John, who blogs over at Chilmark Research, posted a note recently about his conversation with a Massachusetts urologist regarding the latter's experience with HIEs and EHRs (see: Here an HIE, There an HIE, Everywhere an HIE). He makes a lot of interesting points. Below is an excerpt
This urologist’s practice has been using eClinicalWorks (eCW) for several years now and despite their proficiency with using this EHR, the practice has never fully recovered the productivity it once had. Regardless, they have come to accept this hit on productivity as just the cost of doing business....Harry spends two days a week in surgery with most operations taking place in one of four facilities: local, unaffiliated hospital Winchester (who uses Meditech), Beth Israel Deanconess (has a home-grown EHR but encourages affiliate practices to adopt eCW), Childrens’ Hospital (a Cerner shop that also promotes eCW in ambulatory) and Partners which is now moving from its homegrown EHR solution to Epic (BTW, in a recent conversation with a contact at Partners learned that they are spending $1M/day for next five years on Epic switch - ouch!). All of these hospital organizations want a closer affiliation with these urologists in support of future value-based payments and of course just getting these physicians to do more surgeries at their respective institutions. Thus, all of them want the urologist practice to adopt their interoperability model.
Harry stated that Partners, the biggest healthcare organization in metro-Boston and arguably New England, is pushing particularly hard for them to switch to Epic as Epic does not have an HIE offering (Epic Everywhere is not an HIE in our definition nor apparently in ONC’s) and encourages its customers to put all ambulatory affiliates on Epic instead. In addition to these organizations, the Commonwealth is also encouraging this practice to join the statewide HIE. After the pain and suffering Harry’s practice went through to become proficient on eCW, they are loathed to switch to Epic. Besides, switching to Epic would limit their ability to connect with other healthcare organizations they work with as Epic does not play well with others. Harry’s situation is not unique and is likely being played out across the country, especially in urban areas where there may be a number of competing healthcare systems each trying to establish their own HIE. In such a situation what is an independent physician practice to do? Certainly they could sell the practice, as many physicians have already done, to the highest bidder. Not an option for Harry and his physician partners as they like their independence and plan to keep it that way. They could turn to the statewide HIE and hope that it will provide the depth of services (interoperability) to enable them to connect and share records in support of care coordination with all hospital systems they work with. Ideally, this may be the best approach but unfortunately they’ll be waiting a very long time for this to happen, if it happens at all. ill leaves a very critical 20% unresolved.
There's a lot in these paragraphs to discuss and I don't exactly know where to start. The first point that I will take up is John's comment: Epic Everywhere is not an HIE in our definition nor apparently in ONC’s. I agree with him that Epic does not play well with others and made the same point in a previous note: Judith Faulkner, EMR Interoperability, and Washington IT Politics. Here's a quote from it:
[Judith] Faulkner belongs on the Health Information Technology Policy Committee for one important reason. Her company, Epic, holds a near-EMR-monopoly for the larger hospitals in the U.S. (see: Does Epic Exercise a Near-Monoply for EMRs in Larger U.S. Hospitals?) A central tenet of the Epic business model is hospital client control, driven by the fact that the hospitals install the software that Epic delivers when the company delivers it. In essence, she determines the IT strategy for some of the most important hospitals in the U.S. As noted above, the hospitals are also under intense pressure to install the Epic departmental systems (e.g., lab, radiology, oncology), thus ensuring smooth integration and "interoperability" and widening the influence of the company.
A second important point is that the urologist, after installing an office EHR, concluded that his practice would never fully "recover" and achieve its previous level of efficiency and productivity. I have blogged previously about the rapidly disappearance of private physician practices (see: Physician Private Practice Declines; the Last Barrier to Emergence of "Big Medicine"; Why Is the Federal Government Hostile to Private Physician Practices?; More Data on Physicians Leaving Private Practice; Discussion of Findings). There are many reasons why physicians would leave private practice or not enter one at the conclusion of training. However, John't note suggested to me one more reason for this shift. Both physicians and hospitals are being pressured to install EHRs. It may be the case, as with the urologist described above, that physicians in private practice have come to the conclusion that their efficiency is irretrievably harmed by these systems. Therefore, they logically make a decision to take salaried hospital positions where their compensation is less directly linked to their personal productivity and efficiency.
Does John, who blogs over at Chillmark Research, have real-life examples of where Epic is not interfacing with others or is he just propagating stories without real information?
It is interesting that you chose not to select the insightful comment from the same link-back post: "Now comes a little secret that is not discussed much. Hospital CEOs and CIOs have little interest or enthusiasm for multivendor interoperability, either within their hospitals or hospital systems or across the outside hospital systems that they compete with. Customized interfaces between heterogeneous systems are a pain in the rear-end for the CIOs."
It is my experience, dealing directly with Epic, with multiple organizations, that they respond in a timely manner regarding interfaces to other systems. I see the following as the top three issues: 1. It is the lack of standards, mostly in the area of semantics (the content of the interface unless an organization choses to post the interfaced data as a set of non-codified text)
2. Lack of qualified resources in small practices, in the healthcare organization integration teams and in vendors (including aligning timing of resource availability between all parties... at least 4 groups)
3. Syntax issues with the flexibility of HL7 and each healthcare organization's and their vendors' chosen implementation
Thank you for the opportunity to share my experience.
Posted by: Art_Vandelay | June 02, 2013 at 09:28 AM