In celebration of the 30^th anniversary of the film’s original release, Ron Howard’s Apollo 13 was just re-released in IMAX theaters across the country. The film tells the story of the doomed spaceflight, where astronauts Jim Lovell, Fred Haise and Jack Swigert found themselves stranded in outer space after an oxygen tank exploded in April, 1970.

Lovell, the captain of Apollo 13, passed away last month at the age of 97. News of his death broke amidst headlines that the Department of Health and Human Services plans to pull $500 million from funding of research into mRNA vaccines, the novel technology behind the COVID-19 vaccine.

So what does Apollo 13 have to do with mRNA vaccines?

Adrift somewhere between Earth and the moon with their oxygen supply leaking out into space, the Apollo 13 astronauts’ safe return home seemed improbable. But with an all-hands-on-deck response, the nation’s brightest minds at NASA worked tirelessly to develop a plan to defy the odds and bring the astronauts home safely. The story of Apollo 13 is a prime example of what science and engineering can accomplish in short time when lives are on the line.

The story of mRNA vaccine technology and COVID-19 is another example. With an all-hands-on-deck response in 2020, the federal government supported some of today’s brightest minds—in the U.S. and abroad—to develop and distribute novel, highly effective mRNA vaccines to a world in crisis. This is one particularly remarkable story among many other scientific endeavors supported by the federal government that allow Americans to live longer, healthier lives.

Cutting off funding for medical research and sidelining the thousands of dedicated scientists behind it would be akin to NASA laying off its engineers, scientists, and astronauts with Lovell, Haise, and Swigert drifting through space 200,000 miles away before devising a way to bring them home. But instead of leaving the lives of three astronauts hanging in the balance, decisions to gut federal support of biomedical research—into mRNA vaccines and many other areas—imperil millions of Americans.

Secretary of Health and Human Services, Robert F. Kennedy, Jr. stated the reason for pulling funding for mRNA vaccine research is that “the data show these vaccines fail to protect effectively against upper respiratory infections like COVID and flu.”

Both randomized controlled trials and real-world evidence from the COVID-19 pandemic have made it clear, however, that mRNA COVID vaccines were highly effective, saving many thousands of lives here in the U.S. and millions across the globe. Meanwhile, data from clinical trials and real-world COVID mRNA vaccine safety monitoring suggest that risk of serious complication is very low and generally outweighed by the risks of COVID infection. While there is room for more careful evaluation of benefits and risks in some populations, the administration’s concerns about potential risks are often unmoored from the very studies they cite.

There are three things the data have made clear about mRNA vaccine technology:

1. The potential benefits to human health are very large

2. The risks appear to be small compared to the benefits

3. Maximizing benefits and minimizing risks requires more research, not less.

This means that stopping—or even pausing—research into this promising technologies like mRNA vaccines carries its own costs and risks.

Subscribe now

There are signs that Congress understands what’s at stake here, with reports suggesting that a spending package will include funding specifically set aside for mRNA vaccines. To what degree funds will be specifically dedicated to mRNA vaccines—and to scientific endeavors more broadly—will be included in the final legislation remains to be seen.

It’s hard to estimate what the exact costs, in the form of illness and death, might be from stifling research into new technologies, since measuring something that doesn’t happen can be quite challenging.

But take the example of the GLP-1 drugs, which have revolutionized treatment for obesity over the past several years. Research into these drugs was showing promise as early as the late 1980s, yet lack of investment out of reported concern that the drug could only be injected delayed its widespread availability by decades. It’s impossible to know exactly how many patients could have benefitted, but it’s not hard to imagine how life for patients with obesity could have been had the GLP-1 drugs been available ten years sooner.

Fortunately for the stranded crew of Apollo 13, thousands of NASA scientists, engineers, astronauts and others were supported to work around the clock and bring the astronauts home. NASA failed its original mission to land on the moon, but it succeeded in bringing the astronauts home alive after the explosion—earning the Apollo 13 mission the moniker “the successful failure.”

We’ve noted that in some ways, the story of COVID-19 is a “successful failure” as well, since while millions tragically died during the pandemic, the government and scientific community were successful in working tirelessly with NASA-esque ingenuity to rapidly develop and deploy a highly successful vaccine. The rescue of the Apollo 13 astronauts in 1970 and Operation Warp Speed in 2020 show us what scientific endeavors can accomplish when supported to their full potential.

They’re also examples—among many others—of what we have to lose when we don’t take advantage of our nation’s massive scientific talent.

The Apollo 13 film, now back in theaters, tells the story of the nation’s brightest minds working together to save three men stranded in outer space. We don’t know for sure, but today’s brightest minds could be set to save many more lives through federally supported research—let’s not go radio silent on them.

Subscribe now

Share

BUY THE BOOK: Random Acts of Medicine

Let’s start with a simple question: If your doctor’s office called you to schedule an appointment in the morning or the afternoon, does it matter which appointment you pick?

In many scenarios, the answer is probably no—since most of the time, the benefits of picking an appointment that is convenient to you will likely outweigh any differences in care at different appointments.

But there are 3 ways in which the time of day you see your doctor could influence your health outcomes:

• Your body or the treatment you receive could actually function differently at different times of day, impacting outcomes

• Your doctor may provide different care at different times of day

• People who tend to see the doctor in the morning may be different than those who tend to see the doctor in the afternoon. Those differences might impact outcomes.

Does the human body work differently at different types of day?

In our new article out in TIME Ideas, we dive into this first scenario in light of a new study of patients with advanced lung cancer who were randomized to receive an immunotherapy treatment (called a checkpoint inhibitor) at different times of day—with striking results.

Do doctors provide different care at different times of day?

Prior studies have shown that yes, there can be differences in care provided to patients at different times of day. For example, studies of primary care clinics in a Pennsylvania health system showed that that patients seen later in the day were less likely to have their primary care doctor order recommended cancer screening tests, to receive a flu shot, or to be prescribed cholesterol-lowering medication. Meanwhile, surgeons are more likely to have patients experience complications following a non-emergent operation late, compared to early, in the day.

It could be that staffing and resource levels in the facility could differ in the afternoon, leading to differences in care. It might also be the case that doctors are more fatigued or rushing to finish on time at the end of the day, and in turn, make different medical decisions. This idea is reminiscent of a highly publicized study of judges in Israel, which found that judges became progressively less likely to grant parole to prisoners toward the end of a block of cases, and were subsequently more likely to rule favorably for prisoners immediately following a lunch break—more on this later.

But unlike decisions made in the primary care office or operating room, once a cancer patient is started on a treatment, care typically follows protocols with little room for variation. Indeed, in this new study of cancer patients discussed above, the cancer patients received highly protocolized care—suggesting that the care provided was likely similar regardless of the time of day it was given.

Subscribe now

Are the patients who come for morning appointments different than patients who come for afternoon appointments?

In many situations where care is similar regardless of whether patients come in the morning or afternoon, the important next question to ask is about the patients themselves. Are patients who choose morning appointment times similar to those who choose afternoon appointments? If they’re different, it’s possible that any differences seen between morning and afternoon appointments can be explained—at least in part—by factors besides the specific timing of treatments.

Patients who make appointments for treatment of the same condition are going to be different from one another. Some patients might have limited mobility, or the symptoms of their condition (e.g. pain or shortness of breath) may be of different severity. If patients they know they will need more time to come in, patients who might be struggling more—who we would also expect to generally not do as well compared to patients who aren’t struggling as much—may choose afternoon appointments if it’s challenging to come in the morning. It could also be that patients with fewer resources at home (for example, patients who don’t have a car, can’t spare money on a rideshare app, etc.) tend to come later in the day when more assistance might be available to them.

If this is the case, it’s hard to disentangle whether different appointment times are leading to different outcomes for patients, or if it’s because patients who were always going to do better or worse tend to choose appointments at certain times of day. Many studies try to account for these differences, but there are always many potential contributing factors that aren’t measured, like the level of resources at home that might dictate both appointment times and the patient’s overall survival times. Randomized controlled trials—like this new one randomizing cancer patients to appointments at different times of day—are one way to avoid bias from these other potential contributors.

The study of the Israeli judges offers an important lesson in this regard. Subsequent analyses of the judges’ cases found that the cases presented before a lunch break were different than the cases presented after the break, suggesting that other case factors besides the judges’ fatigue and hunger leading up to a break could have explained the findings.

Thank you for reading Random Acts of Medicine! Please share any thoughts you have in the comments section. We look forward to bringing more thought-provoking topics in health care right to your inbox!

Leave a comment

Subscribe now

Share

Buy the book: RANDOM ACTS OF MEDICINE

• In an article and video we put together with TIME for their “Ideas” series, we take the water fluoridation debate as an opportunity to discuss some of the larger issues surrounding how we incorporate science into public health policy.

  

• Out today in TIME Ideas is another article and video about a recent study of ours (with colleagues Vishal Patel and Michael Liu) which was inspired by a 2000 study of London taxi drivers which found that their hippocampus (the part of the brain focused on memory and navigation that is involved in Alzheimer’s dementia) were larger than the general population. In our study, we take a look at what happens to taxi drivers later in life—are they impacted differently by Alzheimer’s disease?

  

• NPR PlanetMoney reporter Greg Rosalsky reached out to us after saw a working paper of ours published with the National Bureau of Economic Research titled “Halloween, ADHD, and Subjectivity in Medical Diagnosis.” In this study, we find that children were more likely to be diagnosed with ADHD if they were seen on Halloween compared to other days. We explored how medical care might be different on holidays in general for the PlanetMoney newsletter.

• Bapu, along with a group of researchers at Harvard and Boston Children’s Hospital, published a study that showed the benefits of “just-in-time coaching” of doctors in training right before they attempted to place breathing tubes in infants undergoing surgery. Such coaching strategies have proven effective in areas other than medical care, too, as Bapu and co-authors wrote in the New York Times.

  

  We hope you all have a happy new year, and we look forward to more Random Acts of Medicine in 2025!

  Share

  Leave a comment

  Subscribe now

Also available in Chinese and Korean!

The 2024 Paris Olympics hosted approximately 40 sports, with gymnastics, track, and swimming all in close competition for “most watched,” attracting over 30 million U.S. viewers—an 82% increase from the 2020 Tokyo Summer Olympics. With today’s video streaming capabilities, viewers could pick and choose which sport to view from the 5,000 hours of coverage. This year, it was easier than ever to discover a new sport that Olympic enthusiasts could find so intriguing that they might be inspired to try it themselves.

Three Olympic sports this year involved guns: riflery, pistol shooting, and shotgun. Silver medalist shooters Yeji Kim from South Korea and Yusuf Dikeç from Turkey popularized their sport by going viral on social media, reaching an even larger audience beyond the their events’ reported viewership.

Kim gained attention for her swagger and eyewear, described by an NPR reporter as “cyborg-esque shooting glasses and casual-yet-confident stance,” which ultimately landed her an acting gig as an assassin.

Meanwhile in the men’s competition, it was the absence of fancy equipment combined with his laid back, hand-casually-in-pocket “nonchalant stance” that led Dikeç to capture the fascination of fellow athletes and fans around the world.

Even if you weren’t an avid Olympic viewer this summer, chances are that you’ve seen images of these two somewhere, as they quickly spread across traditional and social media. And while it would be hard to study this single Olympic season we wondered: if we looked at a bunch of summer Olympics, would there be evidence that the sudden attention on gun sports lead gun sales to shoot through the roof?

Is it even plausible that the Olympics could drive gun sales?

There are a number of factors that research suggests drive gun sales. Most nationwide research on gun sales relies not on actual sales data, but on data generated when a firearms dealer conducts a background check on a prospective buyer. The FBI’s National Instant Criminal Background Check System, or NICS, releases the number of background checks initiated for various types of firearm sales from licensed firearms dealers. Notably, some private sales do not require background checks depending on local laws, but NICS background checks generally serve as a pretty good proxy for actual gun sales.

Gun sales tend to be seasonal, picking up in the fall and winter around hunting season and the holidays. But as far as other events that drive sales, prior research has demonstrated increases in NICS background checks to buy guns following elections, mass shootings, protests, and passage of firearm regulations. So while there is evidence that current events can influence gun sales, these events have generally been ones that spur anxiety about personal safety and/or future restrictions on gun sales.

Our question was a bit different: could an event that might generate new interest in gun sports translate into a measurable increase in gun sales? If the Olympics are randomly timed with respect to interest in firearms, we have a natural experiment on our hands. We can compare gun sales in the months following the summer Olympics to those same times in years without summer Olympics. If there’s a difference in gun sales and the Olympic events are randomly timed with respect to interest in purchasing firearms, we could attribute that difference to the Olympics (these would be assumptions underpinning the analysis—more on this later).

Subscribe now

Gun Sales in Summer Olympic vs. Non-Summer Olympic Years

Below is a plot of the FBI’s NICS background check data, showing the number of monthly background checks run across the country from 1999 to 2024; years with summer Olympic Games are in color, with other years in gray (including 2024, with its incomplete data).

The summer Olympic years don’t particularly stand out from the other years, and they follow the same seasonal trends. Looking at the 2016 summer games—the most recent games before this one that was not impacted by a global pandemic—and the two surrounding years, we see that gun sales followed a similar pattern.

So it didn’t appear that there were substantial increases in gun sales after the summer Olympics. While there’s still a decent chance some small number of people who watched shooting sports were inspired to take up a new sport (or simply purchase a new firearm), there’s several possible explanations why we wouldn’t see any measurable effect. First, while there may be a handful of people inspired to take up sport shooting and buy a new firearm because of the Olympics, their numbers may be so small relative to typical levels of firearm purchases as to not be detectable. Second, those inspired to start sport shooting may already own a firearm and not need to purchase a new one. Third, the pistol shooting event (that Kim and Dikeç competed in) uses an air pistol, which doesn’t require a background check to purchase, so if there was an increase in purchase of these guns, we wouldn’t know about it using FBI data.

Share

What if we had seen a difference in gun sales after Summer Olympics?

Let’s pretend for a moment that we had actually seen the summer Olympic years have higher levels of firearm sales following the games, toward the end of the summer and fall months. In order to attribute those sales to interest from the Olympics, we’d have to assume that there was nothing else happening in those same years that might drive gun sales. If that was the case, then we would have trouble determining whether that increase was due to the Olympics or that other factor.

The summer Olympics have taken place every four years, going back from 2024, to 2020, 2016, 2012, 2008, etc. If those years ring a bell, it’s probably because they’re also the U.S. presidential election years. And as we mentioned earlier, presidential elections have been shown to drive gun sales. Because presidential election years are exactly correlated with summer Olympic years, if we had seen a difference in gun sales these years, we wouldn’t know whether it was because of the election, or because of the Olympics—or some combination of both.

In fact, even though in reality we didn’t see any appreciable difference between the Olympic and other years, it’s possible that there is both an Olympic and an election effect on gun sales that cancel each other out—one that increases and one that decreases sales. This seems pretty unlikely, but it’s possible. More likely, we’d guess the most likely scenario is that neither the average summer Olympic Games nor the average presidential election (at least, in the years we looked at) don’t markedly drive gun sales.

Subscribe now

Leave a comment

BUY THE BOOK: Random Acts of Medicine

The article covered a very clever study by University of Chicago environmental economist Eyal Frank, published in the journal Science. The study—which we’re going to dive into more deeply—illuminates one of the ways our environment and policies surrounding it can impact our health.

Can bats really affect our health?

The only thing we learned about bats in medical school was that they can carry rabies, so if a person is exposed to a bat that has, say, snuck into their home, they might need to receive post-exposure treatments against rabies (a very burdensome ordeal; our condolences to anyone who has gone through this).

But as kids learn in science class, bats play an important role in our ecosystem. This includes ecosystems that support agriculture in the U.S., as Frank describes in the paper:

Existing research in ecology documents that bats provide biological pest control through their high population size and predation rates on a variety of insects, many of which are crop pests. Insectivorous bats consume 40% and above of their body weight in insects each night.

Damages from crop pests can substantially reduce agricultural productivity. In the US, about 13% of crops are estimated to be destroyed by insects each year, which represents a loss of $27.6 billion a year.

So if something were to happen to insect-eating bat populations, it could result in increases in crop pest insect populations, which in turn could result in loss of crops. Without bats to control the pests, farmers would have to use more insecticides to prevent the crop loss. And if those pesticides have negative effects on human health, then we have a connection between the bat population and the health of the human population.

Subscribe now

Starting in about 2006, something did happen to bat populations: a disease called “white nose syndrome,” or WNS, began to spread across insect-eating bat populations in the eastern half of the U.S.

According to the U.S. Fish and Wildlife Service:

Research indicates the fungus that causes WNS, Pseudogymnoascus destructans, is likely exotic, introduced from Europe. What started in New York in 2006 has spread to more than half of the United States and five Canadian provinces by August 2016, leaving millions of dead bats in its path. WNS causes high death rates and fast population declines in the species affected by it…

Researchers call the disease “white-nose syndrome” (WNS) because of the visible white fungal growth on infected bats’ muzzles and wings. This cold-loving fungus infects bats during hibernation, when the bats reduce their metabolic rate and lower their body temperature to save energy over winter. Hibernating bats affected by WNS wake up to warm temperatures more frequently, which results in using up fat reserves and then starvation before spring arrives.

Assuming that the timing of the emergence of WNS in a local bat population is as good as random with respect to human health—a quite reasonable assumption—then we have a natural experiment on our hands. Changes in human health occurring shortly after WNS infects a bat population can reasonably be attributed to WNS.

But since the fungus that causes WNS has never been shown to directly infect or harm humans, it leaves the other changes brought on by WNS—namely, increased use of toxic insecticides because there are fewer bats around to prevent crop loss—as the best explanation for human health problems, as Frank illustrates:

Question 1: Was there more insecticide use when WNS reduced local bat populations?

WNS first emerged in the state of New York in 2006, and spread southward and westward over the subsequent decade. Of course, we would only expect to see increases in insecticide use after WNS started affecting insect-eating bat populations in an area. Using data from the U.S. Fish & Wildlife Service, Frank measured the amount of insecticide used per square kilometer of cropland in the years before and after WNS was detected in a given U.S. county. Combining all counties into a single analysis centered around the time WNS was first detected locally, here’s what he found:

In the year WNS was first detected in a county, insecticide use increased, and continued to increase over the subsequent 6 years. On average, this represented an increase of 2.7 kg of insecticide per square kilometer of cropland. His results were similar in additional analyses where he accounted for things like weather and larger trends in pesticide use. There was also no similar trend in fungicide or herbicide use associated with WNS entering an area, further confirming the notion that insecticides were being used to address the loss of natural pest control from the bats. Indeed, it appeared that as WNS wipes out bat populations in a county, farmers compensate with insecticides to prevent crop loss.

Share

Question 2: What happened to human health after WNS emerged in an area?

If insecticide use increased as a result of WNS, and if insecticides can cause human health problems, then we should expect to see an increase in those health problems. But are insecticides used to prevent crop loss actually toxic to humans?

There are a number of different insecticide chemicals that are used for agricultural purposes that can pose serious risks to the central nervous system (among other organ systems when humans are exposed to sufficient quantities—after all, the purpose of these chemicals is to kill living organisms). Federal regulations on pesticide use for crops focus on ensuring that the amount of chemicals people are exposed to is sufficiently small as to not harm humans. That amount will be smallest for infants, our smallest humans.

Considering there were not widespread reports of insecticide toxicity among humans, it made sense to look at humans where a small increase in chemical exposure could make a difference: among the infants. So using the same process as before, but this time looking at infant mortality (specifically, “internal” infant mortality that excluded external causes like accidents and homicide), Frank looked at infant mortality rates in counties before and after WNS was first detected, again combining them into a single analysis:

Internal infant mortality rates rose in the years following WNS first detection in a county—an average of about 0.5 deaths per 1,000 live births. Notably, there was no increase in external causes of infant mortality (which we would expect since insecticide use shouldn’t impact these types of deaths), nor were there changes in outcomes like infant birth weight (evidence against less healthy babies being born during this period).

Taken together with the insecticide findings, Frank was able to estimate the impact of insecticide use on infant mortality: every 1% increase in insecticide use resulted in an estimated 0.25% increase in internal infant mortality rates.

Subscribe now

Could something else have explained the increases in infant mortality?

Frank’s analyses make it clear that the introduction of WNS to an area resulted in both increases in insecticide use and increases in infant mortality rates. To attribute the increase in infant mortality to the insecticides, we have to assume that WNS cannot lead to infant deaths in other ways.

For example, if there were weather changes that led to both the introduction of WNS to an area and led to increased infant mortality, we couldn’t blame all the infant deaths on the insecticides. But in additional analyses, Frank found that the findings held even when adjusting for weather changes.

Alternatively, if decreasing bat populations led to increases in the mosquito population (perhaps outside of the areas where insecticide was sprayed), which in turn led to more mosquito-born illnesses, we couldn’t blame the insecticides. But no such rise in mosquito-born illness in infants has been reported.

Another possibility might be related to crop destruction itself. If crop destruction affects the supply, and therefore the prices of crops, this might ultimately find its way to families in the form of higher food prices. One would expect this effect to not be localized, however, since the markets for crops are not localized. Relatedly, local economic impacts arising from crop destruction may have impacts on infant mortality, but those economic impacts would need to be large.

So in the absence of any other plausible explanation for increased infant mortality that would have coincided with the emergence of white nose syndrome—and there doesn’t seem to be one—insecticide use remains as the most likely explanation.

What do we do with this information?

One of the reasons this study is remarkable is that it answered two completely different research questions: one about medicine, and one about the environment.

The natural experiment created by WNS provided an opportunity to understand the impact of agricultural insecticides on infants, a very important issue that would otherwise be very difficult to study. This information can inform further development of safe practices in their agricultural use.

But the study also provided solid evidence about our ecosystem—an invasive fungus can have far reaching effects on bat populations, insect populations, our food supply, and infant health. Of course, this is only one example of a single issue on a single health outcome. Surely there are many other connections like these, and taking advantage of natural experiments is a powerful way to study them.

We’ll leave you with some of Eyal Frank’s conclusions from his study: “White-nose syndrome is but one of many threats that bats face, including those that are shared with multiple other species, such as habitat loss and climate change. Improving our understanding of how changes in biodiversity affect human well-being will be important when designing and implementing conservation policies.”

Leave a comment

Share

Subscribe now

Buy the book: RANDOM ACTS OF MEDICINE

Progress is slow, of course, and waiting for biases to go away simply isn’t an option for those who are currently impacted by them. When she’s not doing research, Fragale works with organizations and the professionals within them, focusing on evidence-based ways to avoid some of the harms brought on by gender-based and other biases. This is why she set out to give women—or really anyone whose talents, expertise, or accomplishments go overlooked and undervalued—a guide to navigating workplace bias in her new book Likeable Badass: How Women Get the Success They Deserve.

When I asked if she’d be up for an interview, I warned her that I was particularly interested in picking her brain about women in medicine. We only scratched the surface of some of the issues women physicians face in the Random Acts of Medicine book (now in paperback!) in the chapter titled “What Makes a Good Doctor?”, so I was excited to get to explore the topic with an expert who could also give some practical advice. Our conversation has been edited for brevity and clarity.

Chris Worsham: At the beginning of Likeable Badass, you quickly acknowledge that when it comes to the role of various biases on status and influence within organizations, the current state of affairs is not great, and it should change. But you go on to say that it’s not going to change as fast as anyone would like, so I’m going to give you the current lay of the land, the science behind it, and some actionable steps you can take to deal with it.

Alison Fragale: Yes, and I think a medical analogy here—although it’s not in the book—is pretty apt. Patients inherit or end up with all kinds of problems that are not their fault and that are difficult to manage. We have people in medicine that say that someday, that problem shouldn’t be a problem—e.g. we should end deaths from cancer or heart disease—and they’re working on solving that. But it doesn’t mean that if you’re the person with cancer or heart disease that you’re just going to throw your hands up in the air and wait for someone else to find a cure.

You’ll say, well, what are the things that are within my control right now to take this thing that I sure as hell don’t deserve and have the best possible outcome I can? That’s how I think about any kind of systemic exclusion or bias and helping someone find what is in their control to be seen the way they deserve to be seen.

CW: That sounds like a pretty good analogy to me. The other thing I discovered as I got into the book—and you make this quite clear—is that a lot of what you talk about is not just applicable to women.

AF: Correct, the science is applicable to everybody. I wrote the book for women because it’s an audience I care about, and if I didn’t write this book, it wasn’t going to get written—and I can write for whoever I want! And I didn’t do this in a devious way, but I just thought wouldn’t it be cool if there were some secrets of the universe and we only told women? Wouldn’t that be awesome?

But no, it’s not just applicable to women. Status is a fundamental human need, something that all human beings care about. In any given situation—regardless of one’s gender, race, sexual identity, any of those characteristics—you go into some rooms and you’ll be the high status person, and you’ll go into the next room and you’ll be the low status person.

As a physician, for example, you could walk into a patient’s room and feel like you’re the high status person in that relationship. But then you go to your conference, and maybe you haven’t been practicing that long, and you’re talking to people who are world renowned people in your field, and you feel like a low status person.

So status is something all human beings contend with, and all the same rules ultimately apply. So no, there is nothing in there that is uniquely women specific that only women have to think about or worry about.

Subscribe now

CW: You talk about status, and in the book you delineate the difference between “status” and “power.” What are those differences, broadly speaking?

AF: Status is the extent to which you are respected and highly regarded by other people. Power is the extent to which you control resources that people find valuable. Information is a resource, money is a resource. Car keys for your teenager is a resource. Any time you control a resource that someone else wants, you have power.

Status and power are both sources of influence. Someone will be more likely to be swayed by you and do what you want them to do if they respect your status. But they’ll also do it if you control a resource they value, even if they don’t respect you—like a teenager saying “well I gotta do what mom says, otherwise I don’t get the car.” So they are similar in that regard.

They differ in their consensual nature. Someone’s status is real and can be very far reaching, but status exists only in someone else’s mind. You only get as much status as other people think you should get. It cannot be stolen or taken, only given.

Power is also generally given, but it can also be taken or held by people even if other people don’t necessarily want someone to have it. So the one really important difference between status and power is that everyone gets a say in conferring status to someone, whereas with power, once you have the resources, they may not be able to be take it away from you.

Status has to be continually earned, and I think that’s why it feels amorphous for people and daunting to think about. What I try to do in the book is make it feel more formulaic and less daunting by breaking down the science of how it happens.

CW: What’s a good example of this?

AF: You’re a doctor, right? You’re the only one who can run and interpret a test, the only one who can prescribe medications. You have a lot of resources that patients want. You might also be very respected by your patient. There’s a certain kind of power and status relationship right there.

Where it can start to come apart is when the doctor becomes a patient. When you are a physician treating another physician, they can’t treat themselves but know just enough to be able to put your status into question. They might not honor your usual status because they think “I could be doing this.” You can think of how that situation might be really hard to navigate.

CW: Indeed.

AF: A physician-nurse relationship is another example. The physician is highly dependent on the nurses, so they might have a lot of power over the physician. But then there’s a status differential, since most of the world sees a doctor as a higher status than a nurse. That can put nurses in a situation of being in what we call “low status power holders.” They have a lot of control, but they’re not as highly regarded; this has been a topic of my research. People in these positions do face a lot of incivility—low level mistreatment like having their judgment questioned, cutting them out of information flow, or devaluing their work. And it can even lead to high intensity things like harassment.

And what happens? At least in non-medical contexts, we see very highly qualified people leaving organizations. In fact, senior women leave at higher rates than junior women in organizations, because when you have power and you don’t have status, people mistreat you. And when people mistreat you have options, you’re like “I’m outta here.”

So thinking about honoring people’s status is one of the most controllable, important things we can do to create a workplace that people want to work in. We may be unaware of how our judgments of their power and status are causing us to devalue them, but they are very aware.

Share

CW: You’re reminding me of a study by University of British Columbia economist Heather Sarsons that looked at differences in how male and female surgeons were treated by other doctors who referred their patients to them. In the study, she looked at what happened after a surgeon had a bad outcome—a death following a procedure. If a male surgeon had a bad outcome, it didn’t really change referral rates to that surgeon. But if a female surgeon had a bad outcome, referral rates to her subsequently dropped [see image below].

Sarsons’ study suggested that there was an attitude of “it’s not your fault, bad outcomes sometimes happen” with the male surgeons, but for female surgeons the attitude was more likely to be “this is your fault, and I’m not going to send more patients to you.” I imagine this is subconscious in many cases, but is this a difference between “power”—since the male and female surgeons have control over the same resources—and “status,” since male surgeons were afforded the benefit of the doubt while female surgeons weren’t?

AF: That’s right. Because with status comes assumptions of competence and benevolence. So if you have a person that is assumed to be competent and benevolent—let’s say because of their gender—then confirmation bias kicks in. It’s pretty easy to ride that bias. If I think you’re competent, then even competent people can make mistakes, right?

Without status, we tend to assume either competence or benevolence, but not both. Lacking either one is bad when a mistake is made: either you’re not smart enough, or you don’t care enough.

Now, that happens on average, because gender is a status characteristic on average. But if one were look at the distribution of those male surgeons and female surgeons, there will probably be a tail of female surgeons that are getting referred to just as often as the men. So yes, on average the women aren’t getting referred to as often after a complication, but let’s look at the women who that didn’t happen to. I would bet money that the ones who didn’t lose referrals after a complication are seen as highly respected by referring physicians, for some reason or another, because of how they have shown up over the course of their career. So what I would say to any given woman—surgeon or not—is take steps to put yourself in that tail.

Gender is not anyone’s fate. Status is determined by many, many variables, most of which are within our control. My guess is that the surgeons who didn’t lose referrals were good—perhaps in ways that they couldn’t articulate—at navigating their status, such that by the time that bad outcome happened, they were just as respected as the male surgeons and didn’t suffer the same fate as their female colleagues.

CW: That brings us back to this notion of a “likeable badass” as someone who is competent—a badass—and has high status to go along with it. Now, I’m an internist and so I don’t do major procedures, but I do get to suggest to my patients which surgeons they should go to. And I think many of us find a handful of surgeons that would aptly be described as “likeable badasses;” they’re very good at what they do, but they are also easy to work with professionally and get along with their patients. So the question, it sounds like, is how do we get talented women—or anyone who is struggling with this—to be seen as a likeable badass?

AF: If you have a choice, you’re always going to want to refer to someone who is both likeable and a badass. Anyone who is low status is probably going to be stereotyped as only one or the other. Either you’re really caring and have great bedside manner but I don’t think you’re that good at what you do, or you’re really great at what you do but I think you’re a total bitch.

So what I talk about a lot in the book is acting deliberately and quickly at the outset of a relationship while people aren’t thinking much about you. Whether it’s gender, race, ethnicity, or you speak with an accent, biases are stacked against you. Take people’s minds off that bias and turn it to the way you’re showing up, whether that’s how you’re talking about yourself or how others are saying great things about you.

Here’s an example. Imagine a woman you’ve never met, but someone you think highly of says about her “Hey, you know she is really someone you should be sending your patients to.” Now, all of a sudden when that woman who you’ve never met does show up in your world, she already has status that has been built by a third party. Before you’d even started to think about her, something else had happened to make you think about this person as a highly competent and very caring person.

The advice I give then is that if you can get other people to do that for you, what happens is that you don’t even have to think about how you’re perceived by others, and you can just put your head down and start doing the work. Otherwise, you’re stuck trying to dig yourself out of a reputation problem—even though it was one you don’t deserve—from existing biases.

It’s kind of like the concept of preventing disease being easier than curing disease. If everyone in my family has had a heart attack in their 40s, am I going to wait until I have my heart attack and then say now that I’ve had my heart attack, I’m going to do something about it? Or would I say that statistically, I’m the kind of person that could have a heart attack, so I’m going to exercise more in my 20s?

The same logic applies here. The world has told us that women are at risk of problems related to gender bias, but we haven’t really been coached to do anything about it. So we go about our normal life, and a bad thing happens—a metaphorical heart attack. And we say “this sucks, now I have to try to heal myself.” And you can heal yourself, but it’s a lot easier than never getting sick in the first place. So what I talk about in the book is how to start building your status long before you ever need it, so you don’t end up in the situation where you have to change someone’s mind. Changing someone’s mind is possible, but it takes a tremendous amount of work.

CW: So what you’re saying is that when someone is developing a first impression, they’re going to succumb to various sorts of confirmation bias based on what they’ve heard before, so you should want to do everything you can to make that bias confirm good, positive things.

AF: 100%. One little bit of information can change the way you perceive someone else in just a second. I don’t know if this has ever happened to you where you’ve interacted with someone and you had no idea who they were, but then they say something and you realize that this person is a “big deal.”

For example, I do a lot of work with the military, and there are retired generals out there who are now civilians, wearing civilian clothes, and there are enough of them that no active duty soldier would know who they all are. And they might introduce themselves to someone who is active duty who asks “What do you do?” and they’ll respond “Oh, I’m a retired Lieutenant General,” and all of a sudden you can see their whole demeanor change instantly.

Of course, not everybody has the luxury of just dropping their high status occupation, because they may not have one. But there are things that I cover in the book that you can do to be able to say “I’m a person of value.” Let’s take the simple example of making introductions. Introducing people to each other is a competent and caring move—you’re competent because you’re connected, and you’re caring because you’re using your network to help someone.

Even if you’re in a lowly position in a hierarchy—like a med student—you can show people you have value. “Here’s an article I read that I think everyone could benefit from, so I’m emailing it out.”

Or when I was in grad school, there was the tediousness of reformatting research citations in various formats. It wasn’t hard work, but someone had to do it. So one of the things I did when I was a graduate student was just memorize the various citation formatting rules—should you write your citation this way or that way. It was nothing that someone else couldn’t have done, but nobody was going to take the time. All of a sudden I was a grad student that people needed to talk to. “Go talk to Alison, she knows the answer.”

So you can show up as the expert in something and you can solve someone’s problem. You don’t need to be a retired 3-star general for someone else to say “that’s valuable” or “I’m interested” about what you have to say.

CW: Certainly as a medical resident, one of the best ways to build a good reputation is by knowing how to get X, Y, or Z done around the hospital. And then you build that reputation of “oh yeah, he knows how to get that patient an x-ray sooner” or whatever.

AF: Those are things that you don’t need to have a fancy title to do, and sometimes not having the title can make it a lot easier. People feel sorry for you, they want to do you a favor; they don’t feel like they have to say yes to you so they’re more willing. So the whole idea is figure out ways to use your skills for someone else’s benefit in a way that doesn’t cost you a lot of time, and you’ve figured out a way to gain status. Then you can start to think bout building your status more broadly.

CW: One last question. We talked earlier about doctors and patients. Patients often come in as “low status” and “low power” within the healthcare system, and some patients have even lower status because of existing biases. You don’t have to look too hard to find evidence of bias where doctors are not believing things their female patients are telling them, not taking complaints from patients of certain races as seriously, and so on. In this very fragmented and dysfunctional health care system we have, I think patients are probably wanting both status and power. They want to feel like somebody cares about them and their complaints, and they want to have agency over what happens to them. Off the top of your head, are their any pieces of advice you might have for a patient who has an issue but doesn’t feel like they’re getting what they need out of the health care system?

AF: One simple one that comes to mind is to use an agent. I’m sure you’ve had this situation where a very elderly patient comes in and having trouble communicating, and you’re wondering what’s going on with them. Then all of a sudden, their child comes in who is very on top of things and is able to say “here’s what’s been going on, here’s what’s different than usual, I’m going to give you all of their records” and so on.

That child is their agent, and the patient benefits from their agent’s status as someone who is a trusted source of information. So if you’re not getting the outcomes that you want, ask who might be better: my neighbor, my husband, my child, or somebody else who might be able to help you get your needs met.

Sometimes it’s also helpful to bring in an agent when the patient is someone who is generally not very easy to deal with. If you’re the physician, you want to be able to take the person’s complaint seriously, but at the same time, physicians are human beings. When people are a pain in the ass, it doesn’t really motivate you to want to lean into helping that patient. But when you have those patients you just love and adore, you’re going to want to put in the extra effort to get the test done faster or relief sooner because you like them. So agents can be really helpful in helping with both the competence and the benevolence pieces of the relationship.

CW: In my experience, that agent can really be anybody who is essentially functioning as a third party who is listening to the conversation, watching out for discrepancies where the patient and the doctor might be talking past one another.

AF: One of the things we know about research on negotiation is that women don’t get as good of outcomes when they’re advocating for a benefit for themselves compared to when they’re advocating for someone else; for men the difference isn’t as big.

When you’re a patient, you’re advocating for yourself. You might still feel sick in the hospital and say to the doctor who wants to send you home, “I don’t want to be discharged today.” But it’s a lot easier to make that same request for someone else. Because of the status dynamics, it can be hard to ask for things for yourself. Other people can do it in a much more forceful way, because it’s not their life. When they’re saying “no” to going home, they’re not being selfish because they’re doing it for someone else.

CW: At many hospitals, there is a patient advocacy team, where someone can accompany a patient who could benefit from an agent when they’re talking with doctors. It sounds like we should have a very low threshold to suggest them when it seems like the patient and the medical team have a dynamic that isn’t working. What are some other ways physicians can give status to our patients?

AF: Simple things. People want to know their words matter, that they’ve been heard—so anything that says “I hear you.”

I remember one time my daughter’s back was hurting after she went jumping on a trampoline, and it wasn’t getting better quickly, so I asked if she thought it was bad enough to have to go to the doctor. And she goes “I don’t want to go to the doctor and have them tell me nothing’s wrong with me and feel dumb.” Human beings feel that way all the time—they want to be seen as competent.

So one way to honor people’s status is to say things like “trusting yourself and coming in was a smart thing to do.” Ask yourself: What would make the patient feel respected in this situation? What is this person probably worried about, and what could I say to them so that they feel like they’re a good person—a smart person—who is doing the right thing?

CW: Well thank you so much for chatting with me. I really appreciate your taking the time to focus on these issues in the medical field!

AF: For sure!

“Likeable Badass: How Women Get the Success They Deserve” by Alison Fragale comes out on September 3rd and is available now for pre-order. Also, for more from Dr. Fragale, check out her newsletter, “The Upper Hand.”

Share

Leave a comment

Subscribe now

These efforts are part of a larger push by the International Olympic Committee to focus on minimizing environmental impacts and promoting long term environmental sustainability for the host cities. In addition to the future-minded Olympic Village, Paris took the games as an opportunity to make a variety of long term improvements to the city and its communities. This included efforts to clean up the Seine river, in which swimming has been banned for over a century due to pollution, to the point that bacterial levels were low enough to host the swimming portion of the Olympic triathlon (though some triathaloners fell ill, it’s not believed to be due to bacteria in the Seine).

These, along with a variety of other environmental, social, and economic programs that were part of the Paris 2024 Olympics may indeed improve health for the local population; only time will tell for Paris, but similar programs have had mixed results for prior host cities. But a growing body of research on prior Olympics—natural experiments that take advantage of the rapid changes made in preparation for the games—can tell us what some of those health benefits might be.

What were impacts on air pollution from prior Olympics?

We’re going to focus on air pollution and the measurable impacts it can have on human health, which has been shown to be wide-ranging. There are the more obvious impacts of poor air quality like asthma and other breathing problems, and less obvious ones like heart attacks, strokes, low birth weight, and death from all causes. A number of clever researchers over the past several decades took advantage of the natural experiments brought on by sudden major changes that impact entire populations all at the same time—what we might call an as-good-as-randomly-timed “shock to the system”—because of the Olympics.

Beijing 2008

Air pollution has long been an issue in Beijing, but organizers of the 2008 Beijing Olympics took a number of measures to improve air quality (many of them temporary, like pausing high-emissions industrial operations or limiting who was allowed to drive a car) during the games in August and September of 2008. These efforts worked, with various air pollutant levels dropping by 20-60% in the weeks leading up to the Olympics.

In a study of pregnant mothers that took advantage of the random timing of the Olympics relative to when mothers would be giving birth, researchers in China and the U.S. analyzed birth weights of 83,672 full term births that occurred in Beijing around the time of the 2008 Olympics (birth weight is a common public health metric, since low birth weight is associated with short- and long-term health problems, and air pollution has previously been shown to reduce birth weight). The study found that babies who spent the 8th month of their gestational period—when significant growth occurs—during the 2008 Beijing Olympics had a higher birthweight (by 23 grams) than babies born on those same calendar dates in 2007 or 2009 when pollutant levels were higher.

Another study of about 200 randomly selected Beijing residents measured lung function before, during, and after the Beijing Olympics, finding that airflow on exhalation (which can be impacted by air pollutants) increased in 78% of subjects during the Olympics compared to before the Olympics, and then it worsened in 80% of them after the Olympics were over and pollution control methods had stopped.

Air quality in Beijing has improved significantly since the 2008 Olympics, which seemed to have served as a catalyst for ongoing projects to address air pollution that have continued—such as a large urban forestry project we visited in an earlier post.

Subscribe now

Atlanta 1996

Organizers of the 1996 summer Olympics in Atlanta, Georgia were also worried about air quality, which had historically gotten worse in the summer months in Atlanta—suggesting air quality could be bad that summer even before the estimated 2 million visitors would arrive for the games. In preparation, Atlanta restricted automobile and delivery vehicle travel, encouraged changes in business hours and greater telecommuting (fire up the dial-up modem!), improved existing train systems and public transit, and created new temporary bus lines to both accommodate the crowds and reduce emissions of air pollutants.

In a study published in 2001, a group of university- and CDC-based researchers in Atlanta measured changes in air quality, vehicular traffic, and rates pediatric asthma exacerbations in Atlanta emergency departments in Atlanta in the time before, during, and after the 1996 summer Olympics. They noted significant reduction in the air pollutant ozone—which has been closely tied to increases in asthma exacerbations in children—as well as particulate matter and carbon monoxide (see figure below). They also saw the expected reduction in vehicular traffic. Indeed, the interventions to reduce air pollution levels worked (as they would 12 years later in Beijing).

What happened to the kids with asthma? Depending on which database they used, they saw a reduction in visits to Atlanta emergency departments for asthma ranging from 11% to 44%—with the 44% reduction being among the children covered by Georgia Medicaid (who would be expected to be more susceptible to air pollution levels compared to children covered by commercial insurance policies).

Share

What about Paris?

Compared to Beijing in 2008, the city of Paris was starting with generally better air quality before the Olympics began, and predictions were that air quality would be acceptable based on historical trends for this time of year. Paris already had robust public transit options and is accustomed to receiving millions of tourists every year, but of course, the Olympics is different. As in past Olympics, vehicular travel was restricted, and additional free bus lines were established to get spectators to and from athletic venues.

In the air quality diagram below, with each rectangle representing the daily air quality index measured in the heart of Paris for the year 2024, we can see that after some rough days in January, air quality has been average (“moyenne”) for most of this year, though there happened to be some days with inferior air quality in the weeks leading up to the July 26th opening ceremonies (Jui = Juillet, or July). During the games which ended on August 11th (Aou = Aout, or August), there were some days with poorer air quality.

Now, we won’t read into this too much since some variation in air quality is expected, particularly in the hottest months of the year. Without seeing what happens over the next few weeks, it’s hard to know how this stacks up with prior years. It may be the case that despite some days with poorer air quality, this is was actually better than we otherwise might have expected this time of year.

As far as health outcomes for Parisians, we’ll wait to see if any French researchers take advantage of this natural experiment as more data roll in over the coming years. Beyond air quality, some of the programs focused on urban improvement and promotion of sports among the population may reap benefits—again, only time will tell.

In the meantime, we now look to the future as we prepare to host the 2028 Olympics here in the U.S. in Los Angeles—a city famous for, among other things, its air pollution.

Share

Leave a comment

Subscribe now

Of course, the protection that vaccines offer to individuals relies on them actually receiving the vaccine, and the protection vaccines offer to the larger population relies on a sufficient proportion of the population getting vaccinated.

Sometimes, we might get a recommended vaccine when we’re already at the doctor’s office for another reason - a fortuitous opportunity. But more often, when we get vaccinated, it’s not simply a matter of chance. Typically, we know or are told that we’re due for a vaccine, we remember to get one, and then we take the steps to do so, usually involving a time-consuming and sometimes costly trip to a doctor’s office or pharmacy. Each one of these - lack of knowledge, failure to keep track, time and cost - can create a barrier to vaccination. What can we do to reduce these barriers?

In a new “megastudy” of over 3.6 million Americans published in the journal Nature, researchers based at the University of Pennsylvania (including Katy Milkman and Angela Duckworth) and other institutions across the country collaborated with CVS Pharmacy to study ways to increase uptake of the updated COVID-19 booster in late 2022—when uptake of the recommended booster vaccine was quite low at only 11%.

One intervention they thought might help increase uptake? A free ride to and from the pharmacy.

The Megastudy

Using records from CVS, the researchers identified adults who had previously received the initial COVID-19 vaccination series but who had not yet received the bivalent booster recommended by the CDC. They ultimately randomized 3.6 million patients in 65 different U.S. metropolitan areas into one of 9 different intervention groups, with a plan to measure what percentage of them got vaccinated at a CVS in the next 30 days.

The groups comprised:

• 7 text message-based reminder groups

• 1 group that was offered text reminders and free Lyft rides to and from a local CVS location

• 1 control group without any intervention.

The authors write:

We worked with CVS Pharmacy and a team of nine behavioural science experts to develop eight different intervention messages that were sent to patients by text message in early November 2022 to encourage adoption of the bivalent COVID-19 booster. All eight interventions consisted of an initial set of reminder texts with a follow-up set of reminder texts sent 7 days later, and all the tested text message reminders conveyed to patients that a vaccine was ‘recommended’ and ‘waiting for you’, language that was built on past research. The intervention of focal interest was designed to test the value of free round-trip rides to vaccination sites and it included the aforementioned standard reminder language but also provided people with one free round-trip ride by Lyft (a popular ride-sharing app) to and from a CVS Pharmacy in the month ahead.

Our seven other interventions did not offer free round-trip Lyft rides to a CVS Pharmacy. These interventions instead layered a range of different strategies for encouraging immunization on top of the standard reminder, from conveying current (high) rates of infection in a patient’s county to providing resources to combat misinformation.

Here’s an example series of text messages sent to about 492,000 patients randomized to the “CDC recommended” text-message intervention (all of the text-message interventions can be found here). On day 1 of the study, which was November 3, 2022, they sent this text message:

CVS Pharmacy: Hi [Patient First Name]! The CDC recommends updated COVID boosters to help prevent infection & severe illness. Your booster is waiting for you at CVS. Schedule: [link]

A week later, they sent the following text message:

CVS Pharmacy: Remember, a COVID booster is recommended by the CDC & waiting for you at CVS. Schedule: [link]

For the 50,000 patients who were randomized to getting a free ride, they were sent a similar text message that also included a link with a discount code that would open the Lyft app and provide them with a “Lyft Pass” to bring them to and from a nearby CVS location for free. Here’s what it would look like for a hypothetical patient in San Francisco:

What did the study find?

To best understand the results, let’s start with the control group, who were due for a booster but didn’t get any reminders or free rides. Of these 492,000 patients, about 5% of them got their booster at a CVS over the 30-day study period.

Next, let’s look at the patients who got the various text message reminders but weren’t offered a free ride—about 3.1 million of them in total. Across the various types of text messages and reminders, 6.0-6.3% of them ended up getting vaccinated at a CVS in the next 30 days—a modest, but significant increase over the control group.

Among the 50,000 people offered a free ride (as part of a text reminder), 6.1% of them ended up getting vaccinated at CVS in the next 30 days—no more than the patients who only received a text-message reminder—suggesting that the free Lyft to a CVS didn’t actually lead to more vaccinations above and beyond what was achieved by text messages alone.

Share

Why didn’t a free Lyft increase vaccinations?

This study offered pretty convincing evidence that text-message reminders could modestly increase vaccination rates (a finding consistent with prior research on the topic), but that an additional free ride to the vaccine didn’t make a difference. But impressive as this study was in its size, it doesn’t tell us exactly why the free ride didn’t make a difference.

Looking through the study’s supplemental data, we see that of the 50,000 patients offered a free ride to the vaccine, only 0.1% actually took one (just 50 people!)—and these rides were estimated to be worth about $20 per patient. And while some people may simply not be into ride share apps and/or not have the Lyft app on their phone, this is still a remarkably small number.

At the same time, there are so many CVS locations that in major metro areas, the closest CVS in any given location within a metro area was estimated to be less than a mile away, on average. This means that if CVS locations are purposefully located to be in convenient areas that people might pass in their daily lives—and presumably they are—a free ride just might not be needed for someone interested in getting a vaccine.

While there are many more fascinating details of this study, taken as a whole, it says to us that contrary to intuition, physical transportation to a vaccination site may just not be a meaningful barrier to COVID booster vaccination—at least among the patients in this study who had previously brought themselves to a CVS to get their initial COVID vaccine series.

It may be that for other populations—such as those who usually get vaccinated at their doctor’s office (which might be farther away), those who are less independently mobile, or those living in more rural areas—a free ride might make more of a difference.

Subscribe now

The bigger picture

As we’re sure you’re well aware from our vantage point here in 2024, hitting vaccination goals to maximize public health can be quite challenging. There are many reasons for this that the World Health Organization classifies into three broad categories known as the “3 C’s”: factors of complacency (perceiving risks of the disease or the benefits of a vaccine to be low), confidence (lacking trust in the medical system and vaccine manufacturers), and convenience.

In our book Random Acts of Medicine (now available in paperback to bring to the beach this summer—we had to get the plug in!), we break down a study of ours that focused on that last factor of “convenience.” Our results suggested that the inconvenience of having to bring young children to an extra doctor’s appointment to get a flu shot substantially reduced the chances of their ever receiving one for a given flu season.

But getting vaccinated is inconvenient for adults, too, and there are real costs to doing so. People may lose out on income if they take off work to get the vaccine, have travel expenses like gas or bus fare (a cost the Lyft megastudy tried to address), and for some, the vaccine itself may have a monetary cost as well (fortunately this is an increasingly uncommon occurrence).

But there are also the non-financial economic “costs” of vaccination: the time it takes to go to and return from the vaccination site, time spent waiting and filling out paperwork, pain from the injection itself, discomfort from known side effects from the vaccine (like fever or sore arm), to name a few. A free ride to and from the vaccination site doesn’t ameliorate any of these costs—and it certainly doesn’t address other vaccination barriers of complacency or confidence.

It’s really important, however, to do what these researchers did and actually investigate the various factors if our goal is to increase vaccination rates. Intuition would tell us that a text message reminder might help a little bit (it did) and free ride might help a lot (it didn't). As much as “megastudies” like this one can be highly informative, they’re probably too much of a Lyft (ahem) to study every aspect of vaccination behaviors—which means we’ll have to take advantage of all the data we can to better understand this complex topic.

Share

Leave a comment

Subscribe now

Buy the book: RANDOM ACTS OF MEDICINE

The exams that allow us to graduate medical school, complete residency, obtain a medical license, and maintain board certification are largely based on knowledge of specific facts about how the body works, how it gets sick, and how we can prevent and treat illness and death. Within the profession, we clearly believe medical knowledge is a priority, having medical students spend most of their time in the classroom and lab for two years before spending significant amounts of time with patients in the office or hospital.

Being knowledgeable, however, is necessary but not sufficient—it’s our application of knowledge that defines the expertise that patients want from us and that we owe to anyone who places their life in our hands. (There are of course many other important qualities in a physician, which we explore in Chapter 9 of Random Acts of Medicine.)

A critical part of that expertise is learning what to do with the deficits in our knowledge. While it seems particularly important today when the knowledge base about the human body is exponentially larger than it was in, say, the mid 20th century before the discovery of DNA, this has always been an essential trait of a good doctor. The Socratic lesson that there is wisdom in “knowing that you don’t know” imposes a degree of humility on physicians, which might seem in opposition to the frequent pressure on us to project confidence.

It takes both effort and experience to effectively balance that humility and confidence in our own knowledge. This means it’s important to train our resident doctors in managing what they don’t know—in addition to the more obvious training in using what they do know—when it comes to taking care of patients.

Subscribe now

One way to train medical residents in the unknown

At Mass General Hospital where we practice, the internal medicine residency’s Pathways program takes a unique approach to training physicians in what they don’t know over the course of a two-week training rotation.

I (Chris) sat down with Pathways co-director and endocrinologist, Dr. Marc Wein to learn more about the program (Marc was also one of Bapu’s co-residents, back in the day). I started off my asking him about the origins of the program. (Our interview has been edited for brevity and clarity).

Marc Wein: Around 2016, there were two excited physician-scientist residents, Lauren Zeitels and Victor Fedorov, who felt a little bit frustrated by the fact that they had these really busy lists of patients that they were taking care of in the hospital. They felt there were missed opportunities to learn from their patients who had unexplained things going on, and they wanted to spend more time than they had available to think about those unexplained things.

There were aspects of their patient cases that didn’t make sense, like unusual presentations of common diseases or funny presentations of mystery illnesses. They just really wanted to spend a stretch of time in these proverbial “deep dives” to be able to think about one patient instead of taking care of dozens of patients on a busy inpatient service. They wanted to read as much as they could, talk to experts in the area, and even collect additional samples from the patient to bring to the lab for research.

Chris Worsham: So the idea is if I’m a busy resident that’s responsible for a bunch of patients, we have to do what we can for them in the hospital, get them feeling better and on the right track, and then get them out of the hospital. If there’s something where we don’t really know what’s going on but it’s not a pressing issue, we often just sort of shrug our shoulders and say “it’d be nice to know, but we just can’t deal with this right now, and we’ve got to move on to the next patient.” So with Pathways, the idea is, well, what if you did have the time to actually explore those things that regular patient care just simply does not permit you to do in this day and age?

MW: A big component of this is “how do you maintain your joy in work as a resident?” I think that for many of our residents in the physician-scientist track [who plan on research careers], a lot of that joy comes from discovery and thinking of unknowns in biology and how that applies to medicine. So initially, there was a small group of residents who really wanted to do this, but is has since become a two-week rotation that all residents—physician-scientists or not—all do as part of their training.

CW: So what do the two weeks look like? Where do they find patients to learn more about?

MW: We have a chief resident who is dedicated to Pathways, and they’re responsible for having a bunch of cases at the beginning of the block. Referrals come in from the other residents working on the hospital floors or ICU. Then we take a look at the referrals to find ones that will be a good patient for Pathways, since often a patient with, for example, multi-system organ failure who’s really sick in the ICU isn’t an ideal case because there’s just too much going on.

Here’s a good case. Recently we had a patient who had an inflammatory disorder which caused him to lose 70 pounds over about a year. The question of “why do you stop eating when you get sick?” is a really interesting question. It gets at things like the body’s control of appetite, or whether there is an evolutionary advantage to not eating when you’re sick, or how nutritional intake impacts your ability to heal or fight infections. There are a lot of great questions here about something we see fairly frequently, yet doctors don’t have the chance to deeply investigate and learn about during normal hospital care. Other good questions have been “why does one patient develop delirium [temporary confusion and hallucination] when they’re sick while another similar patient does not?” Residents can look into what is known about what causes delirium. What are the neural pathways? Why are older people more susceptible? All kinds of big questions come up. And so the first part is case selection.

CW: When I was a chief resident, we would always have the temptation to find these really rare or bizarre cases to use for teaching. And of course those are interesting—that’s why we see them on TV—and they have a role in educating residents. But actually spending time looking into understanding on a deeper level things we think we know the answer to, but really don’t—like what is the actual underlying reason why patients lose their appetite or develop delirium—can offer a lot more teaching as residents begin to develop their expertise. [N.b.: This is a guiding principle over at the Curious Clinicians podcast and newsletter].

MW: Every year we have a new chief resident who comes in, and the first thing we tell them is “this is not an episode of House.” We’re not here to figure out some really clever diagnosis that all the specialists have missed. The goal is to really think about why things are happening the way there are.

Another example is a patient who has had a heart transplant, which means that they don’t have any nerves connecting their brain to their transplanted heart. We might ask: How does that affect their body’s physiology? At the beginning, the team of residents will meet the patient, and might even collect some research samples if the patient is interested. They’ll spend the first day or two refining their questions—and we try to emphasize that the questions need to be simple. Next, they’ll start meeting with experts in the area, and in the age of Zoom, these could be people anywhere around the world. They’ll also meet with one or two of the “core scientists” who are part of the Pathways faculty, but we deliberately choose someone who does not have expertise in the area to help them think about how to answer the questions as a non-expert.

Over the course of their block, they put together an educational conference for everyone that doesn’t focus on the patient’s clinical care, but reviews what is known and what is not known about some aspect of that patient’s condition. They also then propose some experiments that could be done to answer those questions. There have even been a couple of examples of residents taking these research questions, inspired by their Pathways block, back into the lab. [Examples here, here, and here].

CW: That’s fascinating. Are patients generally enthusiastic about this? I could imagine someone being a bit wary and not wanting to feel like they’re being experimented upon.

MW: The way we pose it to patients is “We have a group of trainees who want to study aspects of your illness right now and see if we can come to a greater understanding—most likely, we won’t, but there’s a chance that we will. Do we have your permission to do that?” So expectations are low. But there have been a couple of instances where the teams have developed strong relationships with the patients, to the point where they will even attend the presentation that the residents put on. One patient even presented his own case for the first five minutes of the conference, which was tremendously moving, and the team kept going back and forth to him afterwards.

CW: That’s so cool. I think what interests me the most is that this gets at what you’d call metacognition—thinking about the way we think about patient care. I imagine a big part of this is to give residents a way of thinking about the things they don’t know that will last them beyond this two weeks and hopefully for their entire career. What’s your impression of how the residents see their job or see medicine differently coming out the other end of the block compared to going in?

MW: I think when we’re successful, it has a huge impact. The hardest part is for the residents to transition from a busy hospital service where the thinking is “How can we move this patient through?” to a mindset of “How can we slow down?” So for the first few days we often have to tell them to stop worrying about things like the patient’s daily potassium levels and to focus on the questions that we’re really trying to answer. And then the faculty and chief resident will back off for a bit, and check back in, and it is almost always the case where they have a moment where it clicks on what we want them to do. They often come out the other end thinking “we didn’t realize there’s so many directions that we could go thinking about how our patients presented to us and how we might be able to help them.” And every once in a while there’s someone who finds this so compelling that they consider a research career for the first time.

CW: Regardless of the type of research, one thing that is similar at the core is the importance of thinking creatively about research ideas, finding a good source of those ideas, and honing them into actionable research questions. No matter what discipline you’re in, that’s not an innate skill, right? That’s something you need to practice as a researcher, or similarly, as a physician who is looking to improve patient care. You need to learn how to see what’s in front of you, see there’s something different going on here, admit to yourself “I don’t know what this is,” and come up with a plan to figure out what you should do next. That’s a muscle you have to train and build up.

MW: Yes—and figuring out what the right kind of tool to apply to your question is actually what we spend a lot of time on with the groups, because often the investigation they’re proposing is going to be very different depending on the question they’re asking.

CW: I was thinking about an example that came up with my own practice recently. For pulmonologists, getting patients oxygen therapy at home can be quite difficult, and most of the time we shrug it off as a headache as we deal with insurance and oxygen supply companies. But to look into the question of why is this process so difficult ended up being a fascinating examination of how the health care system pays for these services and the financial incentives that are created. So I imagine there are opportunities beyond understanding what’s happening on the level of the organ, the cell, or genetics.

MW: There are all kinds of huge questions. We had one case of a patient with severe anorexia, and initially we were interested in some of the issues surrounding electrolyte imbalances that can come up. But the team really wanted to go into issues of access to care, and the question they ultimately focused on was “Why is it so hard to get a psychiatric hospital bed for patients with eating disorders?”

CW: I tend to think of the health care system as an organism in and of itself—so to me, this is a question of physiology just as if they had they looked into the body’s handling of electrolytes. The approach is the same, whether you look at it from the 30,000-foot view, under an electron microscope, and everything in between.

MW: I remember when Bapu and I were residents together at Mass General, and he would ask questions on rounds about the policy-level issues that were affecting care. And people would be like “we’re just trying to finish rounds here.” But it’s exactly the same kind of thing—there are so many questions, and you’ve got to be able to stop and think them through. And if you want to do that, you need to be really nimble, you need to be able to pick the right questions and find the right tools to answer them.

CW: So in closing, from your standpoint as an educator, what does this program mean to you? Does it scratch any itches you’ve had when it comes to teaching?

MW: One of the goals of the program when it first started was to have it generate research. But it turns out, it’s not always a great research program, because most of the time, we find something interesting and generate questions that for many reasons we still can’t answer. But, over the years with major input from dedicated colleagues including Dr. Mark Fishman, it has turned out to be this incredibly powerful educational tool for teaching residents ways to approach what they don’t know.

What’s exciting for me is that it’s teaching people how to think and exposing smart trainees who came into medicine for good reasons to new ways of thinking. It’s helping them find new sources of joy in their work and giving them tools to be able to answer questions that might come up in their practice. To me, that is the best thing that an educator could possibly do—just having a little bit of ability to do that is very rewarding.

The two residents who started this program, Lauren Zeitels and Victor Fedorov, both died tragically in an avalanche. They were really passionate about this and instrumental in making the Pathways program happen, and Lauren’s parents continue to attend some of the conferences. So it also takes on this additional deep meaning because of that tragic accident, and in many ways we continue to do this in honor of Victor and Lauren.

(Donations in their memory can be made to the MGH Department of Medicine Pathways fund.)

Share

Leave a comment

Share Random Acts of Medicine

In case you somehow haven’t heard, there will be a solar eclipse on Monday, April 8th, with a total eclipse visible in parts of the U.S. inside the “path of totality.” Solar eclipses are undeniably cool, and the hype only increases when there’s the additional factor of having the path of totality cross the mainland U.S. If that sounds familiar, it should: a similar total solar eclipse occurred in 2017.

The 2017 eclipse was welcomed with much fanfare, with a NASA official stating “Never before will a celestial event be viewed by so many and explored from so many vantage points – from space, from the air, and from the ground.” It seemed like everyone wanted to see the eclipse, including the President of the U.S.—who famously broke rule #1 of solar eclipses by looking directly at it without safety glasses (which he later put on).

Some enthusiasts really wanted to see the total eclipse, which was easy if you lived inside the path of totality (the 2017 gray band spreading across the U.S. in the map below).

Anyone else who wanted to see the total eclipse to fully experience the daytime darkness had to travel to the relatively narrow path of totality.

In a new research study published in JAMA Internal Medicine, physician researchers Donald Redelmeier and John Staples wondered what problems that travel might have caused.

The hazards of eclipse tourism

A joint survey study between NASA and the University of Michigan estimated that about 20.5 million Americans traveled to another city to view the eclipse (though only about 31% of them traveled all the way to the path of totality). Unprecedented traffic jams were seen in areas across the country—especially less populated areas like Wyoming where an acute influx of visitors can easily overwhelm the roads.

In their study, Redelmeier and Staples asked a simple question: did all that extra traffic lead to an increase in fatal car accidents?

Using data from the National Highway Traffic Safety Administration, they took a look at the 3-day period comprising the day before, the day of, and the day after the 2017 solar eclipse (August 20 to 22, 2017), figuring that many people would have to spend at least a day traveling to and from their eclipse viewing location. They then compared this three-day period to two other three-day periods occurring 1 week prior (August 13-15, 2017) and 1 week later (August 27-29, 2017)—a reasonable set of controls that could account for some of the typical fluctuations in car accidents based on the season and the day of the week—for a total of 6 control days. For each car accident, they figured out when it occurred relative to the eclipse (which occurs at different times in different locations as it passes across the country) based on the location of the crash.

What did they find? In the 3-day period surrounding the eclipse, there were 31% more people involved in fatal crashes than in the control periods +/- 1 week of the eclipse. Looking at the timing of fatal accidents in the hours surrounding the eclipse (plot below), accidents tended to occur at similar times of day as during the control periods, but there were more of them. The study authors estimate that these increases surrounding the eclipse are similar in magnitude to the increases seen in travel-heavy periods surrounding Thanksgiving, Memorial Day, and July 4th.

Share

Is there more to this story?

It’s quite clear from this study that the eclipse led to additional traffic fatalities. This is a solid natural experiment, since a solar eclipse is randomly timed when it comes to the risk of traffic fatalities, and there is nothing else that occurred at that time that would explain such stark differences between the eclipse and control weeks.

While this study makes the “what” happened pretty clear, it doesn’t tell us as much about the “why.” Obviously, most of this can be explained simply by more people on the road for more hours and in more congestion, and this will translate to higher numbers of accidents.

There are, however, some other ideas that came to mind when reading this study. If you look carefully at the plot above, you’ll notice that after the eclipse (in orange), traffic fatalities are about as high as they were before the eclipse. But they are relatively higher after the eclipse when compared to the control weeks (in dark blue). We don’t just have to eyeball it though; by the numbers, the increase in fatal accidents compared to control weeks was higher in the time period after the eclipse (49% higher) than it was before the eclipse (21% higher). In other words, the highest risk period for eclipse-related traffic fatalities is after the eclipse has already happened.

We think this could be explained by a few things.

The first, and perhaps most likely explanation is that while people may have headed to their eclipse-viewing destination at different times, once the eclipse was over, many would start heading home at the same time—this assumption is backed up by reports of a “post-eclipse rush” in 2017.

Another explanation could be the influence of drugs and alcohol from eclipse-related celebrations. Even though the study found that overall, eclipse-related fatal accidents were more often reported as having “alcohol not detected” compared to the control weeks, to the extent there were eclipse celebrations involving substances, they might impact the post-eclipse period more than pre-eclipse.

Finally, we wondered if more accidents were occurring after the eclipse compared to before the eclipse because of vision issues. While we imagine that someone who is enthusiastic enough to travel for an eclipse would be aware of the necessary eye safety measures, they may not have been able to acquire protective glasses or may have taken an unprotected look at the sun just before or just after the totality. If eye damage from the sun (known as solar retinopathy—more on this below) is impairing vision in significant ways, we’d expect to see more traffic accidents after the eclipse than before it.

Most likely though, the roles of eclipse-induced solar retinopathy or increases in driving under the influence, if any, are far smaller than the role of the post-eclipse rush.

Subscribe now

What does this mean for this year?

The path of totality for Monday’s eclipse is much larger than the previous one, with many more Americans already living in its path, which means fewer people would have to travel to experience the totality. On the flip side, it could be that recent memory from the 2017 eclipse has increased interest in experiencing the total eclipse, spurring more people to travel.

Experts are recommending that those who travel plan to stay a while in their destination and try not to leave immediately to avoid the post-eclipse rush. Eclipse tourists can hopefully find ways to stretch out their trips by exploring new areas (and boosting their local economies) to make the eclipse experience extra special—and a bit safer for everyone on the road.

N.B. on eclipse safety from two doctors who just can’t help themselves

It’s only natural to want to look at the eclipse, but doing so poses a real risk of eye damage in the form of “solar retinopathy,” or burn-like damage to the layer of cells in the back of the eye (the retina) that detect light and send visual signals to the brain.

We all know that you can damage your eyes by staring directly at the sun on a normal day—after all, you can start a fire with sunlight using a magnifying glass, which is not unlike the lens in your eye. But what many don’t realize is that even though the sun is being blocked by the moon during an eclipse, there’s still enough light coming through to cause retinal damage except during the several minutes of totality for people watching from within its path. Staring directly at the non-total eclipse, even for short durations, can lead to solar retinopathy that might affect vision, and for some it will be permanent.

To view the eclipse safely, NASA and the American Astronomical Society require the use of “safe solar viewers” that meet the ISO 12312-2 international standard. They have provided a large list of suppliers that meet this safety standard here: https://eclipse.aas.org/eye-safety/viewers-filters

Subscribe now

Buy the book: RANDOM ACTS OF MEDICINE

Leave a comment

Share

While the arrival of spring means much awaited warmer weather in many parts of the country, for many, it also means the arrival of allergens. Pollens from plant life and molds thriving from rainfall can trigger hay fever and asthma—very common conditions that if you don’t suffer from one yourself, you almost certainly know someone who does.

Asthma, allergies, and eczema tend to run in families, and there are quite clearly genetic factors that contribute to the risk of developing one or more of these conditions. But there are also factors in the living environment contribute to the risk of developing these conditions.

One such factor is exposure to farm animals; studies have shown that children exposed to farm animals—and the environmental bacteria that hang around with them—tend to develop asthma less often. But is it the exposure to farm animals that causes this, or is it just that families who farm simply have a lower risk for asthma to begin with?

Teasing apart what is genetic versus what is a product of the environment is really challenging, since genetically similar people—like extended families—frequently live in the same environment, eat the same foods, and are exposed to the same allergens. So if a child develops asthma, how could we know if it is a result of genetic risk factors, environmental risk factors, or both?

A natural experiment could really help here. If genetically similar people were exposed to different environments for reasons that were unrelated to their genetic risk for asthma, it would be possible to tease apart the influence of environmental factors from the genetic ones.

A 2016 study found one such natural experiment happening right here in the U.S.—in South Dakota and Indiana, to be more precise.

The Amish and The Hutterites

The Amish and the Hutterites are two groups of farming communities in the U.S. who, along with the Mennonites, share a common ethnic and religious background that traces back to the religious movements of 16th century Europe. The study authors—a large group of researchers with expertise in asthma, genetics, immunology, allergy, and environmental science—wrote in the New England Journal of Medicine:

We designed a study that compares two distinctive U.S. farming populations—the Amish of Indiana and the Hutterites of South Dakota… These two particular groups of farmers originated in Europe—the Amish in Switzerland and the Hutterites in South Tyrol [Northern Italy]—during the Protestant Reformation and then emigrated to the United States in the 1700s and 1800s, respectively.

Both groups have since remained reproductively isolated. Their lifestyles are similar with respect to most of the factors known to influence the risk of asthma, including large sibship size, high rates of childhood vaccination, diets rich in fat, salt, and raw milk, low rates of childhood obesity, long durations of breast-feeding, minimal exposure to tobacco smoke and air pollution, and taboos against indoor pets.

However, whereas the Amish practice traditional farming, live on single-family dairy farms, and use horses for fieldwork and transportation, the Hutterites live on large, highly industrialized, communal farms.

So the Amish and the Hutterites share common ancestry and have similar lifestyles when it comes to family life and diet. Where they differ is the amount of exposure they have to animals, since the Amish spend much more time face-to-face with animals while farming compared to the Hutterites, such as by plowing a field with horses (top photo of an Amish farm) rather than industrial machinery (a Hutterite farm below).

What does this have to do with asthma? The study authors continue:

Strikingly, the prevalence of asthma in Amish versus Hutterite schoolchildren is 5.2% versus 21.3% and the prevalence of allergic sensitization is 7.2% versus 33.3%, as previously reported.

Assuming the Amish and the Hutterites are indeed genetically similar after generations of “reproductive isolation” as the authors put it, they would be at a similar genetic risk of asthma. This would leave differences in non-genetic risk factors for asthma as the underlying cause of the differences in the asthma rates between Amish and Hutterite children.

Subscribe now

The data

Researchers collected data on 30 Amish and 30 Hutterite children, matching them by age and sex to ensure they were similar groups to compare. They sampled the children’s blood to analyze their DNA and measure levels of immune reactivity. They then went into 10 Amish and 10 Hutterite homes, collecting dust in the air and vacuuming the floor and mattresses to measure the allergens present in each of the homes.

Their findings? First, they were able to confirm the assumption that after so many generations in the U.S., these children still shared common ancestry. Their genetic profiles were highly similar, suggesting the Amish and Hutterite children would have similar genetic risk for asthma.

Consistent with the previous research on asthma in Amish and Hutterite children, they found that none of the Amish children had asthma, while 6 (20%) of the Hutterite children did.

Amish children had lower levels of IgE—the antibody type most commonly associated with allergy and asthma—against common allergens than Hutterite children; meanwhile, levels of the non-allergic antibody types were similar. Amish children also had lower levels of eosinophils (the white blood cells implicated in severe allergic responses) and differing profiles of overall immune reactivity in their blood.

As far as allergen levels in the home, common allergens (from cats, dogs, dust mites, and cockroaches) were found in 4 out of 10 Amish homes but only 1 of 10 Hutterite homes. Levels of environmental bacteria in the air were 6.8 times higher in Amish compared to Hutterite homes, with differing profiles of bacterial species found in the mattresses.

Taken together, it appears that while as a group, Amish and Hutterite children share genetic risk factors, they live in significantly different environments when it comes to allergen exposure. Amish children are exposed to more animals, more allergens, and more environmental bacteria. Yet they have less asthma, and their blood samples show lower levels of allergy by several measures.

What gives?

The Hygiene Hypothesis

Simply put, the “hygiene hypothesis” reflects current scientific thinking that overly-clean environments can lead to the development of asthma and allergies—that regular exposure to aspects of our environment helps suppress immune responses that lead to allergy. If the immune system regularly encounters certain substances—such as fragments of environmental bacteria—it will be less likely to generate the types of immune responses that generate asthma or severe allergic reactions.

The hygiene hypothesis was sitting right there at the center of the Amish and Hutterite study: the Amish were encountering more animals, more dust, and more bacteria while developing less asthma than the Hutterites. This alone supports the hygiene hypothesis, but the researchers took it one step further.

It’s possible to artificially induce asthma in laboratory mice by exposing them to a chemical called ovalbumin, which tends to induce an allergic immune response. When the mice are exposed to ovalbumin again, they develop signs of asthma.

So the researchers took a group of laboratory mice and exposed them to ovalbumin in an attempt to induce asthma. But they also exposed the mice to extracts of the dust they collected from the Amish and Hutterite homes. They found that the mice who inhaled extracts of Amish house dust were protected against developing ovalbumin-induced asthma, while the mice who inhaled Hutterite house dust weren't and went on to develop asthma.

Share

The bottom line

Because they share common ancestry and remain genetically similar, the Amish and Hutterite communities in the U.S. created a unique opportunity to understand how environmental factors influence the risk of developing allergic conditions like asthma. And even though their study was small with only 60 kids, this clever set of researchers made the most of the natural experiment sitting in front of them.

Simply observing that two genetically-similar groups—with similar diets and similar lifestyles other than their use of animals versus machinery—have different rates of asthma is enough to tell us that there must be something different in their environment contributing to the risk of asthma. These researchers took the initiative to then go out and measure what, exactly, that might be, making a pretty good case for exposure to animal and environmental bacteria.

Leave a comment

Subscribe now

Buy the book: RANDOM ACTS OF MEDICINE

In Chapter 3 of the Random Acts of Medicine book, we dive into a study of ours that took advantage of the random nature of kids’ birthdays. That study showed that the timing of kids’ well-child checkups, which tend to happen around birthdays for young children, impacted whether or not kids would get their annual influenza vaccine (flu shot).

In our new study with , which is out today (no paywall) in the BMJ, we again take advantage of the random nature of birthdays. This time, however, we look at how birthdays impact when vaccinated kids actually get their flu shot—and what this natural experiment can tell us about the optimal time of year to get a flu shot.

We wanted to share this with you the day it came out—always an exciting day for researchers! Check out the abstract below. If reading the abstract or research paper itself is not for you (no judgment there!), here’s some news coverage that’s worth checking out:

• Boston Globe

• Time

• Harvard Medical School News

N.B.: If you haven’t yet had a flu shot this year, it’s still worth getting one now even though it’s late in the flu season!

-Chris & Bapu

Although it’s the place we should be when we’re very sick, the hospital is nevertheless a dangerous place. For starters, simply having an illness that requires hospitalization in the first place puts patients at increased risk for new problems. Then, there’s risks associated with the various treatments we give patients—risks we believe are outweighed by the benefits of such treatments, but risks nonetheless. Then there’s risks of simply being in the hospital, away from routines, in an uncomfortable bed, being awoken for vital signs and blood draws.

It’s no wonder, then, that patients who come into the hospital with one problem frequently develop new, “hospital acquired” conditions. Some of these are benign, but annoying; constipation plagues hospital units around the world. Others are more serious and even life threatening, like blood clots or pneumonias.

But as a new study highlights, there’s one simple intervention that can reduce the risk of pneumonia for hospitalized patients—so simple, a toddler could do it: brushing their teeth.

Subscribe now

What is pneumonia, anyways?

Pneumonia is one of the most common causes of hospitalization and death in the U.S. and across the world. It’s a serious illness that is the result not only of a bacteria or virus infecting the lung, but the inflammatory response that the infection triggers. This can render large portions of the lung unusable for the purposes of adding oxygen to and removing carbon dioxide from the bloodstream—essential functions for life.

To cause pneumonia, bacteria need to make their way into the lungs; the main way this happens is that bacteria work their way down from the mouth with other liquids that enter the airway in an event called “aspiration.” We’re all aspirating in tiny amounts all the time, and occasionally we aspirate in larger amounts (think of that overly-aggressive sip of water you choked on that made everyone ask if you were OK). Our airways have mechanisms to clear out things that shouldn’t be down there, but under the right conditions, once deep in the lung, certain bacteria can cause pneumonia.

For patients who are in the ICU on ventilators, one way we try to prevent pneumonias is by washing their mouths out with a disinfectant called chlorhexadine. Because these patients have a breathing tube going down their throat and many of them are unconscious, the usual mechanisms that prevent bacteria-containing liquid from entering their lungs aren’t working. The idea is that if you kill the bacteria in their mouth, any saliva that does end up down in the lungs won’t have as much bacteria in it. Studies on chlorhexidine for oral care for patients on ventilators, however, are mixed when it comes to their ability to reduce pneumonia and save lives.

Of course, the rest of us reduce the burden of bacteria in our mouths using a toothbrush, a centuries-old tool that, along with toothpaste, helps remove bacteria and debris those bacteria feed on. And while the reason we do this every day is to benefit our teeth and gums, it does make you wonder: can toothbrushing help prevent pneumonia in the hospital?

The data on toothbrushing

A new meta-analysis study by Harvard researchers Selina Ehrenzeller and Michael Klompas combined data from 15 different randomized controlled trials from around the world on toothbrushing to prevent hospital acquired pneumonia. They combined these studies to make one large study that included 2,786 patients randomized to a toothbrushing intervention versus control. (The control arms of these studies varied slightly, though most commonly they were comparing chlorhexidine + toothbrushing to the use of chlorhexidine alone).

They found that across these studies, toothbrushing cut down the risk of pneumonia by about one third, both for regular hospitalized patients and for patients on ventilators (a patient who can’t do it themselves would have their teeth brushed for them). The results were similar whether patients brushed their teeth 2, 3, or 4 times a day.

For patients in the ICU, the toothbrushing was associated with a 19% reduced risk of death in the ICU (studies of other patients did not include data on mortality). These ICU patients who had their teeth brushed also tended to spend less time on the mechanical ventilator and less time in the ICU than those who didn’t get their teeth brushed.

Share

What does this mean for hospital care?

While this study is a departure from the natural experiments we typically focus on, it does demonstrate that a simple, everyday activity can have a real impact on patients. It also highlights how disruptive hospital care can be—toothbrushing is a part of everyday life for most patients, but it can easily be dismissed as unimportant when patients are feeling unwell (or unconscious as many are in the ICU).

We should also touch on the advantage of the meta-analysis research strategy of combining smaller randomized trials into one big one: it preserves the benefits of randomization from each trial. At the same time, it provides more clarity and confidence in the estimate of the effect size by including more patients. In this case, the effect appears pretty sizable, and it comes at very little cost—we have toothbrushes and toothpaste for patients in the hospital already, we just don’t always do it.

For hospital wards and ICUs that were unsure of the benefits of oral care and toothbrushing, we hope this study will help ensure it doesn’t get forgotten as a part of daily patient care. Remembering, of course, is easier said than done, which is why it’s important to make toothbrushing a part of daily protocols and routines in both the busy hospital environment as well as the chaotic environment of a household with two toddlers—though the Worsham family gets some assistance from Daniel Tiger, of course:

Leave a comment

Share

Subscribe now

Hungry judges

In 2011, a highly publicized research study by Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso examining rulings by Israeli judges found that over the course of the work day, judges tended to issue rulings most favorable to prisoners at the beginning of their work day and right after food breaks. Between breaks, the judges’ rulings became progressively less favorable until picking up again after the next food break, as depicted in the figure below:

The pattern persisted when the researchers compared rulings on cases with similar legal characteristics (e.g. severity of the crime, time already served, prisoner’s demographic characteristics), when they looked at each of the 8 judges individually, and after looking at overall time elapsed after a break (as opposed to the number of cases). So the data was very clear about one thing: the farther out from a food break, the less favorable the judges’ rulings.

The interpretation was that important legal rulings were being impacted by judges being hungry and/or in need of a break—factors that should have no bearing on such rulings. The study made headlines around the world, and it is often cited as an example of how arbitrary factors can get in the way of justice.

But was that interpretation correct? We’ll come back to this question, but first, let’s look at a similar study in medicine.

Subscribe now

Tardy doctors

Like judges with a daily docket, doctors seeing outpatients also have a schedule of patients to see throughout the day, usually with a midday break for lunch (or, more realistically, scarfing down some food while attending some meeting or educational conference). For example, a doctor spending a whole day seeing outpatients might have two sessions, one from 8am-12pm and another from 1pm to 5pm.

As many of you have probably experienced, doctors seeing outpatients have schedules that are packed with patients, and it’s easy to start running behind. But if you’re the first patient of the day (or the first patient seen after lunch), you’re going to be less likely to have your doctor running late than if you’re one of the last patients in the session, where previous delays can add up. Doctors don’t like running behind either, so we might try to “catch up” by moving things along more quickly with later patients than we would have had they been the first patient of the day.

If a doctor is running behind, they might be more inclined to jump to a “quick fix” for a patient. For example, if a patient comes in asking for antibiotics for something a doctor is pretty sure is a viral infection, it’s faster and easier to just prescribe antibiotics than it is to have a conversation with the patient about why antibiotics aren’t the best choice at this juncture.

Alternatively, if a patient comes in complaining of pain, it might be easier to just prescribe an opioid than it is to craft a more elaborate plan that involves non-opioid pain medications, physical therapy, and other modalities. These “multi-modal” pain management plans can provide better pain relief than opioids, but it can take a lot of time and energy to put these plans together with a patient. Meanwhile, a potentially unnecessary-but-easy prescription can place a patient at risk for all of the complications associated with opioid use.

A study by Hannah Neprash and Michael Barnett (both collaborators of ours on other projects) looked at what happened to patients who were seeing their primary care doctor with pain complaints. The patients were all coming in with a new pain problem and had not been previously prescribed opioids in the past year. Using data from the electronic health record, they were able to see both when the patients were scheduled to be seen and when they were actually seen by the doctor.

They found that over the course of the day, patients who were seen later for pain complaints were more likely to be prescribed opioids than patients with pain seen earlier in the day—going from 4.0% of visits to 5.3% of visits. They also found that the more the doctor was running late, the more likely they were to prescribe an opioid (this was after accounting for the time of day of the scheduled appointment).

Meanwhile, they didn’t see any differences over the course of the day when it came to prescribing of non-opioid NSAID medications or physical therapy for patients with pain, nor were there trends in prescribing of non-pain related medications for high blood pressure or cholesterol. These conditions essentially served as control groups.

Neprash and Barnett estimated that if opioids were prescribed at the same rate over the whole day as they were during the first three appointments of the day, there would be 4,459 fewer opioid prescriptions among the 678,319 appointments they studied. Each opioid prescription carries risks of adverse outcomes—risks we wouldn’t want patients to have to bear if the only reason they were prescribed the drug was because of when they happened to be scheduled.

It appeared that doctors running late—maybe also tired at the end of the day—could have been putting patients at unnecessary risk from exposure to opioids. If this was the case, it’s evidence of the common complaint from physicians that we can’t practice the best medicine when our schedules are overstuffed with patients. But is this what was happening?

Share

Assumptions

To establish cause and effect in these studies—i.e., to say that hungry judges and tardy doctors are causing problems—we have to rely on an important assumption: that the scheduling of these legal hearings and doctors appointments is as good as random.

Assumptions are impossible to prove (which is why they are assumptions), but we can find evidence to support them. If we’re assuming something is random, then it should divide people up evenly; we should see that judges’ cases and doctors’ patients seen at the beginning of a session are similar to those seen at the end of the session. For the judges, we saw similar crime severity, time served, and demographic characteristics at the beginning and end of sessions. For the doctors, patients were mostly similar, but patients seen earlier in the day were more slightly more likely to be men, to have chronic conditions, and to have longer appointments.

The reason we’re forced to assume is because there are many unmeasured characteristics of the cases for which it is impossible to confirm that they’re evenly distributed between those seen early in the session and those seen later. If there are details we can’t measure that both lead a case to go earlier or later in the day and lead to differences in verdicts or opioid prescriptions, then we can’t conclude that it’s only the scheduling that’s leading to the differences in outcomes. This phenomenon of “residual confounding” means that we can’t be sure that other factors besides timing are impacting the findings, in part or in whole.

While we can’t speak for what judges do, doctors’ offices do sometimes schedule patients they anticipate might put them behind schedule later in the day, to minimize disruptions to other patients. Judges offices might do the same thing—put cases they know will take longer (and that bend toward one end of the favorability scale) toward the end of the day. Identifying such cases in the data is really tricky, making it challenging to find evidence to support these assumptions.

So what’s the answer?

For both studies, we’re left with a big old “maybe.” Each study has some evidence supporting the assumptions. But the judges study only had a few characteristics that could be compared, and the doctors study showed some small differences on a handful of measured characteristics that, even when statistically accounted for, could be a signal of differences in unmeasured ones.

On the judges, a follow up simulation study has suggested that scheduling cases that will take longer and are more likely to be ruled favorably earlier in the day could explain the results of the original study by Danziger, Levav, and Avnaim-Pesso.

On the doctors, Neprash and Barnett note:

The results of this observational study should not be interpreted causally, although our findings were robust in several sensitivity analyses addressing selection bias. There could be unobserved reasons why patients more likely to receive opioids may have appointments later in the day, although we observed similar results in our analyses focusing on appointment lateness, which controls for hour of appointment.

We’ll also add that when looking at non-opioid prescriptions, like those for high blood pressure or cholesterol, you might think that patients seen later in the day would be more likely to be prescribed these medications if selection bias—more complex patients seen later in the day—were a big issue. Neprash and Barnett didn’t find any evidence of that, which is reassuring for their overall conclusions.

Ultimately, many of the studies we’ve visited in Random Acts of Medicine rely on similar assumptions in order to establish cause and effect. Disagreement on the validity of such assumptions is only normal—what is a reasonable assumption is often a matter of opinion. One rule of thumb we follow in our own work and in reading others’ is that while we can’t prove an assumption with certainty, the more evidence we can find supporting that assumption, the better.

Leave a comment

Subscribe now

Share

Buy the book: RANDOM ACTS OF MEDICINE

In his upcoming book, Naomi Osaka: Her Journey to Finding Her Power and Her Voice, journalist and friend of this newsletter Ben Rothenberg illuminates the largely untold story of the early life and rapid ascent of four-time Grand Slam singles tennis champion and one-time #1 ranked women’s tennis player, Naomi Osaka.

In the book (which comes out next week in advance of Osaka’s much anticipated return to the Australian Open after a hiatus to give birth to her daughter) we learn that from a very young age, she was driven by a competitive tennis rivalry with her older sister, Mari.

If that sounds familiar, it should: Naomi Osaka’s greatest role model growing up was (and remains) Serena Williams, who also competed with her older sister, Venus, from a young age before becoming the legend she is today. Children who happen to be born as the younger sibling besting their older sibling in sports, of course, is nothing new. Rothenberg writes:

Many of the world’s greatest athletes started as the second best in their family. Having an older brother to challenge himself against, Michael Jordan said, gave him the fire and ferocity that became his signature on the court. “I don’t think, from a competitive standpoint, I would be here without the confrontations with my brother,” Michael said on The Last Dance. “When you come to blows with someone you absolutely love, that’s igniting every fire within you. I always felt like I was fighting Larry for my father’s attention… When you’re going through it, it’s traumatic. Because I want that approval, I want that type of confidence. So my determination got even greater to be as good—if not better than—my brother.”

Naomi said many of the same things in a 2021 Nike video she filmed alongside Mari. “My sister has been a driving force for me,” Naomi said. “The way my personality is, is probably a lot to do with her, because when I was little she would beat me every day.”

“Yep!” Mari chimed in, stopping Naomi mid-sentence.

“You just had to say something,” Naomi said, glaring with sisterly annoyance until Mari giggled.

“I feel like [losing to Mari] shaped my competitiveness a lot,” Naomi continued. “It just made me really want to win all my matches. That is a job that an older sibling doesn’t get enough credit for. So, definitely, I’m always really grateful for her, even though she’s annoying most of the time.”

Birth order is essentially random; there’s no reason one combination of parental genetic material would be systematically more likely to combine for a first child versus a second or third. Once born, of course, siblings are raised in different environments, with the most notable difference being that a younger sibling has an older sibling around to learn from, play with, and perhaps compete with. Taken as a group, what happens to first-borns, on average, is what we would have expected to happen to the second-borns had they happened to be born first, and vice versa.

Birth order in baseball

While a few examples of sibling rivalries in sports might come to mind, rigorously and systematically studying sibling athletes isn’t easy. But one study by researchers Frank Sulloway and Richard Zweigenhaft took advantage of Major League Baseball’s extensive record keeping to do an analysis of 700 total players, all of whom had brothers who also played in the MLB (the study was highlighted in The Best: How Elite Athletes Are Made by Tim Wigmore and Mark Williams).

After accounting for when they were called up from the minor leagues to the major leagues—where one less-talented brother might make it into the major leagues in part off the reputation of their more-talented brother—Sulloway and Zweigenhaft found that younger brothers outperformed older brothers on a few metrics: they were better batters overall, were more likely to hit home runs, and more likely to take risks and (successfully) steal bases.

Younger brothers also tended to have longer careers by about 2 years. When looking at brothers who played the same position, they found that when players were close in age (less than 5 years apart), younger siblings tended to have longer careers than when they were farther apart in age, suggesting the more highly competitive atmosphere of a close sibling rivalry could really make a difference in a baseball career.

If having siblings around is enough to shape the career of a professional athlete, is it enough to impact health?

Subscribe now

Birth order effects on health

Studying the effect of birth order on health is challenging, since any differences might take decades to manifest, and collecting health data across large numbers of families is no easy feat.

In Sweden, however, such research is a bit easier. The country has been collecting health and genealogical data on its population for generations; their registries have been referred to as “a goldmine for epidemiological research.”

One reason such data are so valuable for studying siblings is that researchers can compare them while being able to easily account for the characteristics those siblings have in common by nature of being raised in the same household—things like a family’s income, geographic location, or parent’s education level that might impact health outcomes—by doing “within-family” comparisons across many families.

In a study looking at Swedish families with more than one child born to the same two parents, researchers Kieron Barclay and Martin Kolk found that among about 1.8 million siblings born between 1938 and 1960, higher birth order (meaning, being a younger sibling) was associated with higher mortality rates in the ensuing decades, more pronounced for third-and-later-born children and women.

Looking at causes of death, increasing birth order in men was associated with higher odds of death by accidents/trauma/suicide. In women, increasing birth order was associated with higher odds of death for those same causes as well as for cancer. Meanwhile, there were insignificant reductions in odds of death of cardiovascular disease with increasing birth order.

In another study using similar data from Sweden, researchers Evelina Björkegren and Helena Svaleryd looked at millions of Swedes across the entire lifespan. They found that relative to first born children, second- and later-born children were less likely to be hospitalized at a young age, but by the time they reached adolescence, they became more likely to be hospitalized—including at older ages, in keeping with the above mortality results.

When it came to specific causes for hospitalization, younger siblings seemed to always be at increased risk for injury visits, but it was in adolescence when disparate rates of mental illness took off for non-first-born children—specifically for substance use disorders. Meanwhile, first-borns were more likely to receive outpatient treatment for ADHD, depression, and anxiety than their later born siblings.

Subscribe now

Why?

We’ve only looked at some small slivers of this question of the effects of birth order over the lifespan, and only a small bite of data from baseball and Sweden.

But outside of health, a growing body of research shows birth order effects on educational achievement and future earnings, with advantages for first-borns. Within health, smaller studies suggest differences in all sorts of diseases. Many of these differences have a number of different possible explanations, ranging from differences in the womb environment to diminishing parental and educational resources for younger siblings as more children enter the household. In all likelihood, differences between siblings are going to be driven by a combination of factors.

To bring this all together, perhaps whatever sets up younger siblings for worse health outcomes are some of the same factors that predispose them to be able to best their older siblings in sports. The original purpose of the baseball study above was actually to look at risk-taking behaviors in sports, also finding that younger siblings were more likely to participate in higher risk sports (e.g. football, rugby, boxing, skiing, scuba diving) than their first-born siblings—stealing bases and slugging for home runs in baseball was identified as part of this risk-taking behavior. That same risk-taking behavior might explain some of the differences in hospitalizations for injuries and substance use seen in the Swedish data.

We won’t pretend to know the answers here, and as parents of two kids for each of us, it’s humbling to think about what the differences in our kids’ upbringing might mean for them later on. Naomi Osaka, Serena Williams, and Michael Jordan made being a younger sibling work to their advantage—at least professionally—but not every younger sibling will be able to do that.

To all of you—many of whom are presumably siblings yourselves—what do you think of all this? Are there any other high quality data that might confidently answer some these questions? Or, what other ideas come to mind? Let us know in the comments below!

Leave a comment

Subscribe now

Share

Buy the book: RANDOM ACTS OF MEDICINE

A reader sent us a recent article in the New York Times UpShot—an interesting section of the newspaper that dedicates itself to a mashup of data analysis and news—that showed a concerning trend: after nearly three decades in decline, the nationwide rate of pedestrian deaths in the U.S. has been steadily increasing since about 2009.

What is happening?

After breaking down the data from the U.S. Department of Transportation, authors Emily Badger, Ben Blatt, and Josh Katz noted two important points:

1. The rise in pedestrian fatalities was almost entirely driven by accidents occurring at night

2. When comparing similar data from other countries, there was no such rise observed in Canada, Britain, France, or Australia.

The data suggest that there is something placing pedestrians at higher risk that is unique to both nighttime and the United States. Some ideas come to mind right off the bat, which the authors at Upshot looked into:

• Late hour, or darkness? The authors defined nighttime as 7pm to 7am, but in different parts of the country at different times of year, that time window doesn’t necessarily mean that it is dark out. To confirm that this was indeed related to darkness, they looked at what specific times of day were seeing the most pedestrian fatalities throughout the year. They found little spikes during morning commuting times during fall and winter, when days are shorter and morning commutes occur in the dark. But the biggest spikes were seen in the evening, with times varying by month based on the time the sun set; the spikes occurred latest in the summer and earliest in the winter, with sudden jumps occurring after daylight savings changes. In sum, the fatalities tracked precisely to darkness.

• Who is getting hit? It’s easy to think that those at greatest risk might be vulnerable groups like children or the elderly. Children may not have the judgment or experience necessary to walk on roads safely at night, and they are smaller and harder to see by motorists. Elderly pedestrians may be at higher risk because they are more likely to suffer from hearing loss, vision loss, or cognitive impairments that make it more dangerous to be a pedestrian. Either way, one would expect those factors to roughly similar over time, which wouldn’t explain the reversing trend in nighttime pedestrian injuries starting around 2009. Regardless of that principle, it turns out that child pedestrian fatalities actually declined over that period, and there was only a small increase among those over 65. Those aged 18-64, however, saw the biggest increase in fatality rate. Why? We don’t know.

• Where is this happening? The highest rates of pedestrian fatalities occurred in the Sun Belt states, but this has been a longstanding trend. The authors mention other research that suggests that pedestrian fatalities in downtown areas are on the decline, while those in suburban areas are on the rise.

  Subscribe now

Why is this happening?

The data alone don’t tell us exactly why this is happening, but there are a number of possible explanations that the authors put forward, none of which seems to be a slam dunk on its own, but taken together, might explain the trend:

• Smartphones? The meteoric rise of the smartphone has taken place over this same time period, which can serve to distract drivers and pedestrians alike. Cars now have what seems to be an ever growing set of internal displays that could be distracting like a smart phone, despite the built-in safety mechanisms that disable some features while in motion. But our phones are distracting both during the day and at night, and smartphone use also has dramatically increased in the comparator countries. One difference between the U.S. and European countries is the less frequent use of manual transmissions—which force drivers to put down their phone to drive. But automatic transmissions are still popular in Canada, which hasn’t seen a rise in nighttime fatalities. (One testable implication of this theory would be to look at changes over time in driving accidents involving automatic versus manual transmission vehicles, under the idea that if growing cell phone use in cars is a cause of the overall observed increase in pedestrian injuries then one would expect sharper increases among automatic transmission vehicles compared with manual transmissions.)

• More dangerous vehicles? Cars have gotten bigger, taller, and heavier over this time period, which makes it more likely that injuries to a pedestrian struck by the vehicle will be fatal. It also means that a driver may not be able to slow down as quickly when seeing a pedestrian in the roadway. The car experts interviewed, however, said that car sizes were getting bigger before 2009. What’s more, deaths from smaller vehicles like sedans have also increased. In sum, bigger cars probably isn’t the sole answer either.

• More dangerous substance use? In the time since 2009, the U.S. has seen a changing drug landscape. While alcohol continues to play a significant role in pedestrian fatalities—the pedestrian is more often intoxicated than the driver—the use of other substances like opioids and cannabis has also risen over this time period with yet-unknown impacts on driver and pedestrian safety.

• More pedestrians on dangerous roads? The authors posit that broader economic trends may be playing a role. “Nationwide,” they write, “the suburbanization of poverty in the 21st century has meant that more lower-income Americans who rely on shift work or public transit have moved to communities built around the deadliest kinds of roads: those with multiple lanes and higher speed limits but few crosswalks or sidewalks. The rise in pedestrian fatalities has been most pronounced on these arterials, which can combine highway speeds with the cross traffic of more local roads.” Additionally, homeless pedestrians may be living near dangerous roadways like freeways or in darkly lit areas, increasing their risk.

Now, as much as turning to the comment sections of major news articles would earn us an F if we were to submit this essay as a homework assignment, we thought there might be some good ideas in the over 2,500 reader comments:

• Worse headlights? Echoing many other commenters, Ben from Boston wrote: “Newer cars have headlights that at their low levels really blind other drivers and make it very difficult to see darker objects at the periphery of the roadway. This is especially true of SUV headlights, as they are higher up.” The increasing popularity of brighter LED bulbs might impact the vision of drivers traveling the opposite direction, especially if the light beam is not aimed correctly (such as when installing an after-market bulb), making it harder to see a pedestrian in the road. While headlight technology has also advanced in the comparison countries, in Europe, adaptive driving beam headlights use cameras and sensors to provide maximal illumination while also preventing the beam from shining into the eyes of an oncoming driver. Adaptive driving beam technology was illegal in the U.S. until last year.

• Worse clothing? A number of commenters noted there might be differences in what pedestrians tend to wear after dark that may be changing over time and be different between countries. If American pedestrians are more likely to dress in dark clothing or less likely to wear something reflective on a dark road, they might be harder for drivers to see. But it doesn’t seem like there was a significant shift in nighttime pedestrian fashion after 2009 that would explain this (if anything, brighter clothing resembling that of the 80s and 90s seems to be gaining in popularity).

• Worse vehicle controls? A number of commenters pointed out that in addition to being a large, visual distraction, touchscreens within newer cars often contain controls to operate the vehicle. Older cars had dedicated buttons and dials that drivers could operate without taking their eyes off the road, such as for climate control or music. Some of these have vanished in favor of more complex touch screens that are less physically intuitive to operate, forcing even a focused driver to take their eyes off the road.

What do you think?

Many of the above explanations would be difficult to study, and there are presumably other contributing factors we haven’t discussed here. We agree with the Upshot authors that this concerning trend is likely due to a combination of factors, but what those factors are and to what extent each one is contributing isn’t clear.

We’d love to hear your thoughts in the comments below. What do you think the primary contributors are? Are there big considerations missing here? Any ideas on how they could be studied?

Leave a comment

Subscribe now

Share

Buy the book: RANDOM ACTS OF MEDICINE

A hot cup of tea can really hit the spot if it’s a cold day, if you’re a bit congested from the umpteenth viral infection your children have brought home from school, or if your vocal cords need a boost (as we found out recording our audiobook, reading out loud for hours on end is a lot harder than it sounds!).

As Americans, it’s hard for us to truly comprehend the role tea plays in daily life in England (and the rest of the UK), where tea has been an extremely popular part of daily life for generations and remains so today. The history of tea in England is long and complex, but as a new study by University of Colorado economist Francisca Antman shows, it turns out that tea’s history in England has something to teach us about public health.

Tea in England

Once an imported luxury good enjoyed mostly by the wealthiest English families, tea had become an easily accessible good across social classes by the end of the 18th century. As Antman writes in The Review of Economics and Statistics:

Why then did tea emerge as the English national beverage? One important factor is the prominent role of the English East India Company (EIC), which had a long-running monopoly over trade with the Far East until 1834. Through its dominance in international markets, the EIC was able to bring so much tea into England that it was able to push other beverages, such as coffee, out of the market.

Another cultural feature that helped solidify England as a nation of tea drinkers was the advent of tea houses, where, unlike all-male coffee houses, women could purchase their own tea. This ensured that tea would become a more accessible drink, available to a wider population, and thus solidify its dominance as the country's national beverage. Tea gardens, which could be enjoyed by men, women, and families together, also enshrined tea as a cultural custom, as did the worker's tea break.

Notably, tea is made using boiling water. The act of boiling water helps make it safer to drink by killing disease-causing bacteria, viruses, and parasites, which at the time were a common cause of illness and death. Compared to other beverages that can be a source of cleaner water—like beer, wine, or coffee—tea was less expensive and had fewer side effects, so it could be drank all day long.

With the popularity of tea on the rise in England, it stood to reason that more people would be consuming safer water, because the water in their tea had been boiled. And if indeed they were drinking safer water, did that translate into reduced rates of death from water-borne illness?

Subscribe now

Steeping in the data

From 1761 to 1834, the English increased their tea imports from China from about 1 pound to 3 pounds of tea per person per year. Over that same time period, death rates in England fell. Of course, tea consumption wasn’t the only change that occurred over that time period that might lead people to live longer—the Industrial Revolution led to many changes in daily life that could also have reasonably impacted mortality in England.

To tease apart any specific contribution of tea drinking to improved mortality, Antman needed to find a natural experiment. If there was an occasion where tea drinking increased suddenly relative to the other gradual changes during the Industrial Revolution, changes in mortality after that event that were out of proportion to what we would have expected from gradual progress could reasonably be attributed to the tea/boiled water.

The year 1784 produced one such event: this was the year import taxes on tea were significantly reduced, from 112% down to just 12.5%. Suddenly, tea was a lot cheaper to import and purchase. Below, we see the relatively sudden increase in tea imports after 1784. Mortality also improves after that year, and it appears to improve faster than it was improving before 1784—but from these data alone, it’s still hard to know if that was because of hot water use with tea, or something else.

Next, Antman figured out which areas of England would have had higher or lower quality water sources, based on geography (e.g. more running water nearby suggests higher quality drinking water), around the time tea’s popularity increased. If tea consumption led to people drinking safer water, then areas with lower quality water to begin with should benefit more from drinking hot tea than areas that already had cleaner water to begin with, right?

In analyses that accounted for changes in the population, she found that prior to the the 1784 drop in the tea tax, mortality rates (deaths per 1,000 people) were similar in areas with higher and lower water quality sources. After the tea tax was lowered, however, the improvements in mortality over the following 50 years were significantly different: in areas with high quality water sources, mortality rates dropped by about 7% over the next 50 years; meanwhile, in areas with low quality water sources, it dropped by about 25%.

Antman also looked at improvements in mortality in areas that imported different amounts of tea, finding that areas with low-quality water sources that imported more tea saw bigger improvements in mortality than when areas with high-quality water sources imported more tea.

Share

So, was it actually the tea/hot water, or something else?

Could there have been something else that happened around 1784 that led to this reduction in mortality? The smallpox vaccine was disseminated around the turn of the 19th century, but Antman did several analyses suggesting that this didn’t explain the changes.

Could it be that areas with low-quality water sources simply benefitted more from the health changes brought on by the Industrial Revolution than people with high-quality water sources? In analyses that accounted for the relationship between changing income—where we might expect to see higher wages in areas thriving from industry—and mortality rates, Antman found no evidence to suggest that increasing incomes explained the observed differences between areas with high- versus low-quality water sources.

So while it’s possible that there was some other change that disproportionately improved health in areas with low-quality water sources, the evidence is all consistent with tea being responsible for improvements in health. Nowadays, of course, the quality of drinking water is no longer a major issue in England—but the traditional “cuppa” lives on.

Leave a comment

Subscribe now

Share

Buy the book: RANDOM ACTS OF MEDICINE

Patients who live at home but require around-the-clock care tend to have a primary caregiver and perhaps some others who look after them. Caregivers are often family members, but they can also include home health aides, visiting nurses and therapists, and others who make it possible to live at home instead of, say, in a nursing home. There are many reasons a patient might require 24/7 care, and the care needs can be quite variable.

Providing such care is a lot of work for caregivers, many of whom gain expertise in managing complex medical conditions. It can be physically and emotionally demanding, stressful, expensive, and socially isolating. Without a doubt, caregivers need and deserve breaks, including vacations—but this can be easier said than done. To find respite care for the patient, caregivers in need of a break can reach out to family and friends, but they may not be able to cover all the necessary care. They can seek professional respite care as well, but it’s expensive and costs can add up quickly.

While likely not preferred, caregivers can also get a “break” if a patient ends up in the hospital with, for example, a complication of their chronic issues or a more serious problem like an infection. While patients are under our care in the hospital, it’s quite common for us to see caregivers taking the opportunity to catch up.

In medical slang, a “pop drop” is when a caregiver drops of a patient in the emergency department with a problem of questionable significance, knowing the patient will be taken care of for a stretch of time and giving them a break (“pop” referring to an elderly man). This break could last a few hours while the patient is evaluated in the emergency department, a few days if the patient ends up being admitted to the hospital, or even longer if the hospital care teams aren’t able to ensure that the patient is able to return home to a safe environment. If, for example, an essential home caregiver dropped off a patient without a serious medical problem at the emergency department on the Friday before a holiday weekend and went out of town, the hospital would have to care for the patient until they could send them home the following Tuesday.

We understand why a caregiver might want—or need—to take advantage of the care available at the hospital. For example, a 2017 study showed that when primary caregivers themselves were experiencing health problems, the patients they cared for used the emergency department more often.

But it did get us wondering: is the pop drop just a one-off story from a Friday evening shift in the ED, or is it a pervasive phenomenon?

Subscribe now

Taking a look at the data

While it’s not hard to find stories of pop drops, identifying them in the data isn’t so straightforward. If we’re looking at insurance claims data, it’s impossible for us to know the motivations for why any individual patient presented for medical care. But if we look at days where we would expect a pop drop to be more likely, we might find evidence of a significant phenomenon.

In a collaboration with Swad Komanduri at RAND and Christopher Whaley at Brown, we started by taking a look at patients with Medicare in an analysis we’ve only shared here.

Specifically, we looked at patients who were admitted to the hospital at least twice in a year—a group of patients we figured could have complex enough medical problems that they could reasonably require care at home from caregivers who need a break. We also figured that a pop drop would be more likely to occur over a holiday weekend with the Monday off than a standard weekend, since caregivers may go out of town for the holiday (i.e., we might see greater-than-typical hospitalizations on the Friday, Saturday, or Sunday before the holiday Monday, i.e., the days of or leading up to the three-day weekend). Here’s what we found when we looked at the number of hospital admissions on any given day of the week:

We saw a few notable things here:

• Not surprisingly, there were fewer hospital admissions on holiday Mondays (in gray, above).

• In the days after that holiday Monday, there were more admissions than usual. This makes sense, as these days are likely compensating for the reduced admissions on the holiday Monday.

• There were some slight differences on the Friday, Saturday, and Sunday before a Monday holiday (in orange, above), but they weren’t dramatic and we certainly didn’t see clear evidence that hospitalization rates were higher on, for example, the Friday going into a three-day weekend. In fact, we were a bit surprised to see a small dip on the Friday, where we would have expected an increase, but perhaps these patients weren’t coming until the Saturday.

We didn’t feel we could make any definitive conclusions about pop drops from this analysis, so we went a little further. The next thing we did was limit our analysis to patients with Alzheimer’s dementia—a group more likely to have regular care at home. We then decided to look at some reasons why they were admitted to the hospital. One of these reasons is “dementia,” which is a chronic issue and is typically not the listed reason for hospital admission—unless there isn’t another more significant problem, which is what we might expect for a pop drop.

Here’s that additional analysis. Here, the colors are different and represent different reasons for hospital admission. The bars show the number of admissions on each day of the week in the week before a Monday holiday relative to a typical week for each condition:

Here, we saw a few more notable things:

• For admissions with dementia as a primary diagnosis—suggesting a more pressing acute problem was not present—we saw a similar pattern as before. A very slight dip on Friday and a slight increase on Saturday before a holiday, but not an impressive pattern (certainly, no evidence of higher hospitalization rates on Friday going into the three-day weekend).

• For GI hemorrhage (dark blue) and heart attacks (AMI in gray), there weren’t any differences in admissions before a holiday Monday.

• Hip fractures and joint dislocation admissions had a curious pattern leading up to and during the holiday weekend. And while we weren’t sure why this pattern existed, it didn’t seem like the pop drop phenomenon would explain it.

  Share

So are pop drops a real problem?

It’s hard to come to any definitive conclusions from this quick look at the data. But we didn’t see evidence of a large impact on holiday weekends, at least in the group of Medicare patients with dementia—a group that’s more likely to have higher demands on caregivers.

But because we were comparing holiday Fridays to typical Fridays, what we couldn’t see from this analysis is whether the pop drop phenomenon is a significant contributor to admissions every Friday; they may be. We also looked only at hospital admissions, since we were most interested in pop drops for a longer duration of respite. If we looked at emergency department visits where people were sent home after a few hours, it might be a different story.

So despite the anecdotes, we still don’t have a great sense of how common it is for caregivers to bring patients to the hospital with minor issues for un-official respite care. While we as medical providers may believe a caregiver is bringing a patient in with a minor complaint just for a break, we shouldn’t assume that there isn’t a serious problem underlying a subtle change. Often times in these complex patients that require around the clock care, a seemingly “minor” symptom that prompted someone who knows the patient well to bring the patient in was the only clue to a serious underlying problem.

But it’s also important to remember that even if pop drops are happening and may be perceived as an abuse of the health care system, caregivers need and deserve a break. Many family caregivers are balancing other work and other family obligations in addition to caring for someone 24/7/365. Official respite services aren’t always available, are often unaffordable, and can’t always meet a patient’s needs—so it’s understandable that some caregivers have to resort to the pop drop.

Subscribe now

Share

Leave a comment

Buy the book: RANDOM ACTS OF MEDICINE

Share

Subscribe now

Buy the book: RANDOM ACTS OF MEDICINE