Ethics in Implementing Change

Problem Solving Skills and Deficits Among Homeless Veterans With Serious Mental Illness

Sonya Gabrielian, Elizabeth Bromley, and Alison B. Hamilton

Veterans Affairs Greater Los Angeles, Los Angeles, California, and University of California, Los Angeles,

David Geffen School of Medicine

Van T. Vu University of California, San Francisco, School of

Medicine

Adrian Alexandrino Jr. Gap Solutions, Inc., Herndon, Virginia

Ella Koosis Veterans Affairs Greater Los Angeles, Los Angeles,

California

Alexander S. Young Veterans Affairs Greater Los Angeles, Los Angeles, California, and University of California, Los Angeles, David

Geffen School of Medicine

Few interventions train homeless consumers in housing-related independent living skills. To inform the development of such interventions for the Department of Veterans Affairs’ Supported Housing consumers with serious mental illness, we examined these consumers’ problem-solving skills and deficits. We performed semistructured interviews and cognitive tests with 20 con- sumers who retained housing for �1 year (“stayers”) and 20 consumers who lost housing in �1 year (“exiters”). Salient types of problems were identified in the qualitative data; we categorized problem-solving approaches by complexity level and identified differences in problem-solving complexity by consumers’ housing outcomes. Instrumental (e.g., money management), interper- sonal, and health-related problems were prominent in consumers’ narratives. Cognition was poor among stayers and exiters. Problem-solving approaches were highly relevant to day-to-day functioning in supported housing. There was a trend toward greater problem-solving complexity in stayers versus exiters. These data explore potential challenges faced in supported housing and help inform the development of a Veterans Affairs-based housing-focused skills training intervention.

Public Policy Relevance Statement Little is known about the problem-solving skills and deficits of formerly homeless consumers engaged in supported housing programs. To inform the development of skills training interventions for Veterans Affairs-supported housing participants, this study highlights this population’s salient types of problems and problem-solving approaches.

This article was published Online First July 16, 2018. Sonya Gabrielian, Elizabeth Bromley, and Alison B. Hamilton, Center

for the Study of Healthcare Innovation, Implementation, and Policy, Men- tal Illness Research, Education, and Clinical Center, Veterans Affairs Greater Los Angeles, Los Angeles, California, and Department of Psychi- atry and Biobehavioral Sciences, University of California, Los Angeles, David Geffen School of Medicine; Van T. Vu, Department of Family and Community Medicine, University of California, San Francisco, School of Medicine; Adrian Alexandrino, Jr., Gap Solutions Inc., Herndon, Virginia; Ella Koosis, Center for the Study of Healthcare Innovation, Implementa- tion, and Policy, Mental Illness Research, Education, and Clinical Center, Veterans Affairs Greater Los Angeles; Alexander S. Young, Center for the

Study of Healthcare Innovation, Implementation, and Policy, Mental Ill- ness Research, Education, and Clinical Center, Veterans Affairs Greater Los Angeles, and Department of Psychiatry and Biobehavioral Sciences, University of California, Los Angeles, David Geffen School of Medicine.

This research was supported by VA HSR&D PPO 13-154 (Principal investigator: Sonya Gabrielian). The authors thank Lillian Gelberg and David Smelson for their valuable contributions to this project.

Correspondence concerning this article should be addressed to Sonya Gabrielian, Mental Illness Research, Education, and Clinical Center, West Los Angeles Veterans Affairs Healthcare Center, 11301 Wilshire Boulevard, Building 210A, Los Angeles, CA 90073. E-mail: sonya.gabrielian@ va.gov

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

American Journal of Orthopsychiatry © 2018 Global Alliance for Behavioral Health and Social Justice 2019, Vol. 89, No. 2, 287–295

http://dx.doi.org/10.1037/ort0000340

287

T hough numerous interventions improve housing and healthfor homeless consumers—including supported housing(Tsemberis & Eisenberg, 2000), Assertive Community Treatment (Hwang & Burns, 2014) and substance use disorder services (Hwang & Burns, 2014)—few interventions train home- less consumers how to obtain and retain housing. Such services are particularly needed for homeless consumers with serious mental illness (SMI), who have cognitive impairments (Paquin, Wilson, Cellard, Lecomte, & Potvin, 2014; Pinkham, 2014; Wykes, Huddy, Cellard, McGurk, & Czobor, 2011) that can hinder the acquisition and maintenance of housing (Backer & Howard, 2007; Depp, Vella, Orff, & Twamley, 2015; Gabrielian et al., 2015; MacReady, 2009). A recent quantitative review revealed signifi- cant neurocognitive impairments among homeless adults, includ- ing a mean full-scale IQ score about one standard deviation below average; these impairments may impede housing rehabilitation efforts (Depp et al., 2015).

Yet, the relationships between neuropsychological assessment scores and real-world performance (ecological validity) are poorly understood (Spooner & Pachana, 2006). In particular, though day-to-day, real-world problem-solving skills have been studied in several populations, including SMI consumers (Bromley, Adams, & Brekke, 2012a) and the general population (Hartley, 1990), this construct is unexplored among homeless consumers with SMI. The Department of Veterans Affairs (VA) boasts a robust system of homeless services (Balshem, Christensen, Tuepker, & Kansagara, 2011) and is an ideal setting to inform intervention development by examining problem solving skills and deficits among homeless consumers with SMI.

Problem-solving therapy (PST) is a cognitive-behavioral inter- vention disseminated within the VA that trains consumers with mental illness in adaptive problem-solving skills. PST is an effec- tive treatment for depression, centered on the assumption that social problem solving is a core mediator and moderator between daily problems and overall functioning (Bell & D’Zurilla, 2009). Particularly in synergy with psychiatric rehabilitation (e.g., voca- tional training), problem-solving interventions effectively improve consumers’ cognition and functioning (e.g., employment, social adjustment, or quality of life; Wykes et al., 2011). However, housing outcomes are absent from the functional measures exam- ined in studies of these interventions, which have rarely been used with homeless consumers.

To inform the development of a VA-based housing-focused skills training intervention—to complement VA homeless services and improve housing attainment and retention—this study drew upon a conceptual model described by Fraser and colleagues (Fraser & Galinsky, 2010). This model depicts a five-step se- quence of intervention development, beginning with the develop- ment of problem and program theories (i.e., the identification of factors that—in this context—are relevant to housing attainment and retention). This initial step informs the development of inter- vention features, followed in turn by specification of program structures and processes (i.e., drafting the intervention, as well as fidelity and outcome measures), refining and confirming in effi- cacy tests (i.e., pilot testing of intervention components), testing effectiveness in practice settings (i.e., larger scale intervention testing), and the dissemination of program findings and materials.

To address the first of these steps—that is, to identify a breadth of factors relevant to housing attainment and retention—this study

used qualitative methods to explore the real-world challenges faced and problem-solving strategies used by VA-supported hous- ing consumers with SMI. We examined these consumers’ day-to- day challenges and categorized their problem-solving approaches by level of complexity, studying differences in problem-solving complexity by supported housing outcomes. We anticipated that better supported housing outcomes would be associated with higher levels of problem-solving complexity.

Method This study was conducted within the VA Supportive Housing

(VASH) program, the largest supported housing initiative in the nation (Pittman, 2013). Data were part of a larger study that aimed to identify factors associated with premature and unwanted exits from VASH (Gabrielian, Hamilton, Alexandrino, Hellemann, & Young, 2017). The parent study used mixed methods to compare demographics, diagnoses, and patterns of health service utilization associated with VASH exits; in contrast, the data presented here focus on challenges and problem-solving approaches described in the qualitative data.

For homeless consumers, VASH combines a financial subsidy for independent rental units with case management services, link- ing program participants to nonmandated mental health, addiction treatment, and medical services (Pittman, 2013). Here, interviews were performed with consumers in the VA Greater Los Angeles VASH program, which serves metropolitan Los Angeles and more homeless veterans than any VA in the nation (i.e., 6,375 Veterans with a history of homelessness). The VA Greater Los Angeles Institutional Review Board approved these study procedures (Study #004), and informed consent was obtained from all participants.

Participants

The parent study used mixed methods to compare consumers with a history of homelessness and SMI (defined broadly, includ- ing depression, bipolar disorder, anxiety disorders, and psychotic disorders; Petzel, 2012), who retained housing for at least 1 year (“stayers”) with those who prematurely their lost housing within 1 year of move-in (“exiters”). Those procedures are detailed else- where (Gabrielian et al., 2017). An online database that tracks use of VA homeless services (LaSalle, 2011)—the Homeless Opera- tions Management Evaluation System—was queried to identify consumers with SMI who received housing though the VA Greater Los Angeles VASH program in 2011 or 2012 (n � 1,558 stayers; n � 85 exiters). On a simple random sample of 85 stayers and all 85 exiters, we used the VA electronic medical record to abstract three variables associated with differential risk for experiencing homelessness (Byrne, Montgomery, & Dichter, 2013; Fargo et al., 2012; Hamilton, Poza, Hines, & Washington, 2012; Nelson, Au- bry, & Lafrance, 2007): age; gender; and the presence or absence of a psychotic disorder, bipolar disorder, or major depressive disorder with psychotic features. Seeking diversity across these three variables, we purposively sampled participants for semistruc- tured qualitative interviews. We enrolled 40 participants, including 20 stayers and 20 exiters. To enroll these 40 participants, we approached 116 consumers from the simple random sample about the data collection procedures.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

288 GABRIELIAN ET AL.

Data Collection

For all participants, demographic variables and housing history were abstracted from the VA electronic medical record. Diagnostic information was obtained from participants’ “problem list” (i.e., a list of diagnoses in the electronic medical record, by patient). Three cognitive tests were administered: the Hopkins Verbal Learning Test–Revised (HVLT-R; Brandt & Benedict, 2001), Letter–Number Span (LNS; Lichtenberger & Kaufman, 2009), and the Symbol Digit Modalities Test (SDMT; A. Smith, 1982). The HVLT-R assesses verbal learning and memory, the LNS measures working memory, and the SDMT captures motor and processing speeds, which is also hailed as one of the most sensitive tests for detecting cognitive dysfunction across a range of disorders (Dick- inson, Ramsey, & Gold, 2007; Joy, Fein, Kaplan, & Freedman, 2000). In short, these tests spanned a breadth of neurocognitive domains, but were selected for their brevity, as this data collection centered around the qualitative interviews.

All individual interviews (�45 min/each) were conducted by one of two study authors. The qualitative data collection was designed to elicit participants’ (n � 40) perceived needs while in VASH (Gabrielian et al., 2017). Exiters were asked to describe the circumstances leading to their housing loss; all participants were asked to describe problems faced during their VASH tenure and how they addressed these problems. All participants were queried for details about problem-solving strategies that they used to solve the problems they described: If an attempt to solve a problem failed, they were asked about alternate strategies used. Moreover, as social problems may be associated with housing outcomes (Gabrielian et al., 2015) and are a core component of PST and other problem-solving interventions (Bell & D’Zurilla, 2009), par- ticipants were specifically queried about interpersonal problems that arose while they were housed.

Analyses

The chi-square test and analysis of variance were used to de- termine how demographics, diagnoses, and cognitive test results varied by housing outcome. Age was considered a covariate in analyses of cognitive test scores. Analyses were performed using Stata software (Version 12.1; StataCorp, 2011).

All interviews were digitally recorded and professionally tran- scribed; written transcripts were checked against audio recordings for accuracy. Analyses were conducted using Atlas.ti (2017), a qualitative data analysis software program; at each stage in the analysis, two authors independently reviewed the transcripts, over- lapping in coding at least 10% of the interviews, comparing responses, reconciling disagreements, and discussing with other authors to refine the codebook until �70% agreement was reached (Campbell, Quincy, Osserman, & Pedersen, 2013).

First, to identify “problems” and participants’ responses to them— experienced in day-to-day life in supported housing— our analyses drew on concepts from the problem-solving code (PRO; Bromley, Mikesell, Mates, Smith, & Brekke, 2012b). Developed in a video ethnography study of individuals with schizophrenia living in the community, the PRO code translates spontaneous, naturalistic instances of everyday problem solving into measurable units of functional performance. Identifying real-world corollaries to tasks performed on validated measures of neurocognition (Kern

et al., 2011), behaviors that can be coded using PRO address challenges faced in everyday life, including managing novel cir- cumstances (e.g., turning on a new appliance), satisfying needs (e.g., preparing food when hungry), planning ahead (e.g., calcu- lating if there is time to stop at the store), or fixing complications (e.g., repairing a broken faucet). This definition of PRO behaviors was used to identify problems in the interview transcripts; the two coders iteratively compared identified problems, discussing rele- vant examples until consensus was reached. In these analyses, only problems actually addressed while in supportive housing were coded, as opposed to hypothetical scenarios that could arise for others (e.g., a participant hypothesizes that his or her peers might struggle with a problem, but did not experience this problem first-hand). Because the PRO code centers on an individual’s approach to challenges, problems that were described without any attempt at a solution— by the participant him/herself or someone (e.g., a case manager) he or she enlisted for help—were not coded. The final definition of this code was applied to all qualitative interviews (n � 40). We identified a total of 324 problems across the 40 interviews.

Second, to categorize these problems, a procedure developed by Berg and colleagues (Berg, Strough, Calderone, Sansone, & Weir, 1998; Blanchard-Fields, Mienaltowski, & Seay, 2007) was em- ployed as a top-level codebook, classifying each problem as in- strumental or interpersonal. Specifically, instrumental problems were scenarios in which individuals had difficulty achieving some- thing relevant to their personal life; these were situations in which participants were trying to accomplish or improve something. In contrast, interpersonal problems involved social concerns (i.e., problems arising when the participant tried to reach an outcome involving other people). This deductive codebook was subse- quently modified (Miles & Huberman, 1994), as health problems emerged inductively from the data (i.e., problems arising that were related to medical or mental health problems). Each coder inde- pendently developed a set of codes to identify subtypes of instru- mental, interpersonal, and health problems. Codes were discussed and refined; a constant comparative approach was used to link codes across and within interviews. To conceptualize a breadth of participants’ problems—to inform the development of future hous- ing skills interventions—two authors applied the entire dataset to the finalized codebook (classifying each problem as one or more subtype of instrumental, interpersonal, or health-related). We also tallied the frequency of each problem subtype.

Third, the PRO code was used to rate the type of problem- solving approaches used by participants, with coders assigning each problem one of three problem-solving strategies (Bromley et al., 2012b): (1) rote and rudimentary, reflecting straightforward solutions to problems (i.e., participants offer rote solutions without considering other options or solely depend on others to solve the problem); (2) anticipatory and additive, involving looking beyond the immediate problem when developing a solution (e.g., in re- sponse to a conflict, a participant considers the future conse- quences of any given solution, rather than a rote solution that works in the moment); and (3) complex and creative, weighing options and developing a multistep action plan to address the problem, often with elaborate preparation or creativity and flexibility in comparing solutions to the problem at hand. Prior work shows good (Cronbach’s � � 0.68) internal consistency

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

289PROBLEM SOLVING IN SUPPORTED HOUSING

reliability— or day-to-day stability— of PRO skill use (Bromley et al., 2012a).

With overlap between the two coders and discussion to refine the codes and resolve discrepancies, all identified problems were coded into one of these three levels. Of note, some problems do not require high-level problem-solving skills (Kimbler, 2013); as a proxy for participants’ highest order problem-solving abilities, individual participants were assigned a ranking corresponding to the highest PRO level coded in their interview transcript. The number of participants achieving each level was tallied and nar- rative examples were generated. A constant comparative approach was used to search for core similarities and differences among participants demonstrating each of the three levels. The chi-square test was used to determine if the highest PRO level differed between stayers and exiters.

Results

Sample Characteristics

Table 1 describes the sample. Demographic characteristics were not significantly different (p � .05) between stayers and exiters.

Prior to entering supportive housing, most participants were chron- ically homeless, (i.e., continuously homelessness for �1 year or four or more episodes of homelessness in the last 3 years; Office of the Assistant Secretary for Community Planning and Develop- ment, HUD, 2015). More exiters than stayers were chronically homeless. The sample had a range of mental health conditions, similar between the two groups. Most participants had one or more substance use disorders.

Cognitive scores were poor in both groups. HVLT-R had a mean raw score of 18.3, similar between groups; a score of 18 is equivalent to the 3.6% percentile in the general population (Nuechterlein & Green, 2006). SDMT had a mean raw score of 41.9, with a statistically significant (p � .01) between-groups difference between stayers (M � 42.8) and exiters (M � 41.0); normative samples of adults aged 45–54 with 12 years of education (similar to our sample means, as SDMT population data is presented by age range) have a mean and standard deviation of 47.3 and 9.6, respectively, indicating that our sample is about one half a standard deviation below the mean (A. Smith, 1982). LNS had a mean raw score of 12.0, also similar between groups, equivalent to the 11% percentile in the general population (Nuechterlein & Green, 2006).

Table 1. Sample Characteristics

Variable Stayers (n � 20) Exiters (n � 20) Total (N � 40) �2 F df P

Age in years (M, SD) 53.5, 7.0 51.2, 7.7 52.3, 7.4 1.04 1 .34 Gender (n, % male) 16, 80.0% 19, 95.0% 35, 87.5% 2.06 1 .15 Race/ethnicity (n, %) 3.14 2 .21

Hispanic 3, 15.0% 1, 5.0% 4, 10.0% Non-Hispanic White 2, 10.0% 6, 30.0% 8, 20.0% Non-Hispanic Black 15, 75.0% 13, 65.0% 27, 67.5%

Marital status (n, %) 1.53 3 .68 Never married 11, 55.0% 11, 55.0% 22, 55.0% Separated 3, 15.0% 2, 10.0% 5, 12.5% Divorced 5, 25.0% 7, 35.0% 12, 30.0% Widowed 1, 5.0% 0, .0% 1, 2.5%

Chronically homeless at VASH entry (n, %) 11, 55.0% 16, 80.0% 27, 67.5% 1.90 1 .09 VASH tenure in days (M, SD)b 137.0, 113.1 Psychiatric diagnoses (n, %)

Depressive disorders 8, 40.0% 7, 35.0% 15, 37.5% .11 1 .74 Bipolar disorder 2, 10.0% 5, 25.0% 7, 17.5% 1.56 1 .21 Psychotic disorders 7, 35.0% 7, 35.0% 14, 35.0% .00 1 1.00 Posttraumatic stress disorder (PTSD) 6, 30.0% 3, 15.0% 9, 22.5% 1.29 1 .26 Other anxiety disorder (not PTSD) 1, 5.0% 0, .0% 1, 2.5% 1.03 1 .31

Substance use disorders (n, %) Alcohol use disorder 11, 55.0% 12, 60.0% 23, 57.5% .10 1 .75 Drug use disorder 13, 65.0% 16, 80.0% 29, 72.5% 1.13 1 .29

Cognitive test scores (raw scores) Hopkins Verbal Learning Test–Revised 18.3, 5.4 18.2, 4.1 18.3, 4.7 1.35 2 .27a

Symbol Digit Modalities Test 42.8, 9.5 41.0, 9.7 41.9, 9.5 7.74 2 .00a�

Letter–Number Span 12.1, 3.1 12.0, 3.8 12.0, 3.4 1.82 2 .18a

PRO (highest level achieved; n, %) 3.18 2 .20 1: Rote and rudimentary 6, 30.0% 10, 50.0% 16, 40.0% 2: Anticipatory and additive 12, 60.0% 10, 50.0% 22, 55.5% 3: Complex and creative 2, 10.0% 0, .0% 2, 5.0%

Note. Defined as continuously homelessness for �1 year or four or more episodes of homelessness in the last 3 years. VASH � Veterans Affairs Supportive Housing; PRO � problem-solving code (see Bromley, Mikesell, Mates, Smith, & Brekke, 2012b). a Age was used as a covariate. b VASH tenure was only calculated for exiters, as stayers were housed at the time of this study. � p � .05.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

290 GABRIELIAN ET AL.

The following text describes and tallies the salient types of problems faced by these participants. Exemplar problem-solving strategies of these participants are also presented, stratified by their highest PRO ranking.

Problems

Table 2 displays the types of problems described by participants. Though each individual problem (n � 324 across 40 interviews) was categorized as instrumental, interpersonal, or health-related, 23 problems were coded into two relevant subtypes (e.g., mental health and substance use disorders). As such, the total number of problem subtypes was 347.

Over one fifth (n � 76, 21.9%) of problems faced by partici- pants were related to acquiring an apartment, including signing up for supported housing services, searching for an apartment, filling out applications, and signing a lease. Financial/money manage- ment problems were the second most common (n � 47, 13.5%) subtype, for example, having difficulty budgeting to pay for rent and groceries. Most financial problems reflected money manage- ment challenges, though resource limitations (i.e., inadequate funds) were often intertwined. Other instrumental problems en- compassed daily needs of apartment living (n � 35, 10.1%), such as acquiring food, clothing, furniture, and cookware; the rental process (n � 27, 7.8%), which included processes and paperwork needed to retain or leave an apartment (e.g., addressing threats of eviction, getting loans when behind on rent, and working with the local housing authority); vocational problems (10, 2.9%; i.e., re- lated to work and/or school); and legal problems (8, 2.3%; i.e., related to criminal justice system involvement).

Most interpersonal problems were related to conflicts (31, 8.9%), including direct interpersonal disagreements (verbal and/or physical) and discrimination/stigma experiences. Fewer (18, 5.2%) problems were related to needs for greater support (e.g., secondary to social isolation or loneliness) and housing communication (20, 5.8%; i.e., needing to convey information about an apartment (e.g., a leaky faucet) to a landlord or property manager). The most common health problem was related to substance use disorders

(36, 10.4%), followed by mental health problems (22, 6.3%), and physical health problems (17, 4.9%).

Problem Solving

Rote and rudimentary. More than one third (n � 15, 37.5%) of participants only employed rote and rudimentary problem-solving approaches. Their approach to finding an apart- ment was concrete; often, participants simply agreed to the first apartment they saw, without weighing alternatives or getting input from others. Others relied entirely on staff to find them an apart- ment, without exploring other alternatives. As one participant described, “I looked in [a newspaper] . . . it had a big ad right there in it, ‘Will accept VASH vouchers’ . . . I seen one that I liked and he basically gave it to me on the spot.” Another participant stated the following:

[My case manager] helped me get it. They pretty much were saving it for me to move in . . . [My case manager] worked with the manage- ment to give me one of their apartments. They gave me the informa- tion and I just took a bus to go to it.

After moving into their apartments, these participants often relied entirely on their case managers to address their everyday needs. One participant described the experience as follows:

[My case manager] helped me with phone services, tutor services [for] my reading. She guided me to everything that I needed as a civilian. She helped me [turn on the] gas. . . . She pretty much walked me through every part of transitioning from living in the streets to . . . living at my own home.

Often participants were so removed from these processes that they could not describe the services used by their case managers. As one participant said, “This other program, they helped me get down payment for that place . . . I know I recently got assistance with my electric bill. I don’t know [which program it is] . . . my worker got it for me.”

Despite this heavy reliance on case management staff, these participants often struggled to use case management to meet their social needs. For example, one participant described feelings of loneliness, wanting “companionship. . . . I was by myself [in my apartment].” Yet, when asked about services he could receive from his case manager, he stated, “She would tell me where food banks was.” Another participant described profound social isolation, naming his case manager as “the only one I talk to.” However, when asked to describe ways his case manager could help improve his social network, he said, “I don’t have no specifics, but every month I look forward to [the case manager’s visit].” He seemingly enjoyed his case manager’s company, but did not receive resources to address his social disconnection.

Anticipatory and additive. More than half (n � 23, 57.5%) of participants described at least one example of anticipa- tory and additive skills. These participants weighed the pros and cons of apartment options, thinking about neighborhood factors and safety concerns. One participant stated the following:

I had to look up a lot of places, and a lot of them I wouldn’t live there if I had all the money, if it was free . . . I went to a few of them . . . where gangbangers are prevalent and crack is everywhere, heroin, you name it . . . I eventually found my place which was clean and whatnot,

Table 2. Subtypes of Problems Addressed by Homeless- Experienced Consumers

Problem Frequency (N � 347)

Instrumental (n, %) 203, 58.5% Apartment acquisition 76, 21.9% Financial 47, 13.5% Daily needs of apartment living 35, 10.1% Rental process 27, 7.8% Vocational 10, 2.9% Legal 8, 2.3%

Interpersonal (n, %) 69, 19.9% Conflicts 31, 8.9% Supportive 18, 5.2% Housing communication 20, 5.8%

Health (n, %) 75, 21.6% Substance use disorders 36, 10.4% Mental health 22, 6.3% Physical health 17, 4.9%

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

291PROBLEM SOLVING IN SUPPORTED HOUSING

and the neighborhood was a little better than most of the ones I had seen.

Another participant reported that his daily transportation needs were a critical consideration in his apartment search:

I was looking at a 10-mile radius, I don’t have transportation, I have a bicycle . . . I [looked] online, the newspaper . . . other websites. I had a list of places I looked at over a year, and it came up to about almost 150 places.

When approaching a problem or need while housed, these partic- ipants were able to brainstorm potential solutions without sole reliance on staff supports. For example, one participant described developing a budget and saving to purchase furniture for his apartment; he said, “I started saving up, and I would just buy like a small coffee table . . . I would go either to Goodwill or Salvation Army. They have reasonable prices, and their furniture is pretty good.” This contrasted from the rote and rudimentary participants that reflexively turned to staff for resources to address apartment- related needs. Similarly, another participant described wanting to move but struggling to find a landlord who would accept a tenant enrolled in a supportive housing program. He developed a strategy to address this problem, saying, “Well, talk to the person. Don’t tell them you’re [in the VASH program] until you can create a rapport with them, then see what happens.”

Complex and creative. Only 2 (5.0%) participants dem- onstrated complex and creative problem-solving skills. Their ap- proach to acquiring housing paralleled their peers with anticipatory and additive skills. However, beyond planning ahead and consid- ering consequences of decisions, they demonstrated resilience when faced with obstacles, often simultaneously pursuing multiple strategies to address problems. For example, one participant de- scribed his multifaceted employment search:

I remember days where I used to get up in the morning . . . to be at the Employment Development computer station. I would motivate myself to go . . . they show you how to make resumes according to your skills, and they have daily postings . . . I [also] tried going into the street from business to business, looking for work . . . now it’s either online or take the forms and mail them.

Similarly, the other participant described moving from one apartment to another (both under the auspices of the supported housing program). He described persevering despite repeated lo- gistical obstacles, “going back and forth to VASH, making phone calls” and taking the initiative to request, in advance, an extension to complete paperwork needed to make the move. He stated the following:

The process was very long, I stayed on top of it. I just don’t see how anybody could not stay on top of it . . . without my case manager. She has a lot of other people too and I need stuff done when I need stuff done. I just cannot wait around on other people to do stuff for me.

Between-Group Comparisons

Table 1 includes the highest PRO level for participants by housing outcome. Though there were no statistically significant between-groups differences (p � .20), fewer stayers than exiters (30.0%/50.0%) only used rote and rudimentary skills. More stayers than exiters (60.0%/50.0%) demonstrated anticipatory and additive

problem solving, and both (n � 2, 10%) examples of complex and creative skills were stayers.

Discussion These findings explore the breadth of challenges and problem-

solving approaches employed by homeless consumers with SMI enrolled in VA’s supported housing program. Instrumental prob- lems (often surrounding apartment acquisition and money man- agement); interpersonal problems (particularly conflicts with others); and health-related problems (especially substance use disorders) were salient in participants’ narratives. Moreover, problem-solving approaches (anticipating consequences of deci- sions, weighing the risks and benefits of multiple solutions to a problem, and maintaining resilience in the face of adversity) ap- peared highly relevant to day-to-day functioning in supported housing.

Prior research describes the challenges of transitioning from homelessness to independent housing (Gabrielian et al., 2017). Within VA, an average of 113 days passes between program enrollment and apartment move-in (O’Connell, Kasprow, & Rosenheck, 2010); this interval is an ideal time for housing-related skills training. For example, processes of apartment acquisition varied significantly between the three PRO groups, often resulting in the rote and rudimentary group reflexively accepting apartments in less than ideal neighborhoods, with little concern for factors like safety. Interventionists could detail the processes of an effective apartment search, such as questions to ask landlords, factors to consider about the neighborhood, and proximity to public trans- portation. Homeless consumers with SMI could role-play these skills with peers, with interventionist input, in preparation for their actual apartment search.

Similarly, rather than solely relying on case management staff for financial concerns that arise after achieving housing, person- alized budgets could be developed with consumers (Hough & Rice, 2010), considering additional expenses (e.g., rent and utili- ties) that are likely to arise once housed. For consumers who have fiduciaries and thus less control over their budgets, conversation and assertiveness skills training—adapted from evidence-based social skills training interventions (Bellack, Mueser, Gingerich, & Agresta, 2004; Liberman et al., 1993)— could facilitate more ef- fective communication and better satisfaction in money manage- ment negotiations. Similarly, to address interpersonal conflicts, conflict management skills training interventions (Bellack et al., 2004; Kopelowicz, Liberman, & Zarate, 2006) could be adapted for homeless consumers with SMI; social skills training interven- tions include modules relevant to health maintenance (e.g., com- municating with health care providers) and could influence problem-solving strategies for health-related problems.

For this vulnerable population, quantitative findings on cogni- tive tests are useful in speculating about the relationships between cognition and housing retention (Backer & Howard, 2007; Gabri- elian et al., 2015; MacReady, 2009; Spence, Stevens, & Parks, 2004). Here, similar to our previous work (Gabrielian et al., 2015), the symbol digit modalities test was significantly associated with housing outcomes; this measure reflects cognitive processing speed, but also employs multiple processes (i.e., perception, work- ing memory, attention, and visuomotor coordination; van Hoof, Jogems-Kosterman, Sabbe, Zitman, & Hulstijn, 1998) and is a

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

292 GABRIELIAN ET AL.

very sensitive test for detecting cognitive dysfunction (Dickinson et al., 2007; Joy et al., 2000; Kern et al., 2011). The between- groups differences seen in SDMT scores suggest the influence of processing speed and other cognitively mediated processes in supported housing retention.

In these qualitative data, the highest achieved PRO was not statistically significantly associated with housing outcomes. How- ever, our qualitative analyses suggest a trend toward greater problem-sophistication in stayers compared to exiters, which aligns with the between-groups differences seen in SDMT scores and further suggests that cognition may be relevant for supported housing outcomes. Of note, some participants encountered more complex problems than others; the highest PRO rating for any given participant is limited by the most complex problem-solving approach that he or she needed to use while housed.

Of note, numerous social and environmental factors interact with cognition to affect problem-solving skills and housing out- comes. For example, experiences of poverty, social isolation, and stigma may contribute to consumers’ learned helplessness (Dixon, 2011), which may manifest as a willingness to accept any available housing option, regardless of quality. Moreover, a lack of mone- tary and social resources can lead to eviction from supported housing, regardless of problem-solving skills. Recognizing the deep connections between person-level variables and social and environmental factors, the greatest value of these data lie in pre- senting a breadth of real-world challenges and problem-solving strategies used by the VA-supported housing consumers with SMI.

To better conceptualize the relationships between cognition and real-world behavior, future research should include more comprehensive cognitive testing (Kern et al., 2011; Zaki, Bolger, & Ochsner, 2008; across a breadth of neurocognitive domains and social cognition), triangulating these data with narratives of cognitive-mediated responses to a standardized set of problems. Performance-based assessments of functional capacity in problem subtypes identified in these data (e.g., money management) may also prove useful in future studies. Returning to the paradigm of Fraser and colleagues (Fraser & Galinsky, 2010), additional spec- ification of problem and program theories (i.e., an improved un- derstanding of the real-world consequences of cognitive deficits and their interplay with social and environmental factors) will facilitate better tailoring of a skills intervention to the distinct needs of homeless consumers with SMI.

Limitations

First, though we tallied the relative frequencies of described problems, these counts may reflect participants’ specific circum- stances, as opposed to the relative importance of described prob- lems. As opposed to quantifying the relative importance of prob- lems faced in supported housing, these analyses aimed capture the breadth of common problems faced by SMI consumers in these settings.

Second, though problem-solving is a high-order process em- ploying multiple cognitive domains (Wang & Chiew, 2010), these qualitative data must be considered alongside findings from psy- chometrically valid and objective measures of cognition in persons with mental health problems (Kern et al., 2011; Zaki et al., 2008), including more detailed cognitive assessments (Kern et al., 2011; Zaki et al., 2008). We selected three objective cognitive assess-

ments for their breadth and brevity, but did not include a formal measure of executive functioning, which may be more directly correlated with our qualitative narratives. In addition, each partic- ipant’s highest order problem-solving approach was limited by his or her most complex problem, along with external factors, for example, the skills set of a given case manager. For example, a participant with a very skilled case manager may actually be employing higher order problem solving to rely on this individual. Future research could ask participants to narrate problem-solving approaches to a predetermined list of problems (for which complex and creative problem solving is possible), with less variability in external factors, for example, available supports.

Third, PRO was developed in a community-based sample of persons with psychotic disorders; this study used a broader defi- nition of SMI that encompassed major mood and anxiety disorders (Petzel, 2012), reflective of the population that engages in sup- ported housing initiatives (Gabrielian, Yuan, Andersen, & Gel- berg, 2016). However, these analytic processes facilitated a nu- anced understanding of PRO that derived from these qualitative data of homeless participants with SMI.

Last, these data were collected from predominantly male Vet- erans with SMI and a history of homelessness in Los Angeles County who were entering supported housing. Their problems— and problem-solving strategies—may differ from homeless per- sons with SMI engaged in disparate rehabilitation efforts (e.g., residential rehabilitation programs) or persons of different back- grounds (e.g., nonveterans or a sample with more women). Con- sumers in a less resource-intense environment (e.g., rural commu- nities) may face additional obstacles in the pursuit of permanent supported housing. In addition, diagnoses presented for this sam- ple were abstracted from the VA medical record, as opposed to standardized clinical interviews that more accurately identify di- agnoses using strict clinical criteria.

Conclusions For consumers with SMI engaged in the VA-supported housing

program, the transition from homelessness to independent housing is laden with challenges. For this vulnerable population, these data suggest the relevance of problem-solving skills and cognition, which interplay with social and environmental factors, for exam- ple, poverty or stigma. Our findings identify major challenges faced in the VA-supported housing program that can inform the development of a housing skills-training intervention in this set- ting. With these data in hand, next steps include the actual spec- ification of the intervention structure and processes; future re- search includes implementation work to tailor, adapt, and evaluate effective interventions for stated challenges (e.g., money manage- ment, social problem solving, communication skills) to this vul- nerable population. Many psychiatric rehabilitation strategies em- ploy social learning-as opposed to neurocognitive-paradigms to address problem-solving skills (Silverstein, 2000). Complement- ing these approaches with cognitive remediation strategies, partic- ularly “compensatory” approaches that compensate for cognitive impairments with aids or by recruiting intact areas of cognition, may provide some benefits. (Leshner, Tom, & Kern, 2013) Though such interventions have precedent in SMI populations (Kopelowicz et al., 2006; Leshner et al., 2013), they are uncom-

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

293PROBLEM SOLVING IN SUPPORTED HOUSING

monly implemented in housing services and hold potential to improve outcomes in supported housing initiatives.

Keywords: homelessness; problem solving; cognition; veterans

References Atlas.ti. (2017). Atlas.ti (Version 7.5). [Computer software]. Retrieved

from http://downloads.atlasti.com/docs/whatsnew7.pdf?_ga�2 .246218564.131834937.1524583362-353849337.1524583362

Backer, T. E., & Howard, E. A. (2007). Cognitive impairments and the prevention of homelessness: Research and practice review. The Journal of Primary Prevention, 28(3– 4), 375–388. http://dx.doi.org/10.1007/ s10935-007-0100-1

Balshem, H., Christensen, V., Tuepker, A., & Kansagara, D. (2011). A critical review of the literature regarding homelessness among veterans. VA-ESP Project #05–225, 1– 64.

Bell, A. C., & D’Zurilla, T. J. (2009). Problem-solving therapy for depres- sion: A meta-analysis. Clinical Psychology Review, 29, 348 –353. http:// dx.doi.org/10.1016/j.cpr.2009.02.003

Bellack, A. S., Mueser, K. T., Gingerich, S., & Agresta, J. (2004). Social skills training for schizophrenia (2nd ed.). New York, NY: Guilford Press.

Berg, C. A., Strough, J. N., Calderone, K. S., Sansone, C., & Weir, C. (1998). The role of problem definitions in understanding age and context effects on strategies for solving everyday problems. Psychology and Aging, 13, 29 – 44. http://dx.doi.org/10.1037/0882-7974.13.1.29

Blanchard-Fields, F., Mienaltowski, A., & Seay, R. B. (2007). Age differ- ences in everyday problem-solving effectiveness: Older adults select more effective strategies for interpersonal problems. The Journals of Gerontology Series B, Psychological Sciences and Social Sciences, 62, 61– 64. http://dx.doi.org/10.1093/geronb/62.1.P61

Brandt, J., & Benedict, H. R. (2001). Hopkins Verbal Learning Test— Revised: Professional manual. Lutz, FL: Psychological Assessment Resources.

Bromley, E., Adams, G. F., & Brekke, J. S. (2012a). A video ethnography approach for linking naturalistic behaviors to research constructs of neurocognition in schizophrenia. The Journal of Neuropsychiatry and Clinical Neurosciences, 24, 125–140. http://dx.doi.org/10.1176/appi .neuropsych.11080201

Bromley, E., Mikesell, L., Mates, A., Smith, M., & Brekke, J. S. (2012b). A video ethnography approach to assessing the ecological validity of neurocognitive and functional measures in severe mental illness: Results from a feasibility study. Schizophrenia Bulletin, 38, 981–991. http://dx .doi.org/10.1093/schbul/sbr002

Byrne, T., Montgomery, A. E., & Dichter, M. E. (2013). Homelessness among female veterans: A systematic review of the literature. Women & Health, 53, 572–596. http://dx.doi.org/10.1080/03630242.2013.817504

Campbell, J. L., Quincy, C., Osserman, J., & Pedersen, O. K. (2013). Coding In-depth semistructured interviews: Problems of unitization and intercoder reliability and agreement. Sociological Methods & Research, 42, 294 –320. http://dx.doi.org/10.1177/0049124113500475

Depp, C. A., Vella, L., Orff, H. J., & Twamley, E. W. (2015). A quanti- tative review of cognitive functioning in homeless adults. Journal of Nervous and Mental Disease, 203, 126 –131. http://dx.doi.org/10.1097/ NMD.0000000000000248

Dickinson, D., Ramsey, M. E., & Gold, J. M. (2007). Overlooking the obvious: A meta-analytic comparison of digit symbol coding tasks and other cognitive measures in schizophrenia. Archives of General Psychi- atry, 64, 532–542. http://dx.doi.org/10.1001/archpsyc.64.5.532

Dixon, J. (2011). Existential poverty: Welfare dependency, learned helpe- lessness and psychological capital. Povery and Public Policy, 3, 279 – 298. http://dx.doi.org/10.2202/1944-2858.1158

Fargo, J., Metraux, S., Byrne, T., Munley, E., Montgomer, A. E., Jones, H., . . . Culhane, D. (2012). Prevalence and risk of homelessness among

U.S. veterans. Preventing Chronic Disease: Public Health Research, Practice and Policy. Advance online publication. http://dx.doi.org/10 .5888/pcd9.110112

Fraser, M. W., & Galinsky, M. J. (2010). Steps in intervention research: Designing and developing social programs. Research on Social Work Practice, 20, 459 – 466. http://dx.doi.org/10.1177/1049731509358424

Gabrielian, S., Bromley, E., Hellemann, G. S., Kern, R. S., Goldenson, N. I., Danley, M. E., & Young, A. S. (2015). Factors affecting exits from homelessness among persons with serious mental illness and substance use disorders. The Journal of Clinical Psychiatry, 76, e469 – e476. http://dx.doi.org/10.4088/JCP.14m09229

Gabrielian, S., Hamilton, A. B., Alexandrino, A., Hellemann, G., & Young, A. S. (2017). “They’re homeless in a home”: Retaining homeless- experienced consumers in supported housing. [in Press]. Psychological Services, 14, 154 –166. http://dx.doi.org/10.1037/ser0000119

Gabrielian, S., Yuan, A. H., Andersen, R. M., & Gelberg, L. (2016). Diagnoses treated in ambulatory care among homeless-experienced veterans: Does supported housing matter? Journal of Primary Care & Community Health, 7, 281–287. http://dx.doi.org/10.1177/ 2150131916656009

Hamilton, A. B., Poza, I., Hines, V., & Washington, D. L. (2012). Barriers to psychosocial services among homeless women veterans. Journal of Social Work Practice in the Addictions, 12, 52– 68. http://dx.doi.org/10 .1080/1533256X.2012.647584

Hartley, A. A. (1990). The cognitive ecology of problem solving. In L. W. Poon, D. C. Rubin, & G. A. Wilson (Eds.), Everday cognition in adulthood and late life (pp. 300 –329). New York, NY: Cambridge University Press.

Office of the Assistant Secretary for Community Planning and Develop- ment, HUD. (2015). Homeless emergency assistance and rapid transi- tion to housing: Defining “chronically homeless.” Retrieved from https://www.federalregister.gov/documents/2015/12/04/2015-30473/ homeless-emergency-assistance-and-rapid-transition-to-housing- defining-chronically-homeless

Hough, J., & Rice, B. (2010). Providing personalised support to rough sleepers (pp. 1– 64). York, UK: Joseph Rowntree Foundation. Retrieved from http://www.jrf.org.uk

Hwang, S. W., & Burns, T. (2014). Health interventions for people who are homeless. Lancet, 384, 1541–1547. http://dx.doi.org/10.1016/S0140- 6736(14)61133-8

Joy, S., Fein, D., Kaplan, E., & Freedman, M. (2000). Speed and memory in WAIS-R-NI Digit Symbol performance among healthy older adults. Journal of the International Neuropsychological Society, 6, 770 –780. http://dx.doi.org/10.1017/S1355617700677044

Kern, R. S., Gold, J. M., Dickinson, D., Green, M. F., Nuechterlein, K. H., Baade, L. E., . . . Marder, S. R. (2011). The MCCB impairment profile for schizophrenia outpatients: Results from the MATRICS psychometric and standardization study. Schizophrenia Research, 126(1–3), 124 –131. http://dx.doi.org/10.1016/j.schres.2010.11.008

Kimbler, K. J. (2013). Everyday problem solving and instrumental activ- ities of daily living: Support for domain specificity. Behavioral Sci- ences, 3, 170 –191. http://dx.doi.org/10.3390/bs3010170

Kopelowicz, A., Liberman, R. P., & Zarate, R. (2006). Recent advances in social skills training for schizophrenia. Schizophrenia Bulletin, 32(Suppl. 1), S12–S23. http://dx.doi.org/10.1093/schbul/sbl023

LaSalle, J. L. (2011). Homeless Operations Management and Evaluation System (HOMES): User manual—Phase 1 (pp. 1–200). Retrieved from http://www.vfwsc.org/homes.pdf

Leshner, A. F., Tom, S. R., & Kern, R. S. (2013). Errorless learning and social problem solving ability in schizophrenia: An examination of the compensatory effects of training. Psychiatry Research, 206, 1–7. http:// dx.doi.org/10.1016/j.psychres.2012.10.007

Liberman, R. P., Wallace, C. J., Blackwell, G., Eckman, T. A., Vaccaro, J. V., & Kuehnel, T. G. (1993). Innovations in skills training for the

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

294 GABRIELIAN ET AL.

seriously mentally ill: The UCLA social and independent living skills modules. Innovations and Research, 2, 43–59.

Lichtenberger, E. O., & Kaufman, A. S. (2009). Essentials of WAIS-IV assessment (6th ed.). Hoboken, NJ: Wiley.

MacReady, N. (2009). Neurological deficits in the homeless: A downward spiral. Lancet Neurology, 8, 228 –229. http://dx.doi.org/10.1016/S1474- 4422(09)70033-X

Miles, M. B., & Huberman, A. (1994). Qualitative data analysis: An expanded source book. Thousand Oaks, CA: SAGE.

Nelson, G., Aubry, T., & Lafrance, A. (2007). A review of the literature on the effectiveness of housing and support, assertive community treat- ment, and intensive case management interventions for persons with mental illness who have been homeless. American Journal of Orthopsy- chiatry, 77, 350 –361. http://dx.doi.org/10.1037/0002-9432.77.3.350

Nuechterlein, K. H., & Green, M. F. (2006). MATRICS (Measurement and Treatment Research to Improve Cognition in Schizophrenia) Consensus Cog- nitive Battery (MCCB) manual. Los Angeles, CA: MATRICS Assessment.

O’Connell, M., Kasprow, W., & Rosenheck, R. A. (2010). National dis- semination of supported housing in the VA: Model adherence versus model modification. Psychiatric Rehabilitation Journal, 33, 308 –319. http://dx.doi.org/10.2975/33.4.2010.308.319

Paquin, K., Wilson, A. L., Cellard, C., Lecomte, T., & Potvin, S. (2014). A systematic review on improving cognition in schizophrenia: Which is the more commonly used type of training, practice or strategy learning? BMC Psychiatry, 14, 139. http://dx.doi.org/10.1186/1471-244X-14-139

Petzel, R. A. (2012). Re-engaging veterans with serious mental illness in treatment. (pp1–5). Washington, DC: Department of Veterans Affairs, Veterans Health Administration. Retrieved from http://www.va.gov/ vhapublications/ViewPublication.asp?pub_ID�2476

Pinkham, A. E. (2014). Social cognition in schizophrenia. The Journal of Clinical Psychiatry, 75(Suppl. 2), 14 –19. http://dx.doi.org/10.4088/JCP .13065su1.04

Pittman, T. (2013). Department of Housing and Urban Development— Department of Veterans Affairs Supportive Housing (pp. 1–25). Re- trieved from http://www.nchv.org

Silverstein, S. M. (2000). Psychiatric rehabilitation of schizophrenia: Un- resolved issues, current trends, and future directions. Applied & Pre- ventive Psychology, 9, 227–247. http://dx.doi.org/10.1016/S0962-1849 (00)80002-5

Smith, A. (1982). Symbol Digits Modalities Test. Los Angeles, CA: West- ern Psychological Services.

Spence, S., Stevens, R., & Parks, R. (2004). Cognitive dysfunction in homeless adults: A systematic review. Journal of the Royal Society of Medicine, 97, 375–379. http://dx.doi.org/10.1177/0141076804097 00804

Spooner, D. M., & Pachana, N. A. (2006). Ecological validity in neuropsychological assessment: A case for greater consideration in research with neurologically intact populations. Archives of Clinical Neuropsychology, 21, 327–337. http://dx.doi.org/10.1016/j.acn.2006.04 .004

StataCorp. (2011). Stata statistical software (Release 12). College Station, TX: Author.

Tsemberis, S., & Eisenberg, R. F. (2000). Pathways to housing: Supported housing for street-dwelling homeless individuals with psychiatric dis- abilities. Psychiatric Services, 51, 487– 493. http://dx.doi.org/10.1176/ appi.ps.51.4.487

van Hoof, J. J., Jogems-Kosterman, B. J., Sabbe, B. G., Zitman, F. G., & Hulstijn, W. (1998). Differentiation of cognitive and motor slowing in the Digit Symbol Test (DST): Differences between depression and schizophrenia. Journal of Psychiatric Research, 32, 99 –103. http://dx .doi.org/10.1016/S0022-3956(98)00057-0

Wang, Y., & Chiew, V. (2010). On the cognitive process of human problem solving. Cognitive Systems Research, 11, 81–92. http://dx.doi .org/10.1016/j.cogsys.2008.08.003

Wykes, T., Huddy, V., Cellard, C., McGurk, S. R., & Czobor, P. (2011). A meta-analysis of cognitive remediation for schizophrenia: Methodology and effect sizes. The American Journal of Psychiatry, 168, 472– 485. http://dx.doi.org/10.1176/appi.ajp.2010.10060855

Zaki, J., Bolger, N., & Ochsner, K. (2008). It takes two: The interpersonal nature of empathic accuracy. Psychological Science, 19, 399 – 404. http://dx.doi.org/10.1111/j.1467-9280.2008.02099.x

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

295PROBLEM SOLVING IN SUPPORTED HOUSING