SCHRES-05857; No of Pages 7

Schizophrenia Research xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Schizophrenia Research
journal homepage: www.elsevier.com/locate/schres

The Lille Apathy Rating Scale (LARS): Exploring its psychometric

properties in schizophrenia
Hanan Yazbek a,c,⁎, Joanna Norton c, Delphine Capdevielle b,c, Aurore Larue c, Jean-Philippe Boulenger b,c,
Marie-Christine Gély-Nargeot a, Stéphane Raffard a,c
a
Epsylon EA 4556 Laboratory of Dynamics of Human Abilities & Health Behaviors, University of Montpellier, France
b
INSERM U-1061, France
c
University Department of Adult Psychiatry, Hospital La Colombière, CHRU Montpellier, 39 avenue Charles Flahault, 34295 Montpellier Cedex 5, France

a r t i c l e i n f o a b s t r a c t

Article history: Background: Apathy in schizophrenia patients is linked to marked functional impairments and can be defined as a
Received 21 October 2013 quantitative reduction of voluntary, goal-directed behaviors. If there are now convincing arguments to consider
Received in revised form 28 March 2014 apathy as a multidimensional psychopathological state (cognitive, emotional, and behavioral), there is a lack of
Accepted 22 April 2014 validated and standardized instruments for detecting apathy and assessing its multidimensional aspects in
Available online xxxx
schizophrenia. The Lille Apathy Rating Scale (LARS) is a semi-structured interview, yielding a global score and
composite subscores for the different domains of apathy and has been validated in several different contexts
Keywords:
Apathy
but not in schizophrenia patients.
Schizophrenia Objective: The aim of this study is to examine the psychometric properties of the LARS and identify the distinct
Lille Apathy Rating Scale (LARS) components of apathy in a sample of schizophrenia patients.
Exploratory factorial analysis Methods: One hundred-and-twelve schizophrenia patients were included and they completed the LARS, The
Calgary Depression Scale in Schizophrenia, the Positive and the Negative Syndrome Scale and the Scale for the
Assessment of Negative Symptoms. The patient group was compared to 51 healthy control subjects.
Results: Principal component analysis showed that the LARS proved a single construct which forms the root of an
oblique factor structure reflecting four dimensions: novelty and social life, behavioral involvement, emotional
involvement, and judgment skills. The main psychometric properties of the LARS were satisfactory.
Conclusions: Our findings show that the LARS has satisfactory psychometric properties when used in a different
setting than the original version. The LARS is a promising instrument to examine apathy in schizophrenia through
a multidimensional framework.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction reduction of voluntary behaviors directed toward one goal (Levy and
Dubois, 2006).
Apathy is a transnosographic psychopathological state frequently Different psychometric tools exist to measure apathy. The most
found in different diseases. Indeed, it is estimated at 55% in Alzheimer's frequently used tools validated in schizophrenia are the Positive and
disease, 27% in Parkinson's disease, 94% in major depressive episode and Negative Syndrome Scale (PANSS; Kay et al., 1987), the Scale to Assess-
53% in schizophrenia (Mulin et al., 2011). In schizophrenia, there is a ment of Negative Symptoms (SANS; Andreasen, 1989), the Apathy
lack of positive results in the treatment of apathy, which contributes Evaluation Scale (AES; Marin et al., 1991) and the Brief Negative Symp-
to poorer functioning and subjective quality of life in both first episode tom Scale (BNSS; Kirkpatrick et al., 2011). The SANS, the PANSS and the
and chronic schizophrenia (Konstantakopoulos et al., 2011; Evensen BNSS each have an “apathy” subscale, but apply different definitions of
et al., 2012; Faerden et al., 2013). In literature, there is strong evidence apathy as a one-dimensional concept. However, apathy requires for its
of the multidimensional nature of apathy: behavioral, cognitive and understanding a multidimensional approach. The AES is the most wide-
emotional (Marin et al., 1991; Levy and Dubois, 2006; Arnould et al., ly used multidimensional tool for assessing apathy in schizophrenia,
2013). Therefore, apathy is defined as a multidimensional psychopath- with good psychometric properties. Nevertheless, limits can be made
ological state (cognitive, emotional, and behavioral) manifesting as a to this scale. Its semi-structured nature induces a lack of standardization
in the administration instructions and scoring method; also the positive
and negative orientation of the question may be a possible source of
⁎ Corresponding author at: Hôpital de la Colombière, Service Universitaire de
Psychiatrie Adulte, CHRU Montpellier, 39 avenue Charles Flahault, 34295 Montpellier
errors. Furthermore, it has not been validated in French and in schizo-
Cedex 5, France. Tel.: +33 4 67 33 97 02; fax: +33 4 67 33 89 95. phrenia. The multidimensional assessment of apathy, in schizophrenia,
E-mail address: yazbek.hanan@gmail.com (H. Yazbek). is limited. A recent scale, The Lille Apathy Rating Scale (LARS; Sockeel

http://dx.doi.org/10.1016/j.schres.2014.04.034
0920-9964/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
2 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx

et al., 2006) was developed to address the limitations of existing tools Likert-type scale ranging from 1 (absent) to 7 (extreme). The PANSS
and to include an assessment of the various components of apathy. Val- can be divided in five factors: positive, negative, cognitive, excitement,
idated initially in French in Parkinson's disease, it has been constructed and dysphoric mood (Levine & Rabinowitz, 2007).
following the main concepts of Marin et al. (1991) and more specifically The Scale for the Assessment of Negative Symptoms (SANS;
the conceptual bases of the AES, the clinical experiences of the authors Andreasen, 1989) was used for evaluating negative symptoms and apa-
(Sockeel et al., 2006) and the concept of apathy of Stuss et al. (2000). thy. This 25-item rating scale was completed during a clinical interview.
Its psychometric properties are good and its structured nature and the Each item is assessed on a Likert-type scale ranging from 0 (absent) to 5
presence of dichotomous items allow more objective administration (severe). The SANS can be divided in three factors: lack of motivation,
instructions and scoring methods. Furthermore, it is important to high- expressivity, and disorganization. Apathy was assessed by using the
light that currently there is no French validated scale that assesses factor “lack of motivation” (items 15, 16, 18, 19, 20, and 21).
apathy in a multidimensional approach in schizophrenia. The Calgary Depression Scale in Schizophrenia (CDSS) developed by
Consequently, the main objective of our research was to examine the Addington et al. (1993) and validated in French by Bernard et al. (1998)
psychometric properties of the LARS and identify distinct components is a 9-item scale measuring depression in schizophrenia patients. A
of apathy in a sample of schizophrenia patients. We also sought to com- score equal to higher than 6 reflects depression (Addington et al., 1994).
pare apathy and its components as measured by the LARS to SANS and The depression and anxiety modules of the PRIME-MD Patient
PANSS symptomatology sub-types and depressive symptomatology. Health Questionnaire (PHQ) (Spitzer et al., 1999) were completed by
We also compare LARS apathy scores to those in a group of healthy the control subjects. The PHQ is a self-report questionnaire designed
controls and propose cut-off values for schizophrenia. to identify common psychiatric disorders applying DSM-IV diagnostic
criteria.
2. Materials and methods
2.3. Statistical analysis
2.1. Subjects
An initial examination of the shared variance between the 9 domains
One hundred and thirty-seven patients with schizophrenia (N = of the LARS revealed a Kaiser–Meyer–Olkin measure of sampling ade-
137), aged between 19 and 59 years, completed the study between quacy of 0.63, demonstrating adequacy of the sample to provide stable
January 2011 and December 2012. Participants were recruited from factor solutions. Principal component analysis (PCA) was carried out
full- and part-time hospitalization and ambulatory care services of the and eigenvalues and scree plots were examined in order to determine
University Department of Adult Psychiatry in Montpellier (N = 98), the structure of the data. Oblique rotation was performed after exami-
Marseille (N = 31) and Nice (N = 8). Diagnosis of schizophrenia was nation of inter-factor correlations.
established via the Structured Clinical Interview for DSM-IV (SCID-I; Internal consistency was determined using Cronbach's standardized
First et al., 1997). Exclusion criteria were: (a) known neurological alpha coefficients both between items and between sub-scales, as well
disease, (b) brain injuries, or (c) Axis II diagnosis of developmental dis- as split-half reliability. Inter-rater reliability was measured on 44 pa-
orders. Written informed consent was obtained from all participants. tients rated by two independent raters, using intra-class correlation.
Participants were tested individually in one session of 1 h. Twenty- Concurrent validity was assessed by comparing LARS apathy factors
five patients (15.4%) refused to participate. to SANS and PANSS factors and depressive symptomatology. Normality
Patients were followed-up at one, 3, 6 and 12 months. The current was tested using the Shapiro–Wilk test. As the LARS factors were not
study was carried out on the one month follow-up data. One hundred normally distributed, correlations with continuous variables were
and twelve patients completed the one-month follow-up. Among the examined using Spearman's correlation coefficient, and with binary
patients lost to follow-up (N = 25), 10 who had participated at inclu- variables using Wilcoxon's rank sum test. Unconditional logistic regres-
sion refused to participate at the 1-month follow-up, 11 were not able sion analysis was used to compare overall LARS score in patients and
to be contacted again and 2 were too ill, one had moved away and one controls, adjusting for age and gender.
returned to work. Compared to the 112 patients in the analysis, those Finally, a receiver operation characteristic (ROC) curve was con-
lost-to-follow-up had a slightly lower educational level (p = 0.04) but structed comparing the different cut-off values of the LARS to the SANS
were not significantly different in terms of gender, age and total factor “lack of motivation” as a binary variable, patients with scores
PANSS, SANS and LARS scores. under 12 being classified as non apathetic (Fig. 1). Specificity and sensi-
A group of 51 healthy control subjects was recruited from the gener- tivity were calculated for different cut-off values as well as positive (PPV)
al population to match the patient group as closely as possible for age, and negative (NPV) predictive values. Youden's index was used to iden-
gender and educational level. Subjects with depressive or anxiety tify the optimal cut-off value [(specificity + sensitivity) − 1]. Statistical
disorders were excluded. analysis was carried out using SAS 9.3.

2.2. Measures 3. Results

The Lille Apathy Rating Scale (LARS; Sockeel et al., 2006) is a subjec- 3.1. Sample description
tive semi-structured interview assessing specific components of apathy.
This scale includes 33 items, divided into nine domains [i.e., everyday The socio-demographic and clinical characteristics of the patient and
productivity (EP), interests (INT), taking initiative (INI), novelty seeking control samples are described in Table 1.
(NS), voluntary actions (VA), emotional responses (ER), concern (C), so-
cial life (SL), and self-awareness (SA)]. The global score ranges from 3.2. Factor analysis
[− 36 to + 36] (no apathetic between [− 36 and − 22]; slight [− 21
and −17]; moderate apathetic [−16 and −10]; and severe apathetic Principal component analysis carried out on the 9 sub-scales of the
[−9 and +36]). The LARS has been validated in French in Parkinson's LARS identified four factors with eigenvalues close to 1, explaining
disease (Sockeel et al., 2006). The scale and its instructions for use are 62.5% of the model. The first factor accounted for 23% of the variance,
shown in the appendix. the second factor 16.7% of the variance, and the third and fourth factors
Symptomatology was evaluated using the Positive and the Negative 12.7% and 10.1% of the variance, respectively.
Syndrome Scale (PANSS; Kay et al., 1987), a 30-item rating scale Correlation coefficients between each of the 9-subscales and the
completed during a clinical interview. Each item is assessing on a total LARS score were all above r = 0.3 (p b 0.001), except for the

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx 3

Sensitivity
1.0

0.9

0.8 -19 cut off score

0.7

0.6 -18 cut off score

-17 cut off score

0.5

0.4

0.3

0.2

0.1

0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
1 - Specificity

Fig. 1. Receiver operating characteristic (ROC) curve of the LARS for the prediction of apathy defined by a score N12 on the SANS lack of motivation factors.

Interests subscale (r = 0.24, p = 0.01). Correlation coefficients Novelty and Social life (Factor 1) is correlated with SANS expressivity
between the 9-subscales reached 0.36 (between VA & TI) suggesting and Judgment skills (Factor 4) with SANS disorganization.
that an oblique rotation was preferable.
Factor loadings after oblique (rotate = oblimin) rotation are pre-
sented in Table 2. Using the 0.40 threshold for factor loading, two sub- 3.5. ROC curve analysis for identifying an optimal LARS cut-off value
scales were found to load on Factor 1 which was subsequently labeled
“Novelty and Social life” (Novelty Seeking, Social Life); two sub-scales ROC curve analysis (Fig. 1) showed that the total LARS score had a
were found to load on Factor 2 labeled “Behavioral Involvement” (Taking moderate to good prediction of apathy as defined by the SANS "lack of
the Initiative, Voluntary Actions); one sub-scale loaded on Factor 3 motivation" with a 12 + cut-off value, with an area under ROC curve
labeled “Emotional Involvement” (Emotional Responses); and finally of 0.73. The three best cut-off values according to the maximum of
two subscales loaded on Factor 4 labeled “Judgment skills” (Interest, the Youden index are given in Table 4, with their corresponding psycho-
Self-awareness). Everyday productivity loaded highly on both Factor 1 metric properties. With the −18 cut-off value on the LARS and a 51.8%
and Factor 3 and Concern loaded similarly on all factors, demonstrated prevalence of apathy in our sample according to SANS lack of motiva-
complex loadings. Consequently, these variables were not taken into ac- tion, the corresponding PPV and NPV are 69.8% and 64.4% respectively.
count in the factor loadings interpretation. The correlation coefficients
between the four oblique factors were low, with a significant correlation
between “Novelty and Social life” and “Behavioral Involvement”, but a low 3.6. Comparison of LARS apathy scores with the control group
correlation coefficient (r = 0.28).
The controls do not differ significantly from the patient sample for
gender and number of years of education (p = 0.06 and p = 0.75 re-
3.3. Internal consistency and inter-rater variability
spectively). The age distribution is significantly younger (p = 0.02).
LARS apathy scores in the two samples are summarized in Fig. 2. The
Internal consistency between the 9 subscales was 0.53. The split-half
difference in apathy score between patients and controls remains
reliability coefficient was 0.34 and reached 0.51 with the Spearman–
unchanged and significant when adjusting for age and gender (p =
Brown “prophecy formula” correction. Regarding inter-rater reliability,
0.002). Applying the − 18 cut-off value proposed for LARS apathy
analysis of variance on repeated measures did not reveal any significant
among schizophrenia patients, 27.4% of the controls are classified as
differences in the scoring (F b 0.13, NS). The intraclass correlation coef-
apathetic, compared to 47.3% in the patient group.
ficient was 0.95, which corresponds to a high level of reliability.
The total LARS scores in the control group were used to further di-
vide apathy into four categories, avoiding the inclusion of control sub-
3.4. Relationship between LARS apathy dimensions and other symptom jects in the severely apathetic category. The following classification
scales was established: [−36; −24] for non apathetic subjects, [−23; −19]
for slightly apathetic subjects, [− 18; − 13] for moderately apathetic
Correlation coefficients between the four LARS factors and SANS, subjects and finally [−12; +36] for severely apathetic subjects. Accord-
PANSS, depressive symptomatology and global functioning are overall ing to this classification, 52.9% of the control group was non-apathetic,
relatively low (Table 3). Both Novelty and Social life (Factor 1) and and 19.7% and 27.4% slightly or moderately apathetic, respectively.
Behavioral Involvement (Factor 2) are positively and significantly corre- None were classified as severely apathetic. In the patient group, 9.8%
lated with SANS “lack of motivation” and PANSS “negative symptoms” of subjects were non-apathetic, 42.9% slightly apathetic, and 25.9% and
meaning symptom severity increases with apathy severity. Only 21.4% moderately or severely apathetic, respectively.

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
4 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx

Table 1
Sample characteristics of the patient group (at the one-month follow-up) and control group.

Patient group (N = 112) % Control group (N = 51) %

Gender (male %) 76.8 62.8

Age (years) (median [min; max]) 38 [19; 59] 29 [19; 56]
Education (years) (median [min; max])⁎ 11 [6; 17] 12 [8; 17]
Marital status ⁎
Single 91.0 –
Married, with partner 4.5 –
Divorced, separated 4.5 –
Living situation⁎
Alone 48.2 –
With partner or family 39.3 –
Other (flat share, hostel, special home etc.) 12.5 –
Cannabis abuse/dependence (yes)⁎ 42.9 –
Abuse/dependence of other substances (mainly cocaine) (yes)⁎ 24.3 –
Duration of psychotic symptoms (years) (median [min; max])⁎ 14 [0; 41] –
Main type of care received
Ambulatory care 50.0 –
Day-hospital 29.5 –
Full-time hospitalization 17.8 –
Private psychiatrist 1.8 –
No follow-up care 0.9 –
Daily chlorpromazine equivalent (CPZeq) drug dosage (mg/day) 612.5 [0; 4250] –
LARS Apathy score (median [min; max]) −19 [−33; 3] −24 [−34; −13]
Scores on each sub-scale (median [min, max])
Everyday productivity (EP) −3 [−4; 4] –
Interests (INT) −2 [−3; 2] –
Taking the initiative (INI) −2 [−4; 4] –
Novelty seeking (NS) −4 [−4; 2] –
Voluntary actions (VA) −2 [−4; 4] –
Emotional responses (ER) −2.5 [−3; 3] –
Concern (C) −2.5 [−4; 2] –
Social life (SL) −2 [−4; 3] –
Self-awareness (SA) −2 [−4; 4] –
LARS Apathy classification⁎⁎
No [−36; −22] 33.0 60.8
Slight [−21; −17] 28.6 25.5
Moderate [−16; −10] 25.9 13.7
Severe [−9; +36] 12.5 0.0
SANS factors (median [min; max])
Lack of motivation 12 [0; 26] –
Expressivity 12 [0; 35] –
Disorganization 4 [0; 13] –
Total score 38 [2; 81] –
PANSS factors (median [min; max])
Negative 22 [10; 40] –
Cognitive 13 [6; 23] –
Dysphoric mood 11 [5; 22] –
Positive 9 [4; 26] –
Excitement 8 [5; 21] –
Total score 68 [39; 106] –
Depressive symptomatology (CDSS)⁎⁎⁎
Score (median [min; max]) 3 [0; 23] –
Score ≥ 6 25.0 –
⁎ At inclusion.
⁎⁎ following Sockeel et al.'s principal components and thresholds.
⁎⁎⁎ CDSS: Calgary Depression Scale for Schizophrenia.

4. Discussion schizophrenia differs. This is in accordance with the results found by

Mulin et al. (2011) who have shown that the different forms of apathy
Our main aim was to investigate the psychometric properties of the (emotional, cognitive, and auto-activation) are not present in the
LARS in a sample of schizophrenia patients in order to provide a validat- same way according to the disease (e.g. Parkinson's disease, Alzheimer's
ed and a specific tool assessing apathy through a multidimensional disease, and schizophrenia), emotional apathy being only predominant
approach specific to schizophrenia. This scale has been used so far in in schizophrenia. For other diseases it is essentially cognitive and auto-
patients with Parkinson's disease and has shown good psychometric activation dimensions that predominate. More specifically in schizo-
properties (Sockeel et al., 2006; Zahodne et al., 2009). It also provides phrenia, cognitive and emotional dimensions are found in 94% of
standardized administration instructions and scoring methods. More- patients while the dimension of auto-activation is found in 84% of
over, the LARS is easy and quick to complete. patients. It is consequently not surprising that the factor structure of
In keeping with Sockeel et al. (2006), exploratory factorial analysis the LARS for schizophrenia is different from that found in Parkinson's
demonstrated that the LARS was composed of four factors, which un- disease.
derline the multidimensional nature of apathy. However, the factorial Whereas the factorial analysis by Sockeel et al. (2006) revealed that
analysis of the LARS in schizophrenia was different from that found apathy in Parkinson was constituted of an Intellectual Curiosity factor,
in Parkinson's disease, suggesting that the structure of apathy in an Action/Initiation factor, an Emotional factor and a Self-Awareness

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx 5

Table 2
Factor loadings after oblique rotation and correlations between factor-based scores (N = 112).

Factor 1 Novelty and Factor 2 Bevarioral Factor 3 Emotional Factor 4 Judgment

Social life (NS + SL) Involvement (INI + VA) Involvement (ER) skills (INT + SA)

(1) Factor loadings:

Everyday productivity (EP) 0.589 0.118 0.518 0.104
Interests (INT) −0.035 0.285 0.117 0.742
Taking the initiative (INI) −0.031 0.856 −0.114 −0.112
Novelty seeking (NS) 0.669 0.094 −0.184 0.119
Voluntary actions (VA) 0.112 0.662 0.248 0.192
Emotional responses (ER) 0.001 0.019 0.838 0.098
Concern (C) 0.323 0.211 0.372 0.259
Social life (SL) 0.786 −0.110 0.303 −0.113
Self-awareness (SA) −0.075 0.294 0.045 −0.732
(2) Correlations between oblique factors:
Factor 1 — Novelty and Social Life 1.000 0.285 0.126 0.118
Factor 2 — Behavioral Involvement 1.000 0.123 0.117
Factor 3 — Emotional Involvement 1.000 0.048
Factor 4 — Judgment Skills 1.000

(1) Numbers in bold type indicate the correlations that determined the clusters for the oblique factors.
(2) Numbers in bold type indicate significant Spearman correlation coefficients between the factors (p b 0.01).

factor using all 9 subscales, only 7 of the 9 subscales were revealed in PANSS negative factor. Only Novelty and Social life factor was associated
our factor structure. In our study, we identified a factor labeled to SANS "expressivity" factor. These results suggest that classical scales
Emotional Involvement which includes only ER. Our Judgment Skills used to assess negative symptoms fail to evaluate the complexity of
factor includes SA and INT. In keeping with Sockeel et al. (2006), our the different dimensions of apathy in schizophrenia. Moreover, only
first factor labeled Novelty and Social life groups both NS and SL; how- Novelty and Social life and Behavioral Involvement factors are positively
ever it does not include INT and VA. Finally our Behavioral Involvement correlated to depressive symptoms, which is not surprising. Indeed, the
factor includes INI as well as VA, but does not include EP as for the Ac- lost of interest for novelty in almost all activities and psychomotor retar-
tion/Initiation factor of Sockeel et al. (2006). Compared to Parkinson's dation marked by an absence of taking initiative and lack of motivation
disease the main difference in our study is that the three forms of apathy are common in depression. Ours results provide elements on the rela-
can be clearly identified from each other in the four factors. Indeed the tive independence of the depressive and apathetic symptomatology in
factors 1 and 4 refer cognitive apathy, the factor 2 to behavioral apathy, schizophrenia.
and the factor 3 to emotional apathy. In Parkinson's disease, the three In addition, for the convergent validity, we compared the LARS total
forms cannot be distinguished in the different factors: factors 1 and 3 score to the SANS factor "lack of motivation" as our gold standard for ap-
refer to cognitive and behavioral apathy, factor 2 to cognitive and athy. We chose a SANS cut-off value of 12 which we considered to be the
emotional apathy, and factor 4 to cognitive apathy. most appropriate as patients with scores of 2 and above on the individ-
The absence of correlations between the LARS factors and positive, ual SANS items have previously been defined as mild apathetic. Accord-
cognitive, dysphoric mood and excitement factors of the PANSS sug- ing to this threshold, 51.8% of the sample is apathetic, which is similar to
gests that apathy as measured by the LARS exists as an independent en- rates found by Faerden et al. (2010) (51%) and Mulin et al. (2011) (53%).
tity regardless of these symptoms. Similarly, Sockeel et al. (2006) have The ROC curve analysis suggested that the total LARS score was a rela-
shown that symptoms induced by Parkinson's disease do not affect tively good predictor of SANS lack of motivation with this cut-off
the assessment of apathy. However, in our study only two of the four value. A LARS threshold of − 18 seemed to be the best compromise
factors of the LARS (Novelty and Social life and Behavioral Involvement) between sensitivity and specificity. Sockeel et al. (2006) showed that
were positively correlated with SANS "lack of motivation" factors and a LARS cut-off value of − 16 represented an excellent specificity and

Table 3
Relationship between LARS apathy factors and PANSS factors, SANS factors, Calgary depression score and global functioning score (Spearman's correlation coefficient), N = 112.

Factor 1 Novelty Factor 2 Behavioral Factor 3 Emotional Factor 4 Total score

and Social life Involvement Involvement Judgment skills

SANS factors:
Lack of motivation 0.437⁎⁎ 0.320 ⁎⁎ 0.018 0.101 0.471⁎⁎
Expressivity 0.272⁎ 0.067 0.020 0.161 0.237 ⁎
Disorganization 0.138 0.177 −0.139 0.196⁎ 0.158
Total score 0.385⁎⁎ 0.272⁎ −0.021 0.193⁎ 0.411⁎⁎
PANSS factors:
Negative 0.294⁎ 0.273⁎ 0.050 0.172 0.391⁎⁎
Cognitive 0.033 0.039 −0.032 0.099 0.042
Dysphoric Mood 0.109 0.130 0.027 −0.044 0.087
Positive 0.021 0.063 −0.037 0.097 0.050
Excitement −0.128 −0.047 −0.065 −0.002 −0.075
Depressive symptomatology (CDSS):
Continuous score 0.212⁎ 0.365⁎⁎ 0.092 −0.00 0.329⁎⁎
Low (score b 6) (n = 84) −5 [−8; 3]a −4 [−8; 5]a −2.5 [−3; 3]a −5 [−6; 2]a −19 [−33; 3]a
High (score ≥ 6) (n = 28) −4.5 [−8; 3]a −2 [−8; 6]a −2.5 [−3; 3]a −4.5 [−6; −2]a −15.5 [−28; −3]a
p = 0.18 p = 0.09 p = 0.62 p = 0.12 p = 0.05

CDSS: Calgary Depression Scale for Schizophrenia.

a
Median [min; max].
⁎ Indicates p value b .05.
⁎⁎ Indicate p value b .01.

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
6 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx

Table 4 nondrug treatment on the altered forms of apathy would be more effec-
Operating characteristics for LARS cut-off scores. tive. It is thus necessary to have tools assessing apathy as objectively as
LARS cut-off scores Sensitivity (%) Specificity (%) PPV (%) NPV (%) possible. Future research is needed to replicate our results and exam-
ined the usefulness of LARS in providing an objective assessment of
−17 60.3 74.1 71.4 63.5
−18 63.8 70.4 69.8 64.4 apathy in schizophrenia. In addition, future studies should explore the
−19 70.7 63.0 67.2 66.7 psychological, biological, and environmental mechanisms underlying
the different forms of apathetic manifestations as revealed by the
LARS in order to develop specific psychosocial and biological interven-
positive predictive value in their study population. However, in a similar tions dedicated to the treatment of the different forms of apathy present
population, Zahodne et al. (2009) identified a LARS cut-off value of −22 in schizophrenia. Moreover, research is clearly needed to confirm the
as having the best psychometric properties. Our findings suggest that a model of apathetic manifestations as proposed by the LARS, notably
LARS score ≥ −18 could indicate the presence of apathy in schizophre- by using longitudinal studies in order to investigate the predictive
nia patients. However further studies with a validated measure of apa- role of these dimensions in daily functioning in individuals with
thy against which to measure the performance of the LARS are needed schizophrenia.
to confirm this.
The first limit of our study is the absence of a validated measure of Role of funding source
apathy in schizophrenia. Using the SANS factor "lack of motivation" The study was financed by a University Hospital clinical research Grant, Montpellier,
2010. The University Hospital had no further role in the study design; in the collection,
item with a cut-off of 12 is questionable because this scale is not recog- analysis and interpretation of data; in the writing of the report; and in the decision to
nized as a validated measure of apathy. On the other hand it is a widely submit the paper for publication.
used validated scale for measuring negative symptoms in schizophre-
nia. It may have been more appropriate to use the judgment of an expert Contributors
clinician based on diagnostic criteria of apathy validated in schizophre- Stéphane Raffard, Delphine Capdevielle, Joanna Norton and Hanan Yazbek contribut-
ed to the study design. Hanan Yazbek & Aurore Larue recruited and assessed the patients.
nia. Sockeel et al. (2006) based their cut-off on the judgment of two clin-
Joanna Norton performed the statistical analysis. Hanan Yazbek, Joanna Norton and
ical experts. However as previously underlined by Zahodne et al. Stéphane Raffard prepared the manuscript, with feedback from the other authors.
(2009), these experts made their judgment based on interview with
the patient and family's member with no information on objective diag- Conflict of interest statement
nostic criteria. The authors declare that they have no competing financial or other interests that
A second limitation relates to the recruitment of our patients from might be perceived to influence the results and discussion reported in this paper.
French public hospitals whatever the type of care received. At the time
of the study, 19 patients were still hospitalized full-time. In their review, Acknowledgments
We would like to thank the Dr. C. Lançon of the Sainte Marguerite Hospital (Marseille),
Capdevielle et al. (2009) explain that the length of hospitalization im- the Dr. M. Benoit of the Pasteur Hospital (Nice) and their teams for their help in collecting
pacts on the quality of life of patients as well as on the presence of data. We are grateful to all of the patients who contributed to this study.
schizophrenic symptoms. It would have been more appropriate to
have a sample of patients treated as outpatients only, in order to avoid Appendix A. Supplementary data
bias related to full-time hospital stay. Also the study did not include
any patients treated in the private sector. Supplementary data to this article can be found online at http://dx.
The control sample could be criticized for its small size and lower doi.org/10.1016/j.schres.2014.04.034.
median age than in the patient sample; however, the difference in over-
all LARS apathy score between the two groups was not modified when References
adjusting for age and gender.
In summary, the originality of our research lies in examining the fac- Addington, D., Addington, J., Maticka-Tyndale, E., 1993. Assessing depression in
schizophrenia: the Calgary Depression Scale. Br. J. Psychiatry 22, 39–44.
torial structure of the LARS in patients with schizophrenia and propos- Addington, D., Addington, J., Maticka-Tyndale, E., 1994. Specificity of the Calgary
ing a validated French multidimensional scale in schizophrenia. Our Depression Scale for schizophrenics. Schizophr. Res. 11, 239–244.
findings show that the LARS has satisfactory psychometric properties Andreasen, N.C., 1989. Scale for the assessment of negative symptoms (SANS). Br. J. Psy-
chiatry 155 (7), 53–58.
when used in a different setting. Schizophrenia is a disease in which Arnould, A., Rochat, L., Azouvi, P., Van der Linden, M., 2013. A multidimensional approach
there is considerable heterogeneity, therefore a specific and focused to apathy after traumatic brain injury. Neuropsychol. Rev. 23 (3), 210–233.

Fig. 2. Diagram of LARS apathy scores among schizophrenia patients and control subjects (p = 0.0004 for the difference in LARS score between both groups, adjusted for age and sex).

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034
 H. Yazbek et al. / Schizophrenia Research xxx (2014) xxx–xxx 7

Bernard, D., Lançon, C., Auquier, P., Reine, G., Addington, D., 1998. Calgary Depression Konstantakopoulos, G., Ploumpidis, D., Oulis, P., Patrikelis, P., Soumani, A., Papadimitriou,
Scale for Schizophrenia: a study of the validity of a French-language version in a G.N., Politis, A.M., 2011. Apathy, cognitive deficits and functional impairment in
population of schizophrenic patients. Acta Psychiatr. Scand. 97, 36–41. schizophrenia. Schizophr. Res. 133, 193–198.
Capdevielle, D., Boulenger, J.-P., Villebrun, D., Ritchie, K., 2009. Durée d'hospitalisation des Levine, S.Z., Rabinowitz, J., 2007. Revisiting the 5 dimensions of the Positive and Negative
patients souffrant de schizophrénie: implication des systèmes de soin et conséquence Syndrom Scale. J. Clin. Psychopharmacol. 27 (5), 431–436.
médicoéconomiques. Encéphale 35, 394–399. Levy, R., Dubois, B., 2006. Apathy and the functional anatomy of the prefrontal cortex–
Evensen, J., Rossenberg, J.I., Barder, H., Haahr, U., ten Velden Hegelstad, W., Joa, I., basal ganglia circuits. Cereb. Cortex 16 (7), 916–928.
Johannessen, J.O., Larsen, T.K., Melle, I., Opjordsmoen, S., Rund, B.R., Simonsen, E., Marin, R.S., Biedrzycki, R.C., Firinciogullari, S., 1991. Reliability and validity of the apathy
Sundet, K., Vaglum, P., Friis, S., McGlashan, T., 2012. Apathy in first episode psychosis evaluation scale. Psychiatry Res. 38, 143–162.
patients: a ten year longitudinal follow-up study. Schizophr. Res. 136 (1–3), 19–24. Mulin, E., Leone, K., Dujardin, K., Delliaux, M., Leentjens, A., Nobili, F., Dessi, B., Tible, O.,
Faerden, A., Finset, A., Friis, S., Agartz, I., Barrett, E.A., Nesvag, R., Andreassen, O.A., Marder, Agüera-Ortiz, L., Osorio, R.S., Yessavage, J., Dachevsky, D., Verhey, F.R., Cruz Jentoft,
S.R., Melle, I., 2010. Apathy in first episode psychosis patients: one year follow up. A.J., Blanc, O., Llorca, P.M., Robert, P.H., 2011. Diagnostic criteria for apathy in clinical
Schizophr. Res. 116 (1), 20–26. practice. Geriatr. Psychiatry 26, 158–165.
Faerden, A., Barrett, E.A., Nesvag, E., Friis, S., Mander, S.R., Ventura, J., Andreassen, O.A., Sockeel, P., Dujardin, K., Devos, D., Denève, D., Destée, A., Defebvre, L., 2006. The Lille
Agartz, I., Melle, I., 2013. Apathy, poor verbal memory and male gender predict apathy rating scale (LARS), a new instrument for detecting and quantifying apathy:
lower psychosocial functioning one year after the first treatment of psychosis. Psychi- validation in Parkinson's Disease. J. Neurol. Neurosurg. Psychiatry 77, 579–584.
atry Res. 210 (1), 55–61. Spitzer, R.L., Kroenke, K., Williams, J.B., 1999. Validation and utility of a self-report version
First, M., Spitzer, R., Gibbon, M., Williams, J., 1997. Structured Clinical Interview for DSM- of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental
IV Axis Disorders (SCID), Clinician Version: User's Guide. Am. Psychiatr. Press, Disorders. Patient Health Questionnaire. JAMA 282 (18), 1737–1744.
Washington, DC. Stuss, D.T., Van Reekum, R., Murphy, K.J., 2000. Differentiation of states and causes of
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) apathy. In: Borod, J. (Ed.), The Neuropsychology of Emotion. Oxford University
for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Press, New York, pp. 340–363.
Kirkpatrick, B., Strauss, G.P., Nguyen, L., Fischer, B.A., Daniel, D.G., Cienfuegos, A., Mander, Zahodne, L.B., Young, S., Kirsch-Darrow, L., Nisenzon, A., Fernandez, H.H., Okun, M.S.,
S.R., 2011. The brief negative symptom scale: psychometric properties. Schizophr. Browers, D., 2009. Examination of the Lille Apathy Rating Scale in Parkinson disease.
Bull. 37 (2), 300–305. Mov. Disord. 24 (5), 667–683.

Please cite this article as: Yazbek, H., et al., The Lille Apathy Rating Scale (LARS): Exploring its psychometric properties in schizophrenia,
Schizophr. Res. (2014), http://dx.doi.org/10.1016/j.schres.2014.04.034