The Independent Living Scales as a Measure of Functional Outcome for Schizophrenia

Article (PDF Available)inPsychiatric Services 55(9):1052-4 · October 2004with1,003 Reads
DOI: 10.1176/ · Source: PubMed
The Independent Living Scales (ILS) measures cognitive skills required for independent living and is intended to provide guidelines for appropriate supervision requirements for persons in residential placement. To assess the validity of the ILS among persons with schizophrenia, the instrument was administered to 162 individuals with schizophrenia who were living in three gradations of care: maximum supervision, moderate supervision, and minimal supervision. Scores on the ILS differed significantly across the three levels of care, whereas scores on the Global Assessment of Functioning (GAF) that were provided by clinicians discriminated only two levels of care. The ILS can be used among patients with schizophrenia to measure cognition as it affects functional outcome.
PSYCHIATRIC SERVICES September 2004 Vol. 55 No. 9
The Independent Living Scales
(ILS) measures cognitive skills
required for independent living
and is intended to provide guide-
lines for appropriate supervision
requirements for persons in resi-
dential placement. To assess the
validity of the ILS among per-
sons with schizophrenia, the in-
strument was administered to
162 individuals with schizophre-
nia who were living in three gra-
dations of care: maximum super-
vision, moderate supervision,
and minimal supervision. Scores
on the ILS differed significantly
across the three levels of care,
whereas scores on the Global As-
sessment of Functioning (GAF)
that were provided by clinicians
discriminated only two levels of
care. The ILS can be used among
patients with schizophrenia to
measure cognition as it affects
functional outcome. (Psychiatric
Services 55:1052–1054, 2004)
t is now widely accepted that cog-
nitive impairment in the areas of
memory, attention, and problem solv-
ing influences functional outcome
among patients with schizophrenia,
accounting for 20 to 60 percent of the
variance in successful psychosocial
rehabilitation, social problem-solving
ability, or community living (1,2). Be-
cause cognition plays such a signifi-
cant role in schizophrenia, clinicians
and researchers need assessment
tools that measure cognition as it af-
fects functional outcome. Several
questionnaires assess performance of
independent living skills on the basis
of information obtained from inform-
ants, self-report, or observation of
simulated everyday functioning tasks
(3–6). Although these measures of
functional capacity discriminate out-
patients from control subjects, they
may not be applicable to inpatients,
and they do not link cognitive skills to
functional ability.
Assessments that focus on skills do
not take cognitive functioning into ac-
count. For example, one can observe
the skill of how a utility bill is paid,
but if the importance of paying utility
bills is not appreciated, or if the per-
son forgets about the bill, payment
will not be sent. Similarly, if an indi-
vidual lacks initiative to perform the
task without the structured demands
of clinical observation, there is no
functional application in a real-life
The Independent Living Scales
(ILS) (7) assesses cognition as it af-
fects daily functioning, but it has not
been widely used in psychiatric popu-
lations. We evaluated whether the
ILS was good at predicting the living
status of inpatients and outpatients
with schizophrenia. We also com-
pared the instrument with the famil-
iar clinician-rated Global Assessment
of Functioning (GAF) to determine
whether the cognitive problem-solv-
ing measure would discriminate func-
tional levels better than a clinically
driven global measure.
The study participants were from
outpatient and chronic inpatient set-
tings in New York City, carried a
DSM-IV axis I diagnosis of schizo-
phrenia or schizoaffective disorder,
and were aged 18 to 55 years. The in-
patients were potential research par-
ticipants in a previous study, recruit-
ed from September 1996 to August
1998 (8). The outpatients were re-
cruited from FEGS (Federation
Employment and Guidance Ser-
vices) from June 1998 to November
1998. Individuals were excluded
from participation if they did not
speak English, had a neurologic or
serious medical disability, had severe
behavioral disturbances, or had an
IQ lower than 70.
The ILS assesses the likelihood of
successful independent community
living. The instrument, which was
originally developed for adults with
dementia, has been used to estimate
the competence of adults with a di-
agnosis of a psychiatric illness, in-
cluding schizophrenia.
The ILS has five subscales and two
factor-analyzed subscales. “Memo-
ry–orientation” contains items that
include orientation to time and
place, recall of a brief shopping list,
and recognition of a missing object.
“Managing money” includes con-
crete tasks designed for monetary
calculations and budgetary precau-
The Independent Living Scales
as a Measure of Functional
Outcome for Schizophrenia
NNaaddiinnee RReevvhheeiimm,, PPhh..DD..
AAlliiccee MMeeddaalliiaa,, PPhh..DD..
Dr. Revheim is affiliated with the pro-
gram in cognitive neuroscience and schiz-
ophrenia of the Nathan S. Kline Institute
for Psychiatric Research, 140 Old Or-
angeburg Road, Orangeburg, New York,
10962 (e-mail,
Dr. Medalia is with the department of
psychiatry and behavioral sciences of
Montefiore Medical Center in the Bronx,
New York.
tions. “Managing home and trans-
portation” tests abilities to use the
telephone and public transportation
as well as home management skills.
“Health and safety” assesses aware-
ness of health problems, medical
emergencies, and potential hazards
around the home. “Social adjust-
ment” reflects the individual’s con-
cerns and attitudes about interper-
sonal relationships. The perform-
ance–information factor subscale re-
flects actual knowledge or skills used
to perform tasks—for example, using
a telephone book or making change.
The problem-solving factor sub-
scale (ILS-PB) comprises 33 items
across all subscales that evaluate ab-
stract reasoning and judgment re-
quired for daily living. Sample items
are “What would you do if your lights
and television went out simultane-
ously?” and “What would you do if
you unintentionally lost ten pounds
in a month?” We chose to focus on
the ILS-PB in order to assess func-
tional capacity as it relates to the un-
derlying cognitive skills.
The ILS-PB takes 20 to 25 min-
utes to administer. As reported in the
ILS manual, standardized scores
ranging from 20 to 39 suggest maxi-
mum (full-time) supervision for dai-
ly living, scores from 40 to 49 suggest
moderate supervision, and scores
from 50 to 63 suggest minimum su-
pervision, or independent living.
Psychometric properties of the ILS-
PB suggest that this subscale main-
tains the integrity of the ILS in its
entirety (alpha coefficient=.86, test-
retest reliability=.90, interrater reli-
ability=.98, concurrent validity
based on tests of social reasoning—
for example, r=.65 with WAIS-R
Comprehension). Discriminant va-
lidity tests found significant differ-
ences between nonclinical and clini-
cal samples.
GAF scores, rated on DSM-IV
axis V, are used to assess the overall
level of psychosocial functioning for
both inpatient and outpatient psy-
chiatric patients. Ratings range
from 0 to 100, based on a hypothet-
ical continuum of mental health
comprising psychological, social,
and occupational functioning based
on behavioral and symptom-orient-
ed descriptions.
We recruited participants as part
of a larger study described elsewhere
(2). Written informed consent was
obtained according to the research
protocols approved by the institu-
tional review boards at the respec-
tive sites. Clinical psychiatrists who
treated study participants assigned
GAF scores in charts but were blind
to the study design. The ILS-PB was
independently administered and
scored by a trained psychologist.
Each participant’s current func-
tional level was coded from chart in-
formation before the ILS was admin-
istered. Three levels of supervision
(living status) were delineated for
community functioning: maximum
supervision (24-hour hospitalization
or staff supervision), moderate super-
vision (daily contact with staff or sig-
nificant others in a residential pro-
gram or family care setting), and min-
imum supervision (intermittent con-
tact with case managers, presence of
roommates, or living alone in an un-
supervised apartment).
A total of 162 patients were inter-
viewed (61 women, or 38 percent,
and 101 men, or 62 percent). The
sample comprised 87 inpatients (54
percent) and 75 outpatients (46 per-
cent). Fifty-three patients had a di-
agnosis of schizoaffective disorder
(33 percent), and 109 had a diagno-
sis of schizophrenia (67 percent).
The mean±SD age of the partici-
pants was 37.2±8.3 years. All partici-
pants were persistently ill and had a
history of continuous care for close
to two years. Further demographic
characteristics of this sample are re-
ported elsewhere (2).
GAF and ILS-PB scores for the
three living status groups are shown
in Table 1. Lower GAF scores were
observed among individuals who re-
quired maximum supervision. How-
ever, no significant differences in
GAF scores were found between the
moderate- and minimum-supervi-
sion groups.
Persons who required maximum
supervision were more deficient in
daily problem-solving skills, whereas
the minimum-supervision group ap-
proached the recommended ILS-PB
cutoff score of 50 for independent
Significant differences on ILS-PB
scores were found between all
Discussion and conclusions
The results of this study indicate that
the ILS-PB successfully discrimi-
nates three levels of functional out-
come. The problem-solving skills
demonstrated on the ILS were sig-
nificantly predictive of living status.
The clinically driven global measure
(GAF) discriminated only two levels
of need for supervision—maximum
supervision and moderate supervi-
sion. This finding suggests that cog-
nitive skills are more sensitive than
symptoms for discriminating func-
tional status, a viewpoint consistent
PSYCHIATRIC SERVICES September 2004 Vol. 55 No. 9
TTaabbllee 11
Mean±SD scores on the Global Assessment of Functioning (GAF) and the prob-
lem-solving factor subscale of the Independent Living Scales (ILS-PB), by living
status, among 162 patients with schizophrenia
Maximum Moderate Minimum
supervision supervision supervision
Variable (N=87) (N=54) (N=21)
Current GAF score
36.6±10.3 52.6±10.3 56.9±9.5
Highest GAF score last year
39.5±10.5 53.3±9.8 57.7±9.6
ILS-PB score
29±10.6 38.4±11.5 48.9±6.2
One-way analysis of variance (ANOVA): F=58.6, df=2, 159, p<.01. Bonferroni post hoc tests were
significant at the .05 level for differences between the maximum care and moderate care groups
only. No significant difference was observed between the moderate and minimum care groups.
One-way ANOVA: F=44.5, df=2, 159, p<.01. Bonferroni post hoc tests were significant at the .05
level for differences between the maximum and moderate care groups only. No significant differ-
ence was observed between the moderate and minimum care groups.
One-way ANOVA: F=35.5, df=2, 159, p<.01. Bonferroni post hoc tests were significant at p<.001
between all groups.
PSYCHIATRIC SERVICES September 2004 Vol. 55 No. 9
with the results of previous research
that found GAF more strongly corre-
lated with clinical symptoms than
functioning per se (9).
Because the ILS-PB specifically
addresses daily problem-solving
skills, it is an appropriate instrument
for aftercare follow-up among per-
sons with severe mental illness as
they make the transition from an in-
stitutional setting (inpatient care) to
community living (outpatient care).
Although it is unlikely that a single
instrument could meet the multiple
demands for obtaining outcomes
data across a variety of settings, we
believe that the ILS-PB warrants
further consideration as a measure
that links cognition to functional
1. Green, MF, Kern, RS, Braff, DL, et al:
Neurocognitive deficits and functional
outcome in schizophrenia: are we measur-
ing the “right stuff”? Schizophrenia Bul-
letin 26:119–136, 2000
2. Revheim N, Medalia A: Verbal memory,
problem-solving skills, and community
status in schizophrenia. Schizophrenia Re-
search 68:149–158, 2004
3. Dickerson F: Assessing clinical outcomes:
the community functioning of persons
with serious mental illness. Psychiatric
Services 48:897–902, 1997
4. Dickerson FB, Origoni AE, Pater A, et al:
An expanded version of the Multnomah
Community Ability Scale: anchors and in-
terview probes for the assessment of
adults with serious mental illness. Com-
munity Mental Health Journal 39:131–
137, 2003
5. Wallace CJ, Liberman RP, Tauber R, et al:
The Independent Living Skills Survey: a
comprehensive measure of the communi-
ty functioning of severely and persistently
mentally ill individuals. Schizophrenia
Bulletin 26:631–658, 2000
6. Patterson TL, Goldman S, McKibbin CL,
et al: UCSD Performance-Based Skills As-
sessment: development of a new measure
of everyday functioning for severely men-
tally ill adults. Schizophrenia Bulletin
27:235–245, 2001
7. Loeb PA: ILS: Independent Living
Scales Manual. San Antonio, Tex, Psycho-
logical Corp, Harcourt Brace Jovanovich,
8. Medalia A, Revheim N, Casey M: Reme-
diation of memory disorders in schizo-
phrenia. Psychological Medicine 30:1451–
1459, 2000
9. Roy-Byrne P, Dagadakis C, Unutzer J, et
al: Evidence for limited validity of the re-
vised Global Assessment of Functioning
Scale. Psychiatric Services 47:864–866,
CCoommiinngg iinn OOccttoobbeerr:: TTrriibbuutteess ttoo
JJoohhnn AA.. TTaallbbootttt,, MM..DD..,, EEddiittoorr EEmmeerriittuuss
Reprints of some of Dr. Talbott’s noteworthy papers
Commentaries by current leaders in the field
Letters from Dr. Talbott’s colleagues and friends
    • "Some assessments, such as screening tools, are not meant to be repeated. For example, the Independent Living Scales (Loeb, 1996; Revheim & Medalia, 2004) and the AMPS (Baron, 1994) are used primarily to determine one's level of autonomy at a given point. However, this does not explain the reasons that an outcome measure, such as the COPM, is not systematically repeated. "
    [Show abstract] [Hide abstract] ABSTRACT: Background. Little is known about assessment practices of occupational therapists working with adults with mental disorders. Purpose. This study investigates the assessment practices of occupational therapists working with clients experiencing symptoms of schizophrenia or major depressive disorder. Method. We conducted a national survey of assessment practices using case vignettes of hypothetical clients. Findings. From 343 vignettes completed by 286 respondents, 68.4% included the use of one or more standardized measures during treatment. Measures were rarely repeated. Results showed that the Canadian Occupational Performance Measure was the most frequently used, suggesting a focus on assessing global functioning, while the Assessment of Motor and Process Skills was listed as the most desired assessment tool. Implementing nonstandardized assessments was common. Implications. Despite wide variations in occupational therapists' assessment practices, the use of standardized assessments is prevalent. The low rate of repeated measures (0% to 25.9%) suggests a need to better monitor changes and treatment outcomes.
    Article · May 2015
    • "However, since it has been argued that its ratings are most strongly correlated with ratings of clinical symptoms than functioning, reliance on the GAF scale as the only tool to assess patients' functioning may be problematic [15]. HoNOS is widely adopted in rehabilitative settings and its ability to predict global functional outcomes has been shown in several studies [2,14,24,32,42,45]. HoNOS has also been shown to perform well against other established clinician-rated instruments that measure constructs related to psychosocial outcome [37]. "
    [Show abstract] [Hide abstract] ABSTRACT: Cognitive dysfunction has been demonstrated in patients with schizophrenia, and this may affect patients' functional outcome. The improvement of such dysfunction by means of cognitive remediation interventions has become a relevant target in the care of schizophrenia. To assess the effectiveness of the cognitive subprograms of Integrated Psychological Therapy (IPT) on symptomatological, neuropsychological and functional outcome variables and to analyze the relationships between cognitive and functional outcome changes in schizophrenia. Thirty-two patients with schizophrenia were assigned to cognitive remediation (IPT-cog) or usual rehabilitative interventions in a naturalistic setting of care. Clinical, neuropsychological and functional outcome variables were assessed at baseline and after 24 weeks of treatment. The IPT-cog group improved significantly more than the comparison group with respect to psychopathological and functional outcome variables. Moreover, only the IPT-cog group improved significantly in the neuropsychological domains of verbal and working memory, with specific significant correlations between neurocognitive performance and functional outcome changes. The results of the study confirm the effectiveness of the cognitive remediation component of IPT in schizophrenia, and indicate that some of the changes in functional outcome may be mediated by improvement in specific cognitive domains.
    Full-text · Article · Jul 2011
    • "Also, the RIL does not provide any indication of hierarchy among items; the RIL does not distinguish complex capabilities (eg, managing one's finances) that AD typically compromises early in the disease process from more basic capabilities (eg, toileting) that typically are affected later on. The Independent Living Scales (ILS) assessment was designed to provide a broad measure of dependence in patients with dementia by incorporating cognitive/reasoning skills in addition to physical measures [28,30]. The ILS is conducted by a clinician and consists of five subscales and two factor-analyzed subscales. "
    [Show abstract] [Hide abstract] ABSTRACT: This article reviews measures of Alzheimer's disease (AD) progression in relation to patient dependence and offers a unifying conceptual framework for dependence in AD. Clinicians typically characterize AD by symptomatic impairments in three domains: cognition, function, and behavior. From a patient's perspective, changes in these domains, individually and in concert, ultimately lead to increased dependence and loss of autonomy. Examples of dependence in AD range from a need for reminders (early AD) to requiring safety supervision and assistance with basic functions (late AD). Published literature has focused on the clinical domains as somewhat separate constructs and has given limited attention to the concept of patient dependence as a descriptor of AD progression. This article presents the concept of dependence on others for care needs as a potential method for translating the effect of changes in cognition, function, and behavior into a more holistic, transparent description of AD progression.
    Full-text · Article · Nov 2010
Show more