001 –a – –a –
Number of inpatient EHR vendors in system −0.052 (−0.148, 0.044) .285 −0.021 .285 0.133 (0.041, 0.225) .005 0.053 .005
Market concentration
 Low Reference Reference Reference Reference
 Moderate −0.040 (−0.185, 0.105) .589 −0.016 .589 −0.093 (−0.274, 0.089) .318 −0.037 .319
 High −0.077 (−0.233, 0.078) .330 −0.030 .330 −0.038 (−0.217, 0.141) .675 −0.015 .676
Year
 2010 Reference Reference Reference Reference
 2011 0.426 (0.181, 0.671) .001 0.117 .001 0.448 (0.203, 0.693) <.001 0.108 .001
 2012 −0.494 (−0.756, -0.231) <.001 −0.180 <.001 −0.558 (−0.856, -0.260) <.001 −0.194 <.001
 2013 −0.584 (−0.846, -0.323) <.001 −0.216 <.001 −0.679 (−0.957, -0.401) <.001 −0.242 <.001
 2014 −1.243 (−1.546, -0.939) <.001 −0.466 <.001 −1.338 (−1.654, -1.022) <.001 −0.494 <.001
Open in a new tab
a

system size omitted from the fully adjusted model because cluster type includes single hospital systems.

In the fully adjusted model (Table 2 column 3), larger hospital size continued to be negatively associated with more intra-system exchange adoption. After controlling for the other factors in the model, compared to centralized system members, hospitals that belonged to decentralized systems were associated with more inter-system exchange (β=−0.314; 95%CI= −0.624, −0.003). That is, compared to centralized systems, decentralized systems had a 12.2 percentage point lower probability of reporting more intra-system exchange than inter-system exchange (Table 3 column 4). Independent hospital systems were also associated with more inter-system exchange (β=−0.407; 95%CI= −0.802, −0.012). Controlling for other factors, as the number of different inpatient EHR vendors in use in health system increased, hospitals tended to share more types of information by intra-system exchange (5.3 percentage points higher) than they could by inter-system exchange (B = 0.133; 95%CI = 0.041, 0.225).

Table 3.

Hospitals’ initial adoptiona of intra-system and inter-system exchange capabilities by type of information and total counts

General acute care hospitals (n = 1076)
 Specialty hospitals (n = 71)
Exchanged intra-system Exchanged inter-system p Exchanged intra-system Exchanged inter-system p
Type of information % % % %
 Patient demographics 93.8 58.9 <.0001 91.2 47.4 <.0001
 Laboratory results 94.6 64.0 <.0001 90.0 51.1 <.0001
 Medication histories 90.2 44.0 <.0001 89.6 35.0 <.0001
 Radiology reports 93.6 60.4 <.0001 90.0 47.0 <.0001
 Clinical care records 86.3 43.4 <.0001 82.3 46.5 <.0001
Number of information types shared (mean, sd) 4.58 (1.47) 2.71 (2.72) <.0001 4.43(2.47) 2.27 (3.64) <.0001
Open in a new tab
a

Among hospitals who had no health information exchange capability at first observation to some exchange at the next observation period.

Comparison of Initial Health Information Exchange Activity

Among the subset of hospitals that advanced from having no health information exchange capability at baseline observation to having new exchange capability for at least one type of information at the next study observation period, adoption of intra-system exchange capabilities was much more common (Table 3). In addition, intra-system exchange was associated with the capability to exchange more types of information. This was true for both general acute care and specialty hospitals. After adopting information exchange, general acute care hospitals on average were able to share more than four of the five types of information by intra-system exchange (mean = 4.6). In comparison, hospitals were able to exchange fewer types of information by inter-system exchange (mean = 2.7; p < .0001). Medication histories (43%) and clinical care records (44%) were the least common information types shared by inter-system exchange. After adopting information exchange, specialty hospitals on average adopted more intra-system information sharing (mean = 4.4) than inter-system information sharing (mean = 2.3; p < .0001).

Discussion

Consistent with the U.S. goals for more widely accessible patient information, such as through the HITECH Act and the Office of National Coordinator (ONC)’s Strategic Plan, hospitals’ adoption of intra-system and inter-system information exchange for both general and specialty hospitals have increased substantially. However, less consistent with the national ideals of widely accessible patient information was hospitals’ prioritization of information exchange capabilities. We hypothesized that health system member hospitals would adopt intra-system exchange more than inter-system exchange and the findings supported this assertion. This study indicates hospitals have adopted intra-system forms more frequently during 2010 to 2014 than inter-system information exchange and hospitals adopting exchange capabilities could exchange more types of data through intra-system exchange than inter-system exchange. The overall trends in information exchange adoption suggest hospitals do not view intra-system and inter-system information exchange as complements, but instead first address information exchange within the organization.

Hospitals’ adoption of an intra-system information exchange first approach does have beneficial features. For one, our findings indicate when hospitals do adopt technology to address information exchange within the organization, their resulting capability to exchange many different types of information is generally comprehensive. For hospitals that had newly adopted intra-system exchange, the vast majority of hospitals reported the capability to share demographics, laboratory results, medication histories, radiology reports, and clinical care records within the organization. The ability to share a large number of different information types has potential benefits for organizational efficiencies41 and can support clinical decision support systems.42 In addition, intra-system information exchange fulfills pressing operational and business needs posed by a multi-vendor EHR environment.19 Similarly, our finding that as the number of different inpatient EHR vendors in use in the system increased, so did the probability that hospitals possessed more intra-system than inter-organization exchange capability is consistent with hospitals’ focus on addressing operational and business needs.

While lagging behind intra-system adoption, the overall increase in inter-system exchange capability over the study period is a positive from a national policy perspective. Current statements like the Medicare Access and CHIP Reauthorization Act of 2015,43 information blocking prohibitions,44 and the 21st Century Cures Act of 201645 clearly favor the adoption of inter-system information exchange. Moreover, inter-system information exchange has the potential to increase quality and create cost savings for the nation as a whole46 and rectifies the fragmented information created by patients seeking care from different health systems, care sought during travel, or displacements created by disasters.47,48 The overall increasing trend towards a higher prevalence of inter-system exchange capabilities, however, may mask important differences. As noted above, hospitals that newly adopted intra-system exchange had the capability to share nearly all types of information. In contrast, when hospitals newly adopted inter-system exchange, they had the capability to exchange far fewer types of information. This was particularly the case for medication histories, which are critical for patient safety and clinical care records which support population health management.49

Hospitals’ prioritizing internal information needs may be explained by several factors. Principally, organizations generally endeavor (or need) to prioritize internal efficiencies prior to engaging in information exchange with any trading partners.50,51 Moreover, adoption and implementation is a multi-step, resource intensive, and complicated process.52 Given difficult choices about the allocation of finite resources,19 these findings suggest that health systems prefer to address intra-system needs over working to foster information sharing capability with other organizations. In addition, trends within health care towards accountable care organizations (ACOs) and population health may be proving to be less powerful drivers than hoped. Researchers and advocates have noted that such payment reform initiatives may spur hospitals and health systems to embrace inter-system exchange as a means of better managing risk by facilitating access to information on patients who seek care from other providers.53 Nevertheless, vertical integration, narrower provider networks, bundled payments, and ACOs also require intra-system information sharing to minimize patient encounters and use of services with external providers.19 These broader trends in health care may indeed lead to more widespread inter-system information sharing, but they do not necessarily eliminate the reasons to prioritize intra-system information sharing. However, it is noteworthy that hospitals with greater autonomy around technology decisions, ie, decentralized system members and independent hospital systems, were more likely to adopt more inter-system information exchange. While these data cannot go beyond showing the association between greater autonomy and inter-system information exchange, it is possible that this reflects differences between system-level and hospital-level strategies and needs (eg, the system’s need for comprehensive patient information from member hospitals for population level analytics and a hospital’s need to exchange information with a neighboring hospital to identify frequent emergency department utilizers). Similarly, other very effective use cases for information exchange, like public health reporting or event notification services, either do not require partnerships with competing organizations or can be accomplished using less-than-interoperable, basic admission-discharge-transfer systems, which again does not necessarily increase inter-system information exchange adoption. U.S. policy goals of “improving the health and well-being of individuals and communities” and having health information “that is accessible when and where it matters most”54 may not be realized without directly addressing the forces encouraging, and trend towards more, intra-system exchange.

This study has several notable limitations. First, our measure of intra-system information exchange is limited to information exchange across hospitals or practices within a single system. Because of the nature of the AHA survey questions, we could not assess intra-system information sharing within organization, such as sharing between departments, nor did we have sufficient information to identify health systems with a single instance of an EHR (compared to those with multiple instances from the same vendor across facilities). Also, we are limited to the categories of information types asked on the survey, and we do not know to what extent these categories were viewed as mutually exclusive by respondents. For example, clinical care records might include medications or other results. In addition, and critically, this analysis measured only the adoption of health information exchange capabilities and on the actual sharing of information. However, even when hospitals have inter-system information exchange capability, the extent of information sharing tends to be low.8 Therefore, the actual amount of inter-system information exchange is likely less than suggested by this study. Moreover, measures of adoption or capability do not address the quality of the information, the number of exchange partners, or the use of the exchanged information. This last feature, the actual integration of exchanged information into the organization’s EHR, is probably the most critical component of interoperability. However, the integration of information from other organizations is difficult, often resisted,55 and not common in health care.56 In contrast, integration may be more frequently realized in the case of intra-system exchanged information, which again pits the two approaches against each other for organizational prioritization.19 Making inter-system exchanged information easier to access and a seamless part of the whole patient record remains a key step to increasing demand and use of inter-system information exchange.57 Lastly, new information sharing agreements and governance within the EHR vendor community came after this study period.58 However, even this may have a muted effect on the implementation of widespread inter-system exchange, because hospitals and health systems have traditionally promoted EHR vendor-mediated information exchange as an alternative to broad inter-system exchange.59

Inter-system information exchange is a goal of U.S. health care policy. However, hospitals are actually prioritizing intra-system information exchange. Current health care policies and trends do encourage inter-system exchange, but they may be insufficient to overcome hospitals’ motivations for pursuing an intra-system-information-exchange-first strategy.

Funding

This work was supported by the Agency for Healthcare Quality & Research grant number 1 R21HS024717-01A1.

Contributors

JV conceived and designed the study, analyzed the data, and drafted the manuscript. KS contributed to the analysis and drafting of the manuscript. Both authors had final approval of the version to be published.

Competing interests

None.

Acknowledgments

The authors thank Brittany Brown-Podgorski and Katy Ellis Hilts for assistance in data collection.

References

  • 1. The White House. Transforming health care: the president’s health information technology plan. Promoting Innovation and Competitiveness. President Bush’s Technology Agenda. 2004. http://georgewbush-whitehouse.archives.gov/infocus/technology/economic_policy200404/chap3.html.
  • 2. Office of the National Coordinator for Health Information Technology. The ONC-Coordinated Federal Health IT Strategic Plan: 2008–2012. Washington, DC: Department of Health & Human Services; 2008. [Google Scholar]
  • 3. Iacovou CL, Benbasat I, Dexter AS.. Electronic data interchange and small organizations: adoption and impact of technology. MIS Q 1995; 194: 465–85. [Google Scholar]
  • 4. Stewart BA, Fernandes S, Rodriguez-Huertas E, Landzberg M.. A preliminary look at duplicate testing associated with lack of electronic health record interoperability for transferred patients. J Am Med Inform Assoc 2010; 173: 341–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. DeSalvo KB, Dinkler AN, Stevens L.. The US office of the national coordinator for health information technology: progress and promise for the future at the 10-year mark. Ann Emerg Med 2015; 665: 507–10. [DOI] [PubMed] [Google Scholar]
  • 6. Heubusch K. Interoperability: what it means, why it matters. J AHIMA 2006; 771: 26–30. [PubMed] [Google Scholar]
  • 7. Brailer DJ. Interoperability: the key to the future health care system. Heal Aff 2005; 24: w19–21. [DOI] [PubMed] [Google Scholar]
  • 8. Holmgren AJ, Patel V, Adler-Milstein J.. Progress in interoperability: measuring US hospitals’ engagement in sharing patient data. Health Aff 2017; 3610: 1820–7. [DOI] [PubMed] [Google Scholar]
  • 9. Furukawa MF, King J, Patel V, Hsiao C-J, Adler-Milstein J, Jha AK.. Despite substantial progress in EHR adoption, health information exchange and patient engagement remain low in office settings. Health Aff 2014; 339: 1672–9. [DOI] [PubMed] [Google Scholar]
  • 10. Rudin RS, Motala A, Goldzweig CL, Shekelle PG.. Usage and effect of health information exchange: a systematic review. Ann Intern Med 2014; 16111: 803. 11. [DOI] [PubMed] [Google Scholar]
  • 11. Department of Health & Human Services. Medicare and Medicaid Programs; Electronic Health Record Incentive Program. Final rule. Fed Regist 2010; 75144: 44314–588. [PubMed] [Google Scholar]
  • 12. Chen C, Garrido T, Chock D, Okawa G, Liang L.. The Kaiser Permanente electronic health record: transforming and streamlining modalities of care. Health Aff 2009; 282: 323–33. [DOI] [PubMed] [Google Scholar]
  • 13. Bahensky JA, Ward MM, Nyarko K, Li P.. HIT implementation in critical access hospitals: extent of implementation and business strategies supporting IT use. J Med Syst 2011; 354: 599–607. [DOI] [PubMed] [Google Scholar]
  • 14. Hermann SA. Best-of-breed versus integrated systems. Am J Health Syst Pharm 2010; 6717: 1406–10. [DOI] [PubMed] [Google Scholar]
  • 15. McMurray J, Hicks E, Johnson H, Elliott J, Byrne K, Stolee P. ‘ Trying to find information is like hating yourself every day’: the collision of electronic information systems in transition with patients in transition. Health Informatics J 2013; 193: 218–32. [DOI] [PubMed] [Google Scholar]
  • 16. Feufel MA, Robinson FE, Shalin VL.. The impact of medical record technologies on collaboration in emergency medicine. Int J Med Inform 2011; 808: e85–95. [DOI] [PubMed] [Google Scholar]
  • 17. Payne T, Fellner J, Dugowson C, Liebovitz D, Fletcher G.. Use of more than one electronic medical record system within a single health care organization. Appl Clin Inform 2012; 0304: 462–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Sherer SA, Meyerhoefer CD, Sheinberg M, Levick D.. Integrating commercial ambulatory electronic health records with hospital systems: an evolutionary process. Int J Med Inform 2015; 849: 683–93. [DOI] [PubMed] [Google Scholar]
  • 19. Vest JR, Kash B.. Differing strategies to meet information sharing needs: the publicly supported community health information exchange versus health systems’ enterprise health information exchanges. Milbank Q 2016; 941: 77–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Fontaine P, Ross SE, Zink T, Schilling LM.. Systematic review of health information exchange in primary care practices. J Am Board Fam Med 2010; 235: 655–70. [DOI] [PubMed] [Google Scholar]
  • 21. Miller AR, Tucker C.. Health information exchange, system size and information silos. J Health Econ 2014; 33: 28–42. [DOI] [PubMed] [Google Scholar]
  • 22. Adler-Milstein J, Jha AK.. Health information exchange among U.S. hospitals: who’s in, who’s out, and why? Healthcare 2014; 21: 26–32. [DOI] [PubMed] [Google Scholar]
  • 23. Conn J. Facilities turning to single vendor for interoperability. Modern Healthcare. 2013. http://www.modernhealthcare.com/article/20130918/NEWS/309189949. Accessed 23 February, 2017.
  • 24. Furukawa MF, Patel V, Charles D, Swain M, Mostashari F.. Hospital electronic health information exchange grew substantially in 2008–12. Health Aff 2013; 328: 1346–54. [DOI] [PubMed] [Google Scholar]
  • 25. Adler-Milstein J, DesRoches CM, Kralovec P, et al. Electronic health record adoption in US hospitals: progress continues, but challenges persist. Health Aff 2015; 3412: 2174–80. [DOI] [PubMed] [Google Scholar]
  • 26. Opoku-Agyeman W, Menachemi N.. Are there differences in health information exchange by health system type? Health Care Manage Rev 2016; 414: 325–33. [DOI] [PubMed] [Google Scholar]
  • 27. Everson J, Lee S-YD, Friedman CP. Reliability and validity of the American Hospital Association’s national longitudinal survey of health information technology adoption. J Am Med Inform Assoc 2014; 21 (e2): e257–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. American Hospital Association. Data collection methods. 2017. http://www.ahadata.com/data-collection-methods/. Accessed November 14, 2017.
  • 29. Adler-Milstein J, DesRoches CM, Furukawa MF, et al. More than half of US hospitals have at least a basic EHR, but stage 2 criteria remain challenging for most. Health Aff 2014; 339: 1664–71. [DOI] [PubMed] [Google Scholar]
  • 30. Jha AK, DesRoches CM, Campbell EG, et al. Use of Electronic Health Records in U.S. Hospitals. N Engl J Med 2009; 36016: 1628–38. [DOI] [PubMed] [Google Scholar]
  • 31. Vest JR. More than just a question of technology: factors related to hospitals’ adoption and implementation of health information exchange. Int J Med Inform 2010; 7912: 797–806. [DOI] [PubMed] [Google Scholar]
  • 32. Walker D, Mora A, Demosthenidy MM, Menachemi N, Diana ML.. Meaningful use of EHRs among hospitals ineligible for incentives lags behind that of other hospitals, 2009-13. Health Aff (Millwood) 2016; 353: 495–501. [DOI] [PubMed] [Google Scholar]
  • 33. Bazzoli GJ, Shortell SM, Dubbs N, Chan C, Kralovec P.. A taxonomy of health networks and systems: bringing order out of chaos. Health Serv Res 1999; 336: 1683–717. [PMC free article] [PubMed] [Google Scholar]
  • 34. Kazley AS, Ozcan YA.. Organizational and environmental determinants of hospital EMR adoption: a national study. J Med Syst 2007; 315: 375–84. [DOI] [PubMed] [Google Scholar]
  • 35. Shen JJ, Moseley CB.. Organisational factors associated with adoption of comprehensive and basic electronic-record systems in US hospitals. Int J Public Policy 2012; 8 (1/2/3): 92. [Google Scholar]
  • 36. Schmit CD, Wetter SA, Kash BA.. Falling short: how state laws can address health information exchange barriers and enablers. Inform Assoc 2018; 256: 635–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. DesRoches CM, Worzala C, Joshi MS, Kralovec PD, Jha AK.. Small, nonteaching, and rural hospitals continue to be slow in adopting electronic health record systems. Health Aff (Millwood) 2012; 315: 1092–9. [DOI] [PubMed] [Google Scholar]
  • 38. Wooldridge JM. Inverse probability weighted estimation for general missing data problems. J Econom 2007; 1412: 1281–301. [Google Scholar]
  • 39.United States Congress. American Recovery and Reinvestment Act of 2009/Division A/Title XIII – Health Information Technology. 111th Congress, 1st Session, Public Law 111-115. Washington, DC: GPO; 2009.
  • 40. Jha AK, DesRoches CM, Kralovec PD, Joshi MS.. A progress report on electronic health records in U.S. hospitals. Health Aff 2010; 2910: 1951–7. [DOI] [PubMed] [Google Scholar]
  • 41. McAfee A. The impact of enterprise information technology adoption on operational performance: an empirical investigation. Prod Oper Manag 2009; 111: 33–53. [Google Scholar]
  • 42. Bates DW, Kuperman GJ, Wang S, et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J Am Med Inform Assoc 2003; 106: 523–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.H.R.2 - Medicare Access and CHIP Reauthorization Act of 2015. Public Law 114-10. 2015–2016 ed 2015. https://www.congress.gov/bill/114th-congress/house-bill/2/text. Accessed February 28, 2018.
  • 44. Office of the National Coordinator for Health Information Technology. Report to Congress: Report on Health Information Blocking. Washington, DC: Department of Health and Human Services; 2015. [Google Scholar]
  • 45. Mello MM, Adler-Milstein J, Ding KL, Savage L.. Legal barriers to the growth of health information exchange-boulders or pebbles? Milbank Q 2018; 961: 110–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Walker J, Pan E, Johnston D, Adler-Milstein J, Bates DW, Middleton B.. The value of health care information exchange and interoperability. Health Aff 2005; 24: w10–8. [DOI] [PubMed] [Google Scholar]
  • 47. Slovis BH, Lowry T, Delman BN, et al. Patient crossover and potentially avoidable repeat computed tomography exams across a health information exchange. J Am Med Inform Assoc 2017; 241: 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Shapiro JS, Mostashari F, Hripcsak G, Soulakis N, Kuperman G.. Using health information exchange to improve public health. Am J Public Health 2011; 1014: 616–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Lobach DF, Kawamoto K, Anstrom KJ, et al. Proactive population health management in the context of a regional health information exchange using standards-based decision support. AMIA Annu Symp Proc 2007; 2007: 473–7. [PMC free article] [PubMed] [Google Scholar]
  • 50. Payton FC. Lessons learned from three interorganizational health care information systems. Inf Manag 2000; 376: 311–21. [Google Scholar]
  • 51. Murphy K. HIE best practices: preparing for health information exchange. EHR Intelligence. 2012. https://ehrintelligence.com/news/hie-best-practices-preparing-for-health-information-exchange. Accessed November 7, 2017.
  • 52. Esmaeilzadeh P, Sambasivan M.. Health information exchange (HIE): a literature review, assimilation pattern and a proposed classification for a new policy approach. J Biomed Inform 2016; 64: 74–86. [DOI] [PubMed] [Google Scholar]
  • 53. Mahajan A. Health information exchange—obvious choice or pipe dream? JAMA Intern Med 2016; 1764: 429–30. [DOI] [PubMed] [Google Scholar]
  • 54. The Office of the National Coordinator for Health Information Technology. Connecting Health and Care for the Nation: a Shared Nationwide Interoperablity Roadmap. Washington, DC: Department of Health & Human Services; 2015. [Google Scholar]
  • 55. Premkumar G, Ramamurthy K.. The role of interorganizational and organizational factors on the decision mode for adoption of interorganizational systems. Decis Sci 1995; 263: 303. [Google Scholar]
  • 56. Patel V, Henry J, Pylypchuk Y, Searcy T.. Interoperability among U.S. Non-federal Acute Care Hospitals in 2015. ONC Data Brief No.36 Washington, DC: Office of the National Coordinator for Health Information Technology; 2016.
  • 57. Thorn SA, Carter MA, Bailey JE.. Emergency physicians’ perspectives on their use of health information exchange. Ann Emerg Med 2014; 633: 329–37. [DOI] [PubMed] [Google Scholar]
  • 58. Commonwell Health Alliance. Carequality and Commonwell Health Alliance agree on connectivity and collaboration to advance interoperability—Commonwell Health Alliance. 2016. http://www.commonwellalliance.org/news/carequality-commonwell-health-alliance-collaboration/. Accessed November 7, 2017.
  • 59. Vest JR, Campion TR Jr, Kaushal R; for the HITEC Investigators. Challenges, alternatives, and paths to sustainability for health information exchange efforts. J Med Syst 2013; 376: 9987. [DOI] [PubMed] [Google Scholar]

Articles from Journal of the American Medical Informatics Association : JAMIA are provided here courtesy of Oxford University Press

RESOURCES