So, if you’re reading this you most likely know that I’m a first-year medical school student in the US. And a month or so ago, we learned a common physical diagnosis technique—a funky one called “tactile fremitus.” Now, this isn’t one I’ve actually had done to me by a doctor, and it’s probably not one you’ve had done to you by a doctor, but it’s commonly taught in medical schools anyway.

Before we dig into the interesting stuff, here are the basics of tactile fremitus. It’s used as part of a pulmonary exam to see if there is air or fluid in your lungs. The doctor places the blades of their hands along your chest wall and asks you to repeat a phrase. As you speak, vibrations travel from your vocal cords through the lung tissue to the chest wall, where they can then be felt by a doctor’s hands. If you’ve got excessive air in the lungs, vibrations are usually depressed. Conversely, if you’ve got fluid in there (like with pneumonia), vibrations are usually increased. In a normal person, the “vocal fremitus” should be felt equally on both sides.

Anyhow, when we learned tactile fremitus, the phrase we were told patients should repeat is “ninety-nine.” So, picture in your head one person sitting down, back partially exposed, and another person in a white coat standing behind them, hands on their back, asking them to repeat “ninety-nine” several times. That’s what a tactile fremitus test looks like. Fun, right?

Naturally, I wondered, why “ninety-nine”? Why not, you know “one hundred”? Or something else entirely?

It turns out that I was not the only one wondering such a thing. In 1973, Dr. William Dock published a study in the Bulletin of the NY Academy of Medicine where he rigorously tested the use of ninety-nine. First, you need to understand that the ideal sounds for this test are low frequency ones, as they transmit more effectively through the lung tissue and onto a doctor’s hands. In English, diphthongs (two vowel sounds smushed together) will do the job. So, Dr. Dock strapped mics to volunteers’ chests and recorded them saying “ninety-nine” as well as a few other choice phrases—“boy boy” and “boogie woogie” (it was the seventies, I guess). It turns out that “ninety-nine” was not a good word to use for tactile fremitus at all. 

Here’s Dr. Dock’s graph. A represents “boogie woogie,” B represents “boy boy,” and C represents “ninety-nine.” The top lines are sounds above 240 Hz, the middle lines sounds between 80 and 240 Hz, and the bottom lines sounds 20-80 Hz. (C only shows the top and bottom lines). As you can see, “boogie woogie” and “boy boy” produce far more low frequency sounds compared with “ninety-nine.”

Why is everyone taught “ninety-nine” then? According to Dr. Dock, when physicians from the US were visiting Austria and Germany, they saw tactile fremitus being used with the phrase neun und neunzig—ninety-nine in German. That worked for them, since ninety-nine in German had a deep, low-frequency-producing “oy” sound. But in English, no dice. A classic case of direct, literal translation that didn’t have the intended practical use. And apparently, no one bothered to question it. Forty-eight years later from Dr. Dock’s revelation, we’re still being taught the bum number rather than the much more juicy “boogie woogie.”

In 2016, Anna Harris, an Associate Professor at Maastricht University in the Netherlands, addressed this lag in an article for the Journal of Graduate Medical Education. Harris mentioned that in the Netherlands (where students are taught mostly in Dutch), some schools hopped eleven digits down to eighty-eight, or achtentachtig, to get the adequate tremors. Even then though, certain accents made the word too soft, and so Amsterdam was suggested. Why do we keep using ninety-nine? According to Harris, “because 99 is a habit, ingrained in the ritual of physical examination”—and rituals are important.

It’s not just tactile fremitus and ninety-nine. Digging into Dr. Dock’s article further made me think about the other ways we use sound in medicine, a focus of Prof. Harris’ own research. For instance, how do we communicate what certain sounds sound like? Such questions are not only intellectually intriguing, but also medically important. According to Dr. Dock, physicians used all sorts of methods to get across the sounds emanating from their patient’s insides, often heard via stethoscope (aka auscultation). “Sounds of sea gulls crying, cranes whooping, geese honking, and horses clattering down dirt roads were once fine analogues for certain cardiac murmurs or sonic rhythms,” Dock wrote. In fact, Dock reports, some docs even mimicked heart sounds by using the hems of their white coats as a kind of guitar string, holding it at different lengths and plucking it to simulate clicks and taps from valves and vessels.

We all know the heart goes “lub dub,” right? According to Dr. Dock, wrong. Unsatisfied with tearing apart just tactile fremitus, he compared recordings of people saying the words “lub dub” with actual heart sounds. Turns out, “lub dub” doesn’t do a good job either. A better fit might be “click click.”

Of course, the evidence for all this is weak, at best. Dr. Dock doesn’t seem to have repeated his tests all that many times, and his method of strapping a mic to a person’s chest might not be the most accurate. But his ragtag endeavors at questioning the status quo are inspiring. What else could physicians be listening for, and how could they do it better? As I’ve seen in just a few months of medical school, sound is essential to understanding and examining the human body. From lung sounds to heart sounds to bowel sounds, we make all sorts of noises—and subtle variations in those noises could mean the difference between healthy and sick.

And how do we teach the art of listening to students? And how do we, as students, learn? The first heart sound I truly listened to was that of my instructor—who happened to have a heart murmur (she was born with it, and she was fine). As I placed my stethoscope over her chest and focused closely, I heard two thumps, with a whoosh in between. Was what I heard the right thing? How could I describe what I was hearing, and compare it with what I was supposed to have heard? I had never really considered how important it was to have the right language to convey a deeply felt sensation like sound.

I’ll get off my soapbox now, but this is a topic I want to keep digging into. I’ve read a lot about the “death of the physical exam” and how telemedicine might be changing doctor’s visits, and I wonder how that will change the use of senses like sound. Professor Anna Harris, of Maastricht, seems to be doing a lot of work on this—and I mention this because I reached out to her for an interview for this piece. She was understandably busy, but if I ever get a hold of her (Dr. Harris, I would still love to speak with you, even briefly!) or one of her collaborators, you can be sure I’ll have a follow-up post with more on this. Till then, I’ll leave you with a wonderful couplet Dr. Dock himself composed:

            Lub dub does not the heart sounds fake

            Nor ninety-nine the thorax shake.

You’re reading Part V (the last part) in a 5-part series about sewage science. Catch up on Parts I, II, III, and IV before diving into this one.

In 2012, Christian Daughton proposed a new concept for wastewater analysis. Rather than analyzing sewage for the breakdown products of foreign substances we encounter or put into our bodies, like drugs or pesticides, researchers could instead seek out endogenous biomarkers – substances produced by the body in response to conditions of health and disease.

Endogenous biomarkers are powerful because they tell you exactly what you want to know. Instead of treating opioid use as a proxy for pain, for instance, you could directly capture the extent of pain by measuring a marker produced during periods of injury or stress. Endogenous biomarkers are also integrative, capturing exposure to various forms of stress rather than a single drug or contaminant.

One example of this is a class of biomarkers known as isoprostanes. Produced by the body in response to oxidative stress and inflammation, they are a widely accepted indicator for conditions such as diabetes, heart disease, and obesity. Yet isoprostane levels can also reflect chronic exposure to social or psychological stressors.

Like much of the rest of WBE, research on isoprostanes and other endogenous biomarkers is still in its early stages. There are also unique challenges. Many biomarkers like isoprostanes are produced at some level in healthy individuals. Thus, it’s important to establish a “reference range” of normal values seen throughout an entire population, already a complicated task for individual patients. Plus, finding a robust biomarker that points to a specific illness or condition is difficult – there is still debate about whether some well-established biomarkers, like cholesterol, have predictive value for the development of disease at all. If these questions can be addressed though, analyzing wastewater for endogenous biomarkers might come the closest to studying a community as an integral “patient”—one that is complex and constantly changing.

Wastewater-based epidemiology is highly interdisciplinary, traditionally operating at the intersection of environmental chemistry and drug epidemiology. The growing interest in analyzing wastewater for SARS-CoV-2 has brought in a third field to the mix: virology and microbiology. “We’re a small bunch, but everyone brings something different to the table,” said Bowes.

The small bunch may soon be getting bigger. “I think Covid has really gotten wastewater into media attention,” Dan Burgard said.  Researchers hope that increased coverage will lead to greater investment and interest in WBE in the future. “We’re seeing a blast of wastewater analysis groups pop up all over now, because people are starting to understand the utility and how useful this can be,” said Devin Bowes.

From drug monitoring to virus testing to emerging uses for studying nutrition and daily stress, wastewater analysis paves the way towards an ideal that Daughton called a “biological passport,” a snapshot of not just the physical, but social, psychological, and economic health of a community. Amid a pandemic, data from wastewater analysis can provide peace of mind for people returning to school and work, empowering communities with data on COVID-19 spread in their own neighborhoods. The coronavirus has without a doubt been devastating for the nation. But its boost for wastewater-based epidemiology may help us confront health crises in the future.

Especially in the U.S., it has been tough to convince people of the value of sewage surveillance, Bowes said. “With Covid, more people are starting to see the value in wastewater testing to understand human behavior and health.”

“In the future, maybe it won’t be so much of an uphill battle.”

You’re reading Part IV in a 5-part series about sewage science. Catch up on Parts I, II, and III before diving into this one.

Many recent applications of wastewater-based epidemiology—including for monitoring nutrition, microbiota, and COVID-19—are still in their infancy. Scientists have yet to solve several key issues. How well does amount of food consumed correlate with the amount of biomarker excreted? What is the role of bacterial species in the microbiome of a single person? How much does the amount of SARS-CoV-2 shed in feces vary from person to person? Add onto these questions ever-present uncertainties about sewer conditions like temperature, flow rate, and the presence of other chemicals or microbes, and the rather simple concept of WBE complicates quickly.

A common problem plaguing WBE scientists might seem at first to be the easiest to solve: that of estimating population size.

Whether measuring the consumption of illicit drugs or leafy green vegetables, the results only have meaning in relation to the number of people who contributed to the sewage in the sample. But the straightforward solution of using population estimates from census data is often not the best—such estimates are frequently inaccurate and don’t account for commuters, tourists, or other day-to-day fluctuations.

“If you have a place that has a high influx of commuters, say Manhattan, that population is really different in the morning of a Monday than it is on a Saturday morning,” Dan Burgard noted. Other methods of estimating population size, like using mobile phone data, can be prohibitively expensive. But there is an alternative. “One of the things we’re working on right now is to try to use compounds in wastewater itself to help us understand how many people are in that actual population,” Burgard told me.

What Burgard means is finding a marker in sewage that can serve as a proxy for population size—preferably one that each person excretes at a similar rate and that remains relatively constant over time. In practice, these are markers of things we consume regularly: artificial sweeteners, nicotine, caffeine.

Yet, the major roadblocks to further advancing wastewater analysis might not be scientific at all, but societal.

Concerns about ethics and privacy have been around for as long as people have been analyzing sewage. In most cases, wastewater in a treatment plant comes from hundreds of thousands of flushes—that is, any data gathered is almost guaranteed to be anonymized. However, when surveying smaller populations, like a school, workplace, or prison, there are legitimate concerns that data could be used to target vulnerable groups. Phil Choi emphasized that WBE researchers adhere to a set of ethical principles to ensure that data is not used to punish or stigmatize a community. But scientists still need to address such concerns if wastewater analysis is going to be used on a larger scale.

Some cities may be reluctant to have their sewage surveilled for fear that a negative outcome— like finding high rates of illicit drug use—will tarnish their reputations. A few WBE researchers speculated that reluctance to air the “dirty laundry” of cities and communities prevents the widespread use of wastewater analysis—particularly in the US. Regardless of the reason, many WBE scientists feel frustrated by the relative lack of investment in wastewater analysis in the States, compared with well-developed programs in places like Europe and Australia. 

The last and only time a government-sponsored effort took place in the U.S. was in 2006, when the Office of National Drug Control Policy began measuring cocaine metabolites in dozens of cities. But the results were never published. 

“I’ve been baffled as to why it has not been embraced more wholeheartedly,” Christian Daughton said. “We have something that could be an extremely valuable tool, and we’re essentially just ignoring it.” 

In the United States, wastewater analysis has typically operated on small, local scales. Some cities partner with nearby universities to analyze their sewage; others work with Biobot, which has contracts with hundreds of cities across the country. But thus far, there has been no national framework for WBE.

“Nothing has been really systemically coordinated at a national level, at least not to my knowledge,” Ryan Newton said. Although WBE is far cheaper than other survey methods, especially when it comes to COVID-19 testing, it still requires significant cost, labor, and resources. Coordinating local, state, and national efforts would make it much easier to share data and best practices—“in a much easier way than an individual academic lab that is just starting from scratch,” Newton said. Only with this investment can there be real progress on the cutting-edge ideas that could catapult the burgeoning field of wastewater-based epidemiology into the mainstream.  

You’re reading Part III in a 5-part series about sewage science. Catch up on Parts I and II before diving into this one.

Devin Bowes, at ASU, was eventually persuaded to change course, signing on as a graduate student in Halden’s lab. Now, she focuses on using wastewater to study something else that comes into—and out of—our bodies: food.

“We know that there’s a connection between what we eat and human health outcomes,” Bowes said. And “there is a connection between what we eat and how we feel.” As a trained nutritionist, Bowes was familiar with the current methods of gathering data on what we eat, like patient food logs or diaries. But those, she says, are subject to biases and omissions. People don’t—or won’t—write down everything. Instead, Bowes and other researchers are proposing to analyze wastewater for the breakdown products of foods, called biomarkers, that can give an objective view of what people are eating.

Bowes gives the example of citrus fruits. Several studies have shown that increased consumption of citrus is correlated with lower body mass index (BMI), lower blood pressure, and better overall diet. Let’s say you want to know how many lemons, oranges, or other citrus fruits people in a community are consuming. Instead of handing out a survey to everyone in town, you could instead measure the amount of a biomarker called proline betaine, a metabolite produced by the body after eating citrus, in wastewater. By doing so, you could tell which communities eat more citrus, and therefore may have healthier diets, and which communities eat less citrus, and therefore may be more at risk for chronic health conditions. Such data allows public health officials to target limited resources to the neediest populations.

Nutrition is still a relatively new application of WBE, and much of the research remains in its early stages. Last year, Bowes coauthored a study with Phil Choi at the University of Queensland that identified a few candidate biomarkers, including those that indicated consumption of whole grains and fiber (enterodiol and enterolactone), meat (1-methylhistidine), cruciferous vegetables (allylisothiocynates), and soy foods (genistein and daidzein). More work is needed to translate these biomarkers from theory to practice.

But once these biomarkers are measured in sewage, they could help to pinpoint social, demographic, and economic disparities. This stems from the perhaps not-so-surprising relationship between socioeconomic status and food consumption, and thus, biomarker levels in wastewater. In a 2019 study, Choi found that consumption of vitamins, citrus, and fiber were associated with indicators of higher socioeconomic status, supporting a known and disturbing trend—of disadvantaged and minoritized groups lacking access to healthy foods. Wastewater analysis can help rectify this disparity by identifying what communities are receiving the appropriate nutrition they need to be well, and which are not, Bowes said. Choi and Bowes hope that their findings can translate to direct interventions, such as opening more grocery stores that serve fresh fruits and vegetables in “food deserts” without places to buy healthy food. These interventions are critical for combating nutrition-related chronic diseases, like Type II diabetes.

“I do think that we need to pay more attention to what people are eating and understand that it is affecting our health,” Bowes said. “That can set us up for success when diseases like COVID-19 arise”—when individuals with underlying conditions like diabetes are at higher risk.

Yet many aspects of nutrition lack unique, identifying biomarkers that can’t be tracked. Choi points out that there is no way to measure how much fast food someone eats from their waste. But there may be another way to study overall dietary patterns: by zeroing in on the microbiome. 

The microbiome consists of the trillions of bacteria that live in and on our bodies, outnumbering our own cells by 10 to 1. Friendly bacteria in our guts help us absorb nutrients from food, but imbalances in the microbial community can lead to disease and inflammation. The unique species composition of our microbiomes plays a huge role in our health; a well-known study in mice found that when the microbiome of a lean animal was transplanted to an obese one, the obese mouse slimmed down—and vice versa.

Scientists have focused on understanding the microbiome of individual people. But what if we could use sewage to capture the microbiome of an entire population? 

This was the impetus for a study coauthored by Ryan Newton and Sandra McLellan, researchers at the University of Wisconsin’s School of Freshwater Sciences. In the study, published in 2015, Newton and McLellan profiled the bacterial communities present in sewage.

To identify what bacterial species were present, they extracted a type of genetic material from over 200 sewage samples taken in 71 U.S. cities. Using data gathered from the Human Microbiome Project, they were able to divide bacteria found in the sewage into human and non-human sources. 

Even though only 10% of the bacteria were determined to be of human origin (with the rest coming from animals, the environment, and other places), the effort still yielded a remarkable finding. Researchers discovered they could predict if a sample came from a city with a higher or lower obesity rate with more than 80% accuracy, supporting the idea that the bacterial profiles of “lean” and “obese” populations were indeed distinct. Furthermore, sewage from “obese” cities had more of a subspecies of bacteria associated with a diet high in animal fat and a microbiome lower in diversity—what the authors called a pro-inflammatory gut type.

McLellan and Newton acknowledged that since the research was published five years ago, the science of microbiomes has gotten more complex than a simple “lean” or “obese” distinction. But the study proved a key concept: that bacteria in sewers can tell us about the collective “gut” of a community, and consequently, collective health.

“I think the idea is certainly something a lot of people are grabbing a hold of now,” Newton told me. “You can use microbiome data in sewage to look at various population-level dynamics in humans.”

You’re reading Part II in a 5-part series about sewage science. Catch up on Part I before diving into this one.

Scientists have long monitored sewage for harmful contaminants with the goal of improving wastewater treatment practices. But no one had ever proposed studying sewage to better understand the people that contribute to it.

That is, not until former EPA scientist Christian Daughton. In the 1990s, the EPA focused solely on identifying “priority pollutants” like pesticides, heavy metals, and solvents in wastewater. But other substances—like drugs, both legal and illegal—could have environmental impacts, too, if they leaked from sewage. In 1999, Daughton coauthored a paper that considered how pharmaceuticals in sewage could be contaminants in the environment. Soon after, he published a book on the topic—and offered a ground-breaking proposal. Rather than studying how pharmaceuticals in sewage impacted the environment, why not see what they can tell us about community-wide drug use?

“That was the very beginning of this idea of monitoring things in sewage, not as contaminants, but as markers for human activity, human behavior, human action,” Daughton said.  

Some of his peers doubted the idea. “Several of the scientists involved with the review said, ‘Why are you doing this?’” Daughton recalled. In their view, it had nothing to do with the environment and the mission of the EPA.

But other scientists saw the value of studying sewage as a proxy for human activities. The first demonstration was in 2005, when a team led by Ettore Zuccato measured cocaine in untreated wastewater in Italy. This proof-of-concept experiment opened the floodgates: Since then, researchers have published a flurry of studies using wastewater to track consumption and use of caffeine, alcohol, tobacco, antidepressants, antibiotics, asthma medications, antihistamines, and more.  

Sewage has also been crucial to studying the opioid crisis. When you take a painkiller, your body metabolizes it, breaking it down into different products as it moves through the digestive system. The metabolites, which are in a different form than the original “parent” compound in the drug, then get secreted through feces or urine. When you flush the toilet, the waste travels through networks of pipes to sewage treatment plants.

Fahad Ahmed, a graduate student at the University of Queensland in Brisbane, Australia, analyzes the raw, untreated sewage by capturing the metabolites onto a special type of column and using a device called a mass spectrometer to obtain their concentrations.

Ahmed then tries to estimate what he calls the “pain burden” of a community. But to get from the concentration of a metabolite to the amount of painkiller consumed per day—the way such figures are typically reported—he needs to know a lot more information. For instance, how the painkiller is processed in the body, how much of the metabolite is excreted, how fast the sewage flows, and the size of the “catchment,” or the population that serves a given treatment plant, could all influence calculations. Ahmed makes reasonable estimates to translate the concentration of a metabolite in the sample of sewage into a pain index. Armed with this knowledge, Ahmed hopes that health officials can develop strategies for alleviating the burden of pain while combating opioid addiction. In the U.S., researchers including those at the startup Biobot and the Human Health Observatory at ASU have partnered with cities to use their wastewater data to assess the effectiveness of drug take-back initiatives.

Such wastewater surveillance programs allow governments to measure the effects of drug policy, identify trends in use, and flag harmful new drugs that enter the market. Since 2011, Sewage Analysis Core Group Europe, or SCORE, has tested sewage in cities across the continent for the presence of cocaine, MDMA, amphetamine, and methamphetamine. A similar monitoring program was started in Australia through the Queensland Alliance for Environmental Health Sciences. In Canada, the national statistical agency Statistics Canada started monitoring marijuana use in six cities in April of 2018.

Dan Burgard, professor of chemistry at the University of Puget Sound, has been a member of SCORE since 2015. Burgard became interested in wastewater analysis after his students told him about the prevalence of “study drugs” like Adderall on campus. Wanting to better understand patterns of use and abuse, he collected samples from a manhole capturing the waste flowing in from four dorms. Unsurprisingly, Burgard found a clear increase in the use of study drugs during periods of high stress, like midterms and finals. During one particular time period—finals week of spring semester—the levels of Adderall were eight times higher than during the first week of school, Burgard said.

Author’s note: To kick off the blog, I’m posting a feature I reported last summer under the auspices of my now alma mater. It’s about what I like to call sewage science. I tried to pitch it to some editors, but none of them bit (and no, I’m not at all bitter about it…). I didn’t want the efforts from my sources, and myself, to go to waste, so…here it is. It’ll come in several parts, this being the first.

When Devin Bowes first heard about the idea of using wastewater to study nutrition, she didn’t believe it was possible. She told Rolf Halden, who suggested the idea, as much. “I was like ‘you’re insane. I don’t know what you’re talking about.’” Bowes had arrived at Arizona State University planning to pursue a Masters in Nutritional Science. Then, in the summer of 2018, she met Halden, director of the ASU’s Biodesign Center for Environmental Health Engineering. Halden tried to persuade her of the untapped value of wastewater for understanding what people eat.

Wastewater has been in the news lately for another reason—as a tool for tracking the spread of SARS-CoV-2, the virus that causes COVID-19. From The Netherlands to the Yosemite Valley, researchers have been analyzing sewage for copies of viral genetic material, estimating the total number of COVID-19 cases in real time. Often, these methods detect cases missed by traditional nasal swab testing, including asymptomatic disease.  

Sewage surveillance can also identify surges of infection. That is what happened late in the spring of 2020 in Tempe, AZ, where the ASU Biodesign Institute had been collaborating with the city to test its sewage. The results, reported in an online dashboard, showed the number of COVID-19 gene copies per liter of wastewater—a proxy for the viral load. After Arizona’s stay-at-home order was lifted in mid-May after months of lockdown, Bowes could see the numbers start creeping up, signaling the surge of infections, hospitalizations, and deaths to come.

COVID-19 testing in sewage is relatively new—but it demonstrates the potential for the burgeoning field known as wastewater-based epidemiology (WBE) to transform the way we understand community health. In addition to diseases like COVID-19, scientists have used sewage to study everything from the consumption of illicit drugs to exposure to environmental contaminants. Researchers like Bowes now intend to mine even more insights from wastewater—tapping it to research population-wide nutrition and markers produced by our own bodies that indicate health, illness, and stress.

“The concept on the surface is very simple,” said Christian Daughton, a former Environmental Protection Agency (EPA) scientist and WBE pioneer. You just take sewage and see what is in it. “But when you start talking about it, and thinking about it, it gets more and more complicated.” If researchers can figure it out, though, the long-ignored treasure trove of information that is in our waste can be unlocked – to the benefit of our health.