CHEST Physician

The question has been lingering for years in medical science circles. Since 2020, when the artificial intelligence (AI) model AlphaFold made it possible to predict protein structures, would the technology open the drug discovery floodgates?

Short answer: No. At least not yet.

The longer answer goes something like this:

A drug target (such as a mutation) is like a lock. The right drug (a protein designed to bind to the mutation, stopping its activity) is the key. But proteins are fidgety and flexible.

“They’re basically molecular springs,” said Gabriel Monteiro da Silva, PhD, a computational chemistry research scientist at Genesis Therapeutics. “Your key can bend and alter the shape of the lock, and if you don’t account for that, your key might fail.”

This is the protein problem in drug development. Another issue making this challenge so vexing is that proteins don’t act in isolation. Their interactions with other proteins, ribonucleic acid, and DNA can affect how they bind to molecules and the shapes they adopt.

Newer versions of AlphaFold, such as AlphaFold Multimer and AlphaFold 3 (the code for which was recently revealed for academic use), can predict many interactions among proteins and between proteins and other molecules. But these tools still have weak points scientists are trying to overcome or work around.

“Those kinds of dynamics and multiple conformations are still quite challenging for the AI models to predict,” said James Zou, PhD, associate professor of biomedical data science at Stanford University in California.

“We’re finding more and more that the only way we can make these structures useful for drug discovery is if we incorporate dynamics, if we incorporate more physics into the model,” said Monteiro da Silva.

Monteiro da Silva spent 3 years during his PhD at Brown University, Providence, Rhode Island, running physics-based simulations in the lab, trying to understand why proteins carrying certain mutations are drug resistant. His results showed how “the changing landscape of shapes that a protein can take” prevented the drug from binding.

It took him 3 years to model just four mutations.

AI can do better — and the struggle is fascinating. By developing models that build on the predictive power of AlphaFold, scientists are uncovering new details about protein activity — insights that can lead to new therapeutics and reveal why existing ones stop working — much faster than they could with traditional methods or AlphaFold alone.

 

New Windows into Protein Dynamics

By predicting protein structural details, AlphaFold models also made it possible to predict pockets where drugs could bind.

A notable step, “but that’s just the starting point,” said Pedro Beltrao, PhD, an associate professor at Institute of Molecular Systems Biology, ETH Zurich in Switzerland. “It’s still very difficult, given a pocket, to actually design the drug or figure out what the pocket binds.”

Going back to the lock-and-key analogy: While he was at Brown, with a team of researchers in the Rubenstein Group, Monteiro da Silva helped create a model to better understand how mutations affect “the shape and dynamics of the lock.” They manipulated the amino acid sequences of proteins, guiding their evolution. This enabled them to use AlphaFold to predict “protein ensembles” and how frequently those ensembles appear. Each ensemble represents the many different shapes a protein can take under given conditions.

“Essentially, it tries to find the most common shapes that a protein will take over an arbitrary amount of time,” Monteiro da Silva said. “If we can predict these ensembles at scale and fast, then we can screen many mutations that cause resistance and develop drugs that will not be affected by that resistance.”

To evaluate their method, the researchers focused on ABL1, a well-studied kinase that causes leukemia. ABL1 can be drugged – unless it carries or develops a mutation that causes drug resistance. Currently there are no drugs that work against proteins carrying those mutations, according to Monteiro da Silva. The researchers used their hybrid AI-meets-physics method to investigate how drugs bind to different ABL1 mutations, screening 100 mutations in just 1 month.

“It’s not going to be perfect for every one of them. But if we have 100 and we get 20 with good accuracy, that’s better than doing four over 3 years,” Monteiro da Silva said.

A forthcoming paper will make their model publicly available in “an easy-to-use graphical interface” that they hope clinicians and medicinal chemists will try out. It can also complement other AI-based tools that dig into protein dynamics, according to Monteiro da Silva.

 

Complementary Tools to Speed Up Discovery 

Another aspect of the protein problem is scale. One protein can interact with hundreds of other proteins, which in turn may interact with hundreds more, all of which comprise the human interactome.

Feixiong Cheng, PhD, helped build PIONEER, a deep learning model that predicts the three-dimensional (3D) structure of interactions between proteins across the interactome.

Most disease mutations disrupt specific interactions between proteins, making their affinity stronger or weaker, explained Cheng. To treat a disease without causing major side effects, scientists need a precise understanding of those interactions.

“From the drug discovery perspective, we cannot just focus on single proteins. We have to understand the protein environment, in particular how the protein interacts with other proteins,” said Cheng, director of Cleveland Clinic Genome Center, Cleveland.

PIONEER helps by blending AlphaFold’s protein structure predictions with next-generation sequencing, a type of genomic research that identifies mutations in the human genome. The model predicts the 3D structure of the places where proteins interact — the binding sites, or interfaces — across the interactome.

“We tell you not only that a binds b, but where on a and where on b the two proteins interact,” said Haiyuan Yu, PhD, director of the Center for Innovative Proteomics, Cornell University, and co-creator of PIONEER.

This can help scientists understand “why a mutation, protein, or even network is a good target for therapeutic discovery,” Cheng said.

The researchers validated PIONEER’s predictions in the lab, testing the impacts of roughly 3000 mutations on 7000 pairs of interacting proteins. Based on their findings, they plan to develop and test treatments for lung and endometrial cancer.

PIONEER can also help scientists home in on how a mutation causes a disease, such as by showing recurrent mutations.

“If you find cancer mutations hitting an interface again and again and again, it means that this is likely to be driving cancer progression,” said Beltrao.

Beltrao’s lab and others have looked for recurrent mutations by using AlphaFold Multimer and AlphaFold 3 to directly model protein interactions. It’s a much slower approach (Pioneer is more than 5000 faster than AlphaFold Multimer, according to Cheng). But it could allow scientists to model interfaces that are not shown by PIONEER.

“You will need many different things to try to come up with a structural modeling of the interactome, and all these will have limitations,” said Beltrao. “Their method is a very good step forward, and there’ll be other approaches that are complementary, to continue to add details.”

 

And It Wouldn’t be an AI Mission Without ChatGPT

Large language models, such as ChatGPT, are another way that scientists are adding details to protein structure predictions. Zou used GPT-4 to “fine tune” a protein language model, called evolutionary scale modeling (ESM-2), which predicts protein structures directly from a protein sequence.

First, they trained ChatGPT on thousands of papers and studies containing information about the functions, biophysical properties, and disease relevance of different mutations. Next, they used the trained model to “teach” ESM-2, boosting its ability “to predict which mutations are likely to have larger effects or smaller effects,” Zou said. The same could be done for a model like AlphaFold, according to Zou.

“They are quite complementary in that the large language model contains a lot more information about the functions and the biophysics of different mutations and proteins as captured in text,” he said, whereas “you can’t give AlphaFold a piece of paper.”

Exactly how AlphaFold makes its predictions is another mystery. “It will somehow learn protein dynamics phenomenologically,” said Monteiro da Silva. He and others are trying to understand how that happens, in hopes of creating even more accurate predictive models. But for the time being, AI-based methods still need assistance from physics.

“The dream is that we achieve a state where we rely on just the fast methods, and they’re accurate enough,” he said. “But we’re so far from that.”

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

The question has been lingering for years in medical science circles. Since 2020, when the artificial intelligence (AI) model AlphaFold made it possible to predict protein structures, would the technology open the drug discovery floodgates?

Short answer: No. At least not yet.

The longer answer goes something like this:

A drug target (such as a mutation) is like a lock. The right drug (a protein designed to bind to the mutation, stopping its activity) is the key. But proteins are fidgety and flexible.

“They’re basically molecular springs,” said Gabriel Monteiro da Silva, PhD, a computational chemistry research scientist at Genesis Therapeutics. “Your key can bend and alter the shape of the lock, and if you don’t account for that, your key might fail.”

This is the protein problem in drug development. Another issue making this challenge so vexing is that proteins don’t act in isolation. Their interactions with other proteins, ribonucleic acid, and DNA can affect how they bind to molecules and the shapes they adopt.

Newer versions of AlphaFold, such as AlphaFold Multimer and AlphaFold 3 (the code for which was recently revealed for academic use), can predict many interactions among proteins and between proteins and other molecules. But these tools still have weak points scientists are trying to overcome or work around.

“Those kinds of dynamics and multiple conformations are still quite challenging for the AI models to predict,” said James Zou, PhD, associate professor of biomedical data science at Stanford University in California.

“We’re finding more and more that the only way we can make these structures useful for drug discovery is if we incorporate dynamics, if we incorporate more physics into the model,” said Monteiro da Silva.

Monteiro da Silva spent 3 years during his PhD at Brown University, Providence, Rhode Island, running physics-based simulations in the lab, trying to understand why proteins carrying certain mutations are drug resistant. His results showed how “the changing landscape of shapes that a protein can take” prevented the drug from binding.

It took him 3 years to model just four mutations.

AI can do better — and the struggle is fascinating. By developing models that build on the predictive power of AlphaFold, scientists are uncovering new details about protein activity — insights that can lead to new therapeutics and reveal why existing ones stop working — much faster than they could with traditional methods or AlphaFold alone.

 

New Windows into Protein Dynamics

By predicting protein structural details, AlphaFold models also made it possible to predict pockets where drugs could bind.

A notable step, “but that’s just the starting point,” said Pedro Beltrao, PhD, an associate professor at Institute of Molecular Systems Biology, ETH Zurich in Switzerland. “It’s still very difficult, given a pocket, to actually design the drug or figure out what the pocket binds.”

Going back to the lock-and-key analogy: While he was at Brown, with a team of researchers in the Rubenstein Group, Monteiro da Silva helped create a model to better understand how mutations affect “the shape and dynamics of the lock.” They manipulated the amino acid sequences of proteins, guiding their evolution. This enabled them to use AlphaFold to predict “protein ensembles” and how frequently those ensembles appear. Each ensemble represents the many different shapes a protein can take under given conditions.

“Essentially, it tries to find the most common shapes that a protein will take over an arbitrary amount of time,” Monteiro da Silva said. “If we can predict these ensembles at scale and fast, then we can screen many mutations that cause resistance and develop drugs that will not be affected by that resistance.”

To evaluate their method, the researchers focused on ABL1, a well-studied kinase that causes leukemia. ABL1 can be drugged – unless it carries or develops a mutation that causes drug resistance. Currently there are no drugs that work against proteins carrying those mutations, according to Monteiro da Silva. The researchers used their hybrid AI-meets-physics method to investigate how drugs bind to different ABL1 mutations, screening 100 mutations in just 1 month.

“It’s not going to be perfect for every one of them. But if we have 100 and we get 20 with good accuracy, that’s better than doing four over 3 years,” Monteiro da Silva said.

A forthcoming paper will make their model publicly available in “an easy-to-use graphical interface” that they hope clinicians and medicinal chemists will try out. It can also complement other AI-based tools that dig into protein dynamics, according to Monteiro da Silva.

 

Complementary Tools to Speed Up Discovery 

Another aspect of the protein problem is scale. One protein can interact with hundreds of other proteins, which in turn may interact with hundreds more, all of which comprise the human interactome.

Feixiong Cheng, PhD, helped build PIONEER, a deep learning model that predicts the three-dimensional (3D) structure of interactions between proteins across the interactome.

Most disease mutations disrupt specific interactions between proteins, making their affinity stronger or weaker, explained Cheng. To treat a disease without causing major side effects, scientists need a precise understanding of those interactions.

“From the drug discovery perspective, we cannot just focus on single proteins. We have to understand the protein environment, in particular how the protein interacts with other proteins,” said Cheng, director of Cleveland Clinic Genome Center, Cleveland.

PIONEER helps by blending AlphaFold’s protein structure predictions with next-generation sequencing, a type of genomic research that identifies mutations in the human genome. The model predicts the 3D structure of the places where proteins interact — the binding sites, or interfaces — across the interactome.

“We tell you not only that a binds b, but where on a and where on b the two proteins interact,” said Haiyuan Yu, PhD, director of the Center for Innovative Proteomics, Cornell University, and co-creator of PIONEER.

This can help scientists understand “why a mutation, protein, or even network is a good target for therapeutic discovery,” Cheng said.

The researchers validated PIONEER’s predictions in the lab, testing the impacts of roughly 3000 mutations on 7000 pairs of interacting proteins. Based on their findings, they plan to develop and test treatments for lung and endometrial cancer.

PIONEER can also help scientists home in on how a mutation causes a disease, such as by showing recurrent mutations.

“If you find cancer mutations hitting an interface again and again and again, it means that this is likely to be driving cancer progression,” said Beltrao.

Beltrao’s lab and others have looked for recurrent mutations by using AlphaFold Multimer and AlphaFold 3 to directly model protein interactions. It’s a much slower approach (Pioneer is more than 5000 faster than AlphaFold Multimer, according to Cheng). But it could allow scientists to model interfaces that are not shown by PIONEER.

“You will need many different things to try to come up with a structural modeling of the interactome, and all these will have limitations,” said Beltrao. “Their method is a very good step forward, and there’ll be other approaches that are complementary, to continue to add details.”

 

And It Wouldn’t be an AI Mission Without ChatGPT

Large language models, such as ChatGPT, are another way that scientists are adding details to protein structure predictions. Zou used GPT-4 to “fine tune” a protein language model, called evolutionary scale modeling (ESM-2), which predicts protein structures directly from a protein sequence.

First, they trained ChatGPT on thousands of papers and studies containing information about the functions, biophysical properties, and disease relevance of different mutations. Next, they used the trained model to “teach” ESM-2, boosting its ability “to predict which mutations are likely to have larger effects or smaller effects,” Zou said. The same could be done for a model like AlphaFold, according to Zou.

“They are quite complementary in that the large language model contains a lot more information about the functions and the biophysics of different mutations and proteins as captured in text,” he said, whereas “you can’t give AlphaFold a piece of paper.”

Exactly how AlphaFold makes its predictions is another mystery. “It will somehow learn protein dynamics phenomenologically,” said Monteiro da Silva. He and others are trying to understand how that happens, in hopes of creating even more accurate predictive models. But for the time being, AI-based methods still need assistance from physics.

“The dream is that we achieve a state where we rely on just the fast methods, and they’re accurate enough,” he said. “But we’re so far from that.”

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

The question has been lingering for years in medical science circles. Since 2020, when the artificial intelligence (AI) model AlphaFold made it possible to predict protein structures, would the technology open the drug discovery floodgates?

Short answer: No. At least not yet.

The longer answer goes something like this:

A drug target (such as a mutation) is like a lock. The right drug (a protein designed to bind to the mutation, stopping its activity) is the key. But proteins are fidgety and flexible.

“They’re basically molecular springs,” said Gabriel Monteiro da Silva, PhD, a computational chemistry research scientist at Genesis Therapeutics. “Your key can bend and alter the shape of the lock, and if you don’t account for that, your key might fail.”

This is the protein problem in drug development. Another issue making this challenge so vexing is that proteins don’t act in isolation. Their interactions with other proteins, ribonucleic acid, and DNA can affect how they bind to molecules and the shapes they adopt.

Newer versions of AlphaFold, such as AlphaFold Multimer and AlphaFold 3 (the code for which was recently revealed for academic use), can predict many interactions among proteins and between proteins and other molecules. But these tools still have weak points scientists are trying to overcome or work around.

“Those kinds of dynamics and multiple conformations are still quite challenging for the AI models to predict,” said James Zou, PhD, associate professor of biomedical data science at Stanford University in California.

“We’re finding more and more that the only way we can make these structures useful for drug discovery is if we incorporate dynamics, if we incorporate more physics into the model,” said Monteiro da Silva.

Monteiro da Silva spent 3 years during his PhD at Brown University, Providence, Rhode Island, running physics-based simulations in the lab, trying to understand why proteins carrying certain mutations are drug resistant. His results showed how “the changing landscape of shapes that a protein can take” prevented the drug from binding.

It took him 3 years to model just four mutations.

AI can do better — and the struggle is fascinating. By developing models that build on the predictive power of AlphaFold, scientists are uncovering new details about protein activity — insights that can lead to new therapeutics and reveal why existing ones stop working — much faster than they could with traditional methods or AlphaFold alone.

 

New Windows into Protein Dynamics

By predicting protein structural details, AlphaFold models also made it possible to predict pockets where drugs could bind.

A notable step, “but that’s just the starting point,” said Pedro Beltrao, PhD, an associate professor at Institute of Molecular Systems Biology, ETH Zurich in Switzerland. “It’s still very difficult, given a pocket, to actually design the drug or figure out what the pocket binds.”

Going back to the lock-and-key analogy: While he was at Brown, with a team of researchers in the Rubenstein Group, Monteiro da Silva helped create a model to better understand how mutations affect “the shape and dynamics of the lock.” They manipulated the amino acid sequences of proteins, guiding their evolution. This enabled them to use AlphaFold to predict “protein ensembles” and how frequently those ensembles appear. Each ensemble represents the many different shapes a protein can take under given conditions.

“Essentially, it tries to find the most common shapes that a protein will take over an arbitrary amount of time,” Monteiro da Silva said. “If we can predict these ensembles at scale and fast, then we can screen many mutations that cause resistance and develop drugs that will not be affected by that resistance.”

To evaluate their method, the researchers focused on ABL1, a well-studied kinase that causes leukemia. ABL1 can be drugged – unless it carries or develops a mutation that causes drug resistance. Currently there are no drugs that work against proteins carrying those mutations, according to Monteiro da Silva. The researchers used their hybrid AI-meets-physics method to investigate how drugs bind to different ABL1 mutations, screening 100 mutations in just 1 month.

“It’s not going to be perfect for every one of them. But if we have 100 and we get 20 with good accuracy, that’s better than doing four over 3 years,” Monteiro da Silva said.

A forthcoming paper will make their model publicly available in “an easy-to-use graphical interface” that they hope clinicians and medicinal chemists will try out. It can also complement other AI-based tools that dig into protein dynamics, according to Monteiro da Silva.

 

Complementary Tools to Speed Up Discovery 

Another aspect of the protein problem is scale. One protein can interact with hundreds of other proteins, which in turn may interact with hundreds more, all of which comprise the human interactome.

Feixiong Cheng, PhD, helped build PIONEER, a deep learning model that predicts the three-dimensional (3D) structure of interactions between proteins across the interactome.

Most disease mutations disrupt specific interactions between proteins, making their affinity stronger or weaker, explained Cheng. To treat a disease without causing major side effects, scientists need a precise understanding of those interactions.

“From the drug discovery perspective, we cannot just focus on single proteins. We have to understand the protein environment, in particular how the protein interacts with other proteins,” said Cheng, director of Cleveland Clinic Genome Center, Cleveland.

PIONEER helps by blending AlphaFold’s protein structure predictions with next-generation sequencing, a type of genomic research that identifies mutations in the human genome. The model predicts the 3D structure of the places where proteins interact — the binding sites, or interfaces — across the interactome.

“We tell you not only that a binds b, but where on a and where on b the two proteins interact,” said Haiyuan Yu, PhD, director of the Center for Innovative Proteomics, Cornell University, and co-creator of PIONEER.

This can help scientists understand “why a mutation, protein, or even network is a good target for therapeutic discovery,” Cheng said.

The researchers validated PIONEER’s predictions in the lab, testing the impacts of roughly 3000 mutations on 7000 pairs of interacting proteins. Based on their findings, they plan to develop and test treatments for lung and endometrial cancer.

PIONEER can also help scientists home in on how a mutation causes a disease, such as by showing recurrent mutations.

“If you find cancer mutations hitting an interface again and again and again, it means that this is likely to be driving cancer progression,” said Beltrao.

Beltrao’s lab and others have looked for recurrent mutations by using AlphaFold Multimer and AlphaFold 3 to directly model protein interactions. It’s a much slower approach (Pioneer is more than 5000 faster than AlphaFold Multimer, according to Cheng). But it could allow scientists to model interfaces that are not shown by PIONEER.

“You will need many different things to try to come up with a structural modeling of the interactome, and all these will have limitations,” said Beltrao. “Their method is a very good step forward, and there’ll be other approaches that are complementary, to continue to add details.”

 

And It Wouldn’t be an AI Mission Without ChatGPT

Large language models, such as ChatGPT, are another way that scientists are adding details to protein structure predictions. Zou used GPT-4 to “fine tune” a protein language model, called evolutionary scale modeling (ESM-2), which predicts protein structures directly from a protein sequence.

First, they trained ChatGPT on thousands of papers and studies containing information about the functions, biophysical properties, and disease relevance of different mutations. Next, they used the trained model to “teach” ESM-2, boosting its ability “to predict which mutations are likely to have larger effects or smaller effects,” Zou said. The same could be done for a model like AlphaFold, according to Zou.

“They are quite complementary in that the large language model contains a lot more information about the functions and the biophysics of different mutations and proteins as captured in text,” he said, whereas “you can’t give AlphaFold a piece of paper.”

Exactly how AlphaFold makes its predictions is another mystery. “It will somehow learn protein dynamics phenomenologically,” said Monteiro da Silva. He and others are trying to understand how that happens, in hopes of creating even more accurate predictive models. But for the time being, AI-based methods still need assistance from physics.

“The dream is that we achieve a state where we rely on just the fast methods, and they’re accurate enough,” he said. “But we’re so far from that.”

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

During a recent walk with my 6-year-old, he told me he smelled marijuana. His comment speaks to its increased (and more open) use since legalization in our state. The macho, misguided part of my dad psyche was proud of his “street cred” but the thinking part of my brain was concerned. He seemed a little young for a talk about drugs. 

I was able to provide a simple, watered-down list of reasons why he shouldn’t smoke marijuana or anything else. The “drugs are bad” aphorism sufficed for my 6-year-old but wasn’t worthy of an academic pulmonologist.

I retired from the military 2 years ago, so marijuana (I’m using the terms “marijuana” and “cannabis” interchangeably here) knowledge wasn’t required for regular practice. I recall one 60-year-old patient who reported smoking four joints a day for years. He had marked emphysema on CT, severe obstruction on spirometry, and he was functionally limited. Buttressed by scattered reports of acute lung injury caused by dabbing or marijuana vaping, this anecdotal “n of 1” led to a predictably pedantic conclusion: Smoking marijuana is bad for the lungs and preaching cessation is worth my time and effort. 

I now work in an inner-city hospital. My 6-year-old could identify the smell permeating the hallways and clinic rooms. I’ve reverted to counseling cessation using little more than my “drugs are bad” speech. When I came across a recent review in Seminars in Respiratory and Critical Care Medicine, I recognized the opportunity to read and do better. This summary is based heavily on that review.

Spoiler alert: The data aren’t great. By federal law, marijuana has been illegal in the United States since 1970, so neither funding nor recruitment has come easy. There’s lots of observational data that depend on self-report and are confounded by cigarette use. A lack of regulation results in variations in composition and concentration. In summary, though, smoking marijuana is associated with changes to the bronchial tree and respiratory symptoms, similar to those seen with chronic bronchitis. These symptoms improve with cessation. 

The relationship between marijuana and airflow obstruction and lung function is complicated. A mix of contradictory data shows a reduction in the ratio of the forced expiratory volume in the first 1 second to the forced vital capacity (FEV₁/FVC), an increase in FVC, and changes in conductance. 

Biologic plausibility, essential to bolster causality but easy to manufacture, seems intuitive for the airway changes (decreased FEV₁/FVC and conductance). The increase in FVC, explained by either the anti-inflammatory properties of delta-9-tetrahydrocannabinol (THC) or the impact from deep inhalations typical of marijuana use, is more difficult to understand. Regardless, I came away from the review less confident about marijuana’s impact on lung structure and function. 

The Seminars review also explores marijuana’s association with lung cancer, emphysema, and other structural changes seen on CT of the chest. There’s certainly noise here but the data at present are underwhelming. 

This all speaks to the general misconception I’ve had, perhaps shared by others, that the well-defined effects on the lung from tobacco abuse can be extrapolated to marijuana. In the past, I’d even gone so far as to equate a pack-year (smoking one pack of cigarettes per day for a year) to a joint-year (smoking one joint per day for a year), a rather dramatic oversimplification. While both are attempts to quantify exposure, the latter connotes far less information. The content of a joint can vary considerably in ways that the content of cigarettes does not, and there have been no formal studies of the comparative impact on the lung. 

 

Final Thoughts

The nuance here matters for several reasons. Legalization means an increase in use and presumably more open reporting by patients. In a vacuum, it seems reasonable to council cessation to reduce symptoms and because additional harms can be assumed, given what we know about smoke inhalation in general. Will cessation drive patients to an increase in tobacco use where harm is better established? 

Given its mixed effects on lung function, is it worth spending behavior change capital, the most precious of patient commodities, on marijuana counseling? Marijuana has numerous effects outside the lung that haven’t been touched on here. How should those be incorporated into our guidance? Legalization and regulation provide the opportunity to obtain the better data that are sorely needed.

Aaron B. Holley, MD, is a professor of medicine at Uniformed Services University in Bethesda, Maryland, and a pulmonary/sleep and critical care medicine physician at MedStar Washington Hospital Center in Washington, DC. He has disclosed the relevant financial relationships with Metapharm, CHEST College, and WebMD.

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

During a recent walk with my 6-year-old, he told me he smelled marijuana. His comment speaks to its increased (and more open) use since legalization in our state. The macho, misguided part of my dad psyche was proud of his “street cred” but the thinking part of my brain was concerned. He seemed a little young for a talk about drugs. 

I was able to provide a simple, watered-down list of reasons why he shouldn’t smoke marijuana or anything else. The “drugs are bad” aphorism sufficed for my 6-year-old but wasn’t worthy of an academic pulmonologist.

I retired from the military 2 years ago, so marijuana (I’m using the terms “marijuana” and “cannabis” interchangeably here) knowledge wasn’t required for regular practice. I recall one 60-year-old patient who reported smoking four joints a day for years. He had marked emphysema on CT, severe obstruction on spirometry, and he was functionally limited. Buttressed by scattered reports of acute lung injury caused by dabbing or marijuana vaping, this anecdotal “n of 1” led to a predictably pedantic conclusion: Smoking marijuana is bad for the lungs and preaching cessation is worth my time and effort. 

I now work in an inner-city hospital. My 6-year-old could identify the smell permeating the hallways and clinic rooms. I’ve reverted to counseling cessation using little more than my “drugs are bad” speech. When I came across a recent review in Seminars in Respiratory and Critical Care Medicine, I recognized the opportunity to read and do better. This summary is based heavily on that review.

Spoiler alert: The data aren’t great. By federal law, marijuana has been illegal in the United States since 1970, so neither funding nor recruitment has come easy. There’s lots of observational data that depend on self-report and are confounded by cigarette use. A lack of regulation results in variations in composition and concentration. In summary, though, smoking marijuana is associated with changes to the bronchial tree and respiratory symptoms, similar to those seen with chronic bronchitis. These symptoms improve with cessation. 

The relationship between marijuana and airflow obstruction and lung function is complicated. A mix of contradictory data shows a reduction in the ratio of the forced expiratory volume in the first 1 second to the forced vital capacity (FEV₁/FVC), an increase in FVC, and changes in conductance. 

Biologic plausibility, essential to bolster causality but easy to manufacture, seems intuitive for the airway changes (decreased FEV₁/FVC and conductance). The increase in FVC, explained by either the anti-inflammatory properties of delta-9-tetrahydrocannabinol (THC) or the impact from deep inhalations typical of marijuana use, is more difficult to understand. Regardless, I came away from the review less confident about marijuana’s impact on lung structure and function. 

The Seminars review also explores marijuana’s association with lung cancer, emphysema, and other structural changes seen on CT of the chest. There’s certainly noise here but the data at present are underwhelming. 

This all speaks to the general misconception I’ve had, perhaps shared by others, that the well-defined effects on the lung from tobacco abuse can be extrapolated to marijuana. In the past, I’d even gone so far as to equate a pack-year (smoking one pack of cigarettes per day for a year) to a joint-year (smoking one joint per day for a year), a rather dramatic oversimplification. While both are attempts to quantify exposure, the latter connotes far less information. The content of a joint can vary considerably in ways that the content of cigarettes does not, and there have been no formal studies of the comparative impact on the lung. 

 

Final Thoughts

The nuance here matters for several reasons. Legalization means an increase in use and presumably more open reporting by patients. In a vacuum, it seems reasonable to council cessation to reduce symptoms and because additional harms can be assumed, given what we know about smoke inhalation in general. Will cessation drive patients to an increase in tobacco use where harm is better established? 

Given its mixed effects on lung function, is it worth spending behavior change capital, the most precious of patient commodities, on marijuana counseling? Marijuana has numerous effects outside the lung that haven’t been touched on here. How should those be incorporated into our guidance? Legalization and regulation provide the opportunity to obtain the better data that are sorely needed.

Aaron B. Holley, MD, is a professor of medicine at Uniformed Services University in Bethesda, Maryland, and a pulmonary/sleep and critical care medicine physician at MedStar Washington Hospital Center in Washington, DC. He has disclosed the relevant financial relationships with Metapharm, CHEST College, and WebMD.

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

During a recent walk with my 6-year-old, he told me he smelled marijuana. His comment speaks to its increased (and more open) use since legalization in our state. The macho, misguided part of my dad psyche was proud of his “street cred” but the thinking part of my brain was concerned. He seemed a little young for a talk about drugs. 

I was able to provide a simple, watered-down list of reasons why he shouldn’t smoke marijuana or anything else. The “drugs are bad” aphorism sufficed for my 6-year-old but wasn’t worthy of an academic pulmonologist.

I retired from the military 2 years ago, so marijuana (I’m using the terms “marijuana” and “cannabis” interchangeably here) knowledge wasn’t required for regular practice. I recall one 60-year-old patient who reported smoking four joints a day for years. He had marked emphysema on CT, severe obstruction on spirometry, and he was functionally limited. Buttressed by scattered reports of acute lung injury caused by dabbing or marijuana vaping, this anecdotal “n of 1” led to a predictably pedantic conclusion: Smoking marijuana is bad for the lungs and preaching cessation is worth my time and effort. 

I now work in an inner-city hospital. My 6-year-old could identify the smell permeating the hallways and clinic rooms. I’ve reverted to counseling cessation using little more than my “drugs are bad” speech. When I came across a recent review in Seminars in Respiratory and Critical Care Medicine, I recognized the opportunity to read and do better. This summary is based heavily on that review.

Spoiler alert: The data aren’t great. By federal law, marijuana has been illegal in the United States since 1970, so neither funding nor recruitment has come easy. There’s lots of observational data that depend on self-report and are confounded by cigarette use. A lack of regulation results in variations in composition and concentration. In summary, though, smoking marijuana is associated with changes to the bronchial tree and respiratory symptoms, similar to those seen with chronic bronchitis. These symptoms improve with cessation. 

The relationship between marijuana and airflow obstruction and lung function is complicated. A mix of contradictory data shows a reduction in the ratio of the forced expiratory volume in the first 1 second to the forced vital capacity (FEV₁/FVC), an increase in FVC, and changes in conductance. 

Biologic plausibility, essential to bolster causality but easy to manufacture, seems intuitive for the airway changes (decreased FEV₁/FVC and conductance). The increase in FVC, explained by either the anti-inflammatory properties of delta-9-tetrahydrocannabinol (THC) or the impact from deep inhalations typical of marijuana use, is more difficult to understand. Regardless, I came away from the review less confident about marijuana’s impact on lung structure and function. 

The Seminars review also explores marijuana’s association with lung cancer, emphysema, and other structural changes seen on CT of the chest. There’s certainly noise here but the data at present are underwhelming. 

This all speaks to the general misconception I’ve had, perhaps shared by others, that the well-defined effects on the lung from tobacco abuse can be extrapolated to marijuana. In the past, I’d even gone so far as to equate a pack-year (smoking one pack of cigarettes per day for a year) to a joint-year (smoking one joint per day for a year), a rather dramatic oversimplification. While both are attempts to quantify exposure, the latter connotes far less information. The content of a joint can vary considerably in ways that the content of cigarettes does not, and there have been no formal studies of the comparative impact on the lung. 

 

Final Thoughts

The nuance here matters for several reasons. Legalization means an increase in use and presumably more open reporting by patients. In a vacuum, it seems reasonable to council cessation to reduce symptoms and because additional harms can be assumed, given what we know about smoke inhalation in general. Will cessation drive patients to an increase in tobacco use where harm is better established? 

Given its mixed effects on lung function, is it worth spending behavior change capital, the most precious of patient commodities, on marijuana counseling? Marijuana has numerous effects outside the lung that haven’t been touched on here. How should those be incorporated into our guidance? Legalization and regulation provide the opportunity to obtain the better data that are sorely needed.

Aaron B. Holley, MD, is a professor of medicine at Uniformed Services University in Bethesda, Maryland, and a pulmonary/sleep and critical care medicine physician at MedStar Washington Hospital Center in Washington, DC. He has disclosed the relevant financial relationships with Metapharm, CHEST College, and WebMD.

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

This transcript has been edited for clarity. 

I would like to briefly talk about a very interesting paper and one that probably has about as much to inform the doctor-patient relationship as any paper you can think of. 

The title itself gives you a little bit of that answer before I even discuss the outcome. The paper is “The Benefits of Quitting Smoking at Different Ages,” recently published in The American Journal of Preventive Medicine.

I’m not going to even begin to attempt to explore the statistics of the analysis, but I think the conclusions are both fascinating and important. I will read the first sentence of the results and then just comment on some of the others because there’s just so much data here and I really want to focus on the punchline. 

The results section said that, compared with people who never smoked, those who smoke currently, aged 35, 45, 55, 65, or 75, (those were all the groups they looked at), and who have smoked throughout adulthood until that age will lose an average of 9.1, 8.3, 7.3, 5.9, and 4.4 years of life, respectively — obviously, it’s a lot — if they continue to smoke for the rest of their lives. 

If somebody is smoking at age 35 and they continue to smoke, they could lose 9 years of life on average. We know that. It’s terrible. That’s why people should never smoke. Period. End of story. There’s no social value. There’s no health value of smoking. It’s a deadly recreational activity for multiple illnesses, and obviously, cancer is prominent among them.

Here’s the conclusion of the paper that I think is interesting. That doctor, whether it’s a primary care doctor, an oncologist, an ob/gyn, or a family doctor, is seeing Mr Smith or Mrs Jones in the office today, whether they know that patient well or not very well, and they’re still smoking. However, if the person we’re describing here quits smoking at these ages, how much life do they add back, compared with if they continued?

They may say: “Oh, I’ve been smoking all my life. What difference does it make? The die is cast.” Wrong! If you’ve been smoking your whole adult life — so let’s just say that you started at age 18, age 20, age 15, or even age 12 — but you quit smoking at the age of 35, you’re going to add 8 years of life on average. If you quit smoking when you’re 65, having smoked your whole adult life, you will add 1.7 years of life. That’s 1.7 years to be with your family, to be with your grandchildren, and enjoy life. If you ask, “Oh, what difference does it make?” It makes a big difference. 

I’ll share another statistic and I’ll be done. I think this is really an interesting one. The chances of gaining at least a year of life among those who quit smoking at the age of 65 was 23.4%. There is a 1 out of 4 chance that you’re going to live an additional year if you stop at age 65. Even if you stop smoking at age 75, you have a 14% chance of living at least an additional year longer than you would have if you didn’t stop smoking. 

There is much to think about here, much to consider, and much to discuss potentially with patients.

Dr. Markman is Professor of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center; President, Medicine & Science, City of Hope Atlanta, Chicago, Phoenix. He reported conflicts of interest with GlaxoSmithKline and AstraZeneca.

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

This transcript has been edited for clarity. 

I would like to briefly talk about a very interesting paper and one that probably has about as much to inform the doctor-patient relationship as any paper you can think of. 

The title itself gives you a little bit of that answer before I even discuss the outcome. The paper is “The Benefits of Quitting Smoking at Different Ages,” recently published in The American Journal of Preventive Medicine.

I’m not going to even begin to attempt to explore the statistics of the analysis, but I think the conclusions are both fascinating and important. I will read the first sentence of the results and then just comment on some of the others because there’s just so much data here and I really want to focus on the punchline. 

The results section said that, compared with people who never smoked, those who smoke currently, aged 35, 45, 55, 65, or 75, (those were all the groups they looked at), and who have smoked throughout adulthood until that age will lose an average of 9.1, 8.3, 7.3, 5.9, and 4.4 years of life, respectively — obviously, it’s a lot — if they continue to smoke for the rest of their lives. 

If somebody is smoking at age 35 and they continue to smoke, they could lose 9 years of life on average. We know that. It’s terrible. That’s why people should never smoke. Period. End of story. There’s no social value. There’s no health value of smoking. It’s a deadly recreational activity for multiple illnesses, and obviously, cancer is prominent among them.

Here’s the conclusion of the paper that I think is interesting. That doctor, whether it’s a primary care doctor, an oncologist, an ob/gyn, or a family doctor, is seeing Mr Smith or Mrs Jones in the office today, whether they know that patient well or not very well, and they’re still smoking. However, if the person we’re describing here quits smoking at these ages, how much life do they add back, compared with if they continued?

They may say: “Oh, I’ve been smoking all my life. What difference does it make? The die is cast.” Wrong! If you’ve been smoking your whole adult life — so let’s just say that you started at age 18, age 20, age 15, or even age 12 — but you quit smoking at the age of 35, you’re going to add 8 years of life on average. If you quit smoking when you’re 65, having smoked your whole adult life, you will add 1.7 years of life. That’s 1.7 years to be with your family, to be with your grandchildren, and enjoy life. If you ask, “Oh, what difference does it make?” It makes a big difference. 

I’ll share another statistic and I’ll be done. I think this is really an interesting one. The chances of gaining at least a year of life among those who quit smoking at the age of 65 was 23.4%. There is a 1 out of 4 chance that you’re going to live an additional year if you stop at age 65. Even if you stop smoking at age 75, you have a 14% chance of living at least an additional year longer than you would have if you didn’t stop smoking. 

There is much to think about here, much to consider, and much to discuss potentially with patients.

Dr. Markman is Professor of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center; President, Medicine & Science, City of Hope Atlanta, Chicago, Phoenix. He reported conflicts of interest with GlaxoSmithKline and AstraZeneca.

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

This transcript has been edited for clarity. 

I would like to briefly talk about a very interesting paper and one that probably has about as much to inform the doctor-patient relationship as any paper you can think of. 

The title itself gives you a little bit of that answer before I even discuss the outcome. The paper is “The Benefits of Quitting Smoking at Different Ages,” recently published in The American Journal of Preventive Medicine.

I’m not going to even begin to attempt to explore the statistics of the analysis, but I think the conclusions are both fascinating and important. I will read the first sentence of the results and then just comment on some of the others because there’s just so much data here and I really want to focus on the punchline. 

The results section said that, compared with people who never smoked, those who smoke currently, aged 35, 45, 55, 65, or 75, (those were all the groups they looked at), and who have smoked throughout adulthood until that age will lose an average of 9.1, 8.3, 7.3, 5.9, and 4.4 years of life, respectively — obviously, it’s a lot — if they continue to smoke for the rest of their lives. 

If somebody is smoking at age 35 and they continue to smoke, they could lose 9 years of life on average. We know that. It’s terrible. That’s why people should never smoke. Period. End of story. There’s no social value. There’s no health value of smoking. It’s a deadly recreational activity for multiple illnesses, and obviously, cancer is prominent among them.

Here’s the conclusion of the paper that I think is interesting. That doctor, whether it’s a primary care doctor, an oncologist, an ob/gyn, or a family doctor, is seeing Mr Smith or Mrs Jones in the office today, whether they know that patient well or not very well, and they’re still smoking. However, if the person we’re describing here quits smoking at these ages, how much life do they add back, compared with if they continued?

They may say: “Oh, I’ve been smoking all my life. What difference does it make? The die is cast.” Wrong! If you’ve been smoking your whole adult life — so let’s just say that you started at age 18, age 20, age 15, or even age 12 — but you quit smoking at the age of 35, you’re going to add 8 years of life on average. If you quit smoking when you’re 65, having smoked your whole adult life, you will add 1.7 years of life. That’s 1.7 years to be with your family, to be with your grandchildren, and enjoy life. If you ask, “Oh, what difference does it make?” It makes a big difference. 

I’ll share another statistic and I’ll be done. I think this is really an interesting one. The chances of gaining at least a year of life among those who quit smoking at the age of 65 was 23.4%. There is a 1 out of 4 chance that you’re going to live an additional year if you stop at age 65. Even if you stop smoking at age 75, you have a 14% chance of living at least an additional year longer than you would have if you didn’t stop smoking. 

There is much to think about here, much to consider, and much to discuss potentially with patients.

Dr. Markman is Professor of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center; President, Medicine & Science, City of Hope Atlanta, Chicago, Phoenix. He reported conflicts of interest with GlaxoSmithKline and AstraZeneca.

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

In preparation for a future international treaty aimed at reducing plastic pollution, the French Parliamentary Office for the Evaluation of Scientific and Technological Choices presented the conclusions of a public hearing on the impact of plastics on various aspects of human health.

Increased Global Plastic Production

Philippe Bolo, a member of the French Democratic Party and the rapporteur for the public mission on the health impacts of plastics, spoke about the latest round of treaty negotiations, held from November 25 to December 1 in South Korea, attended by leading French and global experts about the impact of plastics on human health.

The hearing highlighted a sharp increase in plastic production. “It has doubled in the last 20 years and is expected to exceed 500 million tons in 2024,” Bolo said. This is about 60 kg per person. According to projections from the Organization for Economic Co-operation and Development, on its current trajectory, plastic production will reach 750 million tons by 2040 and surpass 1 billion tons before 2050, he said.

 

Minimal Plastic Waste Recycling

Around one third (32%) of plastics are used for packaging. “Therefore, most plastic production is still intended for single-use purposes,” he said. Plastic waste follows a similar growth trajectory, with volumes expected to rise from 360 million tons in 2020 to 617 million tons by 2040 unless action is taken. Very little of this waste is recycled, even in the most countries that are most advanced in terms of collection, sorting, and processing.

In France, for example, in 2018, only 0.6 million tons of the 3.6 million tons of plastic waste produced was truly recycled. This is less than one fifth (17%). Globally, less than 10% of plastic waste is recycled. In 2020, plastic waste that ended up in the environment represented 81 million tons, or 22% of the total. “Beyond waste, this leads to pollution by microplastics and nanoplastics, resulting from their fragmentation. All environments are affected: Seas, rivers, soils, air, and even living organisms,” Bolo said.

 

Methodological Challenges

However, measuring the impact of plastics on health faces methodological difficulties due to the wide variety of composition, size, and shape of plastics. Nevertheless, the French Standardization Association (Association Française de Normalisation) has conducted work to establish a characterization standard for microplastics in water, which serves as an international reference.

“It is also very difficult to know what we are ingesting,” Bolo said. “A study conducted in 2019 estimated that the average human absorbs 5 grams of plastics per week, the equivalent of a credit card.» Since then, other studies have revised this estimate downward, but no consensus has been reached.

A recent study across 109 countries, both industrialized and developing, found significant exposure, estimated at 500 mg/d, particularly in Southeast Asian countries, where it was due mainly to seafood consumption.

A study concluded that plastic water bottles contain 240,000 particles per liter, 90% of which are nanoplastics. These nanoparticles can pass through the intestinal barrier to enter the bloodstream and reach several organs including the heart, brain, and placenta, as well as the fetus.

 

Changes to the Microbiome

Microplastics also accumulate in organs. Thus, the amount of plastic in the lungs increases with age, suggesting that particles may persist in the body without being eliminated. The health consequences of this are still poorly understood, but exposure to plastics appears to cause changes in the composition of the intestinal microbiota. Pathobionts (commensal bacteria with harmful potential) have been found in both adults and children, which could contribute to dysbiosis of the gut microbiome. Furthermore, a decrease in butyrate, a short-chain fatty acid beneficial to health, has been observed in children’s intestines.

Inhaled nanoplastics may disrupt the mucociliary clearance mechanisms of the respiratory system. The toxicity of inhaled plastic particles was demonstrated as early as the 1970s among workers in the flocking industry. Some developed lung function impairments, shortness of breath, inflammation, fibrosis, and even lung cancer. Similar symptoms have been observed in workers in the textile and polyvinyl chloride industries.

A study published recently in The New England Journal of Medicine measured the amount of microplastics collected from carotid plaque of more than 300 patients who had undergone carotid endarterectomy for asymptomatic carotid artery disease. It found a 4.53 times higher risk for the primary endpoint, a composite of myocardial infarction, stroke, and all-cause mortality, among individuals with microplastics and nanoplastics in plaque compared with those without.

 

Health Affects High

The danger of plastics is also directly linked to the chemical substances they contain. A general scientific review looked at the health impacts of three chemicals used almost exclusively in plastics: Polybromodiphenyl ethers (PBDEs), used as flame retardants in textiles or electronics; bisphenol A (BPA), used in the lining of cans and bottles; and phthalates, particularly diethylhexyl phthalate (DEHP), used to make plastics more flexible.

The review highlighted strong epidemiological evidence linking fetal exposure to PBDEs during pregnancy to low birth weight and later exposure to delayed or impaired cognitive development in children and even a loss of IQ. Statistically significant evidence of disruption of thyroid function in adults was also found.

BPA is linked to genital malformations in female newborns exposed to BPA in utero, type 2 diabetes in adults, insulin resistance, and polycystic ovary syndrome in women. BPA exposure also increases the risk for obesity and hypertension in both children and adults, as well as the risk for cardiovascular disease in adults.

Finally, the review established links between exposure to DEHP and miscarriages, genital malformations in male newborns, delayed or impaired cognitive development in children, loss of IQ, delayed psychomotor development, early puberty in young girls, and endometriosis in young women. DEHP exposure also has multiple effects on cardiometabolic health, including insulin resistance, obesity, and elevated blood pressure.

The economic costs associated with the health impacts of these three substances have been estimated at $675 billion in the United States.

Bolo said that the solution to this plastic pollution is necessarily international. “We need an ambitious and legally binding treaty to reduce plastic production,” he said. “The damage is already done; we need to act to protect human health,” he concluded. The parliamentary office has made nine recommendations to the treaty negotiators.

This story was translated from Medscape’s French edition using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article appeared on Medscape.com.

In preparation for a future international treaty aimed at reducing plastic pollution, the French Parliamentary Office for the Evaluation of Scientific and Technological Choices presented the conclusions of a public hearing on the impact of plastics on various aspects of human health.

Increased Global Plastic Production

Philippe Bolo, a member of the French Democratic Party and the rapporteur for the public mission on the health impacts of plastics, spoke about the latest round of treaty negotiations, held from November 25 to December 1 in South Korea, attended by leading French and global experts about the impact of plastics on human health.

The hearing highlighted a sharp increase in plastic production. “It has doubled in the last 20 years and is expected to exceed 500 million tons in 2024,” Bolo said. This is about 60 kg per person. According to projections from the Organization for Economic Co-operation and Development, on its current trajectory, plastic production will reach 750 million tons by 2040 and surpass 1 billion tons before 2050, he said.

 

Minimal Plastic Waste Recycling

Around one third (32%) of plastics are used for packaging. “Therefore, most plastic production is still intended for single-use purposes,” he said. Plastic waste follows a similar growth trajectory, with volumes expected to rise from 360 million tons in 2020 to 617 million tons by 2040 unless action is taken. Very little of this waste is recycled, even in the most countries that are most advanced in terms of collection, sorting, and processing.

In France, for example, in 2018, only 0.6 million tons of the 3.6 million tons of plastic waste produced was truly recycled. This is less than one fifth (17%). Globally, less than 10% of plastic waste is recycled. In 2020, plastic waste that ended up in the environment represented 81 million tons, or 22% of the total. “Beyond waste, this leads to pollution by microplastics and nanoplastics, resulting from their fragmentation. All environments are affected: Seas, rivers, soils, air, and even living organisms,” Bolo said.

 

Methodological Challenges

However, measuring the impact of plastics on health faces methodological difficulties due to the wide variety of composition, size, and shape of plastics. Nevertheless, the French Standardization Association (Association Française de Normalisation) has conducted work to establish a characterization standard for microplastics in water, which serves as an international reference.

“It is also very difficult to know what we are ingesting,” Bolo said. “A study conducted in 2019 estimated that the average human absorbs 5 grams of plastics per week, the equivalent of a credit card.» Since then, other studies have revised this estimate downward, but no consensus has been reached.

A recent study across 109 countries, both industrialized and developing, found significant exposure, estimated at 500 mg/d, particularly in Southeast Asian countries, where it was due mainly to seafood consumption.

A study concluded that plastic water bottles contain 240,000 particles per liter, 90% of which are nanoplastics. These nanoparticles can pass through the intestinal barrier to enter the bloodstream and reach several organs including the heart, brain, and placenta, as well as the fetus.

 

Changes to the Microbiome

Microplastics also accumulate in organs. Thus, the amount of plastic in the lungs increases with age, suggesting that particles may persist in the body without being eliminated. The health consequences of this are still poorly understood, but exposure to plastics appears to cause changes in the composition of the intestinal microbiota. Pathobionts (commensal bacteria with harmful potential) have been found in both adults and children, which could contribute to dysbiosis of the gut microbiome. Furthermore, a decrease in butyrate, a short-chain fatty acid beneficial to health, has been observed in children’s intestines.

Inhaled nanoplastics may disrupt the mucociliary clearance mechanisms of the respiratory system. The toxicity of inhaled plastic particles was demonstrated as early as the 1970s among workers in the flocking industry. Some developed lung function impairments, shortness of breath, inflammation, fibrosis, and even lung cancer. Similar symptoms have been observed in workers in the textile and polyvinyl chloride industries.

A study published recently in The New England Journal of Medicine measured the amount of microplastics collected from carotid plaque of more than 300 patients who had undergone carotid endarterectomy for asymptomatic carotid artery disease. It found a 4.53 times higher risk for the primary endpoint, a composite of myocardial infarction, stroke, and all-cause mortality, among individuals with microplastics and nanoplastics in plaque compared with those without.

 

Health Affects High

The danger of plastics is also directly linked to the chemical substances they contain. A general scientific review looked at the health impacts of three chemicals used almost exclusively in plastics: Polybromodiphenyl ethers (PBDEs), used as flame retardants in textiles or electronics; bisphenol A (BPA), used in the lining of cans and bottles; and phthalates, particularly diethylhexyl phthalate (DEHP), used to make plastics more flexible.

The review highlighted strong epidemiological evidence linking fetal exposure to PBDEs during pregnancy to low birth weight and later exposure to delayed or impaired cognitive development in children and even a loss of IQ. Statistically significant evidence of disruption of thyroid function in adults was also found.

BPA is linked to genital malformations in female newborns exposed to BPA in utero, type 2 diabetes in adults, insulin resistance, and polycystic ovary syndrome in women. BPA exposure also increases the risk for obesity and hypertension in both children and adults, as well as the risk for cardiovascular disease in adults.

Finally, the review established links between exposure to DEHP and miscarriages, genital malformations in male newborns, delayed or impaired cognitive development in children, loss of IQ, delayed psychomotor development, early puberty in young girls, and endometriosis in young women. DEHP exposure also has multiple effects on cardiometabolic health, including insulin resistance, obesity, and elevated blood pressure.

The economic costs associated with the health impacts of these three substances have been estimated at $675 billion in the United States.

Bolo said that the solution to this plastic pollution is necessarily international. “We need an ambitious and legally binding treaty to reduce plastic production,” he said. “The damage is already done; we need to act to protect human health,” he concluded. The parliamentary office has made nine recommendations to the treaty negotiators.

This story was translated from Medscape’s French edition using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article appeared on Medscape.com.

In preparation for a future international treaty aimed at reducing plastic pollution, the French Parliamentary Office for the Evaluation of Scientific and Technological Choices presented the conclusions of a public hearing on the impact of plastics on various aspects of human health.

Increased Global Plastic Production

Philippe Bolo, a member of the French Democratic Party and the rapporteur for the public mission on the health impacts of plastics, spoke about the latest round of treaty negotiations, held from November 25 to December 1 in South Korea, attended by leading French and global experts about the impact of plastics on human health.

The hearing highlighted a sharp increase in plastic production. “It has doubled in the last 20 years and is expected to exceed 500 million tons in 2024,” Bolo said. This is about 60 kg per person. According to projections from the Organization for Economic Co-operation and Development, on its current trajectory, plastic production will reach 750 million tons by 2040 and surpass 1 billion tons before 2050, he said.

 

Minimal Plastic Waste Recycling

Around one third (32%) of plastics are used for packaging. “Therefore, most plastic production is still intended for single-use purposes,” he said. Plastic waste follows a similar growth trajectory, with volumes expected to rise from 360 million tons in 2020 to 617 million tons by 2040 unless action is taken. Very little of this waste is recycled, even in the most countries that are most advanced in terms of collection, sorting, and processing.

In France, for example, in 2018, only 0.6 million tons of the 3.6 million tons of plastic waste produced was truly recycled. This is less than one fifth (17%). Globally, less than 10% of plastic waste is recycled. In 2020, plastic waste that ended up in the environment represented 81 million tons, or 22% of the total. “Beyond waste, this leads to pollution by microplastics and nanoplastics, resulting from their fragmentation. All environments are affected: Seas, rivers, soils, air, and even living organisms,” Bolo said.

 

Methodological Challenges

However, measuring the impact of plastics on health faces methodological difficulties due to the wide variety of composition, size, and shape of plastics. Nevertheless, the French Standardization Association (Association Française de Normalisation) has conducted work to establish a characterization standard for microplastics in water, which serves as an international reference.

“It is also very difficult to know what we are ingesting,” Bolo said. “A study conducted in 2019 estimated that the average human absorbs 5 grams of plastics per week, the equivalent of a credit card.» Since then, other studies have revised this estimate downward, but no consensus has been reached.

A recent study across 109 countries, both industrialized and developing, found significant exposure, estimated at 500 mg/d, particularly in Southeast Asian countries, where it was due mainly to seafood consumption.

A study concluded that plastic water bottles contain 240,000 particles per liter, 90% of which are nanoplastics. These nanoparticles can pass through the intestinal barrier to enter the bloodstream and reach several organs including the heart, brain, and placenta, as well as the fetus.

 

Changes to the Microbiome

Microplastics also accumulate in organs. Thus, the amount of plastic in the lungs increases with age, suggesting that particles may persist in the body without being eliminated. The health consequences of this are still poorly understood, but exposure to plastics appears to cause changes in the composition of the intestinal microbiota. Pathobionts (commensal bacteria with harmful potential) have been found in both adults and children, which could contribute to dysbiosis of the gut microbiome. Furthermore, a decrease in butyrate, a short-chain fatty acid beneficial to health, has been observed in children’s intestines.

Inhaled nanoplastics may disrupt the mucociliary clearance mechanisms of the respiratory system. The toxicity of inhaled plastic particles was demonstrated as early as the 1970s among workers in the flocking industry. Some developed lung function impairments, shortness of breath, inflammation, fibrosis, and even lung cancer. Similar symptoms have been observed in workers in the textile and polyvinyl chloride industries.

A study published recently in The New England Journal of Medicine measured the amount of microplastics collected from carotid plaque of more than 300 patients who had undergone carotid endarterectomy for asymptomatic carotid artery disease. It found a 4.53 times higher risk for the primary endpoint, a composite of myocardial infarction, stroke, and all-cause mortality, among individuals with microplastics and nanoplastics in plaque compared with those without.

 

Health Affects High

The danger of plastics is also directly linked to the chemical substances they contain. A general scientific review looked at the health impacts of three chemicals used almost exclusively in plastics: Polybromodiphenyl ethers (PBDEs), used as flame retardants in textiles or electronics; bisphenol A (BPA), used in the lining of cans and bottles; and phthalates, particularly diethylhexyl phthalate (DEHP), used to make plastics more flexible.

The review highlighted strong epidemiological evidence linking fetal exposure to PBDEs during pregnancy to low birth weight and later exposure to delayed or impaired cognitive development in children and even a loss of IQ. Statistically significant evidence of disruption of thyroid function in adults was also found.

BPA is linked to genital malformations in female newborns exposed to BPA in utero, type 2 diabetes in adults, insulin resistance, and polycystic ovary syndrome in women. BPA exposure also increases the risk for obesity and hypertension in both children and adults, as well as the risk for cardiovascular disease in adults.

Finally, the review established links between exposure to DEHP and miscarriages, genital malformations in male newborns, delayed or impaired cognitive development in children, loss of IQ, delayed psychomotor development, early puberty in young girls, and endometriosis in young women. DEHP exposure also has multiple effects on cardiometabolic health, including insulin resistance, obesity, and elevated blood pressure.

The economic costs associated with the health impacts of these three substances have been estimated at $675 billion in the United States.

Bolo said that the solution to this plastic pollution is necessarily international. “We need an ambitious and legally binding treaty to reduce plastic production,” he said. “The damage is already done; we need to act to protect human health,” he concluded. The parliamentary office has made nine recommendations to the treaty negotiators.

This story was translated from Medscape’s French edition using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication. A version of this article appeared on Medscape.com.

Long COVID can have an enormous impact on people’s ability to work, particularly if they do not have workplace accommodations and support. Although some patients experience symptoms so severe that they cannot work under any conditions, medical providers and employers can help ensure many patients with long COVID can stay in the workforce.

Long COVID is an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems. By the end of 2023, at least 400 million people worldwide were estimated to have long COVID.

As members of the Patient-Led Research Collaborative, an international group of more than 60 researchers and health advocates living with long COVID and other infection-associated chronic conditions, we have published one of the first research studies of people with long COVID and their desire to work, the specific needs they have, and what doctors and employers can do to create a path for returning to the workforce. 

In our recent paper, we document the barriers and facilitators that individuals living with long COVID experience when attempting to return to work. Our recommendations are based on these findings and include recommendations for both medical providers and employers. 

If you are a medical provider:

• Ensure you adequately document your patients’ COVID cases, any long COVID diagnoses, and the functional impairment that long COVID causes. Remember that you can diagnose a patient with long COVID on the basis of their symptoms, and clinical guidelines do not require a record of a positive COVID-19 test, which many patients lack owing to testing barriers.
• Keep up to date on research on long COVID and related infection-associated chronic conditions — for example, through the Project ECHO Long COVID and Fatiguing Illness Recovery Program — and learn about pacing and other treatment options.

If you are an employer: 

• Utilize a return-to-work model in which any worker with suspected or confirmed COVID discusses support they may need with their employer when they return to work, with additional check-in dates scheduled to reevaluate supports as needed. Planning for this collaborative and iterative evaluation of return-to-work supports for all workers with COVID-19 is important because it may not be immediately clear to a worker whether they have developed long COVID or are generally recuperating from the illness.
• Do not require medical documentation of a SARS-CoV-2 infection or a Long COVID diagnosis to access accommodations — this is owing to disparities in accessing documentation.
• Tailor job responsibilities, provide remote options, allow flexible hours, and provide longer-range deadlines to account for symptoms for people with long COVID and other infection-associated chronic conditions.
• Provide accommodations to any caregivers of people with long COVID in your workplace.
• If requiring in-person work, make the workplace as safe as possible through ventilation and masking requirements, which will help ensure fewer of your workers develop long COVID, and those already with infection-associated chronic conditions will not get worse.

Our findings and recommendations are specific to long COVID, but they can and should apply to other disabilities. Given that our study’s sample was predominately White and working in jobs that did not require substantial physical labor, additional recommendations may be needed for other populations and workers who have labor-intensive jobs.

 

510 Study Participants

Long COVID is characterized as a relapsing-remitting illness, often described as episodic, in which an individual’s symptoms may fluctuate. Symptoms can become more or less severe depending on tasks, exertion, and social support in addition to physiologic processes and medical intervention. In our paper, we illustrate how the long COVID return-to-work experience and individuals’ symptoms can be shaped by workplace, home, and medical environments. 

We randomly selected 510 participants from a global survey of people living with long COVID and systematically analyzed their open-ended responses using established qualitative analysis methods. In this study, we specifically analyzed what patients wrote about their return-to-work experiences, considering how work experiences and relapsing and remitting long COVID symptoms intersected with personal lives and medical care. 

Most of the study participants identified as White, were 30-60 years old (ie, in their key earning years), and had at least a baccalaureate degree. Participants lived in the United States (38%), United Kingdom (25%), continental Europe (8%), Canada (4%), or other countries (25%). Most participants worked in professions that did not require substantial physical labor, and individuals in those fields may experience even greater return-to-work barriers than are reported in this study.

 

Key Findings

Through our qualitative analysis, we identified four primary return-to-work themes: 

1. People living with long COVID have a strong desire and financial need to return to work. 

The participants in our study described how they had experienced financial hardship because they could not successfully return to work and may have incurred new expenses with long COVID. They also often wrote how they wanted to return to work because their jobs provided meaning and structure for their lives. Some people in this study shared how they had tried to return to their jobs but relapsed. As a result, they considered leaving the workforce.

2. Workers’ long COVID symptoms intersect with organization of work and home life.

Most of the people in our study were employed in positions that did not require substantial physical labor. Even so, workers described how their long COVID symptoms were exacerbated by some job tasks. Computer screen time; reading dense material or writing (including emails); and conversations and meetings, regardless of whether they were in-person or via phone or video conferencing, could trigger or make symptoms worse. Workers who needed to stand for long periods of time, such as teachers and healthcare workers, and workers who needed to do lifting as part of their jobs described how these requirements were too taxing and could lead to relapses.

Because of the relapsing and remitting nature of many long COVID symptoms, people reported how it could be difficult to predict how job tasks, long hours, or pressing deadlines may exacerbate symptoms, which would require them to take time off work. For these participants, “pushing through” symptoms only made the symptoms worse. However, people in the study who were allowed to work from home reported how pacing, elevating their legs, and conserving energy (especially by not commuting) was key to doing their jobs well.

Some people in the study described how they were only able to return to work because they had substantial support from family or partners at home. These individuals shared how the people they lived with did most of the cooking, cleaning, and other household tasks so that the person living with long COVID could conserve their energy for work. This reorganization of home life notably shifted household tasks and caregiving to other people in the household, but without this shift, the individual’s long COVID symptoms may be too severe to work.

3. People with long COVID experience disbelief and stigma at work and healthcare settings.

Some people in our study described how their colleagues, supervisors, and human resource managers insinuated that they were fabricating or exaggerating their symptoms. This made it hard for workers to communicate what support they needed and could limit access to necessary work accommodations.

Many people in our study also described how medical providers did not believe that they had long COVID despite experiencing debilitating symptoms, often because they did not have a positive COVID-19 test to prove they had had an acute infection. Many people with long COVID may not have a positive COVID-19 test because:

• They could not access a test because testing access was limited at the start of the COVID-19 pandemic, there are transportation and cost barriers to tests, many health insurance providers no longer cover tests; and there are fewer public testing sites since the World Health Organization declared an end to the public health emergency;
• There is a high probability of false-negative results for viral and antibody tests (especially during the first wave of the pandemic and for individuals with limited immune response); and
• People can develop long COVID after asymptomatic acute infection.

Although healthcare providers can provide a long COVID diagnosis without a positive COVID-19 test on the basis of a patient’s presentation of symptoms and clinical history, many people in our study said that their providers would not provide this diagnosis, which restricted access to worker’s compensation, paid time off, and job accommodations.

Many people in the study also reported that their medical providers misdiagnosed them with a mental health disorder, such as anxiety, instead of long COVID. Although some people with long COVID may experience poor mental health as a natural consequence of dealing with a debilitating medical condition or may have neuropsychiatric symptoms as part of their long COVID, long COVID is not caused by an underlying psychiatric illness.

4. Support of medical providers is key to successful return to work for people living with long COVID.

Some people in our study described how they were able to get workplace accommodations or access workers’ compensation or sick leave because their medical providers recognized they had long COVID and provided them with this documentation. Some of these participants did not have a positive COVID-19 test, but their medical providers were able to diagnose them with long COVID on the basis of symptom presentation and clinical history. This documentation was critical for helping workers remain financially stable and able to return to work.

 

Conclusion

While we continue to search for treatment and cures for long COVID and work to provide a robust social safety net, it is crucial to address the stigma, inaccessibility, and lack of support often experienced by patients in their workplaces. Disabled people have long faced these issues; long COVID may be an opportunity to revolutionize the workplace to ensure an inclusive and accessible environment that can improve the lives of all workers.

For more on how to best be inclusive of employees with long COVID, read Harvard Business Review’s “Long Covid at Work: A Manager’s Guide” and visit the Job Accommodation Network webpage dedicated to long COVID.

Additional discussion about our study and applying the findings to improve work and medical care can be found by listening to the Healthy Work podcast episode titled “Supporting Long COVID at Work.” 

 

Elisabeth Stelson, Gina Assaf, and Lisa McCorkell are members of the Patient-Led Research Collaborative, an international group of more than 60 researchers. Dr Stelson, Postdoctoral Research Fellow, Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, has disclosed no relevant financial relationships. Gina Assaf is Research Lead, Patient-Led Research Collaborative, Washington, DC. Lisa McCorkell is a long COVID patient; Cofounder, Team Lead, Researcher, Patient-Led Research Collaborative, Washington, DC.

A version of this article appeared on Medscape.com.

Long COVID can have an enormous impact on people’s ability to work, particularly if they do not have workplace accommodations and support. Although some patients experience symptoms so severe that they cannot work under any conditions, medical providers and employers can help ensure many patients with long COVID can stay in the workforce.

Long COVID is an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems. By the end of 2023, at least 400 million people worldwide were estimated to have long COVID.

As members of the Patient-Led Research Collaborative, an international group of more than 60 researchers and health advocates living with long COVID and other infection-associated chronic conditions, we have published one of the first research studies of people with long COVID and their desire to work, the specific needs they have, and what doctors and employers can do to create a path for returning to the workforce. 

In our recent paper, we document the barriers and facilitators that individuals living with long COVID experience when attempting to return to work. Our recommendations are based on these findings and include recommendations for both medical providers and employers. 

If you are a medical provider:

• Ensure you adequately document your patients’ COVID cases, any long COVID diagnoses, and the functional impairment that long COVID causes. Remember that you can diagnose a patient with long COVID on the basis of their symptoms, and clinical guidelines do not require a record of a positive COVID-19 test, which many patients lack owing to testing barriers.
• Keep up to date on research on long COVID and related infection-associated chronic conditions — for example, through the Project ECHO Long COVID and Fatiguing Illness Recovery Program — and learn about pacing and other treatment options.

If you are an employer: 

• Utilize a return-to-work model in which any worker with suspected or confirmed COVID discusses support they may need with their employer when they return to work, with additional check-in dates scheduled to reevaluate supports as needed. Planning for this collaborative and iterative evaluation of return-to-work supports for all workers with COVID-19 is important because it may not be immediately clear to a worker whether they have developed long COVID or are generally recuperating from the illness.
• Do not require medical documentation of a SARS-CoV-2 infection or a Long COVID diagnosis to access accommodations — this is owing to disparities in accessing documentation.
• Tailor job responsibilities, provide remote options, allow flexible hours, and provide longer-range deadlines to account for symptoms for people with long COVID and other infection-associated chronic conditions.
• Provide accommodations to any caregivers of people with long COVID in your workplace.
• If requiring in-person work, make the workplace as safe as possible through ventilation and masking requirements, which will help ensure fewer of your workers develop long COVID, and those already with infection-associated chronic conditions will not get worse.

Our findings and recommendations are specific to long COVID, but they can and should apply to other disabilities. Given that our study’s sample was predominately White and working in jobs that did not require substantial physical labor, additional recommendations may be needed for other populations and workers who have labor-intensive jobs.

 

510 Study Participants

Long COVID is characterized as a relapsing-remitting illness, often described as episodic, in which an individual’s symptoms may fluctuate. Symptoms can become more or less severe depending on tasks, exertion, and social support in addition to physiologic processes and medical intervention. In our paper, we illustrate how the long COVID return-to-work experience and individuals’ symptoms can be shaped by workplace, home, and medical environments. 

We randomly selected 510 participants from a global survey of people living with long COVID and systematically analyzed their open-ended responses using established qualitative analysis methods. In this study, we specifically analyzed what patients wrote about their return-to-work experiences, considering how work experiences and relapsing and remitting long COVID symptoms intersected with personal lives and medical care. 

Most of the study participants identified as White, were 30-60 years old (ie, in their key earning years), and had at least a baccalaureate degree. Participants lived in the United States (38%), United Kingdom (25%), continental Europe (8%), Canada (4%), or other countries (25%). Most participants worked in professions that did not require substantial physical labor, and individuals in those fields may experience even greater return-to-work barriers than are reported in this study.

 

Key Findings

Through our qualitative analysis, we identified four primary return-to-work themes: 

1. People living with long COVID have a strong desire and financial need to return to work. 

The participants in our study described how they had experienced financial hardship because they could not successfully return to work and may have incurred new expenses with long COVID. They also often wrote how they wanted to return to work because their jobs provided meaning and structure for their lives. Some people in this study shared how they had tried to return to their jobs but relapsed. As a result, they considered leaving the workforce.

2. Workers’ long COVID symptoms intersect with organization of work and home life.

Most of the people in our study were employed in positions that did not require substantial physical labor. Even so, workers described how their long COVID symptoms were exacerbated by some job tasks. Computer screen time; reading dense material or writing (including emails); and conversations and meetings, regardless of whether they were in-person or via phone or video conferencing, could trigger or make symptoms worse. Workers who needed to stand for long periods of time, such as teachers and healthcare workers, and workers who needed to do lifting as part of their jobs described how these requirements were too taxing and could lead to relapses.

Because of the relapsing and remitting nature of many long COVID symptoms, people reported how it could be difficult to predict how job tasks, long hours, or pressing deadlines may exacerbate symptoms, which would require them to take time off work. For these participants, “pushing through” symptoms only made the symptoms worse. However, people in the study who were allowed to work from home reported how pacing, elevating their legs, and conserving energy (especially by not commuting) was key to doing their jobs well.

Some people in the study described how they were only able to return to work because they had substantial support from family or partners at home. These individuals shared how the people they lived with did most of the cooking, cleaning, and other household tasks so that the person living with long COVID could conserve their energy for work. This reorganization of home life notably shifted household tasks and caregiving to other people in the household, but without this shift, the individual’s long COVID symptoms may be too severe to work.

3. People with long COVID experience disbelief and stigma at work and healthcare settings.

Some people in our study described how their colleagues, supervisors, and human resource managers insinuated that they were fabricating or exaggerating their symptoms. This made it hard for workers to communicate what support they needed and could limit access to necessary work accommodations.

Many people in our study also described how medical providers did not believe that they had long COVID despite experiencing debilitating symptoms, often because they did not have a positive COVID-19 test to prove they had had an acute infection. Many people with long COVID may not have a positive COVID-19 test because:

• They could not access a test because testing access was limited at the start of the COVID-19 pandemic, there are transportation and cost barriers to tests, many health insurance providers no longer cover tests; and there are fewer public testing sites since the World Health Organization declared an end to the public health emergency;
• There is a high probability of false-negative results for viral and antibody tests (especially during the first wave of the pandemic and for individuals with limited immune response); and
• People can develop long COVID after asymptomatic acute infection.

Although healthcare providers can provide a long COVID diagnosis without a positive COVID-19 test on the basis of a patient’s presentation of symptoms and clinical history, many people in our study said that their providers would not provide this diagnosis, which restricted access to worker’s compensation, paid time off, and job accommodations.

Many people in the study also reported that their medical providers misdiagnosed them with a mental health disorder, such as anxiety, instead of long COVID. Although some people with long COVID may experience poor mental health as a natural consequence of dealing with a debilitating medical condition or may have neuropsychiatric symptoms as part of their long COVID, long COVID is not caused by an underlying psychiatric illness.

4. Support of medical providers is key to successful return to work for people living with long COVID.

Some people in our study described how they were able to get workplace accommodations or access workers’ compensation or sick leave because their medical providers recognized they had long COVID and provided them with this documentation. Some of these participants did not have a positive COVID-19 test, but their medical providers were able to diagnose them with long COVID on the basis of symptom presentation and clinical history. This documentation was critical for helping workers remain financially stable and able to return to work.

 

Conclusion

While we continue to search for treatment and cures for long COVID and work to provide a robust social safety net, it is crucial to address the stigma, inaccessibility, and lack of support often experienced by patients in their workplaces. Disabled people have long faced these issues; long COVID may be an opportunity to revolutionize the workplace to ensure an inclusive and accessible environment that can improve the lives of all workers.

For more on how to best be inclusive of employees with long COVID, read Harvard Business Review’s “Long Covid at Work: A Manager’s Guide” and visit the Job Accommodation Network webpage dedicated to long COVID.

Additional discussion about our study and applying the findings to improve work and medical care can be found by listening to the Healthy Work podcast episode titled “Supporting Long COVID at Work.” 

 

Elisabeth Stelson, Gina Assaf, and Lisa McCorkell are members of the Patient-Led Research Collaborative, an international group of more than 60 researchers. Dr Stelson, Postdoctoral Research Fellow, Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, has disclosed no relevant financial relationships. Gina Assaf is Research Lead, Patient-Led Research Collaborative, Washington, DC. Lisa McCorkell is a long COVID patient; Cofounder, Team Lead, Researcher, Patient-Led Research Collaborative, Washington, DC.

A version of this article appeared on Medscape.com.

Long COVID can have an enormous impact on people’s ability to work, particularly if they do not have workplace accommodations and support. Although some patients experience symptoms so severe that they cannot work under any conditions, medical providers and employers can help ensure many patients with long COVID can stay in the workforce.

Long COVID is an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems. By the end of 2023, at least 400 million people worldwide were estimated to have long COVID.

As members of the Patient-Led Research Collaborative, an international group of more than 60 researchers and health advocates living with long COVID and other infection-associated chronic conditions, we have published one of the first research studies of people with long COVID and their desire to work, the specific needs they have, and what doctors and employers can do to create a path for returning to the workforce. 

In our recent paper, we document the barriers and facilitators that individuals living with long COVID experience when attempting to return to work. Our recommendations are based on these findings and include recommendations for both medical providers and employers. 

If you are a medical provider:

• Ensure you adequately document your patients’ COVID cases, any long COVID diagnoses, and the functional impairment that long COVID causes. Remember that you can diagnose a patient with long COVID on the basis of their symptoms, and clinical guidelines do not require a record of a positive COVID-19 test, which many patients lack owing to testing barriers.
• Keep up to date on research on long COVID and related infection-associated chronic conditions — for example, through the Project ECHO Long COVID and Fatiguing Illness Recovery Program — and learn about pacing and other treatment options.

If you are an employer: 

• Utilize a return-to-work model in which any worker with suspected or confirmed COVID discusses support they may need with their employer when they return to work, with additional check-in dates scheduled to reevaluate supports as needed. Planning for this collaborative and iterative evaluation of return-to-work supports for all workers with COVID-19 is important because it may not be immediately clear to a worker whether they have developed long COVID or are generally recuperating from the illness.
• Do not require medical documentation of a SARS-CoV-2 infection or a Long COVID diagnosis to access accommodations — this is owing to disparities in accessing documentation.
• Tailor job responsibilities, provide remote options, allow flexible hours, and provide longer-range deadlines to account for symptoms for people with long COVID and other infection-associated chronic conditions.
• Provide accommodations to any caregivers of people with long COVID in your workplace.
• If requiring in-person work, make the workplace as safe as possible through ventilation and masking requirements, which will help ensure fewer of your workers develop long COVID, and those already with infection-associated chronic conditions will not get worse.

Our findings and recommendations are specific to long COVID, but they can and should apply to other disabilities. Given that our study’s sample was predominately White and working in jobs that did not require substantial physical labor, additional recommendations may be needed for other populations and workers who have labor-intensive jobs.

 

510 Study Participants

Long COVID is characterized as a relapsing-remitting illness, often described as episodic, in which an individual’s symptoms may fluctuate. Symptoms can become more or less severe depending on tasks, exertion, and social support in addition to physiologic processes and medical intervention. In our paper, we illustrate how the long COVID return-to-work experience and individuals’ symptoms can be shaped by workplace, home, and medical environments. 

We randomly selected 510 participants from a global survey of people living with long COVID and systematically analyzed their open-ended responses using established qualitative analysis methods. In this study, we specifically analyzed what patients wrote about their return-to-work experiences, considering how work experiences and relapsing and remitting long COVID symptoms intersected with personal lives and medical care. 

Most of the study participants identified as White, were 30-60 years old (ie, in their key earning years), and had at least a baccalaureate degree. Participants lived in the United States (38%), United Kingdom (25%), continental Europe (8%), Canada (4%), or other countries (25%). Most participants worked in professions that did not require substantial physical labor, and individuals in those fields may experience even greater return-to-work barriers than are reported in this study.

 

Key Findings

Through our qualitative analysis, we identified four primary return-to-work themes: 

1. People living with long COVID have a strong desire and financial need to return to work. 

The participants in our study described how they had experienced financial hardship because they could not successfully return to work and may have incurred new expenses with long COVID. They also often wrote how they wanted to return to work because their jobs provided meaning and structure for their lives. Some people in this study shared how they had tried to return to their jobs but relapsed. As a result, they considered leaving the workforce.

2. Workers’ long COVID symptoms intersect with organization of work and home life.

Most of the people in our study were employed in positions that did not require substantial physical labor. Even so, workers described how their long COVID symptoms were exacerbated by some job tasks. Computer screen time; reading dense material or writing (including emails); and conversations and meetings, regardless of whether they were in-person or via phone or video conferencing, could trigger or make symptoms worse. Workers who needed to stand for long periods of time, such as teachers and healthcare workers, and workers who needed to do lifting as part of their jobs described how these requirements were too taxing and could lead to relapses.

Because of the relapsing and remitting nature of many long COVID symptoms, people reported how it could be difficult to predict how job tasks, long hours, or pressing deadlines may exacerbate symptoms, which would require them to take time off work. For these participants, “pushing through” symptoms only made the symptoms worse. However, people in the study who were allowed to work from home reported how pacing, elevating their legs, and conserving energy (especially by not commuting) was key to doing their jobs well.

Some people in the study described how they were only able to return to work because they had substantial support from family or partners at home. These individuals shared how the people they lived with did most of the cooking, cleaning, and other household tasks so that the person living with long COVID could conserve their energy for work. This reorganization of home life notably shifted household tasks and caregiving to other people in the household, but without this shift, the individual’s long COVID symptoms may be too severe to work.

3. People with long COVID experience disbelief and stigma at work and healthcare settings.

Some people in our study described how their colleagues, supervisors, and human resource managers insinuated that they were fabricating or exaggerating their symptoms. This made it hard for workers to communicate what support they needed and could limit access to necessary work accommodations.

Many people in our study also described how medical providers did not believe that they had long COVID despite experiencing debilitating symptoms, often because they did not have a positive COVID-19 test to prove they had had an acute infection. Many people with long COVID may not have a positive COVID-19 test because:

• They could not access a test because testing access was limited at the start of the COVID-19 pandemic, there are transportation and cost barriers to tests, many health insurance providers no longer cover tests; and there are fewer public testing sites since the World Health Organization declared an end to the public health emergency;
• There is a high probability of false-negative results for viral and antibody tests (especially during the first wave of the pandemic and for individuals with limited immune response); and
• People can develop long COVID after asymptomatic acute infection.

Although healthcare providers can provide a long COVID diagnosis without a positive COVID-19 test on the basis of a patient’s presentation of symptoms and clinical history, many people in our study said that their providers would not provide this diagnosis, which restricted access to worker’s compensation, paid time off, and job accommodations.

Many people in the study also reported that their medical providers misdiagnosed them with a mental health disorder, such as anxiety, instead of long COVID. Although some people with long COVID may experience poor mental health as a natural consequence of dealing with a debilitating medical condition or may have neuropsychiatric symptoms as part of their long COVID, long COVID is not caused by an underlying psychiatric illness.

4. Support of medical providers is key to successful return to work for people living with long COVID.

Some people in our study described how they were able to get workplace accommodations or access workers’ compensation or sick leave because their medical providers recognized they had long COVID and provided them with this documentation. Some of these participants did not have a positive COVID-19 test, but their medical providers were able to diagnose them with long COVID on the basis of symptom presentation and clinical history. This documentation was critical for helping workers remain financially stable and able to return to work.

 

Conclusion

While we continue to search for treatment and cures for long COVID and work to provide a robust social safety net, it is crucial to address the stigma, inaccessibility, and lack of support often experienced by patients in their workplaces. Disabled people have long faced these issues; long COVID may be an opportunity to revolutionize the workplace to ensure an inclusive and accessible environment that can improve the lives of all workers.

For more on how to best be inclusive of employees with long COVID, read Harvard Business Review’s “Long Covid at Work: A Manager’s Guide” and visit the Job Accommodation Network webpage dedicated to long COVID.

Additional discussion about our study and applying the findings to improve work and medical care can be found by listening to the Healthy Work podcast episode titled “Supporting Long COVID at Work.” 

 

Elisabeth Stelson, Gina Assaf, and Lisa McCorkell are members of the Patient-Led Research Collaborative, an international group of more than 60 researchers. Dr Stelson, Postdoctoral Research Fellow, Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, has disclosed no relevant financial relationships. Gina Assaf is Research Lead, Patient-Led Research Collaborative, Washington, DC. Lisa McCorkell is a long COVID patient; Cofounder, Team Lead, Researcher, Patient-Led Research Collaborative, Washington, DC.

A version of this article appeared on Medscape.com.

An electronic nudge explaining the cardiovascular benefits of the influenza vaccine increased vaccination rates, particularly among people who had previously had a heart attack, showed the NUDGE FLU series of clinical trials.

Influenza has the potential to be a dangerous infection on its own, but it increases the risk for cardiovascular events among people with a history of heart attack, said the study’s lead author, Ankeet Bhatt, MD, a cardiologist at Kaiser Permanente San Francisco Medical Center, San Francisco.

“Yearly influenza vaccines help prevent influenza infection and, in patients with a heart attack, are potentially cardioprotective,” he said during his presentation at the American Heart Association (AHA) Scientific Sessions 2024 in Chicago. The NUDGE FLU results were simultaneously published online in JAMA Cardiology.

In Denmark, where the trials were conducted, about 80% of older adults get flu shots, but only about 40% of younger adults with chronic diseases do, Bhatt reported. In the United States, about 45% of adults and 55% of children received at least one dose of the flu vaccine in the 2023/24 flu season, according to the US Centers for Disease Control and Prevention (CDC).

 

The NUDGE FLU Trials

Bhatt and his colleagues conducted three related clinical trials during the 2022/23 and 2023/24 flu seasons: NUDGE-FLU and NUDGE-FLU-2 targeted older adults, whereas NUDGE-FLU-CHRONIC targeted younger adults with chronic diseases. Nearly 2 million people were involved in the three trials.

Participants were randomized to receive one of a series of different behavioral-science-informed letters, delivered through a government-run electronic communication system, or no reminder.

People who received any of the nudges had higher rates of vaccination; among heart attack survivors, there was a 1.8% improvement and among adults without a history of heart attack, there was a 1.3% improvement. But a nudge that explained the potential cardiovascular benefits of flu shots was even more effective, leading to a 3.9% increase among people with a history of heart attack and a 2% increase among those with no heart attack history.

“A simple sentence resulted in a durable improvement in the vaccination rate,” said Bhatt.

The effect was even greater among those who had not been vaccinated in the previous flu season. Among heart attack survivors, nearly 14% more people got the vaccine compared with just 1.5% more survivors who were previously vaccinated. And it was most effective among younger adults who had experienced a recent heart attack, resulting in a 26% increase.

“The impact was larger in patients with a history of acute myocardial infarction, in those who were vaccine-hesitant, and in younger people” — all groups with the most to gain from vaccination in terms of cardiovascular protection — Bhatt reported.

About 25% of people in the United States are unsure about whether to get a flu shot, said Orly Vardeny, PharmD, professor of medicine at the University of Minnesota Medical School in Minneapolis, who was not involved in the study. The fact that previously unvaccinated people were convinced by the nudges is reassuring. “That’s the group where this intervention is most likely to move the needle,” she said.

Around half of all people hospitalized for flu in the United States have cardiovascular disease, CDC data showed, so “even a small increase in the number of patients who get vaccinated has substantial public health benefits,” Vardeny said.

The NUDGE FLU series showed that nudges like this should be employed as a simple tool to improve vaccination rates, but the system would be much more difficult to implement in the United States, Bhatt said.

Denmark has a national health service and a preexisting government electronic communication system, whereas the US system is privately run and more fractured. It would be possible to make it work, he pointed out, but would take some effort.

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

An electronic nudge explaining the cardiovascular benefits of the influenza vaccine increased vaccination rates, particularly among people who had previously had a heart attack, showed the NUDGE FLU series of clinical trials.

Influenza has the potential to be a dangerous infection on its own, but it increases the risk for cardiovascular events among people with a history of heart attack, said the study’s lead author, Ankeet Bhatt, MD, a cardiologist at Kaiser Permanente San Francisco Medical Center, San Francisco.

“Yearly influenza vaccines help prevent influenza infection and, in patients with a heart attack, are potentially cardioprotective,” he said during his presentation at the American Heart Association (AHA) Scientific Sessions 2024 in Chicago. The NUDGE FLU results were simultaneously published online in JAMA Cardiology.

In Denmark, where the trials were conducted, about 80% of older adults get flu shots, but only about 40% of younger adults with chronic diseases do, Bhatt reported. In the United States, about 45% of adults and 55% of children received at least one dose of the flu vaccine in the 2023/24 flu season, according to the US Centers for Disease Control and Prevention (CDC).

 

The NUDGE FLU Trials

Bhatt and his colleagues conducted three related clinical trials during the 2022/23 and 2023/24 flu seasons: NUDGE-FLU and NUDGE-FLU-2 targeted older adults, whereas NUDGE-FLU-CHRONIC targeted younger adults with chronic diseases. Nearly 2 million people were involved in the three trials.

Participants were randomized to receive one of a series of different behavioral-science-informed letters, delivered through a government-run electronic communication system, or no reminder.

People who received any of the nudges had higher rates of vaccination; among heart attack survivors, there was a 1.8% improvement and among adults without a history of heart attack, there was a 1.3% improvement. But a nudge that explained the potential cardiovascular benefits of flu shots was even more effective, leading to a 3.9% increase among people with a history of heart attack and a 2% increase among those with no heart attack history.

“A simple sentence resulted in a durable improvement in the vaccination rate,” said Bhatt.

The effect was even greater among those who had not been vaccinated in the previous flu season. Among heart attack survivors, nearly 14% more people got the vaccine compared with just 1.5% more survivors who were previously vaccinated. And it was most effective among younger adults who had experienced a recent heart attack, resulting in a 26% increase.

“The impact was larger in patients with a history of acute myocardial infarction, in those who were vaccine-hesitant, and in younger people” — all groups with the most to gain from vaccination in terms of cardiovascular protection — Bhatt reported.

About 25% of people in the United States are unsure about whether to get a flu shot, said Orly Vardeny, PharmD, professor of medicine at the University of Minnesota Medical School in Minneapolis, who was not involved in the study. The fact that previously unvaccinated people were convinced by the nudges is reassuring. “That’s the group where this intervention is most likely to move the needle,” she said.

Around half of all people hospitalized for flu in the United States have cardiovascular disease, CDC data showed, so “even a small increase in the number of patients who get vaccinated has substantial public health benefits,” Vardeny said.

The NUDGE FLU series showed that nudges like this should be employed as a simple tool to improve vaccination rates, but the system would be much more difficult to implement in the United States, Bhatt said.

Denmark has a national health service and a preexisting government electronic communication system, whereas the US system is privately run and more fractured. It would be possible to make it work, he pointed out, but would take some effort.

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

An electronic nudge explaining the cardiovascular benefits of the influenza vaccine increased vaccination rates, particularly among people who had previously had a heart attack, showed the NUDGE FLU series of clinical trials.

Influenza has the potential to be a dangerous infection on its own, but it increases the risk for cardiovascular events among people with a history of heart attack, said the study’s lead author, Ankeet Bhatt, MD, a cardiologist at Kaiser Permanente San Francisco Medical Center, San Francisco.

“Yearly influenza vaccines help prevent influenza infection and, in patients with a heart attack, are potentially cardioprotective,” he said during his presentation at the American Heart Association (AHA) Scientific Sessions 2024 in Chicago. The NUDGE FLU results were simultaneously published online in JAMA Cardiology.

In Denmark, where the trials were conducted, about 80% of older adults get flu shots, but only about 40% of younger adults with chronic diseases do, Bhatt reported. In the United States, about 45% of adults and 55% of children received at least one dose of the flu vaccine in the 2023/24 flu season, according to the US Centers for Disease Control and Prevention (CDC).

 

The NUDGE FLU Trials

Bhatt and his colleagues conducted three related clinical trials during the 2022/23 and 2023/24 flu seasons: NUDGE-FLU and NUDGE-FLU-2 targeted older adults, whereas NUDGE-FLU-CHRONIC targeted younger adults with chronic diseases. Nearly 2 million people were involved in the three trials.

Participants were randomized to receive one of a series of different behavioral-science-informed letters, delivered through a government-run electronic communication system, or no reminder.

People who received any of the nudges had higher rates of vaccination; among heart attack survivors, there was a 1.8% improvement and among adults without a history of heart attack, there was a 1.3% improvement. But a nudge that explained the potential cardiovascular benefits of flu shots was even more effective, leading to a 3.9% increase among people with a history of heart attack and a 2% increase among those with no heart attack history.

“A simple sentence resulted in a durable improvement in the vaccination rate,” said Bhatt.

The effect was even greater among those who had not been vaccinated in the previous flu season. Among heart attack survivors, nearly 14% more people got the vaccine compared with just 1.5% more survivors who were previously vaccinated. And it was most effective among younger adults who had experienced a recent heart attack, resulting in a 26% increase.

“The impact was larger in patients with a history of acute myocardial infarction, in those who were vaccine-hesitant, and in younger people” — all groups with the most to gain from vaccination in terms of cardiovascular protection — Bhatt reported.

About 25% of people in the United States are unsure about whether to get a flu shot, said Orly Vardeny, PharmD, professor of medicine at the University of Minnesota Medical School in Minneapolis, who was not involved in the study. The fact that previously unvaccinated people were convinced by the nudges is reassuring. “That’s the group where this intervention is most likely to move the needle,” she said.

Around half of all people hospitalized for flu in the United States have cardiovascular disease, CDC data showed, so “even a small increase in the number of patients who get vaccinated has substantial public health benefits,” Vardeny said.

The NUDGE FLU series showed that nudges like this should be employed as a simple tool to improve vaccination rates, but the system would be much more difficult to implement in the United States, Bhatt said.

Denmark has a national health service and a preexisting government electronic communication system, whereas the US system is privately run and more fractured. It would be possible to make it work, he pointed out, but would take some effort.

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

Vaccines recommended by the Advisory Committee on Immunization Practices (ACIP) offer important protection against severe illness for pregnant people and their babies.¹ However, vaccination coverage estimates among pregnant people remain suboptimal.^2-5 Additionally, some measures indicate that vaccine hesitancy among pregnant people is increasing; for example, 17.5% of surveyed pregnant women reported being very hesitant about influenza vaccination during pregnancy in 2019-2020, compared with 24.7% in 2022-2023.⁶As fall and winter virus season continues, consider opportunities for you and your staff to help communicate the importance of prenatal vaccination to pregnant patients in your care. Explore updated provider toolkits and prenatal vaccination patient education resources, including fact sheets, social media assets, posters, and short videos on respiratory syncytial virus (RSV), Tdap, COVID-19, influenza, and hepatitis B.

In an interview, CDC’s Haben Debessai, MD, an adjunct instructor in obstetrics and gynecology at Emory School of Medicine, Atlanta, Georgia, contextualizes the data to help healthcare professionals communicate effectively with their pregnant patients. 

 

What can practitioners communicate to patients about why it is important to get vaccinated during their pregnancy?

When communicating with their patients, practitioners can consider opportunities to discuss how vaccines work during pregnancy, emphasizing that prenatal vaccinations are beneficial for both the pregnant person and the fetus. It can be helpful to educate patients on how a pregnant person’s immune system can develop antibodies that will then pass to the fetus during the pregnancy and confer protection during the infant’s early months of life — when they are highly susceptible to illnesses that can be severe, such as RSV-associated lower respiratory tract infections. It can also be useful to discuss pregnancy’s impact on the immune system, which contributes to pregnant people being at higher risk for severe illness from infections like COVID-19 and flu, if contracted. The outcomes of severe illness can be dire for both the pregnant person and their pregnancy, which is why vaccination is the best mitigation option. It can also be beneficial to share with patients that some vaccines, like RSV and Tdap, are specifically for neonatal benefit, which could help patients understand why some vaccines are recommended at a specific gestational age and in each pregnancy or subsequent pregnancies. 

What is known about pregnant populations that experience disparities in vaccination coverage? 

While vaccination coverage among pregnant people is suboptimal, coverage estimates are often lowest among Black pregnant people, some of whom report experiencing mistreatment and discrimination during pregnancy and delivery.⁷ It is important to recognize that there are many intersecting factors that may impact vaccination coverage. Systemic and structural factors may prohibit some patient populations from accessing vaccinations (eg, transportation barriers, difficulty accessing adequate healthcare for those on government assistance, language barriers). To be responsive to the intersectional lived realities of each of these communities, the medical and public health community continually strives to increase trustworthiness, which can lead to increased uptake of vaccinations in these populations. 

What vaccines are available and recommended for pregnant people?

Four vaccines are routinely recommended during pregnancy: Tdap, COVID-19, influenza (seasonal), and RSV (seasonal). CDC recommends getting a Tdap vaccine between the 27th and 36th week of each pregnancy, preferably during the earlier part of this time period. CDC recommends that everyone 6 months or older in the United States, including pregnant people, stay up to date on COVID-19 vaccines. A COVID-19 vaccine can be given during any trimester of pregnancy. CDC recommends an annual flu vaccine during each flu season (fall/winter) for everyone 6 months or older in the United States, including pregnant people. A flu vaccine can be given during any trimester of pregnancy. For individuals who will be between 32 and 36 weeks pregnant during September through January, CDC recommends getting an RSV vaccine. RSV season and timing of vaccination may vary depending on geography. If a pregnant patient does not get the RSV vaccine during their pregnancy, CDC recommends that their baby receive an RSV monoclonal antibody (nirsevimab) to provide additional protection during the infant’s first RSV season, if they are younger than 8 months. At this time, pregnant people who received an RSV vaccine during a previous pregnancy (last year) are not recommended to receive another RSV vaccine during pregnancy. The current recommendation is for babies born during subsequent pregnancies to receive nirsevimab. Some pregnant people may also need other vaccines, such as hepatitis B. 

How can practitioners approach conversations about vaccination during pregnancy amid increasing vaccine hesitancy?

Many pregnant people who do get vaccinated describe their provider’s recommendation as an important motivator toward vaccination.^8-11 Communications research suggests that practitioners can further increase trustworthiness by openly discussing potential side effects of prenatal vaccinations and providing patients with a rationale for why each vaccine is recommended. Practitioners can also utilize opportunities to communicate that the risk for severe illness from whooping cough, COVID-19, flu, and RSV in pregnancy and among neonates in the first few months of life is often higher than the risk for an adverse reaction from receiving ACIP-recommended vaccines. Finally, practitioners can consider sharing tested and refined patient education resources at least one appointment prior to the recommended administration of each vaccine, providing individuals with time to process the information they need to facilitate their vaccine decision-making process.

Some patients may be more comfortable with older, well-known prenatal vaccinations but have skepticism about newer vaccines like COVID-19 and RSV. How can practitioners respond to these concerns?

As pregnant people navigate the challenges of making health decisions that could impact their developing baby, practitioners can build trust through empathetically responding to safety concerns and questions, particularly with respect to newly authorized vaccines. Vaccine confidence may be strengthened by communicating to patients that all recommended vaccinations, including those that have been newly authorized, have been rigorously tested prior to being recommended for pregnant people. Additionally, in my clinical practice, I see that patients are often more comfortable accepting vaccines when the benefit for the baby is clearly communicated. I have been pleasantly surprised that most patients I have counseled on the new maternal RSV vaccine have been receptive, making statements like, “If this will help protect my baby from getting sick, then yes, I will get it.”

As you and your staff care for pregnant patients during fall and winter virus season, remember that a provider recommendation remains one of the strongest known predictors of vaccination uptake.¹² As a trusted source of information about prenatal vaccination, consider further incorporating patient education resources to help communicate how prenatal vaccination helps pregnant people share important protection against severe illnesses with their babies. 

Haben Debessai, MD, is a Gilstrap Fellow at the CDC Foundation. Debessai also serves as an Emory Obstetrics/Gynecology Adjunct Instructor at Grady Health System in Atlanta, Georgia. She disclosed no relevant conflicts of interest.

References

1. ACOG Committee Opinion No. 741: Maternal Immunization. Obstet Gynecol. 2018;131:e214-e217. doi:10.1097/AOG.0000000000002662

2. Centers for Disease Control and Prevention. Flu, Tdap, and COVID-19 vaccination coverage among pregnant women – United States, April 2024. 2024 Sep 23. 3. Centers for Disease Control and Prevention. Respiratory syncytial virus (rsv) vaccination coverage, pregnant persons. 2024 Nov 19. 4. Centers for Disease Control and Prevention. COVID-19 vaccination coverage, pregnant persons. 2024 Nov 19. 5. Centers for Disease Control and Prevention. Influenza vaccination coverage, pregnant persons. 2024 Nov 19.6. Razzaghi H et al. IMMWR Morb Mortal Wkly Rep. 2023;72:1065-1071. Published 2023 Sep 29. doi: 10.15585/mmwr.mm7239a4

7. Mohamoud YA et al. MMWR Morb Mortal Wkly Rep 2023;72:961-967. doi: https://dx.doi.org/10.15585/mmwr.mm7235e1.

8. Kiefer MK et al. Am J Obstet Gynecol MFM. 2022;4:100603. doi: 10.1016/j.ajogmf.2022.100603

9. Spires B et al. Obstet Gynecol Clin North Am. 2023;50:401-419. doi: 10.1016/j.ogc.2023.02.013

10. Wales DP et al. Public Health. 2020;179:38-44. doi: 10.1016/j.puhe.2019.10.001

11. Zimmerman M et al. J Natl Med Assoc. 2023;115:362-376. doi:10.1016/j.jnma.2023.04.003

12. Castillo E et al. Best Pract Res Clin Obstet Gynaecol. 2021;76:83-95. doi:10.1016/j.bpobgyn.2021.03.008

Vaccines recommended by the Advisory Committee on Immunization Practices (ACIP) offer important protection against severe illness for pregnant people and their babies.¹ However, vaccination coverage estimates among pregnant people remain suboptimal.^2-5 Additionally, some measures indicate that vaccine hesitancy among pregnant people is increasing; for example, 17.5% of surveyed pregnant women reported being very hesitant about influenza vaccination during pregnancy in 2019-2020, compared with 24.7% in 2022-2023.⁶As fall and winter virus season continues, consider opportunities for you and your staff to help communicate the importance of prenatal vaccination to pregnant patients in your care. Explore updated provider toolkits and prenatal vaccination patient education resources, including fact sheets, social media assets, posters, and short videos on respiratory syncytial virus (RSV), Tdap, COVID-19, influenza, and hepatitis B.

In an interview, CDC’s Haben Debessai, MD, an adjunct instructor in obstetrics and gynecology at Emory School of Medicine, Atlanta, Georgia, contextualizes the data to help healthcare professionals communicate effectively with their pregnant patients. 

 

What can practitioners communicate to patients about why it is important to get vaccinated during their pregnancy?

When communicating with their patients, practitioners can consider opportunities to discuss how vaccines work during pregnancy, emphasizing that prenatal vaccinations are beneficial for both the pregnant person and the fetus. It can be helpful to educate patients on how a pregnant person’s immune system can develop antibodies that will then pass to the fetus during the pregnancy and confer protection during the infant’s early months of life — when they are highly susceptible to illnesses that can be severe, such as RSV-associated lower respiratory tract infections. It can also be useful to discuss pregnancy’s impact on the immune system, which contributes to pregnant people being at higher risk for severe illness from infections like COVID-19 and flu, if contracted. The outcomes of severe illness can be dire for both the pregnant person and their pregnancy, which is why vaccination is the best mitigation option. It can also be beneficial to share with patients that some vaccines, like RSV and Tdap, are specifically for neonatal benefit, which could help patients understand why some vaccines are recommended at a specific gestational age and in each pregnancy or subsequent pregnancies. 

What is known about pregnant populations that experience disparities in vaccination coverage? 

While vaccination coverage among pregnant people is suboptimal, coverage estimates are often lowest among Black pregnant people, some of whom report experiencing mistreatment and discrimination during pregnancy and delivery.⁷ It is important to recognize that there are many intersecting factors that may impact vaccination coverage. Systemic and structural factors may prohibit some patient populations from accessing vaccinations (eg, transportation barriers, difficulty accessing adequate healthcare for those on government assistance, language barriers). To be responsive to the intersectional lived realities of each of these communities, the medical and public health community continually strives to increase trustworthiness, which can lead to increased uptake of vaccinations in these populations. 

What vaccines are available and recommended for pregnant people?

Four vaccines are routinely recommended during pregnancy: Tdap, COVID-19, influenza (seasonal), and RSV (seasonal). CDC recommends getting a Tdap vaccine between the 27th and 36th week of each pregnancy, preferably during the earlier part of this time period. CDC recommends that everyone 6 months or older in the United States, including pregnant people, stay up to date on COVID-19 vaccines. A COVID-19 vaccine can be given during any trimester of pregnancy. CDC recommends an annual flu vaccine during each flu season (fall/winter) for everyone 6 months or older in the United States, including pregnant people. A flu vaccine can be given during any trimester of pregnancy. For individuals who will be between 32 and 36 weeks pregnant during September through January, CDC recommends getting an RSV vaccine. RSV season and timing of vaccination may vary depending on geography. If a pregnant patient does not get the RSV vaccine during their pregnancy, CDC recommends that their baby receive an RSV monoclonal antibody (nirsevimab) to provide additional protection during the infant’s first RSV season, if they are younger than 8 months. At this time, pregnant people who received an RSV vaccine during a previous pregnancy (last year) are not recommended to receive another RSV vaccine during pregnancy. The current recommendation is for babies born during subsequent pregnancies to receive nirsevimab. Some pregnant people may also need other vaccines, such as hepatitis B. 

How can practitioners approach conversations about vaccination during pregnancy amid increasing vaccine hesitancy?

Many pregnant people who do get vaccinated describe their provider’s recommendation as an important motivator toward vaccination.^8-11 Communications research suggests that practitioners can further increase trustworthiness by openly discussing potential side effects of prenatal vaccinations and providing patients with a rationale for why each vaccine is recommended. Practitioners can also utilize opportunities to communicate that the risk for severe illness from whooping cough, COVID-19, flu, and RSV in pregnancy and among neonates in the first few months of life is often higher than the risk for an adverse reaction from receiving ACIP-recommended vaccines. Finally, practitioners can consider sharing tested and refined patient education resources at least one appointment prior to the recommended administration of each vaccine, providing individuals with time to process the information they need to facilitate their vaccine decision-making process.

Some patients may be more comfortable with older, well-known prenatal vaccinations but have skepticism about newer vaccines like COVID-19 and RSV. How can practitioners respond to these concerns?

As pregnant people navigate the challenges of making health decisions that could impact their developing baby, practitioners can build trust through empathetically responding to safety concerns and questions, particularly with respect to newly authorized vaccines. Vaccine confidence may be strengthened by communicating to patients that all recommended vaccinations, including those that have been newly authorized, have been rigorously tested prior to being recommended for pregnant people. Additionally, in my clinical practice, I see that patients are often more comfortable accepting vaccines when the benefit for the baby is clearly communicated. I have been pleasantly surprised that most patients I have counseled on the new maternal RSV vaccine have been receptive, making statements like, “If this will help protect my baby from getting sick, then yes, I will get it.”

As you and your staff care for pregnant patients during fall and winter virus season, remember that a provider recommendation remains one of the strongest known predictors of vaccination uptake.¹² As a trusted source of information about prenatal vaccination, consider further incorporating patient education resources to help communicate how prenatal vaccination helps pregnant people share important protection against severe illnesses with their babies. 

Haben Debessai, MD, is a Gilstrap Fellow at the CDC Foundation. Debessai also serves as an Emory Obstetrics/Gynecology Adjunct Instructor at Grady Health System in Atlanta, Georgia. She disclosed no relevant conflicts of interest.

References

1. ACOG Committee Opinion No. 741: Maternal Immunization. Obstet Gynecol. 2018;131:e214-e217. doi:10.1097/AOG.0000000000002662

2. Centers for Disease Control and Prevention. Flu, Tdap, and COVID-19 vaccination coverage among pregnant women – United States, April 2024. 2024 Sep 23. 3. Centers for Disease Control and Prevention. Respiratory syncytial virus (rsv) vaccination coverage, pregnant persons. 2024 Nov 19. 4. Centers for Disease Control and Prevention. COVID-19 vaccination coverage, pregnant persons. 2024 Nov 19. 5. Centers for Disease Control and Prevention. Influenza vaccination coverage, pregnant persons. 2024 Nov 19.6. Razzaghi H et al. IMMWR Morb Mortal Wkly Rep. 2023;72:1065-1071. Published 2023 Sep 29. doi: 10.15585/mmwr.mm7239a4

7. Mohamoud YA et al. MMWR Morb Mortal Wkly Rep 2023;72:961-967. doi: https://dx.doi.org/10.15585/mmwr.mm7235e1.

8. Kiefer MK et al. Am J Obstet Gynecol MFM. 2022;4:100603. doi: 10.1016/j.ajogmf.2022.100603

9. Spires B et al. Obstet Gynecol Clin North Am. 2023;50:401-419. doi: 10.1016/j.ogc.2023.02.013

10. Wales DP et al. Public Health. 2020;179:38-44. doi: 10.1016/j.puhe.2019.10.001

11. Zimmerman M et al. J Natl Med Assoc. 2023;115:362-376. doi:10.1016/j.jnma.2023.04.003

12. Castillo E et al. Best Pract Res Clin Obstet Gynaecol. 2021;76:83-95. doi:10.1016/j.bpobgyn.2021.03.008

Vaccines recommended by the Advisory Committee on Immunization Practices (ACIP) offer important protection against severe illness for pregnant people and their babies.¹ However, vaccination coverage estimates among pregnant people remain suboptimal.^2-5 Additionally, some measures indicate that vaccine hesitancy among pregnant people is increasing; for example, 17.5% of surveyed pregnant women reported being very hesitant about influenza vaccination during pregnancy in 2019-2020, compared with 24.7% in 2022-2023.⁶As fall and winter virus season continues, consider opportunities for you and your staff to help communicate the importance of prenatal vaccination to pregnant patients in your care. Explore updated provider toolkits and prenatal vaccination patient education resources, including fact sheets, social media assets, posters, and short videos on respiratory syncytial virus (RSV), Tdap, COVID-19, influenza, and hepatitis B.

In an interview, CDC’s Haben Debessai, MD, an adjunct instructor in obstetrics and gynecology at Emory School of Medicine, Atlanta, Georgia, contextualizes the data to help healthcare professionals communicate effectively with their pregnant patients. 

 

What can practitioners communicate to patients about why it is important to get vaccinated during their pregnancy?

When communicating with their patients, practitioners can consider opportunities to discuss how vaccines work during pregnancy, emphasizing that prenatal vaccinations are beneficial for both the pregnant person and the fetus. It can be helpful to educate patients on how a pregnant person’s immune system can develop antibodies that will then pass to the fetus during the pregnancy and confer protection during the infant’s early months of life — when they are highly susceptible to illnesses that can be severe, such as RSV-associated lower respiratory tract infections. It can also be useful to discuss pregnancy’s impact on the immune system, which contributes to pregnant people being at higher risk for severe illness from infections like COVID-19 and flu, if contracted. The outcomes of severe illness can be dire for both the pregnant person and their pregnancy, which is why vaccination is the best mitigation option. It can also be beneficial to share with patients that some vaccines, like RSV and Tdap, are specifically for neonatal benefit, which could help patients understand why some vaccines are recommended at a specific gestational age and in each pregnancy or subsequent pregnancies. 

What is known about pregnant populations that experience disparities in vaccination coverage? 

While vaccination coverage among pregnant people is suboptimal, coverage estimates are often lowest among Black pregnant people, some of whom report experiencing mistreatment and discrimination during pregnancy and delivery.⁷ It is important to recognize that there are many intersecting factors that may impact vaccination coverage. Systemic and structural factors may prohibit some patient populations from accessing vaccinations (eg, transportation barriers, difficulty accessing adequate healthcare for those on government assistance, language barriers). To be responsive to the intersectional lived realities of each of these communities, the medical and public health community continually strives to increase trustworthiness, which can lead to increased uptake of vaccinations in these populations. 

What vaccines are available and recommended for pregnant people?

Four vaccines are routinely recommended during pregnancy: Tdap, COVID-19, influenza (seasonal), and RSV (seasonal). CDC recommends getting a Tdap vaccine between the 27th and 36th week of each pregnancy, preferably during the earlier part of this time period. CDC recommends that everyone 6 months or older in the United States, including pregnant people, stay up to date on COVID-19 vaccines. A COVID-19 vaccine can be given during any trimester of pregnancy. CDC recommends an annual flu vaccine during each flu season (fall/winter) for everyone 6 months or older in the United States, including pregnant people. A flu vaccine can be given during any trimester of pregnancy. For individuals who will be between 32 and 36 weeks pregnant during September through January, CDC recommends getting an RSV vaccine. RSV season and timing of vaccination may vary depending on geography. If a pregnant patient does not get the RSV vaccine during their pregnancy, CDC recommends that their baby receive an RSV monoclonal antibody (nirsevimab) to provide additional protection during the infant’s first RSV season, if they are younger than 8 months. At this time, pregnant people who received an RSV vaccine during a previous pregnancy (last year) are not recommended to receive another RSV vaccine during pregnancy. The current recommendation is for babies born during subsequent pregnancies to receive nirsevimab. Some pregnant people may also need other vaccines, such as hepatitis B. 

How can practitioners approach conversations about vaccination during pregnancy amid increasing vaccine hesitancy?

Many pregnant people who do get vaccinated describe their provider’s recommendation as an important motivator toward vaccination.^8-11 Communications research suggests that practitioners can further increase trustworthiness by openly discussing potential side effects of prenatal vaccinations and providing patients with a rationale for why each vaccine is recommended. Practitioners can also utilize opportunities to communicate that the risk for severe illness from whooping cough, COVID-19, flu, and RSV in pregnancy and among neonates in the first few months of life is often higher than the risk for an adverse reaction from receiving ACIP-recommended vaccines. Finally, practitioners can consider sharing tested and refined patient education resources at least one appointment prior to the recommended administration of each vaccine, providing individuals with time to process the information they need to facilitate their vaccine decision-making process.

Some patients may be more comfortable with older, well-known prenatal vaccinations but have skepticism about newer vaccines like COVID-19 and RSV. How can practitioners respond to these concerns?

As pregnant people navigate the challenges of making health decisions that could impact their developing baby, practitioners can build trust through empathetically responding to safety concerns and questions, particularly with respect to newly authorized vaccines. Vaccine confidence may be strengthened by communicating to patients that all recommended vaccinations, including those that have been newly authorized, have been rigorously tested prior to being recommended for pregnant people. Additionally, in my clinical practice, I see that patients are often more comfortable accepting vaccines when the benefit for the baby is clearly communicated. I have been pleasantly surprised that most patients I have counseled on the new maternal RSV vaccine have been receptive, making statements like, “If this will help protect my baby from getting sick, then yes, I will get it.”

As you and your staff care for pregnant patients during fall and winter virus season, remember that a provider recommendation remains one of the strongest known predictors of vaccination uptake.¹² As a trusted source of information about prenatal vaccination, consider further incorporating patient education resources to help communicate how prenatal vaccination helps pregnant people share important protection against severe illnesses with their babies. 

Haben Debessai, MD, is a Gilstrap Fellow at the CDC Foundation. Debessai also serves as an Emory Obstetrics/Gynecology Adjunct Instructor at Grady Health System in Atlanta, Georgia. She disclosed no relevant conflicts of interest.

References

1. ACOG Committee Opinion No. 741: Maternal Immunization. Obstet Gynecol. 2018;131:e214-e217. doi:10.1097/AOG.0000000000002662

2. Centers for Disease Control and Prevention. Flu, Tdap, and COVID-19 vaccination coverage among pregnant women – United States, April 2024. 2024 Sep 23. 3. Centers for Disease Control and Prevention. Respiratory syncytial virus (rsv) vaccination coverage, pregnant persons. 2024 Nov 19. 4. Centers for Disease Control and Prevention. COVID-19 vaccination coverage, pregnant persons. 2024 Nov 19. 5. Centers for Disease Control and Prevention. Influenza vaccination coverage, pregnant persons. 2024 Nov 19.6. Razzaghi H et al. IMMWR Morb Mortal Wkly Rep. 2023;72:1065-1071. Published 2023 Sep 29. doi: 10.15585/mmwr.mm7239a4

7. Mohamoud YA et al. MMWR Morb Mortal Wkly Rep 2023;72:961-967. doi: https://dx.doi.org/10.15585/mmwr.mm7235e1.

8. Kiefer MK et al. Am J Obstet Gynecol MFM. 2022;4:100603. doi: 10.1016/j.ajogmf.2022.100603

9. Spires B et al. Obstet Gynecol Clin North Am. 2023;50:401-419. doi: 10.1016/j.ogc.2023.02.013

10. Wales DP et al. Public Health. 2020;179:38-44. doi: 10.1016/j.puhe.2019.10.001

11. Zimmerman M et al. J Natl Med Assoc. 2023;115:362-376. doi:10.1016/j.jnma.2023.04.003

12. Castillo E et al. Best Pract Res Clin Obstet Gynaecol. 2021;76:83-95. doi:10.1016/j.bpobgyn.2021.03.008

The story of antimicrobial peptides (AMPs), particularly in tackling antibiotic resistance, has been one of false dawns and unfulfilled promises. But perhaps a new generation of “smarter” compounds could see them find a wider role in clinical practice, said experts.

AMPs may be small molecules, consisting of short chains of amino acids, but these naturally occurring compounds have an important function: They are the “frontline defense” against invasive bacteria, said Henrik Franzyk, MSc Engineering, associate professor in the Department of Drug Design and Pharmacology at the University of Copenhagen in Denmark.

 

Multifunction Line of Defense

AMPs are cationic, meaning they are positively charged. “The reason why nature has maintained these molecules is that all the microbes out there have a negative surface charge,” explained Hans-Georg Sahl, PhD, emeritus professor of pharmaceutical microbiology at the University of Bonn in Germany.

While AMPs are also hydrophobic, they are often amphipathic, with both hydrophobic and hydrophilic regions that allow them to target cell membranes and cause them to rupture similarly to how detergent acts. 

“Thus, the content of a cell gets released, and it destroys the pathogen,” explained Paulina Szymczak, a PhD candidate in the Institute of AI for Health at Helmholtz Munich, Neuherberg, Germany.

“There are variations of that theme,” said Eefjan Breukink, PhD, professor of microbial membranes and antibiotics at Utrecht University in the Netherlands. “And then it depends on the sequence of the particular peptide,” as some can cross the cell membrane and damage the bacterium internally.

Szymczak explained that AMPs can, in this way, target the cell DNA, as both the membrane and the DNA are negatively charged. “That’s also what makes them so powerful because they don’t have just one mechanism of action, as opposed to conventional antibiotics.” 

 

Indiscriminate Killers

But they also have another crucial function. They activate the innate immune system via so-called resident immune cells that are “sitting in the tissues and waiting for bacteria to turn up,” explained Franzyk.

“The problem with antibodies is that they typically need to replicate,” he continued, which takes between 4 and 7 days — a timeline that is much better suited to tackling a viral infection. Bacteria, on the other hand, have a replication cycle of just 30 minutes.

Another big problem is that AMPs kill cells indiscriminately, including our own.

“But the human body is clever in that it only produces these antimicrobial peptides where the bacteria are, so they are not circulating in the blood,” said Franzyk. If a small part of tissue becomes infected, the innate immune cells start producing AMPs, which may kill the bacteria, or call on other immune cells to help.

As part of this process, “they will also kill part of our own tissue, but that’s the price we have to pay,” he said.

 

Local Applications

It is this aspect that has, so far, limited the use of AMPs in clinical practice, certainly as a replacement for conventional antibiotics limited by bacterial resistance. The trials conducted so far have been, by and large, negative, which has dampened enthusiasm and led to the perception that the risk they pose is too great for large-scale investment.

AMPs “are not made for what we need from antibiotics in the first place,” explained Sahl. “That is, a nice, easy distribution in the body, going into abscesses” and throughout the tissues.

He continued that AMPs are “more about controlling the flora in our bodies,” and they are “really not made for being used systemically.” 

Szymczak and colleagues are now working on designing active peptides with a strong antibacterial profile but limited toxicity for systematic use.

However, the “downside with these peptides is that they are not orally available, so you can’t take a pill,” Breukink said, but instead they need to be administered intravenously.

There are, nevertheless, some antibiotics in clinical use that have the same molecular features as AMPs. These include colistin, a last-resort treatment for multidrug-resistant gram-negative bacteria, and daptomycin, which is used in the treatment of systemic infections caused by gram-positive species.

Szymczak added that there have been successes in using AMPs in a more targeted way, such as using a topical cream. Another potentially promising avenue is lung infections, which are being studied in mouse models.

 

Less Prone to Resistance

Crucially, AMPs are markedly less prone to bacterial resistance than conventional antibiotics, partly because of their typical target: the cell membrane.

“Biologically and evolutionarily, it is a very costly operation to rebuild the membrane and change its charge,” Szymczak explained. “It’s quite hard for bacteria to learn this because it’s not a single protein that you have to mutate but the whole membrane.”

This is seen in the laboratory, where it takes around five generations, or passages, for bacteria to develop resistance when grown in the presence of antibiotics, but up to 40 passages when cultured with an AMP.

The limits of the ability of AMPs to withstand the development of bacterial resistance have been tested in the real world.

Colistin has been used widely in Asia as a growth promoter, especially in pig farming. Franzyk explained that farmers have used enormous quantities of this AMP-based antibiotic, which has indeed led to the development of resistance, including contamination of meat for human consumption, leading to resistance spreading to other parts of the world.

“The bad thing about this is it’s not something each individual bacteria needs to acquire,” he said. Because resistance is stored on small, cyclic DNA called plasmids, it “can be transferred from one bacterial species to another.”

 

Novel Avenues

Franzyk suggested that AMPs could nevertheless be used in combination with, or to modify, existing antibiotics to revitalize those for which there is already bacterial resistance, or to allow antibiotics that ordinarily target only gram-positive bacteria to also treat gram-negative infections, for example.

Szymczak and her colleagues are using artificial intelligence to design novel AMP candidates. Instead of manually going through compounds and checking their activity profiles in the lab, those steps are carried out computationally “so that, in the end, you synthesize as few candidates as possible” and can proceed to a mouse model “as fast as possible.”

She personally is looking at the issue of strain-specific activity to design a compound that would target, for example, only multidrug-resistant strains. “What we can do now is something that will target everything, so a kind of last resort peptide. But we are trying to make them smarter in their targets.”

Szymczak also pointed out that cancer cells are “negatively charged, similarly to bacterial cells, as opposed to mammalian cells, which are neutral.”

“So in theory, maybe we could design something that will target cancer cells but not our host cells, and that would be extremely exciting.” However, she underlined that, first, they are trying to tackle antimicrobial resistance before looking at other spaces.

Finally, Breukink is screening for small antibacterial compounds in fungi that are around half the size of a normal peptide and more hydrophobic, meaning there is a much greater chance of them being orally available.

But “you first have to test, of course,” he said, as “if you don’t have specific targets, then you will get problems with toxicity, or other issues that you do not foresee.” 

No funding was declared. No relevant financial relationships were declared.

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

The story of antimicrobial peptides (AMPs), particularly in tackling antibiotic resistance, has been one of false dawns and unfulfilled promises. But perhaps a new generation of “smarter” compounds could see them find a wider role in clinical practice, said experts.

AMPs may be small molecules, consisting of short chains of amino acids, but these naturally occurring compounds have an important function: They are the “frontline defense” against invasive bacteria, said Henrik Franzyk, MSc Engineering, associate professor in the Department of Drug Design and Pharmacology at the University of Copenhagen in Denmark.

 

Multifunction Line of Defense

AMPs are cationic, meaning they are positively charged. “The reason why nature has maintained these molecules is that all the microbes out there have a negative surface charge,” explained Hans-Georg Sahl, PhD, emeritus professor of pharmaceutical microbiology at the University of Bonn in Germany.

While AMPs are also hydrophobic, they are often amphipathic, with both hydrophobic and hydrophilic regions that allow them to target cell membranes and cause them to rupture similarly to how detergent acts. 

“Thus, the content of a cell gets released, and it destroys the pathogen,” explained Paulina Szymczak, a PhD candidate in the Institute of AI for Health at Helmholtz Munich, Neuherberg, Germany.

“There are variations of that theme,” said Eefjan Breukink, PhD, professor of microbial membranes and antibiotics at Utrecht University in the Netherlands. “And then it depends on the sequence of the particular peptide,” as some can cross the cell membrane and damage the bacterium internally.

Szymczak explained that AMPs can, in this way, target the cell DNA, as both the membrane and the DNA are negatively charged. “That’s also what makes them so powerful because they don’t have just one mechanism of action, as opposed to conventional antibiotics.” 

 

Indiscriminate Killers

But they also have another crucial function. They activate the innate immune system via so-called resident immune cells that are “sitting in the tissues and waiting for bacteria to turn up,” explained Franzyk.

“The problem with antibodies is that they typically need to replicate,” he continued, which takes between 4 and 7 days — a timeline that is much better suited to tackling a viral infection. Bacteria, on the other hand, have a replication cycle of just 30 minutes.

Another big problem is that AMPs kill cells indiscriminately, including our own.

“But the human body is clever in that it only produces these antimicrobial peptides where the bacteria are, so they are not circulating in the blood,” said Franzyk. If a small part of tissue becomes infected, the innate immune cells start producing AMPs, which may kill the bacteria, or call on other immune cells to help.

As part of this process, “they will also kill part of our own tissue, but that’s the price we have to pay,” he said.

 

Local Applications

It is this aspect that has, so far, limited the use of AMPs in clinical practice, certainly as a replacement for conventional antibiotics limited by bacterial resistance. The trials conducted so far have been, by and large, negative, which has dampened enthusiasm and led to the perception that the risk they pose is too great for large-scale investment.

AMPs “are not made for what we need from antibiotics in the first place,” explained Sahl. “That is, a nice, easy distribution in the body, going into abscesses” and throughout the tissues.

He continued that AMPs are “more about controlling the flora in our bodies,” and they are “really not made for being used systemically.” 

Szymczak and colleagues are now working on designing active peptides with a strong antibacterial profile but limited toxicity for systematic use.

However, the “downside with these peptides is that they are not orally available, so you can’t take a pill,” Breukink said, but instead they need to be administered intravenously.

There are, nevertheless, some antibiotics in clinical use that have the same molecular features as AMPs. These include colistin, a last-resort treatment for multidrug-resistant gram-negative bacteria, and daptomycin, which is used in the treatment of systemic infections caused by gram-positive species.

Szymczak added that there have been successes in using AMPs in a more targeted way, such as using a topical cream. Another potentially promising avenue is lung infections, which are being studied in mouse models.

 

Less Prone to Resistance

Crucially, AMPs are markedly less prone to bacterial resistance than conventional antibiotics, partly because of their typical target: the cell membrane.

“Biologically and evolutionarily, it is a very costly operation to rebuild the membrane and change its charge,” Szymczak explained. “It’s quite hard for bacteria to learn this because it’s not a single protein that you have to mutate but the whole membrane.”

This is seen in the laboratory, where it takes around five generations, or passages, for bacteria to develop resistance when grown in the presence of antibiotics, but up to 40 passages when cultured with an AMP.

The limits of the ability of AMPs to withstand the development of bacterial resistance have been tested in the real world.

Colistin has been used widely in Asia as a growth promoter, especially in pig farming. Franzyk explained that farmers have used enormous quantities of this AMP-based antibiotic, which has indeed led to the development of resistance, including contamination of meat for human consumption, leading to resistance spreading to other parts of the world.

“The bad thing about this is it’s not something each individual bacteria needs to acquire,” he said. Because resistance is stored on small, cyclic DNA called plasmids, it “can be transferred from one bacterial species to another.”

 

Novel Avenues

Franzyk suggested that AMPs could nevertheless be used in combination with, or to modify, existing antibiotics to revitalize those for which there is already bacterial resistance, or to allow antibiotics that ordinarily target only gram-positive bacteria to also treat gram-negative infections, for example.

Szymczak and her colleagues are using artificial intelligence to design novel AMP candidates. Instead of manually going through compounds and checking their activity profiles in the lab, those steps are carried out computationally “so that, in the end, you synthesize as few candidates as possible” and can proceed to a mouse model “as fast as possible.”

She personally is looking at the issue of strain-specific activity to design a compound that would target, for example, only multidrug-resistant strains. “What we can do now is something that will target everything, so a kind of last resort peptide. But we are trying to make them smarter in their targets.”

Szymczak also pointed out that cancer cells are “negatively charged, similarly to bacterial cells, as opposed to mammalian cells, which are neutral.”

“So in theory, maybe we could design something that will target cancer cells but not our host cells, and that would be extremely exciting.” However, she underlined that, first, they are trying to tackle antimicrobial resistance before looking at other spaces.

Finally, Breukink is screening for small antibacterial compounds in fungi that are around half the size of a normal peptide and more hydrophobic, meaning there is a much greater chance of them being orally available.

But “you first have to test, of course,” he said, as “if you don’t have specific targets, then you will get problems with toxicity, or other issues that you do not foresee.” 

No funding was declared. No relevant financial relationships were declared.

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

The story of antimicrobial peptides (AMPs), particularly in tackling antibiotic resistance, has been one of false dawns and unfulfilled promises. But perhaps a new generation of “smarter” compounds could see them find a wider role in clinical practice, said experts.

AMPs may be small molecules, consisting of short chains of amino acids, but these naturally occurring compounds have an important function: They are the “frontline defense” against invasive bacteria, said Henrik Franzyk, MSc Engineering, associate professor in the Department of Drug Design and Pharmacology at the University of Copenhagen in Denmark.

 

Multifunction Line of Defense

AMPs are cationic, meaning they are positively charged. “The reason why nature has maintained these molecules is that all the microbes out there have a negative surface charge,” explained Hans-Georg Sahl, PhD, emeritus professor of pharmaceutical microbiology at the University of Bonn in Germany.

While AMPs are also hydrophobic, they are often amphipathic, with both hydrophobic and hydrophilic regions that allow them to target cell membranes and cause them to rupture similarly to how detergent acts. 

“Thus, the content of a cell gets released, and it destroys the pathogen,” explained Paulina Szymczak, a PhD candidate in the Institute of AI for Health at Helmholtz Munich, Neuherberg, Germany.

“There are variations of that theme,” said Eefjan Breukink, PhD, professor of microbial membranes and antibiotics at Utrecht University in the Netherlands. “And then it depends on the sequence of the particular peptide,” as some can cross the cell membrane and damage the bacterium internally.

Szymczak explained that AMPs can, in this way, target the cell DNA, as both the membrane and the DNA are negatively charged. “That’s also what makes them so powerful because they don’t have just one mechanism of action, as opposed to conventional antibiotics.” 

 

Indiscriminate Killers

But they also have another crucial function. They activate the innate immune system via so-called resident immune cells that are “sitting in the tissues and waiting for bacteria to turn up,” explained Franzyk.

“The problem with antibodies is that they typically need to replicate,” he continued, which takes between 4 and 7 days — a timeline that is much better suited to tackling a viral infection. Bacteria, on the other hand, have a replication cycle of just 30 minutes.

Another big problem is that AMPs kill cells indiscriminately, including our own.

“But the human body is clever in that it only produces these antimicrobial peptides where the bacteria are, so they are not circulating in the blood,” said Franzyk. If a small part of tissue becomes infected, the innate immune cells start producing AMPs, which may kill the bacteria, or call on other immune cells to help.

As part of this process, “they will also kill part of our own tissue, but that’s the price we have to pay,” he said.

 

Local Applications

It is this aspect that has, so far, limited the use of AMPs in clinical practice, certainly as a replacement for conventional antibiotics limited by bacterial resistance. The trials conducted so far have been, by and large, negative, which has dampened enthusiasm and led to the perception that the risk they pose is too great for large-scale investment.

AMPs “are not made for what we need from antibiotics in the first place,” explained Sahl. “That is, a nice, easy distribution in the body, going into abscesses” and throughout the tissues.

He continued that AMPs are “more about controlling the flora in our bodies,” and they are “really not made for being used systemically.” 

Szymczak and colleagues are now working on designing active peptides with a strong antibacterial profile but limited toxicity for systematic use.

However, the “downside with these peptides is that they are not orally available, so you can’t take a pill,” Breukink said, but instead they need to be administered intravenously.

There are, nevertheless, some antibiotics in clinical use that have the same molecular features as AMPs. These include colistin, a last-resort treatment for multidrug-resistant gram-negative bacteria, and daptomycin, which is used in the treatment of systemic infections caused by gram-positive species.

Szymczak added that there have been successes in using AMPs in a more targeted way, such as using a topical cream. Another potentially promising avenue is lung infections, which are being studied in mouse models.

 

Less Prone to Resistance

Crucially, AMPs are markedly less prone to bacterial resistance than conventional antibiotics, partly because of their typical target: the cell membrane.

“Biologically and evolutionarily, it is a very costly operation to rebuild the membrane and change its charge,” Szymczak explained. “It’s quite hard for bacteria to learn this because it’s not a single protein that you have to mutate but the whole membrane.”

This is seen in the laboratory, where it takes around five generations, or passages, for bacteria to develop resistance when grown in the presence of antibiotics, but up to 40 passages when cultured with an AMP.

The limits of the ability of AMPs to withstand the development of bacterial resistance have been tested in the real world.

Colistin has been used widely in Asia as a growth promoter, especially in pig farming. Franzyk explained that farmers have used enormous quantities of this AMP-based antibiotic, which has indeed led to the development of resistance, including contamination of meat for human consumption, leading to resistance spreading to other parts of the world.

“The bad thing about this is it’s not something each individual bacteria needs to acquire,” he said. Because resistance is stored on small, cyclic DNA called plasmids, it “can be transferred from one bacterial species to another.”

 

Novel Avenues

Franzyk suggested that AMPs could nevertheless be used in combination with, or to modify, existing antibiotics to revitalize those for which there is already bacterial resistance, or to allow antibiotics that ordinarily target only gram-positive bacteria to also treat gram-negative infections, for example.

Szymczak and her colleagues are using artificial intelligence to design novel AMP candidates. Instead of manually going through compounds and checking their activity profiles in the lab, those steps are carried out computationally “so that, in the end, you synthesize as few candidates as possible” and can proceed to a mouse model “as fast as possible.”

She personally is looking at the issue of strain-specific activity to design a compound that would target, for example, only multidrug-resistant strains. “What we can do now is something that will target everything, so a kind of last resort peptide. But we are trying to make them smarter in their targets.”

Szymczak also pointed out that cancer cells are “negatively charged, similarly to bacterial cells, as opposed to mammalian cells, which are neutral.”

“So in theory, maybe we could design something that will target cancer cells but not our host cells, and that would be extremely exciting.” However, she underlined that, first, they are trying to tackle antimicrobial resistance before looking at other spaces.

Finally, Breukink is screening for small antibacterial compounds in fungi that are around half the size of a normal peptide and more hydrophobic, meaning there is a much greater chance of them being orally available.

But “you first have to test, of course,” he said, as “if you don’t have specific targets, then you will get problems with toxicity, or other issues that you do not foresee.” 

No funding was declared. No relevant financial relationships were declared.

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

TOPLINE:

Early treatment with oseltamivir on the same day as hospital admission was associated with fewer severe clinical outcomes, such as worsening pulmonary disease, need for invasive ventilation, organ failure, and in-hospital death in adults hospitalized with influenza. 

METHODOLOGY:

• The 2018 guidelines from the Infectious Disease Society of America recommend prompt administration of oseltamivir to hospitalized patients with suspected or confirmed influenza, regardless of the time of symptom onset; however, variations in treatment practices and circulating virus strains may affect the effectiveness of this practice guideline.
• Researchers conducted a multicenter observational study across 24 hospitals in the United States during the 2022-2023 flu season to assess the benefits of initiating oseltamivir treatment on the same day as hospital admission for adults with acute influenza, compared with late or no treatment.
• They included 840 adults (age, ≥ 18 years) with laboratory-confirmed influenza, of which 415 patients initiated oseltamivir on the same day as hospital admission (early treatment).
• Among the 425 patients in the late/no treatment group, most (78%) received oseltamivir 1 day after admission, while 124 did not receive oseltamivir at all.
• The primary outcome was the peak pulmonary disease severity level that patients experienced during hospitalization, and secondary outcomes included hospital length of stay, ICU admission, initiation of extrapulmonary organ support using vasopressors or kidney replacement therapy, and in-hospital death.

TAKEAWAY:

• Patients in the early treatment group were less likely to experience progression and severe progression of pulmonary disease after the day of hospital admission, compared with those in the late or no treatment group (P < .001 and P = .027, respectively).
• Patients who received early oseltamivir treatment had 40% lower peak pulmonary disease severity than those who received late or no treatment (proportional adjusted odds ratio [paOR], 0.60; 95% CI, 0.49-0.72).
• They also showed lower odds of ICU admission (aOR, 0.25; 95% CI, 0.13-0.49) and use of acute kidney replacement therapy or vasopressors (aOR, 0.40; 95% CI, 0.22-0.67).
• Those in the early treatment group also had a shorter hospital length of stay (median, 4 days vs 4 days) and faced a 64% lower risk for in-hospital mortality (aOR, 0.36; 95% CI, 0.19-0.69) compared with those in the late or no treatment group.

IN PRACTICE:

“These findings support current recommendations, such as the IDSA [Infectious Disease Society of America] Influenza Clinical Practice Guidelines and CDC [Centers for Disease Control and Prevention] guidance, to initiate oseltamivir treatment as soon as possible for adult patients hospitalized with influenza,” the authors wrote.

SOURCE:

The study was led by Nathaniel M. Lewis, PhD, Influenza Division, CDC, Atlanta, Georgia, and was published online  in Clinical Infectious Diseases.

LIMITATIONS:

This study may not be generalizable to seasons when influenza A(H1N1)pdm09 or B viruses are predominant as it was conducted during an influenza A(H3N2) virus–predominant season. The study lacked sufficient power to examine various oseltamivir treatment initiation timepoints or identify a potential maximum time-to-treatment threshold for effectiveness. Moreover, variables such as outpatient antiviral treatment before hospital admission and other treatments using macrolides, statins, corticosteroids, or immunomodulators before or during hospitalization were not collected, which may have influenced the study findings.

DISCLOSURES:

The study received funding from the CDC and the National Center for Immunization and Respiratory Diseases. Some authors reported receiving research support, consulting fees, funding, grants, or fees for participation in an advisory board and having other ties with certain institutions and pharmaceutical companies.

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:

Early treatment with oseltamivir on the same day as hospital admission was associated with fewer severe clinical outcomes, such as worsening pulmonary disease, need for invasive ventilation, organ failure, and in-hospital death in adults hospitalized with influenza. 

METHODOLOGY:

• The 2018 guidelines from the Infectious Disease Society of America recommend prompt administration of oseltamivir to hospitalized patients with suspected or confirmed influenza, regardless of the time of symptom onset; however, variations in treatment practices and circulating virus strains may affect the effectiveness of this practice guideline.
• Researchers conducted a multicenter observational study across 24 hospitals in the United States during the 2022-2023 flu season to assess the benefits of initiating oseltamivir treatment on the same day as hospital admission for adults with acute influenza, compared with late or no treatment.
• They included 840 adults (age, ≥ 18 years) with laboratory-confirmed influenza, of which 415 patients initiated oseltamivir on the same day as hospital admission (early treatment).
• Among the 425 patients in the late/no treatment group, most (78%) received oseltamivir 1 day after admission, while 124 did not receive oseltamivir at all.
• The primary outcome was the peak pulmonary disease severity level that patients experienced during hospitalization, and secondary outcomes included hospital length of stay, ICU admission, initiation of extrapulmonary organ support using vasopressors or kidney replacement therapy, and in-hospital death.

TAKEAWAY:

• Patients in the early treatment group were less likely to experience progression and severe progression of pulmonary disease after the day of hospital admission, compared with those in the late or no treatment group (P < .001 and P = .027, respectively).
• Patients who received early oseltamivir treatment had 40% lower peak pulmonary disease severity than those who received late or no treatment (proportional adjusted odds ratio [paOR], 0.60; 95% CI, 0.49-0.72).
• They also showed lower odds of ICU admission (aOR, 0.25; 95% CI, 0.13-0.49) and use of acute kidney replacement therapy or vasopressors (aOR, 0.40; 95% CI, 0.22-0.67).
• Those in the early treatment group also had a shorter hospital length of stay (median, 4 days vs 4 days) and faced a 64% lower risk for in-hospital mortality (aOR, 0.36; 95% CI, 0.19-0.69) compared with those in the late or no treatment group.

IN PRACTICE:

“These findings support current recommendations, such as the IDSA [Infectious Disease Society of America] Influenza Clinical Practice Guidelines and CDC [Centers for Disease Control and Prevention] guidance, to initiate oseltamivir treatment as soon as possible for adult patients hospitalized with influenza,” the authors wrote.

SOURCE:

The study was led by Nathaniel M. Lewis, PhD, Influenza Division, CDC, Atlanta, Georgia, and was published online  in Clinical Infectious Diseases.

LIMITATIONS:

This study may not be generalizable to seasons when influenza A(H1N1)pdm09 or B viruses are predominant as it was conducted during an influenza A(H3N2) virus–predominant season. The study lacked sufficient power to examine various oseltamivir treatment initiation timepoints or identify a potential maximum time-to-treatment threshold for effectiveness. Moreover, variables such as outpatient antiviral treatment before hospital admission and other treatments using macrolides, statins, corticosteroids, or immunomodulators before or during hospitalization were not collected, which may have influenced the study findings.

DISCLOSURES:

The study received funding from the CDC and the National Center for Immunization and Respiratory Diseases. Some authors reported receiving research support, consulting fees, funding, grants, or fees for participation in an advisory board and having other ties with certain institutions and pharmaceutical companies.

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:

Early treatment with oseltamivir on the same day as hospital admission was associated with fewer severe clinical outcomes, such as worsening pulmonary disease, need for invasive ventilation, organ failure, and in-hospital death in adults hospitalized with influenza. 

METHODOLOGY:

• The 2018 guidelines from the Infectious Disease Society of America recommend prompt administration of oseltamivir to hospitalized patients with suspected or confirmed influenza, regardless of the time of symptom onset; however, variations in treatment practices and circulating virus strains may affect the effectiveness of this practice guideline.
• Researchers conducted a multicenter observational study across 24 hospitals in the United States during the 2022-2023 flu season to assess the benefits of initiating oseltamivir treatment on the same day as hospital admission for adults with acute influenza, compared with late or no treatment.
• They included 840 adults (age, ≥ 18 years) with laboratory-confirmed influenza, of which 415 patients initiated oseltamivir on the same day as hospital admission (early treatment).
• Among the 425 patients in the late/no treatment group, most (78%) received oseltamivir 1 day after admission, while 124 did not receive oseltamivir at all.
• The primary outcome was the peak pulmonary disease severity level that patients experienced during hospitalization, and secondary outcomes included hospital length of stay, ICU admission, initiation of extrapulmonary organ support using vasopressors or kidney replacement therapy, and in-hospital death.

TAKEAWAY:

• Patients in the early treatment group were less likely to experience progression and severe progression of pulmonary disease after the day of hospital admission, compared with those in the late or no treatment group (P < .001 and P = .027, respectively).
• Patients who received early oseltamivir treatment had 40% lower peak pulmonary disease severity than those who received late or no treatment (proportional adjusted odds ratio [paOR], 0.60; 95% CI, 0.49-0.72).
• They also showed lower odds of ICU admission (aOR, 0.25; 95% CI, 0.13-0.49) and use of acute kidney replacement therapy or vasopressors (aOR, 0.40; 95% CI, 0.22-0.67).
• Those in the early treatment group also had a shorter hospital length of stay (median, 4 days vs 4 days) and faced a 64% lower risk for in-hospital mortality (aOR, 0.36; 95% CI, 0.19-0.69) compared with those in the late or no treatment group.

IN PRACTICE:

“These findings support current recommendations, such as the IDSA [Infectious Disease Society of America] Influenza Clinical Practice Guidelines and CDC [Centers for Disease Control and Prevention] guidance, to initiate oseltamivir treatment as soon as possible for adult patients hospitalized with influenza,” the authors wrote.

SOURCE:

The study was led by Nathaniel M. Lewis, PhD, Influenza Division, CDC, Atlanta, Georgia, and was published online  in Clinical Infectious Diseases.

LIMITATIONS:

This study may not be generalizable to seasons when influenza A(H1N1)pdm09 or B viruses are predominant as it was conducted during an influenza A(H3N2) virus–predominant season. The study lacked sufficient power to examine various oseltamivir treatment initiation timepoints or identify a potential maximum time-to-treatment threshold for effectiveness. Moreover, variables such as outpatient antiviral treatment before hospital admission and other treatments using macrolides, statins, corticosteroids, or immunomodulators before or during hospitalization were not collected, which may have influenced the study findings.

DISCLOSURES:

The study received funding from the CDC and the National Center for Immunization and Respiratory Diseases. Some authors reported receiving research support, consulting fees, funding, grants, or fees for participation in an advisory board and having other ties with certain institutions and pharmaceutical companies.

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.

Instead of sitting behind a laptop during patient visits, the pediatrician directly faces the patient and parent, relying on an ambient artificial intelligence (AI) scribe to capture the conversation for the electronic health record (EHR). A geriatrician doing rounds at the senior living facility plugs each patient’s medications into an AI tool, checking for drug interactions. And a busy hospital radiology department runs all its emergency head CTs through an AI algorithm, triaging potential stroke patients to ensure they receive the highest priority. None of these physicians have been sued for malpractice for AI usage, but they wonder if they’re at risk.

In a recent Medscape report, AI Adoption in Healthcare, 224 physicians responded to the statement: “I want to do more with AI but I worry about malpractice risk if I move too fast.” Seventeen percent said that they strongly agreed while 23% said they agreed — a full 40% were concerned about using the technology for legal reasons.  

Malpractice and AI are on many minds in healthcare, especially in large health systems, Deepika Srivastava, chief operating officer at The Doctors Company, told this news organization. “AI is at the forefront of the conversation, and they’re [large health systems] raising questions. Larger systems want to protect themselves.” 

The good news is there’s currently no sign of legal action over the clinical use of AI. “We’re not seeing even a few AI-related suits just yet,” but the risk is growing, Srivastava said, “and that’s why we’re talking about it. The legal system will need to adapt to address the role of AI in healthcare.”

 

How Doctors Are Using AI

Healthcare is incorporating AI in multiple ways based on the type of tool and function needed. Narrow AI is popular in fields like radiology, comparing two large data sets to find differences between them. Narrow AI can help differentiate between normal and abnormal tissue, such as breast or lung tumors. Almost 900 AI health tools have Food and Drug Administration approval as of July 2024, discerning abnormalities and recognizing patterns better than many humans, said Robert Pearl, MD, author of ChatGPT, MD: How AI-Empowered Patients & Doctors Can Take Back Control of American Medicine and former CEO of The Permanente Medical Group.

Narrow AI can improve diagnostic speed and accuracy for other specialties, too, including dermatology and ophthalmology, Pearl said. “It’s less clear to me if it will be very beneficial in primary care, neurology, and psychiatry, areas of medicine that involve a lot of words.” In those specialties, some may use generative AI as a repository of resources. In clinical practice, ambient AI is also used to create health records based on patient/clinician conversations.

In clinical administration, AI is used for scheduling, billing, and submitting insurance claims. On the insurer side, denying claims based on AI algorithms has been at the heart of legal actions, making recent headlines. 

 

Malpractice Risks When Using AI

Accuracy and privacy should be at the top of the list for malpractice concerns with AI. With accuracy, liability could partially be determined by use type. If a diagnostic application makes the wrong diagnosis, “the company has legal accountability because it created and had to test it specific to the application that it’s being recommended for,” Pearl said. 

However, keeping a human in the loop is a smart move when using AI diagnostic tools. The physician should still choose the AI-suggested diagnosis or a different one. If it’s the wrong diagnosis, “it’s really hard to currently say where is the source of the error? Was it the physician? Was it the tool?” Srivastava added.

With an incorrect diagnosis by generative AI, liability is more apparent. “You’re taking that accountability,” Pearl said. Generative AI operates in a black box, predicting the correct answer based on information stored in a database. “Generative AI tries to draw a correlation between what it has seen and predicting the next output,” said Alex Shahrestani, managing partner of Promise Legal PLLC, a law firm in Austin, Texas. He serves on the State Bar of Texas’s Taskforce on AI and the Law and has participated in advisory groups related to AI policies with the National Institute of Standards and Technology. “A doctor should know to validate information given back to them by AI,” applying their own medical training and judgment.

Generative AI can provide ideas. Pearl shared a story about a surgeon who was unable to remove a breathing tube that was stuck in a patients’ throat at the end of a procedure. The surgeon checked ChatGPT in the operating room, finding a similar case. Adrenaline in the anesthetic restricted the blood vessels, causing the vocal cords to stick together. Following the AI information, the surgeon allowed more time for the anesthesia to diffuse. As it wore off, the vocal cords separated, easing the removal of the breathing tube. “That is the kind of expertise it can provide,” Pearl said.

Privacy is a common AI concern, but it may be more problematic than it should be. “Many think if you talk to an AI system, you’re surrendering personal information the model can learn from,” said Shahrestani. Platforms offer opt-outs. Even without opting out, the model won’t automatically ingest your interactions. That’s not a privacy feature, but a concern by the developer that the information may not help the model. 

“If you do use these opt-out mechanisms, and you have the requisite amount of confidentiality, you can use ChatGPT without too much concern about the patient information being released into the wild,” Shahrestani said. Or use systems with stricter requirements that keep all data on site.

 

Malpractice Insurance Policies and AI

Currently, malpractice policies do not specify AI coverage. “We don’t ask right now to list all the technology you’re using,” said Srivastava. Many EHR systems already incorporate AI. If a human provider is in the loop, already vetted and insured, “we should be okay when it comes to the risk of malpractice when doctors are using AI because it’s still the risk that we’re ensuring.”

Insurers are paying attention, though. “Traditional medical malpractice law does require re-evaluation because the rapid pace of AI development has outpaced the efforts to integrate it into the legal system,” Srivastava said.

Some, including Pearl, believe AI will actually lower the malpractice risk. Having more data points to consider can make doctors’ jobs faster, easier, and more accurate. “I believe the technology will decrease lawsuits, not increase them,” said Pearl.

 

Meanwhile, How Can Doctors Protect Themselves From an AI Malpractice Suit?

Know your tool: Providers should understand the tool they’re deploying, what it provides, how it was built and trained (including potential biases), how it was tested, and the guidelines for how to use it or not use it, said Srivastava. Evaluate each tool, use case, and risk separately. “Don’t just say it’s all AI.” 

With generative AI, users will have better success requesting information that has been available longer and is more widely accessed. “It’s more likely to come back correctly,” said Shahrestani. If the information sought is fairly new or not widespread, the tool may try to draw problematic conclusions. 

Document: “Document, document, document. Just making sure you have good documentation can really help you if litigation comes up and it’s related to the AI tools,” Srivastava said.

Try it out: “I recommend you use [generative AI] a lot so you understand its strengths and shortcomings,” said Shahrestani. “If you wait until things settle, you’ll be further behind.” 

Pretend you’re the patient and give the tool the information you’d give a doctor and see the results, said Pearl. It will provide you with an idea of what it can do. “No one would sue you because you went to the library to look up information in the textbooks,” he said — using generative AI is similar. Try the free versions first; if you begin relying on it more, the paid versions have better features and are inexpensive. 

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

Instead of sitting behind a laptop during patient visits, the pediatrician directly faces the patient and parent, relying on an ambient artificial intelligence (AI) scribe to capture the conversation for the electronic health record (EHR). A geriatrician doing rounds at the senior living facility plugs each patient’s medications into an AI tool, checking for drug interactions. And a busy hospital radiology department runs all its emergency head CTs through an AI algorithm, triaging potential stroke patients to ensure they receive the highest priority. None of these physicians have been sued for malpractice for AI usage, but they wonder if they’re at risk.

In a recent Medscape report, AI Adoption in Healthcare, 224 physicians responded to the statement: “I want to do more with AI but I worry about malpractice risk if I move too fast.” Seventeen percent said that they strongly agreed while 23% said they agreed — a full 40% were concerned about using the technology for legal reasons.  

Malpractice and AI are on many minds in healthcare, especially in large health systems, Deepika Srivastava, chief operating officer at The Doctors Company, told this news organization. “AI is at the forefront of the conversation, and they’re [large health systems] raising questions. Larger systems want to protect themselves.” 

The good news is there’s currently no sign of legal action over the clinical use of AI. “We’re not seeing even a few AI-related suits just yet,” but the risk is growing, Srivastava said, “and that’s why we’re talking about it. The legal system will need to adapt to address the role of AI in healthcare.”

 

How Doctors Are Using AI

Healthcare is incorporating AI in multiple ways based on the type of tool and function needed. Narrow AI is popular in fields like radiology, comparing two large data sets to find differences between them. Narrow AI can help differentiate between normal and abnormal tissue, such as breast or lung tumors. Almost 900 AI health tools have Food and Drug Administration approval as of July 2024, discerning abnormalities and recognizing patterns better than many humans, said Robert Pearl, MD, author of ChatGPT, MD: How AI-Empowered Patients & Doctors Can Take Back Control of American Medicine and former CEO of The Permanente Medical Group.

Narrow AI can improve diagnostic speed and accuracy for other specialties, too, including dermatology and ophthalmology, Pearl said. “It’s less clear to me if it will be very beneficial in primary care, neurology, and psychiatry, areas of medicine that involve a lot of words.” In those specialties, some may use generative AI as a repository of resources. In clinical practice, ambient AI is also used to create health records based on patient/clinician conversations.

In clinical administration, AI is used for scheduling, billing, and submitting insurance claims. On the insurer side, denying claims based on AI algorithms has been at the heart of legal actions, making recent headlines. 

 

Malpractice Risks When Using AI

Accuracy and privacy should be at the top of the list for malpractice concerns with AI. With accuracy, liability could partially be determined by use type. If a diagnostic application makes the wrong diagnosis, “the company has legal accountability because it created and had to test it specific to the application that it’s being recommended for,” Pearl said. 

However, keeping a human in the loop is a smart move when using AI diagnostic tools. The physician should still choose the AI-suggested diagnosis or a different one. If it’s the wrong diagnosis, “it’s really hard to currently say where is the source of the error? Was it the physician? Was it the tool?” Srivastava added.

With an incorrect diagnosis by generative AI, liability is more apparent. “You’re taking that accountability,” Pearl said. Generative AI operates in a black box, predicting the correct answer based on information stored in a database. “Generative AI tries to draw a correlation between what it has seen and predicting the next output,” said Alex Shahrestani, managing partner of Promise Legal PLLC, a law firm in Austin, Texas. He serves on the State Bar of Texas’s Taskforce on AI and the Law and has participated in advisory groups related to AI policies with the National Institute of Standards and Technology. “A doctor should know to validate information given back to them by AI,” applying their own medical training and judgment.

Generative AI can provide ideas. Pearl shared a story about a surgeon who was unable to remove a breathing tube that was stuck in a patients’ throat at the end of a procedure. The surgeon checked ChatGPT in the operating room, finding a similar case. Adrenaline in the anesthetic restricted the blood vessels, causing the vocal cords to stick together. Following the AI information, the surgeon allowed more time for the anesthesia to diffuse. As it wore off, the vocal cords separated, easing the removal of the breathing tube. “That is the kind of expertise it can provide,” Pearl said.

Privacy is a common AI concern, but it may be more problematic than it should be. “Many think if you talk to an AI system, you’re surrendering personal information the model can learn from,” said Shahrestani. Platforms offer opt-outs. Even without opting out, the model won’t automatically ingest your interactions. That’s not a privacy feature, but a concern by the developer that the information may not help the model. 

“If you do use these opt-out mechanisms, and you have the requisite amount of confidentiality, you can use ChatGPT without too much concern about the patient information being released into the wild,” Shahrestani said. Or use systems with stricter requirements that keep all data on site.

 

Malpractice Insurance Policies and AI

Currently, malpractice policies do not specify AI coverage. “We don’t ask right now to list all the technology you’re using,” said Srivastava. Many EHR systems already incorporate AI. If a human provider is in the loop, already vetted and insured, “we should be okay when it comes to the risk of malpractice when doctors are using AI because it’s still the risk that we’re ensuring.”

Insurers are paying attention, though. “Traditional medical malpractice law does require re-evaluation because the rapid pace of AI development has outpaced the efforts to integrate it into the legal system,” Srivastava said.

Some, including Pearl, believe AI will actually lower the malpractice risk. Having more data points to consider can make doctors’ jobs faster, easier, and more accurate. “I believe the technology will decrease lawsuits, not increase them,” said Pearl.

 

Meanwhile, How Can Doctors Protect Themselves From an AI Malpractice Suit?

Know your tool: Providers should understand the tool they’re deploying, what it provides, how it was built and trained (including potential biases), how it was tested, and the guidelines for how to use it or not use it, said Srivastava. Evaluate each tool, use case, and risk separately. “Don’t just say it’s all AI.” 

With generative AI, users will have better success requesting information that has been available longer and is more widely accessed. “It’s more likely to come back correctly,” said Shahrestani. If the information sought is fairly new or not widespread, the tool may try to draw problematic conclusions. 

Document: “Document, document, document. Just making sure you have good documentation can really help you if litigation comes up and it’s related to the AI tools,” Srivastava said.

Try it out: “I recommend you use [generative AI] a lot so you understand its strengths and shortcomings,” said Shahrestani. “If you wait until things settle, you’ll be further behind.” 

Pretend you’re the patient and give the tool the information you’d give a doctor and see the results, said Pearl. It will provide you with an idea of what it can do. “No one would sue you because you went to the library to look up information in the textbooks,” he said — using generative AI is similar. Try the free versions first; if you begin relying on it more, the paid versions have better features and are inexpensive. 

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

Instead of sitting behind a laptop during patient visits, the pediatrician directly faces the patient and parent, relying on an ambient artificial intelligence (AI) scribe to capture the conversation for the electronic health record (EHR). A geriatrician doing rounds at the senior living facility plugs each patient’s medications into an AI tool, checking for drug interactions. And a busy hospital radiology department runs all its emergency head CTs through an AI algorithm, triaging potential stroke patients to ensure they receive the highest priority. None of these physicians have been sued for malpractice for AI usage, but they wonder if they’re at risk.

In a recent Medscape report, AI Adoption in Healthcare, 224 physicians responded to the statement: “I want to do more with AI but I worry about malpractice risk if I move too fast.” Seventeen percent said that they strongly agreed while 23% said they agreed — a full 40% were concerned about using the technology for legal reasons.  

Malpractice and AI are on many minds in healthcare, especially in large health systems, Deepika Srivastava, chief operating officer at The Doctors Company, told this news organization. “AI is at the forefront of the conversation, and they’re [large health systems] raising questions. Larger systems want to protect themselves.” 

The good news is there’s currently no sign of legal action over the clinical use of AI. “We’re not seeing even a few AI-related suits just yet,” but the risk is growing, Srivastava said, “and that’s why we’re talking about it. The legal system will need to adapt to address the role of AI in healthcare.”

 

How Doctors Are Using AI

Healthcare is incorporating AI in multiple ways based on the type of tool and function needed. Narrow AI is popular in fields like radiology, comparing two large data sets to find differences between them. Narrow AI can help differentiate between normal and abnormal tissue, such as breast or lung tumors. Almost 900 AI health tools have Food and Drug Administration approval as of July 2024, discerning abnormalities and recognizing patterns better than many humans, said Robert Pearl, MD, author of ChatGPT, MD: How AI-Empowered Patients & Doctors Can Take Back Control of American Medicine and former CEO of The Permanente Medical Group.

Narrow AI can improve diagnostic speed and accuracy for other specialties, too, including dermatology and ophthalmology, Pearl said. “It’s less clear to me if it will be very beneficial in primary care, neurology, and psychiatry, areas of medicine that involve a lot of words.” In those specialties, some may use generative AI as a repository of resources. In clinical practice, ambient AI is also used to create health records based on patient/clinician conversations.

In clinical administration, AI is used for scheduling, billing, and submitting insurance claims. On the insurer side, denying claims based on AI algorithms has been at the heart of legal actions, making recent headlines. 

 

Malpractice Risks When Using AI

Accuracy and privacy should be at the top of the list for malpractice concerns with AI. With accuracy, liability could partially be determined by use type. If a diagnostic application makes the wrong diagnosis, “the company has legal accountability because it created and had to test it specific to the application that it’s being recommended for,” Pearl said. 

However, keeping a human in the loop is a smart move when using AI diagnostic tools. The physician should still choose the AI-suggested diagnosis or a different one. If it’s the wrong diagnosis, “it’s really hard to currently say where is the source of the error? Was it the physician? Was it the tool?” Srivastava added.

With an incorrect diagnosis by generative AI, liability is more apparent. “You’re taking that accountability,” Pearl said. Generative AI operates in a black box, predicting the correct answer based on information stored in a database. “Generative AI tries to draw a correlation between what it has seen and predicting the next output,” said Alex Shahrestani, managing partner of Promise Legal PLLC, a law firm in Austin, Texas. He serves on the State Bar of Texas’s Taskforce on AI and the Law and has participated in advisory groups related to AI policies with the National Institute of Standards and Technology. “A doctor should know to validate information given back to them by AI,” applying their own medical training and judgment.

Generative AI can provide ideas. Pearl shared a story about a surgeon who was unable to remove a breathing tube that was stuck in a patients’ throat at the end of a procedure. The surgeon checked ChatGPT in the operating room, finding a similar case. Adrenaline in the anesthetic restricted the blood vessels, causing the vocal cords to stick together. Following the AI information, the surgeon allowed more time for the anesthesia to diffuse. As it wore off, the vocal cords separated, easing the removal of the breathing tube. “That is the kind of expertise it can provide,” Pearl said.

Privacy is a common AI concern, but it may be more problematic than it should be. “Many think if you talk to an AI system, you’re surrendering personal information the model can learn from,” said Shahrestani. Platforms offer opt-outs. Even without opting out, the model won’t automatically ingest your interactions. That’s not a privacy feature, but a concern by the developer that the information may not help the model. 

“If you do use these opt-out mechanisms, and you have the requisite amount of confidentiality, you can use ChatGPT without too much concern about the patient information being released into the wild,” Shahrestani said. Or use systems with stricter requirements that keep all data on site.

 

Malpractice Insurance Policies and AI

Currently, malpractice policies do not specify AI coverage. “We don’t ask right now to list all the technology you’re using,” said Srivastava. Many EHR systems already incorporate AI. If a human provider is in the loop, already vetted and insured, “we should be okay when it comes to the risk of malpractice when doctors are using AI because it’s still the risk that we’re ensuring.”

Insurers are paying attention, though. “Traditional medical malpractice law does require re-evaluation because the rapid pace of AI development has outpaced the efforts to integrate it into the legal system,” Srivastava said.

Some, including Pearl, believe AI will actually lower the malpractice risk. Having more data points to consider can make doctors’ jobs faster, easier, and more accurate. “I believe the technology will decrease lawsuits, not increase them,” said Pearl.

 

Meanwhile, How Can Doctors Protect Themselves From an AI Malpractice Suit?

Know your tool: Providers should understand the tool they’re deploying, what it provides, how it was built and trained (including potential biases), how it was tested, and the guidelines for how to use it or not use it, said Srivastava. Evaluate each tool, use case, and risk separately. “Don’t just say it’s all AI.” 

With generative AI, users will have better success requesting information that has been available longer and is more widely accessed. “It’s more likely to come back correctly,” said Shahrestani. If the information sought is fairly new or not widespread, the tool may try to draw problematic conclusions. 

Document: “Document, document, document. Just making sure you have good documentation can really help you if litigation comes up and it’s related to the AI tools,” Srivastava said.

Try it out: “I recommend you use [generative AI] a lot so you understand its strengths and shortcomings,” said Shahrestani. “If you wait until things settle, you’ll be further behind.” 

Pretend you’re the patient and give the tool the information you’d give a doctor and see the results, said Pearl. It will provide you with an idea of what it can do. “No one would sue you because you went to the library to look up information in the textbooks,” he said — using generative AI is similar. Try the free versions first; if you begin relying on it more, the paid versions have better features and are inexpensive. 

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