Maaløe et al.

BMC Pregnancy and Childbirth (2016) 16:351

DOI 10.1186/s12884-016-1142-2

RESEARCH ARTICLE Open Access

Stillbirths and quality of care during labour

at the low resource referral hospital of
Zanzibar: a case-control study
Nanna Maaløe1* , Natasha Housseine2,3, Ib Christian Bygbjerg1, Tarek Meguid2,4, Rashid Saleh Khamis2,
Ali Gharib Mohamed2, Birgitte Bruun Nielsen5 and Jos van Roosmalen6

Abstract
Background: To study determinants of stillbirths as indicators of quality of care during labour in an East African
low resource referral hospital.
Methods: A criterion-based unmatched unblinded case-control study of singleton stillbirths with birthweight ≥2000 g
(n = 139), compared to controls with birthweight ≥2000 g and Apgar score ≥7 (n = 249).
Results: The overall facility-based stillbirth rate was 59 per 1000 total births, of which 25 % was not reported in
the hospital’s registers. The majority of singletons had birthweight ≥2000 g (n = 139; 79 %), and foetal heart rate
was present on admission in 72 (52 %) of these (intra-hospital stillbirths). Overall, poor quality of care during
labour was the prevailing determinant of 71 (99 %) intra-hospital stillbirths, and median time from last foetal
heart assessment till diagnosis of foetal death or delivery was 210 min. (interquartile range: 75–315 min.). Of
intra-hospital stillbirths, 26 (36 %) received oxytocin augmentation (23 % among controls; odds ratio (OR) 1.86,
95 % confidential interval (CI) 1.06–3.27); 15 (58 %) on doubtful indication where either labour progress was
normal or less dangerous interventions could have been effective, e.g. rupture of membranes. Substandard
management of prolonged labour frequently led to unnecessary caesarean sections. The caesarean section rate
among all stillbirths was 26 % (11 % among controls; OR 2.94, 95 % CI 1.68–5.14), and vacuum extraction was
hardly ever done. Of women experiencing stillbirth, 27 (19 %) had severe hypertensive disorders (4 % among
controls; OR 5.76, 95 % CI 2.70–12.31), but 18 (67 %) of these did not receive antihypertensives. An additional 33
(24 %) did not have blood pressure recorded during active labour. When compared to controls, stillbirths were
characterized by longer admissions during labour. However, substandard care was prevalent in both cases and
controls and caused potential risks for the entire population. Notably, women with foetal death on admission
were in the biggest danger of neglect.
Conclusions: Intrapartum management of women experiencing stillbirth was a simple yet strong indicator of
quality of care. Substandard care led to perinatal as well as maternal risks, which furthermore were related to
unnecessary complex, time consuming, and costly interventions. Improvement of obstetric care is warranted to
end preventable birth-related deaths and disabilities.
Trial registration: This is the baseline analysis of the PartoMa trial, which is registered on ClinicalTrials.org
(NCT02318420, 4th November 2014).
Keywords: Tanzania, Low resource, Stillbirths, Labour, Quality of care, PartoMa, Caesarean section, Severe
hypertensive disorders, Oxytocin, Criterion-based audit, Case-control study, Guidelines, Partograph

* Correspondence: [email protected]
1
Global Health Section, Department of Public Health, University of
Copenhagen, Øster Farimagsgade 5, Building 9, 1353 Copenhagen K,
Denmark
Full list of author information is available at the end of the article

© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 2 of 12

Background estimates from 2010 suggest a rate of 50 perinatal deaths

More than a quarter of a million women and 2.7 million per 1000 total births, as opposed to 36 per 1000 in main-
newborn babies lose their lives during pregnancy and land Tanzania [17].
childbirth annually [1, 2]. Though often invisible in global The governmental Mnazi Mmoja Hospital is the only
estimates, an additional 2.6 million stillbirths add pro- tertiary care facility on the archipelago, and at the time
foundly to the tragedy, of which half are estimated to of data collection, the only hospital on the biggest island
occur during labour [3]. In all three groups, the vast ma- providing comprehensive obstetric and neonatal care
jority of deaths are caused by largely avoidable obstetric around the clock. In 2014, 13 291 women delivered in
complications with the highest risk at the time of birth the hospital, corresponding with an average of 36 deliv-
[1–3]. Many more women continue to suffer from birth eries daily. Of these, 16 % were caesarean sections (CSs),
related injuries, infections, and disabilities, and an esti- and 44 maternal deaths were counted. According to the
mated one million survivors of birth asphyxia may end up official registers, 41 babies were stillborn in 2014 per
with cerebral palsy, learning difficulties, or other disabil- 1000 total births, and 17 neonatal deaths occurred up to
ities [4, 5]. The world’s highest burden of maternal and discharge per 1000 live births.
perinatal deaths and other birth-related complications The hospital plays a leading role in clinical training of
remains in sub-Saharan Africa and Asia [1–3]. future Tanzanian health care providers. Yearly, approxi-
In the Millenium Development Goals’ era, the global mately 60 intern doctors do their initial clinical rotations
strategy mainly aimed at skilled birth attendance, which at the department, and more than 200 Tanzanian nurs-
resulted in campaigns for women to deliver in health ing and medical students are trained.
facilities. The increasing proportion of facility births, Intrapartum care is located in two rooms only. In
however, has not been matched with improvements in the labour room’s 18 beds, women are assisted during
the quality of intra-facility labour care [6, 7]. Notably, the first stage of labour. Postpartum women needing
reports from referral hospitals in sub-Saharan Africa extra surveillance stay in this room as well. When
suggest ample room for improvement even at the ter- women reach the second stage of labour, they walk 15
tiary level of the health care sector [8–11]. Importantly, m to the delivery room with only three beds. There is
these are the facilities where most of the countries’ one, occasionally two, theatres available for obstetric
future health care workers are trained, and if quality of surgery. For more than a decade, it has been the aim
care was improved, they could possibly be a lever for to apply the World Health Organization’s (WHOs)
achieving nationwide health care improvements. Hence, composite partograph for all women in labour, which
in-depth insight into contextual challenges in delivering is a graphic monitoring sheet including foetal heart
intrapartum quality of care is vital. rate (FHR), labour progress, and maternal vital signs
This paper is part of a baseline study for the PartoMa during latent and active phase of labour [18]. A treat-
project, which aims at improving labour outcome for ment and observation sheet for severe hypertensive
women and their offspring at the referral hospital of disorders has been available since June 2014 and
Zanzibar, Mnazi Mmoja Hospital. The project focuses includes evidence-based guidelines on anticonvulsant
on understanding direct and underlying determinants and antihypertensive treatment [19].
of substandard quality of care as well as strengthening After uncomplicated delivery, the woman and baby
monitoring and decision-making during labour [12]. are usually observed in the labour room for two hours
We here present a case-control study of intrapartum before referral to the postnatal ward for another four
management when the outcome was stillbirth. Although hours where no routine observations are done. Babies
intrapartum stillbirths are considered a sensitive indicator with birth asphyxia, with birthweights <1500 gram (g),
of quality of care at the time of birth, there are few such or delivered by CS are admitted to the Neonatal Inten-
studies from low income settings [3, 8, 11, 13, 14]. sive Care Unit, which is adjacent to the delivery room.
At day time, there are on average three nurse-midwives
and two registrar doctors on duty in the labour and deliv-
Methods ery rooms as well as a fluctuating number of intern doc-
Setting tors. During evenings and nights, there are two or three
The Zanzibar archipelago, a semiautonomous part of nurse-midwives on duty for all obstetric patients and one
Tanzania, struggles with poverty and a resource constraint registrar doctor assisted by two interns are in charge of
health system. Half of the 1.3 million Zanzibarians live the entire department, which also includes an average
below the poverty line [15]. In 2011, the maternal mortality of 30 gynaecological patients. A second doctor is on
ratio was reported at 287 deaths per 100 000 live births, of call from home. At the department, there are two spe-
which the majority occurs during or shortly after childbirth cialist obstetricians, of which one is member of the
[16]. Though little is known about perinatal mortality, study team (TM).
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 3 of 12

Study population (intra-hospital stillbirths). Hence, pre-hospital stillbirths

All stillbirths defined as late foetal deaths ≥1000 g [20] included not only cases in which the stillbirth diagnosis
were identified between 1st October 2014 and 31st January was made on admission, but also cases where there was
2015 in the admission, delivery, and theatre registers, and no documentation of FHR (neither present nor absent)
their case files were searched for. In addition, to give the during the admission. While deliveries from multiple
best possible estimate of the actual hospital-based stillbirth gestations were included in the overall estimation of the
rate, non-registered stillbirths were sought by systematically stillbirth rate, they were not included in the further audit
going through all case files from the four months. Due to a process as their case files were often too ambiguous; e.g.
breakdown in the hospital’s storage system of case files frequently only one FHR was registered throughout
prior to October 2015, four months was the maximum labour. The 2000 g cut-off was decided as this reflects
feasible period for the study. The control group was identi- gestational age of 32–34 weeks, where lung maturity no
fied in the same registers and from the same time period. longer plays a major role in survival, and the newborn is
The first control included in October 2014 was selected by less dependent on dexamethasone treatment prior to de-
throwing dices and afterwards every tenth delivery was livery as well as advanced intensive neonatal care [22].
identified, resulting in a case control ratio of approximately Data on the very preterm stillbirths are only presented
1:4. These tasks were conducted by three research assis- in Additional file 1.
tants (NH, RSK, and AGM), who had weekly meetings with Inclusion criteria for controls were singletons with
TM or NM to assure quality of files retrieval. birthweight ≥2000 g and Apgar score ≥7; hence a group
Afterwards, all case files were checked individually by with immediate good neonatal outcome. It varied whether
NM to determine whether the inclusion criteria were one, five or 10 min (min.) Apgar scores were recorded. It
met for in-depth criterion-based audit (Fig. 1). was therefore decided to use the latest. The case-control
All singleton stillbirths were included for the audit and study was unblinded.
divided into three groups: birthweight 1000–1999 g (very
preterm stillbirths) [21]; birthweight ≥2000 g without Criteria of realistic best quality of care
positive FHR on admission (pre-hospital stillbirths), and Criteria reflecting locally best possible labour manage-
birthweight ≥2000 g with positive FHR on admission ment with the available resources were formulated and

Fig. 1 Sampling of case files. Facility-based stillbirth rate was 59 per 1000 total births. Stillbirths: All late foetal deaths with birthweight ≥1000 g.
Pre-hospital stillbirths: No documented positive foetal heart rate on admission. Intra-hospital stillbirths: Documented positive foetal heart rate on
admission. *Groups compared by the case-control study
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 4 of 12

agreed upon. This developmental process was conducted Background characteristics

by a participatory approach including both local skilled Among intra-hospital stillbirths, 39/72 (54 %) were
birth attendants and seven external specialists in midwifery nulliparous versus 14/64 (22 %) pre-hospital stillbirths
and obstetrics. It was decided that the standards should (odds ratio [OR] 4.22, 95 % confidence interval [CI]
include routine labour care as well as management of fre- 1.99–8.96) and 105/239 (44 %) controls (OR 1.51, 95 %
quent complications of labour, related to FHR and foetal CI 0.89–2.56). All women, except one, had attended ante-
distress, labour progression by dilatation and descent, and natal care at least once. Of women experiencing stillbirth,
maternal vital signs with specific focus on severe hyperten- 69/119 (58 %) had attended at least four visits, versus 103/
sive disorders and fever. Selected criteria were adapted from 214 (48 %) controls (OR 1.49, 95 % CI 0.95–2.34). Cases
the Active management of labour package, modified by with missing information are excluded from these compari-
WHO, and supplemented with other evidence-based sons (Table 1).
guidelines [23–31]. Frequency of routine assessments Of multiparous women experiencing stillbirth, 30/83
was reduced to reflect local reality. For example, FHR (36 %) had also previously experienced the loss of one or
assessments every 30 min. for all women in active labour more children, compared to 30/134 (22 %) controls (OR
were kept as optimal practice, but assessments within in- 1.96, 95 % CI 1.07–3.59). A history of previous CS
tervals of <90 min. were applied as an acceptable audit occurred in 19/83 (23 %) compared to 18/134 (13 %)
criterium. controls (OR 1.91, 95 % CI 0.94–3.90; Table 1).
In addition, information was collected on background Of stillbirths, 35/135 (26 %) were delivered by CS,
and admission characteristics as well as outcome pa- compared to 26/244 (11 %) controls (OR 2.94, 95 % CI
rameters. Finally, if information was available, time 1.68–5.14). While 10/35 (26 %) of the CSs resulting in
from last FHR assessment till delivery or diagnosis of stillbirth were performed in the second stage of labour,
intrauterine foetal death was calculated, as well as the this was never the case for controls. Only one baby was
admission to delivery interval. delivered by vacuum extraction. Of intra-hospital still-
births delivered by CS, 10/20 (50 %) had FHR docu-
Data extraction and analysis mented at the time of deciding on CS. In five of these
Data was extracted into a structured entry form based ten women, FHR were absent. Foetal distress was recorded
on the pre-selected audit criteria, using EPI INFO 7 as an indication for CS in only one case. Indications for
software (Centres for Disease Control and Prevention, CSs resulting in stillbirth were: 21/35 (60 %) for prolonged
Atlanta, GA, USA). Differences were analysed by logistic labour, of which one was also due to foetal distress; 8/35
regression in SAS Enterprise Guide 6.1 (SAS Institute, (23 %) due to antepartum haemorrhage, of which three had
Inc., Cary, NC). P-values <0.05 were considered statisti- severe pre-eclampsia; an additional 2/35 (6 %) solely due to
cally significant. severe pre-eclampsia; 2/35 (6 %) due to cord prolapse; and
2/35 (6 %) due to ≥2 previous CSs. Concerning the pro-
Results longed labour group, 1/21 (5 %) had signs of obstructed
During the four months studied, 216 stillbirths occurred labour on admission and another 6/21 (29 %) developed
out of 3690 total births. This corresponds to an overall uterine rupture; half had signs of rupture on admission, of
hospital-based stillbirth rate of 59 per 1000 total births, which two were referrals, and three ruptures occurred at
of which 53/216 (25 %) were not reported in the hospi- the study site. Four of the women with uterine rupture had
tal’s official registers. Case files of 186/216 (86 %) still- a history of previous CS (Table 2).
births and 301/364 (83 %) controls could be retrieved, There were three maternal deaths in the study popula-
of which 175/186 (94 %) and 293/301 (97 %) were from tion. They were all associated with stillbirth and suffered
singleton gestations. Of the singleton stillbirths, 139/ from severe delays in intrahospital surveillance and
175 (79 %) had a birthweight ≥2000 g and were in- management (Additional file 2).
cluded in the case-control study. Of these, 72/139
(52 %) had a positive foetal heart rate on admission Admission and partograph use
(intra-hospital stillbirths), with only one having a con- Women experiencing intra-hospital stillbirths were
genital abnormality that may have been incompatible admitted particularly early; 42/71 (59 %) before the
with life. Of the 67 pre-hospital stillbirths, 20/67 active phase of labour, compared to 68/246 (28 %)
(30 %) had no documentation of FHR readings during controls (OR 3.79, 95 % CI 2.19–6.57). Their median
the admission. Classification in ‘fresh’ and ‘macerated’ time from admission to delivery was 11 h and 36 min.
stillbirths was not recorded in 77/139 (55 %) stillbirths, (interquartile range (IQR): 5 h and 42 min. to 21 h
and it was therefore not useful for determining the rate and 55 min.). Rates of referrals were similar among
of intrapartum deaths. Of controls, 249 met the inclu- pre- and intra-hospital stillbirths. The referral rate
sion criteria (Fig. 1). among stillbirths (21/139; 15 %) was higher than the
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 5 of 12

Table 1 Characteristics of delivering women

Case-control study
BW ≥2000 g
Cases Cases Controls
Pre-hosp. Stillbirths Intra-hosp. Stillbirths Apgar 7–10
N (%)
Of all women in the study: (n = 67) (n = 72) (n = 249)
Age
<20 years 2 (3.0 %) 7 (9.7 %) 26 (10.4 %)
20–29 years 26 (38.8 %) 35 (48.6 %) 122 (49.0 %)
30–39 years 27 (40.3 %) 28 (38.9 %) 83 (33.3 %)
≥40 years 8 (11.9 %) 2 (2.8 %) 15 (6.0 %)
Information missing 4 (6.0 %) 0 (0.0 %) 3 (1.2 %)
Parity on admission
Para 0a 14 (20.9 %) 39 (54.2 %) 105 (42.2 %)
Para 1–4 33 (49.3 %) 23 (31.9 %) 99 (39.8 %)
Para ≥ 5 17 (25.4 %) 10 (13.9 %) 35 (14.1 %)
Information missing 3 (4.5 %) 0 (0.0 %) 10 (4.0 %)
Antenatal care
≥4 visits 31 (46.3 %) 38 (52.8 %) 103 (41.4 %)
1–3 visits 23 (34.3 %) 26 (36.1 %) 111 (44.6 %)
Not attended 0 (0.0 %) 1 (1.4 %) 0 (0.0 %)
Information missing 13 (19.4 %) 7 (9.7 %) 35 (14.1 %)
HIV
Negative 54 (80.6 %) 62 (86.1 %) 211 (84.7 %)
Positive 0 (0.0 %) 2 (2.8 %) 0 (0.0 %)
Information missing 13 (19.4 %) 8 (11.1 %) 38 (15.3 %)
Gestational age
No information on LMP/gestation weeks 46 (68.7 %) 49 (68.1 %) 181 (72.7 %)
Previous obstetric history
Of multiparous women: (n = 50) (n = 33) (n = 134)
Previous death of child/childrenb,c 18 (36.0 %) 12 (36.4 %) 30 (22.4 %)
1 previous CS 7 (14.0 %) 8 (24.2 %) 8 (6.0 %)
≥2 previous CSs 2 (4.0 %) 2 (6.1 %) 10 (7.5 %)
BW birthweight, CI confidence interval, CS caesarean section, LMP last menstrual period, OR odds ratio
a
Difference between pre- and intra-hospital stillbirths: OR 4.22, 95 % CI 1.99–8.96
b
Documentation was insufficient to clearly distinguish perinatal deaths from deaths later in life
c
Difference between stillbirths and controls: OR 1.96, 95 % CI 1.07–3.59

rate among controls (12/249 (5 %); OR 3.52, 95 % CI Foetal heart rate (FHR)
1.67–7.39; Table 3). In all intra-hospital stillbirths, FHR was reassuring on
Of women reaching active labour and admitted before admission. However, in 60/72 (83 %) >90 min. elapsed
second stage, significantly more in the pre-hospital still- between FHR assessments during active phase of labour,
birth group did not have a partograph filled in, when which was the case for 137/204 (67 %) controls (OR
compared to both intra-hospital stillbirths (OR 9.78, 2.45, 95 % CI 1.23–4.85; Fig. 2). Among 63 intra-hospital
95 % CI 2.56–37.42) and controls (OR 3.39, 95 % CI stillbirths, median time from last FHR recording till
1.52–7.56; Fig. 2). In all groups, 237/276 (86 %) women delivery or detected intrauterine foetal death was 3 h
with a partograph applied had the first cervical dilata- and 30 min. (IQR: 1 h and 15 min.–5 h and 15 min.),
tion appropriately plotted on the alert line (Table 3). compared to 2 h and 0 min. in 176 controls (IQR: 1 h
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 6 of 12

Table 2 Mode of delivery, maternal outcome, and appearance of stillborn babies

Case-control study
BW ≥2000 g
Cases Cases Controls
Pre-hosp. Stillbirths Intra-hosp. Stillbirths Apgar 7–10
N (%)
Mode of delivery
Of all women in the study: (n = 67) (n = 72) (n = 249)
Spontaneous vaginal 45 (67.2 %) 46 (63.9 %) 213 (85.5 %)
Vaginal breech 3 (4.5 %) 5 (6.9 %) 5 (2.0 %)
Vacuum extraction 1 (1.5 %) 0 (0.0 %) 0 (0.0 %)
Caesarean sectiona,b 15 (22.4 %) 20 (27.8 %) 26 (10.4 %)
Mode of delivery unknown 3 (4.5 %) 1 (1.4 %) 5 (2.0 %)
Maternal outcome
Of all women in the study: (n = 67) (n = 72) (n = 249)
Maternal deaths 2 (3.0 %) 1 (1.4 %) 0 (0.0 %)
c
Post partum haemorrhage 7 (10.4 %) 10 (13.9 %) 14 (5.6 %)
Episiotomy/spontaneous tearsd,e 6 (9.0 %) 19 (26.4 %) 79 (31.7 %)
Of vaginal deliveries: (n = 49) (n = 51) (n = 218)
Prolonged admission, ≥1 dayf 9 (18.4 %) 0 (0.0 %) 3 (1.4 %)
Of caesarean sections: (n = 15) (n = 20) (n = 26)
Prolonged admission, ≥5 days 1 (6.7 %) 3 (15.0 %) 2 (7.7 %)
‘Fresh’ versus ‘macerated’ stillbirths
Of all women in the study: (n = 67) (n = 72) (n = 249)
Classification not recorded 36 (53.7 %) 41 (56.9 %) NA
BW birthweight, CI confidence interval, NA not applicable; OR, odds ratio
a
Overall, 9/35 (26 %) of the caesarean sections with stillbirth were done prior to active labour, and 10/35 (29 %) in second stage. Among controls, this was the
case for 13/26 (50 %) and 0/26 (0 %), respectively
b
Difference between stillbirths and controls: OR 2.94, 95 % CI 1.68–5.14
c
Difference between stillbirths and controls: OR 2.34, 95 % CI 1.12–4.90
d
Information was insufficient to distinguish between spontaneous vaginal tears and episiotomies
e
Difference between pre-hospital stillbirths and controls: OR 0.21, 95 % CI 0.09–0.51
f
Difference between pre-hospital stillbirths and controls: OR 16.13, 95 % CI 4.18–62.17

and 3 min.–3 h and 58 min.). For each one-hour in- which was a common complication among intra-hospital
crease in duration from last FHR assessment, the odds stillbirths when compared to controls (Table 5). After
of stillbirth increased 20 % (OR 1.20; 95 % CI 1.08– crossing the alert line, in 18/33 (55 %) and 9/51 (18 %),
1.34). In 58/72 (81 %) of the intra-hospital stillbirths, respectively, ≥3 h elapsed before next vaginal examination
there was no documentation of foetal distress or foetal (OR 5.60, 95 % CI 2.07–15.13). After crossing the action
death prior to delivery (Table 4). line, in 2/16 (13 %) intra-hospital stillbirths and 9/21
(43 %) controls, membranes were still intact, and in an
Labour progress additional 3/16 (19 %) and 5/21 (24 %), there was no infor-
The highest proportion of women admitted in the latent mation regarding membranes. Moreover, severe delays in
phase of labour with no cervical assessments recorded surveillance were found after crossing the action line.
during active labour occurred among pre-hospital still- Oxytocin for labour augmentation was administered in
births: 14/23 (61 %) compared to 24/68 (35 %) controls 26/72 (36 %) of intra-hospital stillbirths, compared to
(OR 2.85, 95 % CI 1.08–7.55; Table 5). 58/249 (23 %) controls (OR 1.86, 95 % CI 1.06–3.27).
In 27/69 (39 %) women experiencing intra-hospital However, in 8/26 (31 %) of those, there was no indication
stillbirth, ≥5 h elapsed between any two vaginal exami- for augmentation, and in an additional 7/26 (27 %) the infu-
nations during active labour, compared to 40/207 (19 %) sion was started between the alert and action line with the
controls (OR 2.68, 95 % CI 1.48–4.86; Fig. 2). This re- membranes still intact. Likewise, 34/58 (59 %) controls had
sulted in delays in diagnosing poor labour progress, the infusion started before crossing the alert line (Fig. 3). In
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 7 of 12

Table 3 Admission and partograph use

Case-control study
BW ≥2000 g
Cases Cases Controls
Pre-hosp. stillbirths Intra-hosp. Stillbirths Apgar 7–10
N (%)
Progress on admission and referrals
Of all women in the study: (n = 67) (n = 72) (n = 249)
a
Before labour pain 5 (7.5 %) 2 (2.8 %) 12 (4.8 %)
Latent phase of laboura,b 18 (26.9 %) 40 (55.6 %) 56 (22.5 %)
First stage of labour 23 (34.3 %) 29 (40.3 %) 153 (61.4 %)
Second stage of labour 15 (22.4 %) 0 (0.0 %) 25 (10.0 %)
Stage of labour on admission unknown 6 (9.0 %) 1 (1.4 %) 3 (1.2 %)
Referral from smaller health centrec 10 (14.9 %) 11 (15.3 %) 12 (4.8 %)
Partograph use
Of women in first stage of labour: (n = 39) (n = 69) (n = 207)
d
The partograph at least partially applied 27 (69.2 %) 66 (95.7 %) 183 (88.0 %)
Of women with the partograph applied: (n = 27) (n = 66) (n = 183)
First cervical dilatation in active labour plotted correctly on the alert line 18 (66.7 %) 53 (80.3 %) 166 (90.7 %)
BW birthweight, CI confidence interval, OR odds ratio
a
Difference in women admitted before active labour between intra-hospital stillbirths and controls: OR 3.79, 95 % CI 2.19–6.57
b
Cervical dilatation <4 cm
c
Difference between intra-hospital stillbirths and controls: OR 3.52, 95 % CI 1.67–7.39
d
Difference between pre-hospital stillbirths and both intra-hospital stillbirths and controls: OR 9.78, 95 % CI 2.56–37.42, and OR 3.39, 95 % CI
1.52–7.56, respectively

none of the cases, information on oxytocin titration Maternal vital signs

and maintenance dose was documented, and FHR and Prevalence of severe hypertensive disorders was signifi-
contractions were never assessed half hourly after starting cantly higher among women experiencing stillbirth (27/
the infusion. Preceeding 13/35 (37 %) of CSs resulting in 139, 19 %) than in controls (10/249, 4 %; OR 5.76, 95 % CI
stillbirth, oxytocin was started for labour augmentation. 2.70–12.31). An additional 33/130 (25 %) and 81/235
This was also the case in four of six uterine ruptures, of (35 %) of women reaching active phase of labour, had no
which three had a history of one previous CS. recordings of blood pressure during active labour (OR

Fig. 2 Proportion of labouring women reaching each of six criteria for minimal acceptable routine surveillance during labour. Significant
differences were found in FHR (OR 0.41, 95 % CI 0.21–0.81), cervical dilatation (OR 0.37, 95 % CI 0.21–0.68), and contractions (OR 0.26, 95 % CI
0.14–0.47). Intra-hospital stillbirths: documented positive FHR on admission, birthweight ≥2000 g. Controls: Apgar score ≥7, birthweight ≥2000 g.
* Of all women at the hospital during active first stage of labour (n = 69 and n = 207, respectively). ** Of women with at least one FHR reading
(n = 72 and n = 204, respectively). *** Of women reaching active phase of labour (n = 70 and n = 235, respectively). **** Of all women in the study
(n = 72 and n = 249, respectively). FHR, foetal heart rate; BP, blood pressure; Temp, temperature
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 8 of 12

Table 4 Intrapartum surveillance of the foetus severe pre-eclampsia cases, 4/13 (31 %) had no documen-
Case-control study tation of having received magnesium sulphate.
BW ≥2000 g In 43/139 (31 %) stillbirths and 104/249 (42 %) con-
Cases Controls trols, there was no maternal temperature recording from
Intra-hosp. Apgar 7–10 admission till delivery (OR 0.63, 95 % CI 0.40–0.97;
Stillbirths
Fig. 2). Intrapartum fever or infection were rare diagno-
N (%)
ses with five stillbirths related to infection and none
Of women with at least one FHR reading: (n = 72) (n = 204)
among controls.
FHR in normal range on admission 72 (100.0 %) 202 (99.0 %)
(110–160 beats per min.)
Discussion
Foetal distress detected prior to delivery 15 (20.8 %) 0 (0.0 %) The overall facility-based stillbirth rate was 59 per
<90 min. between any 2 recordings 12 (16.7 %) 67 (32.8 %) 1000 total births. Approximately 80 % of the singleton
of FHRa
stillbirths had a birthweight ≥2000 g. In half of these,
Median time from last FHR till delivery 210 120 the FHR was still present after admission to hospital.
or detected IUFD (min.)b,c
In all groups, major challenges were identified in intrapar-
BW birthweight, CI confidence interval, FHR foetal heart rate, min. minutes, OR
odds ratio
tum surveillance, timely decision-making, and documenta-
a
Difference between intra-hospital stillbirths and controls: OR 0.41, 95 % tion. This resulted in stillbirths as well as unacceptable
CI 0.21–0.81 maternal and neonatal risks for all women and babies. The
b
It was possible to calculate average time from last FHR till delivery in 63
(86 %) cases and 176 (86 %) controls. The interquartile ranges were 75–315 min. findings are largely in line with the limited number of other
and 63–238 min., respectively
c
stillbirth studies from sub-Saharan Africa [3, 11, 13, 14].
For each one-hour increase in duration from last FHR assessment, the odds of
stillbirth increased 20 % (OR 1.20; 95 % CI 1.08–1.34)
Our study provides a more in-depth assessment of intrapar-
tum care, which may contribute to effectively target inter-
ventions to reduce risks through improved quality of care
0.65, 95 % CI 0.40–1.04; Fig. 2). Overall, 13/27 (48 %) of (Table 6).
all stillbirth cases with severe hypertension had significant
proteinuria (≥2+ on urine dipstick). However, urine ana- Causes of stillbirths
lysis was not recorded in an additional 6/27 (22 %) cases, As suggested in other studies, the high number of intra-
and information about clinical symptoms were too sparse hospital stillbirths appeared primarily to be a sensitive
to analyse for signs of organ failure. In 18 (67 %) of all 27 indicator of substandard quality of care [3, 32, 33]. For
stillbirth cases with severe hypertension, there were no re- instance, primigravid women suffered an increased risk
cordings of relevant antihypertensive treatment. Of the of intra-hospital stillbirths, which may be associated with

Table 5 Intrapartum surveillance of labour progress

Case-control study
BW ≥2000 g
Cases Cases Controls
Pre-hosp. Stillbirths Intra-hosp. Stillbirths Apgar 7–10
N (%)
Surveillance in latent phase of labour
Of women admitted before active labour: (n = 23) (n = 42) (n = 68)
Assessment of cervical dilatation during active laboura,b 9 (39.1 %) 37 (88.1 %) 44 (64.7 %)
Assessment of labour progression
Of women in first stage of active labour: (n = 39) (n = 69) (n = 207)
c
<5 h. between any 2 recordings of cervical dilatation in active labour 39 (100.0 %) 42 (60.9 %) 167 (80.3 %)
<3 h. between any 2 recordings of uterine contractionsd 33 (84.6 %) 18 (26.1 %) 120 (58.0 %)
e
Alert line crossed 2 (5.1 %) 33 (47.8 %) 51 (24.5 %)
Action line crossedf 1 (2.6 %) 16 (23.2 %) 21 (10.1 %)
BW birthweight, CI confidence interval, OR odds ratio
a
If a vaginal examination was done in latent phase ≤4 h prior to delivery, this was registered as acceptable
b
Difference between pre-hospital stillbirths and controls: OR 0.35, 95 % CI 0.13–0.93
c
Difference between intra-hospital stillbirths and controls: OR 0.37, 95 % CI 0.21–0.68
d
Difference between intra-hospital stillbirths and controls: OR 0.26, 95 % CI 0.14–0.47
e
Difference between intra-hospital stillbirths and controls: OR 2.80, 95 % CI 1.59–4.95
f
Difference between intra-hospital stillbirths and controls: OR 2.67, 95 % CI 1.30–5.49
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 9 of 12

at the study site, and 58 % were admitted before active

labour.
As found in other East African studies [8, 11, 35], not
only was correct management often delayed, unnecessary
and even harmful management was sometimes initiated.
The presence of under- as well as over-treatment became
particularly apparent when reviewing cases with poor pro-
gress of labour, where CSs often were found unnecessary.
Yet, as only 4 % of foetuses had a documented positive
FHR when deciding to perform CS, it is doubtful how big
an impact e.g. possible delays in the decision-to-delivery
interval for CS had on causing stillbirths; the unnecessary
CSs may rather have caused avoidable maternal risks. In
contrast, the under-use of vacuum extraction may be seen
as an indicator of poor FHR monitoring, leading to undis-
closed foetal distress in the second stage of labour.
Fig. 3 Initiation of oxytocin for labour augmentation, according to Likewise, there appeared to be a dangerous over-use of
labour progress. The difference in overall use of oxytocin for labour oxytocin for labour augmentation. This is similar to
augmentation between intra-hospital stillbirths and controls was
studies from Bangladesh and Pakistan, where misuse of
significant with the stillbirth cases receiving the treatment more
often (OR 1.86, 95 % CI 1.06–3.27). Intra-hospital stillbirths: Documented oxytocin was associated with stillbirth and birth asphyxia
positive foetal heart rate on admission, birthweight ≥2000 g. Controls: [36, 37]. Notably, the Pakistani study draws attention to
Apgar score ≥7, birthweight ≥2000 g the danger of insufficiently trained healthcare workers
administrating this highly potent drug [36].
Similar to intrapartum care, antenatal visits did not ap-
their admission earlier in labour, and longer and more pear consistent with effective antenatal surveillance and
complicated labour duration [34]. In addition, increased treatment, and as in other studies this appeared to be a
vulnerability and need for quality care among all women central determinant of both pre- and intra-hospital still-
experiencing intra-hospital stillbirth is emphasized by births [33]. For instance, while nearly all women
the association with severe hypertension and prior loss attended antenatal care at least once, a severe hyperten-
of child. sion prevalence of 19 % among stillbirths suggests
While staff often referred to late referral of women as missed opportunities [38]. Furthermore, the fact that less
a major cause to adverse labour outcome, 85 % of women than half of the study population had attended four
experiencing intra-hospital stillbirth directly sought care antenatal care visits indicate a lost chance for continuity
in care [39].

Table 6 Seven target areas for improving intrapartum quality of Maternal risks
care at the study site Substandard quality of risk assessment on admission as well
1. Strengthened risk assessment on admission, with particular focus on foetal as poor intrapartum surveillance and decision-making were
heart rate, blood pressure, temperature, and previous obstetric history. associated with profound maternal risks and appeared to be
2. Improved routine surveillance during latent and active phase of labour, major determinants of the death of three women
regarding all key parameters (foetal heart rate, dilatation of cervix and (Additional file 2). Women with foetal death on ad-
descent, contractions, maternal vital signs, and urinary output). mission were the most neglected. While they were in
3. Increased prioritization of women with already diagnosed intrauterine particular high intrapartum risk due to the often
foetal death for routine assessments during labour.
underlying morbidity and further at increased postpartum
4. Timely prevention and management of prolonged labour, with focus risk of e.g. obstetric fistula, labour progress and vital signs
on alternative and less harmful interventions than oxytocin infusion
for labour augmentation (e.g. artificial rupture of membranes and were often undocumented throughout active labour
emptying of bladder), and more restrictive dosages and improved [40, 41].
surveillance when oxytocin is administered. CS is generally not indicated when there is foetal death
5. Reduction of caesarean sections after intrauterine foetal death, by [42]. However, a high proportion of CSs were done on
improved management of prolonged labour, and enforcement of doubtful indications, and many were related to insufficient
vacuum extraction and craniotomy use.
management of prolonged labour. Thus, 26 % CSs among
6. Improved management of severe hypertensive disorders, with
particular focus on antihypertensive treatment.
stillbirths is unacceptably high; in particular as the vast
majority had either foetal death diagnosed or did not have
7. Better intrapartum documentation as well as record keeping.
FHR recorded prior to surgery. Except for an even higher
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 10 of 12

rate found at three hospitals in Mozambique [11], this is based guidelines, it would be premature to conclude inef-
markedly higher than other studies from low- and middle- fectiveness of the WHO partograph. However, for the par-
income countries [33]. A high proportion of CSs were tograph to assist in surveillance and management, it must
done in the second stage of labour without an attempt of be coupled to a locally achievable and relevant labour
operative or destructive vaginal delivery. While short- and management protocol. Although often not prioritised in
long-term maternal risks of suboptimally treated pro- evaluations of partograph use, this has previously proven
longed labour and unnecessary CSs are widely established effective [14, 18]. For instance, when considering the low
[34, 43, 44], lack of transparency as to when to perform resources at the study site, it seems unrealistic to assure
CS is found in other African studies too [11, 45, 46]. close monitoring and titration of oxytocin augmentation if
Six women suffered from uterine rupture. When con- more than a few women are treated simultaneously [53].
sidering the low level of surveillance in 14 % of stillbirth This study identified 25 % underreporting of stillbirths
cases with one or more previous CSs, and the misuse of in the official hospital registers, and even though a sys-
oxytocin, many more appeared at risk of rupture. Fur- tematical surge was conducted through all piles of case
thermore, while foetal bradycardia is an early sign of files, a considerable number of files remained missing
impending rupture [47], substandard FHR assessments (Fig. 1). Initially, it was the intention also to include early
made it less useful in timely detection. neonatal deaths in the study. However, data collection
Of the 19 % with severe hypertensive disorders experi- revealed frequent default record keeping between the
encing stillbirth, more than half had severe pre-eclampsia obstetric and neonatal units as well as substantial under-
or eclampsia. A Nigerian study found a similar prevalence reporting of very early neonatal deaths in all registers,
and comparable insufficient antenatal and intrapartum which resulted in reluctance to include them. Further-
surveillance and treatment of these dangerous conditions more, missing documentation in medical records – or
[48]. Recent data from well-resource settings emphasize “blanks” – was a frequent finding, which is likely to have
suboptimally treated severe hypertension as an important affected patient care and labour outcomes. Incomplete
contributor to maternal mortality [38]; in our study, 67 % health information systems are notoriously linked with
of women with severe hypertension did not receive any poor health outcomes [3, 54]. It is warranted that the
antihypertensive treatment. underlying factors for these “blanks” in medical record-
ing are evaluated, and that quality of documentation and
Clinical implications record keeping as well as use of the data are improved.
At this East African referral hospital, facility births were
frequently not accompanied by skilled intrapartum at- Strengths and limitations
tendance. While widespread insufficiency in quality of The present pragmatic study was found suitable as a
routine and emergency labour care may partly be caused structured, simple, and low-cost method to identify
by massive structural constraints, suboptimal care often central challenges in intrapartum care at this real-world set-
resulted in more risk associated, complicated, and re- ting with limited information available. Classification in
source draining interventions. Some of the revealed defi- pre- and intra-hospital stillbirths was a useful, more achiev-
ciencies may be addressed even without high costs in able, and simple alternative to ‘fresh’ versus ‘macerated’
manpower and other resources, and the main risks and stillbirths, which, as in a study from Ghana, was found
determinants are crucial in effectively designing low cost unreliable [55]. Moreover, intra-hospital stillbirths may
interventions (Table 6). be seen as an even stronger indicator of intra-hospital
For many years, effectiveness of using the WHO quality of care than ‘fresh’ stillbirths. However, in 30 %
partograph has been questioned [49, 50]. Yet, when of pre-hospital stillbirths there was no FHR documen-
analysing quality of intrapartum care at low resource tation on or after admission, which may potentially hide
facilities, partograph use for timely surveillance and an even higher proportion of intra-hospital stillbirths.
decision-making appears central in ending preventable Selected audit criteria were unambiguously applicable.
complications [11, 14, 18, 35]. In 86 % of all cases where Yet, though intensive efforts were made for adapting
the partograph was applied, first cervical dilatation in ac- international evidence-based guidelines to reach local
tive phase of labour was plotted correctly on the alert line, reality, some criteria, such as <90 min. between FHR
and knowledge on accurate recording did not appear to recordings, might be too optimistic as a sensitive audit
be a major challenge. However, similar to other studies, standard for detecting quality improvements at this setting.
WHO’s recommendations for frequency of recordings A central limitation to the study is that a criterion-
were not followed and also did not seem achievable with based audit does not allow exploration of underlying de-
the resources available [35, 51, 52]. In the present study, terminants of substandard care, such as structural needs
even though the majority of intrapartum decision-making for supplies, space, and knowledge/skills among staff.
did not seem influenced by partograph use or evidence- Another limitation is the varying quality of data, which
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 11 of 12

might bias results; there may be a tendency of staff to in criticaly revising the paper draft. All authors have approved to the final
forget reporting given care or to under-report mismanage- version to be published and agree to be accountable for all aspects of the
work in ensuring that questions related to the accuracy or integrity of any
ment. Participant observations during the study period part of the work are appropriately investigated and resolved.
also identified the issues presented in the current paper,
and qualitative analysis opened up to a complex tangle of Competing interests
both structural and process related underlying challenges The authors declare that they have no competing interests.
influencing health providers’ ability to deliver acceptable Jos van Roosmalen is a Section Editor for BMC Infectious Diseases. Tarek
Meguid is an Associate Editor for BMC Infectious Diseases.
quality of care.
Consent for publication
Conclusion Not applicable.
Stillbirths are both a devastating burden of avoidable lost
lives in itself and a strong and easy to assess indicator of Ethics approval and consent to participate
quality of antenatal and intrapartum care. Substandard Ethical approval was obtained from the Zanzibar Medical and Research
Ethical Committee (ZAMREC/0001/JUNE/014) and Mnazi Mmoja Hospital, and
care led to substantial maternal and perinatal risks, which
the PartoMa project is registered with ClinicalTrials.org (NCT02318420, 4th
furthermore were related to resource draining interventions November 2014). In this baseline study of case files, all patient identities
that were not always necessary. Furthermore, 25 % underre- were anonymized and assigned a research number. Individually obtained
informed consent was therefore not required.
porting of stillbirths in hospital registers indicates a poor
health information system. These findings are largely in line Author details
1
with other reports from sub-Saharan Africa, and improve- Global Health Section, Department of Public Health, University of
Copenhagen, Øster Farimagsgade 5, Building 9, 1353 Copenhagen K,
ment of intra-hospital obstetric knowledge, care, and docu-
Denmark. 2Mnazi Mmoja Hospital, Zanzibar, Tanzania. 3Julius Center for
mentation is central to end preventable birth-related deaths Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht,
and disabilities. Considering referral hospitals’ major The Netherlands. 4School of Health & Medical Sciences, State University of
Zanzibar, P.O.Box:146, Zanzibar, Tanzania. 5Department of Obstetrics,
teaching tasks for future health providers, it is war-
Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, 2100
ranted to address the tertiary level in order to achieve Copenhagen Ø, Denmark. 6Athena Institute, VU University of Amsterdam, De
quality improvement of the entire health care sector. Boelelaan 1105, 1081 HV Amsterdam, The Netherlands.

Received: 28 June 2016 Accepted: 1 November 2016

Additional files

Additional file 1: Very preterm stillbirths. Criterion-based audit data on References

very preterm stillbirths (birthweight <2000 g). (DOCX 49 kb) 1. Alkema L, Chou D, Hogan D, Zhang S, Moller A-B, Gemmill A, et al. Global,
Additional file 2: Maternal deaths. Intra-hospital management preceding regional, and national levels and trends in maternal mortality between 1990
the three maternal deaths within the study population. (DOCX 39 kb) and 2015, with scenario-based projections to 2030: a systematic analysis by
the UN Maternal Mortality Estimation Inter-Agency Group. Lancet. 2015;387:
462–74.
Abbreviations 2. UNICEF, WHO, The World Bank, United Nations. Levels & Trends in Child
CI: Confidence interval; CS: Caesarean section; FHR: Foetal heart rate; g: Grams; Mortality: Report 2015. New York: World Health Organization; 2015.
IQR: Interquartile range; min.: Minutes; OR: Odds ratio; WHO: World Health 3. Lawn JE, Blencowe H, Waiswa P, Amouzou A, Mathers C, Hogan D, et al.
Organization Stillbirths: rates, risk factors, and acceleration towards 2030. Lancet. 2016;
387:587–603.
Acknowledgements 4. Lawn JE, Blencowe H, Oza S, You D, Lee ACC, Waiswa P, et al. Every
The research team would like to thank the many health providers at the newborn: progress, priorities, and potential beyond survival. Lancet. 2014;
study site who have kindly assisted in collecting the data here presented. 384:189–205.
The research team is also grateful for the fellow research assistants Mbweni 5. Requejo J, Bryce J, Victora C. Building a Future for Women and Children:
Makame Ali and Said Mzee Said’s high quality work. Finally, many thanks to The 2012 Report. Geneva: WHO; 2012.
Aksel Karl Georg Jensen, Section of Biostatistics, University of Copenhagen, 6. UNICEF, WHO, The World Bank, United Nations. Levels & Trends in Child
for statistical consultation and review. Mortality: Report 2015. New York: USAID; 2015.
7. Puchalski Ritchie LM, Khan S, Moore JE, Timmings C, van Lettow M, Vogel
Funding JP, et al. Low- and middle-income countries face many common barriers to
The present study was funded by grants from the Lundbeck, Laerdal, and implementation of maternal health evidence products. J. Clin. Epidemiol.
Augustinus foundations. 2016; In Press.
8. Oladapo O, Adetoro O, Ekele B, Chama C, Etuk S, Aboyeji A, et al. When
Availability of data and materials getting there is not enough: a nationwide cross-sectional study of 998
The datasets supporting the conclusions of this article are included within maternal deaths and 1451 near-misses in public tertiary hospitals in a low-
the article and its additional files. income country. BJOG. 2015;123:928–38.
9. Kidanto HL, Mogren I, van Roosmalen J, Thomas AN, Massawe SN, Nystrom
Authors’ contributions L, et al. Introduction of a qualitative perinatal audit at Muhimbili National
NM formulated the study design, managed data acquisition, analysed and Hospital, Dar es Salaam, Tanzania. BMC Pregnancy Childbirth. 2009;9:45.
interpreted data, and drafted the paper. NH participated substantialy in 10. Nyamtema AS, Urassa DP, Massawe S, Massawe A, Lindmark G, van
acquisition, analysis, and interpretation of data, and in criticaly revising the Roosmalen J. Partogram use in the Dar es Salaam perinatal care study. Int J
paper draft. ICB, JvR, TM, and BBN contributed substantially to the study Gynaecol Obstet. 2008;100:37–40.
design, analysis and interpretation of data, and they critically revised the 11. Geelhoed D, Stokx J, Mariano X, Mosse Lázaro C, Roelens K. Risk factors for
paper draft. RSK and AGM participated substantialy in acquisition of data and stillbirths in Tete, Mozambique. Int J Gynaecol Obstet. 2015;130:148–52.
Maaløe et al. BMC Pregnancy and Childbirth (2016) 16:351 Page 12 of 12

12. University of Copenhagen. The PartoMa Project for saving lives at birth with poor drug regulatory control: characteristics and risk patterns. Int
[Internet]. Available from: http://publichealth.ku.dk/sections/global/project/ Health. 2015;8:89–95.
partoma/ 37. Begum B, Zaman RU, Afza NS. Understanding the magnitude of still birth in
13. Aminu M, Unkels R, Mdegela M, Utz B, Adaji S, van den Broek N. Causes of Mymensingh Medical College Hospital. Mymensingh Med J. 2010;19:340–2.
and factors associated with stillbirth in low- and middle-income countries: 38. Von Dadelszen P, Firoz T, Donnay F, Gordon R, Hofmeyr GJ, Lalani S, et al.
a systematic literature review. BJOG. 2014;121(Suppl):141–53. Preeclampsia in low and middle income countries-health services lessons
14. Van Roosmalen J. Perinatal mortality in rural Tanzania. BJOG. 1989;96:827–34. learned from the PRE-EMPT (PRE-Eclampsia-Eclampsia Monitoring, Prevention
15. WHO, MoH Z. Millenium Development Goals Progress Tracking Report - and Treatment) project. J Obstet Gynaecol Canada. 2012;34:917–26.
Zanzibar. Stone Town: Ministry of Health and Social Welfare; 2007. 39. Afnan-Holmes H, Magoma M, John T, Levira F, Msemo G, Armstrong CE, et
16. HMIS Z. Zanzibar Annual Health Bulletin 2011. Stone Town: Ministry of al. Tanzania’s Countdown to 2015: an analysis of two decades of progress
Health and Social Welfare; 2012. and gaps for reproductive, maternal, newborn, and child health, to inform
17. National Bureau of Statistics. Tanzania Demographic and Health Survey. Dar es priorities for post-2015. Lancet Glob Heal. 2015;3:e396–409.
Salaam: NBS and ICF Macro; 2010. 40. Tebeu PM, Fomulu JN, Khaddaj S, de Bernis L, Delvaux T, Rochat CH. Risk
18. Kwast BE, Lennox CE, Farley TMM. World Health Organization partograph in factors for obstetric fistula: a clinical review. Int Urogynecol J. 2012;23:387–94.
management of labour. World Health Organization Maternal Health and 41. Cowgill KD, Bishop J, Norgaard AK, Rubens CE, Gravett MG. Obstetric fistula
Safe Motherhood Programme. Lancet. 1994;343:1399–404. in low-resource countries: an under-valued and under-studied problem–
19. Drost E, van Lonkhuijzen LRCW, Meguid T, Landis Lewis D, Zeeman GG. systematic review of its incidence, prevalence, and association with stillbirth.
Implementing safe motherhood: a low-cost intervention to improve the BMC Pregnancy Childbirth. 2015;15:193.
management of eclampsia in a referral hospital in Malawi. BJOG. 2010;117: 42. Stanton C, Ronsmans C. Recommendations for routine reporting on indications
1553–7. for cesarean delivery in developing countries. Birth. 2008;35:204–11.
20. World Health Organization. International Classification of Diseases 10th 43. Villar J, Carroli G, Zavaleta N, Donner A, Wojdyla D, Faundes A, et al.
revision (ICD-10) [Internet]. 2010. Available from: http://apps.who.int/ Maternal and neonatal individual risks and benefits associated with
classifications/icd10/browse/2016/en. Accessed 27 Jan 2016 caesarean delivery: multicentre prospective study. BMJ. 2007;335:1025.
21. RCOG. Green-top Guideline No. 1b: Tocolysis for Women in Preterm Labour. 44. O’Neill SM, Kearney PM, Kenny LC, Khashan AS, Henriksen TB, Lutomski JE,
London: RCOG; 2011. et al. Caesarean delivery and subsequent stillbirth or miscarriage: systematic
22. RCOG. Green-top Guideline No. 7: Antenatal Corticosteroids to Reduce review and meta-analysis. PLoS One. 2013;8, e54588.
Neonatal Morbidity and Mortality. London: RCOG; 2010. 45. Maaløe N, Bygbjerg IC, Onesmo R, Secher NJ, Sorensen BL. Disclosing
23. O’Driscoll K, Stronge JM, Minogue M. Active management of labour. Br Med doubtful indications for emergency cesarean sections in rural hospitals in
J. 1973;3:135–7. Tanzania: a retrospective criterion-based audit. Acta Obstet Gynecol Scand.
24. World Health Organization. Managing Complications in Pregnancy and 2012;91:1069–76.
Childbirth: A Guide for Midwives and Doctors. Geneva: WHO, Integrated 46. Litorp H, Kidanto HL, Nystrom L, Darj E, Essén B. Increasing caesarean
Management of Pregnancy And Childbirth; 2007. section rates among low-risk groups: a panel study classifying deliveries
25. National Collaborating Centre for Women’s and Children’s Health. NICE according to Robson at a university hospital in Tanzania. BMC Pregnancy
Guideline: Intrapartum care for healthy women and babies. London: NICE; Childbirth. 2013;13:107.
2014. 47. Ridgeway JJ, Weyrich DL, Benedetti TJ. Fetal heart rate changes associated
26. National Collaborating Centre for Women’s and Children's Health. NICE with uterine rupture. Obstet Gynecol. 2004;103:506–12.
Guideline: Caesarean section. Manchester: NICE; 2011. 48. Guerrier G, Oluyide B, Keramarou M, Grais R. High maternal and neonatal
27. National Collaborating Centre for Women’s and Children's Health. NICE Guideline: mortality rates in northern Nigeria: an 8-month observational study. Int J
Hypertension in pregnancy: diagnosis and management. London: NICE; 2011. Womens Health. 2013;5:495–9.
49. Souza JP, Oladapo OT, Bohren MA, Mugerwa K, Fawole B, Moscovici L, et al.
28. Lewis D, Downe S, FIGO Intrapartum Fetal Monitoring Expert Consensus
The development of a Simplified, Effective, Labour Monitoring-to-Action
Panel. FIGO consensus guidelines on intrapartum fetal monitoring:
(SELMA) tool for Better Outcomes in Labour Difficulty (BOLD): study
Intermittent auscultation. Int J Gynaecol Obstet. 2015;131:9–12.
protocol. Reprod Health. 2015;12:49.
29. Ayres-de-Campos D, Spong CY, Chandraharan E, FIGO Intrapartum Fetal
50. Lavender T, Hart A, Smyth RMD. Effect of partogram use on outcomes for
Monitoring Expert Consensus Panel. FIGO consensus guidelines on
women in spontaneous labour at term. Cochrane Database Syst Rev. 2013;7,
intrapartum fetal monitoring: Cardiotocography. Int J Gynaecol Obstet.
CD005461.
2015;131:13–24.
51. Braddick L, Tuckey V, Abbas Z, Lissauer D, Ismail K, Manaseki-Holland S, et al.
30. Baxley E, Deutchman M, Atwood L, Murphy N, Yu J. ALSO Advanced Life
A mixed-methods study of barriers and facilitators to the implementation of
Support in Obstetrics - Course Syllabus. 5th ed. Leawood: American
postpartum hemorrhage guidelines in Uganda. Int J Gynaecol Obstet. 2016;
Academy of Family Physicians; 2014.
132:89–93.
31. Ameh CA, Ekechi CI, Tukur J. Monitoring severe pre-eclampsia and
52. Gans-Lartey F, O’Brien BA, Gyekye FO, Schopflocher D. The relationship
eclampsia treatment in resource poor countries: skilled birth attendant
between the use of the partograph and birth outcomes at Korle-Bu
perception of a new treatment and monitoring chart (LIVKAN chart). Matern
teaching hospital. Midwifery. 2013;29:461–7.
Child Health J. 2012;16:941–6.
53. Jonsson M. Use and Misuse of Oxytocin During Delivery. Uppsala:
32. Goldenberg RL, McClure EM, Kodkany B, Wembodinga G, Pasha O, Esamai F,
Uppsala University; 2009.
et al. A multi-country study of the “intrapartum stillbirth and early neonatal
54. Kerber KJ, Mathai M, Lewis G, Flenady V, Erwich JJHM, Segun T, et al.
death indicator” in hospitals in low-resource settings. Int J Gynecol Obstet.
Counting every stillbirth and neonatal death through mortality audit to
2013;122:230–3.
improve quality of care for every pregnant woman and her baby. BMC
33. McClure EM, Saleem S, Goudar SS, Moore JL, Garces A, Esamai F, et al.
Pregnancy Childbirth. 2015;15 Suppl 2:S9.
Stillbirth rates in low-middle income countries 2010–2013: a population-
55. Gold KJ, Abdul-Mumin A-RS, Boggs ME, Opare-Addo HS, Lieberman RW.
based, multi-country study from the Global Network. Reprod Health. 2015;
Assessment of “fresh” versus “macerated” as accurate markers of time since
12 Suppl 2:S7.
intrauterine fetal demise in low-income countries. Int J Gynaecol Obstet.
34. Harrison MS, Ali S, Pasha O, Saleem S, Althabe F, Berrueta M, et al. A 2014;125:223–7.
prospective population-based study of maternal, fetal, and neonatal outcomes
in the setting of prolonged labor, obstructed labor and failure to progress in
low- and middle-income countries. Reprod Health. 2015;12 Suppl 2:S9.
35. Maaløe N, Sorensen BL, Onesmo R, Secher NJ, Bygbjerg IC. Prolonged
labour as indication for emergency caesarean section: a quality assurance
analysis by criterion-based audit at two Tanzanian rural hospitals. BJOG.
2012;119(5):605–13.
36. Shah S, Van den Bergh R, Prinsloo JR, Rehman G, Bibi A, Shaeen N, et al.
Unregulated usage of labour-inducing medication in a region of Pakistan