Key words postpartum depression - gynaecologists - pregnancy - screening - representative survey
Schlüsselwörter postpartale Depression - Gynäkologen - Schwangerschaft - Screening - repräsentative
Befragung
Introduction
With a prevalence of 10 to 15 % [1 ], [2 ], [3 ] postpartum depression (PPD) is one of the most common psychological disorders of
women during pregnancy and after birth. Halbreich and Karkun [2 ] evaluated 143 studies on the frequency of PPD world-wide and determined the prevalence
to be between 0 % and 60 % depending on the tool, chosen cut-off and time period after
birth investigated. In Germany von Ballestrem et al. [4 ] found a prevalence for PPD of 3.6 % in a random sample of 722 mothers; however,
there are only very few empirical studies on the frequency of PPD in Germany. Postpartum
(or postnatal) depression or PPD, coded in ICD-10 as a mild mental and behavioural
disorder in childbed, not classified elsewhere (F53.0), is a maternal depressive disorder
characterised by fear of failure, emotional ambivalence and insensitivity that often
begins in the first to sixth week after birth of the baby but can also occur already
during pregnancy [5 ]. PPD is often identified as a predictor for attachment disorders [6 ], infanticide [7 ], developmental disabilities [8 ] and a later depressive disease of the child [9 ]. Mental problems of the parents such as affective disorders also belong to the most
important risk factors for endangerment of the childʼs welfare [10 ]. Against the background of an increased risk for impairment of the childʼs development,
the treatment of PPD is of particular relevance.
With regard to PPD in Germany there is a clear management deficit due to the lack
of psychotherapeutic options for mothers with new-born babies. PPD often leads to
a feeling of shame among the afflicted mothers, they fear stigmatisation, separation
from the baby or they are not consciously aware of their disorder or the available
possibilities for help [11 ]. According to the study of Le Strat et al. [12 ] women with postpartum psychological disorders less frequently take advantage of
available help than do mentally ill women who are not in the postpartum period. In
addition, the need for therapeutic options to treat PPD in mother-child facilities
that, besides the treatment of PPD, also address the relationship to the baby without
requiring separation from the child is only covered in Germany to about 21 % [13 ]. Since as yet no highly promising results have been achieved for the primary prevention
of PPD [14 ], the early recognition and treatment of PPD is especially important [15 ]. Available as screening tools for PPD are the Bromley Postnatal Depression Scale
[16 ], the Postpartum Depression Checklist (PDC; [17 ]), the Postpartum Depression Screening Scale (PDSS; [18 ]) as well as the Edinburgh Postnatal Depression Scale (EPDS; [19 ], [20 ]), a self-assessment questionnaire (10 items) which is also widely distributed in
Germany. Since the afflicted women are often not able on their own to describe their
psychological problems, a more active handling of psychological disorders in the period
of pregnancy and after birth is particularly important [15 ]. On account of their sensitivity and specificity as confirmed in a meta-analysis
[21 ], the EPDS and the Patient Health Questionnaire (PHQ; [22 ]) are recommended as screening tools for PPD after birth in the S3 guidelines for
unipolar depression (consultation version) [23 ]. During the childbed period in which the PPD symptoms frequently occur for the first
time, together the midwife and the gynaecologist are the primary contact persons for
the young mother and play a significant role in the earliest possible recognition
of PPD and, if necessary, referral of the afflicted patient during pregnancy and in
the first six to twelve months after birth to psychotherapeutic services, for example,
as part of the post-natal care [24 ]. The recognition and treatment of pregnancy-related psychological diseases such
as PPD is as yet in Germany not a regular component of general medical education or
specialist training in the field of gynaecology and obstetrics.
In this German nation-wide survey of local gynaecologists the following questions
were posed: What diagnostics do gynaecologists in Germany carry out for pregnant women
and mothers with a suspicion of PPD? What management possibilities do gynaecologists
see for patients with PPD? From the gynaecologistʼs point of view what barriers are
there to the diagnosis and management of mothers with PPD? What factors from the gynaecologistʼs
point of view influence the active diagnosis and consultations of patients suffering
from PPD?
Materials and Methods
Execution
In the framework of a nation-wide questionnaire survey, n = 3000 practicing gynaecologists
were asked about their handling of patients with a suspicion of PPD. The sample was
chosen at random from a basic population of n = 9823 addresses of registered gynaecologists
in Germany recorded in the database of the address service provider “MediAdressen
Select” in January 2013. In a first step every third practice ordered according to
post code was selected and from this sample in a second step every 12th practice as
well as those practices that had already participated in a preliminary study were
excluded ([Fig. 1 ]). According to the physician statistics of the German Medical Association there
were n = 9784 specialists for gynaecology and obstetrics registered in Germany in
2012 [25 ]. The difference between the number of registered physicians in the employed address
list and in the physician statistics of the German Medical Association (n = 39) can
be explained by different reporting dates and different means of data acquisition
of the two sources. In contrast to the German Medical Association, the address provider
service uses only public sources for address research in accord with the Federal Data
Protection Law so that, for example, the closing of medical practices may have been
recorded only after a delay of several months.
Fig. 1 Flow diagramme of the participants in the German nation-wide survey of practicing
gynaecologists. NB: * = according to the database of the address service provider
“MediAdressen Select” (as of January 2013).
A preliminary study was carried out in autumn 2012 in n = 15 gynaecology practices
in Hamburg and served to test the questionnaire with respect to its comprehensibility,
relevance and practicability as well as to investigate appropriate incentives and
formulation of the accompanying letter.
In spring 2013 two survey waves (invitation, reminder) at an interval of one month
were undertaken in which the n = 3000 selected gynaecologists were contacted by post.
Reasons for non-participation such as, for example, closure of the practice were acquired
by means of a pre-stamped postcard sent with the second letter that could be returned
gratis by the non-participants. In addition, demographic data about the non-participant
could be extracted from the practice stamp on the postcard, e.g., type of practice
and location.
Instruments
For this survey of practicing gynaecologists a questionnaire encompassing 28 items
was used, it was developed on the basis of the findings from surveys of comparable
samples (e.g., Berner et al. [26 ]). The questionnaire contained details of a) practice and person, b) diagnostics
and handling of patients with a suspicion of PPD as well as c) system- and patient-related
barriers to management in both open and closed answer formats. Multiple answers were
possible for many items.
Sample
Of the 3000 contacted gynaecologists n = 1034 took part in the questionnaire survey.
The response rate thus amounted to 35 %. The proportion of male gynaecologists of
36 % was lower than the average proportion of practicing gynaecologists in the year
2012 (41 % male; [25 ]). On average the participating gynaecologists were 52 years old (SD = 7.5) and in
the age range of 35 to 73 years. The duration of professional activity was on average
24 years (SD = 8.4, range: 1–55 years. More than half of the participants (58 %; n = 621)
reported that they had gained additional qualifications, 74 % with psychosomatic or
therapeutic further training (n = 448), 44 % with additional medical qualifications
(n = 264) and 15 % with alternative methods such as acupuncture or natural therapies
(n = 90). The majority were active in one-person practices (61 %) and in the “old”
federal states including Berlin (85 %). For 38 % (n = 385) of the participating gynaecologists
their catchment areas were urban, for 25 % (n = 251) rural and for 37 % (n = 367)
both urban and rural. The average number of pregnant women treated per year was 150;
75 % of the participating gynaecologists managed up to 300 pregnant women per year
in their own practices.
Demographic data for the participating gynaecologists were compared with representative
data for the registered gynaecologists in Germany [24 ] ([Table 1 ]). The regional distribution of the participants in the federal states or, respectively,
in the “new” and “old” federal states is comparable with that of the entire cohort
of practicing gynaecologists in Germany. Male and gynaecologists aged more than 50
years are underrepresented in the sample.
Table 1 Sociodemographic characteristics of the sample of registered gynaecologists in the
study group (n = 1 034) compared with the entire cohort of registered gynaecologists
in Germany according to physician statistics of the German Medical Association (GBE,
2015).
Sample of the nation-wide survey
Registered gynaecologists in Germanya
pb
a Data from the physicians statistics of the German Medical Association for the year
2012 (2015); b χ2 test according to Pearson; * p < 0.05, ** p < 0.01, *** p < 0.001
(n = 1 034)
(n = 9 784)
Gender
**
366 (36 %)
4 054 (41 %)
656 (64 %)
5 730 (59 %)
Age (years)
***
28 (4 %)
288 (3 %)
256 (34 %)
2 918 (30 %)
332 (44 %)
4 147 (43 %)
112 (15 %)
1 675 (17 %)
22 (2 %)
756 (8 %)
Federal states
n. s.
122 (12 %)
1 234 (13 %)
143 (14 %)
1 529 (16 %)
49 (5 %)
498 (5 %)
21 (2 %)
266 (3 %)
14 (1 %)
110 (1 %)
35 (4 %)
278 (3 %)
66 (7 %)
683 (7 %)
20 (2 %)
185 (2 %)
105 (10 %)
917 (9 %)
220 (22 %)
2 269 (23 %)
57 (6 %)
429 (4 %)
12 (1 %)
120 (1 %)
63 (6 %)
467 (5 %)
34 (3 %)
237 (2 %)
35 (4 %)
334 (3 %)
16 (2 %)
228 (2 %)
New vs. old Federal states
n. s.
154 (15 %)
1 383 (14 %)
858 (85 %)
8 401 (86 %)
Non-participants
Only 7 % (n = 143 of n = 1966) of the non-participants returned the postcard giving
reasons for not taking part in the analysis. Most of these non-participants were active
in single-person practices (75 %) and located in the old federal states (84 %). About
47 % of the non-participants were male. The non-participants differed significantly
with regard to gender and practice from the participants in this nation-wide survey;
with regard to location in the old or new federal states there was no significant
difference between the participants and the non-participants.
Reasons given for not participating were for 44 % (n = 62) lack of time; 23 % (n = 33)
reported that they had other specialist fields of activity or, respectively, focused
on other target groups. Seventeen percent (n = 24) were not interested in the survey,
whereas 15 % (n = 21) had previously had bad experiences with surveys. About 11 %
(n = 16) of the non-participants stated that they had closed their practices, e.g.,
due to retirement or change of location. Thus, it can be assumed that these physicians
were still recorded in the address database of the address service provider due to
a delayed data actualisation. This may in part explain the difference in physician
statistics. Miscellaneous reasons for non-participation were given by 7 % (n = 10).
Representativeness of the sample
For an assessment of the representativeness of the present cross-sectional survey,
three factors that could limit the representativeness of the survey were examined:
The target population is not sufficiently clearly defined,
Selection of the sample was not random (e.g., for reasons of practicability) and
there is a non-response bias, i.e., differences between participants and non-participants.
Under consideration of the preformed target group definition, the procedure chosen
to select the sample and a comparison of the observed characteristics of the sample
with the characteristics of the target population or, respectively, the non-participants
it can be assumed that the sample is very probably representative with regard to the
regional distribution between the new and the old federal states for the registered
gynaecologists in Germany at the time of the survey in the Spring of 2013.
Statistical evaluation
The results of the survey were analysed descriptively. In order to determine the outcome-influencing
factors in the performance of an active diagnosis and consultation in cases of PPD,
binary logistic regressions were undertaken.
Results
Diagnostics of PPD
Half of the surveyed gynaecologists estimated the incidence of PPD among patients
in their own practices to be 3 %, 75 % of the participants estimated the incidence
of PPD to be up to 5 %. The gynaecologists gained the suspicion of PPD mainly in consultations
with the patients (99 %) or through informative tips from third persons, e.g., midwifes,
medical assistants or relatives (72 %), whereby multiple answers were possible ([Table 2 ]). About 50 % of the surveyed physicians specifically sought suggestions for PPD
during the consultations. A questionnaire as screening tool for PPD was employed by
16 % of the gynaecologists. A questionnaire for PPD was routinely used by 62 of the
1034 participants (6 %), whereas 10 % used a questionnaire when needed. More than
half of the participating gynaecologists reported the use of more than one method
to detect evidence for PPD.
Table 2 Procedures used by the participating gynaecologists in the diagnostics for PPD (n = 1 034).
Diagnostic methods*
n
%
* = multiple answers possible
Detection during consultations
Overall frequency of consultations
1 003
99 %
of which:
specific questioning
510
50 %
coincidental procedure
523
52 %
Detection using a questionnaire
Overall frequency of questionnaire use
165
16 %
of which:
routine use
62
6 %
use as needed
100
10 %
Detection through tips for third persons
Overall frequency
733
72 %
Number of diagnostic methods
None of the 3 methods (consultation, questionnaire, third person)
7
1 %
1 method
234
23 %
2 methods
651
64 %
3 methods
119
12 %
Active anamnesis
Targeted questioning or routine use of questionnaire
539
53 %
Interventions and mediation for mental burdens
One of the most frequent interventions for patients with PPD is the consultation with
a gynaecologist, which was reported as an intervention by 84 % of the participating
physicians. Patients with PPD are referred to a therapist or a hospital by the majority
of the gynaecologists (86 %). About 1 % of the registered gynaecologists did not mention
any interventions for the management of patients with PPD.
Of the n = 1034 surveyed registered gynaecologists, 96 % felt themselves to be responsible
for the recognition of PPD in their patients (“yes, of course” or “generally yes”).
Barriers for diagnostics and management
Among the most frequent problems from the registered gynaecologistsʼ point of view
in the care of patients with PPD are the limited time available for consultation (74 %),
the low reimbursement for consultations (53 %), the lack of effective treatment and
management options (50 %), the lack of recognition of the diagnosis by the partner
(48 %) and the rejection or non-utilisation of referrals by the patients (45 %). Further
barriers mentioned frequently are the lack of psychotherapy places, long waits, communication
problems and anxiety from stigmatisation and shame. Whereas 96 % of the gynaecologists
saw at least one system-related barrier, 79 % of the participants also saw at least
one patient-related barrier.
Factors influencing an active anamnesis and consultation for PPD
An active PPD anamnesis, i.e., the specific questioning or routine control by means
of a questionnaire was carried out by 53 % (n = 539) and consultations in the presence
of PPD were performed by 82 % (n = 852) of the gynaecologists. The results of the
binary logistic regressions for active anamnesis and consultation are presented in
[Tables 3 ] and [4 ].
Table 3 Logistic regression of active anamnesis for PPD by gynaecologists.
AV: active anamnesis for PPD (n = 885)
Model accuracy
B
Significance
Odds ratio
* p < 0.05, ** p < 0.01. Active anamnesis: targeted questioning or routine use of
questionnaire
Complete model
χ2
46.118 ***
R2 (Nagelkerkeʼs)
0.068
Correct prediction
59 %
Predictors
Federal state (new)
− 0.563
0.004**
0.57
Catchment area (urban)
0.018*
− 0.481
0.007**
0.62
− 0.315
0.051
0.73
Gender (female)
0.373
0.017*
1.45
Length of professional experience
− 0.003
0.760
1.00
Additional qualifications (none)
0.008**
0.604
0.001**
1.83
0.021
0.933
1.16
0.149
0.396
0.56
Number of system-related barrier
0.131
0.056
1.14
Number of patient-related barriers
0.137
0.118
1.15
Table 4 Logistic regression of consultations in PPD by gynaecologists.
AV: Consultation for PPD (n = 885)
Model accuracy
B
Significance
Odds ratio
*p < 0.05, **p < 0.01, *** p < 0.001
Complete model
χ2
51.297 ***
R2 (Nagelkerkeʼs)
0.098
Correct prediction
85 %
Predictors
Active anamnesis for PPD (yes)
0.608
0.190
1.84
Federal state (new)
− 0.150
0.559
0.86
Catchment area (urban)
0.181
0.128
0.599
1.14
0.423
0.067
1.53
Gender (female)
− 0.188
0.393
0.83
Length of professional experience
− 0.013
0.277
0.99
Additional qualifications (none)
0.097
0.647
0.022*
1.91
− 0.142
0.667
0.87
0.191
0.427
1.21
Number of system-related barriers
0.43
< 0.001***
1.54
Number of patient-related barriers
0.072
0.572
1.08
It can be seen that an active diagnosis for PPD is more often performed when the physician
is female, possesses an additional psychosomatic qualification, and has a practice
in an urban catchment area or in the old federal states.
A consultation for an existing PPD was associated with the physicianʼs possession
of an additional psychosomatic or therapeutic qualification and a higher subjective
perception of system-related barriers to the management of patients with PPD from
the physicianʼs point of view.
Discussion
Since PPD is a maternal psychological disease that can seriously impair the childʼs
development, the gynaecologists are an important instance in the period after birth,
in the sense of pre-emptive child protection, to recognise the presence of PPD and
to enable access to existing support facilities. In the S3 guidelines for unipolar
depression [23 ], this observation has recently been taken into consideration by the extension and
differentiation of the recommendations for the recognition and treatment of PPD thus
placing PPD more strongly in the field of view of physicians and psychologists. The
present survey was intended to illustrate the health-care situation for mothers with
PPD in Germany from the point of view of registered gynaecologists. It was able to
show that gynaecologists have a good access to pregnant women and young mothers and
include, in addition to the somatic treatment options, the management of psychological
irregularities of their patients also in view of the associated risks for the childʼs
development. The response rate of 35 % can be considered as high in the light of the
response rates in comparable surveys, this can be evaluated as a sign for the current
relevance of the investigated topic.
In this nation-wide and, with respect to regional distribution, representative survey
the majority of the responding gynaecologists stated that the recognition of PPD belonged
in their area of responsibility (97 %) although their medical education and specialist
training had not prepared them for it. Fifty percent of the responding gynaecologists
estimated the incidence of PPD among their own patients on the basis of their preponderantly
subjective opinions in the absence of a standardised assessment method to be up to
3 %, 75 % estimated the incidence to be up to 5 %. This result can be taken as an
indication of the great sensitivity and expertise of gynaecologists for PPD. It can
be assumed that standardised PPD diagnostics in gynaecological practices would enable
a more reliable detection of patients with psychological disorders.
For an adequate management of PPD a cornerstone could be a screening (e.g., by means
of EPDS or PHQ) by gynaecologists and, in the case of a positive result, a consultation
and, whenever necessary, referral would be recommended [23 ]. It was found, however, that only half of the surveyed gynaecologists actively looked
for PPD. Merely 16 % used a questionnaire for this purpose although there are empirically
supported recommendations for the use of screening tools such as the EPDS in the management
of PPD [23 ]. While an active diagnostic procedure was used only by a minority of the gynaecologists,
in cases of recognition of PPD the majority of the gynaecologists do, in accord with
the guidelines, undertake a consultation or make a referral to other facilities.
Even when the clarified variance is rather small due to the predictors employed in
the model, it can be deduced from this survey that PPD is more frequently addressed
actively when the gynaecologist is female, has a practice in the old federal states,
in an urban catchment area and possesses an additional qualification. Besides the
possession of an additional qualification, a consultation is associated with the number
of system-related barriers. The lower amount of active diagnostics in the new federal
states could be due to the lower density of physicians there [27 ], which leads to a larger number of patients to be cared for and thus less time available
per patient [28 ]. Also in rural regions there is often a shortage of physicians [27 ]. That female gynaecologists are more active in the diagnosis of PPD is in accord
with the results from studies with general practitioners and paediatricians in which
female physicians more frequently employ screenings to detect PPD and are more liable
to address psychological topics [29 ], [30 ]. An additional qualification, e.g., in psychosomatic or therapeutic fields, can
lead to a stronger sensibilisation for topic-related diseases and to increased competence
in professional consultations and counselling [31 ].
These results provide suggestions as to where improvement is necessary from the gynaecologistsʼ
point of view in order to optimise the management of patients with PPD. Above all,
structural barriers such as the lack of time for consultations, the low reimbursement
for consultations and the lack of effective treatment and care options were mentioned.
This study reflects the subjective perceptions of the care situation from the gynaecologistʼs
point of view. Content, quality and efficacy of consultations and referrals were not
addressed within the framework of this survey. Similarly, aspects that could improve
the predictive strength of the model for active diagnostics and counselling in suspected
cases of PPD were not considered since they have already proved to be relevant for
the prediction for screening and treatment results in PPD (e.g., subjective importance
of screening and treatment, trust in oneʼs own diagnostic and therapeutic abilities)
[32 ].
The non-participant analysis showed that there were significantly more men and more
gynaecologists in one-person practices in the group of non-participants than in the
participant group. In addition to the usual selection effects for participation in
voluntary surveys, this could be due to interest and attitudes as well as possible
answering tendencies in the direction of social desirability that led to selection
effects among the investigated population with regard to attitudes and the mentioned
procedures.
Practical Conclusions
The diagnosis and care of patients with PPD in the framework of gynaecological management
is already accepted as one of their tasks by many registered gynaecologists in Germany.
In the light of the empirically confirmed relationship between maternal PPD and impaired
development of the child through to endangerment of the childʼs welfare, improvements
are necessary at the following levels:
disorder-specific qualification of gynaecologists within the framework of medical
studies and advanced specialist training,
use of a systemic diagnostic work-up for detecting PPD to increase the sensitivity
for PPD symptoms,
consideration of risk factors for PPD during gestation and after birth, as well as
establishment of suitable psychotherapeutic care options for mother and child.
The recognition and management of PPD in pregnant women and young mothers by registered
gynaecologists appears to be a promising strategy to offer adequate aids, in the sense
of pre-emptive child protection, for families and to avoid or reduce the possible
negative consequences for the childʼs development. Empirical studies are needed to
check the efficacy of this strategy.
Acknowledgements
We thank Dr. Levente Kriston for methodological advice.
