George Weldon is the chair of the Alarm Committee of Northern Westchester Hospital (part of Northwell Health, the largest IDN in New York State). Since March of 2015 his day job is Director of Telecommunications, but over the past four years what might normally be a pure operational role has mushroomed into something much larger. George has leveraged the telecommunications systems and technologies to spearhead an hospital-wide push to improve both the patient and the caregiver experience. In collaboration with the CareSight team, George and his colleagues use data harvested from their nurse call, mobile voice, EMR, and related systems to evaluate their progress against target goals 24 hours a day. It is proving to be a strong foundation for continuous improvement.
I recently chatted with George via Slack (appropriately an online messaging/collaboration platform) to check in on how he’s doing, but also to see how others can take advantage of what he’s learned along the way. Our conversation follows.
Looking Across Systems and Technologies to Improve Outcomes
Northern Westchester is a relatively small community hospital, but its size is an advantage. They have not skimped on their technology investment, especially commensurate to their size, and they have been able to more easily innovate. Based on what we sometimes see with bigger customers, they are more able to “think out of the box.” One way they’ve done that is by thinking of their patient care systems as non-vendor specific platforms, each with important data that can be correlated.
KS: Thanks for agreeing to participate in our Blog. Let’s start by sharing some about you and your work. Give me a quick snapshot of your background and describe the work you do with customers at NWH.
GW: I started in a completely unrelated role of Business Manager, which focused heavily on finance. Quickly after, I moved into a Communications Manager role where I adopted several systems and broken/absent processes to support them the systems I was responsible for.
In this role, I identified opportunities to streamline work and manage inventory, and then began looking at our analytics. As my role developed, I became a Director and began taking on the responsibility for newly acquired technology such as Rauland Responder 5 Nurse Call.
KS: You seem to have taken what is usually a very cut and dry operational role and made that into something else.
GW: I tried to get abstract and explore ways we can look at data to improve outcomes.
KS: Were you being asked to do that? Why did you seize the opportunity?
GW: I was never asked to pursue analytics. It was my own doing. As an administrator, I was supporting multiple systems that, operationally, became a well-oiled machine. I had an opportunity to analyze the systems I supported from a new angle—with data. It peaked my curiosity.
KS: So are you saying that you needed to look at all the systems together to work on outcomes?
GW: Yes, and once I was able to get my hands on varying disparate systems, I knew I could eventually work toward something valuable.
KS: And in so doing you could align with bigger organizational goals?
GW: Correct. We had to reframe how we looked at the data so that we could place greater focus and effort on staff satisfaction. Instead of viewing data only in terms of how it affected patients, we flipped the paradigm and also began using the data to see how we could help our staff. We knew that addressing staff satisfaction would have a positive domino effect overall.
KS: So you could help them do their jobs better by giving them supporting data and the ability to make decisions about workflow?
GW: Yes, that is correct. Our focus became very much about staff satisfaction and identifying the low hanging fruit that could decrease alarm fatigue.
KS: OK...good pivot. Let's talk about the low hanging fruit.
GW: I began combing through what I believed to be our opportunities, or low hanging fruit: codes, clinical call flow, and alarms from nurse call.
Based on personal experience with a family member in the hospital combined with my professional exposure, I formed a list and pitched it to leadership.
KS: And the list of opportunities to improve staff workflow, it expanded across the facility and clinical roles?
GW: Yes. Also, I knew going at this alone wouldn’t get me far, so I lobbied to form my own committee. In March 2018, the Alarms Committee was made a reality.
KS: And is the Alarms Committee well represented from across the hospital?
GW: The committee is composed of nursing leadership, quality, education, Biomed, Telecom, ED, line staff nursing/PCAs as well as industrial engineers. Initially, it was co-chaired with the life safety officer, whose support and understanding of the low hanging fruit was huge in getting lift off.
KS: It’s given you an opportunity to talk the same language.
GW: We were all able to begin speaking the same language and understand what the varying alerts/alarms were. We took that list of low hanging fruit I had previously identified to a multidisciplinary team of professionals and we dove into the data.
We started with Codes. At NWH these were going to all users on wireless communication devices (and in many cases, also paged overhead the old fashioned way). We quickly decided that instead of blasting everybody, we would only send Codes to first responders.
KS: Why did you target the Codes? What was the benefit?
GW: Our goal was to target alarms to staff that needs them with the concept of “if you are not responsible to respond to the code type, you shouldn't be bothered with hearing it.”
We quickly went from 415 simultaneous users receiving an urgent Code to an average of 45 or so targeted staff members. There was an immediate reduction in noise and mass disruption of all roles across the organization that wear the wireless communication technology.
Keep in mind that this had no impact on response time and no impact on patient outcomes, but the byproduct is that the bystander effect disappeared. Observers/onlookers who had no business being there were focusing on their own work instead of being distracted by codes that were not relevant to their function.
KS: OK, #2 on your low hanging fruit list?
GW: Again, working towards targeting communication, mass mobile device audio conferencing was #2.
The use of conferencing behaves similarly to a walkie talkie—push a button and get a group. The “push to talk” function that was initially targeted to many was now focused on a few, crucial first responders.
KS: You had some hunches about this, you observed it, and you had data to support your hunches. Then you considered the impact and planned to overcome it?
GW: Enhancing call flow was the goal. Calls targeted by floor meant only staff associated with the floor where a patient was at risk were to be notified.
We then proposed how the call flow should be built. And with the data to support the argument, the changes were implemented—resulting not only in dramatic decreases in nuisance alerts but also reductions of what was often pure noise on the floors. This obviously impacted patients, but clinician distraction also decreased.
KS: Hospitals worry a lot about stuff being missed, so the tendency is to over communicate. Overcommunication has an impact, a load...
So the recommendation is to put tools in place to understand the load, otherwise, you’re flying blind?
GW: Correct. We love automation and love to keep everyone “in the know,” but that is a double-edged sword that wasn't obvious or visible until we assessed with data. Before we started tweaking, the decision was to nearly always to notify all.
Making Decisions About When to Notify Caregivers
At the tail end of a major patient care redesign, NWH looked to Connexall, Vocera, and Rauland to reinforce their performance goals. NWH added in integration between these technologies to close the gap between patient needs and responses. Vocera and Connexall created greater visibility to patients, but six years later the organization was in the position to revisit the impact of those flow decisions.
GW: Another challenge pervasive across the country is alarm fatigue. To help reduce alarm fatigue at NWH, we used CareSight’s analytics to determine safe and appropriate parameters to adjust unnecessary alarms. For example, a Cord Out alarm will sound if housekeeping inadvertently unplugs a cord while cleaning a room. Even transporters moving our patients can trigger an alarm. Implementing a five-second delay for Bed Exit and Cord Out alarms allowed those alarms to self-correct in certain situations. It also allowed us to better determine when those alarms should be dispatched to mobile communication devices.
KS: How did you determine which alarms would benefit from the delay?
GW: CareSight analytics told us how many of these alarms were physically cancelled within the room, by a staff member, in five seconds or less. In order to turn that alarm off, staff had to physically cancel it at the bedside. However, even if they canceled in under a second, that alarm went to all RNs/ PCAs associated with the pod where the room was alarming.
KS: So it seems that the same elements came into play here...you did observation, had some hunches, validated it with data, then did education. Then you gave management a practical way to attack it.
GW: Exactly. With data, we were able to validate what we already knew: unnecessary alarms exist. We were able to correct it by implementing a simple five-second delay. The result is approximately 25,000 alarms being silenced per year. For a small hospital, that’s a lot. Add in how incredibly disruptive this can be at night, and it’s a major improvement to the patient’s experience at NWH.
CareSight Analysis of Nurse Calls via Time Banding showed NWH that the implementation of a five-second delay could reduce false alarms by 33%
Reordering Who Responds to Calls
Nursing is extremely difficult work, whether one is an RN or a Patient Care Assistant. Orchestrating the wide range of priorities in real time is challenging, especially if the staff-to-patient ratio is low on any given shift. Hospitals are trying super hard to be responsive, but who should respond and what should happen when there is no response can be a major point of contention.
KS: Let’s wrap with one more "low hanging" fruit. Number 4?
GW: Clinical call flow. The original call flow was that notifications were sent to RN (Nurse), RN Buddy (partnering nurse) and the Charge nurse.
All of these roles had their own caseload on the unit to attend to and their own calls to respond.
KS: Lots of redundancy. No matter what, make sure someone responds.
GW: The PCA (an integral member of the patient care dyad) was never involved.
The need for redundancy made complete sense, but the “who” in the storyline did not.
We implemented a log in my telecom department capturing every single call that came down and pooling them into buckets. Calls for TV help, bathroom help, help out of bed, or to order food were very much PCA-driven tasks and were often, if not always, delegated from the RN to the PCA.
A sample size of just under 10,000 calls produced jaw dropping results.
Over 85% of the calls were for PCA-driven responses.
Anything that was related to pain/nausea or a request specifically for the RN or MD was categorized as true RN response need.
Leadership could not argue with the data.
The decision to change the call flow was agreed to in October 2018, whereby the flow became RN, PCA, RN Buddy.
KS: So you became a true partner to the clinical staff. Unusual, as often someone in your role is purely an executor/implementor.
GW: Correct. This was one of the first times a process designed by our leadership was actually revamped, but it was supported by data and made it an easier sell (but still no easy feat).
KS: But you were able to make a real case.
GW: This change in call flow has resulted in nearly a 20% reduction in call load to RN on our inpatient units.
KS: And you educated leadership and the line level staff as part of the process.
GW: Very much so—to the point they are now discussing putting the PCA in the front and RNs following.
KS: It's a great story to tell.
GW: While our response time has remained fairly consistent (we have not shaved minutes/seconds off the response time stamp so to speak), our HCAHPS in both quiet at night and response to call bell has consistently improved since July (the first implementation/change of the low hanging fruit). Quiet has gone from the 50th percentile to the 63rd. Response to call bell has gone from the 79th percentile to the 86th.
KS: And you have a foundational framework, organizational buy-in, and credibility.
GW: Yes, administratively—a pre-/post-implementation survey was given to all RN/PCA staff.
KS: And?
GW: 59% of staff (N size of 142) reported feeling less overwhelmed/more focused on the task at hand since the reduction in noise.
In terms of what had the greatest positive impact, 38% of staff reported the reduction in codes, 34% indicated the reduction in conference blasts, 16% reported the delay and 12% noted the call flow change.
The majority of staff have noticed a difference.
KS: The improvement in patient satisfaction scores and staff surveys is a winning combo.
GW: Over the last six months we have had great success and it seems to be a sustained impact. It was super important that our efforts show results and that we have ways to keep our eyes on the ball.
KS: We are very excited about the work you are doing and the collaboration.
GW: And we are excited about using our foundation and expanding our reach :-)
Kenny Schiff is the Founder and CEO of CareSight. A 20-year veteran of the healthcare technology business, Kenny is considered by his customers and peers to be a no-nonsense, trusted resource who can be counted on to deliver complex solutions with high impact. His team pioneered managed services to clinical communications customers starting in 2003. Visionary always, but never afraid to be hands-on, CareSight is a great creative platform for Kenny’s entrepreneurial and technical passions.
George Weldon is the Director of Telecommunications for Northern Westchester Hospital. After starting his career as a licensed clinical therapist, George gained a unique clinical perspective of health technology and communications. On a day-to-day basis he oversees the department responsible for wireless mobile communications (Vocera, Tiger Text), nurse call, clinical alarm responsiveness data analytics, and several integrated technologies. He also supervises the administration of their PBX system and all audiovisual related systems.
George holds a Masters Degree in Psychology and Mental Health Counseling from Pace University. George was a top “40 Under 40” awardee of Westchester County. He continues to prove himself in the world of healthcare technology and analytics.
