It's Not Just About Software

Software IS eating the world, but technology is more than SaaS

The Teardown

Monday :: March 15th, 2021

Like this newsletter? Subscribe here

---

What Is “Healthy” Anyways

The pandemic was challenging for our health. Notwithstanding the impact of COVID-19, many lost access to our routines that covered activities such as exercising, eating healthy (e.g., KALE), and importantly — going to the doctor. The latter is that nuisance thing that you have to do, but you sometimes put off. The yearly physical is chronically ignored by some. Telehealth has changed the traditional visit-focused model, but not completely. For example, you can check your blood pressure at home, but you can also let a professional handle it during your physical. There are lots of these minor trade-offs.

On the other hand, a doctor might tell you something that causes an irrational reaction. You're there, and you didn't internalize the risks and probabilities of the thing you just heard. Your response is, in some ways, disconnected from the reality of the information you just received. But doctors are also human and aren't always giving you information rooted in pure and proven science. The proliferation of data on the web has both helped and worsened the problem. We're all doctors now.

I recently went to my first in-person physical in over a year. For obvious reasons, in-person visits weren't possible during much of last year. The various professionals surrounding me drew blood and promised to email the results through the portal. I expected to read that I am a model of supreme health, a god of fitness, perfect in all ways, obviously.

My doctor zeroed in on one blood test metric with this statement:

Your lab results (attached) were notable for a hemoglobin a1c of 6.2, which puts you in the range of pre-diabetes. Hemoglobin a1c represents a rolling 3 month average of your blood sugar. >5.7 is considered prediabetes, and >6.5 is considered diabetes

Um, what? So I am a borderline diabetic? But what about the miles I run, the Peloton I ride, the weights I lift, and the spinach I eat that goes straight to my biceps like Popeye?

I was, to say the least, surprised by this result. My doctor's text is well-structured and logical. He's a doctor. He spent lots of time training to provide me with information that guides my health and alerts me to problems.

But part of my job is to be inherently distrustful of data. I try to triangulate and validate and ensure to some reasonable extent that the data I use to make decisions is credible. Thus, I’ve arrived at the fundamental question: is the data credible, and is the doctor’s statement actually troublesome?

I can’t break it down in an email that is short enough to keep you awake. That said, I started digging into the question.

From The New York Times:

A few years ago, routine lab tests showed that Susan Glickman Weinberg, then a 65-year-old clinical social worker in Los Angeles, had a hemoglobin A1C reading of 5.8 percent, barely above normal.

“This is considered prediabetes,” her internist told her. A1C measures how much sugar has been circulating in the bloodstream over time. If her results reached 6 percent — still below the number that defines diabetes, which is 6.5 — her doctor said he would recommend the widely prescribed drug metformin

Ok, so I’m not alone. Other people receive lab results that tell them they are on the precipice of a serious health problem.

However, at least one study indicates that the Hga1c test isn’t as meaningful as it seems:

“I felt like Patient Zero,” she said. “There were a lot of unknowns.”

Now, there are fewer unknowns. A longitudinal study of older adults, published online this month in the journal JAMA Internal Medicine, provides some answers about the very common in-between condition known as prediabetes.

The researchers found that over several years, older people who were supposedly prediabetic were far more likely to have their blood sugar levels return to normal than to progress to diabetes. And they were no more likely to die during the follow-up period than their peers with normal blood sugar.

“In most older adults, prediabetes probably shouldn’t be a priority,” said Elizabeth Selvin, an epidemiologist at the Johns Hopkins Bloomberg School of Public Health in Baltimore and the senior author on the study.

I might be at peak health, just another average statistic, or in need of a serious revamp of my diet and health profile. A retrospect look at my diet provided some obvious clues: I don't gobble literal blocks of sugar, but actually, I sort of do - sometimes cruising through as many as three or four RX Bars per day. I considered them more of a protein source with incidental sugar, but they're loaded with natural sugar. Like many of you, I drink a cocktail when there is an occasion to imbibe.

This isn't really a technology-focused story until you think about the role technology plays in our health. We wear Apple Watches, track heart rate, measure steps, count hours of REM sleep, and somehow aggregate all of this data into some amorphous blob. Whether that blob is valuable is debatable, and the proliferation of health information across the internet - both scientifically-grounded and proven and not — makes it easier and more challenging to pick a path.

One blog says keto is terrible for you, but a doctor on Twitter discusses how the same diet reduces diabetes risk. I'm just making stuff up! But it's yet another data point — accurate or not — on the internet. Meanwhile, your resting heart rate is 50 bpm. You're an elite athlete! Now get up at 4:30 am like a real gangster and crush your unread emails.

I don't want to go back to relying purely on medical journals to give me the facts. Nor do I want to go back to relying on U.S. government agencies like the USDA to tell me that certain types of fat are harmful even if there isn't overwhelming scientific consensus. But, how should we sort out the best answer?

Build Highways During Tupperware Parties

Seriously. This newsletter touches directly or tangentially on technology and its place in our lives, but what does that mean? We all use apps, we all use lots of software in our professional lives, and SaaS is just so crazy-ex hot, right? Marc Andreessen told us in 2011 that software was eating the world.

But technology isn't only software. Technology is also building tangible objects that we use, like toasters, cars, and plastic items of all shapes, sizes, and uses. I still marvel at what lacks sexiness these days: we derived the ideas, process, and technology to build labyrinth subway networks in many cities.

Back to plastic. It changed the world for all conceivable good and bad reasons. We store, transport, and preserve food in enormous quantities because of plastic. Delicious children's nuggets often come wrapped in sealed plastic bags. Devouring these tasty morsels presents a classic problem: what to do with the plastic? Can it be recycled? Is it the right number for recycling? What does your local town or city do, and what will they accept?

You might throw that plastic bag in your recycling, or you might throw it in the trash. One option assumes that sorting your plastic means it is recycled? Well, no, not really. It's probably not recycled if it's not clean enough, and it's probably not recycled if it isn't the correct number encased in that cute little triangle that lives on lots of plastic materials. Even after all the appropriate steps, it might be cheaper for your town to send the whole pile to an incinerator to avoid spending a fortune being green.

Yale’s Environment360 project published an interesting article about what we could be doing instead:

A road running through Accra, Ghana’s capital, looks like any other blacktop. Yet what most drivers don’t realize is that the asphalt under them contains a slurry of used plastics — shredded and melted bags, bottles, and snack wraps — that otherwise were destined for a landfill.

Plastic recycling isn't just a problem here in the United States. The pain is acute in other countries that both consume plastic and receive "recycling" from other countries in return for cash.

Barely 5 percent of the 5,000 tons of plastic that Ghanaians discard each day makes it to recycling facilities. The rest winds up in landfills, illegal dumps, streets, and waterways, or is burned in open pits, poisoning the air. In a developing nation, “it’s difficult to recycle plastic,” noted Heather Troutman, program manager of the Ghana National Plastic Action Partnership. “It’s expensive, complicated, technical, and much easier just to burn it. But if you could put value on recycled plastic,” by turning it into fishing nets, fuel, or paving material, “it won’t get buried; it won’t get burned; it won’t make it to the ocean.”

Ghanaians toss 5,000 tons away each year and recycle 5%, maybe. We're not much better here in the U.S., recycling just 8% of what we generate every year.¹.

Let’s put that plastic into our roads:

A growing number of studies say that roads containing waste plastic have the potential to perform as well or better than traditional roads. They can last longer, are stronger and more durable in respect to loads and rutting, can tolerate wide temperature swings, and are more resistant to water damage, cracking, and potholes. The technology also has the potential to reclaim anywhere from a small to a sizable amount of plastics from landfills and random dumping, researchers are finding, while providing a significant amount for road paving and repair. In a small nation like Ghana, where only 23 percent of roads are presently paved, waste plastic could go a long way.

Waste plastic could be huge in a large country with an aging infrastructure. This use case is perfect for all of our delivery containers from DoorDash, shopping bags from the grocery store and FreshDirect, and soft-shell zip-ups full of organic granola.

Collecting Your Poop: It’s Science

This newsletter's genesis was a short email about a product called Text Rex, a messaging-based recommendation service from The Infatuation. Text Rex provided recommendations using ordinary text messages. That's right! Not some AI virtual assistant! The service was unlike anything I had encountered and instantly became my go-to for recommendations. Yelp was a dumpster fire. The New York Times restaurant section couldn't cover every possible filter. Text Rex allowed me to describemy needs and receive recommendations rather than filter websites designed for broader use.

Text Rex wasn't life-changing technology, but it was an innovative use of existing technology. So much of what I look at in tech is, well, boring. So much of the reporting and Twitter hot-takes feels predictable: "hot new SV app raises [X] round." Twitter and Tiktok light the app on fire and, viola, news!

Yet, in Cambridge, Massachusetts, home to esteemed institutions like the hotel where I got married and forgot the last third of my wedding, scientists collect sewage to spot health and disease trends and recommend action to the city. From The New Yorker:

“It’s called ‘stool’ within the medical community, or ‘biological sample,’ ” Jennings Heussner, a business-development associate with Biobot Analytics, who accompanied Randall, said. Biobot is a company that specializes in wastewater epidemiology, or the study of sewage for the purpose of tracking the spread of disease. “We are great at euphemisms in this business,” Heussner added.

Biobot has a contract with Cambridge to analyze wastewater for COVID-19. The city and the school district used the data to help make decisions about whether schools should stay open; last month, the district switched to remote learning, after high wastewater virus levels and daily cases rose. Water samples are collected every Tuesday morning. Randall hooked the crowbars under the manhole cover and pried it off. A yellow harness hung inside, attached to what looked like a small submarine dangling below. Randall pulled it up, opened the lid, and removed a bottle. “It’s a combination of sink water and toilet water,” he said, holding the bottle up to the light. “It’s not too cloudy.” He carefully poured the contents into specimen tubes, which would be shipped to a lab and analyzed.

Now we’re talking! Last night’s Mexican food informs public health! All kidding aside, this is awesome in every way. Relatively boring technology and analytics are blended to understand viral loads and increasing drug use in city populations. All through sewage.

1. Source: https://www.epa.gov/facts-and-figures-about-materials-waste-and-recycling/plastics-material-specific-data. Used 2018 data as basis for my math.