Cohort Study Of Depressed Mood During Pregnancy And After Childbirth

Abstract

Objective: To follow mothers' mood through pregnancy and after childbirth and compare reported symptoms of depression at each stage.
Design: Longitudinal cohort study.
Setting: Avon.
Participants: Pregnant women resident within Avon with an expected date of delivery between 1 April 1991 and 31 December 1992.
Main outcome measures: Symptom scores from the Edinburgh postnatal depression scale at 18 and 32 weeks of pregnancy and 8 weeks and 8 months postpartum. Proportion of women above a threshold indicating probable depressive disorder.
Results: Depression scores were higher at 32 weeks of pregnancy than 8 weeks postpartum (difference in means 0.88, 95% confidence interval 0.79 to 0.97). There was no difference in the distribution of total scores or scores for individual items at the four time points. 1222 (13.5%) women scored above threshold for probable depression at 32 weeks of pregnancy, 821 (9.1%) at 8 weeks postpartum, and 147 (1.6%) throughout. More mothers moved above the threshold for depression between 18 weeks and 32 weeks of pregnancy than between 32 weeks of pregnancy and 8 weeks postpartum.
Conclusions: Symptoms of depression are not more common or severe after childbirth than during pregnancy. Research and clinical efforts need to be moved towards understanding, recognising, and treating antenatal depression.

Full text

Cohort study of depressed mood during pregnancy and
after childbirth
Jonathan Evans, Jon Heron, Helen Francomb, Sarah Oke, Jean Golding, on behalf of the
Avon Longitudinal Study of Parents and Children Study Team
Abstract
Objective To follow mothers’ mood through
pregnancy and after childbirth and compare reported
symptoms of depression at each stage.
Design Longitudinal cohort study.
Setting Avon.
Participants Pregnant women resident within Avon
with an expected date of delivery between 1 April
1991 and 31 December 1992.
Main outcome measures Symptom scores from the
Edinburgh postnatal depression scale at 18 and 32
weeks of pregnancy and 8 weeks and 8 months
postpartum. Proportion of women above a threshold
indicating probable depressive disorder.
Results Depression scores were higher at 32 weeks of
pregnancy than 8 weeks postpartum (difference in
means 0.88, 95% confidence interval 0.79 to 0.97).
There was no difference in the distribution of total
scores or scores for individual items at the four time
points. 1222 (13.5%) women scored above threshold
for probable depression at 32 weeks of pregnancy,
821 (9.1%) at 8 weeks postpartum, and 147 (1.6%)
throughout. More mothers moved above the
threshold for depression between 18 weeks and 32
weeks of pregnancy than between 32 weeks of
pregnancy and 8 weeks postpartum.
Conclusions Symptoms of depression are not more
common or severe after childbirth than during
pregnancy. Research and clinical efforts need to be
moved towards understanding, recognising, and
treating antenatal depression.
Introduction
Women are more vulnerable to psychiatric illness dur-
ing the postnatal period. The rate of psychiatric admis-
sion is increased postnatally, mostly because of the
raised risk of psychosis in the first month after
childbirth.1In 1968, Pitt described a syndrome of
“atypical depression following childbirth,”2although
there is now no evidence that there is a categorical dif-
ference between depression after childbirth and
depression at other times. The prevalence of non-
psychotic depressive illness in the postnatal period is
similar to that in the general population.34
Nevertheless, postnatal depression has become a
focus of concern. General practitioners, health visitors,
and others are exhorted to recognise and treat this
condition. The consequences of postnatal depression
to the child, mother, and family may include neglect of
the child, family breakdown, self harm, and suicide.
However, the more common consequences include
emotional and behavioural problems, and cognitive
delay in the children of depressed mothers.56
In contrast, depression during pregnancy has been
relatively neglected. Indeed, pregnancy was thought to
protect women against depression. Studies of antenatal
psychopathology have mostly examined antenatal
mood as a predictor of postnatal depression.7–10 Watson
et al found that in 23% of those who had postnatal
depression this had started during pregnancy.11
Depressed mood during pregnancy has also been
associated with poor attendance at antenatal clinics,
substance misuse, low birth weight, and preterm deliv-
ery.12 13 Psychopathological symptoms during preg-
nancy have physiological consequences for the fetus,
which may explain some of these effects.14
We studied mood through pregnancy and after
childbirth using prospectively gathered data from a
cohort of 14 000 women. We compared depressive
symptom score, the pattern of reported symptoms, and
the proportion of mothers above a threshold
indicating probable depressive disorder at each stage.
Participants and methods
The Avon longitudinal study of parents and children
enrolled women resident in Avon who were in the
early stages of pregnancy with an expected date of
delivery between 1 April 1991 and 31 December 1992.
We recruited 14 541 women, of whom 13 799 had off-
spring surviving to 12 months old. Further details of
the study aims and design are available (www.ich.
bris.ac.uk/alspacext/). Ethical approval was obtained
from the study’s ethics committee and local ethics
committees.
Women completed the Edinburgh postnatal
depression scale15 and the Crown Crisp experiential
index16 as part of a series of postal questionnaires. We
present here data for questionnaires completed at 18
weeks and 32 weeks of pregnancy and at 8 weeks and 8
months postpartum. The Edinburgh postnatal depres-
sion scale focuses on the cognitive and affective
features of depression rather than somatic symptoms.
It is the only self report scale that has been validated
for use postnatally and during pregnancy.17 18 The scale
cannot in itself confirm a diagnosis of depression; how-
ever, a score above 12 is widely used to indicate prob-
able depressive disorder. Validation of the scale showed
that all those found to have definite major depression
when interviewed, had scored above 12 on the scale.
Use of this threshold gave an overall sensitivity of 86%
and specificity of 78% for all forms of depression.15
Statistical methods
We calculated mean Edinburgh postnatal depression
scale scores for all responders and those responding at
all four time points. Although the scores were
negatively skewed, the differences in scores were
symmetric and normal enough for analysis with paired
ttests. The 95% confidence intervals for the differences
in the mean values were corrected for multiple (six)
comparisons. We plotted frequency histograms of total
scores for each period and compared changes in
Papers
Division of
Psychiatry,
University of
Bristol, Bristol
BS2 8DZ
Jonathan Evans
consultant senior
lecturer
Unit of Paediatric
and Perinatal
Epidemiology,
Division of Child
Health, University
of Bristol, Bristol
BS8 1TQ
Jon Heron
research assistant
Jean Golding
professor of paediatric
epidemiology
Department of
Women’s Health
and Care of the
Newborn, North
Bristol NHS Trust,
Southmead
Hospital,
Westbury-on-Trym,
Bristol BS10 5NB
Helen Francomb
midwife
Mother and Baby
Unit, Barrow
Hospital, Barrow
Gurney, Bristol
BS48 3SG
Sarah Oke
consultant psychiatrist
Correspondence to:
JEvans
j.evans@bristol.
ac.uk
BMJ 2001;323:257–60
257BMJ VOLUME 323 4 AUGUST 2001 bmj.com

symptom score between 18 weeks and 32 weeks of
pregnancy with changes between 32 weeks of
pregnancy and 8 weeks postpartum using a paired
ttest. Mean scores for each item of the Edinburgh
postnatal depression scales at each time point were
plotted to compare the frequency with which each
symptom was reported. We repeated the analyses on
the data from the Crown Crisp experiential index
depression subscale to investigate whether a different
scale produced any substantial differences in results.
We divided the cohort between probable cases of
depression and probable non-cases at each time point
using a threshold of 12/13 on the Edinburgh postnatal
depression scale. We then compared changes across
this threshold from 18 weeks to 32 weeks of pregnancy
and 32 weeks of pregnancy to 8 weeks postpartum. We
made a simplifying assumption that the two groups
were independent, although some of the women would
have been common to both groups. Under this
assumption, using a normal approximation to the
binomial distribution, we derived a confidence interval
for the difference in the proportion of women rising
above the threshold for probable depression between
each time point.
Results
Of the 13 799 eligible mothers, 12 059 (87%)
completed at least one of the four questionnaires and
9028 (65%) completed all four. Table 1 shows the mean
Edinburgh postnatal depression scale scores for each
period. Mean scores were higher in pregnancy than
postnatally, with a peak at 32 weeks of pregnancy of
6.72 (SD 4.94) and a lowest value at 8 months post-
partum (5.25 (4.61)). The mean change in depression
score from that at 18 weeks of pregnancy was −0.097
(95% confidence interval −0.18 to −0.01, P = 0.025) at
32 weeks of pregnancy, 0.78 (0.69 to 0.88, P < 0.001) at
8 weeks postpartum, and 1.37 (1.27 to 1.46, P< 0.001)
at 8 months postpartum. Mean change in score was
0.88 (0.79 to 0.97, P < 0.001) between 32 weeks of
pregnancy and 8 weeks postpartum, 1.46 (1.37 to 1.56,
P < 0.001) between 32 weeks of pregnancy and 8
months postpartum, and −0.58 ( −0.50 to −0.67,
P <0.001) between 8 weeks and 8 months postpartum.
Women who did not complete all four questionnaires
had higher depression symptom scores than women
who completed all four. The mean depression scores
were therefore higher when partial responders were
included (table 1).
There was a small rise in depression symptom
score during pregnancy (mean change 0.10; SE 0.043)
and a small drop after childbirth ( −0.88; 0.047). These
changes were significantly different (difference 0.98,
95% confidence interval 0.83 to 1.13; P< 0.001).
Table 2 shows the cumulative frequency of women
with increasing Edinburgh postnatal depression scores
at the four time points. The distribution did not differ
between the time points.
Table 3 gives the scores for each item in the
questionnaire. A similar pattern was seen at all four
time points, but question 3 (I have blamed myself
unnecessarily when things went wrong) was rated
higher at 18 weeks of pregnancy that at other times
and question 6 (things have been getting on top of me)
was rated lower at 8 months postpartum than at other
times.
When we repeated the above analyses using the
Crown Crisp experiential index depression subscale
there was no substantial differences in the findings.
The proportion of women with probable depres-
sion (Edinburgh postnatal depression score >13) was
11.8% at 18 weeks of pregnancy, 13.5% at 32 weeks of
pregnancy, 9.1 % at 8 weeks postpartum, and 8.1% at 8
months postpartum (table 1). In all, 147 (1.6%) women
had probable depression at all four time points and
6771 (75%) scored below the threshold at all time
points; 436 (4.8%) had probable depression at 32
weeks of pregnancy only and 240 (2.7%) at 8 weeks
postpartum only.
Of the 7966 women who were below the threshold
for probable depression at 18 weeks of pregnancy,
673 (8.4%) were above the threshold at 32 weeks of
pregnancy. Of the 7806 below the threshold for prob-
able depression at 32 weeks of pregnancy, 410 (5.3%)
were above the threshold at 8 weeks postpartum.
There were 253 (3.2%) fewer women newly rising
above the threshold for probable depression after
childbirth than during pregnancy (95% confidence
interval 2.4% to 4.0%).
Table 1 Edinburgh postnatal depression score during pregnancy and after childbirth
18 weeks
pregnant
32 weeks
pregnant
8 weeks
postpartum
8 months
postpartum
Women completing all four time points (n=9028):
Mean (SD) score 6.62 (4.66) 6.72 (4.94) 5.84 (4.65) 5.25 (4.61)
No (%) depressed (score>13) 1062 (11.8) 1222 (13.6) 831 (9.2) 731 (8.1)
Women completing any time points (n=12 059):
No 12 059 11 968 11 579 11 063
Mean (SD) score 6.99 (4.87) 7.07 (5.08) 6.06 (5.79) 5.40 (4.68)
No (%) depressed (score>13) 1676 (13.9) 1819 (15.2) 1181 (10.2) 973 (8.8)
Table 2 Cumulative number (percentage) of women with various Edinburgh postnatal
depression scores at each time point.
Score
18 weeks
pregnant
32 weeks
pregnant
8 weeks
postpartum
8 months
postpartum
4 3412 (37.8) 3492 (38.7) 4163 (46.1) 4730 (52.4)
12 7966 (88.2) 7806 (86.5) 8207 (90.9) 8290 (91.8)
16 8742 (96.8) 8689 (96.2) 8766 (97.1) 8779 (97.2)
20 8966 (99.3) 8951 (99.1) 8951 (99.1) 8972 (99.4)
24 9022 (99.9) 9014 (99.8) 9010 (99.8) 9014 (99.8)
28 9027 (100) 9028 (100) 9028 (100) 9021 (99.9)
Table 3 Mean score for Edinburgh postnatal depression scale
items at each time point among 9028 women with complete
data. (Maximum score for each question is 3)
Question
No
18 weeks
pregnant
32 weeks
pregnant
8 weeks
postpartum
8 months
postpartum
1 0.33 0.41 0.31 0.29
2 0.31 0.39 0.27 0.26
3 1.20 1.02 1.01 1.01
4 1.22 1.16 1.01 0.92
5 0.59 0.69 0.55 0.49
6 1.21 1.25 1.22 0.88
7 0.32 0.39 0.23 0.25
8 0.80 0.79 0.73 0.67
9 0.49 0.54 0.44 0.40
10 0.14 0.08 0.07 0.08
Papers
258 BMJ VOLUME 323 4 AUGUST 2001 bmj.com

Discussion
We found that mothers have higher scores on the
Edinburgh postnatal depression scale in pregnancy
than postnatally and that the distribution of total
scores and individual symptoms did not differ before
and after childbirth. These data suggest that depression
is no more likely after childbirth than it is after events
during pregnancy.
Validity of instrument
Self report instruments do not provide a clinical diagno-
sis of depression. Some mothers scoring above the
threshold will not have a depressive illness and some
below the threshold will. However, a validation study
found that a score of >13 gave the best estimate of
prevalence of depression.15 This threshold may overesti-
mate depression during pregnancy.17 However, even if
we used a higher threshold in pregnancy than
postnatally, depression was still common during
pregnancy. Furthermore, the proportion of women rat-
ing themselves as severely depressed was similar before
and after childbirth. We found no evidence that the
question asking whether the woman feared she might
harm herself was misinterpreted or overrated during
pregnancy, despite suggestions that this could happen.19
Patterns of depression
We found no evidence to support the existence of a
subgroup of women with a specific type or severity of
symptoms. It has been suggested that a small subgroup
of women with postnatal depression have abnormal
thyroid function leading to depressed mood.20 Women
who did not respond at all time points had higher
depression scores, and the non-responders probably
included some of the most depressed women. Our
results may therefore be biased by selective non-
response among women with severe postnatal depres-
sion. Another explanation for our lower postnatal
scores is that incidence of depression may peak at a
time not measured in this study. Our data support
research suggesting that there is no difference in the
pattern of symptoms of depression during pregnancy
or postnatally. Thus, postnatal depression does not
seem to be a distinct syndrome.
The scales that we used were developed using only
the core symptoms of depression because they are
applicable to different contexts.Women’s experience of
depression during pregnancy and after childbirth,
however, may be very different. Childbirth is an impor-
tant biological, social, and psychological event.
Although caution is needed when equating self
report data with a clinical diagnosis of depression, our
data suggest that childbirth is less likely than the events
of pregnancy to be followed by depression in women
who are not depressed, and more likely to be followed
by improvement in women who are depressed.The fall
in probable cases of depression after childbirth has
been reported in a study using this dataset and exam-
ining change and stability in depression using latent
variable modelling.21 However, there is some evidence
that childbirth may be a non-specific stressor for
depression in most women and a specific stressor in a
subgroup of women.22 We will be able to test this
hypothesis by following mothers through to their next
pregnancy.
Implications
The consequences of antenatal depression are not well
understood. Psychopathology during pregnancy may
have an important effect on the uterine environment,14
and research is urgently needed into the consequences
for the child of antenatal depression.
It is also important to study the potential benefits of
screening for, and treating, depression during preg-
nancy. Although there are concerns about the
widespread use of antidepressants during pregnancy,
the benefits may outweigh the risks for women with
severe depression. Non-pharmacological inter ventions
that have been found effective in mild to moderate
depression23 could be evaluated for treating depression
during pregnancy.24 Our results show depression
during pregnancy is more common than postnatal
depression. Offering treatment may be important for
both the mother and the future wellbeing of the child
and family.
We thank the mothers and fathers who took part and the mid-
wives for their cooperation and help in recruitment. The study
team comprises interviewers, computer technicians, laboratory
technicians, clerical workers, research scientists, volunteers, and
managers, who continue to make the study possible.
Contributors: JE wrote the paper and planned the data
analyses, JH analysed the data, and HF planned the analyses and
commented on the final draft of the paper. SO contributed to
writing the paper and commented on the final draft. JG
designed the study, advised on data analysis,and commented on
the final draft. The ALSPAC study team designed the study and
collected and entered the data. JE will act as guarantor.
Funding: Medical Research Council, Wellcome Trust,
Department of Health, Department of the Environment, and
various charitable organisations and commercial companies.
The ALSPAC study is part of the WHO initiated European
longitudinal study of pregnancy and childhood.
Competing interests: None declared.
1 Kendall RE, Wainwright S, Hailey A, Shannon B. The influence of child-
birth on psychiatric morbidity. Psychol Med 1976;6:297-302.
2 Pitt B. Atypical depression following childbirth. Br J Psychiatr y
1968;114:1325-35.
3 Cooper PJ, Campbell EA, Day A, Kennerley H, Bond A. Non-psychotic
psychiatric disorder after childbirth: a prospective study of prevalence,
incidence, course and nature. Br J Psychiatry 1988;152:799-806.
4 O’Hara MW, Zekoski EM, Phillips LH, Wright EJ. Controlled prospective
study of postpartum mood disorders: comparison of childbearing and
non-childbearing women. J Abnormal Psychol 1990;99:3-15
5 Sinclair D, Murray L. Effects of postnatal depression on children’s adjust-
ment to school. Br J Psychiatry 1998;172:58-63.
6 Murray L, Sinclair D, Cooper P, Ducournau P, Turner P. The
socioemotional development of 5 year olds with postnatally depressed
mothers. J Child Psychol Psychiatry 1999;40:1259-71.
What is known on this topic
Postnatal depression is common
Recognising and treating depression is
emphasised in postnatal care
Depression is also common in pregnancy
What this study adds
Self reported symptom scores for depression are
higher in pregnancy than postnatally
The severity and nature of depressed mood does
not differ before and after childbirth
More mothers have scores that rise above a
threshold for probable depression during
pregnancy than after childbirth
Papers
259BMJ VOLUME 323 4 AUGUST 2001 bmj.com

7 O’Hara MW, Neunaber DJ, Zekoski EM. Prospective study of postpartum
depression: prevalence, course and predictive factors. J Abnormal Psychol
1984:93:158-71.
8 Green MJ. Who is unhappy after childbirth? Antenatal and intrapartum
correlates from a prospective study.J Reprod Infant Psychol 1990:8:175-83.
9 Hobfoll SE, Ritter C, Lavin J, Hulsizer MR, Cameron RP. Depression
prevalence and incidence among inner-city pregnant and postpartum
women. J Consult Clin Psychol 1995;65:445-53.
10 Cox JL, Connor Y, Kendell RE.Prospective study of the psychiatric disor-
ders of childbirth. Br J Psychiatry 1982;140:111-7.
11 Watson JP, Elliott SA, Rugg AJ, Brough DI. Psychiatric disorder in
pregnancy and the first postnatal year. Br J Psychiatry 1984;144:453-62.
12 Pagel MD, Smilkstein G, Regen H, Montano D. Psychosocial influences on
new born outcome: a controlled prospective study. Soc Sci Med
1990;30:597-604.
13 Hedegaard M, Henriksen TB, Sabroe S, Secher NJ. Psychological distress
in pregnancy and preterm delivery. BMJ 1993;307:234-9.
14 Teixeira JMA,Fisk NM, Glover V.Association between maternal anxiety in
pregnancy and increased uterine artery resistance index: cohort based
study. BMJ 1999;318:153-7.
15 Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression.
Development of the 10-item Edinburgh postnatal depression scale. Br J
Psychiatry 1987;150:782-6.
16 Crown S, Crisp AH. Manual of the Crown-Crisp experiential index .London:
Hodder and Stoughton, 1979.
17 Murray D, Cox JL. Screening for depression during pregnancy with the
Edinburgh depression scale (EPDS). J Reprod Infant Psychol 1990;8:99-
107.
18 Thorpe K. A study of the Edinburgh postnatal depression scale for use
with parent groups outside the postpartum period. J Reprod Infant Psychol
1993;11:119-25.
19 Green, JM, Snowdon C, Statham H. EPDS by post. Br J Psychiatr y
1991;158:865.
20 Harris B, Othman S, Davies JA, Weppner GJ, Richards CJ, Newcombe
RG, et al. Association between postpartum thyroid antibodies and
depression. BMJ 1992;305:152-6.
21 Fergusson D, Horwood J, Thorpe K, ALSPAC Study Team. Changes in
depression during and following pregnancy. Paediatric Perinatal Epidemiol
1996;10:279-93.
22 Cooper PJ, Murray L. Course and recurrence of postnatal depression.
evidence for the specificity of the diagnostic concept. Br J Psychiatry
1995;166:191-5.
23 Mynors-Wallis LM, Gath DH, Day A, Baker F. Randomised controlled
trial of problem solving treatment, antidepressant medication, and com-
bined treatment for major depression in primary care. BMJ
2000;320:26-3.
24 Spinelli MG. Interpersonal psychotherapy for depressed antepartum
women: a pilot study. Am J Psychiatry 1997;154:1028-30.
(Accepted 19 April 2001)
Impact of published clinical outcomes data: case study in
NHS hospital trusts
Russell Mannion, Maria Goddard
Abstract
Objective To examine the impact of the publication
of clinical outcomes data on NHS Trusts in Scotland
to inform the development of similar schemes
elsewhere.
Design Case studies including semistructured
interviews and a review of background statistics.
Setting Eight Scottish NHS acute trusts.
Participants 48 trust staff comprising chief executives,
medical directors, stroke consultants, breast cancer
consultants, nurse managers, and junior doctors.
Main outcome measures Staff views on the benefits
and drawbacks of clinical outcome indicators
provided by the clinical resource and audit group
(CRAG) and perceptions of the impact of these data
on clinical practice and continuous improvement of
quality.
Results The CRAG indicators had a low profile in the
trusts and were rarely cited as informing internal
quality improvement or used externally to identify
best practice. The indicators were mainly used to
support applications for further funding and service
development. The poor effect was attributable to a
lack of professional belief in the indicators, arising
from perceived problems around quality of data and
time lag between collection and presentation of data;
limited dissemination; weak incentives to take action;
a predilection for process rather than outcome
indicators; and a belief that informal information is
often more useful than quantitative data in the
assessment of clinical performance.
Conclusions Those responsible for developing
clinical indicator programmes should develop robust
datasets. They should also encourage a working
environment and incentives such that these data are
used to improve continuously.
Introduction
The public dissemination of standardised data on clini-
cal outcomes is now established practice in many
health systems. In the United States, where public
reporting is most advanced, comparative information
on performance in the form of report cards, provider
profiles, and consumer reports has been released for
over a decade.1–3 In Europe, Scotland has been at the
forefront of public disclosure. Since 1994 the Scottish
Executive has published clinical outcome indicators
collected by the clinical resource and audit group
(CRAG) for all Scottish NHS acute trusts and health
boards. More recently, clinical performance data have
been published for trusts in England and Wales as part
of the NHS performance assessment framework.
A postal questionnaire survey conducted by the
Scottish Executive in 1997 indicated that the CRAG
indicators published in Scotland were of some
practical value to health professionals and, in a few
instances, had helped to bring about a change in clini-
cal practice. On the whole, however, the survey found
that the indicators had little effect on behaviour.4We
present the key findings of a study designed to examine
the impact of the publication of these data on provider
organisations. The CRAG indicators are similar to
those now published more widely within the rest of the
United Kingdom, and therefore an analysis of the
Scottish experience may help with the implementation
of such programmes elsewhere.
The CRAG indicators are compiled and dissemi-
nated by the Scottish Executive. Seven reports have
been published detailing 38 clinical indicators for
named trusts and health boards in Scotland. It is
important to note that the CRAG indicators are not
part of a formal framework of performance assess-
ment. Since the indicators were first published the
Papers
Centre for Health
Economics,
University of York,
Heslington, York
YO10 5DD
Russell Mannion
senior research fellow
Maria Goddard
assistant director
Correspondence to:
R Mannion
rm15@york.ac.uk
BMJ 2001;323:260–3
260 BMJ VOLUME 323 4 AUGUST 2001 bmj.com