[Fwd: Subsidized Housing Ends Homelessness: 20 to 1 odds]

Graeme Bacque (gbacque@idirect.com)
Sat, 21 Nov 1998 17:26:11 -0500


-------- Original Message --------
Subject: Subsidized Housing Ends Homelessness:  20 to 1 odds
Date: Sat, 21 Nov 1998 16:29:29 +0000
From: "David Hulchanski, University of Toronto"
<david.hulchanski@utoronto.ca>
Reply-To: david.hulchanski@utoronto.ca
To: "HOMELESSNESS -- Toronto Research Working Group"
<david.hulchanski@utoronto.ca>,"Members, Toronto City Council"
<david.hulchanski@utoronto.ca>


>From

PREDICTORS OF HOMELESSNESS AMONG FAMILIES IN NY CITY:
FROM SHELTER REQUEST TO HOUSING STABILITY
 
American Journal of Public Health, Nov 1998.

--    "Subsidized housing was virtually the only predictor of
residential stability after shelter."

--    "The odds of stability were 20.6 times greater for those who
received subsidized housing than for those who did not. Among
families who received subsidized housing, 97% were in their own
apartment, and 80% were stable..   Of those who did not receive
subsidized housing, only 38% were in their own apartment, and 18%
were stable. The contribution of age was comparatively trivial."

--    "If homeless families had been randomly assigned to receive
subsidized housing in a controlled experiment, we could now conclude
that receipt of subsidized housing, rather than individual
characteristics, was key to their long-term stability."

--    "Mental or physical health problems did not appreciably cause 
family homelessness or impede later stability."

--    "Conclusions. Housing subsidies are critical to ending 
homelessness among families."


===========================


TITLE:  

"Predictors of homelessness among families in New York City: From
shelter request to
housing stability"

American Journal of Public Health, 88(11), 1651-1657,  November 1998

by Marybeth Shinn; Beth C Weitzman; Daniela Stojanovic; James R
Knickman; et al;   New York University

ABSTRACT

Objectives. This study examined predictors of entry into shelter and
subsequent housing stability for a cohort of families receiving
public assistance in New York City.

Methods. Interviews were conducted with 266 families as they
requested shelter and with a comparison sample of 298 families
selected at random from the welfare caseload. Respondents were
reinterviewed 5 years later. Families with prior history of shelter
use were excluded from the follow-up study. 

Results.  Demographic characteristics and housing conditions were
the most important risk factors for shelter entry; enduring poverty
and disruptive social experiences also contributed. Five years
later, four fifths of sheltered families had their own apartment.
Receipt of subsidized housing was the primary predictor of housing
stability among formerly homeless families (odds ratio 20.6,  95%
confidence interval). 

Conclusions. Housing subsidies are critical to ending homelessness
among families. 

----------------------------------------------------

[Selections from the article]

Converging evidence suggests that 3% of Americans have been
literally homeless over the course of a 5-year period.1,2   Far
fewer are without homes on any given night, indicating that, for
many, homelessness is a temporary state. Reducing or ending
homelessness requires knowledge of why people become homeless and
how most manage to return to conventional housing. Because many
putative causes of homelessness cannot be manipulated, researchers
typically infer causes of homelessness by comparing cross-sectional
samples of homeless people with a comparison group. These studies
confound the causes and consequences of homelessness. Furthermore,
if exits from homelessness are not evenly distributed across all
homeless people, cross-sectional designs confound positive
correlates of entry with negative correlates of exit from
homelessness. The present study, also correlational in nature,
avoided these 2 biases. We examined predictors of seeking shelter
among a sample of poor families in New York City with no prior
shelter experience at the time of initial assessment and
reinterviewed them 5 years later to determine predictors of
residential stability. 

Families, this study's focus, make up about 40% of those who become
homeless, although they represent fewer of those who are homeless on
any given night.3 

Explanations for Homelessness 

Researchers have proposed that at least 4 classes of variables
contribute to homelessness: persistent poverty, behavioral
disorders, impoverished social networks, and loss of affordable
housing.4,5 We tested predictor variables from each of these
domains. 

Methods 

Sample.  This article describes predictors of shelter requests and
later housing stability for a cohort of New York City families
initially interviewed in 1988 (time 1) and followed up an average of
4.85 years later (time 2). Shelter requesters represented a census
of eligible families applying for shelter during the study period at
3 of New York's 4 emergency assistance units, the entry point for
all family shelters except for a few small ones specializing in
domestic violence; comparison families were selected at random from
the public assistance caseload via multistage cluster sampling.


Postshelter Stability 

At the time of the follow-up interview, 79% of the 256 families who
used a shelter were housed in their own residence, typically a
rented apartment. Only 4% were in a shelter. Most others were
doubled up with relatives or friends. 

Receipt of subsidized housing between time 1 and time 2 was the
primary predictor of stability at time 2 (seeTable 3). Indeed, the
best model arrived at by the same procedure as for Table 2 included
only age and receipt of subsidized housing. 

The odds of stability were 20.6 times greater for those who received
subsidized housing than for those who did not. Among families who
received subsidized housing, 97% were in their own apartment, and
80% were stable at time 2, figures comparable to those in the
comparison group (92% in their own apartment, 80% stable). Of those
who did not receive subsidized housing, only 38% were in their own
apartment, and 18% were stable. The contribution of age was
comparatively trivial. A 10-year increase in the age of the mother
was associated with a 79% increase in the odds of stability (1.06 10
= 1.79). An alternative analysis, using measures from time 2 rather
than time 1, yielded the same 2-variable model. 

If homeless families had been randomly assigned to receive
subsidized housing in a controlled experiment, we could now conclude
that receipt of subsidized housing, rather than individual
characteristics, was key to their long-term stability.

Discussion 

Homelessness was a stage families passed through, and not a
permanent state: four fifths of families who entered shelter had
their own apartments 5 years later, and three fifths were stably
housed, having been in their own residence at least I year and an
average of nearly 3 years. Individual characteristics associated
with shelter entry did not prevent most families from becoming
rehoused. Of note, families had access to income supports, primarily
Aid to Families with Dependent Children, and nearly half received
some form of subsidized housing; with new time limits on welfare and
fewer new subsidized units available, later cohorts of homeless
families may not fare as well. 

Mental or physical health problems did not appreciably cause family
homelessness or impede later stability. Substance abuse at time 1
had a small association with shelter entry, and both substance abuse
and imprisonment had zero-order relationships with stability that
did not hold up in the context of subsidized housing. Consistent
with other research, levels of mental illness and substance abuse
were lower than among single homeless individuals. Levels of
substance abuse, in particular, were higher after shelter than
before. Substance abuse, like domestic violence, may disrupt
families' lives. We make no claim that housing solved any of these
problems among poor families, but the problems contributed little to
residential instability. 

The importance of subsidized housing to stability among homeless
families may well generalize beyond New York. In 6 cities, 88% of
families who received Section 8 housing and case management services
remained in permanent housing during an 18-month follow-up period.12
Of note, these families had experienced long-term patterns of
recurrent homelessness and needed a variety of health and support
services. Generalization to adults with serious mental illnesses is
less clear; a San Diego study33 found housing more important than
intensive services to stability, while a New York study found
timely, specialized services critical in a group with varied housing
options.34 

A final disturbing finding is that the housing characteristics that
predicted homelessness were widespread in the welfare caseload.
Thirteen percent of families on the welfare caseload in New York in
1988 did not have an apartment of their own but were doubled up with
others. Almost half lived in crowded conditions, with more than 2
people per bedroom. Only 31% lived in subsidized housing.   National
data also suggest a large pool of ill-housed, poor people. Rates of
young adults between 25 and 35 years of age who headed independent
households declined 7.1% in the 1980s,35 and rates of doubling up
reached levels not seen since the aftermath of World War II.36
Relatively few poor households (26% of those with incomes below the
area median in 1993) receive federal housing subsidies.22 Under
these conditions, our results suggest, family homelessness will
endure."

Acknowledgments 

This research was supported by a contract from the New York City
Human Resources Administration and National Institute of Mental
Health research grant MH46116. 

AUTHORS

Marybeth Shinn, Daniela Stojanovic, and Susan James are with the
Department of Psychology, and Beth C. Weitzman and Lisa Duchon are
with the Wagner Graduate School of Public Service, New York
University. At the time this study was conducted, James R. Knickmnan
and Lucila Jimenez were with New York University. James R. Knickman
is now with The Robert Wood Johnson Foundation, Princeton, NJ, and
Lucila Jimenez is now with the Lincoln Medical and Mental Health
Center, Bronx, NY. David H. Krantz is with the Departments of
Psychology and Statistics, Columbia University, New York City. 

Requests for reprints should be sent to Beth C. Weitzman, PhD,
Wagner Graduate School of Public Service, New York University, 40 W
4th St, New York, NY 10003 (e-mail: weitzman@is2.nyu.edu). 

=======================



---------------------------------------------
J. David Hulchanski
     Dr. Chow Yei Ching Professor of Housing
     North American Editor, HOUSING STUDIES

Centre for Applied Social Research
Faculty of Social Work, University of Toronto
Toronto, Ontario, Canada  M5S 1A1

Tel 416 978-1973; Fax 416 978-7072
david.hulchanski@utoronto.ca
---------------------------------------------