Preventing homelessness using a computer model

When her phone rang in February, Mashawn Cross was skeptical of the friendly voice offering help at the end of the line.

‘You said you’re doing what? And you’re with who?” the 52-year-old remembered to say.

Cross, who was out of work because of her ailing back and knees, scraped around $200 a month in aid plus whatever she could earn recycling bottles and cans. Her gas and electricity bills chewed on her checks. She’d been in and out of the emergency room, her doctor said she might need to get a colostomy bag, and depression plagued her day by day.

Kourtni Gouché listened and began to help. The LA County employee helped Cross get some household supplies so she could save money and pay her utility bills. She offered to buy her a new bed to soothe her aching back. She began linking Cross to programs to relieve her depression and get her off cigarettes, something Cross had long wanted but struggled with.

“I feel like I’ve got a friend here,” Cross said, breaking into tears now and then as she praised the social worker who kept standing up for her. In her apartment in a duplex in south LA, she suddenly remembered a question she had forgotten to ask Gouché during their regular conversations.

“How did you get my name to begin with?” asked Cross.

The answer is an unusual mobilization of data analytics to prevent homelessness before it starts.

Cross Is Part Of LA County’s Rare Effort To Marry Predictive Models — a tool used to predict events by tracking patterns in current and historical data — with the highly personalized work of homelessness prevention.

Cross, left, with Vanderford and Gouché.

(Robert Gauthier/Los Angeles Times)

The county found Cross and dozens of other people through a predictive tool developed by UCLA researchers, which collects data from eight LA ​​County agencies to help emergency responders focus their attention and help on people believed to be most at risk. to lose their home.

LA County struggles to keep up with the number of people who become homeless each year, even as it ramps up its efforts to get people into their homes. It’s critical to figure out who to help because millions of residents seem vulnerable yet avoid homelessness, says Janey Rountree, founder of California Policy Law at UCLA.

“You would never have enough money to provide prevention to anyone who appears to be at risk,” Rountree said. “You really need a different strategy to figure out who’s actually going to be homeless if they don’t get help right away.”

Researchers have found it’s surprisingly complicated to guess who will be homeless and who will avoid it: In a report released three years ago, the California Policy Lab and the University of Chicago Poverty Lab said a decent forecast involves at least 50 factors. would require – and that the best models would require “somewhere between 150 and 200”.

The predictive model now being used in LA County uses an algorithm that contains about 500 functions, according to the UCLA team.

It pulls data from eight provincial agencies to determine who needs help, looking at a wide range of data in provincial systems: who ended up in the emergency room. It’s booked in jail. Who has had a psychiatric crisis that has led to hospitalization. Who has received monetary aid or food benefits — and who has listed a county office as their “home address” for such programs, an indicator that often means they were homeless at the time.

Rountree and her team, when it started using such models to identify who was most at risk, found that “they were not the people who enrolled in typical homeless prevention programs.” As of 2020, the UCLA team found that few of the people identified by its predictive modeling — fewer than two dozen in two years — were receiving services specifically designed to prevent homelessness under Measure H, a LA County sales tax imposed by voters approved.

That is why the municipality decided to call them. In July, the newly formed Homeless Prevention Unit began reaching out to people most at risk according to the predictive model, cold calling residents like Cross. The UCLA analysis is performed with data stripped of identifying details, which combines the county with names and information to find potential customers.

The work is being performed in the Health Housing Division of the LA County Department of Health Services, which focuses on homeless and vulnerable patients. People in the program can receive financial aid and referrals to other services to help with their overall health and housing maintenance. Single adults receive up to $4,000 or $6,000 in aid, and families are allocated according to their size; the program analyzes whether higher amounts make a difference in the results.

For Anthony Padilla Cordova, it was “perfect timing” when the call came in in February. “I just didn’t know where I was going.”

Padilla, 29, had been released from state prison during the pandemic and tried to stay off alcohol and drugs. He had finally entered rehab after being re-arrested for his alcohol use and moved from there to a frugal house, he said. But his life still felt precarious.

Many of his housemates, struggling with their own addictions, seemed “still stuck in the prison mindset,” he said, and Padilla feared that if he lost his cool and started fighting someone, he would be kicked out. thrown. Padilla had started working as a prep cook and dishwasher, but he didn’t have his driver’s license back, which could mean spending hours on the bus getting to work as far as Irvine.

Fabian Barajas said he could help. The employee arranged the cost of a breathalyzer that had to be installed and activated in a vehicle before Padilla could legally drive again. He was given gift cards to cover daily expenses such as groceries. He gave him clothes and shoes to wear to work.

And “if I feel like I’ve lost hope…I’ve got Fabian to call,” Padilla said.

Case managers work with each participant for four months, although they limit that period of time with up to . be able to extend another two months if necessary. As Padilla nears the end of the four-month program, he now lives in his own studio in MacArthur Park. He has remained sober. He has his driver’s license again. And he’s saved enough for a down payment on a car, Barajas said.

“It looks good as long as you can stay consistent and sober,” Barajas told him on a recent Tuesday.

Padilla thought about the help he had received from the province and other programs. “If I didn’t have any of these resources,” he said, “I’d probably be homeless,”

About 150 people have completed or are participating in the LA County program. It started with $3 million in funding — half from Measure H, half from the Conrad N. Hilton Foundation — and quickly raised nearly $14 million in federal money under the U.S. bailout plan, which will allow the program to run through 2024, it said. Dana Vanderford, associate director of homelessness prevention. The staff, once “a small and mighty team of seven,” will soon expand to 28, including 16 case managers, Vanderford said.

So far, about 90% of participants have retained their housing during the program, Vanderford said. Rountree and her team are still assessing how such results compare to similar people who did not participate in the program, which will help determine whether it was successful.

But based on past patterns and analysis, it is estimated that the first 54 people enrolled in the program would have had a 33% risk of becoming homeless had the program not existed and “the world looked exactly the same.” if when the analysis was done, Rountree said.

Yet “the world has probably changed. Therefore, we will have to conduct the full causal investigation,” she said.

At first glance, Cross may seem like a surprising candidate for homelessness prevention. She has a housing voucher for her South LA unit. She is not fighting an eviction or rent increase. But Cross has been homeless before — and without Gouché’s help and the program, she’s worried about the choices she’d have to make between food and the electric bill.

“If she didn’t come, I’d be sitting here saying, ‘Should I do this, or should I do this?’ said Cross. Now “I don’t have to stress if I get enough recycling this month.”

In the front room of her house, seated by a television stand decorated with elephant figurines meant to bring good luck, postcards from a trip to Catalina Island, and an open Bible near the Book of Psalms, she and Gouché discussed her next steps.

They discussed getting a new bed and pillows to relieve her back pain; the fear that Cross said was “nothing to play with”; her unease at the many pills she had been prescribed, which she feared would leave her with a new addiction.

Cross also wanted to quit smoking, saying it cost her money she couldn’t afford. Gouché offered to get her phone numbers for smoking cessation programs, as well as additional resources that would help her with mental health and substance use problems.

Cross nodded. “Everything can be tried at least once,” she said. “If it works, I’ll continue.”

Anger and depression had previously sent her back to cigarettes, she said.

Gouché said: “The most important thing is to be aware – you pretty much said it – that you essentially deal with difficult situations that way.”

The programs can help build on that awareness and help her find other ways to deal with it, the caseworker said.

“You’re definitely on the right track,” she told Cross.

Leave a Comment