SCCM Pod-471: Mitigating Diagnostic Delays and Errors With Emphasis on Sepsis

---

Diagnostic delays and errors are significant contributors to patient illness, injury, and death in the United States. According to the Society to Improve Diagnosis in Medicine, diagnostic errors impact nearly 12 million Americans every year, leading to prolonged hospital stays, increased nonreimbursed healthcare costs, and even more harm when combined with other medical errors.

Maureen Madden, DNP, RN, CPNP-AC, CCRN, FCCM, is joined by Mary Jo C. Grant, ACNP, PhD, FAAN, to discuss how to reduce diagnostic delays and errors, with an emphasis on sepsis. This podcast is funded by the Gordon and Betty Moore Foundation through a grant program administered by the Council of Medical Specialty Societies.

*If you are unable to play the podcast please click here to download the file.

Transcript:

This podcast is funded by the Gordon and Betty Moore Foundation through a grant program administered by the Council of Medical Specialty Societies.

Dr. Madden: Hello and welcome to the Society of Critical Care Medicine Podcast. I’m your host, Maureen Madden. Today I’ll be speaking with Dr. Mary Jo Grant, ACNP, PhD, FAAN. We will be talking about how to reduce diagnostic delays and errors with an emphasis on sepsis. Dr. Grant is a pediatric ICU nurse practitioner and pediatric cardiac ICU practitioner and a researcher at Primary Children’s Hospital in Salt Lake City. Welcome, Dr. Grant. Before we start, do you have any disclosures to report?

Dr. Grant: No, I have no disclosures.

Dr. Madden: Excellent. You’re going to focus on reducing diagnostic delays and errors, and we’ll talk about it some in the context of sepsis. Part of what had speared this conversation is CMS and the Agency for Healthcare Research and Quality estimate that the U.S. healthcare system spends about 38 billion annually on sepsis, with costs rising about 8% a year and most hospitals losing money on every case. Do you care to comment on that statistic?

Dr. Grant: First, I’d like to thank you and thank the Society for the opportunity to speak on diagnostic errors and delays today. I think that, to start out, the Surviving Sepsis recommendations that were created in 2021 mentioned applying diagnostic delays and errors to the guidelines. That’s really the starting point for doing this, this recognition within the Society that diagnostic delays and errors do lead to problems with sepsis and lead to further morbidity and mortality and, in relation to that, expense as well.

Dr. Madden: For the United States, just coming out of COVID, and we know that there were huge impacts on our healthcare system and the economics of that, initially first with so many things being shut down and items that were not emergent were pushed back. Now, we have so many things going on in terms of the nursing shortage and technology pieces, and that we have delays in terms of providing items in the pipeline. So, when we look at trying to improve the quality of our care, specifically when we’re looking at sepsis, and each hospital now has to pay for these items, we talked about that the amount is astronomical when you really look at it. But they really say at this point in time that national compliance with bundle standards is stuck at about 50%. Do you care to comment on that?

Dr. Grant: I think that there are a number of issues that prevent us from being able to apply the bundles in a real-time way; a lot of those reasons are related to diagnostic delays and errors. First, just to lay the groundwork for our discussion here, let me talk about some definitions of what I mean when I say diagnostic delays and errors. There are two primary definitions of diagnostic errors. The first is from the National Academy of Medicine; their definition is the failure to establish an accurate explanation of the patient’s health problem or, two, the failure to communicate that explanation to the patient and within the electronic health record.

In the diagnostic research community, the definition is a little bit different. It is defined as a missed opportunity to make a timely or correct diagnosis or take the next diagnostic action or step based on availability of evidence at the time, so, a little more time sensitive in what’s used in the research community. There is a difference between delays and errors. We’ll just use the term “errors” as it encompasses delays. But generally, diagnostic delays are secondary to systemic and coordination issues, where diagnostic errors are typically secondary to individual factors. We can go over some of those individual factors.

Dr. Madden: I love the fact that you brought up definitions because I think it always puts us all in the same understanding of what we’re speaking about. As a practicing clinician, and I know you are a practicing clinician, but you also have your research focus, the second definition really resonates with me as missed opportunities. I think a lot of that is what we see as practicing clinicians, the lack of recognition. When you say errors, I almost want to change it and to fit it into that capacity of missed opportunities rather than something that potentially has been deliberate and, I don’t know if that’s the right word, created that aspect. But I would love for you to talk about some of the other factors that fit within both errors and diagnostic delays.

Dr. Grant: Just to start with the incidence, we know that, from postmortem studies, about 34,000 ICU patients may die annually from a diagnostic error. We know from some other literature that as many as 18% of patients in a rapid response experience a diagnostic error. So, this is a real issue. The origins of diagnostic errors really go back to the ability to evaluate the electronic medical record. It is also a leading cause of claims-associated death and disability. Until a decade ago, much of what was known about diagnostic errors was learned through the evaluation of data from malpractice claims. In recent years, investigators have used different approaches, including autopsy studies, case reviews, surveys of patients and physicians, voluntary reporting systems, secondary reviews, diagnostic testing audits, and closed claims reviews. But each of these methods has limitations, and none are very well suited for establishing the presence of diagnostic errors in real time when used alone.

The Surviving Sepsis Campaign really focuses on an appropriate diagnosis for that patient, so the treatment can be done in a timely fashion. Diagnostic errors tend to lead us down a path that’s not appropriate for this patient that’s experiencing sepsis. That’s why we really need to focus on diagnostic errors and limit the cognitive biases that come along with clinicians’ evaluation of a septic patient at a specific point in time.

Dr. Madden: Very true. As you were speaking, I’m thinking about this, and no disrespect to you or to myself, but both of us have been practicing as clinicians for a long time and have been part of the transition and the implementation of technology where we’ve converted from all paper charting and documentation and have now moved into this electronic health record, which is supposed to have, as you said, really provided opportunities to improve our care and improve the timing of details related to diagnostics and then hopefully reduce the delays and the errors. I’m not so sure that anybody’s found the perfect system yet. I know, for the various EMRs that I’ve had the opportunity to work with, we’re still trying to improve the sensitivity and the timeliness of the documentation, so you can extrapolate those pieces so you get the trigger or it populates the alarm for screening somebody for suspicion of sepsis. What’s your views on that?

Dr. Grant: I think the electronic medical record is really not optimized for the communication of diagnostic reasoning or certainty. It gives an opportunity to present a diagnosis, but it does not facilitate an understanding of the clinician’s thinking about their diagnostic differential as well as their diagnostic certainty and that lack of transparency in the electronic medical record makes it both difficult to evaluate whether there was a diagnostic error or not but also creates a misunderstanding as things are cut and pasted from day to day through the electronic medical record of what exactly this clinician was thinking and how certain they are about the appropriateness of the diagnosis. It perpetuates an error through time and does a great disservice to our critically ill patients.

Dr. Madden: So many people who are far more intelligent than I am, sophisticated in developing this technology of the EMR, talk about it as the components and structure, that it’s like artificial intelligence, reliance upon the technology, that it’s going to be able to create an understanding of the factors that we’re looking for so that we have the optimal time to, first of all, quickly identify the sepsis cases, thereby minimizing false alerts in that the alarm fatigue is not there and the team then trusts what the flag is, but also the opportunity then to use all these bundles that we’ve created, that we know if we follow the steps, that it should help optimize care, the timeliness of it, and potentially also optimize outcomes.

I just don’t know at times if a fully automated clinical surveillance tool will be able to do that because, as you said, the emphasis on the clinician and their diagnostic reasoning and how they implement items into the EMR also creates that challenge or that disconnect where they’re not having the appropriate recognition. Thoughts about how we might be able to create that and improve the trigger tool or something? You and I both work with the pediatric population and we talked a little bit about that before we started this conversation, that our population is so diverse. How do we create a tool that would look at all the ages with the variability of their vital signs and cognitive, how that would be able to do that?

Dr. Grant: I think the sepsis tool needs to go beyond the trigger of sepsis and definition of sepsis and diagnosis of sepsis. But an EHR tool could be used for defining proper antimicrobials or proper dosing based on pharmacokinetic properties and based on the patient’s clinical characteristics, which will make them more likely or less likely to be susceptible to a certain infection, and based on the propensity of a certain organism within an institution, because we do know that there’s institutional variability on the presence of organisms. An EMR in terms of sepsis needs to go beyond just the trigger alert and beyond helping the clinician to recognize sepsis, but more to give assistance to the clinicians when it comes to deciding the various treatment modalities on sepsis and really implementing the Surviving Sepsis recommendations that were updated in 2021. That’s where I see some real utility in the EMR and where the focus has not yet been created.

I know at Boston Children’s, they’re using a checklist for patients as they round on them. The checklist has embedded some triggers to help them to decide appropriateness of different therapies. This isn’t necessarily focused on sepsis, but it’s on the daily checklist, and it gives some real-time feedback to what’s going on with the patients and what’s in their electronic medical record. I think that using the Surviving Sepsis recommendations and embedding those recommendations within the EMR and giving real-time feedback to the clinician, the clinical characteristics of the patients is the future of using the EMR to help with our ability for patients to survive sepsis.

Dr. Madden: You’ve raised a couple points that I’d like to ask about. I’m not aware of what Boston Children’s is doing. Part of the question is whether their checklist is embedded within their EMR or is it paper? Because I’ve done a lot of work and implementation in terms of interdisciplinary rounding tools with the context behind that to try and ensure that AHI and nursing-sensitive factors are all addressed in a timely fashion, same concept. Do they need that Foley catheter? Do they need that central line, antibiotic stewardship, as you said? My experience and the research I have, it’s all done on paper. It’s intriguing if Boston has implemented electronically and then it talks to the rest of the data to correlate it, as you said, to drive care. Do you know what they’re doing?

Dr. Grant: I know, just from reading a recent manuscript about this, so I’m not best to speak about what they’re doing, but they are using an electronic trigger tool, so it’s not paper based. But there are a number of causes of diagnostic errors, and one of the causes, when you think about the patient’s diagnostic journey, it starts with clinical information gathering. Then there’s a synthesis of key information, and then there’s decision-making and communication of that diagnosis. Those three factors on a patient’s diagnostic journey are affected by cognitive failures, the admission of critical information on the physical examination, the admission of critical information in the EMR, problems with lab values, problems with interpretation of lab values, and problems in terms of cognitive failures, anchoring with implicit bias and with premature closure.

The place where an EMR can be used with eliminating these cognitive failures could be in implementing checklists that will look for unknown information to the clinicians at the time to help them to make the diagnosis because, as you know, the diagnosis of sepsis is not always clear. Patients are very complex, particularly in the pediatric ICU. They tend to have multiple comorbidities. It’s often difficult to hone down which of these comorbidities is having the greatest impact in leading to this patient becoming septic. I think if there’s premature closure of the diagnosis, then you can limit the ability to look at maybe one of these other comorbidities, which may be leading to the patient’s clinical deterioration.

The other causes of diagnostic errors are systems failures, which are poor coordination of care or inadequate mechanisms to convey critical results. Again, the EMR may be able to have an impact there. Then there’s just that we work in a very chaotic clinical work environment, and this chaos leads to the ability to think through a problem and to use cognitive processes to avoid distractions and interruptions because an ICU is a busy place, and those distractions and interruptions prevent us from being able to really think through a diagnosis and make sure that we have eliminated all the secondary potential diagnoses that are present in this patient.

Dr. Madden: From that, we keep coming back to the human factor. We’re all fallible, we know that, as you talk about premature closure or not looking at potentially the differential diagnoses in their entirety, we can try and minimize human error. But as you said also, the coordination of the systems and the technology and the personnel always needs to be improved. We’re constantly improving it. One of the challenges that I’ve seen with the EMR also is that different facilities have different EMRs and they don’t necessarily talk to each other, or there’s a reliance upon opening the ability to see other charts, even if it’s in an EMR that can pull in outside encounters. Because the challenge, as you talk, I see all these patients that I’ve cared for and everything, that the value of having all that historical data, including, as you said, they have comorbidities and if they have had past events of sepsis or even of infections and what those were, and looking at antibiotics and the sensitivities, not just from within the institution, but to the specific patient.

I think there’s so much work still to be done and I am not the person, I don’t have that skill set, I don’t have that knowledge of improving the EMR. We can give them the feedback but comment on the concept that people still aren’t following the bundle. They say there’s 50% compliance on the sepsis bundle. We’ve seen it in the research data and the publications that the more components of the bundle that are followed, the more optimal care and outcomes can occur. How can we try and engage people to follow the bundle?

Dr. Grant: I think a lot of that comes down to individual clinician characteristics. I know in medical school, they are starting to teach about cognitive biases and focus on critical thinking. If there’s a lack of critical thinking, then that’s where the bundles will help us to move forward in the application of the bundle to a specific patient. But the diagnosis has to be made first before the bundle is going to be initiated. This lack of critical thinking; problems with handoff causing a degradation of the information; problems with exhaustion and fatigue from the caregivers; varied training, background, and experience; and ego of the clinicians and their willingness to follow the bundle and apply it to a certain patient population. And there is a difference in work ethic and responsibility.

I think that not every clinician is the same, and we need to think about these individual clinician characteristics, which may lead to the limited application of the bundle. It’s important that we think through that there are clinician failures. There can be failures in history-taking, failures in assessment and physical exam factors, which prevent them from coming up with an appropriate diagnosis, which would lead to implementation of the sepsis bundle. So, this is a big issue and a big problem. There’s been a lot of effort made to further the clinician’s implementation of the bundle earlier on, but not a lot of movement has been made on this front.

Dr. Madden: Right. So as this podcast is from the Society of Critical Care Medicine and the organization’s membership is multidisciplinary, that’s a huge focus. You commented on physicians about cognitive bias, and I think it really is, since we’re multidisciplinary at the bedside, that it involves all the disciplines, everybody needs to have that education and training. It’s the same concept when we talk about the sepsis bundle. In-training individuals, where are they receiving the exposure and the education about these bundles? Not just sepsis, but we have them in place for multiple other entities. They’ve proven, when you follow the bundle, whether it’s the implementation or the maintenance of it, that we have improved care and decreased infections and improved outcomes. I think we really need to make sure, when we’re having these conversations, that every person who’s involved in the care of a patient is included in the conversation and that they are given the responsibility to have the same critical thinking skills in place and to speak up. Maybe as a multidisciplinary group practicing together, we can try and take away some of the holes in the Swiss cheese model.

Dr. Grant: Yes. Sarah Redman and Amelia Barwise and their colleagues have a recent publication where they surveyed acute care stakeholders about different factors that contributed to diagnostic errors. They found that physicians perceive that cognitive failures happen more frequently than those that are in other roles, which I thought was an interesting finding. They surveyed nurse practitioners, PAs, and a small subset of nurses and found that they didn’t recognize cognitive failures as a cause of diagnostic errors.

Dr. Madden: I think that research has value and it’s not spoken about enough because, when we talk about the concept of also diagnostic errors and diagnostic delays because of cognitive thinking, we’ve made progress with the recognition from the IOM and in terms of lives that were lost in regard to medical errors and we focused on that. Then it brought in the advent of checklists, Atul Gawande had brought it in. In that process, the electronic technology for documentation has come in and the coordination of it. But I think communication is a key element and everybody needs to recognize that they have the ability and the right to speak up when there may be a discrepancy. The fact that physicians feel that cognitively, that’s where they’re making the diagnostic error, somehow, I want to see that improve. I don’t know what you think about that.

Dr. Grant: I agree that this is a care team that manages these patients, and I think it’s important to have a flattened hierarchy and the ability for anyone to speak up when they are worried about whether this is an appropriate or inappropriate diagnosis, and whether there has been incorrect elimination of multiple other reasons why this diagnosis of sepsis happened. Sarah Redmond and Amelia Barwise published extensively on diagnostic errors and have done some mixed-method studies looking at the presence and the characteristics in diagnostic error causes. I would turn the listeners to her literature.

Dr. Madden: Thank you for that too. The focus of this podcast has really been looking at diagnostic delays and errors with the emphasis on sepsis, partly because of the enormous expense that’s associated with it within the facilities, within our entire healthcare system. I think there’s a lot of awareness about our healthcare system and things that need to improve. We could go on for a long time, but I think our time is coming to a close. I just wanted to ask you if you had any additional points that you would like to make before we close this out.

Dr. Grant: I think that one of the key points that I would like to make is an emphasis on creating a diagnostic time-out after a patient has had something like a rapid response or an unplanned ICU admission. This diagnostic time-out gives an opportunity to look through the differential diagnosis. This diagnostic time-out should be documented in the EMR so it’s clear to the next clinicians that come forward what exactly the clinicians at the time were thinking. I think it’s important that everyone on the care team is part of this discussion and decision-making when you do the diagnostic time-out. That would be one of my recommendations that I would make that would help us as we move forward in applying the Surviving Sepsis recommendations.

Dr. Madden: I think that’s a great point. At this point, I just want to thank you, Dr. Mary Jo Grant, for your time and the opportunity to speak on this. I’ve really enjoyed it. This concludes another edition of the Society of Critical Care Medicine Podcast. For the Society of Critical Care Medicine Podcast, I’m Maureen Madden.

This podcast is funded by the Gordon and Betty Moore Foundation through a grant program administered by the Council of Medical Specialty Societies.

Maureen a Madden, DNP, RN, CPNP-AC, CCRN, FCCM, is a professor of pediatrics at Rutgers Robert Wood Johnson Medical School and a pediatric critical care nurse practitioner in the pediatric intensive care unit at Bristol-Myers Squibb Children’s Hospital in New Brunswick, New Jersey, USA.

Join or renew your membership with SCCM, the only multiprofessional society dedicated exclusively to the advancement of critical care. Contact a customer service representative at +1 847 827-6888 or visit sccm.org/membership for more information. The SCCM Podcast is the copyrighted material of the Society of Critical Care Medicine. All rights are reserved. Find more episodes at sccm.org/podcast.

This podcast is for educational purposes only. The material presented is intended to represent an approach, view, statement, or opinion of the presenter that may be helpful to others. The views and opinions expressed herein are those of the presenters and do not necessarily reflect the opinions or views of SCCM. SCCM does not recommend or endorse any specific test, physician, product, procedure, opinion, or other information that may be mentioned.

Some episodes of the SCCM Podcast include a transcript of the episode’s audio. Although the transcription is largely accurate, in some cases it is incomplete or inaccurate due to inaudible passages or transcription errors and should not be treated as an authoritative record.

Disclaimer