Maternal near miss and quality of maternal health care in Baghdad, Iraq.

0.2 MB·English

by  Jabir, Maysoon

Jabiretal.BMCPregnancyandChildbirth2013,13:11
http://www.biomedcentral.com/1471-2393/13/11
RESEARCH ARTICLE Open Access
Maternal near miss and quality of maternal health
care in Baghdad, Iraq
Maysoon Jabir1, Imad Abdul-Salam2, Dhikra M Suheil3, Wafa Al-Hilli4, Sana Abul-Hassan5, Amal Al-Zuheiri6,
Rasha Al-Ba'aj7, Abeer Dekan1, Özge Tunçalp8* and Joao Paulo Souza9
Abstract
Background: The maternal near-miss concept has been developed as aninstrumentfor assisting health systemsto
evaluate and improve their quality of care. Our study aimed at studying the characteristicsand quality of care
provided to women withseverecomplicationsin Baghdad through theuse of theWorld Health Organization
(WHO) near-miss approach for maternal health.
Methods: This is a facility-based, cross-sectional study conducted in6 public hospitals inBaghdad between March
1, 2010 and the June 30,2010. WHO near-miss approach was utilized to analyze the data interms of indicators of
maternal near miss and accessto and quality of maternal care.
Results: The maternal near-missrate was low at5.06 per 1,000 live births, whiletheoverall maternal near miss:
mortality ratio was 9:1.One thirdof the near-miss cases were referredfrom other facilities and themortality index
was the same for referred women and for in-hospital women (11%). The intensive care unit (ICU) admission rate
was 37% for women with severe maternal outcomes(SMO), while theoverall admission rate was 0.28%. Anemia
(55%)and previous cesarean section (45%)were the mostcommon associated conditions with severe maternal
morbidity. The use of magnesium sulfate for treatment of eclampsia, oxytocin for prevention and treatment of
postpartum hemorrhage, prophylactic antibiotics during caesarean section, and corticosteroids for inducing fetal
lung maturation inpreterm birth is suboptimum.
Conclusions: The WHO near-miss approach allowed systematic identification ofthe roadblocks to improve quality
ofcare and then monitoring the progress. Criticalevidence-based practices,relevant to the management of
women experiencinglife-threatening conditions, are underused. In addition, possible limitations in thereferral
system result in a very high proportion ofwomen presenting atthe hospital already ina severehealth condition(i.
e. withorgan dysfunction).A shortage of ICU beds leading to women taken care ofwithout admissionto ICU may
also contribute to a highproportion of maternal deaths and organ dysfunction.
Keywords: Maternal morbidity, Obstetric complications,WHO near-miss approach, Quality improvement,
Developing countries, Baghdad
Background towards achievement of MDGs has been challenged by
In2000,theleadersofallUnitedNationsMemberStates two major wars in the last two decades. On top of the
agreed that policies conducive to development and to unbearably high human cost, these wars have produced
theeliminationofextremepoverty wouldbeputinplace an important damage to the country’s infrastructure and
in global scale. A set of goals has been established and affected variouscomponentsofthehealth systems.
manycountrieshavedonea substantialprogresstowards Social determinants and the health system perform-
those goals, which became known as the Millennium ance play a major role in the occurrence of maternal
Development Goals (MDGs) [1]. In Iraq, progress deaths.OneoftheMDGs,reductionofmaternalmortal-
ity is a robust indicator of development. In this context,
*Correspondence:[email protected]
even with the two wars, since 1990 there is a trend to-
8DepartmentofPopulation,FamilyandReproductiveHealth,JohnsHopkins
BloombergSchoolofPublicHealth,615NWolfeSt,Baltimore,MD21205,USA wards reduction of maternal mortality in Iraq. The
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Jabiretal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page2of9
http://www.biomedcentral.com/1471-2393/13/11
World Health Organization estimates that in 1990, the maternal mortality and morbidity classifications devel-
maternal mortality in Iraq was around 93 deaths per oped a consensus on maternal near-miss identification
100,000 live births, while in 2008, it was estimated to be (Additional file 1: Table S1), which is based on two
around 75 deaths per 100,000 live births [2]. Recent esti- components[8]:
mates are compatible with this trend [3], which is most
likely due to the resilience of Iraqi people in keeping 1.Identificationofpotentiallylife-threatening
basic infrastructure and services functional under ad- conditions,which may ormay notbenear-miss cases
verse conditions and, possibly, the substantial efforts for (i.e. specificcomplicationssuch assevere
reconstruction haveproducedsome visible effect. preeclampsiaand/orcriticalinterventions suchas
Confidential enquiries and similar clinical audits of ma- blood transfusion).
ternal deaths have been used to evaluate and improve the 2.Identificationofnear-miss casesbasedonorgan
quality of maternal health care in many countries. How- system dysfunction and organ-dysfunctionproxies
ever,asmaternaldeathsbecomelessfrequentorarefound including clinical,laboratoryand management
to be relatively rare in individual health facilities, the as- criteria.
sessmentofqualityofcareperformedexclusivelybasedon
maternal deaths becomes more challenging. The concept Our study aims at assessing the prevalence and evalu-
of maternal near miss has been evolving during the past ating the management of severe maternal morbidity
two decades as number of women dying from complica- using the WHO near-miss approach in Baghdad, Iraq
tionsofpregnancyandchildbirthisprogressivelydecreas- [9]. This is the first multi-center study conducted in the
ing in many countries, while the number of those with city of Baghdad to assess the near-miss cases in obstetric
life-threatening complications who are treated and dis- practiceinIraq.
charged home exceeds the number of those who die.
Therefore,studyingthosewomenwhonearlydiedbutsur- Methods
vived,identifiedasthenear-misscases,wouldgiveabetter This is a facility-based, cross-sectional study conducted
indicationofcareprovidedforthosewomenwhosurvived in six public hospitals in Baghdad between March 1,
thenear-missevent[4]. 2010 and the June 30, 2010. It consists of a near-miss
Overall, there had been three major approaches to the criterion-based clinical audit implemented according the
identificationofnearmisscases:1)Clinicalcriteriarelated WHO near-miss approachfor maternalhealth[6]. Selec-
toa specificdiseaseentity(i.e.,pre-eclampsia,postpartum tion criteria for the facilities included were 1) to be a
hemorrhage), 2) Management-based criteria (i.e., admis- public hospital and 2) to have more than one thousand
sion to ICU, need for a blood transfusion), or 3) Organ deliveriesperyearaccordingtotheinformationprovided
system dysfunction based criteria [5,6]. Depending on by the health and biostatistics department in the Iraqi
these differentapproaches, prevalenceof nearmissvaries. Ministry of Health. We included all of the hospitals fit-
According to a recent systematic review, prevalence rates tingthese criteria inBaghdad:atotalofsix.Thesepublic
of near miss varied between 0.6 and 14.98% for disease- hospitals were distributed all over Baghdad city to serve
specificcriteria,between0.04and4.54%formanagement- a population of approximately seven millions. Three of
based criteria and between 0.14 and 0.92% for organ- these hospitals were general hospitals with obstetric
baseddysfunctionbasedonMantelcriteria[7].Womenin units and the rest were major maternity hospitals. Four
resource-poor settings experience a higher prevalence in out of six hospitals had intensive care units (ICUs),
all these categories. However, due to wide variation in although the remaining two hospitals had a "close
identification of cases as well as the variation within each observation unit" to monitor and treat women with
category, it has not been possible to pool the data and post-operative and post-delivery complications, run by
make a summary estimate [4,7]. Currently, a maternal specialized obstetricians. These hospitals receive refer-
near-misscaseisdefinedas"awomanwhonearlydiedbut rals from midwives, health centers, private hospitals as
survived a complication that occurred during pregnancy, well as unbooked patients (patients self-referring them-
childbirth or within 42 days of termination of pregnancy” selves to the hospitals). No fees are paid by patients for
[8]. In order to overcome these challenges, the maternal theservicesprovided.
near-miss concept has been developed as an instrument A study coordinator among the hospital coordinators
for assisting health systems to evaluate and improve their was designated in each of these six facilities and the
qualityofcare.Inrecentyears,WHOhasdevelopedatool overall study coordination was performed by the Centre
forevaluatingthequalityofmaternalhealthcarebasedon ofTrainingandHumanDevelopmentintheIraqiMinis-
thenear-missconcept[9]. try of Health. These six selected study coordinators were
In an attempt to standardize the identification of ma- trained on a two-day course in January 2010 on identify-
ternal near-miss events, the WHO working group for ing severe life-threatening conditions, maternal near-
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page3of9
http://www.biomedcentral.com/1471-2393/13/11
miss events and deaths, and how to implement data system. This online data management system was based
collection. on the Google platform and data entered in the online
TheWHOconsensusonmaternal near-missdefinition form were stored in an online spreadsheet, which incor-
was used to define the cases [8]. Definitions and abbre- porated a comprehensive set of consistency rules to pro-
viations used in this paper are found in Table 1. Data vide concurrent data quality check. The inconsistencies
were collected on a daily basis by the coordinators using that were identified generated queries to the study coor-
hospital records or staff interviews and the forms were dinators. The standard data tables based on the WHO
filled while the women were still in the hospital. Cases near-miss approach was automatically generated using
were defined according to potentially life-threatening Microsoft Excel as the data were entered. The online
conditions including severe postpartum hemorrhage, data entrysystemwaspasswordprotected.
severe preeclampsia, eclampsia, sepsis and ruptured The study was approved by the ethics committee of
uterus, whereas organ or system failure depending on the local supervising committee of the Arab Board for
certain clinical criteria, laboratory markers and manage- Health Specializations (ABHS). All data were obtained
ment proxies were used to identify the near-miss cases from medical records and did not identify participants,
amongpotentiallylife-threateningconditions(Additional therefore each site was granted a waiver of individual
file 1: Table S1). The maternal outcome, gestational age informedconsent.
and neonatal outcome were also collected during the
hospital stay or by the 7th day postpartum, whichever Results
came first. During the four-month data collection period in the six
Using the data collected, various indicators have been study facilities, overall there were 25,472 live births, 212
calculated. In line with maternal mortality ratio, mater- women with potentially life threatening conditions and
nal near-miss ratio was the number of near-miss cases 145 severe maternal outcomes (129 near-miss cases and
per 1000 live births. Moreover, maternal near-miss mor- 16 maternal deaths). Our results will be presented as
tality ratio, which is the ratio between maternal near- suggested in the WHO handbook on the near-miss ap-
miss cases and maternal deaths, was calculated. For this proach formaternalhealth[9].
indicator, higher ratios indicate better care, meaning Cases were identified according to underlying causes of
more women survived as a near miss rather than be- morbidityandorgansystemdysfunctioninTables2and3,
coming maternal deaths. Also, mortality index was cal- respectively. Most common organ dysfunctions reported
culated, where the number of maternal deaths was among near-miss cases were cardiovascular dysfunction
divided by the number of women with life-threatening followed by uterine dysfunction leading to hysterectomy,
conditions (maternal near miss and maternal deaths) 55.8 and 53.5, respectively (Table 2). It should be noted
and was expressed as a percentage. Higher indices indi- that almost 50% of the women with maternal near-miss
cate that more women with life-threatening conditions morbidityhad multiple organdysfunction. Mortalityindex
die (low quality of care), whereas lower indices signify washighestforrenaldysfunction,40%.Maternalnear-miss
better quality ofcare. mortality ratio was highest for uterine dysfunction (11:1),
Access to hospital and intra-hospital care were assessed followed by cardiovascular and coagulation/ hematologic
by the proportion of near-miss cases and maternal deaths disorders (10:1). Of all the women with potentially life-
presenting within 12 hours of hospital stay versus after threateningconditions,181women(85.4%)underwentthe
12hours,thelatter indicatingthequalityofcareprovided following critical interventions: 118 (55.7%) women used
within the hospital. Intensive care unit (ICU) use among bloodproducts,78(36.8%)womenhadlaparotomyand75
ourstudypopulationwascollectedaswell. (34.4%)womenwereadmittedtoICU.
We also collected data on the coverage of selected Underlying causes were similar among potentially life-
evidence-based interventions used for prevention and threatening conditions, near-miss cases and mater-
treatment of the main causes of maternal deaths. This nal deaths, most common being severe postpartum
was part of a criterion-based clinical audit approach hemorrhagefollowedbyhypertensivedisorders(Table3).
used to assess the quality of care and included interven- Associated conditions have also been assessed and the
tions related tothe prevention and treatment ofpostpar- most common reported conditions were anemia and
tum hemorrhage, severe preeclampsia and eclampsia, previous C-section, 55% and 44.5% among near-miss
useofantibioticsforinfectionprophylaxisduringcaesar- women, respectively. Despite the small numbers, the
ean section (C-section) and treatment of sepsis. In highest maternal near-miss mortality ratio among these
addition, we measured the use ofcorticosteroids for fetal underlying causes was observed among women with
lungmaturity. pregnancy-relatedinfections(25%).
Data were sent monthly to the country coordinator Majority of women with potentially life-threatening
and subsequently entered into an online data entry conditions, near-miss cases and maternal deaths had
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page4of9
http://www.biomedcentral.com/1471-2393/13/11
Table1Maternalnear-missterminologyandindicators[8,9]
MaternalNearMiss(MNM) Awomanwhonearlydiedbutsurvivedacomplicationthatoccurredduringpregnancy,
childbirthorwithin42daysofterminationofpregnancy.
MaternalDeath(MD) Deathofawomanwhilepregnantorwithin42daysofterminationofpregnancyorits
management,butnotfromaccidentalorincidentalcauses.
LiveBirth(LB) Thebirthofanoffspring,whichbreathesorshowsevidenceoflife.
Severematernaloutcome(SMO) Alife-threateningcondition(i.e.organdysfunction),includingallmaternaldeathsandmaternal
near-misscases.
Womenwithlife-threatening Allwomenwhoeitherqualifiedashavingmaternalnearmissorwhodied.Itisthesumof
conditions(WLTC) maternalnearmissandmaternaldeaths.
MaternalNearMissRatio(MNMR) Thenumberofmaternalnearmisscasesper1,000livebirths.
SevereMaternalOutcomeRatio(SMOR) Thenumberofwomenwithlifethreateningconditionsper1,000livebirths.Thisindication
givesanestimationoftheamountofcareandresourcesthatwouldbeneededinan
areaorfacility.
MaternalNearMissMortalityRatio: Theratiobetweenmaternalnear-misscasesandmaternaldeaths.Higherratiosindicate
bettercare.
MortalityIndex Thenumberofmaternaldeathsdividedbythenumberofwomenwithlifethreateningconditions,
expressedasapercentage.Thehighertheindexthemorewomenwithlife-threateningconditions
die(lowqualityofcare),whereasthelowertheindexthefewerwomenwithlife-threatening
conditionsdie(betterqualityofcare).
Perinataloutcomeindicators (e.g.perinatalmortality,neonatalmortalityorstillbirthrates)inthecontextofmaternalmiss
couldbeusefultocomplementthequalityofcareevaluation.
HospitalAccessIndicators:
Thefollowingindicatorsareusedtoexplore (cid:129)SMO12:Casespresentingtheorgandysfunctionormaternaldeathwithin12hoursof
theaccesstothefacilityintermsof hospitalstay
functioningreferralsystems.
(cid:129)ProportionofSMO12casesamongallSMOcases
(cid:129)ProportionofSMO12casescomingfromotherfacilities
(cid:129)SMO12mortalityindex:ThenumberofSMO12casesdividedbythenumberofallSMOcases
expressedasapercentage.
Intra-hospitalCare:
Thefollowingindicatorsareusedtoexplore (cid:129)Intra-hospitalSMO:Casespresentingtheorgandysfunctionormaternaldeathafter12hours
accesstoqualitycareinthefacility: ofhospitalstay.
(cid:129)Intra-hospitalSMOrate(per1000livebirths):Thenumberofintra-hospitalSMOcasesper1000
livebirths.
(cid:129)Intra-hospitalmortalityindex:Thenumberofintra-hospitalSMOcasesdividedbythenumberof
allSMOcasesexpressedasapercentage.
(cid:129)ICUadmissionrate:ThenumberofwomenadmittedtoICUamongtotalnumberof
womengivingbirth.
(cid:129)ICUadmissionrateamongwomenwithSMO:ThenumberofwomenwithSMOdividedby
theICUadmissionsamongtotalnumberofwomengivingbirth.
ProcessIndicators:
Thefollowingindicatorsareusedtoassess (cid:129)Preventionofpostpartumhemorrhage:Thenumberofwomenwhoreceivedasingledoseof
thecoverageofselectedevidence-based oxytocindividedbythenumberofallwomengivingbirth(vaginaldelivery+cesareansection)
interventionsusedforpreventionand
(cid:129)Treatmentofseverepostpartumhemorrhage:ThenumberofwomenwithseverePPH
treatmentofthemaincausesof
whoreceivedtherapeuticoxytocindividedbythenumberofallwomenwith
maternaldeaths.
postpartumhemorrhage.
(cid:129)Eclampsia:Thenumberofwomenwitheclampsiawhoreceivedmagnesiumsulfatedivided
bythenumberofallwomenwitheclampsia.
(cid:129)Preventionofseveresystemicinfections/sepsis:Thenumberofwomenhavingacesarean
sectionandreceivingprophylacticantibioticsdividedbythenumberofallwomenhaving
cesareansections.
(cid:129)Treatmentofsevereinfectionsandsepsis:Thenumberofwomenwithseveresystemicinfections
orsepsiswhoreceivedIVantibioticsdividedbythenumberofallwomenwithseveresystemic
infectionsorsepsis.
(cid:129)Fetallungmaturation:Thenumberofwomenhavingalivebirthafter3hoursofhospital
stayandreceivingcorticosteroidsforfetallungmaturationdividedbyallwomenhavinga
livebirthafter3hoursofhospitalstay.
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page5of9
http://www.biomedcentral.com/1471-2393/13/11
Table2Morbidityconditionsleadingtoinclusionina outcomes and near-miss indicators can be found in
sampleofwomenwithpotentiallylife-threatening Table 5. The maternal near-miss ratio (MNMR) and se-
conditionsandseverematernaloutcomes vere maternaloutcomeratio (SMOR)were 5.06 and5.69
n % per1,000livebirths,respectively(Table5).
Womenwithpotentiallylife-threateningconditions 212 100.00%
Womenwithseverecomplications 174 82.08% AccesstoHospitalandCareIndicators
Of the 145 cases with severe maternal outcomes (near
Severepostpartumhemorrhage 84 39.62%
miss and maternal deaths), 127 cases (87.6%) presented
SeverePreeclampsia 4 1.89%
with the organ dysfunction or maternal death within the
Eclampsia 43 20.28% first 12 hours of hospital admission and 34% of these
Sepsisorseveresystemicinfection 10 4.72% cases were referred from other facilities (Table 5). The
Ruptureduterus 29 13.68% mortality indicesfor thefirst12hoursof hospitaladmis-
sion and after 12 hours (intra-hospital) are 11.02% and
Othercomplicationsassociatedwithseverematernal 22 10.38%
outcome 11.11%, respectively.
WomenundergoingcriticalInterventions 181 85.38% Among all the women giving birth at our study facil-
ities, the overall ICU admission rate was 0.28%, whereas
Useofbloodproducts 118 55.66%
ICU admission rate among women with severe maternal
Interventionalradiology(uterinearteryembolization) 2 0.94%
outcomes was 37.2% and proportion of maternal deaths
Laparotomy 85 40.09%
occurredwith ICU admission was50%(Table 6).
AdmissiontoIntensiveCareUnit 73 34.43%
Organdysfunctioninmaternalnear-misscases 129 100.00% ProcessIndicators
Cardiovasculardysfunction 72 55.81% Amongwomenwithseverematernaloutcomes,oxytocin
was used in 86% of the women for the prevention of
Respiratorydysfunction 30 23.26%
postpartum hemorrhage, whereas only 68% of the
Renaldysfunction 6 4.65%
women with severe PPH received oxytocin as a treat-
Coagulation/hematologicdysfunction 30 23.26%
ment agent among other treatment regimens. In 93% of
Hepaticdysfunction 5 3.88% the eclampsia cases an anticonvulsant was used and
Neurologicdysfunction 23 17.83% MgSO4 was used in 67% of these cases. Antibiotics were
Uterinedysfunction/hysterectomy 69 53.49% used for all women with an established infection. In case
of caesarean section, 61% of women received prophylac-
Multipleorgandysfunction 64 49.61%
tic antibiotics intra-operatively and the rest 39% were
Organdysfunctioninmaternaldeaths 16 100.00%
given antibiotics post-operatively. Corticosteroids for
Cardiovasculardysfunction 7 43.75%
fetal lung maturity were used in only 55% of pregnant
Respiratorydysfunction 8 50.00% women eligible forthis type oftreatment(Table7).
Renaldysfunction 4 25.00%
Coagulation/hematologicdysfunction 3 18.75% Discussion
ThisstudyshowsthatinurbanIraq,theprevalenceofma-
Hepaticdysfunction 0 0.00%
ternaldeathsandnear-misscasesisrelativelylow.Despite
Neurologicdysfunction 4 25.00%
that, this study also highlights some opportunities to im-
Uterinedysfunction/hysterectomy 6 37.50%
provecare,bothatthefacilitylevelandattheorganization
Unspecifiedorgandysfunction 1 6.25% of care / health system level. Some evidence-based prac-
Multipleorgandysfunction 9 56.25% tices,relevanttothemanagementofwomenexperiencing
life-threatening conditions, are underused. In addition,
C-sections, 51%, 61% and 69%, respectively. Neonatal possible limitations in the referral system result in a very
outcomes (preterm births, stillbirths and perinatal high proportion of women presenting at the hospital
deaths) were worst among maternal deaths and compar- already in a severe health condition (i.e. with organ dys-
able between women with potentially life-threatening function). A shortage of ICU beds may also contribute to
conditions and maternal near misses (Table 4). Overall, a high proportion of maternal deaths and organ dysfunc-
there were no women with complete abortion reported tiontakingcarewithoutadmissiontoICU.
and ectopic pregnancy was highest among women with Comparing the major causes of near-miss cases and
potentially life-threatening conditions (18.4%) followed maternal deaths, obstetric hemorrhage and hypertension
bymaternal deaths (6.3%). were the most common underlying causes of severe ma-
The total maternal mortality ratio for the hospitals ternal outcomes, whichis comparable to otherstudies in
was 62.8 per 100,000 live births. Severe maternal developing countries [10-12]. In our study, anemia is
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page6of9
http://www.biomedcentral.com/1471-2393/13/11
Table3Underlyingcausesofpotentiallylife-threateningconditionsandseverematernaloutcomes
Womenwithpotentially Maternal Maternaldeaths
life-threateningconditions near-misscases
N= 212 N= 129 N= 16
n % n % n %
Underlyingcauses
Pregnancywithabortiveoutcome 44 20.75% 6 4.65% 1 6.25%
ObstetricHemorrhage 99 46.70% 85 65.89% 7 43.75%
Hypertensivedisorders 49 23.11% 27 20.93% 4 25.00%
Pregnancy-relatedinfection 9 4.25% 4 3.10% 1 6.25%
Otherobstetricdiseaseorcomplication 28 13.21% 22 17.05% 1 6.25%
Medical/Surgical/Mentaldiseaseorcomplication 38 17.92% 26 20.16% 5 31.25%
Unanticipatedcomplicationsofmanagement 14 6.60% 11 8.53% 1 6.25%
Coincidentalconditions 10 4.72% 8 6.20% 2 12.50%
Unknown 4 1.89% 3 2.33% 1 6.25%
Contributorycauses/associatedconditions
Anemia 96 45.28% 71 55.04% 8 50.00%
HIVinfection 0 0.00% 0 0.00% 0 0.00%
Previouscaesareansection 74 34.91% 58 44.96% 7 43.75%
Prolonged/obstructedlabor 28 13.21% 21 16.28% 3 18.75%
found in more than 50% of women with severe maternal develop in the hospital setting as the former indicates a
outcomes, and when compared with population level failure in access to the facilities and/or to the referral
data among Iraqi pregnant women (37.9%), this is a sta- chain where such hospitals would need adequate
tistically significant difference, underlining the vulner- resources and organization to deal with such emergen-
ablestatus ofthis sub-population[13]. cies [14]. In our study two thirds of women with SMO
In under-resourced settings there is a need to separate developing within the first twelve hours of admission
thenear-misscasesonarrivaltohospitalfrom thosethat were admitted without referrals, which may indicate an
Table4Endofpregnancyandpregnancyoutcomesinourstudypopulation
Potentiallylife-threateningconditions Maternalnear-misscases Maternaldeaths
N= 212 N= 129. N= 16
n % n % n %
Endofpregnancy
Vaginaldelivery 46 21.70% 37 28.68% 3 18.75%
CaesareanSection 107 50.47% 78 60.47% 11 68.75%
Completeabortion 0 0.00% 0 0.00% 0 0.00%
Curettage/vacuumaspiration 7 3.30% 5 3.88% 0 0.00%
Medicalmethodsforuterineevacuation 0 0.00% 0 0.00% 0 0.00%
Laparotomyforectopicpregnancy 39 18.40% 1 0.78% 1 6.25%
Other 4 1.89% 2 1.55% 0 0.00%
Unknown 2 0.94% 1 0.78% 0 0.00%
Womenstillpregnantathospitaldischargeordeath 7 3.30% 5 3.88% 1 6.25%
Caesareansectionrate* n/a 69.93% n/a 67.83% n/a 78.57%
Pretermbirths 57 37.25% 42 36.52% 8 57.14%
Stillbirths 32 20.92% 23 20.00% 6 42.86%
Perinataldeaths** 65 42.48% 50 43.48% 7 50.00%
*Caesareandeliveriesdividedbyalldeliveries/**fetaldeaths+intra-hospitalearlyneonatalmortality.
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page7of9
http://www.biomedcentral.com/1471-2393/13/11
Table5Severematernaloutcomesandnearmiss Table7Processoutcomeindicatorsrelatedwithspecific
indicators conditionsamongwomenwithseverematernal
Alllivebirthsinthepopulationundersurveillance 25,472 outcomes(maternalnearmissandmaternaldeaths)
Severematernaloutcomes(SMO)cases(n) 145 n %
Preventionofpostpartumhemorrhage
MaternalDeaths(n) 16
Targetpopulation:womengivingbirthinhealthfacilities 132 100.00%
Maternalnearmisscases(n) 129
Oxytocin 114 86.36%
Overallnear-missindicators
Anyuterotonic(includingoxytocin) 115 87.12%
SevereMaternalOutcomeratio(per1000livebirths) 5.69
Treatmentofseverepostpartumhemorrhage
Maternalnearmissincidenceratio(per1000livebirths) 5.06
Targetpopulation:womenwithseverePPH 84 100.00%
Maternalnearmissmortalityratio 9:1
Oxytocin 57 67.86%
Mortalityindex 11.03%
Ergometrine 52 61.90%
Hospitalaccessindicators
Misoprostol 35 41.67%
SMOcasespresentingtheorgandysfunctionormaternaldeath 127
within12hoursofhospitalstay(SMO12)(n) Otheruterotonics 3. 3.57%
ProportionofSMO12casesamongallSMOcases 87.59% Anyoftheaboveuterotonics 57 67.86%
ProportionofSMO12casescomingfromotherhealthfacilities 34.65% Tranexamicacid 6 7.14%
SMO12mortalityindex 11.02% Removalofretainedproducts 16 19.05%
Intra-hospitalcare Balloonorcondomtamponade 1. 1.19%
Intra-hospitalSMOcases(n) 18 Arteryligation 4 4.76%
Intra-hospitalSMOrate(per1000livebirths) 0.71 Hysterectomy 50 59.52%
Abdominalpacking 4 4.76%
Intra-Hospitalmortalityindex 11.11%
ProportionofcaseswithSMO 80 95.24%
Mortality 6. 7.14%
issue in the referral system and/or the detection of preg-
nant women with life-threatening complications outside AnticonvulsantsforEclampsia
these hospitals due to delay in seeking care or reaching Targetpopulation:womenwitheclampsia 43 100.00%
care. However it should be noted that in our study, the Magnesiumsulfate 29 67.44%
overall mortality index is very similar between women Otheranticonvulsant 22 51.16%
with severematernaloutcomeswho were referred versus
Anyanticonvulsant 40 93.02%
in-hospital patients. This is in contrast to a similar study
ProportionofcaseswithSMO 26 60.47%
where SMO was relatively higher among referred
Mortality 4 9.30%
women [12].
One of the key principles of effective management of Preventionofcaesareansectionrelatedinfection
complications related to pregnancy and childbirth is Targetpopulation:womenundergoingcaesareansection 107 100.00%
matching the level of care to the severity of the clinical Prophylacticantibioticduringcaesareansection 65 60.75%
conditions.Womenpresentingcomplicationsmayrequire Treatmentforsepsis
different levels of care, from basic obstetric care to inten-
Targetpopulation:womenwithsepsis 10 100.00%
sive care, including in this continuum comprehensive ob-
Parenteraltherapeuticantibiotics 10 100.00%
stetric care and surgery. Most of the women presenting
ProportionofcaseswithSMO 6 60.00%
organ dysfunction would be more appropriately managed
Mortality 2 20.00%
at the ICU level [15]. The low ICU admission rates
observed in this study suggests an important shortage of Ruptureduterus
ICU beds, which is corroborated by the substantial pro- Targetpopulation:womenwithruptureduterus 29 100.00%
portion of women experiencing organ dysfunction or Laparotomy 28 96.55%
Laparotomyafter3hoursofhospitalstay 5 17.24%
Table6Intensivecareuseinourstudypopulation
ProportionofcaseswithSMO 24 82.76%
Allwomengivingbirth 25,841
Mortality 3 10.34%
ICUadmissionrate 0.28%
Pretermbirth
ICUadmissionrateamongwomenwithSMO 37.24%
Targetpopulation:womenhavingapretermdeliveryafter 18 100.00%
SMOrateamongwomenadmittedtoICU 73.97% 3hoursofhospitalstay
ProportionofmaternaldeathsoccurredwithoutICUadmission 50.00% Corticosteroidsforfetallungmaturation 10 55.56%
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page8of9
http://www.biomedcentral.com/1471-2393/13/11
dyingwithoutaccesstoICU bed.Inaddition,it shouldbe Additional file
noted that two of the six hospitals had “close observation
units” instead of a proper ICU. Overall, the provision of Additionalfile1:TableS1.Criteriatoidentifypotentiallylife-
threateningconditionsandnearmiss[9].
adequate critical care, with appropriate staffing, equip-
ment and management strategies can contribute to a bet-
ter outcome among women with life-threatening Competinginterests
conditions[15]. Theauthorsdeclarethattheyhavenocompetinginterests.
The use of a criterion-based clinical audit method-
ology within WHO near-miss approach revealed oppor- Authors’contributions
MJhadtheideaoftestingtheWHOMaternalnear-missapproachinIraq.MJ
tunities to improve care, where a target population with
andJPSdevelopedthestudyprotocolanddatacollectionprocesses.MJ,IA,
aclearindicationofaneffectiveinterventionisidentified DMS,WA,SA,AA,RA,ADimplementationtheStudyinIraqandprovided
and then the use of this specific intervention is assessed. criticalinputduringthestudydevelopment,interpretationoffindingsand
reportingofthestudyresults.MJ,OTandJPSdraftedthemanuscriptand
Among the hospitals in our study, magnesium sulfate
leadtheinterpretationandpresentationoftheresults.Allauthors
for treatment of eclampsia, oxytocin for prevention and contributedtodraftingandrevisingthemanuscriptandhavereadand
treatment of postpartum hemorrhage, prophylactic anti- approvedthefinalversion.
biotics during cesarean section, and corticosteroids for
inducing fetal lung maturation in preterm birth were all Acknowledgements
ThisstudywaspresentedasthefinalassignmentfortheGFMER/2010course
underused. Of note, there were 5 cases of ruptured
inReproductiveHealthResearch.ThestudyauthorsaregratefultoDr.Hani
uterus that had the laparotomy being performed after M.Badr(GDMedicalCityTeachingComplex)forsponsoringthetwoday
three hours of hospital stay, suggesting an intra-hospital trainingcourse,Dr.OhanF.Yonan(CentreofTrainingandDevelopment
/MOH)forco-coordinatingandcommunicatingbetweentheHospitalsand
delay inthemanagement ofobstructed labor.
Ms.AmalEstephan(Dept.ofHealthandBiostatistics)forprovidingbaseline
This study has several strengths that deserve noting. informationrequiredforhospitalselection.
This is the first study assessing the quality of care in the
Authordetails
Iraqi facilities using the recent WHO maternal near-
1BaghdadTeachingHospital,Baghdad,Iraq.2DepartmentofHealthand
miss definition and criteria. We collaborated with the Biostatistics/MinistryofHealth,Baghdad,Iraq.3Al-YermoukTeaching
Iraqi Ministry of Health and by conducting the study Hospital,Baghdad,Iraq.4Al-KadimeyahTeachingHospital,Baghdad,Iraq.
5Al-ElwiyahMaternityTeachingHospital,Baghdad,Iraq.6IbnAl-BildiHospital
with the participation of the hospital staff and local cap-
forWomenandChildren,Baghdad,Iraq.7FatimaAl-Zahraa'Maternity
acity strengthening, we aimed to create a long lasting Hospital,Baghdad,Iraq.8DepartmentofPopulation,FamilyandReproductive
surveillance and quality improvement mechanism in Health,JohnsHopkinsBloombergSchoolofPublicHealth,615NWolfeSt,
Baltimore,MD21205,USA.9DepartmentofReproductiveHealthandResearch,
those hospitals, which can be replicated in other
WorldHealthOrganization,Geneva,Switzerland.
resource-poor settings. There are some limitations as
well. Our study was conducted in only six hospitals in Received:15February2012Accepted:26December2012
Published:16January2013
Baghdad; therefore our results cannot be generalized to
the overall country. Also, we presented our data in ag-
References
gregate, however an in-depth analysis would be more
1. MillenniumDevelopmentGoals;http://www.un.org/millenniumgoals/%5D.
beneficial to assess the causes and contributory factors 2. WorldHealthOrganization,UNICEF,UNFPA,TheWorldBank:Trendsin
in individual hospitals to improve quality of care. More maternalmortality:1990to2008.EstimatesdevelopedbyWHO,UNICEF,
UNFPAandTheWorldBank.Geneva,Switzerland:WorldHealth
importantly this manuscript only describes our work to
Organization;2010.
evaluate the quality of care and our subsequent findings, 3. WorldHealthOrganization,UNICEF,UNFPA,TheWorldBank:Trendsin
but does not cover the ongoing efforts to improve care maternalmortality:1990to2010.EstimatesdevelopedbyWHO,UNICEF,
UNFPAandTheWorldBank.Geneva,Switzerland:WorldHealth
based on these results. Also, it should be noted that
Organization;2012.
we only collected data up to 7th day postpartum, 4. SayL,PattinsonRC,GulmezogluAM:WHOsystematicreviewofmaternal
whereas the definition of near miss includes cases up morbidityandmortality:theprevalenceofsevereacutematernal
morbidity(nearmiss).ReprodHealth2004,1(1):3.
to 42 days.
5. WaterstoneM,BewleyS,WolfeC:Incidenceandpredictorsofsevere
obstetricmorbidity:case–controlstudy.BMJ2001,322(7294):1089–1093.
Conclusions discussion1093–1084.
6. MantelGD,BuchmannE,ReesH,PattinsonRC:Severeacutematernal
The use of the WHO maternal near-miss approach
morbidity:apilotstudyofadefinitionforanear-miss.BrJObstet
enabled the identification of important roadblocks to Gynaecol1998,105(9):985–990.
improve quality of maternal care in Baghdad, Iraq. Our 7. TuncalpO,HindinM,SouzaJ,ChouD,SayL:Theprevalenceofmaternal
nearmiss:asystematicreview.BJOG2012,119(6):653–661.
results suggest that severe maternal outcomes can be 8. SayL,SouzaJP,PattinsonRC:Maternalnearmiss–towardsastandardtool
potentially reduced by fostering the use of evidence- formonitoringqualityofmaternalhealthcare.BestPractResClinObstet
based interventions for life-threatening complications, Gynaecol2009,23(3):287–296.
9. WorldHealthOrganization:Evaluatingthequalityofcareforsevere
improving referral systems, and optimizing the use of
pregnancycomplications:theWHOnear-missapproachformaternal
criticalcare. health.Geneva:WHO;2011.
Jabiretal.BMCPregnancyandChildbirth2013,13:11 Page9of9
http://www.biomedcentral.com/1471-2393/13/11
10. OladapoOT,Sule-OduAO,OlatunjiAO,DanielOJ:"Near-miss"obstetric
eventsandmaternaldeathsinSagamu,Nigeria:aretrospectivestudy.
ReprodHealth2005,2:9.
11. AlmerieY,AlmerieMQ,MatarHE,ShahrourY,AlChamatAA,AbdulsalamA:
Obstetricnear-missandmaternalmortalityinmaternityuniversity
hospital,Damascus,Syria:aretrospectivestudy.BMCPregnancyChildbirth
2010,10:65.
12. AdisasmitaA,DevianyPE,NandiatyF,StantonC,RonsmansC:Obstetric
nearmissanddeathsinpublicandprivatehospitalsinIndonesia.
BMCPregnancyChildbirth2008,8:10.
13. MinistryofHealth,WorldHealthOrganization,EuropeanUnion:IraqFamily
HealthSurvey2006/7(IFHS2006/7).Baghdad,Iraq:MinistryofHealth;2008.
14. FilippiV,RonsmansC,GohouV,GoufodjiS,LardiM,SahelA,SaizonouJ,De
BrouwereV:Maternitywardsoremergencyobstetricrooms?Incidence
ofnear-misseventsinAfricanhospitals.ActaObstetGynecolScand2005,
84(1):11–16.
15. ZeemanGG:Obstetriccriticalcare:ablueprintforimprovedoutcomes.
CritCareMed2006,34(9Suppl):S208–S214.
doi:10.1186/1471-2393-13-11
Citethisarticleas:Jabiretal.:Maternalnearmissandqualityof
maternalhealthcareinBaghdad,Iraq.BMCPregnancyandChildbirth
201313:11.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit