[ad_1]
Jocelyn Escanuela was at Walmart checkout line when an unknown number appeared on her phone. She still can’t explain why she picked it up, then hears the pitch of the cold character, which sounds like a scam.
She was chosen to receive the $6,000 grant, the caller told her. And she will have a personal assistant to help her get through her “crisis.”
Did they even know that she was in crisis?
The caller turns out to be legal. She came from the Homelessness Prevention Unit. This is an experimental Los Angeles County program that is testing whether stopping homelessness is viable before starting it alone at a time by choosing them from a mountain of data.
The Escanuela crisis was detected not by humans but by predictive statistical models developed to solve the challenges that have made homelessness prevention an appetizing but lacking strategy.
Despite sound evidence that services such as eviction defense and financial aid can prevent people from becoming homeless, it is impossible to know if a particular person has become homeless without help. Research shows that only a small percentage do so. The elusive goal of prevention is to identify a small percentage.
“There are real consequences for not taking them to those who need them most because there are limited preventive resources to work with them,” said Steve Berg, chief policy officer for the National Alliance to End Homelessness, which has raised eyebrows at historically costly prevention programs.
But Berg said, “If these emerging technologies prove to be effective in predicting who is most likely to become homeless without help, that would be good news.”
Achieving that elusive accuracy will become increasingly important as both the city’s ULA “mansion tax” and countywide measurements begin to lead millions of dollars to prevent homelessness.
The model that chose Escanuela as a high risk has been tested to see how effective it is.
This was created by UCLA’s California Policy Lab. It is a laboratory that has access to data from county agencies, such as the health and social services sector, interacting with the most vulnerable people. Policy Lab sifts through all its data and evaluates around 500 markers to generate a list of individuals and families that it predicts is at a higher risk of becoming homeless. The list will be handed over to the Homeless Prevention Unit and its housing stabilization team.
“We meet people who have just left the hospital and those who have just lost their jobs,” said Dana Rayvanderford, associate director of the Homelessness Prevention Unit. “When we lose our only provider, our family, we meet people. We see people because they have been verbally eviction warnings from their landlords.”
Escanuela runs her own eyelash service business in an apartment she shares with her mother.
(Allen J. Scheven/Los Angeles Times)
Analysts in the Homeless Prevention Unit will randomly move the names of high-risk lists to come up with two candidate groups. Half will be offered intervention – cash scholarships and case managers for four months. The other half has received nothing and never knows that they have been chosen, but they are monitored through contacts with the county and homeless agencies.
Escanuela has landed in the target group (lucky half) of random clinical trials.
The Holy Grail of Prevention is a model that allows you to identify who will become homeless and avoid spending money on people who will never do.
A 2023 report found that in the University of Notre Dame’s Economic Opportunity Lab, people serving in the Santa Clara County Prevention Program are almost 80% less likely to become homeless than the control group after receiving services.
That’s not impressive as only 4.1% of those who didn’t get help became homeless.
“We’ve learned about the home-based prevention program in New York City,” said Bethsin, a research professor at Vanderbilt University.
Her research found that models are moderately superior to outreach workers at prediction.
“Even the urban way, it was cost-effective and reasonably successful,” Singh said.
These studies involved people seeking preventive services. Policy Labs and the Homelessness Prevention Unit are taking the next step to find people who are not seeking services using predictive analytics.
The initial findings are promising. The data used to build the model predicts that around 47,000 people receiving county services, 24% of those predicted to be at risk, are homeless compared to only 7% of the entire sample.
It has also been proven effective in finding people who may become chronically homeless.
“Our clients live with a really high level of risk,” Vanderford said. “They have complex health and mental health conditions. They meet us at the real moment of crisis. The timing of our clients seems magical to me.”
After sufficient people have been followed for 18 months after completing the four-month program, the full results of the exam will not be final until 2027.
The Homelessness Prevention Unit was created with funding from the American Rescue Plan Act, supplemented by the county’s funding. We have around 250 active clients and sales of 4-6 months can process 750 people a year. About 90% said they either kept their homes or found a new home, Vanderford said.
It’s labor-intensive work. Four analysts pass through a RAW list that randomly screens unqualified candidates. There is a delay before Policy Labs retrieves county data, so many of the final lists are already homeless and prove that their predictions are accurate.
“There’s a real challenge to staying in touch with people,” Vanderford said. “The phone goes off. The client may be hospitalized or put in prison. Clients may wonder that this call sounds a little too good to make it true. Voicemail doesn’t respond.”
“I would never answer such calls,” Escanuela said. “I don’t know what forced me to answer.”
Neither Escanuela nor Vanderford knows the specific factors that made her on the high-risk list, except that she has access to county services.
But the call was timely. She and her family were in a long battle of eviction, and she was afraid to become homeless again.
She said as a child she spent a long time at the church in parks and in the evenings, and later lived at the Union Rescue Mission downtown and the Hope Garden Shelter in Sylmer.
“I didn’t want to go back to it, especially as an adult,” she said.
Once registered, Escanuela received a call from Chris Schuchert, one of the program’s 20 contract case managers. Over the program’s four-month and two-month extension, they communicated over phone and text. He helped her in many ways, from getting a grocery debit card to dealing with her emotional needs.
“Chris managed to find me a therapist,” Escanuela said. “I was going through so many things when I had to talk to someone.”
The program’s case manager processes client expenses and pays directly to the vendor and landlord. Case managers need to request, which often follows negotiations with clients. Status items such as $250 shoes touch on the $50 model, but you can approve a $800 bed to improve your sleep, Schuchert said.
The case manager will provide referrals to health and mental health institutions. And if the client is thousands of dollars in rent, they can refer them and stay in LA or other groups that help the suffering tenants.
In Escanuela’s case, that wasn’t necessary as the landlord stopped accepting rent during the eviction process. Shuchert thought it would be best to leave the apartment she shared with her mother and siblings and avoid eviction of her records.
Today, she lives with her mother in a comfortable apartment in Pomona. She paid the move-in cost for a home business providing eyelash services and saved enough money to buy the equipment.
She said she pays her way.
[ad_2]Source link
