Background
Exchange of health information through interoperable systems is an essential goal as providers transition from hard to digital copies of medical records [
1]. Interoperable systems ensure that electronic health information can be used and exchanged without any special effort from the sender or receiver through the use of a common language [
2]. Without interoperable systems, the full potential benefits of adopting Electronic Health Records cannot be achieved [
3].
The Health Information Technology for Economic and Clinical Health (HITECH) Act enacted in 2009 [
4], and the associated State Health Information Exchange Cooperative Program [
5], provide monetary incentives to eligible providers and hospitals to support the adoption of EHRs and health information exchange. To receive these incentives, hospitals and physicians must meet usability criteria also known as meaningful use (MU) objectives (core and menu) that ensure EHRs are used to support health policy priorities [
6]. EHR products that are purchased through the Meaningful Use incentives are certified by the Department of Health and Human Services [
7]. While certification criteria changed the supply side of the EHR market, the stated incentives allowed for a greater demand for EHR certified products [
8].
A core objective of Meaningful Use’s second stage is the exchange of clinical summaries when patients transition between providers. Until the year 2014, certification requirements stated that both the Continuity of Care Record (CCR) and the Continuity of Care Document (CCD) standardized formats could be used for said exchange [
9]. Current regulation, published in 2015, requires that vendors demonstrate that they are able to use the second version of the Consolidated Clinical Document Architecture (C-CDA 2.1) markup standard. Aside from the format standard, MU incentives foster an open approach to health information exchange, allowing for direct exchange among EHR vendors as well as enterprise and community solutions [
10].
Despite these efforts to promote interoperability and meaningful use, information exchange has remained below expectations set by HITECH [
1,
11,
12]. Qualitative and quantitative studies identify several operational and economic barriers to health information exchange. Qualitative studies have found that operational barriers include the use of information as a competitive advantage, lack of cost-efficiency, limited incentives for staff and diminished trust of other providers [
13‐
15]. Quantitative studies have shown that certain hospital characteristics are related to the probability that hospitals exchange health information with other hospitals. For example, Adler-Milstein and Jha [
16] found that hospitals with a larger market share within a region were more likely to participate in information exchange but that hospitals in competitive markets were less likely to share information. Furthermore, Miller and Tucker [
17] found that hospitals that are part of larger health systems are less likely to exchange information with hospitals outside their system. In sum, several studies find that information exchange is a function of a hospital’s strategic reasons for sharing [
16‐
20].
Although most research has looked at the characteristics of hospitals associated with information exchange, recently more focus has been directed toward vendors and how they may also use information exchange strategically [
21]. While EHR products must be capable of transforming from proprietary architectures to the semantics and structure used in CCR or CCD formats at least once during the certification process, there are still reports of both cost and technical barriers for the process of exchanging clinical care summaries [
22,
23]. Hence, EHR vendors could knowingly and unjustifiable interfere with health information exchange by engaging in what is known as information blocking [
21].
We hypothesize that vendors can leverage proprietary software to make it easier for hospitals to share clinical care summaries with other hospitals that have the same software while making it challenging to share with hospitals that use a different EHR vendor. This imposes costs on hospitals that need to share information and creates an incentive for them to adopt the dominant vendor. Specifically, research has shown that hospitals that use dominant vendors likely face fewer technical obstacles and engage in a higher number of HIE activities [
24].
Empirical work in other domains supports this. For example, Shapiro and Varian find that there are network externality benefits to being connected to a larger communication network [
25]. The value of connecting to a network depends on the number of others that are already connected, which means that from the perspective of a user, being connected to a bigger network is better [
25]. When an EHR vendor has a large closed network, in this case a large number of hospitals that use its product, a user will have an easier time sharing information. This creates a competitive advantage for the EHR vendor that provides the closed network.
Even if policy incentivizes the exchange of clinical care summaries, there is significant variation in the use of HIE across EHR vendors. Some vendors have been at the forefront by facilitating exchange through private proprietary networks. The most prominent of these networks is Care Everywhere, a system incorporated into EPIC EHR products since 2005 [
26]. Although Care Everywhere is meant to be able to connect to EHR systems from other vendors, it is most successfully used to connect with other EPIC users [
27]. Additionally, connection even within the Care Everywhere network may require additional customization [
28].
The present study aims to identify the effect of vendor choice and vendor network size on whether a hospital reports participating in the exchange of clinical care summaries. As a measure of the size of an EHR vendor’s network we will use EHR vendor market share and market concentration. We hypothesize that the probability of a hospital engaging in the exchange of clinical care summaries with another hospital outside its health systems is associated with the market share of the EHR vendor in the immediate region where exchanges are more likely to occur.
To determine this association, we use a logistic regression model at the hospital level using the response from each provider about whether they exchange clinical care summaries outside of their system as a dependent variable. EHR vendor market share is measured by the percentage of hospitals that have the same vendor in the hospital referral region (HRR) where a hospital operates. These regions, or markets for tertiary medical care, consolidate zip codes where the majority of patients are referred to a specific hospital for cardiovascular surgery and neurosurgery [
29]. We expect that in HRRs where intra-vendor sharing occurs the probability of a hospital engaging in information exchange increases as the market share of this hospital’s EHR vendor increases. This effect is due to the increase of the number of opportunities to engage in intra-vendor exchange. We also test for differences that might be unique to large EHR vendors that have established proprietary information networks, such as EPIC, by testing the different interactions in logistic regressions for the three largest EHR vendors.
A second analysis is done at the state level. The aim of this second model is to further analyze the propensity to share in the context of state level policies that incentivize health information exchange. Our hypothesis is that a higher EHR vendor market concentration, measured by the Herfindahl-Hirschman Index (HHI), is associated with a positive change in the percentage of hospitals that participate in information exchange within each state. We also expect to find differences in the propensity of this exchange depending on the strategies adopted by each state to incentivize HIE.
Discussion
Through this analysis we have found empirical evidence that, among other factors, vendor market share and concentration are related to the likelihood of a hospital sharing clinical care summaries and the percentage of hospitals within a state that exchange such information. These factors remain important even when we take into account policies that incentivize information exchange such as the requirement for the use of standardized formats (CCR, CDA and CCD) and State level programs. While the capability to use a common format to send clinical care summaries is significant in increasing the likelihood of participating in the exchange of these documents, this ability is not enough to guarantee exchange outside a hospital’s network. In fact, 72% of hospitals that do not share clinical care summaries with other vendors are capable of using these common formats. Furthermore, almost 30% of hospitals that can use CCD and CDA continue to claim that they are not capable of exchanging clinical care summaries with hospitals using a different certified EHR vendor. This supports the notion that even when EHR systems are certified to comply with this MU requirement, exchange with outside vendors remains a challenge.
In this context, EHR vendor market share and concentration become relevant topics of analysis. Of the hospitals that exchange clinical care summaries with hospitals outside their system 23% assert that they cannot exchange with hospitals using a different EHR vendor (despite the fact that only 10% of hospital EHR systems don’t support CCD or CDA exchange standards), suggesting that exchange in this subgroup is happening directly between hospitals using the same EHR vendor. Although we cannot conclude from the available data if exchange for the rest of the sample is taking place through proprietary or standards-based methods, we can presume that there are benefits to exchanging using proprietary methods that are strengthened when the vendors are more concentrated. These benefits may include reduced technical difficulty and ease of access to specific interfaces, which might remain influential even if a hospital is technically able to exchange using standard formats.
When we control for each of the three largest EHR Vendors in the market we find relevant differences in the propensity for information exchange. We analyze the interactions with these EHR vendors in our sample and find that the positive relationship between HIE and market share is very strong for hospitals that use Epic. Hospitals using EHR vendor Epic are much more likely to exchange clinical care summaries than the rest of the hospitals in our sample. Conversely, hospitals that use Meditech and Cerner are less likely to exchange this type of information. By promoting proprietary sharing, larger players strengthen the network externality benefit of choosing an EHR from a large player. From these results we can infer that the availability of Epic’s Care Everywhere has important implications for hospitals looking to participate in information exchange. In fact, Epic becomes an interesting case study for the effects of having a proprietary network for health information exchange. Our analysis shows that Epic users might overcome some of the barriers for information exchange when other hospitals in the same region use Epic. However, when there are no users nearby that use this same EHR vendor the net benefits for exchange are diminished. This suggests that when removing the incentive of a geographically close Epic user for exchange, additional customization could act as a deterrent for developing further HIE capabilities [
27,
28].
Due to the competitive nature of the EHR market, a larger player such as Epic could leverage its network size by facilitating intra-vendor sharing in an effort to enlist new users interested in sharing within its existing network. Smaller practices and hospitals interested in exchanging clinical care summaries with larger hospitals that use said EHR vendor would need to join the network. The decision to choose a specific EHR product involves a lock-in factor because of the sizeable costs of implementation. Not only does this make it unlikely that smaller hospitals could then change to a different vendor, it may involve additional unforeseen costs that could discourage them from implementing usable exchange capabilities [
10,
36].
We have similar results at the state level. We find that higher Vendor HHI is positively correlated with the percentage of hospitals within the state that share information, even when controlling for different policies that incentivize or hinder information exchange. The different strategies applied through the State Health Information Exchange Cooperative Agreement Program (State HIE) do not show a significant effect on the percentage of hospitals that exchange clinical care summaries within a state. Hence, in states with highly concentrated markets measured by the Herfindahl-Hirschman Index (where one or two EHR vendors are used by the majority of the hospitals) there are more hospitals engaging in information exchange.
Part of the objective of the State HIE program was to fill HIE service gaps and build capacity for every eligible provider [
5]. The fulfillment of this goal could be an important contribution toward overcoming some of the limitations of vendor facilitated exchange and the possible failures of closed proprietary networks. Unfortunately, our current research shows that none of the state level strategies seem to be successful in reducing this effect. In states where there are less concentrated markets, none of the different implementations were significant in incentivizing exchange. This might be a symptom of misaligned incentives, as there have been reports of current regulation undermining the role of community health information exchanges supported by State HIE by allowing EHR vendor mediated exchange that cuts out public exchanges [
10].
As more hospitals transition to the second stage of meaningful use, data from recent years shows that similar challenges for HIE persist. While the percentage of hospitals that report that they have the capability to send clinical care summaries has increased, the percentage of hospitals that send them during transitions of care remains low. Data from Meaningful Use attestations between 2014 and 2016 shows that a median hospital sends clinical care summaries electronically for 33% of transitions, while the use of Epic as an EHR provider positively increases this probability [
37]. Furthermore, qualitative work evidences that the number of EHR providers in the market, and the need for different interfaces to exchange clinical information between them, is still reported as an important barrier for HIE [
10]. A recent survey of third party HIE organizations supports the issues of vendor information blocking, with half of those surveyed reporting that they had experienced information blocking by an EHR vendor [
38]. Finally, vendor choice remains an important determinant in the successful implementation of MU objectives [
11,
39].
Conclusions
Identifying the barriers for information exchange is a necessary step to achieve the goals of the HITECH Act in creating a more efficient and effective healthcare system. Our research finds a relationship between the existence of dominant EHR networks and the exchange of clinical care summaries, which has important policy implications as the meaningful use program continues to transition to future stages. In fact, there is some evidence that information blocking could be partly the result of vague policies that undermine public exchanges.
Even though the current certification process for EHR products requires the use of a common language, there are several gaps that permit variability in its implementation. These gaps allow EHR vendors to implement information exchange capabilities in different ways. A clear example is the implementation of Care Everywhere, which has been successful in increasing sharing among Epic users. Nevertheless, the existence of isolated networks means that many hospitals are left out. In the case of Epic, this affects smaller and rural hospitals disproportionally (only 21% of hospitals that use EHR vendor Epic are rural which is significantly less than the sample mean).
In order to avoid proprietary exchange networks that foreclose some hospitals, it is important for the current regulation attempt to be more inclusive of hospitals that do not use large vendors and are therefore unable to use proprietary methods for exchange. Incentives could be tied to open exchange using previously defined standards rather than metrics that just measure if HIE occurs. For state level incentives, it might be necessary that state programs identify hospitals that are being left out of the exchange networks and offer technical and financial support. In our analysis at the state level we find no significant relationship between the percentage of hospitals that participate in health information exchange and the policies implemented through the State Health Information Exchange Cooperative Agreement Program. Our research suggest that future state level policies should take into account the different market conditions of EHR vendors in order to accommodate hospitals that may be left out of large proprietary networks.
Finally, although our findings suggest the importance of a network where information is exchanged only among hospitals that use a specific EHR vendor within a region and a state, further research is necessary to validate this relationship. Current information collection efforts only ask if information exchange occurs. More work needs to be done to determine the methods of exchange, including interviews with hospital staff that might give us some insight on if and why proprietary methods of exchange are being used.