Boombostic Health · 2026-07-21 · 14 min
Key moments - from our scoring
Substance score
59 / 100
Five dimensions, 20 points each
Hospital alarm fatigue represents a critical patient safety blind spot: 90% of bedside monitor alarms are non-actionable, causing clinicians to ignore alerts regardless of actual clinical significance. Dr. Darren Klugman, Chief Clinical Officer at CommWave, explains how the company's platform unifies historically siloed data from electronic health records and bedside middleware into a single structured cloud-based system. Rather than adding more data to overwhelmed providers, CommWave applies mathematical modeling and data science to deliver actionable clinical recommendations while automating monitoring parameters. The platform addresses not just patient safety but hospital operations - providing clinical engineering with real-time device visibility, enabling nursing staffing optimization, and using patient recovery forecasting to accelerate ICU discharge and optimize bed utilization. Klugman emphasizes that vendors must stop creating point solutions and instead offer integrated platforms that work across care delivery spectrums, while health systems must prioritize ROI and ensure data is being meaningfully applied to practice.
90% of patient-generated bedside alarms are non-actionable, meaning no clinical action is taken in response. This drives alarm fatigue where clinicians stop believing and responding to alarms, causing important safety signals to be ignored when they do occur.
CommWave applies real mathematical modeling and data science to structured, normalized hospital data rather than using generic LLMs like Claude or Anthropic; this produces actionable clinical recommendations tailored to actual patient monitoring rather than generic insights.
It uses mathematical modeling on integrated clinical data to identify when patients are ready for ICU discharge or step-down care, enabling hospitals to right-size bed utilization, accelerate patient throughput, and reduce patients' exposure to ICU-acquired harm.
Monitoring adjustments shouldn't vary based on which provider is at the bedside; data-driven parameters ensure consistent, safe, objective monitoring rather than subjective clinical decisions that waste provider cognitive time on tasks that can be safely automated.
Patient safety improvements and faster patient throughput (reduced length of stay), which combine to reduce harm exposure and increase revenue through optimized bed utilization and improved patient satisfaction.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode delivers several substantive claims about hospital alarm fatigue and data-driven care optimization, particularly the '90% non-actionable alarms' statistic and the cognitive overload problem for clinicians. However, there is significant repetition of the same core message throughout (alarm fatigue, data silos, workflow integration) without proportional depth. The second half devolves into general industry commentary and mutual backslapping with the host rather than drilling into new territory.
90% of patient generated bedside alarms are what are called non actionable, meaning nobody does anything about that
all of the data we have has eclipsed the human capacity to analyze it
The core insight about alarm fatigue leading to alert desensitization is well-documented in patient safety literature, not novel. The framing of data unification as a solution is reasonable but not counterintuitive - integrating silos and applying math to data are familiar moves in healthtech. The 'human in the loop AI' discussion recycles standard vendor positioning language without fresh angles or contrarian thinking.
90% of patient generated bedside alarms are what are called non actionable
we have data. The problem with healthcare is that nobody, many people are unaware of what data we have
Dr. Klugman is the Chief Clinical Officer of CommWave and has clinical training, which provides credibility on hospital workflows and patient safety. However, he is also a vendor selling into the market, which creates inherent bias. He speaks with operational authority but is not presented as an independent clinician evaluating systems - he's a founder-operator with skin in the game on his own product. Relevant but not a neutral practitioner perspective.
I'm the Chief Clinical Officer at CommWave and we are a patient safety and hospital operations platform
As a practicing clinician, maybe from your perspective
The episode cites the '90% non-actionable alarms' statistic repeatedly but provides no source, sample size, or context. Beyond that anchor number, claims are largely abstract: 'improved bed utilization,' 'accelerate patient throughput,' 'reduce length of stay,' and 'every day in ICU reduces IQ points' - all without supporting data, customer examples, or quantified outcomes. No named hospital deployments, no ROI figures, no before/after metrics.
90% of patient generated bedside alarms are what are called non actionable
every day you're in an ICU, two things happen that the public isn't aware about. Your IQ points are reduced and you're exposed to harm
Host asks soft, open-ended questions that allow the guest to deliver prepared pitches rather than probing claims. When the guest makes provocative statements (e.g., ICU stays reduce IQ), there is no follow-up questioning. The host agrees frequently ('Yeah,' 'Right') and redirects to align with his own company's messaging rather than challenging or digging deeper. This reads as a friendly peer conversation, not a rigorous interview.
Yeah, it's so great having you. And Darren and I went to undergrad together
Yeah. Uh, we agree. I mean at HC1 we built...
Computed from the transcript - who did the talking, and the words that came up most.
Too Many Alarms, Not Enough Signal: How AI Can Make Hospitals Safer | Dr. Darren Klugman Show Notes Hospital alarms are supposed to protect patients. But when nearly every signal becomes background noise, the warning that actually matters may be the one no one hears. Live from ViVE, Marcus Gordon sits down with Dr. Darren Klugman, Chief Clinical Officer at CalmWave, to examine one of healthcare's most overlooked risks: the growing gap between the volume of hospital data and the human capacity to interpret it. Dr. Klugman explains why alarm fatigue is more than an operational nuisance - it is a patient safety problem. Clinicians are being asked to manage increasingly complex patients, devices, scores, and alerts while still delivering the human care that technology cannot replace. The answer is not another dashboard or another point solution. It is a smarter operational foundation that unifies electronic health record and bedside device data, separates meaningful signals from noise, and turns information into recommendations clinicians can act on.
Transcribed and scored by The B2B Podcast Index.
Speaker A: Every one of us has been in the hospital and hears the incessant noise and the beeps constant. And if you've ever been in the hospital as a patient or a caregiver, those times when a provider walks in while the bedside monitor is alarming and hits silent and walks out and uh, you're sitting there as a patient or a caregiver thinking, wait, that noise was supposed to alert you that something's happening and you look at the provider and say, no, no, don't worry, that's nothing. 90% of patient generated bedside alarms are what are called non actionable, meaning nobody does anything about that. Non actionable alarms by in turn therefore lead to patient safety issues. Because when nobody believes the noises, the important ones are ignored.
Speaker B: Welcome to Boom Bostic Health, where we challenge the business of healthcare and explore bold ideas that drive meaningful change. Well, uh, good afternoon live at Vive, another edition of the Bombastic health podcast. And it's have the pleasure and honor of my good friend, Dr. Darren Klugman. Darren, how are you?
Speaker A: Great, thanks for having me. Really excited to be here.
Speaker B: Yeah, it's so great having you. And Darren and I went to undergrad together, uh, and excited really, uh, to kind of talk about our healthcare journeys and what we're seeing today. So first, Darren, kind of tell us about what you're doing at CommWave.
Speaker A: Sure, absolutely. So I'm the Chief Clinical Officer at CommWave and we are a patient safety and hospital operations platform where we bring together historically siloed data sets from the electronic health record and the bedside middleware, which is all of the patient vital data, all of the hardware that's fed into the bedside monitor into a unified, structured, normalized and de identified cloud based data storage platform. We then use that data to improve patient care, reduce patient harm, and ultimately have data informed hospital operations so that everything is streamlined from nursing, staffing, clinical engineering and ultimately. Now we're rolling out our newest use of this data set which is forecasting patient recovery.
Speaker B: That's great to hear. You know, um, uh, at Vive, there's just a lot around data and safety and interoperability. But what you don't often hear about is patient safety. Right, right. So it's a very unique angle. And as a practicing clinician, maybe from your perspective, what do you kind of, you know, through that lens, how do you see AI really being able to accelerate how we take care of patients within the four walls of the hospital?
Speaker A: I think there's two, there's two things that I'd like To point out the first is a story. Every one of us has been in a hospital and here's the incessant noise and the beeps constantly. And if you've ever been in the hospital as a patient or a caregiver, those times when a, uh, provider walks in while the bedside monitor is alarming and hits silent and walks out and you're sitting there as a patient or a caregiver thinking, wait, that noise was supposed to alert you that something's happening. And you look at the provider and say, I'm sorry, Dr. Gordon. And oh, no, no, don't worry, that's nothing. 90%, uh, of patient generated bedside alarms are what are called non actionable, meaning nobody does anything about that. Non actionable alarms by, in turn therefore lead to patient safety issues because when nobody believes the noises, the important ones are ignored. Right. And that's what we're tackling. But as you said, uh, we shouldn't tackle problems in silos. We have data.
Speaker B: Yeah.
Speaker A: The problem with healthcare is that nobody, many people are unaware of what data we have. And so we are very dogmatic and live in a world that this is all how we've always done.
Speaker B: Right.
Speaker A: So why is it that we should have a massive data set in the EHR and a massive data set that is significantly bigger than the EHR from the patient and not bring those together to inform care?
Speaker B: Yeah.
Speaker A: So you're 100% right. It should be data informed care delivery and there should be a role for AI. What we do is math and data science. So we apply mathematical modeling and data science to this huge data set. So it's not an LLM wrap on data that you can get off the shelf and claude or anthropic. This is real deep data science. I think one of the other things I'd say to your question is the following, which is that all of the data we have has eclipsed the human capacity to analyze it.
Speaker B: Right.
Speaker A: Patients are more complicated, care is more complicated. There's more and more devices integrated into the care delivery. And as that has expanded, we keep saying to providers, utilize this new data element. Here's a new score that's going to help you predict the providers can't utilize what they're seeing now. Right. So one of the fundamental things that we do at Comm Wave is we say we're going to take the data structure and normalize it and then apply mathematical modeling data science to deliver you recommendations. You can then decide to use it or not. But what we don't want you doing is Using cognitive time to think about things that you shouldn't have to think about.
Speaker B: Yeah, I think that's one of the biggest, that's always been one of the biggest barriers to technology adoption in the provider environment. Right. As a provider, you're certainly keenly aware of that. Um, often there's been technology there, but how do we use it effectively? How do we get it within the workflow? How does it augment what we're doing? And we like to talk about it, uh, with our agentic AI platform as kind of human in the loop. But what you're helping to do is actually go one step beyond that and say, okay, we're going to have your human, but your human is that AI agent that you can trust, and that verifies that. So then that next level, that task that you might have normally had to do, you don't have to do. You can trust us to help you go move on to your next task and move through your patient queue more effectively.
Speaker A: And all of those things are true, right? Yeah. And what we believe is that if you as a health system believe that you've hired the best and brightest to do this work.
Speaker B: Yeah.
Speaker A: We should empower them to make these decisions and we should want them practicing at the top of their degrees and spending the time thinking about things that really require their expertise.
Speaker B: Yeah.
Speaker A: So, for example, bedside patient monitoring and alarm adjustments. Why should it change dependent upon who the provider is at the bedside? Uh, why should it be a one size fits all solution?
Speaker B: Right.
Speaker A: If shouldn't, and we have the data to inform those, uh, parameters rather than making it subjective. And by the way, this is not cognition that people should use. They should be touching the patient, talking to the families, analyzing the data. And how patients are monitored most safely should be automated and should be data driven.
Speaker B: And I think too, it doesn't only have impact directly on the patient care that's being delivered at that moment in time. You have patient sat and all these other things that are factors. And how hospitals get paid, how patients view the provider or the care delivery team that they're interacting with. So it's such an important part of the equation in that acute care environment that often gets overlooked.
Speaker A: And on the other side of it, from the system side, we can't assume that systems are going to pay for 80 different solutions that are peripheral to a particular problem. And so what we have built is a single data set that is the foundation for hospital operations. So our data set gets you improved patient safety, improved data, informed care, uh, and a platform for clinical Engineering to see real time visibility into device performance. Is there an alarm going off on the bedside monitor on the ventilator? Historically, they've never gotten it. It's all from me calling and saying, hey, this ventilator doesn't work. Can you please come here? Now we give clinical engineering real time visibility into those alarms. They can show up and say, marcus, your ventilator is not working properly. There's an alarm. We're going to take it and give you a new one. We give it to nursing. Nursing can utilize this data to inform nursing staffing. And ultimately now with patient recovery, we're looking at mathematical modeling and data to understand when patients are ready, um, to either discharge from an icu, step down unit, we can right size bed utilization, accelerate patient throughput and ultimately optimize revenue and reduce patient exposure to harm by being in a bed that they don't need to be in.
Speaker B: So within that framework, what are the top two or three RRI metrics that you focus on or you're seeing your provider providers come to you or how systems come to you to really kind of acutely look at?
Speaker A: So I think there's two, right? The first is we focus, as I said, over and over again on patient safety and operations, right? So by using data to inform these decisions, we're making care safer. By making care safer and reducing harm, you're immediately getting patients through the system more quickly. And ultimately our patient recovery platform is going to allow systems to get patients out of ICUs quicker. And we know that every day you're in an ICU, two things happen that the public isn't aware about. Your IQ points are reduced and you're exposed to harm. So getting patients out of the ICU is very good for patients. And ultimately we're in a business. And what hospital systems want is they want optimal bed utilization and patient throughput. And our data stack now informs that. So we can help hospital systems get patients through much more quickly. And ah, that does two things. It reduces length of stay and increases revenue by allowing improved bed utilization.
Speaker B: Well, thanks for very clearly articulating that. And I think, um, when we think about the value prop, right, we're here at Vive and there are literally hundreds of companies that are talking about value props. Um, you have a very clear, articulate story. Um, you know, and as you know, you sit on the practitioner side, but you also sit on the business side selling into practitioners. So what have you seen and maybe what's some of your advice and guidance? And our audience is, you know, health systems Providers, entrepreneurs, executives. So we have an entire suite of people listed on both sides of the coin. What does that look like from your lens relative to a business trying to sell into a hospital system? And then for those executives at a hospital system buying this impertinent technology across their entire suite, what should they be looking at?
Speaker A: I think there's a couple of things. I think number one is we have to embrace the time that we're in. Right. The time that we're in is one where the utilization of data is unlike it's ever been before and we should maximize that potential. Right. So in doing that, I think there are fundamental things that we need to understand. We should not be asking the bedside providers to interpret additional data than they already have. They can't use what they have. So we need to make it the data actionable and we need to make it relevant rather than giving them additional data to interpret. I think the second thing is what we are really focused on is providing a single solution for these systems across different care paradigms. So our system is very well designed for single ICU care. It's exceptionally well designed for remote ICU monitoring because everything starts with full visibility of the ICU and we have different lenses. You can see the patient lens and the ICU on a unit level, on a hospital level, on a system level. Right. So as, as vendors, we need to provide solutions.
Speaker B: Uh, yeah.
Speaker A: That offer a lot of value across an entire care delivery spectrum.
Speaker B: Yeah.
Speaker A: Right. That's our responsibility. On the terms of the health systems they are now inundated with, everywhere they turn there's another solution. Right. And I think it's incumbent upon them to understand where the ROI sits and how their data is being utilized.
Speaker B: Yeah.
Speaker A: In order to really inform care. And as I said before, it's a business and there's a way to do this together. And I think that's what we're really excited about is how we have been able to take this data driven approach to really inform care across the spectrum. I think that's at the end of the day what we're all going to have to answer for. We can't have a million point solutions for one problem. It's not going to work.
Speaker B: Yeah. Uh, we agree. I mean at HC1 we built an agentic longitudinal platform, you know, really with the lens through the lab that I think ultimately, you know, if you take and you put piece together these longitudinal pieces, whether it sits within an ICU or a lab or another piece of the hospital and you're able to put them together effectively and generate that RRI use case. Um, I think, you know, there's unlimited potential. And it's all about, you know, accelerating patient care. Right. Reducing the burdens on the providers, um, and enabling, you know, all the practitioners to, to practice at top of their license. And for this, this to get better. I mean, really, that's, you know, it's, it's, it's incredible. Um, but it's, it's all about, you know, making the system better for everyone. Yeah.
Speaker A: And, you know, it's that old adage, never let a good crisis go to waste. And I think we're all seeing it. We're in the midst of a healthcare crisis. They're not growing new nurses, they're not growing new physicians. In fact, they're leaving at a rate unlike we've ever seen before.
Speaker B: Yeah.
Speaker A: And we're not getting more beds.
Speaker B: Right.
Speaker A: And as the care delivery environment shrinks and rural hospitals are closing, it's incumbent upon all of us to make sure that access to care remains the same and the quality of care, as you said, really should be elevated. And that's the goal here.
Speaker B: Absolutely. Great. Well, Dr. Klugman, thank you so much. Appreciate your insights today and, uh, continued success to you and the team. Will be in touch.
Speaker A: Thanks a lot.
Speaker B: All right, thank you.
Speaker A: Appreciate the time. Okay, thanks.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
The $19,000 Bill That Exposed Healthcare's Biggest Problem73 / 100
Healthcare AI Has to Prove Itself Now
AI Won't Save Healthcare If the Data Layer Is Broken
The $15 Billion Problem Hiding Inside Hospital Workflows
From Chart Review to Clinical Capacity Duke's AI Strategy
