Journal of Psychosomatic Obstetrics & Gynecology

ISSN: 0167-482X (Print) 1743-8942 (Online) Journal homepage: http://www.tandfonline.com/loi/ipob20

Validation of the Edinburgh Postnatal Depression

Scale (EPDS) in Chilean Postpartum Women

E. Jadresic, R. Araya & C. Jara

To cite this article: E. Jadresic, R. Araya & C. Jara (1995) Validation of the Edinburgh Postnatal
Depression Scale (EPDS) in Chilean Postpartum Women, Journal of Psychosomatic Obstetrics &
Gynecology, 16:4, 187-191, DOI: 10.3109/01674829509024468

To link to this article: http://dx.doi.org/10.3109/01674829509024468

Published online: 07 Jul 2009.

Submit your article to this journal

Article views: 42

View related articles

Citing articles: 5 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ipob20

Download by: [Deakin University Library] Date: 16 March 2016, At: 05:55
 J. Psychosom. Obstet. Gynecol. 16 (1995) 187- 191

Validation of the Edinburgh

postnatal depression scale (EPDS) in
Chilean postpartum women
E.Jadresic', R. Araya2 and C.Jara'

'Departamento de Psiquiatria y Salud Mental, Divisi6n Ciencias Medicas Norte, Universidad de

Chile; and 2Unidad de Evaluacion, Divisi6n de Programas, Ministerio de Salud, Chile

Key words: VALIDATION,

DEPRESSION,
POSTPARTUM
DEPRESSION,
EDINBURGH
POSTNATAL
DEPRESSION
SCALE
Downloaded by [Deakin University Library] at 05:55 16 March 2016

ANALEIS

ABSTRACT physical check-ups throughout their pregnancies,

A validity study of the Edinburgh Postnatal Depression
Scale (EPDS) against the Research Diagnostic Criteria
(RDC)was carried out on a sample ofwomen attending a
health care center in Santiago. One hundred and eight
middle-class mothersfilled in the EPDS and were later
interviewed by the main author using the Psychiatric
Assessment Schedule (PAS). The internal consistency of
the EPDS was reasonably good (Cronbach's a 0.77).
Validity coefiientsfor the scale were calculated to determine
the best caselnon-case threshold which was found to be
9/10. Sensitivity, spec8city and positive predictive value
for this threshold were loo%, 80% and 37% correspond-
ingly. The same best cut-ojpoint (9/10) was found by
other Chilean investigators in a sample of working-class
women. T h e EPDS was shown to be a usejhl screening
instrumentf o r postnatal depression in these settings.
INTRODUCTION
In Chile, a country with a population of nearly 14
million inhabitants, 250 OOO deliveries take place
every year. Maternal and infant mortality and mor-
bidity rates have declined markedly over the last
decades. Today most Chilean women have routine
delivery (97.4% of all deliveries take place in
hospital) and the puerperium'. However, the
psychological well-being of mothers has not been a
matter of concern.
Two recent Chilean studies found incidence rates
of 9.2%* and for postpartum depression,
similar to incidence rates found in developed
countries, which range from 7.1%4 to 14%5.A pre-
vious validation study of a Spanish version of the
Edinburgh Postnatal Depression Scale (EPDS) con-
ducted in Chilean working-class women6, found it
to be a useful instrument to detect postnatal
depression with a sensitivity of 86%, specificity of
80% and a positive predictive value of 56% at a
threshold of 9/10. However, this study did not use
a standardized interview and did not apply
sufficiently strict diagnostic criteria. One of these
Chilean studies* also suggests that health profes-
sionals fail to detect depressed mothers, despite
being in regular contact with them.
The purpose of this study was to estimate the
external validity of our own translation into Spanish
of the EPDS against the criterion of the Research
Diagnostic Criteria (RDC) in a sample of middle-
class women.

Correspondence to: Dr Enrique Jadresic, Departamento de Psiquiatria y Salud Mental, Divisi6n Ciencias Medicas Norte, Universidad
de Chile, Avenida La Paz 1003,Santiago, Chile

187
 EPDS in Chile
METHOD
Instruments
The EPDS’ is a simple, self-administered scale,
which can be completed in about 5 min. It consists
of ten questions, each having four possible answers
which are scored 0, 1 , 2 and 3 according to increas-
ing severity of the symptom. Seven of the items are
worded in such a manner that the presence of the
symptom is acknowledged by a positive response.
The other three items are worded in the opposite
manner, and, therefore, their answering codes were
reversed before the analysis. The EPDS was
translated into Spanish by E.J. and then back
translated into English by R.A. Both are Chilean-
born but British-trained psychiatrists, working at
Downloaded by [Deakin University Library] at 05:55 16 March 2016
present in Chile. The back-translation of our
Spanish version was found to be almost identical to
the English version by a native English-speaking
psychiatrist (Prof Rumar Kamesh, Institute of
Psychiatry, London). In addition, pilot interviews
with our final EPDS version (see Appendix) in 12
childbearing women, showed its acceptability and
ease of administration.
A standardized psychiatric interview, the
Psychiatric Assessment Schedule (PAS)8, based on
the Present State Examination (PSE)9, was used as
the interviewing instrument to derive the RDC
criteriat0 by the principal author who had been
trained to use it at the Institute of Psychiatry in
London.
Procedure
The study was conducted on a sample of middle-
class women who attended the antenatal clinic at a
University Hospital Out-patient Clinic (CEDIUC)
in Santiago. All pregnant women who attended the
Clinic for their regular check-ups on Mondays and
Thursdays over a period of 2 months were invited
to take part in this study. These women were
followed up and interviewed again 2-3 months
after giving birth. Data presented here correspond
to this period. All interviews were conducted
blindly to the EPDS scores by E.J.
Cases of depressive illness (minor or major) were
defined according to RDC which was used as the
‘gold standard’ throughout. The following indices
were calculated: Cronbach’s a (as a measure of
internal consistency), sensitivity (the proportion of
‘true’ cases correctly identified by the question-
188
Jadresic, Araya and Jara
naire), specificity (the proportion of correctly identi-
fied normals), overall misclassification rate (the
percentage of misclassified respondents), positive
predictive value (the probability that a high scorer
will be found to be case at subsequent examination)
and negative predictive value (the probability that a
low scorer will be found to be a non-case).
RESULTS
Characteristics of the sample
All of the 108 mothers meeting the inclusion
criteria agreed to participate. The mean age of the
sample was 27.7 years (SD 5.3) and most women
had a stable partner (90%). Twelve per cent were
professionals and almost half (44%) were
housewives. Virtually all (99%) women included in
the sample had secondary education and 16% of
them had further education. Forty-eight per cent
were primiparous and only 2% had three or more
children. Eighty-six per cent had normal deliveries,
11% Caesarean sections and a further 3% had
forceps deliveries.
The EPDS as a screening instrument
The internal consistency of the scale estimated by
Cronbach’s a was 0.7722, increasing to 0.7727 if
item 10was deleted and decreasing to 0.7290 if item
8 was deleted. The lowest corrected (excluding the
item itself from the total) item-total correlation
was for item 10 (0.254) whereas the highest was for
item 8 (0.649). Altogether, 11 women met RDC
criteria for depressive illness. The validity
coefficients for various cut-off points according to
our criterion (RDC) are shown on Table 1.
The comparison of these indices suggests that the
optimum case/non-case threshold score (best trade-
off between sensitivity and specificity, favoring the
former) is 9/10. Using this threshold, the sensitivity
and specificity were found to be 100% and 80%
respectively. If the threshold score was lowered to
8/9 all mothers with depression would be correctly
identified but the proportion of correctly identified
normals would decrease considerably (67%). If the
chosen threshold was set at 10/11 the sensitivity
would decrease to 82% and specificity would
increase to 87%.
DISCUSSION
The value of the EPDS as a tool for the identi-
fication of depression in postpartum women has
J. Psychosom. Obstet. Gynecol.
 EPDS in Chile
Table 1 Validity coefficients (%) of the Chilean translation into Spanish of the Edinburgh postnatal depression scale
(EPDS) according to Research Diagnostic Criteria (RDC)
Threshold 8/9 9/10
Sensitivity 100 100
Specificity 67 80
Positive predictive value 26 37
Negative predictive value 100 100
Misclassification rate 30 18
been underlined in a number of studies conducted
in English-speaking c ~ u n t r i e s ~ ~To
- ' ~our
. knowl-
edge, apart from a previous Chilean translation into
Spanish6, no other translation into Spanish of the
Downloaded by [Deakin University Library] at 05:55 16 March 2016
EPDS has been specifically validated. As increasing
evidence23vt4shows, the incidence and prevalence of
postpartum depression does not differ substantially
in develop-ing and developed nations. This latter
finding together with the lack of resources to deal
with psychiatric problems in poorer countries,
empha-sizes the need for brief, simple and
inexpensive means of screening for psychiatric
morbidity in the puerperium.
Once a scale is translated into a language
different from the original, it is necessary to carry
out validity studies to establish the optimal
threshold for that newly translated instrument used
in a particular sociocultural setting. Although it was
not the purpose of this study to assess the reliability
for this scale, the internal consistency as measured
by the Cronbach a was reasonably good. The
lowest total-item correlation for item 10 ('The
thought of harming myself has occurred to me')
might be explained by the content of this item,
assessing a symptom indicative of great severity. It is
noteworthy that the optimum threshold score
found in our study involving middle-class women
(9/10) was the same as the cut-off point found in a
previous study with Chilean working-class women6.
Thus, the cut-off point for case definition with the
EPDS was found to be the same for women from
different socioeconomic backgrounds in Chile.
The validity coeficients found in our study are
REFERENCES
1. Jadresic E. Depresih Postparto. Rev Chil Neuro-
Psiquiat 1990;28:147-58.
2. Jadresic E, Jara C, Miranda M, Arrau B, Araya R.
Trastornos emocionales en el embarazo y el puer-
perio: estudio prospectivo de 108 mujeres. Rev Chil
J. Psychosom. Obstet. Gynecol.
Jadresic, Araya and Jara
~~
10/11 11/12 12/13
82 73 55
87 91 94
43 47 50
98 97 95
14 11 10
fairly good and similar to the corresponding values
found by Murray and Carothers12 using the same
diagnostic criteria and threshold score in their
English validation study.
It is important to consider some methodological
limitations in the present study. The sample and the
number of depressed women were small, all
interviews were conducted by the same person and
our sample only included middle-class women.
However, interviewing middle-class people is
rather difficult in this country and, therefore, it was
not possible to increase the sample size within the
budget available for this study. O n the other hand,
the interviewer had extensive training with the
instruments used and as he was the only rater this
reduced a possible inter-rater variability
In conclusion, the EPDS appears to be a useful
and valid instrument to screen for postpartum
depression in Chilean women. Its widespread use
could contribute to the detection of those women
who would benefit from help; thus diminishing
their suffering and the consequences of this
disorder upon themselves and their families.
ACKNOWLEDGEMENTS
This work was supported by a Grant from
FONDECYT (1130/90).We thank the personnel of
the health center involved in this study for all the
help and support provided throughout. Finally our
special thanks to the mothers for giving us some of
their precious time during a period of many
demands and life changes.
Neuro-Psiquiat 1992;30:99- 106.
3. Alvarado R, Rojas M, Monardes J, Neves E, Olea E,
Perucca E, Vera A. Cuadros depresivos en el post-
parto y variables asociadas en una cohorte de 125
mujeres embarazadas. Rev Psiquiat 1992;3-4:
1168-76.
189
 EPDS in Chile
4. Cutrona CE. Causal attributions and perinatal
depression. J Abnorm Psychol 1983;92:161-72.
5. &mar R, Robson KM. A prospective study of
emotional disorders in childbearing women. Br J
Psychiatr 1984; 1M35-47.
6. Alvarado R, Vera A, Rojas M, Olea E, Monardes J,
Neves E. La Escala de Edinburgo para la detecci6n de
cuadros depresivos en el postparto. Rev Psiquiat
1992;3 - 4: 1177- 81.
7. Cox JL, Holden JM, Sagovsky R. Detection of
postnatal depression: Development of the 10-item
Edinburgh Postnatal Depression Scale. Br J Psychiatr
1987;1m 7 8 2 - 6.
8.Dean C , Surtees PG, Sashidharan SP Comparison of
Research Diagnostic Systems in an Edinburgh Com-
munity Sample. Br J Psychiatr 1983;142:247-56.
9. Wing JK, Nixon JN, Mann SA, Leff JI! Reliability of
Downloaded by [Deakin University Library] at 05:55 16 March 2016
the PSE (9th Edition) used in a population survey.
Received 29 August 1993; accepted 21 December 1994
APPENDIX
Edinburgh Postnatal Depression Scale’
As you have recently had a baby, we would like to
know how you are feeling. Please UNDERLINE
the answer which comes closest to how you have
felt IN T H E PAST 7 DAYS, not just how you feel
today
In the past 7 days:
I have been able to laugh and see the funny
(1)
side of things
as much as I always could
not quite so much now
definitely not so much now
not at all
(2) I have looked forward with enjoyment to
things
as much as I ever did
rather less than I used to
definitely less than I used to
hardiy at all
* ( 3 ) I have blamed myself unnecessarily when
things went wrong
yes, most of the time
yes, some of the time
not very often
no, never
1Yt9
Jadresu, Araya and Jara
Psychol Med 1977;7:505- 16.
10. Spitzer R, Endicott J, Robins E. Research Diagnostic
Criteria. Instrument No. 58. New York New York
State Psychiatric Institute 1975.
11. Harris B, Huckle P, Thomas R, Johns S, Fung H.
The use of rating scales to identify postnatal
depression. Br J Psychiatr 1989;154:813- 17.
12. Murray L, Carothers A. The validation of the
Edinburgh Postnatal Depression Scale on a
community sample. Br J Psychiatr 1990;157:288-90.
13. Glaze R, Cox JL. Validation of a Computerized
Version of the 10-Item (Self-Rating) Edinburgh
Posmatal Depression Scale. J Affect Disord 1991;
22( 1-2):73 -7.
14. Cox JL. Postnatal depression: a comparison of
African and Scottish women. SOC Psychiatr 1983;
18:25 - 8.
(4) I have been anxious or worried for no good
reason
no, not at all
hardly ever
yes, sometimes
yes, very often
*(5) I have felt scared or panicky for no very
good reason
yes, quite a lot
yes, sometimes
no, not much
no, not at all
*(6) Things have been getting on top of me
yes, most of the time I haven’t been able
to cope at all
yes, sometimes I haven’t been coping as
well as usual
no, most of the time I have coped quite
well
no, I have been coping as well as ever
*(7) I have been so unhappy that I have had
dificulty sleeping
yes, most of the time
yes, sometimes
not very often
no, not at all
*(8) I have felt sad or miserable
yes, most of the time
J. Psychosom. Obstet. Gynecol.
 EPDS in Chile Jadresu, Araya andJara

yes, quite often (4) He estado nerviosa o inquieta sin tener

not very often motivo
no, not at all no, nunca
casi nunca
*(9) I have been so unhappy that I have been si, a veces
crying si, con mucha frecuencia
yes, most of the time
yes, quite often (5) He sentido miedo o he estado asustadiza sin
only occasionally tener motivo
no, never si, bastante
si, a veces
*(lo) The thought of harming myself has
no, no mucho
occurred to me
no. nunca
yes, quite often
sometimes (6) Las cosas me han estado abrumando
hardly ever si, la mayor parte del tiempo no he podido
never hacer las cosas en absoluto
Downloaded by [Deakin University Library] at 05:55 16 March 2016

si, a veces no he podido hacer las cosas tan

Response categories are scored 0, 1, 2 and 3
bien como siempre
according to increased severity of the symptom.
no, la mayor parte del tiempo he hecho
Items marked with an asterisk are reverse scored
las cosas bastante bien
(i.e. 3, 2, 1 and 0). The total score is calculated by
no, he estado haciendo las cosas tan bien
adding together the scores for each of the ten items.
como siempre

Escala de Edinburgo (Spanish version) (7) Me he sentido tan desdichada que he tenido
dificultades para dormir
Como usted hace poco tuvo un bebC, nos gustaria si, la mayor parte del tiempo
saber como se ha estado sintiendo. Por favor si, a veces
SUBRA.. la respuesta que mas se acerca a como se no con mucha frecuencia
ha sentido en 10s tiltimos 7 dias. no, nunca
En 10s ~ l t i m o 7s dias: (8) Me he sentido triste o desgraciada
si, la mayor parte del tiempo
(1) He sido capaz de reirme y ver el lado si, bastante a menudo
divertido de las cosas
no con mucha frecuencia
tanto como siempre
no, nunca
no tanto ahora
mucho menos ahora (9) Me he sentido tan desdichada que he estado
no, nada llorando
si, la mayor parte del tiempo
(2) He disfrutado mirar hacia adelante si, bastante a menudo
tanto como siempre
s610 ocasionalmente
menos que antes
no, nunca
mucho menos que antes
casi nada (10) Se me ha ocurrido la idea de hacerme daiio
si, bastante a menudo
(3) Cuando las cosas han salido ma1 me he a veces
cuipado a mi misma innecesariamente
casi nunca
si, la mayor parte del tiempo
nunca
si, a veces
no con mucha frecuencia
no, nunca

J. Psychosom. Obstet. Gynecol. 191