DOI: 10.1111/1471-0528.

14259
www.bjog.org

Post-traumatic stress symptoms in Swedish

obstetricians and midwives after severe obstetric
events: a cross-sectional retrospective survey
Å Wahlberg,a,b M Andreen Sachs,c K Johannesson,d G Hallberg,a,e M Jonsson,a,e
A Skoog Svanberg,a U Högberga
a
Department of Women’s and Children’s Health/Obstetrics & Gynaecology, Uppsala University, Uppsala, Sweden b Department of
ane University Hospital, Malmö, Sweden c LIME/Medical Management Centre, Karolinska Institutet,
Gynaecology and Obstetrics, Sk

Stockholm, Sweden d Neuroscience/Psychiatry, Uppsala University, Uppsala, Sweden e Department of Gynaecology and Obstetrics, Akademiska
Hospital, Uppsala, Sweden
Correspondence: Å Wahlberg, Department of Obstetrics and Gynaecology, Jan Waldenströms gata 47, S-214 66 Malmö, Sweden.
Email asa.wahlberg@kbh.uu.se

Accepted 2 June 2016. Published Online 24 August 2016.

Objective To examine post-traumatic stress reactions among both professions reported symptoms indicative of partial PTSD,
obstetricians and midwives, experiences of support and whereas 7% of the obstetricians and 5% of the midwives
professional consequences after severe events in the labour ward. indicated symptoms fulfilling PTSD criteria. Having experienced
emotions of guilt or perceived insufficient support from friends
Design Cross-sectional online survey from January 7 to March 10,
predicted a higher risk of suffering from partial or probable
2014.
PTSD. Obstetricians and midwives with partial PTSD
Population Members of the Swedish Society of Obstetrics and symptoms chose to change their work to outpatient care
Gynaecology and the Swedish Association of Midwives. significantly more often than colleagues without these
symptoms.
Methods Potentially traumatic events were defined as: the child
died or was severely injured during delivery; maternal near-miss; Conclusions A substantial proportion of obstetricians and
maternal mortality; and other events such as violence or threat. midwives reported symptoms of partial or probable PTSD after
The validated Screen Questionnaire Posttraumatic Stress Disorder severe traumatic events experienced on the labour ward. Support
(SQ-PTSD), based on DSM-IV (1994) 4th edition, was used to and resilience training could avoid suffering and consequences for
assess partial post-traumatic stress disorder (PTSD) and probable professional carers.
PTSD.
Keywords midwives, obstetricians, perinatal, post-traumatic stress
Main outcome measures Partial or probable PTSD. disorder, survey.
Results The response rate was 47% for obstetricians (n = 706) Tweetable abstract In a survey 15% of Swedish obstetricians and
and 40% (n = 1459) for midwives. Eighty-four percent of the midwives reported PTSD symptoms after their worst obstetric
obstetricians and 71% of the midwives reported experiencing at event
least one severe event on the delivery ward. Fifteen percent of

Please cite this paper as: Wahlberg Å, Andreen Sachs M, Johannesson K, Hallberg G, Jonsson M, Skoog Svanberg A, Högberg U. Post-traumatic stress
symptoms in Swedish obstetricians and midwives after severe obstetric events: a cross-sectional retrospective survey. BJOG 2016; DOI: 10.1111/1471-
0528.14259.

professionals who have experienced serious incidents may

Introduction
encounter psychological reactions such as stress, distress,
The concept of ‘second victim’ was coined by Wu in 2000 panic, anxiety, insomnia and feelings of shame and guilt.2–4
to describe the pain that doctors may experience after hav- Long-lasting effects manifested as burnout or post-trau-
ing been involved in a medical error.1 Since then research matic stress disorder (PTSD) have been described among
in the area has provided us with more knowledge. Both surgeons and midwives.5–7 In addition to psychological
quantitative and qualitative studies have shown that consequences there are also reports of professional

ª 2016 Royal College of Obstetricians and Gynaecologists 1

 Wahlberg et al.

consequences such as change of occupation.1,4,8 Patient care under 40 years (55.6%), than in those aged over 40
might also be negatively affected. Emotional barriers such (39.6%), but there were no differences between the sexes.
as distancing have been shown to inhibit healthcare staff For the assessment of post-traumatic stress symptoms
from providing effective care to parents after stillbirth.9 the Swedish version of Screen Questionnaire Post-traumatic
Only a few studies have reported on the prevalence Stress Disorder (SQ-PTSD) was selected. The SQ-PTSD is
and effects of work-related PTSD, and the problem may based on the diagnostic criteria for PTSD, according to the
be more common than is generally envisaged.7,10 Expo- Diagnostic and Statistical Manual of Mental Disorders,
sure to severe events in maternity care is part of the Fourth Edition (DSM-IV).13,14 The reliability, validity, sen-
professional experience of obstetricians and midwives. sitivity and specificity of the SQ-PTSD have been tested
Hall and Scott11 conclude that one unanswered question with satisfactory results.13 For this study the SQ-PTSD was
about second victims of adverse events concerns the rela- adopted appropriately. The A1 criterion, ‘Confrontation
tionship between severe second-victim reactions and with the stressor should involve actual or threatened death
post-traumatic stress symptoms. A recent systematic liter- or serious injury, or a threat to the physical integrity of self
ature review on stillbirths stated that ‘Acknowledgement or others’, was defined as exposure to a serious and poten-
of the personal and professional cost of stillbirth on staff tially traumatic event on the delivery unit, such as: (1) The
is essential, for their professional wellbeing and to enable child died during birth or had severe asphyxia, neonatal
health workers to deliver effective care to bereaved death due to asphyxia or other birth-related injury; (2)
parents.’12 maternal near-miss, maternal death during delivery; and
To our knowledge there are no studies on post-traumatic (3) other difficult and threatening events during obstetric
stress reactions among healthcare professionals after experi- care, as defined by the respondents, for example violence
encing a severe obstetric incident. The objectives of this or threat. The respondents were asked to recall the event
study were to assess among obstetricians and midwives: perceived as the most difficult and then answer according
1 Traumatic experiences in conjunction with a severe event to the A2 criterion, which contains questions on: (1) emo-
such as emotions of fear, helplessness, panic, threats to tions of intense fear, helplessness or panic at the time of
professional identity, guilt and the development of post- the event; and (2) feelings of threat to their professional
traumatic stress symptoms; identity/role during the event. A positive answer was com-
2 Risk factors for developing post-traumatic stress symp- pulsory for the question about intense fear, helplessness or
toms: experience of negative reactions from parents, sup- panic (A2:1) for the partial and probable PTSD conditions,
port received from local managerial staff, colleagues, whereas a confirmation regarding the threat towards the
friends or partner, experiences of ‘reassembly’ (debrief- professional identity (A2:2) was not.13 Criteria B–E of the
ing), being involved in complaints procedures and SQ-PTSD questionnaire were not changed from the origi-
reporting to the national authority; nal version.13 The B criterion related to persistent re-
3 Professional consequences in the short and long term experiencing of the traumatic event in intrusive thoughts,
after having experienced symptoms of post-traumatic nightmares or flashbacks; the C criterion to persistent
stress symptoms. avoidance of stimuli associated with the event and emo-
tionally numbing symptoms, described as an inability to
experience any positive feelings such as love, contentment,
Methods
satisfaction and happiness; the D criterion to hyperarousal
A cross-sectional, retrospective web survey (Survey-Mon- symptoms such as difficulties in sleeping, concentrating
key) of registered members of the Swedish Society of and controlling anger; and the E criterion to duration of
Obstetrics and Gynaecology (SFOG) and the Swedish Asso- the disturbance, i.e., symptoms of criteria B, C and D for
ciation of Midwives (SBF) was conducted between January more than 1 month. The questions for criteria A2:1, A2:2,
7 and March 10, 2014. The survey was sent to all members B, C and D are presented in Table S1. The definition of
<67 years old (official retirement age in Sweden) having an partial PTSD was used in accordance with that of Breslau
e-mail address. Members with no known e-mail address et al.15; one criterion from each symptom group (B, C and
did not receive the survey (28% of SBF members and 3% D), and a duration of more than 1 month (E criterion).
of SFOG members). The questionnaire went to 3849 mid- Because PTSD is a clinical diagnosis for which screening
wives and 1498 obstetricians, followed by three reminders. tools are only indicative, we used the term ‘probable’ PTSD
The response rate was 47.1% (n = 706) for obstetricians when all criteria for the diagnosis were fulfilled according
and 39.9% (n = 1459) for midwives. Five hundred and to the SQ-PTSD.
thirty-eight female doctors responded and 168 male. Only The questionnaire included demographic and profes-
four out of 1459 midwives in the survey were men. Among sional characteristics. Additionally information on feelings
the obstetricians the response rate was higher in those aged of guilt in conjunction with the event was collected. Having

2 ª 2016 Royal College of Obstetricians and Gynaecologists

 Post-traumatic stress in Swedish obstetricians and midwives

feelings of guilt is not part of the PTSD diagnosis, accord- criterion), three or more symptoms of avoidance (C crite-
ing to DSM-IV, but is a commonly described emotion in rion) and two or more symptoms of increased arousal (D
conjunction with the condition. There was also a question criterion) (Table S1).
about potential negative reactions from the parents. Ques- The chi-square test and Fisher’s test were used to analyse
tions about perceived support by the local managerial staff, the associations. Crude and adjusted odds ratios were cal-
colleagues, friends and partner were included in the survey. culated by logistic regression analysis; the statistical soft-
There was a question on potential reassembly (debriefing) ware package IBM SPSS version 20 was used to perform all
after the severe event and perceptions thereof. We also statistical analyses.
requested that participants should provide information on The study was approved by the Regional Ethical Board
event analysis, as well as reports to the National Board of of Uppsala on November 8. 2013 (2013/351).
Health and Welfare or the Health and Social Care Inspec-
torate (after June 1 2013) by the hospital and complaints
Results
by the patient or the family of the patient to one of these
government agencies. Out of 706 obstetricians and 1459 midwives who responded,
Personal and professional consequences, such as sick 84% (n = 594) and 71% (n = 1034), respectively, had expe-
leave, were covered, as well as information about the rienced one or more severe events on the labour ward, and
respondents’ work status, i.e. remaining at the same work- fulfilled the A1 criterion in the PTSD-screening.
place or leaving emergency obstetric services for other Of those, 43% reported emotions of intense fear, help-
branches of obstetrics and gynaecology or places of work. lessness or panic (A2:1 criterion) in connection with the
Face validity of the questionnaire was checked through severe event, with no differences between obstetricians and
pilot testing, and minor changes were made. midwives. A threat to their professional role/identity (A2:2
For this study, those who met all the DSM-IV items for criterion) in the moment of exposure was reported by 25%
criteria A–E were classified as having symptoms of PTSD of the obstetricians and 17% of the midwives (P = 0.001).
(Table 1). The F criterion, ‘the disturbance causes clinically Emotions of guilt were experienced by 47% of the obstetri-
significant distress or impairment in social and occupa- cians and 28% of the midwives after the traumatic experi-
tional, or other important areas of functioning’, was not ence (P = 0.001) (Table S1). Seventy percent of the
included in the applied diagnosis in this study. Partial obstetricians and 66% of the midwives had at least one
PTSD was defined as prevalence of a traumatic event (A1 symptom of re-experiencing the event (B criterion) and
criterion) in conjunction with emotions of intense fear, 40% of the obstetricians and 35% of the midwives had at
helplessness or panic (A2:1 criterion) as well as at least one least one symptom of avoidance or numbing (C criterion)
symptom of each B, C and D criteria; re-experiencing, after the event. Twelve percent of the obstetricians and 7%
avoidance and hyperarousal symptoms with a duration of of the midwives had experienced three or more symptoms
at least 1 month (E criterion). PTSD was defined as the of avoidance or numbing (C criterion). One or more
prevalence of a traumatic event (A1), the subjective symptoms of arousal (D criterion) were reported by 57%
response of fear, helplessness or panic (A2:1 criterion) and of the obstetricians and 58% of the midwives after a trau-
a duration of more than 1 month (E criterion) in combi- matic experience. Two or more of the arousal symptoms
nation with at least one symptom of re-experience (B (D criterion) were experienced by 36% of the obstetricians
and 29% of the midwives.
Because partial PTSD requires a minimum of one crite-
Table 1. Post-traumatic stress symptoms (partial or probable PTSD
and probable PTSD) for obstetricians and midwives after exposure to rion from each symptom group (B, C and D) and a dura-
severe events in the maternity ward. tion of more than a month (E criterion), respondents with
more than one symptom in each group were also included
Symptoms of PTSD [%(95% CI)]
in the partial PTSD data. In all, 15% reported partial PTSD
All Obstetricians Midwives P*
or PTSD, while all the symptom criteria for ‘full’ PTSD
(n = 1628) (n = 594) (n = 1034) were fulfilled by 7% of the obstetricians and 5% of the
midwives (Table 1).
Probable 15 (13–17) 15 (12–18) 15 (13–17) 1.00 The distribution in terms of working experience among
symptoms of the 241 obstetricians and midwives with symptoms of at
PTSD) least partial PTSD was relatively even (Figure 1).
Symptoms of 5 (4–7) 7 (5–9) 5 (3–6) 0.076 Table 2 shows the risk of developing partial PTSD/PTSD
probable PTSD after exposure to severe events in the delivery ward. Fifty-
*chi2 test. four percent of the exposed obstetricians had participated in
a reassembly (debriefing) after the event, of whom 27%

ª 2016 Royal College of Obstetricians and Gynaecologists 3

 Wahlberg et al.

100% exposure to an event that was perceived as the worst dur-

90% ing their working life on the labour ward. As a result, trau-
80% matised midwives took sick leave significantly more often
70%
than did traumatised doctors. However, healthcare provi-
60%
ders with symptoms of partial PTSD or probable PTSD in
50%
40%
both groups changed their work to a significantly higher
30% degree.
20%
10% Strengths and limitations of the study
0% Our study’s main strength is the relatively high numbers of
5

10

15

20

25

30

35

40
participants, including obstetricians, compared with those
to

to

to

to

to

to

to

to
0

10

15

20

25

30

35
of similar studies from the labour ward.7,16,17 At the same
>

>

>

>

>

>

>

Partial or probable PTSD No partial or probable PTSD

time less than half of the obstetricians and midwives who
Figure 1. Symptoms of partial or probable PTSD reported by were invited participated, causing probable selection bias.
obstetricians and midwives by years of professional experience The type and extent of selection bias could not be evaluated
(n = 241).
in our study, but a non-response study on a population-
based post-disaster (2004 tsunami) postal questionnaire
described this intervention as a negative experience. For showed that non-participants reported a lower perceived
midwives, 42% had participated in a reassembly (debriefing), threat of death and lower levels of post-traumatic stress
of whom 25% were unsatisfied with the intervention. The reactions, and stated their reasons for not responding as
crude odds ratios indicate that, for both obstetricians and ‘lack of interest or time’, ‘lack of relevant experiences’ and
midwives, emotions of guilt, negative reactions from parents, ‘too personal or emotionally disturbing’.18 Our results must
the experience of insufficient support from local managerial therefore be interpreted with some caution, since they could
staff, colleagues, friends and partner as well as bad experi- not claim to represent the majority of Swedish obstetricians
ences during reassembly (debriefing) are significant factors and midwives, but rather those finding the subject relevant.
in the development of PTSD/partial PTSD. A higher risk was Another important limitation was recall bias. We asked
also seen in female obstetricians, but this heightened risk did about the ‘worst’ severe incident that the respondents had
not remain after adjustment. Odds ratios for developing par- experienced, which might have happened decades ago. Fur-
tial PTSD/ PTSD remained statistically significant after thermore memory research has shown that there are many
adjusting for emotions of guilt and insufficient support from ways in which our memory can go awry. In this context it is
friends for both obstetricians and midwives, while negative of interest that people’s memory can be influenced when
reactions from parents remained significant only for obste- they are interrogated in a suggestive fashion or when they
tricians, and experiencing insufficient support from col- talk to other people who give their version of an event.19
leagues remained significant only for midwives. Both reassemblings and healthcare providers’ responses to
The obstetricians reporting partial or probable PTSD complaints procedures may play a role here.
more often changed their job situation by stopping work-
ing on the delivery unit, stopping being on call or changing Interpretation
their role to outpatient care. Such work-related effects were Having emotions of guilt was a major risk factor for devel-
not seen among those not affected by post-traumatic stress oping partial or probable PTSD. Experiencing negative
symptoms after exposure to a severe event. Similar effects reactions from the involved parents, which was a significant
were seen among midwives reporting partial PTSD or risk factor for obstetricians after adjustment, could be mul-
PTSD, i.e. they changed their work to outpatient care more tifactorial, such as actual clinical errors and the doctor’s
often than did those without post-traumatic stress symp- inability to give the required emotional response to those
toms after exposure to a severe event. However, midwives affected by the event. Patients wish for honest information
who reported symptoms of partial or probable PTSD fol- and apologies when appropriate.20–24 Insufficient support
lowing exposure to a severe event reported significantly from friends remained significant on professional group
higher levels of sick leave than did obstetricians (Table 3). level as well as after adjustment. Lack of social support, of
which friends could be considered a major part, is the
highest ranked risk factor predicting PTSD.25
Discussion
Our study also showed that reassembly (debriefing) fol-
Main findings lowing the severe event was associated with a crude risk
In this study, 7% of obstetricians and 5% of midwives increase in post-traumatic stress reactions, which is in line
reported symptoms in accordance with those of PTSD after with other studies that have concluded that debriefing

4 ª 2016 Royal College of Obstetricians and Gynaecologists

 Table 2. Risk of developing post-traumatic stress symptoms (partial PTDS) after exposure to severe events on the delivery ward for obstetricians (n = 594) and midwives (n = 1034).
Percentage (of all exposed), crude odds ratio (OR) and adjusted odds ratio (aOR).

All Obstetricians Midwives

% OR (95% CI) aOR (95% CI)* % OR (95%CI) aOR (95% CI)** % OR (95% CI) aOR (95% CI)***

ª 2016 Royal College of Obstetricians and Gynaecologists

Female 75.4 1.99 (1.07‒3.69) 1.84 (0.94‒3.60)
Single 10.7 1.37 (0.90–2.07) 8.1 0.86 (0.30–1.92) 12.3 1.51 (0.94–2.42)
Childless 8.5 1.25 (0.79‒1.99) 8.4 0.67 (0.30‒1.52) 8.6 1.45 (0.89‒2.36)
Emotions of guilt for something you did or did 35.3 2.70 (2.03‒3.58) 2.01 (1.49‒2.72) 47.3 4.63 (2.79‒7.67) 2.52 (1.45‒4.36) 28.4 3.65 (2.58–5.17) 2.04 (1.40–2.98)
not do in conjunction with the event
Negative reactions from parents 14.6 2.03 (1.45‒2.83) 1.53 (1.05‒2.23) 22.1 3.45 (2.13‒5.57) 2.22 (1.27‒3.89) 10.3 2.11 (1.30‒3.42) 1.23 (0.71‒2.13)
Experience of insufficient support from manager 42.3 2.15 (1.15–3.91) 1.50 (1.02‒2.22) 44.3 4.03 (2.48–6.54) 1.17 (0.56‒2.46) 41.1 5.50 (3.79–8.00) 1.55 (0.97–2.48)
Experience of insufficient support from colleagues 28.3 3.17 (2.39–4.20) 1.51 (1.04‒2.20) 35.4 4.92 (3.06‒7.91) 1.81 (0.85‒3.84) 24.3 4.90 (3.44–7.00) 1.57 (1.01‒2.44)
Experience of insufficient support from friends 26.2 3.48 (2.62–4.62) 2.24 (1.63‒3.08) 29.3 4.65 (2.92‒7.41) 1.82 (1.04–3.18) 24.4 5.64 (3.96–8.03) 2.34 (1.57–3.48)
Experience of insufficient support from partner 30.7 1.41 (1.06‒1.87) 1.25 (0.91‒1.69) 13.1 3.81 (2.22‒6.56) 1.78 (0.95‒3.34) 32.2 1.96 (1.39–2.80) 0.96 (0.67‒1.40)
Negative experience of reassembly (debriefing) 11.9 2.35 (1.65–3.35) 1.43 (0.97–2.10) 14.5 3.58 (2.10–6.08) 1.91 (1.05–3.48) 11.4 2.82 (1.76‒4.53) 1.21 (0.72–2.03)
Reported to national authority**** by hospital 26.6 1.33 (0.99‒1.79) 1.06 (0.76–1.47) 39.6 1.69 (1.07‒2.67) 0.82 (0.48–1.41) 19.0 1.99 (1.35–2.94) 1.36 (0.88–2.10)
organisation, patient or patient’s family

*aOR for all (guilt, negative reactions from parents, reassembly, support from manager, colleagues, friends, reported to authority by hospital organisation).
**aOR for obstetricians (sex, guilt, negative reactions from parents, reassembly, support from manager, colleagues, friends, reported to authority by hospital organisation).
***aOR for midwives (guilt, negative reactions from parents, reassembly, support from manager, colleagues, friends, reported to authority by hospital organisation).
****National Board of Health and Welfare or Health and Social Care Inspectorate (after 2013).
Post-traumatic stress in Swedish obstetricians and midwives

5
 Wahlberg et al.

Table 3. Professional long-term consequences for obstetricians and midwives exposed to severe events in the delivery ward by experience of
post-traumatic symptoms (partial PTSD/PTSD) or not.

Post-traumatic stress symptoms (partial PTSD/PTSD) [n (%)]

Obstetricians Midwives

No (n = 467) Yes (n = 88) P* No (n = 764) Yes (n = 153) P*

Sick leave 8 (1.7) 4 (4.5) 0.202 8 (1.0) 11 (7.2) <0.001

Midwives stopped working in the delivery ward — — 33 (4.3) 26 (17.0) <0.001
Obstetricians stopped working with emergency 11 (2.4) 9 (10.2) <0.001 — —
obstetrics daytime
Obstetricians stopped being on call 16 (3.4) 9 (10.2) 0.011 — —
Midwives and obstetricians moved to outpatient care 14 (3.0) 11 (12.5) <0.001 23 (3.0) 16 (10.5) <0.001

*Fisher’s exact test.

cannot be recommended for the prevention of PTSD.26 We In contrast to the midwives in our study, the obstetri-
used the term ‘reassembly’ in the survey since it could be cians with partial or probable PTSD did not take sick leave
questioned to what extent this intervention qualified for to a greater extent than did their less traumatised col-
the term ‘debriefing’, referring to the concepts of ‘critical leagues. This might give an indication of different ‘cultures’
incident stress debriefing’ by Mitchell27 or ‘psychological between the professions or age groups concerning the
debriefing’ by Dyregrov.28 However, the word ‘debriefing’ acceptance of psychological distress as a valid reason for
is commonly used by Swedish healthcare providers, which sick leave. When assessing the consequences after healthcare
is why we have used it in this paper. Similar proportions providers have been traumatised by a severe event there are
among obstetricians and midwives were unsatisfied with other aspects to consider apart from the suffering of those
the reassembly (debriefing) but a risk increase for partial or involved. It has been shown that depressed and burned-out
probable PTSD after adjustment was seen only for doctors. physicians make up to six times more medical errors than
One interpretation is the vulnerability in such reassembly do their non-depressed colleagues, and the most important
situations, in which the doctor bears the final responsibility contributing factor is lapse of judgement.5,30 It has also
and guilt more frequently than does the midwife. The been established that physicians in the role of second vic-
crude risk increase of post-traumatic stress reactions fol- tims tend to respond with a projection of blame towards
lowing reassembly in our study could, however, be related the patient or other members of the healthcare team.1,31
to the severity of the incident, i.e. professionals who had The prevalence of PTSD in our study is lower than the
experienced very severe events were offered reassembly to a 15% seen in a study on American trauma surgeons using a
greater extent than were those who had experienced events different screening test, the Posttraumatic Stress Disorder
of lesser severity. Checklist-C (PLC-C).32 In that study the number of years
Having been reported to the national authority by the in practice was not predictive of the development of PTSD,
hospital organisation was not associated with an increased in line with our results. Often, being a woman is correlated
risk of symptoms of partial or probable PTSD. One inter- to a heightened risk of PTSD, considered to be caused by
pretation of this finding might lie within the Swedish sys- increased vulnerability to emotional stress and formation of
tem itself, in which healthcare professionals are not memory, where neurobiological sex differences may play a
financially burdened by patient complaints or reports made role.13 Among trauma surgeons, however, being a man was
by hospitals. correlated to an increased risk of PTSD and in our study
In our study the sample sizes of obstetricians and mid- there were no significant differences between the genders
wives with probable PTSD were too small to make calcula- for obstetricians, after adjustment.32 Our study also found
tions of professional consequences in terms of sick leave a lower level of PTSD than the levels of secondary trau-
and changed working conditions. Our results, however, matic stress seen among American nurse-midwives (29%)
showed that when including cases of both partial and prob- and UK midwives (32%), but the response rates were con-
able PTSD, a significant proportion of obstetricians and siderably lower in these studies.7,16,33 Our results are in line
midwives changed their work situation and left emergency with a meta-analysis on work-related incidents and post-
obstetrics, a finding that confirms the results of previous traumatic stress symptoms showing that doctors and nurses
studies.4,29 often cope by working part-time or by changing jobs.29

6 ª 2016 Royal College of Obstetricians and Gynaecologists

 Post-traumatic stress in Swedish obstetricians and midwives

There is increasing acknowledgement of the dynamic Table S1. Experiences, emotions and symptoms of post-
process of resilience, for the individual healthcare provider traumatic stress for Swedish obstetricians (n = 594) and
as well as for the organisation.34–36 Awareness of vulnera- midwives (n = 1034) after exposure to severe events (A1
bility as well as realistic expectations, resilience training criterion) on the delivery ward. Chi2 test.&
including acceptance that things go wrong for most people
from time to time as well as support improving the group’s
trust and bonds may play important roles for obstetricians References
and midwives facing potentially traumatic events in their 1 Wu AW. Medical error: the second victim. The doctor who makes
work.34,35,37 the mistake needs help too. BMJ 2000;320:726–7.
2 Sirriyeh R, Lawton R, Gardner P, Armitage G. Coping with medical
error: a systematic review of papers to assess the effects of
Conclusions involvement in medical errors on healthcare professionals’
psychological well-being. Qual Saf Health Care 2010;19:e43.
This study provides evidence that obstetricians and mid- 3 Sachs MA, Baehrendtz S, Sellgren SF, Ullström S, Øvretveit J. Staff
wives, after experiencing a severe traumatic event while who have been involved in adverse events is left without help.
working on the delivery ward, may develop symptoms of Systematic support from colleagues and managers is desirable,
partial or probable PTSD. Resilience training with a poten- according to interview study. La€kartidningen 2013;110:550–2.
4 Scott SD, Hirschinger LE, Cox KR, McCoig M, Brandt J, Hall LW. The
tial to increase individual as well as group resilience may
natural history of recovery for the healthcare provider “second
prevent healthcare providers from developing symptoms of victim” after adverse patient events. Qual Saf Health Care
severe anxiety, and promote professional retaining. How- 2009;18:325–30.
ever, unanswered questions remain as to how to organise 5 Shanafelt TD, Balch CM, Bechamps G, Russell T, Dyrbye L, Satele D,
and implement resilience training for all obstetricians and et al. Burnout and medical errors among American surgeons. Ann
Surg 2010;251:995–1000.
midwives on the labour ward.
6 Hildingsson I, Westlund K, Wiklund I. Burnout in Swedish midwives.
Sex Reprod Healthc 2013;4:87–91.
Disclosure of interests 7 Beck CT, LoGiudice J, Gable RK. A mixed-methods study of
None declared. Completed disclosure of interests form secondary traumatic stress in certified nurse-midwives: shaken
available to view online as supporting information. belief in the birth process. J Midwifery Womens Health
2015;60:16–23.
8 Wu AW, Steckelberg RC. Medical error, incident investigation and
Contribution to authorship the second victim: doing better but feeling worse? BMJ Qual Saf
UH raised the initial idea of the study. The survey was 2012;21:267–70.
developed by all authors: ÅW, MAS, KBJ, GH, MJ, ASS 9 Ellis A, Chebsey C, Storey C, Bradley S, Jackson S, Flenady V, et al.
and UH. Statistical analyses were conducted by ÅW and Systematic review to understand and improve care after stillbirth: a
review of parents’ and healthcare professionals’ experiences. BMC
UH and writing was done by ÅW with support from the
Pregnancy Childbirth 2016;16:16.
co-authors; MAS, KBJ, GH, MJ, ASS and UH. 10 Glynn SM, Drebing C, Penk K. Psychosocial Rehabilitation. Effective
Treatment for PTSD. Practice guidelines from the Internatuional
Details of ethics approval Society for Traumatic Stress Studies, 2nd edn. New York, NY:
The study was approved by the Regional Ethical Board of Guilford Press, 2009.
11 Hall LW, Scott SD. The second victim of adverse health care events.
Uppsala on 8 Nov. 2013 (2013/351).
Nurs Clin North Am 2012;47:383–93.
12 Heazell AE, Siassakos D, Blencowe H, Burden C, Bhutta ZA,
Funding Cacciatore J, et al. Stillbirths: economic and psychosocial
This work was supported by a grant from AFA Insurances, consequences. Lancet 2016;387:604–16.
owned by the Confederation of Swedish Enterprise, the 13 Frans O, Rimmo PA, Aberg L, Fredrikson M. Trauma exposure and
post-traumatic stress disorder in the general population. Acta
Swedish Trade Union Confederation and the Council for
Psychiatr Scand 2005;111:291–9.
Negotiation and Cooperation, and by the Medial Faculty of 14 American Psychiatric Association. Diagnostic and Statistical Manual of
Uppsala University. Mental Disorders. Washington, DC: American Psychiatric Press, 1994.
15 Breslau N, Lucia VC, Davis GC. Partial PTSD versus full PTSD: an
Acknowledgement empirical examination of associated impairment. Psychol Med
€
© SQ-PTSD 1997 Fredrikson & Frans revised by Orjan 2004;34:1205–14.
16 Sheen K, Spiby H, Slade P. Exposure to traumatic perinatal
Frans in 2006. experiences and posttraumatic stress symptoms in midwives:
prevalence and association with burnout. Int J Nurs Stud
2015;52:578–87.
Supporting Information 17 Wallbank S, Robertson N. Predictors of staff distress in response to
professionally experienced miscarriage, stillbirth and neonatal loss: a
Additional Supporting Information may be found in the
questionnaire survey. Int J Nurs Stud 2013;50:1090–7.
online version of this article:

ª 2016 Royal College of Obstetricians and Gynaecologists 7

 Wahlberg et al.

18 Hussain A, Weisaeth L, Heir T. Nonresponse to a population-based 29 de Boer J, Lok A, Van’t Verlaat E, Duivenvoorden HJ, Bakker AB,
postdisaster postal questionnaire study. J Trauma Stress Smit BJ. Work-related critical incidents in hospital-based health care
2009;22:324–8. providers and the risk of post-traumatic stress symptoms,
19 Loftus EF. Make-believe memories. Am Psychol 2003;58:867–73. anxiety, and depression: a meta-analysis. Soc Sci Med 2011;73:316–
20 Carranza L, Lyerly AD, Lipira L, Prouty CD, Loren D, Gallagher TH. 26.
Delivering the truth: challenges and opportunities for error 30 Fahrenkopf AM, Sectish TC, Barger LK, Sharek PJ, Lewin D, Chiang
disclosure in obstetrics. Obstet Gynecol 2014;123:656–9. VW, et al. Rates of medication errors among depressed and burnt
21 Gallagher TH, Waterman AD, Ebers AG, Fraser VJ, Levinson W. out residents: prospective cohort study. BMJ 2008;336:488–91.
Patients’ and physicians’ attitudes regarding the disclosure of 31 Engel KG, Rosenthal M, Sutcliffe KM. Residents’ responses to
medical errors. JAMA 2003;289:1001–7. medical error: coping, learning, and change. Acad Med
22 Gallagher TH, Garbutt JM, Waterman AD, Flum DR, Larson EB, 2006;81:86–93.
Waterman BM, et al. Choosing your words carefully: how physicians 32 Joseph B, Pandit V, Hadeed G, Kulvatunyou N, Zangbar B, Tang A,
would disclose harmful medical errors to patients. Arch Intern Med et al. Unveiling posttraumatic stress disorder in trauma surgeons: a
2006;166:1585–93. national survey. J Trauma Acute Care Surg 2014;77:148–54
23 O’Connor E, Coates HM, Yardley IE, Wu AW. Disclosure of patient discussion 54.
safety incidents: a comprehensive review. Int J Qual Health Care 33 Creedy DK, Gamble J. A third of midwives who have experienced
2010;22:371–9. traumatic perinatal events have symptoms of post-traumatic stress
24 Wu AW, Huang IC, Stokes S, Pronovost PJ. Disclosing medical errors disorder. Evid Based Nurs 2016;19:44.
to patients: it’s not what you say, it’s what they hear. J Gen Intern 34 Livingstone C. Misunderstanding the meaning of resilience. BMJ
Med 2009;24:1012–17. 2015;350:h1423.
25 Brewin CR, Andrews B, Valentine JD. Meta-analysis of risk factors 35 Howe A, Smajdor A, Stockl A. Towards an understanding of
for posttraumatic stress disorder in trauma-exposed adults. J Consult resilience and its relevance to medical training. Med Educ
Clin Psychol 2000;68:748–66. 2012;46:349–56.
26 Rose S, Bisson J, Churchill R, Wessely S. Psychological debriefing for 36 Hegney DG, Rees CS, Eley R, Osseiran-Moisson R, Francis K. The
preventing post traumatic stress disorder (PTSD). Cochrane Database contribution of individual psychological resilience in determining the
Syst Rev 2002;CD000560. professional quality of life of Australian nurses. Front Psychol
27 Mitchell JT. When disaster strikes. . .the critical incident stress 2015;6:1613.
debriefing process. JEMS 1983;8:36–9. 37 Firth-Cozens J. Doctors, their wellbeing, and their stress. BMJ
28 Dyregrov A. The process in psychological debriefings. J Trauma 2003;326:670–1.
Stress. 1997;10:589–605.

8 ª 2016 Royal College of Obstetricians and Gynaecologists