ORIGINAL ARTICLE

Beck Depression Inventory (BDI):

A Reliability and Validity Test in the
Malaysian Urological Population

K F Quek, M.Phi1*, W Y Low,PhD*, A H Razack, FRCS**, C S Loh, FRCS (Urol)***, *Health

Research Development Unit, **Department of Surgery, Faculty of Medicine, University of
Malaya, 50603 Kuala Lumpur, ***Sunway Medical Centre and Gleneagles Intan Medical Centre

Introduction used in other countries and therefore need to be

validated for the local population. The Beck
Benign prostatic hyperplasia (BPH) is a common
Depression Inventory (BDI) has thus become a
disease, which is rarely life-threatening and
commonly used instrument in multicentre,
impose a psychological impact on an individual
international clinical trials to assess psychiatric
such as depression ' . It is no doubt that lower
disorders. The BDI, developed by Beck & Steer,
urinary tract symptoms (LUTS) can be a hindrance
(986) has been widely used in the studies of
to patients and interfere with their daily activities.
depreSSion in many countries both in the community-
The physical symptoms such as the frequency,
based studies as well as the clinical studies4•
nocturia, urgency, urge incontinence and
dribbling can have a profound psychological The study was conducted at the University
affect on the patient's quality of life which can Hospital Kuala Lumpur and was aimed to assess
cause tremendous emotional burden, namely the reliability and validity of the Beck Depression
,
anxiety, stress and a high level of depression ,2. Inventory (BDI) among urological patients.

There are several measures of depression such as the

Beck Depression Inventory (BDI), Hamilton Materials and Methods
Depression Rating Scale3• All these have been widely Study sample
The patients were selected based on the inclusion
and exclusion criteria. For patients with LUTS, the
This article was accepted: 11 June 2001 inclusion criteria were patients who are stable,

Med J Malaysia Vol 56 No 3 Sept 2001 285

ORIGINAL ARTICLE

whereas the exclusion criteria were patients who The psychometric properties of the BDI were
were not treated with surgical and medical assessed in three different samples. Validity and
treatment for lower urinary tract symptoms prior reliability were studied in a group of patients with
to this study. Patients less than 40 years old who LUTS (N=108) and control patients without LUTS
were illiterate were excluded in the study, as were (N=50) and responsiveness was assessed in a group
patients with any chronic and acute diseases. For of patients admitted for transurethral resection of
the control group, the inclusion criteria include the prostate (TURP)(N=79). Management of cases
patients who were free from all major chronic and based on clinical criteria such as medical history,
acute diseases while the exclusion criteria, those physical and rectal examinations was entirely done
with urological problems except renal stone with by a Urologist (AH.R).
minimal severity.

Procedure
Instrument
The study protocol was approved by the Ethics
The BDI is designed to assess the severity of Committee, University Hospital Kuala Lumpur.
depression among the psychiatric patients as well Written informed consent was obtained from all
as possible depression in normal population. This participants after being explained the nature of
inventory measures cognitive, affective, somatic the study. The patients were then required to
symptoms, neurovegetative and endogenous complete the BD!. All questionnaires were self-
aspects of depression 4 • The self-report administered as well as assisted guidance if
questionnaire is rated on a four-point scale necessary by one of the authors (K.F.Q). All
ranging from 0 (no symptom) to 3 (severe patients included in the validity study were
symptom). The items were chosen to assess only scheduled for twelve weeks after the first
the severity of depression, and were not selected administration of the BD!. In the sensitivity to
to reflect any particular theory of depression. The change study, patients completed the
21 symptoms and attitudes assessed by the BDI questionnaires a week prior to surgical treatment
include: (1) Mood; (2) Pessimism; (3) Sense of and were retested at three months follow up.
failure; (4) Self-dissatisfaction; (5) Guilt; (6)
Punishment; (7) Self-dislike; (8) Self-accusation;
Data Analysis
(9) Suicidal ideas; (0) Crying; (1) Irritability;
(12) Social withdrawal; (3) Indecisiveness; (4) The Cronbach's alpha coefficient was used to
Body image change; (5) Work difficulty; (6) assess the internal consistency of the BDI6 • Test-
Insomnia; (7) Fatigability; (8) Loss of appetite; retest reliability was assessed. using the intraclass
(9) Weight loss; (20) Somatic preoccupation; and correlation coefficient (ICC), which is derived
(21) Loss of libido. from the analysis of variance (ANOVA) model.
Values of ICC varies from 1 (perfectly reliable) to
The BDI score was obtained by summing the o (totally unreliableY. Responsiveness was
ratings given by the interviewer for each of the 21 analyzed by calculating the mean difference
items. The overall depression scores range from 0 between BD! before and after TURP and dividing
to 63 and normally divided into four categories. it by the mean standard deviation of the scores
Scores of 0 to 9 were considered within the before TURP (effect size)8.
normal range or asymptomatic, scores of 10 to 15
indicate mild depression, scores of 16 to 23 Mean differences in BDI scores before and after
indicate moderate depression and scores of 24 to TURP was also calculated for each individual item
63 indicate extremely severe depression. The by means of a paired t test or by dividing it by the
questionnaire is easily administered and takes mean standard deviation of stable patients
about 5 - 10 minutes to complete5 • (Guyatt statistic)9.

286 Med J Malaysia Vol 56 No 3 Sept 2001

 BECK DEPRESSION INVENTORY (BDI)

Sensitivity of the BDI was assessed by comparing score (~=10.04, sd=5.4l), glvmg an average
between the means of pre-treatment and post- improvement after TURP on the depression level of
treatment item scores of patients who have 4.39. Overall effect size and responsiveness were
undergone TURF whereas specificity was assessed found to be high indicating improvement in TURP-
by comparing the pre-treatment and post-treatment induced urinary symptoms in these patients.
item scores in subjects rated as control.
In terms of treatment responsiveness, sensitivity
and specificity of the instrument was evaluated by
Results comparing the change between baseline and end
point scores following treatment. All items of the
237 patients consisting of medical, surgical and
BDI demonstrated a high degree of sensitivity and
control group participated in this study. The mean
specificity to the effects of treatment (Table III).
age for the medical group was 63.67 years old
Significant changes were observed across more
(SD=8.57), surgical group was 70.01 years old
than half of the BDI items in the LUTS group. The
(SD=8.17) and control group was 50.04 years old
lowest magnitude of change was noted in item
(SD=12.29). By ethnicity, the Chinese formed the
21. In contrast, except for item 19 (p<O.OOl), none
largest ethnic groups in all the three groups.
of the comparison in the treatment of the control
Patients with LUTS in the surgical group were
subjects approached significance.
significantly more depressed as compared to
control patients (p<O.OOOl). Of the 237, 108 LUTS
The result of the discriminant validity between the
patients has the total BDI mean scores of 8.07
surgical and the control group is shown in Table
(SD=5.58) compared to the 79 patients
IV. Significant differences between the two
undergoing TURP (~=14.43, SD=7.87)(p<0.000l).
groups were noted in the total scores of BDI. The
This showed that prior to treatment, the surgical
surgical group tend to be more depressed (mean
group was significantly more depressed than the
BDI=14.43) than the control group (mean
medical and control patients.
BDI=7.76) who were asymptomatic.
Internal consistency for the BDI was high for all
the items of BDI indicating a high level of
Discussion
homogeneity among items in the scale. The
internal consistency showed the resulting values The psychometric properties of the BDI validated
of Cronbach's alpha for the scale when individual here and other countries showed that they are
item were excluded from the analysis. Test-retest virtually identical with respect to their
reliability was assessed in 108 patients after 12 measurement properties 4 • Although test-retest
weeks interval and the total scores of BDI had an exhibited statistical significant at p<0.05 at some
ICC of 0.85 (p<O.OOl) (Table 1). The total scores items of BDI, this is not surprising because the
of BDI before (8.07) and after (7.67) showed most test-retest was done at 12 weeks interval, thus
of the patients were not depressed according to allowing the urinary symptoms and depression to
the classification of the severity of BDI. improve or worsen. In contrast, the minimal
changes of symptoms and depression would
Responsiveness was assessed in patients occur if test-retest were done at one week, two
undergoing TURP. Table II showed the pre and week or the latest one month after the initial
post scores, mean difference, effect size, and the assessment. The reasons for the test-retest to be
Guyatt statistic for individual items, and for total carried out at twelve weeks interval because most
scores. Before treatment, patients in the surgical patients in the TURP group would then have
group were depressed and subsequently improved achieved the maximum benefit or total symptom
after treatment. The mean pre intervention score on improvement and this would make comparisons
the total BDI (~= 14.43, sd=7.87) was significantly between the patients with LUTS and patients
higher (p<O.OOOl) than the mean post intervention undergoing TURP much easier.

Med J Malaysia Vol 56 No 3 Sept 2001 287

I'V
<X> Table I 0
;>0
<X>
Validity and Reliability: G)

Mean Test-Retest Score, Intraelass Correlation CoeHicient and z

;!::
Internal Consistency for Individual BDI items »
;>0
-I
BDI Intraelass Internal Mean SD Mean SD Mean SD 95% Confidence Interval nr -
Items Correlation Consistency** Test Retest DiHerence Lower Upper rn

Coefficient* (ICC) Score Score

1 0.67 0.67 0.32 0.47 0.30 0.50 0.028 0.48 -0.12 0.06
2 0.87 0.87 0.32 0.47 0.45 0.85 0.13 0.88 -0.04 0.30
3 0.82 0.82 0.14 0.44 0.18 0.47 0.037 0.36 -0.03 0.11
4 0.73 0.73 0.54 0.60 0.56 0.59 0.019 0.55 -0.08 0.12
5 0.69 0.69 0.35 0.50 0.34 0.53 ·0.0093 0.50 -0.10 0.09
6 0.75 0.75 0.28 0.53 0.21 0.63 0.065 0.52 -0.16 0.03
7 0.66 0.66 0.27 0.44 0.21 0.41 0.056 0.43 -0.14 0.03
8 0.73 0.73 0.46 0.62 0.39 0.56 0.074 0.54 -0.18 0.03
9 0.56 0.56 0.028 0.16 0.019 0.13 0.0093 0.17 -0.04 0.02
10 0.69 0.69 0.34 0.95 0.33 0.95 0.0093 0.92 -0.18 0.17
11 0.70 0.70 0.72 1.05 0.78 0.98 0.056 0.97 -0.13 0.24
12 0.68 0.68 0.21 0.41 0.29 0.45 0.074 0.43 -0.007 0.15
13 0.72 0.72 0.29 0.53 0.35 0.62 0.065 0.53 -0.04 0.17
14 0.79 0.79 0.16 0.44 0.18 0.43 0.019 0.36 -0.05 0.09
15 0.84 0.84 0.47 0.63 0.56 0.63 0.093 0.46 0.004 0.18
16 0.77 0.77 0.59 0.71 0.55 0.67 0.046 0.60 -0.16 0.07
~
CD 17 0.84 0.84 0.43 0.60 0.51 0.66 0.074 0.47 -0.01 0.16
CL
'-
18 0.60 0.61 0.24 0.45 0.16 0.37 0.083 0.43 -0.17 -0.0003
~
Q
0"
19 0.83 0.84 0.20 0.51 0.13 0.43 0.074 0.35 -0.14 -0.006
'<
V>
c' 20 0.76 0.77 0.53 0.70 0.39 0.61 0.14*** 0.57 -0.25 -0.03
Q: 21 0.82 0.82 1.02 0.96 0.86 0.95 0.16*** 0.74 -0.30 -0.02
tJ1 Total scores 0.85 0.85 8.07 5.58 7.67 5.69 0.41 *** 4.08 -0.37 -0.37
0-
z0 p<O.OOI for aJlICCsI*)
w Cronbach's alpha: note that Cronbach's alpha value given for each item represent the effect of removing that item from the calculation of the alpha value leg if
C/)
CD item 1 is omitted, the resulting value for the scale is 0.67 if item 2 is omitted, it is 0.87, and so forth)(**)
~
I'V t test for paired comparisons significant 1***)
0
0
 BECK DEPRESSION INVENTORY (BDI)

Table II
Responsiveness: Mean Scores Before and After TURP, Effect Size and GuyattStatistic
BDI PreTURP PostTURP
Items Mean SD Mean SD Mean Differences* SD Effect Size Guyatt statistic
1 0.63 0.48 0.32 0.47 0.32 0.67 0.67 0.68
2 0.77 0.95 0.57 0.89 0.20* 1.36 0.21 0.43
3 0.27 0.55 0.076 0.31 0.19 0.64 0.35 0.43
4 1.01 0.61 0.66 0.57 0.35 0.89 0.57 0.58
5 0.68 0.54 0.38 0.49 0.30 0.79 0.56 0.60
6 0.49 0.57 0.23 0.50 0.27 0.73 0.47 0.51
7 0.59 0.49 0.21 0.41 0.38 0.67 0.78 0.86
8 0.67 0.63 0.37 0.56 0.30 0.79 0.48 0.48
9 0.025 0.16 o o 0.025* 0.16 0.16 0.16
10 0.63 1.18 0.39 0.97 0.24* 1.59 0.20 0.25
11 0.81 1.03 0.51 0.84 0.30 1.36 0.29 0.29
12 0.82 0.69 0.63 0.60 0.19* 0.95 0.28 0.46
13 0.80 0.67 0.67 0.59 0.13* 0.95 0.19 0.25
14 0.28 0.48 0.21 0.41 0.063* 0.63 0.13 0.14
15 1.02 0.72 0.90 0.65 0.13* 0.94 0.18 0.21
16 1.15 0.75 0.77 0.68 0.38 0.94 0.56 0.54
17 0.61 0.54 0.47 0.57 0.14* 0.84 0.26 0.23
18 0.46 0.59 0.29 0.46 0.16* 0.77 0.27 0.36
19 0.51 0.77 0.23 0.42 0.28 0.90 0.36 0.55
20 0.77 0.75 0.33 0.55 0.44 0.84 0.59 0.63
21 1.51 1.01 1.54 0.94 0.038* 1.44 0.038 0.04
Total scores 14.43 7.87 10.04 5.41 4.39 5.90 0.56 0.79
Effect size=Mean difference/SD PreTURP
Guyatt statistics=Mean difference/SD of stable LUTS patients (medication group)
* t test for paired comparisons not significant

BDI with its ability to discriminate between These findings provide substantial assurance that
patients with LUTS and those without showed high scores obtained using the BDI are reliable. The
levels of sensitivity and specificity likewise with the validity and reliability scores were consistent with
discriminant validity. The reasonably large effect scores of BDI validated in various countries5 •
size obtained indicated a high degree of sensitivity
to change, likewise with other studies8 •
Conclusion
This study showed that the LUTS significantly This study showed that the intraclass
contributed to the increased of the correlation coefficient for total scores of the
depression level especially those with severe items of the BDI indicates high intraclass
LUTS. The improvement of the symptoms reliability and the high Cronbach's alpha
following treatment resulted in the coefficient suggest that BDI exhibits good
improvement of the depression. internal consistency. BDI seems to be a valid,

Med J Malaysia Vol 56 No 3 Sept 2001 289

ORIGINAL ARTICLE

Table III
BDI Items Characteristics of Patients UndergoinPi TURP
and the Control Group: Sensitivity and Sped icity
Sensitivity N Mean Changes SEM t statistics p value
Items
1 79 0.32 0.075 4.19 0.0001
2 79 0.20 0.15 1.32 0.19
3 79 0.19 0.072 2.63 0.01
4 79 0.35 0.10 3.53 0.001
5 79 0.30 0.089 3.42 0.001
6 79 0.27 0.082 3.24 0.002
7 79 0.38 0.075 5.07 0.0001
8 79 0.30 0.089 3.42 0.001
9 79 0.025 0.018 1.42 0.159
10 79 0.24 0.18 1.35 0.182
11 79 0.30 0.15 1.98 0.05
12 79 0.19 0.11 1.78 0.079
13 79 0.13 0.11 1.18 0.241
14 79 0.063 0.071 0.90 0.373
15 79 0.13 0.11 1.20 0.234
16 79 0.38 0.11 3.60 0.01
17 79 0.14 0.095 1.47 0.146
18 79 0.16 0.087 1.89 0.063
19 79 0.28 0.10 2.74 0.008
20 79 0.44 0.095 4.67 0.001
21 79 0.038 0.16 0.235 0.815
Total scores 79 4.39 0.66 6.61 0.0001
1 50 0.02 0.073 0.275 0.785
2 50 0.12 0.84 1.43 0.159
3 50 0 0.081 0 1
4 50 0.04 0.075 0.531 0.598
5 50 0.04 0.049 0.814 0.42
6 50 0.14 0.099 1.41 0.164
7 50 0.04 0.057 0.704 0.485
8 50 0.16 0.072 2.22 0.031
9 50 0.02 0.02 1 0.322
10 50 0.18 0.12 1.46 0.151
11 50 0 0.16 0 1
12 50 0.10 0.059 1.70 0.096
13 50 0.06 0.053 1.14 0.261
14 50 0 0.064 0 1
15 50 0.10 0.082 1.22 0.229
16 50 0.04 0.075 0.53 0.598
17 50 0 0.076 0 1
18 50 0.02 0.053 0.375 0.709
19 50 0.22 0.082 2.67 0.01
20 50 0.06 0.088 0.685 0.497
21 50 0.08 0.075 1.07 0.29
Total scores 50 0.68 0.73 0.925 0.36

290 Med J Malaysia Vol 56 No 3 Sept 2001

 BECK DEPRESSION INVENTORY (BDI)

Table IV
BDI Items Characteristics: Discriminant Validity
Items Pre TURP Patients Control
Mean SEM Mean SEM Mean SEM 95% Confidence Interval P value
Difference Lower Higher
1 0.63 0.055 0.46 0.082 0.17 0.099 0.022 0.37 0.08
2 0.77 0.11 0.42 0.099 0.35 0.15 0.064 0.64 0.05
3 0.27 0.062 0.22 0.077 0.046 0.099 0.15 0.24 0.64
4 1.01 0.069 0.54 0.087 0.47 0.11 0.25 0.69 0.0001
5 0.68 0.061 0.30 0.065 0.38 0.093 0.20 0.57 0,0001
6 0.49 0.065 0.32 0.097 0.17 0.11 0.048 0.39 0.12
7 0.59 0.056 0.28 0.064 0.32 0.085 0.15 0.48 0.0001
8 0.67 0.071 0.50 0.096 0.17 0.12 0.062 0.40 0.15
9 0.025 0.018 0 0 0.025 0.018 0.01 0.061 0.16
10 0.63 0.13 0.26 0.11 0.37 0.17 0.037 0.71 0.05
11 0.81 0.11 0.70 0.15 0.11 0.19 0.26 0.48 0.56
12 0.82 0.078 0.30 0.065 0.52 0.11 0.30 0.74 0.0001
13 0.80 0.075 0.30 0.66 0.50 0.11 0.28 0.71 0.0001
14 0.28 0.054 0.10 0.051 0.18 0.074 0.031 0.33 0.05
15 1.02 0.08 0.48 0.077 0.54 0.12 0.31 0.78 0.0001
16 1.15 0.085 0.52 0.087 0.63 0.13 0.38 0.88 0.0001
17 0.61 0.061 0.48 0.077 0.13 0.098 0.066 0.32 0.19
18 0.46 0.067 0.16 0.052 0.30 0.085 0.13 0.46 0.001
19 0.51 0.086 0.36 0.086 0.15 0.13 0.11 0.40 0.26
20 0.77 0.084 0.62 0.11 0.15 0.14 0.12 0.42 0.26
21 1.51 0.11 0.50 0.10 1.01 0.15 0.70 1.31 0.0001
Total scores 14.43 0.88 7.76 0.74 6.67 1.15 4.39 8.95 0.0001

useful and a reliable instrument for assessing Acknowledgements

the severity of LUTS in the Malaysian
The authors would like to express their gratitude
urological population.
to the University Malaya's Research and
Development Management Unit for providing the
research grant for this study.

Med J Malaysia Vol 56 No 3 Sept 2001 291

ORIGINAL ARTiClE

1. Quek KF, Low WY, Razack AH, Loh CS. 6. Cronbach L]. Coefficient alpha and the internal.
Psychological effects of treatment for lower urinary structure of test. Psychometrika 1951;16: 297.
tract symptoms. British Journal Urology
7. Deyo RA, Dichr P, Patrick DL. Reproducibility and
International 2000; 86: 630-33.
responsiveness of health status measures. Control
2. Lukacs B, Leplege A, McCarthy C, Comet D. Clinical Trials 1991; 12 (supp 1): 142S-158S.
Construction and validation of a BPH specific
8. Cohen]. Statistical Power analysis for the
health related quality of life scale, for medical
Behavioral Sciences. New York: Academic Press.
outcome research studies. In Proceedings of the
1977; 75-105.
Second International Consultation on Benign
Prostatic Hyperplasia (BPH) (eds ATK. Cockett , S. 9. Guyatt G, Walter S, Norman G. Measuring
Khoury, Y. Aso et at.), Paris.: SCI. 1994; 139-43. changes overtime: assessing the usefulness of
evaluative instruments. Journal Chronic Disease
3. Hamilton M. A rating scale for depression.
1987; 40: 171-78.
Journal of Neurology, Neurosurgery and
Psychiatry; 1960; 23: 56-61.
4. Steer RA, Beck AT and Garrison B. Applications of
the Beck Depression Inventory; in Sartorius N, Ban
TA (eds): Assessment of Depression. Berlin,
Springer. 1986; 121-42.
5. Beck AT and Steer RA. Beck Depression Inventory:
Manual. The Psychological Corporation, Harcourt
Brace Jovanich Inc., San Diego. 1986.

292 Med J Malaysia Vol 56 No 3 Sept 2001