I’ve been working on a project involving laboratory interfaces for a mid-sized multispecialty practice that is trying to integrate with multiple local hospitals. They’re valiantly trying to stay independent, which is quite a challenge given the rampant consolidation that is going on in nearly every healthcare market.
The practice’s leadership figures that if they interface with the hospitals in addition to the reference labs they already send to, it will make it easier to manage patients regardless of where they are admitted. As they were putting together this plan, however, they didn’t understand the complexity of working with organizations that aren’t entirely focused on earning the practice’s laboratory business like the national reference labs are.
Since the practice’s previous interface projects took 30 to 45 days, they assumed that working with the hospital would be the same. They also assumed that the hospital laboratory representatives who regularly come to the practice to tell them about new tests would be able to assist them in navigating the entire process, not realizing that those reps were more sales agents than true account managers.
The first surprise came when Hospital One told them it would be a minimum of three months before they could even talk about a timeline for starting a lab interface project, so they would have to stay on paper for the foreseeable future. It would be a fairly straightforward process to create a printable laboratory requisition so we could move the practice away from the hospital’s carbon-paper form and into EHR-based ordering. However, the lack of an interface had already created a significant amount of extra work for the nursing staff who was expected to manually key all lab results that were related to reportable clinical quality measures.
Even though we couldn’t fix the interface problem, I helped them create a new workflow for keying the results, which involved their medical records staff in addition to the nursing staff, so the workload could be better distributed. Cross-training is always a good thing, and assuming adequate training and quality assurance review, there was no reason why the medical records staff couldn’t be part of the workflow. Still, given the nature of the one-off workflow to key results, compared to the interfaces with the reference labs, I didn’t foresee the practice sending any more orders to Hospital One than they had been with handwritten orders.
Hospital Two was a significantly more accommodating, probably in part due to the fact that the practice hadn’t been sending business to its lab previously. Although they didn’t have available staff to assist with a bi-directional interface project, they were willing to set up a results-only interface that would at least allow discrete results to come into the patient chart without the staff needing to be involved.
Unfortunately, the client’s EHR handles this type of situation by creating two orders in the patient chart — one for the actual order and one that is created when the unsolicited result hits the system. This leads to extra work because someone has to reconcile the orders and match them up, and it would leave the practice with the same amount of extra work as the first hospital. When I mentioned the inconvenience and asked if they were willing to help us implement a workaround that would function as a semi-solicited interface, they were eager to hear about what it would take.
Having done it with other clients, I knew the hospital’s lab system was capable of holding the client’s internal accession number, and that keying it on each order would solve the problem. Usually only about half the hospitals I interact with are willing to do this, often citing the risk of error or the magnitude of the extra work for their lab staff. However, this facility jumped at the chance to see if they could make it work in order to obtain a piece of the practice’s business.
They were so eager to move the project forward that they agreed to send someone to the practice to key in the orders for testing so that the practice didn’t have to hardly expend any resources. Once the orders were keyed, they resulted them promptly, faster than almost any hospital lab I’ve ever worked with. The entire testing phase took barely more than a week and they resolved any issues that were found by the end of the next business day. I have to admit, it was a dream project and the entire thing was done in less than four weeks.
Many of us in healthcare are a tiny bit superstitious (never say the word “quiet” in the emergency department) so I knew that given the success of the project with Hospital Two that the next project was likely to be a nightmare. My vague suspicion grew into actual worry when I met the IT project manager the hospital had assigned to the interface project. I could sense the rarified air around him as soon as I walked in the room and had to suffer through his overly complicated explanation of what an interface project entails. I think he assumed that as a physician I didn’t know anything and he totally missed the part where the practice administrator explained that I was their consultant and had assisted multiple clients with interface projects.
He went on for a good 20 or 30 minutes that seemed like a lifetime, talking about all the important work the hospital IT team would be doing to make the interface happen and how little the lab and practice teams would impact the process. When I finally was able to jump in and explain my experience and the practice’s goals and objectives, I was treated to a rainbow of colors on his face as he went from angry red to bilious green to white. I think it had honestly never occurred to him that anyone on the practice side could have a clue how things should be done.
Since he claimed he didn’t have a sample project plan to review with us, I provided him with my own, which produced an outstanding level of pallor as he realized he wasn’t going to be able to put one over on us. We asked him to review the proposed timeline and comment on it and he said he would be able to get back with us in the next couple of weeks. That’s never a good sign, but I couldn’t tell if he was actually backlogged or just being passive aggressive. As time went on and he haggled about everything from the selection of components for the test scripts to the way in which labs would be resulted, I knew it was the latter. The project has been stalled in every imaginable way, with various resources being unavailable or on vacation at various times despite the hospital having agreed to a project plan and timeline.
The practice’s pleas to hospital leadership have fallen on deaf ears. This week I’ll have to have a serious discussion about halting the project. We’ve been using too many resources with little return, and if this is how a hospital acts when a practice wants to send them their business, I doubt they’ll be responsive if there are issues. The other hospital’s semi-solicited interface has been working like a dream, and to the end users, it functions just like the reference labs’ bi-directional interfaces. There are a couple of kinks for the practice’s IT staff every now and then, but overall, it’s been a big success. There simply isn’t much reason to continue working with a competitor hospital that just puts roadblocks in the way.
It will be interesting to see whether the first hospital ever circles back to us or whether a halted project will bring the third one in line. I suppose some hospitals are simply so big that they forget about their base, or maybe leadership just lets certain constituencies run amok. I can’t say that healthcare IT will ever be dull and am grateful that organizations like this create job security for people like me.
How does your hospital earn business from independent practices? Email me.
Email Dr. Jayne.