On my commute into work one night this past January, I could tell that something unusual was happening: Several police cars with full lights and sirens raced past me, as did a pair of ambulances. I followed their wails through downtown Seattle to King County’s public hospital, where I work the night shift as the physician responsible for treating patients with general medical issues. The hospital serves many members of Seattle’s homeless population; on any given night, at least half of my patients stay in migrating “tent cities,” crisis shelters, empty vehicles, abandoned buildings, or directly on the streets.  

When I arrived, several news vans were parked outside the door to the emergency department, a sure sign that some sort of tragedy had just occurred in the city. Their lights were off, but their antennae were up and their engines were idling; it appeared that the evening news live-spots had just wrapped up. I soon learned that five shooting victims had been brought into the hospital; they’d been shot as they huddled around a campfire in “the jungle,” an illegal homeless encampment nestled near the southern tip of downtown Seattle. Two had died, while three remained critically injured.

From the outside of the building, it might have felt like an unusual night, but inside it was a fairly typical scene. The ER was humming with the electric air of urgency: hospital stretchers and their inhabitants filled the available curtained-off bays; patients in better condition slept on stretchers lining the hallways. Several patients in the waiting room–triaged as the least sick in the building–sat waiting in various stages of boredom.

I reviewed the chart of a patient I had seen recently, a man who just couldn’t seem to keep his diabetes under control. He had been in last month,  and three other times last year, for treatment of a potentially life-threatening complication of his high blood sugar. When I asked how things had been going, he made a disgusted face. “I hate it, man.” Frankly, it had surprised me to see him again so soon after his last discharge. He had seemed to understand the details of how and when to take his insulin the last time he left the hospital—and the gravity of what would happen if he didn’t. “Can’t keep no needles around at the shelter,” he said simply, when I asked what went wrong. For him, and for many homeless people with diabetes, the challenges of managing a chronic illness on the streets—finding secure storage for medications, reliably obtaining nutritious food, appropriately dosing and timing injections—can be insurmountable.

Later, I checked in on another patient of mine, a man who had come in several nights prior wearing a pair of tattered, sopping-wet socks and an old dirt-encrusted pair of jeans. His feet were the same color as the socks, a not-quite-biologic shade of grey. The jeans had essentially eroded into his shins, to the point they had to be cut away from his skin. He was diagnosed with trenchfoot, a condition caused by prolonged exposure to cold and dampness; over time, the damp skin can progress to maceration, ulceration, and gangrene. Prevention is simple: clean, dry socks and shoes that can be removed when necessary; the ability to elevate extremities; and access to basic medical supplies like gauze and tape.

But for homeless patients—even those who are able to access temporary living situations—this can be a daily struggle. Many shelters require residents to vacate in the hours of the morning and line up again for reentry around dusk. In the meantime, individuals often pass the time on wet, dirty streets—a situation not conducive to the elevation, rest, and frequent wound care required for healing extremities. This particular patient’s case was especially severe and required amputation of the irreversibly injured tissue. He told me a bit about his life on the streets, including some time passing through “the jungle” over the last few years. “I knew it was turning into a real bad place,” he said. And then, echoing a line that Seattle’s mayor, Ed Murray, had told reporters following the shooting: “People are dying out there.”

It's true. Across the country, homeless people die disproportionately young. In the Seattle area in 2015, the average age of death for homeless individuals was 48 years old. Other major American cities report similar averages: 48 in Los Angeles, 53 in Philadelphia. When recently I told another one of my patients, , a 42 year-old man who had lived most of his adult life sleeping in doorways and under bridges, that his liver was starting to show some damage from a combination of drinking and Hepatitis C, he shrugged: “Not like I got much time left anyway.”

This human calculus can often be a driving force for decisions that are otherwise seemingly inexplicable. One night, I was called by a bedside nurse to talk to a 49-year-old woman who had been homeless for several years, ever since escaping a domestic partner who beat her so brutally that she still walked with a limp. Years of battling depression and chronic pain led to a heroin addiction and, now, a dangerous bloodstream infection. She had a prolonged course of IV antibiotics ahead of her, but she desperately wanted to leave. The hospital social worker had helped her get in touch with her daughter, who she had lost in the drawn-out custody battle, and, it turned out, there was an unmet grandchild to visit. As I tried to reason with her to stay, thinking I had the ultimate trump card, I announced authoritatively: “But you could die!” Puzzled for a moment, she stared back at me and calmly replied: “Well, of course. That’s why I want to visit her now.”

My warning wasn’t an exaggeration: Addiction is substantially deadlier for homeless as compared to non-homeless populations. According to a recent paper in The American Journal of Public Health, mortality rates linked to tobacco were three to five times higher for homeless individuals, alcohol-attributable mortality rates were six to 10 times higher; and drug-attributable mortality rates were eight to 17 times higher. But overall, though substance abuse was a major factor, it only contributed around 57 percent of the difference in mortality rates between the study’s homeless and non-homeless cohorts. The difference may be explained by the challenges that patients like mine face in managing chronic medical issues with unstable living situations and limited resources.

A day or so after the shooting, the news vans packed up and left the hospital. I hope there are no more violent tragedies to bring them back anytime soon; I also hope the patients I saw that night don't return with new problems. Even if they don’t, though, I know I'll continue to see many others like them—people with similar struggles, living reminders of the ongoing brutality of what it is to be sick without a home.