Curbside Consult with Dr. Jayne 6/26/17

A reader recently reached out with some thoughts on life after a large go-live:

Our large academic medical center went live with ambulatory EHR several years ago. The clinicians and residents were used to many of the system features already from inpatient, but we still had a lot of configuration decisions, setup, training, reduced volumes, then a fair amount of post-live elbow-to-elbow support in decreasing amounts. Though there were a few frantic phone calls with crying and screaming clinicians or administrative staff at the time, it went fairly well all things considered.

However, many post-live optimizations were never completed and it was assumed that new hires could just be trained by existing staff. There is minimal formal training and no discussion of the individual configuration options that we helped people set up during rigorous pre-live training. We lack discussion of workflows and regulatory requirements that have shifted or are no longer tracked, and other changes have been made to the system that have broken prior customizations. Documentation of our individual decisions was by vendor consultants and I don’t think any coherent documentation was left behind at the end of the engagement. We aren’t even alerted to processes that have obviously become broken because the front-line clinicians and staff don’t know any different, assuming that it’s just the poorly designed software at fault. And the further we are from go-live, the worse it gets. It’s like throwing the frog in boiling water or turning the heat up gradually.

Do other systems or consultants do a better job of managing this as they find themselves several years post go-live?

At least in my experience, many organizations struggle with this. However, I see it more acutely in organizations that treated their EHR projects like IT projects instead of operational or clinical projects. The go-live itself is often seen as the endpoint, with little vision around the ongoing efforts needed to maintain a system and its users at a top tier level of performance. There is a lot of money spent to support the go-live, so groups tend to economize on ongoing support.

It sounds like your approach leading up to the migration was fairly tried and true, making the most of existing knowledge from the inpatient system while tending to the decisions that needed to be made specific to the ambulatory system. You had a good amount of elbow support, which many clinicians appreciate. Beyond that, many groups find a greater level of success spending more resources upfront to encourage (and/or force) providers to complete a set number of test patient scenarios prior to the go-live, which potentially makes for an easier go-live with less reductions to the schedule or less elbow-to-elbow support.

I personally like requiring physicians and their care teams to document a good number of patients with their most common chief complaints, along with documenting sample visits on some of their most complicated patients. That tends to prepare them a bit better and they have better mastery than if they try to learn during go-live. I’ve found the stress of the go-live itself tends to make learning difficult.

As you mentioned, post-live optimization is where things often fall apart. Some organizations don’t even budget a post-live optimization program into their implementation, which is a grave mistake. Budget permitting, I like to perform circle-back visits at two weeks, 30 days, 60 days, and 90 days after go-live. This allows the support or implementation team to see what processes are working well in the office and what processes have become ripe for bad habits. Even with the most rigorous training and practice, it’s hard to retain all the nuances of different EHR workflows, especially for patient care situations that you don’t see every day.

For those groups that did budget a post-live optimization program, I frequently see those resources shifted to other initiatives that have taken priority for one reason or another. Maybe the group shifted into acquisition mode, maybe they joined an ACO, but optimizing the EHR and practice operations seems to frequently fall by the wayside.

You mention shifting regulatory workflows and that is an issue I see frequently, especially with practices that participate in multiple grant programs. Once I worked with a group that was insistent that they needed to document the date of the last dental exam on all patients. I continued to ask “why” to every reason they gave until we distilled down to the fact that it was originally mandated for a grant in which they hadn’t participated in more than three years. They had been on the brink of customizing a template to capture that date, not knowing that it wasn’t important except for a sub-group of patients for whom that information was already captured in the system’s health promotion templates.

Institutional memory can be a blessing and a curse in situations like this, the latter when people remember things being one way but not the underlying reason and are so dedicated to keeping things the same that they lose sight of what they are doing. It can be a blessing when you have a stable workforce that can do things like train new workers, but that is certainly the exception in many ambulatory workplaces today.

The idea that workers will just train the new people as part of their ongoing daily duties doesn’t tend to produce desired outcomes. In practices where I’ve worked, on-the-job training has been a bust as trainers don’t have time to focus and trainees don’t understand what is best practice and what is their trainer just making it through the day. Fortunately, in my current practice situation, our version of on the job training actually has a rigorous schedule behind it with checklists and skill proficiency. The trainer and trainee are added to the office schedule on top of the normal staff, so that the training process can be focused. It costs more up front to take this approach, but it’s been more than worth it.

Training of new employees has to include training for user-level preferences and configurations because these are the things that make EHR workflows efficient and personal. When I perform EHR optimizations (or EHR clean-up missions, as the case may be), these are the first elements I emphasize. They’re often the proverbial low-hanging fruit that gets users into a more receptive state of mind for when you come back to cover more challenging workflows.

I cringed when I read the comment about the documentation of decisions being done by consultants who didn’t leave coherent documentation. That’s one of the things that pushes me over the edge. Documentation and hand-off should be part of every engagement, to ensure that your client hasn’t simply been handed a fish, but rather taught to tie his own flies, cast the line, reel it in, fillet it, and cook it over a fire that they have built.

In my consulting engagements, the decisions are documented not only in a spreadsheet-style matrix, but in a corresponding executive summary slide deck. It’s not enough to know that a customization was made, but you need to know why so that you can determine whether it needs to be maintained. Customizations should be reviewed with every major upgrade and evaluated to see if they need to be retained or if they can be retired in favor of new functionality. It’s also a great opportunity to make sure the physicians for whom they were built still work in the organization. Otherwise, as a general rule, the customizations can be put to rest as long as no one else has adopted them.

In those situations, I like to use database queries to determine if the customizations are even used. I once worked with a physician who was ready to fight tooth and nail to keep a customization until I showed her the queries that proved that out of every 100 times she used the template in question, she only used the “have-to-have-it” checkbox one time. In that situation, free-texting would not have killed her.

The comment that users assume the software is at fault rather than looking at the process also resonated. I’ve found that the organizations that handle long-term sustainable process improvements the best do so because they have dedicated teams that continue to work with practices to make sure changes are adopted and incorporated in an ongoing fashion. They make sure users have ready access to training in a variety of formats, whether written, recorded, live, or 1:1. They recognize that users have different learning styles and often crazy schedules and may need accommodation to become truly proficient with an application. And they’re willing to challenge whether it’s a problem with the user, the training, the content, or the technology. They’re not afraid to ruffle feathers getting to a root cause or trying to do the right thing for patient care and user satisfaction.

I work daily with clients who aren’t aware that their vendors have documentation around not only best-practice EHR workflows, but best practices for running the office in general. Several vendors have in-house consultants who are available to help clients with these issues, although I’ve seen come clients give them the cold shoulder because the feel the vendor-employed consultants are inherently biased. I’ve seen them argue with vendor educators who are trying to emphasize well-documented and published clinical best practices, belittling them and dismissing their wisdom just because their paycheck comes from a vendor.

The best example I’ve seen is a group that argued with the vendor about hanging signs to encourage diabetic patients to remove their shoes and socks for a foot exam. They told the vendor it was outside the vendor’s scope, despite the vendor rep being a registered nurse and having citations from articles proving the approach as effective in improving foot exam performance metrics.

The bottom line is that some groups do handle the ongoing maintenance of a system better than others. Those that have a plan accompanied by leadership buy-in and a corresponding budget do best. Others that don’t meet those criteria often become easy prey for vendors trying to sell replacement systems. It’s amazing to me when a client won’t sign a $50,000 proposal for optimization, but ends up paying millions for a new system when their previous system would have been just fine had they maintained it. It’s like never changing the oil in your car and then being surprised when the engine seizes.

How does your organization handle post-live support and optimization? Email me.

Email Dr. Jayne.