We Track Stocks by the Second—But Can’t Measure Our Health

Why the most important trends in medicine remain surprisingly hard to see

We can watch markets move in milliseconds—but when it comes to our health, we often don’t know what’s changing until years later.

I find it striking that we can track stock prices second by second, watching markets flicker in real time. A trader in New York or Singapore can see the value of a company change in milliseconds. Algorithms respond instantly. Billions of dollars move in response to tiny shifts in information.

Yet when it comes to major health conditions—from autism to colon cancer—our surveillance systems are often so weak that we cannot even agree on whether the problem is rising or falling. Sometimes the uncertainty persists for years. Sometimes for decades.

The contrast is not just technological. It is temporal. Markets generate immediate signals. Health unfolds slowly, often with long delays between cause and effect.

Counting Death, Missing Causes

Few documents seem as definitive as a death certificate. As Kathryn Schulz shows in a New Yorker essay, that certainty is often an illusion. She traces their origins not to scientific discovery, but to necessity.

During the Great Plague of London in 1665, officials compiled “Bills of Mortality” to track the spread of disease—counting the dead and, eventually, assigning causes. The causes were often imaginative—“grief,” “teething,” “falling from a horse”—but the goal was not precision. It was order.

Over time, this crude system evolved into the modern death certificate, with its tidy lines and standardized codes. The form suggests clarity: a single cause of death, written down, coded, counted. But that clarity is often an illusion. Physicians are asked to compress a complex chain of events into a linear narrative—one underlying cause, a few contributing factors—often with incomplete information. Training is uneven. Judgment varies. Sometimes the cause is inferred; sometimes it is guessed.

In practice, death certificates record the final event—the heart that stops, the lungs that fail—rather than the deeper forces that set those events in motion.

That limitation is easy to miss until you see it in a single life. Ella Kissi-Debrah, a nine-year-old girl in London, died after a severe asthma attack. For years, her death was attributed simply to asthma. Only after a seven-year inquiry did a coroner conclude that air pollution had made a material contribution—the first time this had been formally recognized on a death certificate. The underlying cause had been there all along. It simply wasn’t counted.

What we are left with is a paradox. Death certificates tell us how many people die with remarkable precision. But when it comes to why, they are far less certain.

And that uncertainty follows a pattern. We are good at counting the final events—the diseases, the organ failures, the last link in the chain. We are far less consistent at measuring the upstream causes—the exposures and conditions that shape who gets sick in the first place.

Autism and the Limits of Data

Autism offers a telling example of how fragile our measurements can be—much like other common conditions, where even basic trends are hard to measure.

Over the past three decades, rates of autism spectrum disorder have risen dramatically. Some researchers argue that the increase reflects broader diagnostic criteria and better recognition. Others suspect that environmental factors are contributing to a real rise in prevalence. Both explanations likely contribute.

But the uncomfortable reality is that our surveillance systems are not strong enough to settle the debate.

Data come from fragmented sources—school records, clinical diagnoses, insurance claims. Definitions change. Reporting practices evolve. Even in the case of autism, where the CDC has built a more rigorous surveillance system over the past two decades, important uncertainties remain. Differences in ascertainment, diagnostic practices, and data sources still make it difficult to determine how much of the observed increase reflects a true rise.

Scientists often spend as much time debating the data as interpreting it. Beneath the disagreement lies a simpler problem: we are trying to answer large questions with incomplete information.

This is not just a problem of autism. It reflects a broader limitation in how we measure health.

We tend to think of surveillance as a technical exercise—collecting data, building systems, improving efficiency. At its core, it is about seeing where disease is emerging, how exposures are shifting, and whether health is improving or deteriorating. Without connected, continuous data, even a modern healthcare system can become surprisingly blind—rich in information, but poor in awareness.

Surveillance is not just a public health tool. It is a form of national security. A country that cannot see what is happening to its population cannot protect it.

What Other Countries Do Differently

Other countries have made that investment.

Denmark and Finland, for example, maintain national registries that track health conditions across entire populations. Researchers can examine decades of data on schizophrenia, autism, or congenital anomalies. Because these records are systematic and continuous, they allow scientists to detect slow-moving changes—the kind that accumulate over years and only become visible at the population level.

Some of these systems are extraordinary. The UK Biobank has enrolled hundreds of thousands of participants, collecting detailed data on genetics, clinical biomarkers, and disease outcomes. Nordic registries link health, education, and social data with a level of precision that was unimaginable a generation ago.

These systems have transformed our ability to measure disease.

But they were not designed to capture the environmental conditions that give rise to it. Information on chemical exposures and other environmental risks remains sparse, inconsistent, and often entirely missing. We can track diagnoses, hospitalizations, and deaths with increasing accuracy, yet we often lack comparable data on the exposures that precede them.

We can see the outlines of an epidemic, but its causes often remain just out of view.

A Patchwork of Surveillance

The United States faces a different problem.

We do not have a single, unified system. Instead, we rely on a patchwork of surveys, insurance claims, hospital records, and regional registries. Each provides a partial view. Together they offer useful information, but rarely the continuity needed to detect subtle change across an entire population.

There is one notable exception.

The National Health and Nutrition Examination Survey—NHANES—does something few other systems even attempt. It measures not just disease, but exposure. It collects data on toxic chemicals in blood and urine, along with detailed information on nutrition and other risk factors. In that sense, it comes closer than most systems to capturing the environmental conditions that shape health.

But NHANES has its own limitations.

It is conducted continuously, but still captures only a small slice of the population at any given time. Its sample, while representative, is not large enough to track trends for many conditions with precision. And for some of the most important health outcomes, it lacks the longitudinal depth needed to detect slow, population-wide changes.

We are left with a paradox.

Some countries can see disease clearly, but not its causes. The United States can measure exposures but struggles to track their consequences over time. Nowhere do we do both well.

What We Value

The result is an imbalance. We have exquisitely sensitive instruments for measuring the pulse of the economy, yet remarkably blunt ones for tracking the health of our population.

That imbalance reflects our priorities.

Markets demand immediacy. Investors want to know, instantly, whether a company’s value has changed by a fraction of a percent. Governments and corporations have built enormous systems to satisfy that demand.

Public health operates on a different clock. Diseases unfold over years or decades. Exposures accumulate quietly. The signals are faint, often buried in the noise of everyday life.

Detecting those signals requires patience, continuity, and a tolerance for delayed answers.

Political systems are not especially good at that.

When Signals Are Hard to See

The consequences become apparent whenever a new condition emerges.

Recent studies have reported rising cancer rates among younger adults. The increases are modest for many cancers but striking for a few, including colorectal cancer. Researchers are now scrambling to understand whether the pattern reflects changes in lifestyle, environmental exposures, screening practices, or something else entirely.

But the first question remains: Is the increase real?

Even in a country with advanced medical systems, the answer is not always clear. Surveillance systems take time to detect trends—and even longer to confirm them. By the time the signal becomes unmistakable, the underlying causes may already have been shaping health for decades.

This pattern repeats across public health. The signal is there—but it is spread thin, across millions of people, where it is easy to overlook. We miss these problems not because they aren’t there, but because we look for large effects in individuals instead of small shifts across populations.

A Different Kind of Attention

We have built systems that can measure some aspects of disease with increasing precision, and others that can measure exposure. But we have not built systems that can do both—consistently, continuously, and at scale.

What we measure becomes visible; what we fail to measure remains obscured. And that gap shapes how we understand health.

The most important threats today—chronic diseases and mental health conditions driven by subtle shifts in risk—develop slowly across entire populations, with effects that appear in individuals at the extremes but are often explained away as genetic or behavioral.

Markets fluctuate by the second, and we have built systems to capture every movement. But the health of a population unfolds over years and decades, demanding a different kind of attention—one that values long-term observation as much as real-time reaction.

Our systems are precise—but not complete.

Until we make that shift, we may continue to know the price of everything in our economy while remaining uncertain about the forces that shape our health.