Just over a decade ago, Boston doctors began monitoring a population of 119 homeless people with health problems. The subjects’ average age was 47. Today roughly half of them are dead.
That toll is not atypical: a homeless person of any medical background is roughly four times more likely to die than a housed person of the same age. These deaths are often lonely, anonymous affairs. After being warehoused in a city coroner’s office for months, the body may be cremated and buried in a pauper’s field.
“Somebody dying on our streets—I think that’s as bad as it gets in America,” says Rebecca Kanis, director of the 100,000 Homes Campaign, a movement of more than a hundred community groups aiming to house most of the nation’s 110,000 chronically homeless by 2014. “We can do better than this.”
The campaign is introducing an unlikely tool to prevent these tragedies: a potentially life-saving mobile app being tested in several communities this summer. The “Homeless Connector” will eventually allow ordinary Americans on their way to class or home from work to identify the people most at risk of dying on the street, and to find them help.
The app is based on the research of Jim O’Connell, an internist with Boston’s Health Care for the Homeless program who earned the trust of the city’s street people over decades in part by doing shifts on a sandwich wagon.
O’Connell (often working with another doctor, Stephen Hwang) realized gradually that certain widespread theories about homeless people’s health didn’t hold up. His patients didn’t die more often in the winter, as was commonly supposed; they died throughout the year, and fall was actually the more lethal season. “It was in the transition between fall and winter,” he says, because that’s when people who check out of homeless shelters after the summer are exposed to cold for the first time.
Also, the dead weren’t people who avoided institutional treatment and “fell through the cracks,” as previously believed. Many had checked into emergency rooms and detox centers just days before death. And certain health conditions that are relatively common in the homeless population marked patients for greatly increased risk of dying. For instance, frostbite doesn’t typically kill people, but, in part because it suggests that the patient isn’t aware of his surroundings, it is a key indicator of more catastrophic troubles to come.
In the mid-2000s, Kanis mined O’Connell’s research to develop a questionnaire called the Vulnerability Index. Along with basic biographical questions, the survey asks a homeless person about eight risk factors that lead to an elevated risk of dying: Are you 60 or older? Have you been hospitalized more than three times in the last year? Have you visited the emergency room more than three times in the last three months? Do you suffer from cirrhosis of the liver? End-stage renal disease? HIV/AIDS? Do you have any other chronic medical conditions combined with psychiatric and substance abuse problems? Do you have a history of hypothermia or frostbite?
Roughly 43 percent of the homeless answer yes to at least one question. These medically fragile people become the 100,000 Homes Campaign’s priorities. Finding them housing fast can lengthen their lives (many homeless people don’t take vital medication, for instance, because it dulls their senses, making it harder to stay vigilant on the dangerous streets). Others, already dying, are able to die with dignity in a home of their own.
The Vulnerability Index was first used in New York City. Now more than 60 communities across the country affiliated with the campaign have adopted it. Typically, volunteers canvass an area between 4 a.m. and 6 a.m. three days in a row. They wake everyone they see sleeping on the streets; about 70 percent agree to be surveyed. In addition to collecting the medical and biographical data, volunteers take a picture of the person. Back at headquarters, this information becomes the basis for future strategy: “they can write the names on a dry erase board: this is who we’re getting this week,” Kanis says.
Being launched in five communities in June, the “Homeless Connector” mobile app is a version of the Vulnerability Index that, through the 100,000 Homes Campaign website, will eventually be available to anyone with a tablet or smartphone. Encountering a homeless person on the street, the volunteer electronically enters the survey answers and (with permission) snaps a cell phone picture. The information (along with the homeless person’s geographic coordinates, recorded via smartphone) is routed to the headquarters of the nearest community group participating in the 100,000 Homes Campaign.
“We want to create a country in which every homeless person is known by name and health condition,” Kanis says. “We want to strip away the anonymity of homelessness.”