If the job description for the CMIO role doesn’t include the ability to diffuse tensions and de-escalate situations, perhaps it’s time for a change. In fact, with clinician burnout becoming so pervasive, a willingness to “dig a little deeper and understand” their concerns — particularly around alert fatigue and customization requests — has become paramount for IT and clinician leaders.
“These are things that are bothering them for a reason that’s very real to them,” said Bill Manard, MD, CMIO at SCL Health. By simply having brief conversations around why policies exist “can help reduce frustration significantly.
During a recent panel discussion, Manard and co-panelists David Allard, MD (CMIO, Henry Ford Hospital and Health Network) and Jose Barreau, MD (Co-Founder & CEO, Halo Health), shared their perspectives on the biggest challenges organizations face in presenting data to providers, the measures that can be taken to reduce alert fatigue, and the keys to building and maintaining a successful vendor partnership.
Identifying the Issues
Before anything can be done to resolve issues, leaders must first uncover them, said Allard. For his team, that means appointing clinical liaisons to do rounds, and having nursing and provider advisory groups meet with EHR builders and analysts to ensure their voices are heard.
Manard’s team employs a similar approach by getting out and talking to clinicians to identify the pain points and try to determine whether a system-based solution already exists. Because SCL Health is an Epic shop, they’re able to leverage their own data, as well as data from other users, to help determine which interventions to utilize.
The ultimate goal, according to Allard, is to present data in an informational way, rather than overwhelming users with “a slew of numbers that needs to be digested.”
According to Barreau, it starts by stripping down to the basics and asking a simple question: how difficult is it for physicians and nurses to do the physical work of documentation and real-time communication? Chances are, it’s pretty difficult. “Every time you get interrupted with an alert, it’s an interruption of patient care that causes a break in concentration,” which can affect both care quality and patient satisfaction.
Halo Health’s approach is to look at what steps need to be taken to deal with the interruption, how intuitive those steps are,” he noted. “The less steps you have to take, and the more intuitive they are, the easier it is to do your job.”
And while technologies can certainly help curb some of the issues, educating users and implementing best practices are just as critical, if not more. One of those best practices, according to Manard, is applying the five rights framework to ensure the right information is going to the right person, at the right time, using the right tools, and fitting into the right workflow. Many alerts, however, don’t meet those criteria.
As a result, his team is undertaking a complete review of all alerts that appear within the system, and working to eliminate those that cause unnecessary interruptions, which can break the train of through and increase cognitive burden. “It’s simply intolerable to allow these interruptions to prevent our clinicians from providing excellent care.”
The problem is, but cutting off all interruptions, organizations can limit communication and collaboration, which isn’t ideal. The answer is to determine which alerts are, in fact, necessary, and which can wait.
Let the Data Flow
The big question, of course, is how that can be done. During the discussion, the panelists shared practices they use within their organizations.
- Everything isn’t urgent. For some, the instinct is to transmit a piece of information as soon as they receive it — even if it isn’t necessarily timely, said Barreau. The majority of the time, it can be sent through the EHR and looked at when a clinician is seeing that particular patient. In these situations, the onus is on the health system to decide what’s urgent and must be viewed in real-time, and what isn’t,” he noted, adding that a very small percentage falls into the former category.
- Periodic reviews. At Henry Ford, the policy is to review all alerts periodically, according to Allard. “We have dashboards to say which ones fire the most, which ones are ignored the most, when they fire and to whom, and what’s the false positive rate.” It requires patience, as it is work that “never ends,” but is well worth the effort.
- Say no to zero tolerance. Although in theory, a zero tolerance policy might work, in reality it doesn’t, for one reason: “You’re going to miss some important things,” said Barreau. Keeping patients safe, which of course is the main objective, “You have tolerate a little bit of alert fatigue for the benefit of the patient.”
- Configure, don’t customize. What leaders don’t want to do is present a blank slate, he noted. “When you build an enterprise platform across the system, you have to make it configurable — you can’t make it customizable.” Allard agreed, noting that although some tweaks should be allowed as to where or how providers receive alerts, it should be done on a system level, and not an individual level, especially when it involves clinical decision support. “I don’t want to customize a whole lot; either it’s important enough or it’s not.”
- Think system-wide. It’s important to remember that alerts are a component of decision support, according to Manard. “We’re talking about decision support tools, and data presentation is a huge part of that.” As part of SCL Health’s shift from “a purely user customizable mentality to more of a system mentality,” they’ve implemented a system in which, rather than present users with 300 reports that can present the information clinicians need, they provide five reports.
- Appoint committees. Like many organizations, Henry Ford has established clinical decision support committee that continuously evaluate alerts and review requests that come in, said Allard. “The goal is to make a reasonably quantitative assessment and determine the benefit of firing an alert, what’s the damage if it’s ignored or missed, and what are the opportunities.”
- Use guard rails. Finally, realize that there are always going to be alerts that don’t fit into the workflow, but are still necessary, such as regulatory requirements. “You need to have those guard rails in place,” he added.
The final – and perhaps most important – piece is communicating with vendors around change requests; something that has been a sticking point for most organizations, including SCL Health.
“We have historically allowed a lot of changes to go through without formal reviews,” said Manard, which has led to quite a bit of friction. “Now when we engage with a vendor in this space, we rely on their expertise.” Part of that is flipping the script when presenting to operational stakeholders to say, why are the processes that work for similar organizations not working for us?
Another key component is being willing to deny requests, but do it in a way that’s both respectful and informative, he noted. “It’s really helpful to explain the rationale behind it, while also validating their frustrations.” Doing that, however, isn’t always easy. In fact, it’s not uncommon for clinical and IT leaders to face harsh criticisms. But, unpleasant as it may be, “de-escalation” is part of the job. “You have to be able to diffuse tensions and explain to our clinician colleagues why decisions are made, and why systems function the way that they do.”
Of course, if the request isn’t feasible — or stands to benefit a small number of people, it’s important to be transparent, said Allard, adding that “empathy, understanding and education are critical.”
To view the archive of this webinar — Analyzing IT’s Opportunities to Reduce Clinician Burnout (Sponsored by Halo Health) — please click here.