WHEN I BEGAN MY SEARCH FOR THE STATE’S deadliest place about a year ago, I assumed I’d discover it among our biggest cities, with their violent crime and resurgent incidences of AIDS; or along the border, with its unsanitary air and water and old nemeses like tuberculosis; or someplace in the Golden Triangle (Beaumont, Port Arthur, Orange), with its notorious high cancer rates. I never dreamed I’d find it in Potter County (population: 113,546), an unremarkable quadrant of the Panhandle that is home to half of the city of Amarillo, the Alibates Flint Quarries, and the infamous if-you-can-eat-it-all-it’s-free steak.

But as the epidemiologist’s favorite mantra goes, the numbers don’t lie. Using the average of 1997, 1998, and 1999 Texas Department of Health vital statistics, which reveals a true trend rather than just a snapshot, I found that people die at a faster rate out here on the High Plains than in any of Texas’ 34 counties with populations of more than 100,000. No one has died of anthrax recently in these parts, but people are dying of everything else. Potter County ranked first in deaths from all causes per 100,000 (1,206, compared with the state average of 897). It was first in infant mortality, heart disease, lower respiratory disease, and accidents; second in deaths from cancer. Potter was also among the state’s leaders in such varied health “indicators” as violent crime, sexually transmitted diseases, and low birth weights. Just as confounding as Potter’s deadliness was the fact that its sister county, Randall, which sits just across Interstate 40, ranked near the bottom of those same 34 counties, with an overall mortality rate of just 701 deaths per 100,000.

When you drive Potter’s narrow, generally treeless and chuckholed residential streets, the first thing you notice is its poverty. The houses are mostly small, slab-foundation, wood-frame structures, and many of them are boarded up and surrounded by empty lots. There is also a conspicuous lack of the accoutrements of modern middle-class civilization—condo developments and shopping malls and a Starbucks on every other corner. But you’d never think of the word “deadly” to describe this place. It’s too quiet.

In determining the death-rate capital of Texas, I excluded counties with fewer than 100,000 people. That’s because small numbers are anathema to good epidemiology. To qualify as a truly sick, deadly place, a county needed to have many generations, ethnicities, social classes, and the basic dangers of urban life (crime, traffic accidents, AIDS) represented. It also had to have enough health-care resources available to be able to judge whether they were part of the problem. Real County, in Central Texas, actually has the highest death rate in the state. But its statistics are skewed by its aging population of 2,500 who have only two doctors to take care of them.

Why they’re dying in Potter County is not so transparent. When I first posed the question to the Amarillo director of public health, Claudia Blackburn, last January, she said she knew that Potter’s mortality numbers were high but seemed as puzzled as I about why they were chart-topping. To her credit, she then went public with my finding, telling the Amarillo Daily News, “We know that this is an issue.”

Blackburn also pointed out that since Amarillo is almost equally divided between two counties, Potter County’s numbers should be viewed like those from the poorer half of any city. Potter County had twice as many residents living below the poverty line as Randall did and three times the number of police-dispatch calls. “If you combine the counties and look at the whole city, we look more like everyone else,” she said.

I saw what she was getting at. But then, how much better would Dallas County’s death rate look if it could combine its statistics with neighboring Collin County, which is nothing but affluent suburbs like Plano? Besides, poverty per se clearly wasn’t Potter County’s only problem, as there were some counties like Webb, which contains the city of Laredo, that had higher percentages of low-income residents but substantially lower rates of death. This disparity may be due in part to what state epidemiologist Dennis Perrotta calls the “Mexican paradox”—the remarkable longevity of even the poorest Mexican Americans. But in any case, the real question was, What was it particularly about Potter’s poverty that was making the county so lethal?

One possibility was its exposure to environmental toxins over the years. Of the two Amarillo counties, Potter has always been where heavy industry has settled: a Texaco refinery, a smelter, a helium plant (all long since closed), and a huge meat-processing plant. Could any of these have produced some sort of occult toxicity that was still killing people? That’s doubtful since no cancer clusters or disease syndromes have occurred, and a 2000 federal study of the Pantex nuclear weapons facility in contiguous Carson County produced no smoking gun either. Federal studies frequently say that, I know, but if there were some silent killer in the air, soil, or water around here, wouldn’t it be affecting Randall’s residents as lethally as Potter’s?

Another possibility was lack of adequate health care. But I found that Potter denizens have five hospitals at their disposal and a doctor-resident ratio that was below the state average. I looked at whether Potter’s large minority population (nearly 40 percent black, Hispanic, and Asian compared with Randall’s 12 percent) could be inflating the mortality figures. But I found that all races in Potter were dying about twice as fast as their counterparts in Randall. And the high number of nursing homes in Potter County contributed only slightly to its outsized mortality rate.

So I was back to square one. The main thing that seemed to distinguish Potter from Randall—and from many other counties—was its poverty. And no question about it, poverty is an established killer, shorthand for a host of “at risk” behaviors: poor eating habits, smoking, substance abuse, nonuse of condoms and seat belts and smoke alarms, lack of prenatal care. In that sense, Potter and Randall counties may be seen as a terrific case study of the oldest public health maxim of them all: that at least half of us die each year because of the most elementary “health behaviors”—and those are inextricably linked to income and education levels.

But there clearly had to be something that was making Potter not just more deadly than Randall but more so even than any of Texas’ other poor counties. Potter County commissioner Iris Lawrence, an African American who grew up in the area, theorizes that the isolation of the poor and the ethnic in Amarillo is even more oppressive than that suffered by their counterparts in big cities. The county’s slide into obscurity seems to have started in the late sixties. Up until then, it was regarded as the “real” Amarillo: It contained the downtown area, the main rail yards, major industries, the local military installation. But when Amarillo Air Force Base closed, in 1968, thousands of homes here were suddenly abandoned and scores of businesses were left without a clientele. “The place just never did snap back,” says Amarillo’s director of community services, J. D. Smith.

So the empty homes were eventually demolished, the businesses moved south of I-40, and the Potter County part of Amarillo became a kind of urban outback where the African American community, various immigrant classes such as Hispanics (old) and Vietnamese (new), and the working and the retired poor tended to aggregate. “Most who could afford to move out did,” says Smith, pointing out that between 1960 and 1970, Potter County’s population dropped from 115,000 to 90,000.

This “left behind” mentality, coupled with the endemic geographic isolation of Amarillo, may make poor health behaviors that much more difficult to reform in Potter than elsewhere. For example, one obvious reason for Potter’s mortality rate is simply that many of its residents still smoke at rates higher than the state and national averages. Smoking especially increases the rate of cardiovascular and lower respiratory deaths—two key components of Potter’s overall death rate.

In a similar way, Potter’s high infant mortality rate may be a case of the right message not getting through. Infant deaths tend to result from teenage pregnancies (of which Potter also has a huge number), poor prenatal care, short interpregnancy intervals, and inadequate maternal education—all problems governed by education and income levels. Why is everybody dying in Potter County? Well, for one thing, they’re not taking very good care of themselves.

In the face of all of this, Blackburn has done the right studies and initiated many of the right education programs for sexually transmitted diseases, smoking, and teenage pregnancy, with some measure of success. But her real foe here is much more insidious and elusive than any bioterrorism plot. It is the particular malaise known as “being stuck.” Aside from contributing to poor self-care, this condition is now recognized as a major health risk in and of itself. Recent studies have suggested that the stress produced by socioeconomic isolation and inequality can have specific pathological effects such as cardiovascular disease and depression. And while it’s impossible to prove, it seems to me that this deadly dynamic can only be more profound if you’re reminded of your inequality every time you cross the interstate and see the shiny new face of Randall County or if every time you approach the edge of town and take in that infinite Panhandle horizon and listen to the deafening quiet, you realize that, depending on which side of the tracks you’re from, there’s everywhere—or nowhere—to go from here.