Internal Medicine News

In a fiercely competitive healthcare landscape, hospitals are pulling out all the stops to lure top nursing talent through their doors. From offering eye-popping sign-on bonuses to implementing flexible work schedules, today’s hospitals are transforming the nurse hiring process into a high-stakes game where only the most innovative strategies win.

As the nursing shortage intensifies, the creative recruitment approach isn’t just about the perks — it’s becoming an essential tool in the race to build a skilled nursing workforce.

Nursing vacancies are as high as 17% — more than double prepandemic levels — and hospitals scrambling to fill them need to do more than raise salaries and bolster benefits packages to entice nurses.

“I am very thankful when I hear of creative ideas that nurse administrators come up with to try to get their ultimate goal, which is enough qualified nurses to take care of patients,” said Linda Plank, dean of the Louise Herrington School of Nursing at Baylor University in Dallas, Texas.
 

Signing Bonuses, Tuition Reimbursement, and Self-Scheduling, Please

Signing bonuses were among the top perks offered to healthcare workers, with almost 18% of job openings advertising the incentive for new nurse hires; the average signing bonus for registered nurses (RNs) topped $11,000. In 2023, California-based Palomar Health made headlines when it offered eligible RNs a $100,000 signing bonus paid over a 3-year period.

“We are seeing a variety of incentives, like sign-on bonuses, that can be effective at getting the attention of potential new hires,” said Deborah Trautman, PhD, RN, FAAN, president, and CEO of the American Association of Colleges of Nursing. “With the growing competition for registered nurses, especially those prepared in baccalaureate programs, employers should consider what’s most important to nurses entering the field.”

Hospitals have also invested in benefits ranging from tuition reimbursement, student loan forgiveness, and professional development opportunities to expanded parental leave and onsite childcare. Flexible scheduling is a sought-after perk that benefits both new and experienced RNs and could also help with recruitment.

“In the past, [hospitals said], ‘our shift starts at 7 and ends at 7,’ ” Ms. Plank said. “Now, hospitals are a little bit more flexible ... and being open to flexible shifts has merit. If we’re willing to look at things differently, it could get more people involved in patient care.”

An American Nurses Foundation report found that nurses preferred variable and flexible shift lengths, flexible start times, and self-scheduling options over set schedules. In fact, 45% of nurses who left clinical practice would consider returning to work if hospitals switched to a self-scheduling model.

The Cleveland Clinic in Ohio introduced staggered shifts that start at nontraditional times, including 11 AM to 11 PM, flexible shift lengths, and split RN positions that allow clinical care nurses to divide their time between different departments. Last year, Hackensack Meridian Health in New Jersey also piloted a self-scheduling program for its nursing staff.
 

Hiring Outside the United States

Despite the prevalence of signing bonuses and premium perks, some hospitals still struggled to fill open positions with nurses recruited from outside the United States. Data from The Kaiser Family Foundation show that 32% of hospitals hired foreign-educated RNs in 2022 — more than double the number hired in 2010.

Jennifer Mensik Kennedy, PhD, MBA, RN, NEA-BC, FAAN, president of the American Nurses Association (ANA), is concerned about that trend. The ANA supports the International Council of Nurses and their call for “stronger codes for ethical recruitment of nurses” because international nurse recruitment practices can negatively affect the quality of healthcare in countries that are depleted of nurses.

“Recruiting international nurses as a key strategy for building core staffing is not sustainable in the long term,” Ms. Kennedy said in an interview. “We need to redirect our focus on how to retain staff through fostering healthy work environments and addressing antiquated payment models.”
 

Reinforcing Retention

Recruiting nurses is just one element of addressing the nursing shortage. Prioritizing job satisfaction is essential to retaining nursing staff. Currently, 33% of nurses who enter the profession quit within the first 2 years.

A growing number of hospitals have implemented programs focused on increasing retention. Lifepoint Health, a national network of 60-plus acute care hospitals, launched a Nurse Residency Program in 2023. The 12-month program, which offers training and mentorship to help recent nursing school graduates, has recruited 750 new nurses and helped them transition to clinical practice.

The Nurse Residency Program has been so successful that the hospital system plans to introduce a 2-year fellowship program in the fall of 2024 that supports the professional development of nurses who want to specialize in areas like acute care, obstetrics, or the intensive care unit.

“We are more focused than ever on increasing partnerships and alignment with our local nursing programs, expanding clinical education opportunities for nursing students, owning and driving a nursing culture, and creating an environment where employees want to work,” said Michelle Watson, MSN, RN, CENP, chief nurse executive and senior vice president of clinical operations at Lifepoint Health.

Ms. Watson also credits their facilities’ chief nursing officers as being “highly engaged and visible leaders” who spend time with RNs to learn about their career aspirations and help them understand how the organization can support their desires for ongoing professional development.

The 2022 Nurse Staffing Task Force, a collaborative initiative by the ANA and other prominent national nursing and healthcare organizations, has developed and widely disseminated a set of recommendations for hospitals focusing on investing in nurse staffing, safe and supportive work environments, and competitive wages.

In addition, the Reimagining Nursing Initiative, started by the American Nurses Foundation, is striving to help nurses feel valued and compensated by creating pilot programs that can help modernize nurses’ reimbursement structure so that they can direct bill.

In the end, attracting and retaining top nursing talent is about more than filling positions — it’s about building a healthcare system where nurses thrive. “In the United States and abroad, we owe it to nurses and the communities they serve to have sustainable and appropriate solutions to staffing and work environment challenges,” said Ms. Kennedy.

A version of this article appeared on Medscape.com.

In a fiercely competitive healthcare landscape, hospitals are pulling out all the stops to lure top nursing talent through their doors. From offering eye-popping sign-on bonuses to implementing flexible work schedules, today’s hospitals are transforming the nurse hiring process into a high-stakes game where only the most innovative strategies win.

As the nursing shortage intensifies, the creative recruitment approach isn’t just about the perks — it’s becoming an essential tool in the race to build a skilled nursing workforce.

Nursing vacancies are as high as 17% — more than double prepandemic levels — and hospitals scrambling to fill them need to do more than raise salaries and bolster benefits packages to entice nurses.

“I am very thankful when I hear of creative ideas that nurse administrators come up with to try to get their ultimate goal, which is enough qualified nurses to take care of patients,” said Linda Plank, dean of the Louise Herrington School of Nursing at Baylor University in Dallas, Texas.
 

Signing Bonuses, Tuition Reimbursement, and Self-Scheduling, Please

Signing bonuses were among the top perks offered to healthcare workers, with almost 18% of job openings advertising the incentive for new nurse hires; the average signing bonus for registered nurses (RNs) topped $11,000. In 2023, California-based Palomar Health made headlines when it offered eligible RNs a $100,000 signing bonus paid over a 3-year period.

“We are seeing a variety of incentives, like sign-on bonuses, that can be effective at getting the attention of potential new hires,” said Deborah Trautman, PhD, RN, FAAN, president, and CEO of the American Association of Colleges of Nursing. “With the growing competition for registered nurses, especially those prepared in baccalaureate programs, employers should consider what’s most important to nurses entering the field.”

Hospitals have also invested in benefits ranging from tuition reimbursement, student loan forgiveness, and professional development opportunities to expanded parental leave and onsite childcare. Flexible scheduling is a sought-after perk that benefits both new and experienced RNs and could also help with recruitment.

“In the past, [hospitals said], ‘our shift starts at 7 and ends at 7,’ ” Ms. Plank said. “Now, hospitals are a little bit more flexible ... and being open to flexible shifts has merit. If we’re willing to look at things differently, it could get more people involved in patient care.”

An American Nurses Foundation report found that nurses preferred variable and flexible shift lengths, flexible start times, and self-scheduling options over set schedules. In fact, 45% of nurses who left clinical practice would consider returning to work if hospitals switched to a self-scheduling model.

The Cleveland Clinic in Ohio introduced staggered shifts that start at nontraditional times, including 11 AM to 11 PM, flexible shift lengths, and split RN positions that allow clinical care nurses to divide their time between different departments. Last year, Hackensack Meridian Health in New Jersey also piloted a self-scheduling program for its nursing staff.
 

Hiring Outside the United States

Despite the prevalence of signing bonuses and premium perks, some hospitals still struggled to fill open positions with nurses recruited from outside the United States. Data from The Kaiser Family Foundation show that 32% of hospitals hired foreign-educated RNs in 2022 — more than double the number hired in 2010.

Jennifer Mensik Kennedy, PhD, MBA, RN, NEA-BC, FAAN, president of the American Nurses Association (ANA), is concerned about that trend. The ANA supports the International Council of Nurses and their call for “stronger codes for ethical recruitment of nurses” because international nurse recruitment practices can negatively affect the quality of healthcare in countries that are depleted of nurses.

“Recruiting international nurses as a key strategy for building core staffing is not sustainable in the long term,” Ms. Kennedy said in an interview. “We need to redirect our focus on how to retain staff through fostering healthy work environments and addressing antiquated payment models.”
 

Reinforcing Retention

Recruiting nurses is just one element of addressing the nursing shortage. Prioritizing job satisfaction is essential to retaining nursing staff. Currently, 33% of nurses who enter the profession quit within the first 2 years.

A growing number of hospitals have implemented programs focused on increasing retention. Lifepoint Health, a national network of 60-plus acute care hospitals, launched a Nurse Residency Program in 2023. The 12-month program, which offers training and mentorship to help recent nursing school graduates, has recruited 750 new nurses and helped them transition to clinical practice.

The Nurse Residency Program has been so successful that the hospital system plans to introduce a 2-year fellowship program in the fall of 2024 that supports the professional development of nurses who want to specialize in areas like acute care, obstetrics, or the intensive care unit.

“We are more focused than ever on increasing partnerships and alignment with our local nursing programs, expanding clinical education opportunities for nursing students, owning and driving a nursing culture, and creating an environment where employees want to work,” said Michelle Watson, MSN, RN, CENP, chief nurse executive and senior vice president of clinical operations at Lifepoint Health.

Ms. Watson also credits their facilities’ chief nursing officers as being “highly engaged and visible leaders” who spend time with RNs to learn about their career aspirations and help them understand how the organization can support their desires for ongoing professional development.

The 2022 Nurse Staffing Task Force, a collaborative initiative by the ANA and other prominent national nursing and healthcare organizations, has developed and widely disseminated a set of recommendations for hospitals focusing on investing in nurse staffing, safe and supportive work environments, and competitive wages.

In addition, the Reimagining Nursing Initiative, started by the American Nurses Foundation, is striving to help nurses feel valued and compensated by creating pilot programs that can help modernize nurses’ reimbursement structure so that they can direct bill.

In the end, attracting and retaining top nursing talent is about more than filling positions — it’s about building a healthcare system where nurses thrive. “In the United States and abroad, we owe it to nurses and the communities they serve to have sustainable and appropriate solutions to staffing and work environment challenges,” said Ms. Kennedy.

A version of this article appeared on Medscape.com.

In a fiercely competitive healthcare landscape, hospitals are pulling out all the stops to lure top nursing talent through their doors. From offering eye-popping sign-on bonuses to implementing flexible work schedules, today’s hospitals are transforming the nurse hiring process into a high-stakes game where only the most innovative strategies win.

As the nursing shortage intensifies, the creative recruitment approach isn’t just about the perks — it’s becoming an essential tool in the race to build a skilled nursing workforce.

Nursing vacancies are as high as 17% — more than double prepandemic levels — and hospitals scrambling to fill them need to do more than raise salaries and bolster benefits packages to entice nurses.

“I am very thankful when I hear of creative ideas that nurse administrators come up with to try to get their ultimate goal, which is enough qualified nurses to take care of patients,” said Linda Plank, dean of the Louise Herrington School of Nursing at Baylor University in Dallas, Texas.
 

Signing Bonuses, Tuition Reimbursement, and Self-Scheduling, Please

Signing bonuses were among the top perks offered to healthcare workers, with almost 18% of job openings advertising the incentive for new nurse hires; the average signing bonus for registered nurses (RNs) topped $11,000. In 2023, California-based Palomar Health made headlines when it offered eligible RNs a $100,000 signing bonus paid over a 3-year period.

“We are seeing a variety of incentives, like sign-on bonuses, that can be effective at getting the attention of potential new hires,” said Deborah Trautman, PhD, RN, FAAN, president, and CEO of the American Association of Colleges of Nursing. “With the growing competition for registered nurses, especially those prepared in baccalaureate programs, employers should consider what’s most important to nurses entering the field.”

Hospitals have also invested in benefits ranging from tuition reimbursement, student loan forgiveness, and professional development opportunities to expanded parental leave and onsite childcare. Flexible scheduling is a sought-after perk that benefits both new and experienced RNs and could also help with recruitment.

“In the past, [hospitals said], ‘our shift starts at 7 and ends at 7,’ ” Ms. Plank said. “Now, hospitals are a little bit more flexible ... and being open to flexible shifts has merit. If we’re willing to look at things differently, it could get more people involved in patient care.”

An American Nurses Foundation report found that nurses preferred variable and flexible shift lengths, flexible start times, and self-scheduling options over set schedules. In fact, 45% of nurses who left clinical practice would consider returning to work if hospitals switched to a self-scheduling model.

The Cleveland Clinic in Ohio introduced staggered shifts that start at nontraditional times, including 11 AM to 11 PM, flexible shift lengths, and split RN positions that allow clinical care nurses to divide their time between different departments. Last year, Hackensack Meridian Health in New Jersey also piloted a self-scheduling program for its nursing staff.
 

Hiring Outside the United States

Despite the prevalence of signing bonuses and premium perks, some hospitals still struggled to fill open positions with nurses recruited from outside the United States. Data from The Kaiser Family Foundation show that 32% of hospitals hired foreign-educated RNs in 2022 — more than double the number hired in 2010.

Jennifer Mensik Kennedy, PhD, MBA, RN, NEA-BC, FAAN, president of the American Nurses Association (ANA), is concerned about that trend. The ANA supports the International Council of Nurses and their call for “stronger codes for ethical recruitment of nurses” because international nurse recruitment practices can negatively affect the quality of healthcare in countries that are depleted of nurses.

“Recruiting international nurses as a key strategy for building core staffing is not sustainable in the long term,” Ms. Kennedy said in an interview. “We need to redirect our focus on how to retain staff through fostering healthy work environments and addressing antiquated payment models.”
 

Reinforcing Retention

Recruiting nurses is just one element of addressing the nursing shortage. Prioritizing job satisfaction is essential to retaining nursing staff. Currently, 33% of nurses who enter the profession quit within the first 2 years.

A growing number of hospitals have implemented programs focused on increasing retention. Lifepoint Health, a national network of 60-plus acute care hospitals, launched a Nurse Residency Program in 2023. The 12-month program, which offers training and mentorship to help recent nursing school graduates, has recruited 750 new nurses and helped them transition to clinical practice.

The Nurse Residency Program has been so successful that the hospital system plans to introduce a 2-year fellowship program in the fall of 2024 that supports the professional development of nurses who want to specialize in areas like acute care, obstetrics, or the intensive care unit.

“We are more focused than ever on increasing partnerships and alignment with our local nursing programs, expanding clinical education opportunities for nursing students, owning and driving a nursing culture, and creating an environment where employees want to work,” said Michelle Watson, MSN, RN, CENP, chief nurse executive and senior vice president of clinical operations at Lifepoint Health.

Ms. Watson also credits their facilities’ chief nursing officers as being “highly engaged and visible leaders” who spend time with RNs to learn about their career aspirations and help them understand how the organization can support their desires for ongoing professional development.

The 2022 Nurse Staffing Task Force, a collaborative initiative by the ANA and other prominent national nursing and healthcare organizations, has developed and widely disseminated a set of recommendations for hospitals focusing on investing in nurse staffing, safe and supportive work environments, and competitive wages.

In addition, the Reimagining Nursing Initiative, started by the American Nurses Foundation, is striving to help nurses feel valued and compensated by creating pilot programs that can help modernize nurses’ reimbursement structure so that they can direct bill.

In the end, attracting and retaining top nursing talent is about more than filling positions — it’s about building a healthcare system where nurses thrive. “In the United States and abroad, we owe it to nurses and the communities they serve to have sustainable and appropriate solutions to staffing and work environment challenges,” said Ms. Kennedy.

A version of this article appeared on Medscape.com.

This discussion was recorded on August 2, 2024. This transcript has been edited for clarity. 

Robert D. Glatter, MD: Today, we’ll be discussing the results of a new study published in The Journal of Emergency Medicine, looking at the incidence of delayed intracranial hemorrhage among older patients taking preinjury anticoagulants who present to the emergency department (ED) with blunt head trauma.

Joining me today is the lead author of the study, Dr. Richard Shih, professor of emergency medicine at Florida Atlantic University. Also joining me is Dr. Christina Shenvi, associate professor of emergency medicine at the University of North Carolina (UNC) Chapel Hill, with fellowship training in geriatric emergency medicine. 

Welcome to both of you.

Richard D. Shih, MD: Thanks, Rob. 

Christina L. Shenvi, MD, PhD, MBA: Thanks. Pleasure to be here. 
 

ICH Study Methodology

Dr. Glatter: It’s a pleasure to have you. Rich, this is a great study and targeted toward a population we see daily in the emergency department. I want you to describe your methodology, patient selection, and how you went about organizing your study to look at this important finding of delayed intracranial hemorrhage, especially in those on anticoagulants.

Dr. Shih: This all started for our research team when we first read the 2012 Annals of Emergency Medicine paper. The first author was Vincenzo Menditto, and he looked at a group of patients that had minor head injury, were anticoagulated, and had negative initial head CTs. 

There were about 100 patients, of which about 10 of them did not consent, but they hospitalized all these patients. These were anticoagulated, negative-first head CTs. They hospitalized the patients and then did a routine second CT at about 24 hours. They also followed them for a week, and it turned out a little over 7% of them had delayed head CT. 

We were wondering how many delayed intracranial hemorrhages we had missed because current practice for us was that, if patients had a good physical exam, their head CT was normal, and everything looked good, we would send them home.

Because of that, a number of people across the country wanted to verify those findings from the Menditto study. We tried to design a good study to answer that question. We happen to have a very large geriatric population in Florida, and our ED census is very high for age over 65, at nearly 60%. 

There are two Level I trauma centers in Palm Beach County. We included a second multicenter hospital, and we prospectively enrolled patients. We know the current state of practice is not to routinely do second CTs, so we followed these patients over time and followed their medical records to try to identify delayed bleeding. That’s how we set up our methodology.
 

Is It Safe to Discharge Patients With Trauma After 24 Hours?

Dr. Glatter: For the bulk of these patients with negative head CTs, it’s been my practice that when they’re stable and they look fine and there’s no other apparent, distracting painful trauma, injuries and so forth, they’re safe to discharge. 

The secondary outcome in your study is interesting: the need for neurosurgical intervention in terms of those with delayed intracranial hemorrhage.

Dr. Shih: I do believe that it’s certainly not the problem that Menditto described, which is 7%. There are two other prospective studies that have looked at this issue with delayed bleeding on anticoagulants. Both of these also showed a relatively low rate of delayed bleeding, which is between like 0.2% and 1.0%. In our study, it was 0.4%. 

The difference in the studies is that Menditto and colleagues routinely did 24-hour head CTs. They admitted everybody. For these other studies, routine head CT was not part of it. My bet is that there is a rate of delayed bleeding somewhere in between that seen in the Menditto study and that in all the other studies.

However, talking about significant intracranial hemorrhage, ones that perhaps need neurosurgery, I believe most of them are not significant. There’s some number that do occur, but the vast majority of those probably don’t need neurosurgery. We had 14 delayed bleeds out of 6000 patients with head trauma. One of them ended up requiring neurosurgery, so the answer is not zero, but I don’t think it’s 7% either. 

Dr. Glatter: Dr. Shenvi, I want to bring you into the conversation to talk about your experience at UNC, and how you run things in terms of older patients with blunt head trauma on preinjury anticoagulants.

Dr. Shenvi: Thanks, Rob. I remember when this paper came out showing this 7% rate of delayed bleeding and the question was, “Should we be admitting all these people?” Partly just from an overwhelming need for capacity that that would bring, it just wasn’t practical to say, “We’re going to admit every patient with a negative head CT to the hospital and rescan them.” That would be hundreds or thousands of patients each year in any given facility. 

The other thing is that delayed bleeds don’t always happen just in the first 24 hours. It’s not even a matter of bringing patients into observation for 24 hours, watching them, and rescanning them if they have symptoms. It can occur several days out. That never, in almost any institution that I know of, became standard practice. 

The way that it did change my care was to give good return precautions to patients, to make sure they have somebody with them to say, “Hey, sometimes you can have bleeding several days out after a fall, even though your CT scan here today looks perfect,” and to alert them that if they start having severe headaches, vomiting, or other symptoms of intracranial hemorrhage, that they should come back. 

I don’t think it ever became standard practice, and for good reason, because that was one study. The subsequent studies that Richard mentioned, pretty quickly on the heels of that initial one, showed a much lower rate of delayed ICH with the caveats that the methodology was different. 
 

Shift in Anticoagulants

Dr. Shenvi: One other big change from that original study, and now to Richard’s study, is the shift in anticoagulants. Back in the initial study you mentioned, it was all warfarin. We know from other studies looking at warfarin vs the direct oral anticoagulants (DOACs) that DOACs have lower rates of ICH after a head injury, lower rates of need for neurosurgical intervention, and lower rates of discharge to a skilled nursing facility after an intracranial hemorrhage.

Across the board, we know that the DOACs tend to do better. It’s difficult to compare newer studies because it’s a different medication. It did inform my practice to have an awareness of delayed intracranial hemorrhage so that I warn patients more proactively. 

Dr. Glatter: I haven’t seen a patient on warfarin in years. I don’t know if either of you have, but it’s all DOACs now unless there’s some other reason. That shift is quite apparent.

Dr. Shih: The problem with looking at delayed bleeding for DOACs vs warfarin is the numbers were so low. I think we had 13 people, and seven were in the no-anticoagulant group. The numbers are even lower, so it’s hard to say. 

I just wanted to comment on something that Dr. Shenvi said, and I pretty much agree with everything that she said. Anticoagulants and warfarin, and that Menditto study, have a carryover effect. People group DOACs with warfarin similarly. When a patient is brought in, the first thing they talk about with head trauma is, “Oh, they’re on an anticoagulant” or “They’re not on an anticoagulant.” It’s so ingrained.

I believe that, in emergency medicine, we’re pressed for space and time and we’re not as affected by that 24-hour observation. Maybe many of our surgeons will automatically admit those patients. 

I haven’t seen a guideline from the United States, but there are two international guidelines. One is from Austria from 2019, and one is from Scandinavia. Both recommended 24-hour observation if you’re on an anticoagulant.

There is a bit of controversy left over with that. Hopefully, as more and more of information, like in our study, comes out, people will be a little bit more clear about it. I don’t think there’s a need to routinely admit them. 

I do want to mention that the Menditto study had such a massive impact on everybody. They pointed out one subgroup (and it’s such a small number of patients). They had seven cases of delayed bleeding; four or five of them were within that 24 hours, and a couple were diagnosed later over the next couple days.

Of those seven people, four of them had international normalized ratios (INRs) greater than 3. Of those four patients, I’ve heard people talk about this and recommend, “Okay, that’s the subgroup I would admit.” There’s a toss-up with what to do with DOAC because it’s very hard to tell whether there’s an issue, whether there are problems with their dosing, and whatever. 

We actually recently looked at that. We have a much larger sample than four: close to 300 patients who were on warfarin. We looked at patients who had INRs below 3 and above 3, and we didn’t show a difference. We still don’t believe that warfarin is a big issue with delayed bleeding.
 

Should We Be Asking: ‘Are They on Blood Thinners?’

Dr. Shenvi: One of the interesting trends related to warfarin and the DOACs vs no anticoagulant is that as you mentioned, Dr Shih, the first question out of people’s mouths or the first piece of information emergency medical services gives you when they come in with a patient who’s had a head injury is, “Are they on blood thinners or not?”

Yet, the paradigm is shifting to say it’s not actually the blood thinners themselves that are giving older patients the higher risk for bleeding; it’s age and other comorbidities.

Certainly, if you’re on an anticoagulant and you start to bleed, your prognosis is much worse because the bleeding doesn’t stop. In terms of who has a bleeding event, there’s much less impact of anticoagulation than we used to think. That, in part, may be due to the change from warfarin to other medications.

Some of the experts I’ve talked to who have done the research on this have said, “Well, actually, warfarin was more of a marker for being much older and more frail, because it was primarily prescribed to older patients who have significant heart disease, atrial fibrillation, and so on.” It was more a marker for somebody who is at risk for an intracranial hemorrhage. There are many changes that have happened in the past 10 years with medications and also our understanding. 
 

Challenges in Patient Follow-up

Dr. Glatter: That’s a great point. One thing, Rich, I want to ask you about is in terms of your proxy outcome assessment. When you use that at 14 and 60 days with telephone follow-up and then chart review at 60 and 90 days (because, obviously, everyone can’t get another head CT or it’s difficult to follow patients up), did you find that worked out well in your prospective cohort study, in terms of using that as a proxy, so to speak? 

Dr. Shih: I would say to a certain extent. Unfortunately, we don’t have access to the patients to come back to follow up all of them, and there was obviously a large number of patients in our study. 

The next best thing was that we had dedicated research assistants calling all of the patients at 14 days and 60 days. I’ve certainly read research studies where, when they call them, they get 80%-90% follow-up, but we did not achieve that.

I don’t know if people are more inundated with spam phone calls now, or the older people are just afraid of picking up their phone sometimes with all the scams and so forth. I totally understand, but in all honesty, we only had about a 30%-35% follow-up using that follow-up pathway. 

Then the proxy pathway was to look at their charts at 60 and 90 days. Also, we looked at the Florida death registry, which is pretty good, and then finally, we had both Level I trauma centers in the county that we were in participating. It’s standard practice that if you have an intracranial hemorrhage at a non–Level I trauma center, you would be transferred to a Level I trauma center. That’s the protocol. I know that’s not followed 100% of the time, but that’s part of the proxy follow-up. You could criticize the study for not having closer to 90% actual contact, but that’s the best we could do. 

Dr. Glatter: I think that’s admirable. Using that paradigm of what you described certainly allows the reader to understand the difficulty in assessing patients that don’t get follow-up head CT, and hardly anyone does that, as we know.

To your point of having both Level I trauma centers in the county, that makes it pretty secure. If we’re going to do a study encompassing a similar type of regional aspect, it would be similar.

Dr. Shenvi: I think your proxies, to your credit, were as good as you can get. You can never get a 100% follow-up, but you really looked at all the different avenues by which patients might present, either in the death registry or a Level I center. Well done on that aspect.

 

Determining When to Admit Patients for Observation

Dr. Glatter: In terms of admissions: You admit a patient, then you hear back that this patient should not have been admitted because they had a negative head CT, but you put them in anyway in the sense of delayed bleeding happening or not happening.

It’s interesting. Maybe the insurers will start looking at this in some capacity, based on your study, that because it’s so infrequent that you see delayed bleeding, that admitting someone for any reason whatsoever would be declined. Do you see that being an issue? In other words, [do you see] this leading to a pattern in terms of the payers?

Dr. Shih: Certainly, you could interpret it that way, and that would be unfortunate. The [incidence of] delayed bleeding is definitely not zero. That’s the first thing. 

The second thing is that when you’re dealing with an older population, having some sense that they’re not doing well is an important contributor to trying to fully assess what’s going on — whether or not they have a bleed or whether they’re at risk for falling again and then hitting their head and causing a second bleed, and making sure they can do the activities of daily life. There really should be some room for a physician to say, “They just got here, and we don’t know him that well. There’s something that bothers me about this person” and have the ability to watch them for at least another 24 hours. That’s how I feel. 

Dr. Shenvi: In my location, it would be difficult to try to admit somebody purely for observation for delayed bleeding. I think we would get a lot of pushback on that. The reasons I might admit a patient after a fall with a negative head CT, though, are all the things that, Rob, you alluded to earlier — which are, what made them fall in the first place and were they unable to get up? 

I had this happen just this week. A patient who fell couldn’t get off the ground for 12 hours, and so now she’s dehydrated and delirious with slight rhabdomyolysis. Then you’re admitting them either for the sequelae of the fall that are not related to the intracranial hemorrhage, or the fact that they are so debilitated and deconditioned that they cannot take care of themselves. They need physical therapy. Often, we will have physical and occupational therapists come see them in the ED during business hours and help make an assessment of whether they are safe to go home or whether they fall again. That can give more evidence for the need for admission.

Dr. Glatter: To bring artificial intelligence into this discussion, algorithms that are out there that say, “Push a button and the patient’s safe for discharge.” Well, this argues for a clinical gestalt and a human being to make an assessment because you can use these predictive models, which are coming and they’re going to be here soon, and they already are in some sense. Again, we have to use clinical human judgment. 

Dr. Shih: I agree. 
 

Advice for Primary Care Physicians

Dr. Glatter: What return precautions do you discuss with patients who’ve had blunt head trauma that maybe had a head CT, or even didn’t? What are the main things we’re looking for?

Dr. Shenvi: What I usually tell people is if you start to have a worse headache, nausea or vomiting, any weakness in one area of your body, or vision changes, and if there’s a family member or friend there, I’ll say, “If you notice that they’re acting differently or seem confused, come back.”

Dr. Shih: I agree with what she said, and I’m also going to add one thing. The most important part is they are trying to prevent a subsequent fall. We know that when they’ve fallen and they present to the ED, they’re at even higher risk for falling and reinjuring themselves, and that’s a population that’s already at risk.

One of the secondary studies that we published out of this project was looking at follow-up with their primary care physicians, and there were two things that we wanted to address. The first was, how often did they do it? Then, when they did do it, did their primary care physicians try to address and prevent subsequent falls?

Both the answers are actually bad. Amazingly, just over like 60% followed up. 

In some of our subsequent research, because we’re in the midst of a randomized, controlled trial where we do a home visit, when we initially see these individuals that have fallen, they’ll schedule a home visit for us. Then a week or two later, when we schedule the home visit, many of them cancel because they think, Oh, that was a one-off and it’s not going to happen again. Part of the problem is the patients, because many of them believe that they just slipped and fell and it’s not going to happen again, or they’re not prone to it.

The second issue was when patients did go to a primary care physician, we have found that some primary care physicians believe that falling and injuring themselves is just part of the normal aging process. A percentage of them don’t go over assessment for fall risk or even initiate fall prevention treatments or programs. 

I try to take that time to tell them that this is very common in their age group, and believe it or not, a fall from standing is the way people really injure themselves, and there may be ways to prevent subsequent falls and injuries. 

Dr. Glatter: Absolutely. Do you find that their medications are a contributor in some sense? Say they’re antihypertensive, have issues of orthostasis, or a new medication was added in the last week. 

Dr. Shenvi: It’s all of the above. Sometimes it’s one thing, like they just started tamsulosin for their kidney stone, they stood up, they felt lightheaded, and they fell. Usually, it’s multifactorial with some changes in their gait, vision, balance, reflex time, and strength, plus the medications or the need for assistive devices. Maybe they can’t take care of their home as well as they used to and there are things on the floor. It’s really all of the above.
 

‘Harder to Unlearn Something Than to Learn It’

Dr. Glatter: Would either of you like to add any additional points to the discussion or add a few pearls? 

Dr. Shenvi: This just highlights the challenge of how it’s harder to unlearn something than to learn it, where one study that maybe wasn’t quite looking at what we needed to, or practice and prescribing patterns have changed, so it’s no longer really relevant. 

The things that we learned from that, or the fears that we instilled in our minds of, Uh oh, they could go home and have delayed bleeding, are much harder to unlearn, and it takes more studies to unlearn that idea than it did to actually put it into place. 

I’m glad that your team has done this much larger, prospective study and hopefully will reduce the concern about this entity. 

Dr. Shih: I appreciate that segue. It is amazing that, for paramedics and medical students, the first thing out of their mouth is, “Are they on an anticoagulant?”

In terms of the risk of developing an intracranial hemorrhage, I think it’s much less than the weight we’ve put on it before. However, I believe if they have a bleed, the bleeds are worse. It’s kind of a double-edged sword. It’s still an important factor, but it doesn’t come with the Oh my gosh, they’re on an anticoagulant that everybody thinks about.
 

No. 1 Cause of Traumatic Injury Is a Fall from Standing

Dr. Glatter: These are obviously ground-level falls in most patients and not motor vehicle crashes. That’s an important part in the population that you looked at that should be mentioned clearly. 

Dr. Shih: It’s astonishing. I’ve been a program director for over 20 years, and geriatrics is not well taught in the curriculum. It’s astonishing for many of our trainees and emergency physicians in general that the number-one cause for traumatic injury is a fall from standing.

Certainly, we get patients coming in the trauma center like a 95-year-old person who’s on a ladder putting up his Christmas lights. I’m like, oh my God. 

For the vast majority, it’s closer to 90%, but in our study, for the patients we looked at, it was 80% that fall from standing. That’s the mechanism that causes these bleeds and these major injuries. 

Dr. Shenvi: That’s reflective of what we see, so it’s good that that’s what you looked at also. 

Dr. Glatter: Absolutely. Well, thank you both. This has been a very informative discussion. I appreciate your time, and our readers will certainly benefit from your knowledge and expertise. Thank you again.

Dr. Glatter, assistant professor of emergency medicine at Zucker School of Medicine at Hofstra/Northwell in Hempstead, New York, is a medical adviser for this news organization. He disclosed having no relevant financial conflicts. Dr. Shih is professor of emergency medicine at the Charles E. Schmidt College of Medicine at Florida Atlantic University, Boca Raton. His current grant funding and area of research interest involves geriatric emergency department patients with head injury and fall-related injury. He disclosed receiving a research grant from The Florida Medical Malpractice Joint Underwriting Association Grant for Safety of Health Care Services). Dr. Shenvi, associate professor of emergency medicine at the University of North Carolina at Chapel Hill, disclosed ties with the American College of Emergency Physicians, Institute for Healthcare Improvement, AstraZeneca, and CurvaFix.

A version of this article appeared on Medscape.com.

This discussion was recorded on August 2, 2024. This transcript has been edited for clarity. 

Robert D. Glatter, MD: Today, we’ll be discussing the results of a new study published in The Journal of Emergency Medicine, looking at the incidence of delayed intracranial hemorrhage among older patients taking preinjury anticoagulants who present to the emergency department (ED) with blunt head trauma.

Joining me today is the lead author of the study, Dr. Richard Shih, professor of emergency medicine at Florida Atlantic University. Also joining me is Dr. Christina Shenvi, associate professor of emergency medicine at the University of North Carolina (UNC) Chapel Hill, with fellowship training in geriatric emergency medicine. 

Welcome to both of you.

Richard D. Shih, MD: Thanks, Rob. 

Christina L. Shenvi, MD, PhD, MBA: Thanks. Pleasure to be here. 
 

ICH Study Methodology

Dr. Glatter: It’s a pleasure to have you. Rich, this is a great study and targeted toward a population we see daily in the emergency department. I want you to describe your methodology, patient selection, and how you went about organizing your study to look at this important finding of delayed intracranial hemorrhage, especially in those on anticoagulants.

Dr. Shih: This all started for our research team when we first read the 2012 Annals of Emergency Medicine paper. The first author was Vincenzo Menditto, and he looked at a group of patients that had minor head injury, were anticoagulated, and had negative initial head CTs. 

There were about 100 patients, of which about 10 of them did not consent, but they hospitalized all these patients. These were anticoagulated, negative-first head CTs. They hospitalized the patients and then did a routine second CT at about 24 hours. They also followed them for a week, and it turned out a little over 7% of them had delayed head CT. 

We were wondering how many delayed intracranial hemorrhages we had missed because current practice for us was that, if patients had a good physical exam, their head CT was normal, and everything looked good, we would send them home.

Because of that, a number of people across the country wanted to verify those findings from the Menditto study. We tried to design a good study to answer that question. We happen to have a very large geriatric population in Florida, and our ED census is very high for age over 65, at nearly 60%. 

There are two Level I trauma centers in Palm Beach County. We included a second multicenter hospital, and we prospectively enrolled patients. We know the current state of practice is not to routinely do second CTs, so we followed these patients over time and followed their medical records to try to identify delayed bleeding. That’s how we set up our methodology.
 

Is It Safe to Discharge Patients With Trauma After 24 Hours?

Dr. Glatter: For the bulk of these patients with negative head CTs, it’s been my practice that when they’re stable and they look fine and there’s no other apparent, distracting painful trauma, injuries and so forth, they’re safe to discharge. 

The secondary outcome in your study is interesting: the need for neurosurgical intervention in terms of those with delayed intracranial hemorrhage.

Dr. Shih: I do believe that it’s certainly not the problem that Menditto described, which is 7%. There are two other prospective studies that have looked at this issue with delayed bleeding on anticoagulants. Both of these also showed a relatively low rate of delayed bleeding, which is between like 0.2% and 1.0%. In our study, it was 0.4%. 

The difference in the studies is that Menditto and colleagues routinely did 24-hour head CTs. They admitted everybody. For these other studies, routine head CT was not part of it. My bet is that there is a rate of delayed bleeding somewhere in between that seen in the Menditto study and that in all the other studies.

However, talking about significant intracranial hemorrhage, ones that perhaps need neurosurgery, I believe most of them are not significant. There’s some number that do occur, but the vast majority of those probably don’t need neurosurgery. We had 14 delayed bleeds out of 6000 patients with head trauma. One of them ended up requiring neurosurgery, so the answer is not zero, but I don’t think it’s 7% either. 

Dr. Glatter: Dr. Shenvi, I want to bring you into the conversation to talk about your experience at UNC, and how you run things in terms of older patients with blunt head trauma on preinjury anticoagulants.

Dr. Shenvi: Thanks, Rob. I remember when this paper came out showing this 7% rate of delayed bleeding and the question was, “Should we be admitting all these people?” Partly just from an overwhelming need for capacity that that would bring, it just wasn’t practical to say, “We’re going to admit every patient with a negative head CT to the hospital and rescan them.” That would be hundreds or thousands of patients each year in any given facility. 

The other thing is that delayed bleeds don’t always happen just in the first 24 hours. It’s not even a matter of bringing patients into observation for 24 hours, watching them, and rescanning them if they have symptoms. It can occur several days out. That never, in almost any institution that I know of, became standard practice. 

The way that it did change my care was to give good return precautions to patients, to make sure they have somebody with them to say, “Hey, sometimes you can have bleeding several days out after a fall, even though your CT scan here today looks perfect,” and to alert them that if they start having severe headaches, vomiting, or other symptoms of intracranial hemorrhage, that they should come back. 

I don’t think it ever became standard practice, and for good reason, because that was one study. The subsequent studies that Richard mentioned, pretty quickly on the heels of that initial one, showed a much lower rate of delayed ICH with the caveats that the methodology was different. 
 

Shift in Anticoagulants

Dr. Shenvi: One other big change from that original study, and now to Richard’s study, is the shift in anticoagulants. Back in the initial study you mentioned, it was all warfarin. We know from other studies looking at warfarin vs the direct oral anticoagulants (DOACs) that DOACs have lower rates of ICH after a head injury, lower rates of need for neurosurgical intervention, and lower rates of discharge to a skilled nursing facility after an intracranial hemorrhage.

Across the board, we know that the DOACs tend to do better. It’s difficult to compare newer studies because it’s a different medication. It did inform my practice to have an awareness of delayed intracranial hemorrhage so that I warn patients more proactively. 

Dr. Glatter: I haven’t seen a patient on warfarin in years. I don’t know if either of you have, but it’s all DOACs now unless there’s some other reason. That shift is quite apparent.

Dr. Shih: The problem with looking at delayed bleeding for DOACs vs warfarin is the numbers were so low. I think we had 13 people, and seven were in the no-anticoagulant group. The numbers are even lower, so it’s hard to say. 

I just wanted to comment on something that Dr. Shenvi said, and I pretty much agree with everything that she said. Anticoagulants and warfarin, and that Menditto study, have a carryover effect. People group DOACs with warfarin similarly. When a patient is brought in, the first thing they talk about with head trauma is, “Oh, they’re on an anticoagulant” or “They’re not on an anticoagulant.” It’s so ingrained.

I believe that, in emergency medicine, we’re pressed for space and time and we’re not as affected by that 24-hour observation. Maybe many of our surgeons will automatically admit those patients. 

I haven’t seen a guideline from the United States, but there are two international guidelines. One is from Austria from 2019, and one is from Scandinavia. Both recommended 24-hour observation if you’re on an anticoagulant.

There is a bit of controversy left over with that. Hopefully, as more and more of information, like in our study, comes out, people will be a little bit more clear about it. I don’t think there’s a need to routinely admit them. 

I do want to mention that the Menditto study had such a massive impact on everybody. They pointed out one subgroup (and it’s such a small number of patients). They had seven cases of delayed bleeding; four or five of them were within that 24 hours, and a couple were diagnosed later over the next couple days.

Of those seven people, four of them had international normalized ratios (INRs) greater than 3. Of those four patients, I’ve heard people talk about this and recommend, “Okay, that’s the subgroup I would admit.” There’s a toss-up with what to do with DOAC because it’s very hard to tell whether there’s an issue, whether there are problems with their dosing, and whatever. 

We actually recently looked at that. We have a much larger sample than four: close to 300 patients who were on warfarin. We looked at patients who had INRs below 3 and above 3, and we didn’t show a difference. We still don’t believe that warfarin is a big issue with delayed bleeding.
 

Should We Be Asking: ‘Are They on Blood Thinners?’

Dr. Shenvi: One of the interesting trends related to warfarin and the DOACs vs no anticoagulant is that as you mentioned, Dr Shih, the first question out of people’s mouths or the first piece of information emergency medical services gives you when they come in with a patient who’s had a head injury is, “Are they on blood thinners or not?”

Yet, the paradigm is shifting to say it’s not actually the blood thinners themselves that are giving older patients the higher risk for bleeding; it’s age and other comorbidities.

Certainly, if you’re on an anticoagulant and you start to bleed, your prognosis is much worse because the bleeding doesn’t stop. In terms of who has a bleeding event, there’s much less impact of anticoagulation than we used to think. That, in part, may be due to the change from warfarin to other medications.

Some of the experts I’ve talked to who have done the research on this have said, “Well, actually, warfarin was more of a marker for being much older and more frail, because it was primarily prescribed to older patients who have significant heart disease, atrial fibrillation, and so on.” It was more a marker for somebody who is at risk for an intracranial hemorrhage. There are many changes that have happened in the past 10 years with medications and also our understanding. 
 

Challenges in Patient Follow-up

Dr. Glatter: That’s a great point. One thing, Rich, I want to ask you about is in terms of your proxy outcome assessment. When you use that at 14 and 60 days with telephone follow-up and then chart review at 60 and 90 days (because, obviously, everyone can’t get another head CT or it’s difficult to follow patients up), did you find that worked out well in your prospective cohort study, in terms of using that as a proxy, so to speak? 

Dr. Shih: I would say to a certain extent. Unfortunately, we don’t have access to the patients to come back to follow up all of them, and there was obviously a large number of patients in our study. 

The next best thing was that we had dedicated research assistants calling all of the patients at 14 days and 60 days. I’ve certainly read research studies where, when they call them, they get 80%-90% follow-up, but we did not achieve that.

I don’t know if people are more inundated with spam phone calls now, or the older people are just afraid of picking up their phone sometimes with all the scams and so forth. I totally understand, but in all honesty, we only had about a 30%-35% follow-up using that follow-up pathway. 

Then the proxy pathway was to look at their charts at 60 and 90 days. Also, we looked at the Florida death registry, which is pretty good, and then finally, we had both Level I trauma centers in the county that we were in participating. It’s standard practice that if you have an intracranial hemorrhage at a non–Level I trauma center, you would be transferred to a Level I trauma center. That’s the protocol. I know that’s not followed 100% of the time, but that’s part of the proxy follow-up. You could criticize the study for not having closer to 90% actual contact, but that’s the best we could do. 

Dr. Glatter: I think that’s admirable. Using that paradigm of what you described certainly allows the reader to understand the difficulty in assessing patients that don’t get follow-up head CT, and hardly anyone does that, as we know.

To your point of having both Level I trauma centers in the county, that makes it pretty secure. If we’re going to do a study encompassing a similar type of regional aspect, it would be similar.

Dr. Shenvi: I think your proxies, to your credit, were as good as you can get. You can never get a 100% follow-up, but you really looked at all the different avenues by which patients might present, either in the death registry or a Level I center. Well done on that aspect.

 

Determining When to Admit Patients for Observation

Dr. Glatter: In terms of admissions: You admit a patient, then you hear back that this patient should not have been admitted because they had a negative head CT, but you put them in anyway in the sense of delayed bleeding happening or not happening.

It’s interesting. Maybe the insurers will start looking at this in some capacity, based on your study, that because it’s so infrequent that you see delayed bleeding, that admitting someone for any reason whatsoever would be declined. Do you see that being an issue? In other words, [do you see] this leading to a pattern in terms of the payers?

Dr. Shih: Certainly, you could interpret it that way, and that would be unfortunate. The [incidence of] delayed bleeding is definitely not zero. That’s the first thing. 

The second thing is that when you’re dealing with an older population, having some sense that they’re not doing well is an important contributor to trying to fully assess what’s going on — whether or not they have a bleed or whether they’re at risk for falling again and then hitting their head and causing a second bleed, and making sure they can do the activities of daily life. There really should be some room for a physician to say, “They just got here, and we don’t know him that well. There’s something that bothers me about this person” and have the ability to watch them for at least another 24 hours. That’s how I feel. 

Dr. Shenvi: In my location, it would be difficult to try to admit somebody purely for observation for delayed bleeding. I think we would get a lot of pushback on that. The reasons I might admit a patient after a fall with a negative head CT, though, are all the things that, Rob, you alluded to earlier — which are, what made them fall in the first place and were they unable to get up? 

I had this happen just this week. A patient who fell couldn’t get off the ground for 12 hours, and so now she’s dehydrated and delirious with slight rhabdomyolysis. Then you’re admitting them either for the sequelae of the fall that are not related to the intracranial hemorrhage, or the fact that they are so debilitated and deconditioned that they cannot take care of themselves. They need physical therapy. Often, we will have physical and occupational therapists come see them in the ED during business hours and help make an assessment of whether they are safe to go home or whether they fall again. That can give more evidence for the need for admission.

Dr. Glatter: To bring artificial intelligence into this discussion, algorithms that are out there that say, “Push a button and the patient’s safe for discharge.” Well, this argues for a clinical gestalt and a human being to make an assessment because you can use these predictive models, which are coming and they’re going to be here soon, and they already are in some sense. Again, we have to use clinical human judgment. 

Dr. Shih: I agree. 
 

Advice for Primary Care Physicians

Dr. Glatter: What return precautions do you discuss with patients who’ve had blunt head trauma that maybe had a head CT, or even didn’t? What are the main things we’re looking for?

Dr. Shenvi: What I usually tell people is if you start to have a worse headache, nausea or vomiting, any weakness in one area of your body, or vision changes, and if there’s a family member or friend there, I’ll say, “If you notice that they’re acting differently or seem confused, come back.”

Dr. Shih: I agree with what she said, and I’m also going to add one thing. The most important part is they are trying to prevent a subsequent fall. We know that when they’ve fallen and they present to the ED, they’re at even higher risk for falling and reinjuring themselves, and that’s a population that’s already at risk.

One of the secondary studies that we published out of this project was looking at follow-up with their primary care physicians, and there were two things that we wanted to address. The first was, how often did they do it? Then, when they did do it, did their primary care physicians try to address and prevent subsequent falls?

Both the answers are actually bad. Amazingly, just over like 60% followed up. 

In some of our subsequent research, because we’re in the midst of a randomized, controlled trial where we do a home visit, when we initially see these individuals that have fallen, they’ll schedule a home visit for us. Then a week or two later, when we schedule the home visit, many of them cancel because they think, Oh, that was a one-off and it’s not going to happen again. Part of the problem is the patients, because many of them believe that they just slipped and fell and it’s not going to happen again, or they’re not prone to it.

The second issue was when patients did go to a primary care physician, we have found that some primary care physicians believe that falling and injuring themselves is just part of the normal aging process. A percentage of them don’t go over assessment for fall risk or even initiate fall prevention treatments or programs. 

I try to take that time to tell them that this is very common in their age group, and believe it or not, a fall from standing is the way people really injure themselves, and there may be ways to prevent subsequent falls and injuries. 

Dr. Glatter: Absolutely. Do you find that their medications are a contributor in some sense? Say they’re antihypertensive, have issues of orthostasis, or a new medication was added in the last week. 

Dr. Shenvi: It’s all of the above. Sometimes it’s one thing, like they just started tamsulosin for their kidney stone, they stood up, they felt lightheaded, and they fell. Usually, it’s multifactorial with some changes in their gait, vision, balance, reflex time, and strength, plus the medications or the need for assistive devices. Maybe they can’t take care of their home as well as they used to and there are things on the floor. It’s really all of the above.
 

‘Harder to Unlearn Something Than to Learn It’

Dr. Glatter: Would either of you like to add any additional points to the discussion or add a few pearls? 

Dr. Shenvi: This just highlights the challenge of how it’s harder to unlearn something than to learn it, where one study that maybe wasn’t quite looking at what we needed to, or practice and prescribing patterns have changed, so it’s no longer really relevant. 

The things that we learned from that, or the fears that we instilled in our minds of, Uh oh, they could go home and have delayed bleeding, are much harder to unlearn, and it takes more studies to unlearn that idea than it did to actually put it into place. 

I’m glad that your team has done this much larger, prospective study and hopefully will reduce the concern about this entity. 

Dr. Shih: I appreciate that segue. It is amazing that, for paramedics and medical students, the first thing out of their mouth is, “Are they on an anticoagulant?”

In terms of the risk of developing an intracranial hemorrhage, I think it’s much less than the weight we’ve put on it before. However, I believe if they have a bleed, the bleeds are worse. It’s kind of a double-edged sword. It’s still an important factor, but it doesn’t come with the Oh my gosh, they’re on an anticoagulant that everybody thinks about.
 

No. 1 Cause of Traumatic Injury Is a Fall from Standing

Dr. Glatter: These are obviously ground-level falls in most patients and not motor vehicle crashes. That’s an important part in the population that you looked at that should be mentioned clearly. 

Dr. Shih: It’s astonishing. I’ve been a program director for over 20 years, and geriatrics is not well taught in the curriculum. It’s astonishing for many of our trainees and emergency physicians in general that the number-one cause for traumatic injury is a fall from standing.

Certainly, we get patients coming in the trauma center like a 95-year-old person who’s on a ladder putting up his Christmas lights. I’m like, oh my God. 

For the vast majority, it’s closer to 90%, but in our study, for the patients we looked at, it was 80% that fall from standing. That’s the mechanism that causes these bleeds and these major injuries. 

Dr. Shenvi: That’s reflective of what we see, so it’s good that that’s what you looked at also. 

Dr. Glatter: Absolutely. Well, thank you both. This has been a very informative discussion. I appreciate your time, and our readers will certainly benefit from your knowledge and expertise. Thank you again.

Dr. Glatter, assistant professor of emergency medicine at Zucker School of Medicine at Hofstra/Northwell in Hempstead, New York, is a medical adviser for this news organization. He disclosed having no relevant financial conflicts. Dr. Shih is professor of emergency medicine at the Charles E. Schmidt College of Medicine at Florida Atlantic University, Boca Raton. His current grant funding and area of research interest involves geriatric emergency department patients with head injury and fall-related injury. He disclosed receiving a research grant from The Florida Medical Malpractice Joint Underwriting Association Grant for Safety of Health Care Services). Dr. Shenvi, associate professor of emergency medicine at the University of North Carolina at Chapel Hill, disclosed ties with the American College of Emergency Physicians, Institute for Healthcare Improvement, AstraZeneca, and CurvaFix.

A version of this article appeared on Medscape.com.

This discussion was recorded on August 2, 2024. This transcript has been edited for clarity. 

Robert D. Glatter, MD: Today, we’ll be discussing the results of a new study published in The Journal of Emergency Medicine, looking at the incidence of delayed intracranial hemorrhage among older patients taking preinjury anticoagulants who present to the emergency department (ED) with blunt head trauma.

Joining me today is the lead author of the study, Dr. Richard Shih, professor of emergency medicine at Florida Atlantic University. Also joining me is Dr. Christina Shenvi, associate professor of emergency medicine at the University of North Carolina (UNC) Chapel Hill, with fellowship training in geriatric emergency medicine. 

Welcome to both of you.

Richard D. Shih, MD: Thanks, Rob. 

Christina L. Shenvi, MD, PhD, MBA: Thanks. Pleasure to be here. 
 

ICH Study Methodology

Dr. Glatter: It’s a pleasure to have you. Rich, this is a great study and targeted toward a population we see daily in the emergency department. I want you to describe your methodology, patient selection, and how you went about organizing your study to look at this important finding of delayed intracranial hemorrhage, especially in those on anticoagulants.

Dr. Shih: This all started for our research team when we first read the 2012 Annals of Emergency Medicine paper. The first author was Vincenzo Menditto, and he looked at a group of patients that had minor head injury, were anticoagulated, and had negative initial head CTs. 

There were about 100 patients, of which about 10 of them did not consent, but they hospitalized all these patients. These were anticoagulated, negative-first head CTs. They hospitalized the patients and then did a routine second CT at about 24 hours. They also followed them for a week, and it turned out a little over 7% of them had delayed head CT. 

We were wondering how many delayed intracranial hemorrhages we had missed because current practice for us was that, if patients had a good physical exam, their head CT was normal, and everything looked good, we would send them home.

Because of that, a number of people across the country wanted to verify those findings from the Menditto study. We tried to design a good study to answer that question. We happen to have a very large geriatric population in Florida, and our ED census is very high for age over 65, at nearly 60%. 

There are two Level I trauma centers in Palm Beach County. We included a second multicenter hospital, and we prospectively enrolled patients. We know the current state of practice is not to routinely do second CTs, so we followed these patients over time and followed their medical records to try to identify delayed bleeding. That’s how we set up our methodology.
 

Is It Safe to Discharge Patients With Trauma After 24 Hours?

Dr. Glatter: For the bulk of these patients with negative head CTs, it’s been my practice that when they’re stable and they look fine and there’s no other apparent, distracting painful trauma, injuries and so forth, they’re safe to discharge. 

The secondary outcome in your study is interesting: the need for neurosurgical intervention in terms of those with delayed intracranial hemorrhage.

Dr. Shih: I do believe that it’s certainly not the problem that Menditto described, which is 7%. There are two other prospective studies that have looked at this issue with delayed bleeding on anticoagulants. Both of these also showed a relatively low rate of delayed bleeding, which is between like 0.2% and 1.0%. In our study, it was 0.4%. 

The difference in the studies is that Menditto and colleagues routinely did 24-hour head CTs. They admitted everybody. For these other studies, routine head CT was not part of it. My bet is that there is a rate of delayed bleeding somewhere in between that seen in the Menditto study and that in all the other studies.

However, talking about significant intracranial hemorrhage, ones that perhaps need neurosurgery, I believe most of them are not significant. There’s some number that do occur, but the vast majority of those probably don’t need neurosurgery. We had 14 delayed bleeds out of 6000 patients with head trauma. One of them ended up requiring neurosurgery, so the answer is not zero, but I don’t think it’s 7% either. 

Dr. Glatter: Dr. Shenvi, I want to bring you into the conversation to talk about your experience at UNC, and how you run things in terms of older patients with blunt head trauma on preinjury anticoagulants.

Dr. Shenvi: Thanks, Rob. I remember when this paper came out showing this 7% rate of delayed bleeding and the question was, “Should we be admitting all these people?” Partly just from an overwhelming need for capacity that that would bring, it just wasn’t practical to say, “We’re going to admit every patient with a negative head CT to the hospital and rescan them.” That would be hundreds or thousands of patients each year in any given facility. 

The other thing is that delayed bleeds don’t always happen just in the first 24 hours. It’s not even a matter of bringing patients into observation for 24 hours, watching them, and rescanning them if they have symptoms. It can occur several days out. That never, in almost any institution that I know of, became standard practice. 

The way that it did change my care was to give good return precautions to patients, to make sure they have somebody with them to say, “Hey, sometimes you can have bleeding several days out after a fall, even though your CT scan here today looks perfect,” and to alert them that if they start having severe headaches, vomiting, or other symptoms of intracranial hemorrhage, that they should come back. 

I don’t think it ever became standard practice, and for good reason, because that was one study. The subsequent studies that Richard mentioned, pretty quickly on the heels of that initial one, showed a much lower rate of delayed ICH with the caveats that the methodology was different. 
 

Shift in Anticoagulants

Dr. Shenvi: One other big change from that original study, and now to Richard’s study, is the shift in anticoagulants. Back in the initial study you mentioned, it was all warfarin. We know from other studies looking at warfarin vs the direct oral anticoagulants (DOACs) that DOACs have lower rates of ICH after a head injury, lower rates of need for neurosurgical intervention, and lower rates of discharge to a skilled nursing facility after an intracranial hemorrhage.

Across the board, we know that the DOACs tend to do better. It’s difficult to compare newer studies because it’s a different medication. It did inform my practice to have an awareness of delayed intracranial hemorrhage so that I warn patients more proactively. 

Dr. Glatter: I haven’t seen a patient on warfarin in years. I don’t know if either of you have, but it’s all DOACs now unless there’s some other reason. That shift is quite apparent.

Dr. Shih: The problem with looking at delayed bleeding for DOACs vs warfarin is the numbers were so low. I think we had 13 people, and seven were in the no-anticoagulant group. The numbers are even lower, so it’s hard to say. 

I just wanted to comment on something that Dr. Shenvi said, and I pretty much agree with everything that she said. Anticoagulants and warfarin, and that Menditto study, have a carryover effect. People group DOACs with warfarin similarly. When a patient is brought in, the first thing they talk about with head trauma is, “Oh, they’re on an anticoagulant” or “They’re not on an anticoagulant.” It’s so ingrained.

I believe that, in emergency medicine, we’re pressed for space and time and we’re not as affected by that 24-hour observation. Maybe many of our surgeons will automatically admit those patients. 

I haven’t seen a guideline from the United States, but there are two international guidelines. One is from Austria from 2019, and one is from Scandinavia. Both recommended 24-hour observation if you’re on an anticoagulant.

There is a bit of controversy left over with that. Hopefully, as more and more of information, like in our study, comes out, people will be a little bit more clear about it. I don’t think there’s a need to routinely admit them. 

I do want to mention that the Menditto study had such a massive impact on everybody. They pointed out one subgroup (and it’s such a small number of patients). They had seven cases of delayed bleeding; four or five of them were within that 24 hours, and a couple were diagnosed later over the next couple days.

Of those seven people, four of them had international normalized ratios (INRs) greater than 3. Of those four patients, I’ve heard people talk about this and recommend, “Okay, that’s the subgroup I would admit.” There’s a toss-up with what to do with DOAC because it’s very hard to tell whether there’s an issue, whether there are problems with their dosing, and whatever. 

We actually recently looked at that. We have a much larger sample than four: close to 300 patients who were on warfarin. We looked at patients who had INRs below 3 and above 3, and we didn’t show a difference. We still don’t believe that warfarin is a big issue with delayed bleeding.
 

Should We Be Asking: ‘Are They on Blood Thinners?’

Dr. Shenvi: One of the interesting trends related to warfarin and the DOACs vs no anticoagulant is that as you mentioned, Dr Shih, the first question out of people’s mouths or the first piece of information emergency medical services gives you when they come in with a patient who’s had a head injury is, “Are they on blood thinners or not?”

Yet, the paradigm is shifting to say it’s not actually the blood thinners themselves that are giving older patients the higher risk for bleeding; it’s age and other comorbidities.

Certainly, if you’re on an anticoagulant and you start to bleed, your prognosis is much worse because the bleeding doesn’t stop. In terms of who has a bleeding event, there’s much less impact of anticoagulation than we used to think. That, in part, may be due to the change from warfarin to other medications.

Some of the experts I’ve talked to who have done the research on this have said, “Well, actually, warfarin was more of a marker for being much older and more frail, because it was primarily prescribed to older patients who have significant heart disease, atrial fibrillation, and so on.” It was more a marker for somebody who is at risk for an intracranial hemorrhage. There are many changes that have happened in the past 10 years with medications and also our understanding. 
 

Challenges in Patient Follow-up

Dr. Glatter: That’s a great point. One thing, Rich, I want to ask you about is in terms of your proxy outcome assessment. When you use that at 14 and 60 days with telephone follow-up and then chart review at 60 and 90 days (because, obviously, everyone can’t get another head CT or it’s difficult to follow patients up), did you find that worked out well in your prospective cohort study, in terms of using that as a proxy, so to speak? 

Dr. Shih: I would say to a certain extent. Unfortunately, we don’t have access to the patients to come back to follow up all of them, and there was obviously a large number of patients in our study. 

The next best thing was that we had dedicated research assistants calling all of the patients at 14 days and 60 days. I’ve certainly read research studies where, when they call them, they get 80%-90% follow-up, but we did not achieve that.

I don’t know if people are more inundated with spam phone calls now, or the older people are just afraid of picking up their phone sometimes with all the scams and so forth. I totally understand, but in all honesty, we only had about a 30%-35% follow-up using that follow-up pathway. 

Then the proxy pathway was to look at their charts at 60 and 90 days. Also, we looked at the Florida death registry, which is pretty good, and then finally, we had both Level I trauma centers in the county that we were in participating. It’s standard practice that if you have an intracranial hemorrhage at a non–Level I trauma center, you would be transferred to a Level I trauma center. That’s the protocol. I know that’s not followed 100% of the time, but that’s part of the proxy follow-up. You could criticize the study for not having closer to 90% actual contact, but that’s the best we could do. 

Dr. Glatter: I think that’s admirable. Using that paradigm of what you described certainly allows the reader to understand the difficulty in assessing patients that don’t get follow-up head CT, and hardly anyone does that, as we know.

To your point of having both Level I trauma centers in the county, that makes it pretty secure. If we’re going to do a study encompassing a similar type of regional aspect, it would be similar.

Dr. Shenvi: I think your proxies, to your credit, were as good as you can get. You can never get a 100% follow-up, but you really looked at all the different avenues by which patients might present, either in the death registry or a Level I center. Well done on that aspect.

 

Determining When to Admit Patients for Observation

Dr. Glatter: In terms of admissions: You admit a patient, then you hear back that this patient should not have been admitted because they had a negative head CT, but you put them in anyway in the sense of delayed bleeding happening or not happening.

It’s interesting. Maybe the insurers will start looking at this in some capacity, based on your study, that because it’s so infrequent that you see delayed bleeding, that admitting someone for any reason whatsoever would be declined. Do you see that being an issue? In other words, [do you see] this leading to a pattern in terms of the payers?

Dr. Shih: Certainly, you could interpret it that way, and that would be unfortunate. The [incidence of] delayed bleeding is definitely not zero. That’s the first thing. 

The second thing is that when you’re dealing with an older population, having some sense that they’re not doing well is an important contributor to trying to fully assess what’s going on — whether or not they have a bleed or whether they’re at risk for falling again and then hitting their head and causing a second bleed, and making sure they can do the activities of daily life. There really should be some room for a physician to say, “They just got here, and we don’t know him that well. There’s something that bothers me about this person” and have the ability to watch them for at least another 24 hours. That’s how I feel. 

Dr. Shenvi: In my location, it would be difficult to try to admit somebody purely for observation for delayed bleeding. I think we would get a lot of pushback on that. The reasons I might admit a patient after a fall with a negative head CT, though, are all the things that, Rob, you alluded to earlier — which are, what made them fall in the first place and were they unable to get up? 

I had this happen just this week. A patient who fell couldn’t get off the ground for 12 hours, and so now she’s dehydrated and delirious with slight rhabdomyolysis. Then you’re admitting them either for the sequelae of the fall that are not related to the intracranial hemorrhage, or the fact that they are so debilitated and deconditioned that they cannot take care of themselves. They need physical therapy. Often, we will have physical and occupational therapists come see them in the ED during business hours and help make an assessment of whether they are safe to go home or whether they fall again. That can give more evidence for the need for admission.

Dr. Glatter: To bring artificial intelligence into this discussion, algorithms that are out there that say, “Push a button and the patient’s safe for discharge.” Well, this argues for a clinical gestalt and a human being to make an assessment because you can use these predictive models, which are coming and they’re going to be here soon, and they already are in some sense. Again, we have to use clinical human judgment. 

Dr. Shih: I agree. 
 

Advice for Primary Care Physicians

Dr. Glatter: What return precautions do you discuss with patients who’ve had blunt head trauma that maybe had a head CT, or even didn’t? What are the main things we’re looking for?

Dr. Shenvi: What I usually tell people is if you start to have a worse headache, nausea or vomiting, any weakness in one area of your body, or vision changes, and if there’s a family member or friend there, I’ll say, “If you notice that they’re acting differently or seem confused, come back.”

Dr. Shih: I agree with what she said, and I’m also going to add one thing. The most important part is they are trying to prevent a subsequent fall. We know that when they’ve fallen and they present to the ED, they’re at even higher risk for falling and reinjuring themselves, and that’s a population that’s already at risk.

One of the secondary studies that we published out of this project was looking at follow-up with their primary care physicians, and there were two things that we wanted to address. The first was, how often did they do it? Then, when they did do it, did their primary care physicians try to address and prevent subsequent falls?

Both the answers are actually bad. Amazingly, just over like 60% followed up. 

In some of our subsequent research, because we’re in the midst of a randomized, controlled trial where we do a home visit, when we initially see these individuals that have fallen, they’ll schedule a home visit for us. Then a week or two later, when we schedule the home visit, many of them cancel because they think, Oh, that was a one-off and it’s not going to happen again. Part of the problem is the patients, because many of them believe that they just slipped and fell and it’s not going to happen again, or they’re not prone to it.

The second issue was when patients did go to a primary care physician, we have found that some primary care physicians believe that falling and injuring themselves is just part of the normal aging process. A percentage of them don’t go over assessment for fall risk or even initiate fall prevention treatments or programs. 

I try to take that time to tell them that this is very common in their age group, and believe it or not, a fall from standing is the way people really injure themselves, and there may be ways to prevent subsequent falls and injuries. 

Dr. Glatter: Absolutely. Do you find that their medications are a contributor in some sense? Say they’re antihypertensive, have issues of orthostasis, or a new medication was added in the last week. 

Dr. Shenvi: It’s all of the above. Sometimes it’s one thing, like they just started tamsulosin for their kidney stone, they stood up, they felt lightheaded, and they fell. Usually, it’s multifactorial with some changes in their gait, vision, balance, reflex time, and strength, plus the medications or the need for assistive devices. Maybe they can’t take care of their home as well as they used to and there are things on the floor. It’s really all of the above.
 

‘Harder to Unlearn Something Than to Learn It’

Dr. Glatter: Would either of you like to add any additional points to the discussion or add a few pearls? 

Dr. Shenvi: This just highlights the challenge of how it’s harder to unlearn something than to learn it, where one study that maybe wasn’t quite looking at what we needed to, or practice and prescribing patterns have changed, so it’s no longer really relevant. 

The things that we learned from that, or the fears that we instilled in our minds of, Uh oh, they could go home and have delayed bleeding, are much harder to unlearn, and it takes more studies to unlearn that idea than it did to actually put it into place. 

I’m glad that your team has done this much larger, prospective study and hopefully will reduce the concern about this entity. 

Dr. Shih: I appreciate that segue. It is amazing that, for paramedics and medical students, the first thing out of their mouth is, “Are they on an anticoagulant?”

In terms of the risk of developing an intracranial hemorrhage, I think it’s much less than the weight we’ve put on it before. However, I believe if they have a bleed, the bleeds are worse. It’s kind of a double-edged sword. It’s still an important factor, but it doesn’t come with the Oh my gosh, they’re on an anticoagulant that everybody thinks about.
 

No. 1 Cause of Traumatic Injury Is a Fall from Standing

Dr. Glatter: These are obviously ground-level falls in most patients and not motor vehicle crashes. That’s an important part in the population that you looked at that should be mentioned clearly. 

Dr. Shih: It’s astonishing. I’ve been a program director for over 20 years, and geriatrics is not well taught in the curriculum. It’s astonishing for many of our trainees and emergency physicians in general that the number-one cause for traumatic injury is a fall from standing.

Certainly, we get patients coming in the trauma center like a 95-year-old person who’s on a ladder putting up his Christmas lights. I’m like, oh my God. 

For the vast majority, it’s closer to 90%, but in our study, for the patients we looked at, it was 80% that fall from standing. That’s the mechanism that causes these bleeds and these major injuries. 

Dr. Shenvi: That’s reflective of what we see, so it’s good that that’s what you looked at also. 

Dr. Glatter: Absolutely. Well, thank you both. This has been a very informative discussion. I appreciate your time, and our readers will certainly benefit from your knowledge and expertise. Thank you again.

Dr. Glatter, assistant professor of emergency medicine at Zucker School of Medicine at Hofstra/Northwell in Hempstead, New York, is a medical adviser for this news organization. He disclosed having no relevant financial conflicts. Dr. Shih is professor of emergency medicine at the Charles E. Schmidt College of Medicine at Florida Atlantic University, Boca Raton. His current grant funding and area of research interest involves geriatric emergency department patients with head injury and fall-related injury. He disclosed receiving a research grant from The Florida Medical Malpractice Joint Underwriting Association Grant for Safety of Health Care Services). Dr. Shenvi, associate professor of emergency medicine at the University of North Carolina at Chapel Hill, disclosed ties with the American College of Emergency Physicians, Institute for Healthcare Improvement, AstraZeneca, and CurvaFix.

A version of this article appeared on Medscape.com.

A new analysis suggests one reason doctors are wary of value-based care arrangements: Overkill.

Researchers found that primary care physicians in one large integrated health system were required to track an average of 57 different quality measures across multiple insurers that linked outcomes to payments under value-based contracts.

Medicare contracts were the most likely to pile quality measures on physicians with an average of 13.42 measures vs 10.07 for commercial insurer contracts and 5.37 for Medicaid contracts, reported Claire Boone, PhD, of the University of Chicago in Illinois and Providence Research Network, Portland, Oregon, and colleagues in JAMA Health Forum. The analysis, which may be the first of its kind, tracked 890 primary care physicians from 2020 to 2022.

The average of 57 quality measures per physician was unexpectedly high, Dr. Boone said in an interview.

“The magnitude of that number surprised us,” Dr. Boone said. “Primary care physicians and their practices have a lot on their plate. Now we know that one of those things is a very large number of different quality metrics to pay attention to, measure, report on, and implement.

Value-based care programs use quality measures to evaluate how well clinicians are doing their jobs and adjust reimbursement accordingly. A payer, for example, may raise reimbursements if a clinician has higher numbers of patients who meet quality measure standards for depression screening or blood pressure.

Dr. Boone said her research group is studying the impact of quality measures and was surprised that data showed individual primary care physicians had to deal with a high number of value-based contracts.

The researchers tracked value-based contracts for 890 physicians (58.3% women, 41.7% men) in an unidentified West Coast Health system. (Several study authors work for the Providence Health System, which serves several Western States and Texas.) The average number of patients per physician was 1309.

The physicians were part of an average of 11.18 value-based contracts (commercial insurers, 49.50%; Medicaid, 21.49%; and Medicare, 29.01%). This number grew from 9.39 in 2020 to 12.26 in 2022. Quality measure data weren’t available for 29% of contracts.

Quality measures were considered unique if they referenced different conditions.

For example, colorectal cancer screening is unique from depression screening. The researchers also considered measures for the same condition unique if the target value differed — for example, blood pressure control defined as < 140/90 vs blood pressure control defined as < 130/80, Dr. Boone said.

Dr. Boone said she expected payers to coordinate quality measures.

“The fact that they largely are not is really the main finding of this paper. Without coordination, the use of value-based contracts and quality measures at scale leads to many unique measures being used. This may reflect the fact that there are so many important tasks to do in primary care, and there’s no consensus on which ones should be included in quality-based contracts.”

Ronald N. Adler, MD, an associate professor in the Department of Family Medicine and Community Health at UMass Chan Medical School, Worcester, Massachusetts, who’s familiar with the findings but didn’t take part in the research, said the study offers something new — the quantification of quality measures.

He said in an interview that physicians deal with quality measures in different ways. Some clinicians “don’t really care,” and have an attitude of “this is not why I got into medicine.” But others “are very competitive around this, and it leads to a lot of a lot of stress. Trying to address 50-plus measures is impossible and demoralizing.”

The metrics may measure things like mammogram screening that are out of the physician’s control, Dr. Adler said. “I can recommend a mammogram, and my patient can choose not to do it. Or maybe my patient is homeless; she doesn’t have transportation, and it’s not a priority for her, even though she wants to do it.”

Patients may not take medication as prescribed, or they may be unable to afford it, he said. “Can they afford to eat healthy foods? Or is ramen all they can afford, and their sugars are through the roof? There are a lot of factors at play here that are independent of the quality of care provided by the doctor.”

As for his own approach, Dr. Adler said he worries about some measures more than others. “I’m very proactive about screening my patients for colon cancer and maybe a little less so about mammography.”

For colon cancer screening, “there are a lot of benefits and not that many harms as opposed to mammography, which has harms such as false positives and overdiagnosis of breast cancer.”

Dr. Adler is a member of the Quality Measure Alignment Taskforce in Massachusetts, which is trying to establish consensus on appropriate quality measures. But payer participation is voluntary. “Our health systems are too siloed ... so there is no readily available mechanism for enforcing such recommendations.”

Wayne Altman, MD, chair of Family Medicine at Tufts University School of Medicine, Boston, Massachusetts, is also familiar with the study findings but didn’t take part in the research. He said in an interview that clinicians shouldn’t have to deal with more than 5-10 quality measures in total.

He pointed out that many measures don’t make sense in certain populations. Titrating blood pressure to < 140/90 isn’t ideal for elderly patients because aggressive control can send their blood pressure dangerously low. “They’re going to fall down, break a hip, and likely die within a year. You have to have the right population and be aware of unintended consequences.”

Still, Dr. Adler noted, there’s an important role for quality measures in healthcare.

“We need data to inform our quality improvement activities, but they need to be the right measures. People can’t respond reasonably to improve on 50-plus measures,” he said. “They need to be consolidated and prioritized. It would be really helpful if we could have a much lower number of measures that are meaningful, safe, and connect to things that matter.”

No funding has been reported. Dr. Boone disclosed a grant from the National Institute on Aging. Dr. Adler and Dr. Altman had no disclosures.

A version of this article first appeared on Medscape.com.

A new analysis suggests one reason doctors are wary of value-based care arrangements: Overkill.

Researchers found that primary care physicians in one large integrated health system were required to track an average of 57 different quality measures across multiple insurers that linked outcomes to payments under value-based contracts.

Medicare contracts were the most likely to pile quality measures on physicians with an average of 13.42 measures vs 10.07 for commercial insurer contracts and 5.37 for Medicaid contracts, reported Claire Boone, PhD, of the University of Chicago in Illinois and Providence Research Network, Portland, Oregon, and colleagues in JAMA Health Forum. The analysis, which may be the first of its kind, tracked 890 primary care physicians from 2020 to 2022.

The average of 57 quality measures per physician was unexpectedly high, Dr. Boone said in an interview.

“The magnitude of that number surprised us,” Dr. Boone said. “Primary care physicians and their practices have a lot on their plate. Now we know that one of those things is a very large number of different quality metrics to pay attention to, measure, report on, and implement.

Value-based care programs use quality measures to evaluate how well clinicians are doing their jobs and adjust reimbursement accordingly. A payer, for example, may raise reimbursements if a clinician has higher numbers of patients who meet quality measure standards for depression screening or blood pressure.

Dr. Boone said her research group is studying the impact of quality measures and was surprised that data showed individual primary care physicians had to deal with a high number of value-based contracts.

The researchers tracked value-based contracts for 890 physicians (58.3% women, 41.7% men) in an unidentified West Coast Health system. (Several study authors work for the Providence Health System, which serves several Western States and Texas.) The average number of patients per physician was 1309.

The physicians were part of an average of 11.18 value-based contracts (commercial insurers, 49.50%; Medicaid, 21.49%; and Medicare, 29.01%). This number grew from 9.39 in 2020 to 12.26 in 2022. Quality measure data weren’t available for 29% of contracts.

Quality measures were considered unique if they referenced different conditions.

For example, colorectal cancer screening is unique from depression screening. The researchers also considered measures for the same condition unique if the target value differed — for example, blood pressure control defined as < 140/90 vs blood pressure control defined as < 130/80, Dr. Boone said.

Dr. Boone said she expected payers to coordinate quality measures.

“The fact that they largely are not is really the main finding of this paper. Without coordination, the use of value-based contracts and quality measures at scale leads to many unique measures being used. This may reflect the fact that there are so many important tasks to do in primary care, and there’s no consensus on which ones should be included in quality-based contracts.”

Ronald N. Adler, MD, an associate professor in the Department of Family Medicine and Community Health at UMass Chan Medical School, Worcester, Massachusetts, who’s familiar with the findings but didn’t take part in the research, said the study offers something new — the quantification of quality measures.

He said in an interview that physicians deal with quality measures in different ways. Some clinicians “don’t really care,” and have an attitude of “this is not why I got into medicine.” But others “are very competitive around this, and it leads to a lot of a lot of stress. Trying to address 50-plus measures is impossible and demoralizing.”

The metrics may measure things like mammogram screening that are out of the physician’s control, Dr. Adler said. “I can recommend a mammogram, and my patient can choose not to do it. Or maybe my patient is homeless; she doesn’t have transportation, and it’s not a priority for her, even though she wants to do it.”

Patients may not take medication as prescribed, or they may be unable to afford it, he said. “Can they afford to eat healthy foods? Or is ramen all they can afford, and their sugars are through the roof? There are a lot of factors at play here that are independent of the quality of care provided by the doctor.”

As for his own approach, Dr. Adler said he worries about some measures more than others. “I’m very proactive about screening my patients for colon cancer and maybe a little less so about mammography.”

For colon cancer screening, “there are a lot of benefits and not that many harms as opposed to mammography, which has harms such as false positives and overdiagnosis of breast cancer.”

Dr. Adler is a member of the Quality Measure Alignment Taskforce in Massachusetts, which is trying to establish consensus on appropriate quality measures. But payer participation is voluntary. “Our health systems are too siloed ... so there is no readily available mechanism for enforcing such recommendations.”

Wayne Altman, MD, chair of Family Medicine at Tufts University School of Medicine, Boston, Massachusetts, is also familiar with the study findings but didn’t take part in the research. He said in an interview that clinicians shouldn’t have to deal with more than 5-10 quality measures in total.

He pointed out that many measures don’t make sense in certain populations. Titrating blood pressure to < 140/90 isn’t ideal for elderly patients because aggressive control can send their blood pressure dangerously low. “They’re going to fall down, break a hip, and likely die within a year. You have to have the right population and be aware of unintended consequences.”

Still, Dr. Adler noted, there’s an important role for quality measures in healthcare.

“We need data to inform our quality improvement activities, but they need to be the right measures. People can’t respond reasonably to improve on 50-plus measures,” he said. “They need to be consolidated and prioritized. It would be really helpful if we could have a much lower number of measures that are meaningful, safe, and connect to things that matter.”

No funding has been reported. Dr. Boone disclosed a grant from the National Institute on Aging. Dr. Adler and Dr. Altman had no disclosures.

A version of this article first appeared on Medscape.com.

A new analysis suggests one reason doctors are wary of value-based care arrangements: Overkill.

Researchers found that primary care physicians in one large integrated health system were required to track an average of 57 different quality measures across multiple insurers that linked outcomes to payments under value-based contracts.

Medicare contracts were the most likely to pile quality measures on physicians with an average of 13.42 measures vs 10.07 for commercial insurer contracts and 5.37 for Medicaid contracts, reported Claire Boone, PhD, of the University of Chicago in Illinois and Providence Research Network, Portland, Oregon, and colleagues in JAMA Health Forum. The analysis, which may be the first of its kind, tracked 890 primary care physicians from 2020 to 2022.

The average of 57 quality measures per physician was unexpectedly high, Dr. Boone said in an interview.

“The magnitude of that number surprised us,” Dr. Boone said. “Primary care physicians and their practices have a lot on their plate. Now we know that one of those things is a very large number of different quality metrics to pay attention to, measure, report on, and implement.

Value-based care programs use quality measures to evaluate how well clinicians are doing their jobs and adjust reimbursement accordingly. A payer, for example, may raise reimbursements if a clinician has higher numbers of patients who meet quality measure standards for depression screening or blood pressure.

Dr. Boone said her research group is studying the impact of quality measures and was surprised that data showed individual primary care physicians had to deal with a high number of value-based contracts.

The researchers tracked value-based contracts for 890 physicians (58.3% women, 41.7% men) in an unidentified West Coast Health system. (Several study authors work for the Providence Health System, which serves several Western States and Texas.) The average number of patients per physician was 1309.

The physicians were part of an average of 11.18 value-based contracts (commercial insurers, 49.50%; Medicaid, 21.49%; and Medicare, 29.01%). This number grew from 9.39 in 2020 to 12.26 in 2022. Quality measure data weren’t available for 29% of contracts.

Quality measures were considered unique if they referenced different conditions.

For example, colorectal cancer screening is unique from depression screening. The researchers also considered measures for the same condition unique if the target value differed — for example, blood pressure control defined as < 140/90 vs blood pressure control defined as < 130/80, Dr. Boone said.

Dr. Boone said she expected payers to coordinate quality measures.

“The fact that they largely are not is really the main finding of this paper. Without coordination, the use of value-based contracts and quality measures at scale leads to many unique measures being used. This may reflect the fact that there are so many important tasks to do in primary care, and there’s no consensus on which ones should be included in quality-based contracts.”

Ronald N. Adler, MD, an associate professor in the Department of Family Medicine and Community Health at UMass Chan Medical School, Worcester, Massachusetts, who’s familiar with the findings but didn’t take part in the research, said the study offers something new — the quantification of quality measures.

He said in an interview that physicians deal with quality measures in different ways. Some clinicians “don’t really care,” and have an attitude of “this is not why I got into medicine.” But others “are very competitive around this, and it leads to a lot of a lot of stress. Trying to address 50-plus measures is impossible and demoralizing.”

The metrics may measure things like mammogram screening that are out of the physician’s control, Dr. Adler said. “I can recommend a mammogram, and my patient can choose not to do it. Or maybe my patient is homeless; she doesn’t have transportation, and it’s not a priority for her, even though she wants to do it.”

Patients may not take medication as prescribed, or they may be unable to afford it, he said. “Can they afford to eat healthy foods? Or is ramen all they can afford, and their sugars are through the roof? There are a lot of factors at play here that are independent of the quality of care provided by the doctor.”

As for his own approach, Dr. Adler said he worries about some measures more than others. “I’m very proactive about screening my patients for colon cancer and maybe a little less so about mammography.”

For colon cancer screening, “there are a lot of benefits and not that many harms as opposed to mammography, which has harms such as false positives and overdiagnosis of breast cancer.”

Dr. Adler is a member of the Quality Measure Alignment Taskforce in Massachusetts, which is trying to establish consensus on appropriate quality measures. But payer participation is voluntary. “Our health systems are too siloed ... so there is no readily available mechanism for enforcing such recommendations.”

Wayne Altman, MD, chair of Family Medicine at Tufts University School of Medicine, Boston, Massachusetts, is also familiar with the study findings but didn’t take part in the research. He said in an interview that clinicians shouldn’t have to deal with more than 5-10 quality measures in total.

He pointed out that many measures don’t make sense in certain populations. Titrating blood pressure to < 140/90 isn’t ideal for elderly patients because aggressive control can send their blood pressure dangerously low. “They’re going to fall down, break a hip, and likely die within a year. You have to have the right population and be aware of unintended consequences.”

Still, Dr. Adler noted, there’s an important role for quality measures in healthcare.

“We need data to inform our quality improvement activities, but they need to be the right measures. People can’t respond reasonably to improve on 50-plus measures,” he said. “They need to be consolidated and prioritized. It would be really helpful if we could have a much lower number of measures that are meaningful, safe, and connect to things that matter.”

No funding has been reported. Dr. Boone disclosed a grant from the National Institute on Aging. Dr. Adler and Dr. Altman had no disclosures.

A version of this article first appeared on Medscape.com.

TOPLINE: 

Compared with other urine analysis methods, urine Gram stain has a moderate predictive value for detecting gram-negative bacteria in urine culture but does not significantly improve urinary tract infection (UTI) diagnosis in the emergency department (ED).

METHODOLOGY:

• Researchers conducted an observational cohort study at the University Medical Center Groningen in the Netherlands, encompassing 1358 episodes across 1136 patients suspected of having a UTI.
• The study included the following predefined subgroups: patients using urinary catheters and patients with leukopenia (< 4.0×10⁹ leucocytes/L). Urine dipstick nitrite, automated urinalysis, Gram stain, and urine cultures were performed on urine samples collected from patients presenting at the ED.
• The sensitivity and specificity of Gram stain for “many” bacteria (quantified as > 15/high power field) were compared with those of urine dipstick nitrite and automated bacterial counting in urinalysis.

TAKEAWAY:

• The sensitivity and specificity of Gram stain for “many” bacteria were 51.3% and 91.0%, respectively, with an accuracy of 76.8%.
• Gram stain showed a positive predictive value (PPV) of 84.7% for gram-negative rods in urine culture; however, the PPV was only 38.4% for gram-positive cocci.
• In the catheter subgroup, the presence of monomorphic bacteria quantified as “many” had a higher PPV for diagnosing a UTI than the presence of polymorphic bacteria with the same quantification.
• The overall performance of Gram stain in diagnosing a UTI in the ED was comparable to that of automated bacterial counting in urinalysis but better than that of urine dipstick nitrite.

IN PRACTICE:

“With the exception of a moderate prediction of gram-negative bacteria in the UC [urine culture], urine GS [Gram stain] does not improve UTI diagnosis at the ED compared to other urine parameters,” the authors wrote.

SOURCE:

The study was led by Stephanie J.M. Middelkoop, University of Groningen, University Medical Center Groningen, the Netherlands. It was published online on August 16, 2024, in Infectious Diseases.

LIMITATIONS: 

The study’s limitations included a small sample size within the leukopenia subgroup, which may have affected the generalizability of the findings. Additionally, the potential influence of refrigeration of urine samples on bacterial growth could have affected the results. In this study, indwelling catheters were not replaced before urine sample collection, which may have affected the accuracy of UTI diagnosis in patients using catheters.
 

DISCLOSURES:

No conflicts of interest were disclosed by the authors.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article first appeared on Medscape.com.

TOPLINE: 

Compared with other urine analysis methods, urine Gram stain has a moderate predictive value for detecting gram-negative bacteria in urine culture but does not significantly improve urinary tract infection (UTI) diagnosis in the emergency department (ED).

METHODOLOGY:

• Researchers conducted an observational cohort study at the University Medical Center Groningen in the Netherlands, encompassing 1358 episodes across 1136 patients suspected of having a UTI.
• The study included the following predefined subgroups: patients using urinary catheters and patients with leukopenia (< 4.0×10⁹ leucocytes/L). Urine dipstick nitrite, automated urinalysis, Gram stain, and urine cultures were performed on urine samples collected from patients presenting at the ED.
• The sensitivity and specificity of Gram stain for “many” bacteria (quantified as > 15/high power field) were compared with those of urine dipstick nitrite and automated bacterial counting in urinalysis.

TAKEAWAY:

• The sensitivity and specificity of Gram stain for “many” bacteria were 51.3% and 91.0%, respectively, with an accuracy of 76.8%.
• Gram stain showed a positive predictive value (PPV) of 84.7% for gram-negative rods in urine culture; however, the PPV was only 38.4% for gram-positive cocci.
• In the catheter subgroup, the presence of monomorphic bacteria quantified as “many” had a higher PPV for diagnosing a UTI than the presence of polymorphic bacteria with the same quantification.
• The overall performance of Gram stain in diagnosing a UTI in the ED was comparable to that of automated bacterial counting in urinalysis but better than that of urine dipstick nitrite.

IN PRACTICE:

“With the exception of a moderate prediction of gram-negative bacteria in the UC [urine culture], urine GS [Gram stain] does not improve UTI diagnosis at the ED compared to other urine parameters,” the authors wrote.

SOURCE:

The study was led by Stephanie J.M. Middelkoop, University of Groningen, University Medical Center Groningen, the Netherlands. It was published online on August 16, 2024, in Infectious Diseases.

LIMITATIONS: 

The study’s limitations included a small sample size within the leukopenia subgroup, which may have affected the generalizability of the findings. Additionally, the potential influence of refrigeration of urine samples on bacterial growth could have affected the results. In this study, indwelling catheters were not replaced before urine sample collection, which may have affected the accuracy of UTI diagnosis in patients using catheters.
 

DISCLOSURES:

No conflicts of interest were disclosed by the authors.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article first appeared on Medscape.com.

TOPLINE: 

Compared with other urine analysis methods, urine Gram stain has a moderate predictive value for detecting gram-negative bacteria in urine culture but does not significantly improve urinary tract infection (UTI) diagnosis in the emergency department (ED).

METHODOLOGY:

• Researchers conducted an observational cohort study at the University Medical Center Groningen in the Netherlands, encompassing 1358 episodes across 1136 patients suspected of having a UTI.
• The study included the following predefined subgroups: patients using urinary catheters and patients with leukopenia (< 4.0×10⁹ leucocytes/L). Urine dipstick nitrite, automated urinalysis, Gram stain, and urine cultures were performed on urine samples collected from patients presenting at the ED.
• The sensitivity and specificity of Gram stain for “many” bacteria (quantified as > 15/high power field) were compared with those of urine dipstick nitrite and automated bacterial counting in urinalysis.

TAKEAWAY:

• The sensitivity and specificity of Gram stain for “many” bacteria were 51.3% and 91.0%, respectively, with an accuracy of 76.8%.
• Gram stain showed a positive predictive value (PPV) of 84.7% for gram-negative rods in urine culture; however, the PPV was only 38.4% for gram-positive cocci.
• In the catheter subgroup, the presence of monomorphic bacteria quantified as “many” had a higher PPV for diagnosing a UTI than the presence of polymorphic bacteria with the same quantification.
• The overall performance of Gram stain in diagnosing a UTI in the ED was comparable to that of automated bacterial counting in urinalysis but better than that of urine dipstick nitrite.

IN PRACTICE:

“With the exception of a moderate prediction of gram-negative bacteria in the UC [urine culture], urine GS [Gram stain] does not improve UTI diagnosis at the ED compared to other urine parameters,” the authors wrote.

SOURCE:

The study was led by Stephanie J.M. Middelkoop, University of Groningen, University Medical Center Groningen, the Netherlands. It was published online on August 16, 2024, in Infectious Diseases.

LIMITATIONS: 

The study’s limitations included a small sample size within the leukopenia subgroup, which may have affected the generalizability of the findings. Additionally, the potential influence of refrigeration of urine samples on bacterial growth could have affected the results. In this study, indwelling catheters were not replaced before urine sample collection, which may have affected the accuracy of UTI diagnosis in patients using catheters.
 

DISCLOSURES:

No conflicts of interest were disclosed by the authors.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article first appeared on Medscape.com.

This transcript has been edited for clarity.

Hi. I’m Art Caplan. I’m at the Division of Medical Ethics at the NYU Grossman School of Medicine in New York City. 

We have many debates and arguments that are swirling around about the out-of-control costs of Medicare. Many people are arguing we’ve got to trim it and cut back, and many people note that we can’t just go on and on with that kind of expenditure.

People look around for savings. Rightly, we can’t go on with the prices that we’re paying. No system could. We’ll bankrupt ourselves if we don’t drive prices down. 

There’s another area that is driving up cost where, despite the fact that Medicare doesn’t pay for it, we could capture resources and hopefully shift them back to things like Medicare coverage or the insurance of other efficacious procedures. That area is the wellness industry. 

I looked up a number recently, and I was shocked to see that worldwide, $1.8 trillion is being spent on wellness, including billions in the US. Again, Medicare doesn’t pay for that. That’s money coming out of people’s pockets that we could hopefully aim at the payment of things that we know work, not seeing the money drain out to cover bunk, nonsense, and charlatanism.

Does any or most of this stuff work? Do anything? Help anybody? No. We are spending money on charlatans and quacks. The US Food and Drug Administration (FDA), which you might think is the agency that could step in and start to get rid of some of this nonsense, is just too overwhelmed trying to track drugs, devices, and vaccines to give much attention to the wellness industry.

What am I talking about specifically? I’m talking about everything from gut probiotics that are sold in sodas to probiotic facial creams and the Goop industry of Gwyneth Paltrow, where you have people buying things like wellness mats or vaginal eggs that are supposed to maintain gynecologic health.

We’re talking about things like PEMF, or pulse electronic magnetic fields, where you buy a machine and expose yourself to mild magnetic pulses. I went online to look them up, and the machines cost $5000-$50,000. There’s no evidence that it works. By the way, the machines are not only out there as being sold for pain relief and many other things to humans, but also they’re being sold for your pets.

That industry is completely out of control. Wellness interventions, whether it’s transcranial magnetism or all manner of supplements that are sold in health food stores, over and over again, we see a world in which wellness is promoted but no data are introduced to show that any of it helps, works, or does anybody any good.

It may not be all that harmful, but it’s certainly financially toxic to many people who end up spending good amounts of money using these things. I think doctors need to ask patients if they are using any of these things, particularly if they have chronic conditions. They’re likely, many of them, to be seduced by online advertisement to get involved with this stuff because it’s preventive or it’ll help treat some condition that they have. 

The industry is out of control. We’re trying to figure out how to spend money on things we know work in medicine, and yet we continue to tolerate bunk, nonsense, quackery, and charlatanism, just letting it grow and grow and grow in terms of cost.

That’s money that could go elsewhere. That is money that is being taken out of the pockets of patients. They’re doing things that may even delay medical treatment, which won’t really help them, and they are doing things that perhaps might even interfere with medical care that really is known to be beneficial.

I think it’s time to push for more money for the FDA to regulate the wellness side. I think it’s time for the Federal Trade Commission to go after ads that promise health benefits. I think it’s time to have some honest conversations with patients: What are you using? What are you doing? Tell me about it, and here’s why I think you could probably spend your money in a better way. 
 

Dr. Caplan, director, Division of Medical Ethics, New York University Langone Medical Center, New York, disclosed ties with Johnson & Johnson’s Panel for Compassionate Drug Use (unpaid position). He serves as a contributing author and adviser for Medscape.

A version of this article appeared on Medscape.com.

This transcript has been edited for clarity.

Hi. I’m Art Caplan. I’m at the Division of Medical Ethics at the NYU Grossman School of Medicine in New York City. 

We have many debates and arguments that are swirling around about the out-of-control costs of Medicare. Many people are arguing we’ve got to trim it and cut back, and many people note that we can’t just go on and on with that kind of expenditure.

People look around for savings. Rightly, we can’t go on with the prices that we’re paying. No system could. We’ll bankrupt ourselves if we don’t drive prices down. 

There’s another area that is driving up cost where, despite the fact that Medicare doesn’t pay for it, we could capture resources and hopefully shift them back to things like Medicare coverage or the insurance of other efficacious procedures. That area is the wellness industry. 

I looked up a number recently, and I was shocked to see that worldwide, $1.8 trillion is being spent on wellness, including billions in the US. Again, Medicare doesn’t pay for that. That’s money coming out of people’s pockets that we could hopefully aim at the payment of things that we know work, not seeing the money drain out to cover bunk, nonsense, and charlatanism.

Does any or most of this stuff work? Do anything? Help anybody? No. We are spending money on charlatans and quacks. The US Food and Drug Administration (FDA), which you might think is the agency that could step in and start to get rid of some of this nonsense, is just too overwhelmed trying to track drugs, devices, and vaccines to give much attention to the wellness industry.

What am I talking about specifically? I’m talking about everything from gut probiotics that are sold in sodas to probiotic facial creams and the Goop industry of Gwyneth Paltrow, where you have people buying things like wellness mats or vaginal eggs that are supposed to maintain gynecologic health.

We’re talking about things like PEMF, or pulse electronic magnetic fields, where you buy a machine and expose yourself to mild magnetic pulses. I went online to look them up, and the machines cost $5000-$50,000. There’s no evidence that it works. By the way, the machines are not only out there as being sold for pain relief and many other things to humans, but also they’re being sold for your pets.

That industry is completely out of control. Wellness interventions, whether it’s transcranial magnetism or all manner of supplements that are sold in health food stores, over and over again, we see a world in which wellness is promoted but no data are introduced to show that any of it helps, works, or does anybody any good.

It may not be all that harmful, but it’s certainly financially toxic to many people who end up spending good amounts of money using these things. I think doctors need to ask patients if they are using any of these things, particularly if they have chronic conditions. They’re likely, many of them, to be seduced by online advertisement to get involved with this stuff because it’s preventive or it’ll help treat some condition that they have. 

The industry is out of control. We’re trying to figure out how to spend money on things we know work in medicine, and yet we continue to tolerate bunk, nonsense, quackery, and charlatanism, just letting it grow and grow and grow in terms of cost.

That’s money that could go elsewhere. That is money that is being taken out of the pockets of patients. They’re doing things that may even delay medical treatment, which won’t really help them, and they are doing things that perhaps might even interfere with medical care that really is known to be beneficial.

I think it’s time to push for more money for the FDA to regulate the wellness side. I think it’s time for the Federal Trade Commission to go after ads that promise health benefits. I think it’s time to have some honest conversations with patients: What are you using? What are you doing? Tell me about it, and here’s why I think you could probably spend your money in a better way.