monitoring emergency obstetric care

Monitoring emergency obstetric care

a handbook

Monitoringemergency obstetric care

a handbook

WHO Library Cataloguing-in-Publication Data :

Monitoring emergency obstetric care: a handbook.

1.Obstetrics - standards. 2.Emergency services, Hospital - statistics and numerical data. 3.Data collection - methods. 4.Quality indicators, Health care. 5.Maternal health services - supply and distribution. 6.Maternal mortality - prevention and control. 7.Handbooks. I.World Health Organization. II.United Nations Population Fund. III.UNICEF. IV.Mailman School of Public Health. Averting Maternal Death and Disability.

ISBN 978 92 4 154773 4 (NLM classification: WA 310)

© World Health Organization 2009

All rights reserved. Publications of the World Health Organization can be obtained from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: [email protected]). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press, at the above address (fax: +41 22 791 4806; e-mail: [email protected]).

The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concern-ing the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use.

Printed in

ContentsAbbreviations iv

Acknowledgements v

Preface vi

Executivesummary vii

1. INTRODUCTION 1

1.1Overviewofindicators 4

1.2SignalfunctionsofEmOC 6

1.3UseoftheEmOCindicators 9

2. INDICATORSFOREmOC 10

2.1Indicator1:AvailabilityofEmOCservices 10

2.2Indicator2:GeographicaldistributionofEmOCfacilities 13

2.3Indicator3:ProportionofallbirthsinEmOCfacilities 16

2.4Indicator4:MetneedforEmOC 19

2.5Indicator5:Caesareansectionsasaproportionofallbirths 25

2.6Indicator6:Directobstetriccasefatalityrate 31

2.7Indicator7:Intrapartumandveryearlyneonataldeathrate 34

2.8Indicator8:ProportionofdeathsduetoindirectcausesinEmOCfacilities 36

2.9Summaryandinterpretationofindicators1–8 38

3. COLLECTINGDATAFORTHEINDICATORS 43

3.1Typesofdatarequired 43

3.2Preparation 43

3.3Form1:AllpotentialEmOCfacilitiesinselectedareas 46

3.4Form2:ReviewofEmOCatfacilities 48

3.5Form3:SummaryofdataonEmOCfacilitiesinanarea 50

3.6Form4:Calculationofindicatorsforeacharea 51

3.7Form5:Calculationofindicatorsforthecountry 51

3.8Monitoringatthearealevel 51

REFERENCES 54

APPENDIXA: FormsandworksheetsfordatacollectionandcalculationofEmOCindicators 61

Form1.ListofpossibleEmOCfacilities 63

Form2.ReviewofpossibleEmOCfacilities 69

Form3.SummaryofdataonEmOCfacilitiesinthearea 85

Form4.Calculationofindicatorsforgeographicalarea 107

Form5.Calculationofindicatorsforacountry 123

APPENDIXB: Informationonregistersanddatacollection 145

APPENDIXC: Randomnumbertable 151

Monitoringemergencyobstetriccare:ahandbook iv

AbbreviationsAMDD AvertingMaternalDeathandDisabilityProgram

EmOC EmergencyObstetricCare

HIV Humanimmunodeficiencyvirus

UNFPA UnitedNationsPopulationFund

UNICEF UnitedNationsChildren’sFund

WHO WorldHealthOrganization

5Monitoringemergencyobstetriccare:ahandbook v

AcknowledgementsMonitoring emergency obstetric care: a handbook

waspreparedbyDeborahMaine(BostonUniversity,

Boston, Massachusetts, United States of America,

and the Averting Maternal Death and Disability

Program (AMDD),MailmanSchoolofPublicHealth,

ColumbiaUniversity,NewYorkCity,NewYork,United

States), Patsy Bailey (Family Health International,

Research Triangle Park, North Carolina, United

States, and AMDD), Samantha Lobis (AMDD) and

JudithFortney(AMDD).

ThehandbookisbasedonthepublicationGuidelines

for monitoring the availability and use of obstet-

ric services (1997) prepared by Deborah Maine,

TessaWardlaw(UNICEF)andateamfromColumbia

University(VictoriaWard,JamesMcCarthy,Amanda

Birnbaum,MuratAlkalinandJenniferBrown),andon

recommendations made during a technical consul-

tationheld in2006atWHOinGeneva, inwhichthe

following persons participated: Patsy Bailey (Family

Health International and AMDD), Shelah Bloom

(University of North Carolina, Chapel Hill, North

Carolina,UnitedStates),DavidBraunholtz (Initiative

for Maternal Mortality Programme Assessment

(IMMPACT) Project, University of Aberdeen,

Aberdeen, Scotland), Vincent de Brouwere (Prince

Leopold Institute of Tropical Medicine, Antwerp,

Belgium),MarcDerveeuw(UNFPA),HemantDwivedi

(UNFPA), Øystein Evjen Olsen (Institute for Health

ResearchandDevelopmentandPrimaryHealthCare,

Iringa,UnitedRepublicofTanzania),VincentFauveau

(UNFPA), Judith Fortney (AMDD), Lynn Freedman

(AMDD), Joan Healy (Ipas, Chapel Hill, North

Carolina, United States) Justus Hofmeyr (University

of the Witwatersrand, Johannesburg, South Africa),

Samantha Lobis (AMDD), Deborah Maine (Boston

University, Boston, Massachusetts, United States,

and AMDD), Saramma Mathai (UNFPA), Affette

McCaw-Binns(UniversityoftheWestIndies),Isabelle

Moreira(UNFPA),LuweiPearson(UNICEF),Rosalind

Raine(UniversityCollegeLondon,London,England),

Geetha Rana (UNICEF), Judith Standley (UNICEF),

Nancy Terreri (UNICEF), Kanako Yamashita-Allen

(World Bank, Washington DC, United States), Jelka

Zupan (WHO), Katherine Ba-Thike (WHO), Alexis

Ntabona(WHO),MatthewsMatthai(WHO).

LaleSay(WHO)helpedinrevisionofthehandbookby

facilitating the technicalconsultation, reviewingdraft

versions, and coordinating the publishing process.

VincentFauveau (UNFPA), JudithStandley (UNICEF)

and Lynn Freedman (AMDD) reviewed many drafts

within their organizations. Jennifer Potts (AMDD)

and Vincent de Brouwere reviewed several versions

andmadesubstantivecontributionstothetext.Yves

Bergevin (UNFPA), Luc de Bernis (UNFPA), Juliana

Bol (RAISE Initiative, Columbia University, New York

City, New York, United States), Sara Casey (RAISE

Initiative),FranceDonnay(UNFPA),MetinGulmezoglu

(WHO),JoanHealy(Ipas),RitaKabra(WHO),Barbara

Kwast(AMDD),CarineRonsmans(LondonSchoolof

HygieneandTropicalMedicine,London,England)and

CynthiaStanton(JohnsHopkinsBloombergSchoolof

PublicHealth,Baltimore,Maryland,UnitedStates)also

reviewed the handbook. Lucy Anderson, Alexandra

DelValle, Gina Gambone, Laura Harris, and Christen

Mullenhelped incompilationof the references.Paul

VanLookreviewedthetext.

ConflictofinterestThe participants of the technical consultation were

primarily independent experts from academia. No

conflicts of interest were declared. Other partici-

pantsincludedstafffromWHO,UNFPA,UNICEF,and

Columbia University who have been engaged in in-

countryapplicationoftheindicatorsreviewedatthe

consultation.

Monitoringemergencyobstetriccare:ahandbook vi

PrefaceEfforts to improve the lives of women and children

aroundtheworldhaveintensifiedsinceworldleaders

adoptedtheUnitedNationsMillenniumDeclarationin

September2000andcommittedthemselvestoreach-

ingMillenniumDevelopmentGoals4and5,onchild

mortalityandmaternalhealth.Theoriginaltargetsfor

theseGoalswereatwo-thirdsreductioninthemortal-

ityofchildrenunder5andathree-quartersreduction

inthematernalmortalityratiobetween1990and2015.

Thereisworldwideconsensusthat,inordertoreach

thesetargets,good-qualityessentialservicesmustbe

integratedintostronghealthsystems.Theadditionin

2007ofanewtarget inGoal5—universalaccess to

reproductivehealthby2015—reinforcesthisconsen-

sus:allpeopleshouldhaveaccesstoessentialmater-

nal,newborn,childand reproductivehealthservices

providedinacontinuumofcare.

In order to reduce maternal mortality, Emergency

ObstetricCare(EmOC)mustbeavailableandacces-

sible toallwomen.Whileallaspectsof reproductive

healthcareincludingfamilyplanninganddeliverywith

the help of a skilled health professional also plays

an important role in reducingmaternal andneonatal

mortality, this handbook focuses on the critical role

of EmOC in saving the lives of women with obstet-

riccomplicationsduringpregnancyandchildbirthand

savingthe livesofnewborns intrapartum.Thehand-

bookdescribesindicatorsthatcanbeusedtoassess,

monitorandevaluatetheavailability,useandquality

ofEmOC.

Whilst this handbook focuses on emergency care,

abroadersetof indicatorsshouldbeused tomoni-

tor fundamental aspects of reproductive health pro-

grammes designed to reduce maternal mortality,

ensure universal access to reproductive health care

andreducechildmortality.

7 This page has been left blank

ExecutivesummaryReducing maternal mortality has arrived at the top

of health anddevelopment agendas.Toachieve the

MillenniumDevelopmentGoalofa75%reduction in

thematernalmortalityratiobetween1990and2015,

countries throughout the world are investing more

energy and resources into providing equitable, ade-

quate maternal health services. One way of reduc-

ingmaternalmortalityisbyimprovingtheavailability,

accessibility,qualityanduseofservicesforthetreat-

ment of complications that arise during pregnancy

andchildbirth.Theseservicesarecollectivelyknown

asEmergencyObstetricCare(EmOC).

Sound programmes for reducing maternal mortality,

likeallpublichealthprogrammes,shouldhaveclear

indicators in order to identify needs, monitor imple-

mentation and measure progress. In order to fulfil

these functions, thedatausedtoconstruct the indi-

catorsshouldbeeitheralreadyavailableorrelatively

easyandeconomicaltoobtain.Theindicatorsshould

beabletoshowprogressoverarelativelyshorttime,

insmallaswellaslargeareas.Most importantly,the

indicators should provide clear guidance for pro-

grammes—showing which components are working

well,whichneedmore inputorneed tobechanged

andwhatadditionalresearchisneeded.

For a variety of technical and financial reasons, the

maternalmortalityratiodoesnotmeettheserequire-

ments.Consequently,in1991,UNICEFaskedColumbia

University(NewYorkCity,NewYork,UnitedStatesof

America)todesignanewsetofindicatorsforEmOC.

Thefirstversionwastestedin1992.In1997,theindi-

cators were published as Guidelines for monitoring

theavailabilityanduseofobstetricservices,issuedby

UNICEF,WHOandUNFPA(1).Theseindicatorshave

beenusedbyministriesofhealth,internationalagen-

cies and programme managers in over 50 countries

aroundtheworld.

InJune2006,aninternationalpanelofexpertspartici-

patedinatechnicalconsultationinGenevatodiscuss

modificationstotheexistingindicatorsforEmOCand

revisions to the Guidelines, taking into account the

accumulatedexperienceandincreasedknowledgein

theareaofmaternalhealthcare.Thepresenthand-

book contains the agreed changes, including two

newindicatorsandanadditionalsignalfunction,with

updated evidence and new resources. In addition,

the Guidelines were renamed as the Handbook, to

emphasizethepracticalpurposeofthispublication.

The purpose of this handbook is to describe the

indicatorsandtogiveguidanceonconductingstud-

iestopeopleworkinginthefield.Itincludesalistof

life-savingservices,or ‘signal functions’, thatdefine

a health facility with regard to its capacity to treat

obstetricandnewbornemergencies.Theemphasisis

onactual rather thantheoretical functioning.Onthe

basisoftheperformanceoflife-savingservicesinthe

past3months,facilitiesarecategorizedas‘basic’or

‘comprehensive’.Thesectiononsignalfunctionsalso

includesanswerstofrequentlyaskedquestions.

TheEmOCindicatorsdescribedinthishandbookcan

beusedtomeasureprogressinaprogrammaticcon-

tinuum:fromtheavailabilityofandaccesstoEmOC

totheuseandqualityofthoseservices.Theindica-

torsaddressthefollowingquestions:

• ArethereenoughfacilitiesprovidingEmOC?

• Arethefacilitieswelldistributed?

• Areenoughwomenusingthefacilities?

• Aretherightwomen(i.e.womenwithobstetric

complications)usingthefacilities?

• Areenoughcriticalservicesbeingprovided?

• Isthequalityoftheservicesadequate?

Monitoringemergencyobstetriccare:ahandbook vii

Monitoringemergencyobstetriccare:ahandbook

Thehandbookprovidesadescriptionofeachindica-

torandhowitisconstructedandhowitcanbeused;

the minimum and/or maximum acceptable level (if

appropriate); the background of the indicator; data

collection and analysis; interpretation and presenta-

tionoftheindicator;andsuggestionsforsupplemen-

tarystudies.Thereisafurthersectiononinterpretation

ofthefullsetofindicators.Sampleformsfordatacol-

lectionandanalysisareprovided.

Useof theseEmOC indicators toassessneedscan

help programme planners to identify priorities and

interventions. Regular monitoring of the indicators

alerts managers to areas in which advances have

beenmadeandthosethatneedstrengthening.Close

attention to the functioningofkeyservicesandpro-

grammescansubstantiallyandrapidlyreducemater-

nalmortalityindevelopingcountries.

viii

Monitoringemergencyobstetriccare:ahandbook 1

1.IntroductionOverthepasttwodecades,theinternationalcommu-

nityhasrepeatedlydeclareditscommitmenttoreduce

the high levels of maternal mortality in developing

countries, starting with the 1987 Safe Motherhood

Conference in Nairobi, Kenya, followed by the 1990

World Summit for Children at United Nations head-

quarters, the 1994 International Conference on

Population and Development in Cairo, Egypt, the

1995FourthWorldConferenceonWomeninBeijing,

China,‘Nairobi10YearsOn’inSriLankain1997,and

theMillenniumDevelopmentGoalsestablishedbythe

UnitedNationsin2000.In2007,anumberofevents

marked the20thanniversaryof the launchingof the

Safe Motherhood Initiative, including the Women

DeliverConferenceinLondon,England,atwhichcalls

were made for renewed commitment, programmes

and monitoring. Most importantly, over the past 20

years,consensushasbeenreachedontheinterven-

tions that are priorities in reducing maternal mortal-

ity (2). Stakeholders agree that good-quality EmOC

shouldbeuniversallyavailableandaccessible,thatall

womenshoulddelivertheirinfantsinthepresenceof

aprofessional,skilledbirthattendant,andthatthese

keyservicesshouldbeintegratedintohealthsystems.

It becameclear earlyon, however, that itwouldnot

be simple to measure progress in this area. The

conventional approach was to monitor the number

of maternal deaths with ‘impact’ indicators such

as the maternal mortality ratio. In theory, repeated

measurements of this ratio over time can be used

to monitor trends. This approach has a number of

serious drawbacks, both technical and substantive.

Maternal mortality is extremely difficult and costly

to measure when vital registration systems are

weak, and even when systems are strong (3). Even

innovativemethodspresentdifficulties.Forexample,

the direct ‘sisterhood’ method provides information

forareferenceperiodof7yearsbeforeasurvey;thus,

the informationgathereddoesnotreflectthecurrent

situationorprogressmaderecently.Recentadvances

in sampling procedures for the sisterhood method

have, however, greatly increased its efficiency and

havedecreasedcosts.Thesechangesallowforlarger

samplesandconsequentlyashorterreferenceperiod

andnarrowerconfidenceintervalsthanthetraditional

approach. Even this method, however, is known to

give underestimates of the maternal mortality ratio

(4,5).

Another approach is use of ‘process,’ ‘output’ or

‘outcome’ indicators, to measure the actions that

prevent deaths or illness. Widely used process

indicatorsincluderatesofchildhoodimmunizationand

contraceptiveprevalence.Thishandbookpresentsa

seriesofindicatorsdesignedtomonitorinterventions

that reduce maternal mortality by improving the

availability, accessibility, use and quality of services

for the treatmentofcomplicationsduringpregnancy

andchildbirth.Theindicatorsarebasedoninformation

from health facilities with data on population and

birth rates. There are several advantages to this

approach. First, the indicators can be measured

repeatedlyatshortintervals.Secondly,theindicators

provide information that isdirectlyuseful forguiding

policies and programmes and making programme

adjustments.Itisimportanttorememberthatalthough

‘process,’‘output’and‘outcome’indicatorsaremore

useful,practicaland feasible than impact indicators,

formanyreasons, thesemeasurescannotsubstitute

formaternalmortalityratiosasadirectmeasureofthe

overalllevelofmaternalmortalityinapopulation.

TheGuidelinesformonitoringtheavailabilityanduseof

obstetricserviceswereinitiallydevelopedbyColumbia

University’sSchoolofPublicHealth,supportedbyand

incollaborationwithUNICEFandWHO.Adraftversion

wasissuedin1992,andtheguidelineswereformally

publishedbyUNICEF,WHOandUNFPA in1997 (1).

Since then, they have been used in many countries

(Table 1). The present document is a revision of the

1997versionoftheguidelines,incorporatingchanges

based on monitoring and assessment conducted

worldwide.

The recommendations related to measuring the

indicatorswerereviewedandupdatedonthebasisof

existingevidence,aswellasexperienceinusingthe

indicatorswithincountryprogrammes.

These recommendations will be updated regularly

usingstandardWHOprocedures. It isexpectedthat

thenextupdatewillbein2014.

1


Regionandcountry Useofindicators References

Africa

Angola Nationalneedsassessment(reportinprogress)

Benin Nationalneedsassessment (6,7)

Burundi NeedsassessmentplannedwithUNICEF

Cameroon Subnationalneedsassessment (8-10)

Chad Nationalneedsassessment (7,11)

Comoros (12)

Côted’Ivoire Nationalneedsassessment (10,13)

Eritrea Needsassessmentwithpartialcoverage (14)

Ethiopia Programmemonitoringandevaluation;needsassessmentwithpartialcoverage1

(15)

Gabon Nationalneedsassessment (16,17)

Gambia Nationalneedsassessment (17,18)

Ghana Subnationalneedsassessment (19)

Guinea Subnationalneedsassessment (20)

GuineaBissau Nationalneedsassessment (17,21)

Kenya Subnationalneedsassessments2 (22-24)

Lesotho Nationalneedsassessment (25)

Madagascar Subnationalneedsassessments (26)

Malawi Nationalneedsassessment;programmemonitoringandevaluation

(27-30)

Mali Nationalneedsassessment;programmemonitoringandevaluation

(31,32)

Mauritania Nationalneedsassessment (10,33)

Mozambique Nationalneedsassessment;programmemonitoringandevaluation(datanotyetanalysed)

(34-37)

Namibia Needsassessment (38)

Niger Needsassessment (10,39)

Rwanda Subnationalneedsassessment;programmemonitoringandevaluation

(15,23,39-42)

Senegal Nationalneedsassessment (10,37,43)

SierraLeone Nationalneedsassessment (44)

Uganda Nationalneedsassessment (23,45,46)

UnitedRepublicofTanzania Nationalneedsassessment;programmemonitoringandevaluation

(15,39,47-51)

Zambia Nationalneedsassessment (52)

Zimbabwe Nationalneedsassessment (53,54)

Americas

Bolivia Nationalneedsassessment3 (55,56)

Ecuador NationalneedsassessmentwithUNFPA,2006

ElSalvador Nationalneedsassessment (56-58)

Guatemala Needsassessment (59)

Honduras Nationalneedsassessment (56,60)

Nicaragua Nationalandsubnationalneedsassessments;programmemonitoringandevaluation

(61,62)

Peru Needsassessmentswithpartialcoverage;programmemonitoringandevaluation4

(63-65)

UnitedStates Nationalneedsassessment (66)

Table 1. Selected countries in which emergency obstetric care indicators were used in assessing needs or for monitoring and evaluation (2000–2007)


Regionandcountry Useofindicators References

EasternMediterranean

Afghanistan Needsassessmentswithpartialcoverage (67)

Djibouti Nationalneedsassessment (68)

Iraq Needsassessmentplanned

Morocco Nationalneedsassessment;programmemonitoringandevaluation

(62,69)

Pakistan Needsassessmentswithpartialcoverage;programmemonitoringandevaluation

(70-73)

Somalia Subnationalneedsassessment (74)

Sudan Nationalneedsassessment (23,75)

SyrianArabRepublic Nationalneedsassessment5

Yemen Needsassessmentswithpartialcoverage

Europe

Kyrgyzstan Nationalneedsassessment6

Tajikistan Nationalneedsassessment;programmemonitoringandevaluation7

(76)

South-EastAsia

Bangladesh Nationalandsubnationalneedsassessments;programmemonitoringandevaluation

(77-79)

Bhutan Needsassessment;programmemonitoringandevaluation (9,80)

India Needsassessmentswithpartialcoverage;programmemonitoringandevaluation

(9,81-85)

Nepal Subnationalneedsassessment;programmemonitoringandevaluation

(37,86-88)

SriLanka Subnationalneedsassessment;programmemonitoringandevaluation

(62,89)

Thailand Needsassessmentwithpartialcoverage (90)

WesternPacific

Cambodia Planned

Mongolia Planned

VietNam Needsassessmentwithpartialcoverage;programmemonitoringandevaluation

(91,92)

1CARE.Unpublisheddata.2000.2DoctorsoftheWorld.WestPokotfacilityneedsassessment—maternalandnewborncare.Unpublisheddata.Nairobi,2007.3EngenderHealthAcquireProject.Unpublisheddata.2007.4CARE.Unpublisheddata.2004:Huancavelicaregion,Peru.5MinistryofHealthandUNICEF,Unpublisheddata.2004:Syria.6MinistryofHealthofKyrgyzstanandUNICEF,StatusofEmergencyObstetricCare(EOC)intheKyrgyzRepublic.Unpublished.2005.7MinistryofHealthofTajikistanandUNICEF,Unpublisheddata.Dushanbe,2005.


Inthisnewedition,theindicatorshavebeenrevisedto

reflect10years’wealthofexperience.Otherchanges

reflect the broadening of programmes; e.g. a signal

function on treatment of complications in newborns

and new indicators on perinatal mortality and on

maternaldeaths reportedasdue to indirect causes,

such as HIV and malaria, have been added. These

changes were discussed and agreed by an interna-

tionalpanelofexpertsatthetechnicalconsultationin

June2006(93).Duringthereview,itwasalsodecided

tochangethetitle.Weusetheterm‘handbook’rather

than ‘guidelines,’ because ‘handbook’ reflects more

accurately the practical nature of this document.

Anotherchangemadeinthiseditionisreplacementof

‘essentialobstetriccare’by‘EmOC’.1Overtheyears,

theterminologyhasbeenadjustedsothattheindica-

torsrelatespecificallytotreatmentoftheemergency

obstetric complications that cause most maternal

deaths.

Thishandbookincludesanexplanationofthecurrent

indicatorsforEmOCandtheirimplications,suggests

supplementarystudiesthatcanimproveunderstanding

ofthesituationinagivenarea,andprovidesanswers

to common questions that arise when using the

indicators.Thisisfollowedbyworksheetsandtables

toillustratestudyquestionsandcalculations.

Theindicatorsdescribedcanbeusedatanystageof

thedesignandimplementationofEmOCprogrammes

and can be incorporated into routine health

managementinformationsystems.Inmanycountries,

theseindicatorshaveprovidedtheframeworkformore

detailed assessments of national needs for EmOC,

establishingtheavailability,useandqualityofservices

and the specific information needed for detailed

programmeplanning,suchasequipmentinventories.2

Modules for conducting needs assessments can be

foundat:www.amddprogram.org.

1.1Overviewofindicators

Inthesectionsbelow,wepresentaseriesofindicators

formonitoringprogressinthepreventionofmaternal

andperinataldeaths.Theirorderisbasedonthelogic

that,forwomentoreceiveprompt,adequatetreatment

forcomplicationsofpregnancyandchildbirth,facilities

forprovidingEmOCmust:

• existandfunction,

• begeographicallyandequitablydistributed,

• beusedbypregnantwomen,

• beusedbywomenwithcomplications,

• providesufficientlife-savingservices,and

• providegood-qualitycare.

Thus,theindicatorsanswerthefollowingquestions:

• ArethereenoughfacilitiesprovidingEmOC?

• Arethefacilitieswelldistributed?

• Areenoughwomenusingthefacilities?

• Aretherightwomenusingthefacilities?

• Areenoughcriticalservicesbeingprovided?

• Isthequalityofservicesadequate?

Thefirstindicatorthereforefocusesontheavailability

ofEmOCservices.Adequatecoveragemeansthatall

pregnantwomenhaveaccesstofunctioningfacilities.

Once availability is established, questions of use

can be addressed. Even if services are functioning,

if women with complications do not use them (for

whateverreason),theirlivesareindanger.Finally,the

indicators cover theperformanceof health services.

Afterall,manywomendieinhospital:someofthemdie

because theywerenotadmitteduntil theircondition

wascritical;manyothers,however,diebecausethey

didnotreceivetimelytreatmentatahealthfacilityor

becausethetreatmenttheyreceivedwasinadequate.

Table2showsthesixEmOCindicatorsissuedin1997,

withsomeminormodificationssuggestedbythe2006

technicalconsultationonthebasisoftheparticipants’

expertiseandexperienceinvariouscountries:

1

‘Emergency obstetric care’ or ‘EmOC’ is being used in thisdocument rather than ‘emergency obstetric and newborn care’or ‘EmONC’ because this set of indicators focus primarily onobstetric complications and procedures. While there is one newsignalfunctiononneonatalresuscitationandonenewindicatoronintrapartumcare fromtheperspectiveof thenewborn, thesetofindicatorsdonot represent the full rangeofemergencynewbornprocedures.2Theseassessmentsalsoincludemoreinformationonemergencynewborncare,andareoftencalledEmONCneedsassessments.


• Therecommendationforthemixtureofbasic

andcomprehensiveEmOCfacilitiesper500000

populationhasbeenchangedfrom‘atleastone

comprehensiveandfourbasicEmOCfacilitiesper

500000population’to‘atleastfiveEmOCfacili-

tiesincludingatleastonecomprehensivefacility

per500000population’.

Table 2. The original six emergency obstetric care indicators, with modifications

Indicator Acceptablelevel

1. Availabilityofemergencyobstetriccare:basicandcomprehensivecarefacilities

Thereareatleastfiveemergencyobstetriccarefacilities(includingatleastonecomprehensivefacility)forevery500000population

2. Geographicaldistributionofemergencyobstetriccarefacilities

Allsubnationalareashaveatleastfiveemergencyobstetriccarefacilities(includingatleastonecomprehensivefacility)forevery500000population

3. Proportionofallbirthsinemergencyobstetriccarefacilitiesa

(Minimumacceptableleveltobesetlocally)

4. Metneedforemergencyobstetriccare:proportionofwomenwithmajordirectobstetriccomplicationswhoaretreatedinsuchfacilitiesa

100%ofwomenestimatedtohavemajordirectobstetriccomplicationsbaretreatedinemergencyobstetriccarefacilities

5. Caesareansectionsasaproportionofallbirthsa Theestimatedproportionofbirthsbycaesareansectioninthepopulationisnotlessthan5%ormorethan15%c

6. Directobstetriccasefatalityratea Thecasefatalityrateamongwomenwithdirectobstetriccomplicationsinemergencyobstetriccarefacilitiesislessthan1%

• Theminimumacceptablelevelforindicator3was

removed,andcountriesareadvisedtousetheir

owntargets.

• Thenameofindicator6hasbeenupdatedfrom:

‘casefatalityrate’to‘directobstetriccasefatality

rate’.

Adaptedfromreference(1).aWhiletheseindicatorsfocusonservicesprovidedinfacilitiesthatmeetcertainconditions(andthereforequalifyas‘emergencyobstetriccarefacilities’),westronglyrecommendthattheseindicatorsbecalculatedagainwithdatafromallmaternityfacilitiesintheareaeveniftheydonotqualifyasemergencyobstetriccarefacilities.bTheproportionofmajordirectobstetriccomplicationsthroughoutpregnancy,deliveryandimmediatelypostpartumisestimatedtobe15%ofexpectedbirths.cSeesection2.5foradiscussionofthisrange.

These indicators refer to the availability and use of

facilitiesandtheperformanceofhealth-caresystems

insavingthelivesofwomenwithobstetriccomplica-

tions.Theacceptablelevelsofmostoftheindicators

are specified as minimum and/or maximum and are

necessarilyapproximate.Theyarebasedonthebest

data,estimates,andassumptionscurrentlyavailable.

The acceptable levels can be adapted according to

countries’circumstances;however, if theyaremodi-

fied,itisimportanttoreportthefindingsinrelationto

thestandardlevelssuggestedhere,sothattheresults

canbecomparedwiththosefromotherstudies.

Theseindicatorscanbeusedtosetprioritiesforpro-

grammes as well as to monitor them. Programme

plannersandmanagers responsible for reducing the

numberofmaternaldeathscanstartatthetopofthe

listandworkdown.Whentheyreachanindicatorfor

whichthecountrydoesnotmeettheacceptablelevel,

appropriateinterventionsareneeded.Forexample,if

a country meets the acceptable levels for the num-

ber and distribution of EmOC facilities but not for

theiruse,interventionsareneededtounderstandand

improveuse.


Table3setsouttwonewindicatorsthatwereadopted

at the2006technicalconsultationontheguidelines.

Thesereflecttheevolutionofthematernalhealthfield:

indicator7reflectstherenewedfocusonthequalityof

obstetriccareandtheassociationbetweenmaternal

andneonatalhealth,and indicator8 reflects indirect

causesofmaternaldeathsinsomecountries,suchas

malaria.

Table 3. New indicators for emergency obstetric care

Indicator Acceptablelevel

7. Intrapartumandveryearlyneonataldeathrate Standardstobedetermined

8. Proportionofmaternaldeathsduetoindirectcausesinemergencyobstetriccarefacilities

Nostandardcanbeset

Theseindicatorsshouldalsobecalculatedwithdataforallfacilitiesinthearea,ifpossible.

1.2SignalfunctionsofEmOC

Forthepurposesofassessingandmonitoringthelevel

ofcarethatafacilityisactuallyproviding,itishelpful

touseashortlistofclearlydefined‘signalfunctions’.

Thesearekeymedical interventionsthatareusedto

treatthedirectobstetriccomplicationsthatcausethe

vastmajorityofmaternaldeathsaroundtheglobe.The

listofsignalfunctionsdoesnotincludeeveryservice

thatoughttobeprovidedtowomenwithcomplicated

pregnancies or to pregnant women and their new-

bornsingeneral;thatinformationisprovidedinother

publications (94-96).Thesignal functionsare indica-

torsofthelevelofcarebeingprovided.Furthermore,

somecriticalservicesaresubsumedwithinthesesig-

nalfunctions.Forexample,ifcaesareansectionsare

performedinafacility,thisimpliesthatanaesthesiais

being provided. While the signal functions are used

toclassify facilitiesonthebasisthat thesefunctions

havebeenperformedinthepast3months,itishelpful

touseamoreinclusivelistoffunctionsandsupplies

whenassessingneedforEmOCinordertoplanpro-

grammes.

Thelistofsignalfunctionsinthiseditionofthehand-

bookhasbeenupdatedwiththeadditionofthenew

function:‘performneonatalresuscitation’atbasicand

comprehensivelevels.Inaddition,thenameofthesec-

ondsignal functionhasbeenchanged from ‘admin-

ister parenteral oxytocics’ to ‘administer uterotonic

drugs’.ThelistofsignalfunctionsinTable4includes

afewexamplesofdrugsorequipmentthatcouldbe

usedwhenperformingthesignalfunctions;however,

the drugs and procedures mentioned are illustrative

and not exhaustive. For a complete list of recom-

mendedproceduresanddrugs,pleaserefertoWHO’s

Managingcomplicationsinpregnancyandchildbirth:

aguideformidwivesanddoctors(95)andManaging

newborn problems: a guide for doctors, nurses and

midwives(96).


Table 4. Signal functions used to identify basic and comprehensive emergency obstetric care services

Basicservices Comprehensiveservices

(1) Administerparenteral1antibiotics Performsignalfunctions1–7,plus:

(2)Administeruterotonicdrugs2(i.e.parenteraloxytocin)

(8)Performsurgery(e.g.caesareansection)

(3)Administerparenteralanticonvulsantsforpre-eclampsiaandeclampsia(i.e.magnesiumsulfate).

(9)Performbloodtransfusion

(4) Manuallyremovetheplacenta

(5)Removeretainedproducts(e.g.manualvacuumextraction,dilationandcurettage)

6) Performassistedvaginaldelivery(e.g.vacuumextraction,forcepsdelivery)

(7) Performbasicneonatalresuscitation(e.g.withbagandmask)

Abasicemergencyobstetriccarefacilityisoneinwhichallfunctions1–7areperformed.Acomprehensiveemergencyobstetriccarefacilityisoneinwhichallfunctions1–9areperformed.

Pleaserefertothefollowingwebsitesforrecommendedproceduresforeachsignalfunctionlistedabove:- Managingcomplicationsinpregnancyandchildbirth:aguideformidwivesanddoctors: http://www.who.int/making_pregnancy_safer/documents/9241545879/en/index.html- Cochranereviews:http://www.cochrane.org/reviews

Adaptedfromreference(1).1Injectionorintravenousinfusion.2Uterotonicdrugsareadministeredbothtopreventandtotreatpostpartumhaemorrhage.ArecentWHOtechnicalconsultation(Nov2008)todevelopguidelinesforinterventionsforpreventingpostpartumhaemorrhage,reviewedallavailableevidence,andidentifiedparenteraloxytocinastherecommendedchoiceofdrugforpreventionofpostpartumharemorrhage.Parenteralergometrine(2ndline)andmisoprostol(3rdline)areoptionsthatshouldonlybeusedwhereoxytocinisnotavailable.

Table 5. Signal functions and related complications

Majorobstetriccomplication Signalfunction

Haemorrhage Antepartum:PerformbloodtransfusionPerformsurgery(e.g.caesareansectionforplacentapraevia)Postpartum:AdministeruterotonicdrugsPerformbloodtransfusionPerformmanualremovalofplacentaPerformremovalofretainedproductsPerformsurgery(hysterectomy)foruterinerupture

Prolongedorobstructedlabour PerformassistedvaginaldeliveryPerformsurgery(caesareansection)AdministeruterotonicdrugsPerformneonatalresuscitation

Postpartumsepsis AdministerparenteralantibioticsRemoveretainedproductsPerformsurgeryforpelvicabscess

Table5 showswhichsignal functionsareused to treat themajordirectobstetric complications that cause

mostmaternaldeaths.Box1listsanumberofquestionsfrequentlyaskedaboutthesignalfunctions,withtheir

answers.


Majorobstetriccomplication Signalfunction

Complicationsofabortion Forhaemorrhage:PerformbloodtransfusionRemoveretainedproductsForsepsis:AdministerparenteralantibioticsRemoveretainedproductsForintra-abdominalinjury:AdministerparenteralantibioticsPerformbloodtransfusionPerformsurgery

Pre-eclampsiaoreclampsia AdministerparenteralanticonvulsantsPerformneonatalresuscitationPerformsurgery(caesareansection)

Ectopicpregnancy PerformsurgeryPerformbloodtransfusion

Ruptureduterus PerformsurgeryPerformbloodtransfusionAdministerparenteralantibiotics

Newborndistress(intrapartum) PerformnewbornresuscitationPerformsurgery(caesareansection)

Adaptedfromreference(97).

Box 1. Frequently asked questions about signal functions

• Whyuseparenteraladministration,ratherthanoral?Inanemergency,theremustbeaquickphysiologicalresponsetoantibiotics,oxytocicsandanticonvulsantswhenneeded.Inaddition,thekeylifesavingdrugsformaincomplicationscanonlybeadministeredparenteral.Therefore,thedefinitionspecifiesparenteralratherthanoraladministration.

• Whyweretheseitemsselectedassignalfunctionsandnotothers?Other itemshavebeendiscussedassignalfunctions,suchasuseofthepartograph,activemanagementofthethirdstageoflabour,availabilityofservices24h/day,7days/week,intravenousfluids,anaesthesiaandplasmaexpanders.Useofthepar-tographandactivemanagementofthethirdstageoflabourarebothpartofgoodobstetricpracticeandshouldbeusedforallwomeninlabourtopreventprolonged,obstructedlabouranditssequelae,suchasobstetricfistula.Availabilityofservices24h/day,7days/weekisafunctionofmanagementandplanningratherthana life-savingskill. Intravenousfluidsare implicit inthesignal functionsthatrequireparenteraldrugs.Anaesthesiaandplasmaexpandersarealsoimplicitintheavailabilityofobstetricsurgery,e.g.cae-sareansection.Althoughtheeightoriginalobstetricsignal functionsdonot formanexhaustive list, theywerechosenbecauseoftheroletheyplayinthetreatmentofthefivemajorcausesofmaternaldeath.

• WherecanIobtainamorecompletelistoffunctionsandequipmentformaternalandnewbornhealth?Thewebsites of WHO (http://www.who.int/reproductive-health/publications/pcpnc/) (98), the Johns HopkinsProgramfor InternationalEducationinGynecologyandObstetrics(http://www.jhpiego.org/scripts/pubs/category_detail.asp?category_id=24) (99) and AMDD (http://www.amddprogram.org/resources/DesignEvalMM-EN.pdf) (100)provide linkstomanualswithmorecomplete inventoriesofdrugs,suppliesandequipmentforhealthcentresandhospitals.


• Whydon’tthesignalfunctionsincludespecificdrugsorequipment?Wehopethatinternationalstandardsofcarewillbeusedtodetermineinpracticewhichdrugsandtypesofequipmentareusedtoperformthesignalfunctions.Thesestandardsaredynamicandcanchangeoverlongperiodswithtechnologicalprog-ress.WeencourageuseoftheWHOguidelinesofcare,theReproductiveHealthLibrary(http://www.who.int/rhl),theCochraneCollaborationsystematicreviewsandotherinternationalresources.ThelistofsignalfunctionsinTable4doesincludeafewexamplesofdrugsorequipmentthatcouldbeused,butthelistofoptionsisnotexhaustive.

• Whyusethe3–monthreferenceperiodasopposedtoalongertime?The3–monthreferenceperiodwaschosenbecauseitprovidesasnapshotofthecurrentfunctioningofafacility.Itwasalsoselectedbecauserecallislessaccurateoverlongerperiodsandbecauseskills(suchasvaginaldeliverywithavacuumextrac-tor,caesareansectionormanualremovaloftheplacenta)aremorelikelytobemaintainediftheyareusedfrequently.Monitoringthedeliveryofservicesandstockoutsareconsiderationsforhealthserviceplanners.

• Whatshouldwedowhena facility that isbeingmonitoredprovidesbasicorcomprehensiveemergencyobstetriccareirregularlybecauseofoneortwomissingsignalfunctions?Thisisnotaprobleminafacility-basedsurveyoraneedsassessment,asthetechnicalguidelineistoassesstheperformanceofthesignalfunctionsinthemostrecent3–monthperiod.Itbecomesanissuewhenmonitoringemergencyobstetriccarestatusovertime.Itisnotuncommonforafacilitytochangeitsstatuswhenithasasmallcaseloadorfrequentstaffturnover.Forpragmaticandprogrammaticreasonsinregionalornationalmonitoring,werec-ommendannualreclassification.Districtmanagerscanmonitortheirownperformancemorefrequentlyandshouldbeencouragedtodosoinordertoassesstheirfunctioningandtoprovidedatafordecision-makingtoimproveservices.

• Whatdowedoifasignalfunctionisperformedduringthe3–monthreferenceperiodbutnotinanobstetriccontext?Mostofthesignalfunctionsarelikelytobeperformedonlyinanobstetriccontext,butparenteralantibioticsoranticonvulsantsandbloodtransfusionscanbeadministeredinothercontexts.Inanassess-mentofaninstitution’scapacityandperformancefordeliveringemergencyobstetriccare,thesignalfunc-tionsshouldhavebeenperformedinanobstetriccontext.

1.3UseoftheEmOCindicators

Asshown inTable1, the indicators forEmOChave

been used in more than 50 countries to plan pro-

grammes and to monitor and evaluate progress in

reducing maternal mortality. Some countries have

conducted more detailed needs assessments that

also include other indicators and information use-

ful for planning safe motherhood programmes. (For

sample data collection forms, refer to: http://www.

amddprogram.org/).Inothercountries,morefocused

needsassessmentshavebeenconducted,datacol-

lectionbeinglimitedtotheindicatorsonformssimilar

to those in Appendix A. The more focused compo-

nentsofneedsassessmentsdescribedinthishand-

bookcanbe integrated intoneedsassessments for

otherhealthissues,suchaspreventionofmother-to-

childtransmissionofHIVinfection,orforahealthsys-

temoverall.RegardlessofwhethertheEmOCneeds

assessment ismoredetailedormore focused, itwill

yielddata thatcanbeused tomonitorandevaluate

progress in reducing maternal mortality and provide

valuable information for health ministries and health

managerstoshapestrategiesandactivitiestoimprove

maternalhealthoutcomes.

Inmoreandmorecountries,theEmOCindicatorshave

beenintegratedintoroutinehealthmanagementinfor-

mationsystemstotrackprogressatdistrict,regional

andnationallevels.Whileperiodicneedsassessments

anddatacollectionsystemssetupoutsidehealthman-

agement informationsystemsmayplayanimportant

role, integration of the EmOC indicators into health

management informationsystems isamoreefficient

wayofmonitoringtheavailabilityanduseofsuchcare

overtime.Countriesthatareintentonreducingmater-

nalmortalityshouldstrivetoincludetheseindicators

intotheirhealthmanagementinformationsystems.

10

2.IndicatorsforEmOCBelow, the explanation of each EmOC indicator

includes a description, the recommended minimum

or maximum acceptable level (if appropriate), back-

groundinformation,adviceondatacollection,analy-

sis, interpretationandpresentation,andsuggestions

for supplementary studies related to the indicator.

Worksheets are provided in Appendix A to facilitate

thecalculations.

2.1Indicator1:AvailabilityofEmOCservices

Description

TheavailabilityofEmOCservicesismeasuredbythe

numberof facilities thatperformthecompletesetof

signal functions inrelationto thesizeof thepopula-

tion.Whenstaffhascarriedoutthesevensignalfunc-

tions of basic EmOC in the 3-month period before

theassessment,thefacilityisconsideredtobeafully

functioning basic facility. The facility is classified as

functioningatthecomprehensivelevelwhenitoffers

thesevensignal functionsplussurgery (e.g.caesar-

ean)andbloodtransfusion(Table4).

Todeterminetheminimumacceptablenumberofbasic

andcomprehensiveEmOC facilities for a countryor

region (dependingon thescopeof theassessment),

beginbydividingthetotalpopulationby500000.This

istheminimumacceptablenumberofcomprehensive

facilities.Then,multiplythatnumberby5tocalculate

the overall minimum number of facilities, both basic

andcomprehensive.Thesenumbersshouldbecom-

pared with the actual number of facilities found in

ordertoclassifytheservicesasfullyfunctioningbasic

orcomprehensiveEmOCfacilities.

Theresultsofthisexercisecanalsobeexpressedas

apercentageof theminimumacceptablenumberof

basic or comprehensive care facilities. To calculate

thepercentageof therecommendedminimumnum-

beroffacilitiesthatisactuallyavailabletothepopu-

lation, divide the number of existing facilities by the

recommendednumberandmultiplyby100.Asimilar

exercisewilldeterminewhatpercentageoftherecom-

mendedminimumnumberofcomprehensivefacilities

isavailable.

Minimumacceptablelevel

Forevery500000population, theminimumaccept-

ablelevelisfiveEmOCfacilities,atleastoneofwhich

providescomprehensivecare.

Backgroundanddiscussion

To save women with obstetric complications, the

healthsystemmusthavefacilitiesthatareequipped,

staffed and actually provide EmOC. The composite

natureof this indicator tellsusnotonlywhether the

signalfunctionswereperformedrecently;italsoindi-

rectlytellsusabouttheavailabilityofequipmentand

drugsandtheavailabilityandskillofthestaff.

ThenumberofEmOCfacilitiesrequiredtotreatcom-

plications depends on where facilities are located,

wherepeopleliveandthesizeandcapabilitiesofthe

facilities.Onecouldcountonlyfacilitieswhereallnine

EmOCproceduresareperformed,butthatwouldgive

thewrongmessage,implyingthatonlyhospitalswith

sophisticated equipment and specialist physicians

canreducematernalmortality.Apromising interven-

tionistheupgradingofhealthcentresandothersmall

facilitiestoenablethemtoprovidebasicEmOC(36,

65). The ‘health centre intrapartum care strategy’,

proposed in the Lancet series on maternal health,

suggeststhatallbirthstakeplaceinafacility;thisis

likelytobeoneofthemorecost-efficientstrategiesfor

reducingmaternalmortality,providedthatthequality

ofcareisadequate(101).

A health centre that provides basic EmOC can pre-

vent many maternal and perinatal deaths. For some

conditions (e.g. some cases of postpartum haemor-

rhage), basic care will be sufficient. For other com-

plications (e.g.obstructed labour), higher-level treat-

ment is required.Even then, firstaidcansave lives,

becauseawoman’sconditioncanbestabilizedbefore

sheisreferred.Forexample,awomanwithobstructed

labourcannotbetreated inahealthcentrethatpro-

videsonlybasiccare:sheneedsacaesareansection.


Thechancesofthemotherandhernewbornofsurviv-

ingacaesareansectionare,however,greatlyimproved

ifshedoesnotarriveatthehospitaldehydratedand

infected.Topreventthis,intravenousfluidsandantibi-

oticscanbeadministeredatthehealthcentre,espe-

ciallywhenthetriptothehospital is long.TheWHO

guidelines for primary health care, Pregnancy, child-

birth,postpartum,newborncare:aguidetoessential

practice(98)recommendsthatwomenwithcomplica-

tionsbegiventhefirstdoseofantibiotics,oxytocin,or

magnesiumsulfate(asrequired)beforereferral.

In the previous edition of this document, the rec-

ommended minimum ratio of EmOC facilities to

500000populationwasonecomprehensiveandfour

basic facilities.Since1997,experience inmore than

40 countries has shown that health systems often

haveatleastonecomprehensivefacilityper500000

population and sometimes more. Fully functioning

basicfacilities,however,aremuchlesscommon.On

thebasisof thisexperience, thegroupdecided that

the ratio of one comprehensive to four basic facili-

tiesmightbelessimportantthanhavingatleastone

comprehensive facility and emphasizing the number

offacilitiesper500000population.

Arecentanalysisof24nationalornear-nationalneeds

assessments showed that all but two countries met

theminimumacceptablelevelofonecomprehensive

EmOCfacilityper500000population.Thecountries

includedsomewithhighmaternalmortalityratios,but

theyhadveryfewfullyfunctioningbasicfacilities(102).

IntheUnitedStates(theonlycountrywitharelatively

lowmaternalmortalityratio inwhichtheEmOCindi-

catorshavebeenmeasured),nobasicfacilitieswere

identified,buttherearemanycomprehensivefacilities,

witharatioofonecomprehensivefacilityfor100000

population(66).

ImplicitinthedefinitionofanEmOCfacilityisthatthe

signal functions be available to women at any hour

oftheday,everydayoftheweek.Ifawomanneeds

a caesarean section at midnight on a Saturday, she

should have the same quality of care as a woman

requiringthesameserviceat10:00onaWednesday

morning. The primary obstacle to the provision of

EmOC24h/day,7days/weekinmanycountriesisa

lack of essential cadres of health workers (i.e. mid-

wives, practitioners who can operate anaesthetists

and laboratory technicians). When facilities are not

abletoprovidethesignalfunctions24h/day,7days/

week, local and other management must search for

creativesolutions.Somemayinvolvesimplerotation

ofpersonnel,butothersmay requireapolicy review

ofwhatcadreofproviderisauthorizedandtrainedto

provideEmOC,oradditionalbudgetaryallocations.In

somesituations,accommodationforhealthpractitio-

nershasbeenbuiltonhospitalgroundstoallowcon-

tinuousservice.

Datacollectionandanalysis

Thisindicatordependsontheclassificationofafacili-

ty’sEmOCstatusafterdirectinspection.Often,afacil-

ityisassumedtobefunctioning,butavisitshowsthat

therealityisquitedifferent.Theimportantdistinction

betweenthewayafacilityissupposedtofunctionand

whatitactuallydoesisillustratedbyacasestudyin

Uganda.In2003,theneedforEmOCwasassessed,

inordertoprovidetheGovernmentwithbackground

for drawing up an operational strategy to reduce

maternaldeaths.Withinthehealthinfrastructureplan

in Uganda, district hospitals and health centres IV

shouldbeabletoprovidecomprehensiveEmOC.The

assessmentshowed,however,thatonly21ofthe32

hospitalsassessed(65%)werecomprehensive,while

theother11 functionedat thebasic level.Of the36

healthcentresIVvisited,onlytwo(6%)functionedat

thecomprehensivelevelandanothertwoatthebasic

level. Health centres III theoretically provide basic

EmOC,butonly5(4%)ofthe129assessedfunctioned

attheirintendedlevel.Theresults—particularlywhich

signalfunctionsweremissing—wereusedtoprepare

theannualplan for thesector-wideapproach,which

calledforanationalefforttoimproveEmOC(46).

Incalculatingthisindicator,thenumberoffunctioning

facilitiesiscomparedwiththesizeofthepopulation.

Themostrecentcensusshouldbeusedtodetermine

thepopulationsizeinagivenarea.Ifthelastcensus

ismorethan5yearsold,nationalinstitutesofstatis-

ticsarelikelytohaveprojectionsthatthegovernment

(including the ministry of health) uses for planning.

Recent heavy in- or out-migration might have to be

takenintoconsideration.


The minimum acceptable level for Indicator 1 has

beendefinedinrelationtothepopulationratherthan

birthsbecausemosthealthplanningisbasedonpop-

ulationsize.If,however,itisjudgedmoreappropriate

toassesstheadequacyofEmOCservicesinrelation

tobirths, the comparableminimumacceptable level

wouldbefivefacilitiesforevery20000annualbirths

(includingatleastonecomprehensivefacility).

Ifacountryhasamixofpublicandprivatefacilities,a

decisionmustbemadeaboutwhethertocollectdata

fromallofthemortofocusononesector (generally

thepublicsector).Onlybyincludingtheprivatesector,

however,willtherebeacompletepictureofhowwell

thehealthsystemfunctionsand theoverall levelsof

availability,useandqualityofcare.Becausetheindi-

catorsarebasedonpopulationestimates(totalpopu-

lation,forexample),itmakessensethatallhealthfacil-

ities (ora representativesample)beselected for the

assessment.Themoreacountryreliesonprivatefacil-

itiesforEmOC,themoreimportantitistoincludethe

privatesector.Asanillustrationofthispoint,aneeds

assessmentconductedinBeninin2003showedthat

onefourthoffacilitiesprovidingcomprehensiveEmOC

and almost all the facilities functioning at the basic

levelwereprivatelyoperated(7).

Interpretationandpresentation

If,intheaggregate,acountryorregiondoesnothave

fiveEmOCfacilities(includingatleastonecomprehen-

sivefacility)per500000population,theoverallmini-

mumacceptable levelofEmOCservices isnotmet.

Inthiscase,ahighpriorityistoincreasethenumber

offunctioningfacilitiesuntilatleasttheminimumlevel

is met. This may be done in different ways, e.g. by

upgradingexistingfacilitiesorbuildingnewfacilities,

orsomecombinationofthetwo.

IftheoverallminimumacceptablelevelofEmOCser-

vicesismet, it isreasonabletoconcludethat, inthe

aggregate,anacceptableminimumnumberoffacilities

currentlyexists.Thenextstepwouldbetolookatthe

geographicaldistributionofthefacilities(Indicator2).

We strongly recommend that, in addition to looking

attheratiooffacilitiestopopulation,dataonperfor-

manceof thesignal functionsbepresented in terms

oftheproportionoffacilitiesprovidingeachofthesig-

nalfunctions,asillustratedinFigure1.Suchdataare

extremelyusefulforplanningandsettingprioritiesfor

interventions.Figure1showsthatinBeninin2003,not

allhospitalsthatprovidedobstetricsurgeryalsohad

thecapacitytotransfuseblood.Furthermore,only9%

ofhealthcentresbutalmost90%ofhospitalsremoved

retainedproducts.Today,manual vacuumaspiration

is often used to treat complications of abortion by

mid-levelprofessionalsathealthcentresanddistrict

hospitals (103).Thisprocedurereducestheneedfor

referral,whichoftenentailsconsiderableexpensefor

thefamily,life-threateningdelaysandevendeaths.

Insomecountriescertainsignalfunctionsarevirtually

missingbecausetheyarenotincludedinpre-service

trainingofhealthpersonnelornationaltreatmentpro-

tocols. If a signal function is systematically absent

in a country, it is possible to use the designation

‘Comprehensiveminus1’or‘Basicminus1’asatem-

porarymeasure,whilepoliciesarereviewedandpro-

grammaticinterventionsplannedtoremedythelack.

Supplementarystudies

Reasonsfornotperformingsignalfunctions

Thereareanumberofpossiblereasonsthatahealth

centre or small hospital does not qualify as a basic

EmOC facility. Very often, it is the result of some

management problem. When determining a facility’s

EmOCstatus,consider the following foreachsignal

function:

• Isstaffatthefacilitytrainedandconfidentintheir

skillstoperformtheservice?

• Arethecadresofstaffworkingatthefacilityor

thefacilityitselfauthorizedtoperformthesignal

function?

• Aretherequisitesuppliesandequipmentinplace

andfunctioning?

• Weretherecasesforwhichtheuseofaparticular

signalfunctionwasindicated?


0%88%

81%0%

58%

9%

2%

88%

0%

92%

86%98%

78%

100%

59%

10%

86%

20% 30%

96%

40% 50% 60% 70% 80% 90% 100%

Hospitals (N=48) Health Centres (N=234)

Cesarean delivery

Blood transfusion

Assisted vaginal delivery

Removal of retained products

Manual removal of placenta

Parenteral anticonvulsants

Parenteral oxytocics

Parenteral antibiotics

Figure 1. Proportion of health facilities in which each signal function was performed during the past 3 months,

Benin, 2003

The last explanation refers to the fact that a facility

may have a low caseload, with the result that there

mighthavebeennoneedforoneofthesignalfunc-

tionsduringthe3-monthperiod.Thequestionofcase

load, in turn, could be investigated by determining

whetherthecatchmentpopulation istoosmallgiven

theincidenceofthecomplicationinquestion,ifaccess

is a serious problem for reasons related to informa-

tion,cost,distance,transportorculturalpractices,or

ifbypassingthisfacilityforanother,better-functioning

facilityiscommonpractice.

Whendataonsignalfunctionsarepresentedasshown

inFigure1,itmaybepossibletoseeapatternatthe

countryordistrictlevel,e.g.whetheraparticularsignal

functionisnotbeingperformed.Itwouldbeusefulto

enquire further, for example by discussing the issue

withfacilitystafftolearnwhattheyperceivetheprob-

lemstobe.Thatwillnotelucidatewhywomenuseor

donotuseaparticularfacility;thatkindofinformation

can be derived only from women in the community.

Focus groups are often used to collect this kind of

information.Communitysurveysmightalsobe infor-

mative, but they are more difficult and expensive to

conductthanfocusgroups.

2.2Indicator2:GeographicaldistributionofEmOCfacilities

Description

The second indicator is calculated in the same way

as the first, but it takes into consideration the geo-

graphicaldistributionandaccessibilityof facilities. It

can help programme planners to gather information

aboutequityinaccesstoservicesatsubnationallevel.

To determine the minimum acceptable number of

basic and comprehensive facilities, begin by divid-

ingthesubnational(e.g.provinces,statesordistricts)

population by 500 000. This will give you the mini-

mum acceptable number of comprehensive EmOC

facilities for thesubnationalarea.Then,multiply that

numberby5 tocalculate theoverallminimumnum-

beroffacilities,bothbasicandcomprehensive,forthe

subnationalarea.Tocalculate thepercentageof the

recommended minimum number of facilities that is

actuallyavailabletothesubnationalpopulation,divide

thenumberoffunctioningEmOCfacilitiesbytherec-

ommended number and multiply by 100. A similar

exercisewilldeterminewhatpercentageoftherecom-

mended minimum number of comprehensive EmOC

facilitiesisavailable.

FromMinistèredelaSantéPubliqueduBénin,2003,citedinreferences(6,104).


To determine the percentage of subnational areas

thathavetherecommendednumberofEmOCfacili-

ties(includingtheminimumnumberofcomprehensive

facilities)fortheirpopulationsize,thenumberofsub-

nationalareaswiththerecommendedminimumnum-

berisdividedbythenumberofsubnationalareasand

multipliedby100.


To ensure equity and access, 100% of subnational

areasshouldhavetheminimumacceptablenumbers

ofEmOCfacilitiesoratleastfivefacilities(includingat

leastonecomprehensivefacility)per500000popula-

tion.

Backgroundanddiscussion

Facilities that offer EmOC must be distributed so

thatwomencanreachthem.Iffacilitiesareclustered

aroundacapitalcityoronlyinlargecommercialcen-

tres, women in more remote regions will experience

delayingettingtreatment,whichmightthreatentheir

survival and the survival of their newborns. Table 6

showstheestimatedaveragetimefromonsetof the

majorobstetriccomplicationstodeath.Itcanbeseen

that theaverage time todeath is 12hoursormore,

although postpartum haemorrhage can kill faster.

Therefore,livescouldbesavedatruralhealthfacilities

withinjectableuterotonicsandrehydrationwithintra-

venousfluids.

Complication Hours Days

Haemorrhage•Postpartum•Antepartum

212

Ruptureduterus 1

Eclampsia 2

Obstructedlabour 3

Infection 6

Table 6. Estimated average interval between onset of major obstetric complications and death, in the absence of medical interventions

From Maine, D. Prevention of Maternal Deaths in DevelopingCountries: Program Options and Practical Considerations, inInternational Safe Motherhood Conference. 1987. Unpublisheddata:Nairobi.

In view of the urgency of maternal complications,

EmOC services must be distributed throughout a

country. The distribution can be checked efficiently

by calculating the number of facilities available in

subnationalareas.Ananalysisatregional,state,pro-

vincial, district or other level often reveals discrep-

ancies in health services equity. The ratio of EmOC

facilitiestothetotalpopulationisoftenhigherthanfor

smallergeographicalareas.InNicaraguain2001,for

instance, thecoverageof thecombinedpopulations

of nine administrative regions with comprehensive

EmOCfacilitiesmorethanmettherequiredminimum

(225%).Whentheregionswereexaminedindividually,

however,onlyfourhadtheminimumacceptablelevel

ofcomprehensivecare(102).Aneedsassessmentin

Mauritaniain2000showedthatthenumberanddis-

tributionoffacilitiesprovidingEmOCwerebothinsuf-

ficient. Only eight of the 67 facilities surveyed pro-

videdsuchcare(sevenprovidedcomprehensivecare

and one provided basic care). More than half of all

thecomprehensiveEmOCfacilitieswereinthecapital

city,Nouakchott,and9of13regionshadnoEmOC

facilities(105).

Insomesituations,especiallywherethepopulationis

widelydispersedandtravelisdifficult,itmaybeadvis-

ableforgovernmentstoexceedtheminimumaccept-

ablelevel.InBhutan,forexample,anassessmentof

needsforEmOCrevealedproblemsinthegeographi-

cal distribution of facilities, and the Government

promptlyupgradedfacilitiestoimprovetheavailability

ofcare(Figure2).


Figure 2. Emergency obstetric care facilities in Bhutan

FromUNICEF,DepartmentofHealthServices,andMinistryofHealthandEducation.Semi-annualreportstoAMDD,Jan–June2002&July–Dec2002.Unpublisheddata.2002:Bhutan,citedinreference(104).

Samdrup Jongkha

Resirboo

TrashigangMongar

Trashi Yangtse

Pemagatshel

WangduePhodrang

Zhemgang

Trongsa

BumthangLhuentse

SarpangTsirang

Punakha

Gasa

Thimphu

DaganaChukha

Phuentsholing

Samtse

Haa

Paro

Comprehensive EmOC centres – 4

Basic EmOC centres – 4



Manyofthesameissuesindatacollectionthatexist

for Indicator1arealsorelevant for Indicator2.One

issueis,however,morelikelytoariseinsubnational

thaninnationalcoverage:Howmanyandwhattype

ofEmOCfacilitiesarerecommendedforpopulations

smaller than 500 000? No one answer fits all situ-

ations, but ‘prorating’ would be advised, e.g. if the

populationiscloseto250000,threefacilitieswould

beacceptable(roundingupisthemoreconservative

response).Whetheroneofthethreeshouldbecom-

prehensive depends on the location and proximity

(distanceintermsoftime)ofcomprehensivefacilities

inneighbouringareas.

Emergency obstetric care facilities in subnational

areascanalsobestratifiedbymanagement,todeter-

mine thedistributionofpublic andprivate facilities.

Thisanalysiscanbeparticularlyrevealinginanarea

withprivatebutnogovernmentfacilities,wheregov-

ernmentfacilitiesofferfreeservicesandprivatefacili-

tieschargeuserfees,orwheregovernmentfacilities

chargeandmissionhospitalsarefree.


If subnational geographical areas do not meet the

minimumacceptableratio,underservedareasshould

betargetedandresourcesdevotedtoimprovingthe

availabilityofservices.

The numbers of comprehensive and basic EmOC

facilitiespersubnationalpopulationcanbepresented

ineithertablesormapsonwhichsubnationalareas

areshadedaccordingtothelevelofcoverage(100%

ormoreandatincrementsoflessthan100%).

Supplementarystudy

Indicatorsofaccess toEmOCincludedistanceand

time.Asdigitalmappingandgeographical informa-

tion systems become more widely available, use of

thisindicatorislikelytoincrease.Areasonablestan-

dardfortheavailabilityofservicescanbeestablished,

such as having basic and comprehensive facilities

availablewithin2–3hoursoftravelformostwomen.In

thepast,determiningthedistancebetweenfacilities

and where people live was cumbersome; however,

geographical information systems make calculations

ofdistanceandtraveltimemucheasier,andmeasure-

mentmethodswillbecomemoreconsistent(106).

MapsthatshowtheEmOCstatusoffacilities,thedis-

tanceofcommunitiesfrombasicandcomprehensive

facilities(bothintraveltimeandinrelationtoroadnet-

works), populationdispersionanddensity andother

features that show inequities in terms of access to

carecanbeeffectiveadvocacyandplanningtools.

2.3Indicator3:ProportionofallbirthsinEmOCfacilities

Description

Indicator3istheproportionofallbirthsinanareathat

take place in EmOC health facilities (basic or com-

prehensive).Thenumerator isthenumberofwomen

registeredashavinggivenbirth infacilitiesclassified

asEmOCfacilities.Thedenominatorisanestimateof

all the livebirthsexpected in thearea, regardlessof

wherethebirthtakesplace.

Westronglyrecommendaparallelindicator:thepro-

portionofbirths inall health facilities in thearea,or

‘institutionalbirths’or‘institutionaldeliveries’.Werec-

ommend this in order to give a more complete pic-

tureof thepatternsofuseof thehealthsystem(see

Figure 3). The numerator is always service statistics

fordeliveriesinthefacilities,whilethedenominator—

theexpectednumberoflivebirths—isusuallycalcu-

latedfromthebestavailabledataandbymultiplying

thetotalpopulationoftheareabythecrudebirthrate

of thesamearea.Othermethods forcalculating the

expectednumberoflivebirthscanalsobeused.


Nominimumacceptablelevelisproposed.Inthepre-

viouseditionofthishandbook,theminimumaccept-

able levelwassetat15%ofexpectedbirths. In the

intervening years, many governments have commit-

tedthemselvestoincreasingtheproportionofwomen

whogivebirthinhealthfacilities,andsomeareaiming

for100%.Therefore,theminimumtargetforthisindi-

catorshouldbesetbynationalorlocalgovernments.


Background

Indicator 3 was originally proposed to determine

whetherwomenareusingtheEmOCfacilitiesidenti-

fiedbyindicators1(AvailabilityofEmOCservices)and

2(GeographicaldistributionofEmOCfacilities),andit

servesasacrudeindicatoroftheuseofobstetricser-

vicesbypregnantwomen.Insituationswhererecord

systemsareinadequatetocollectdataforIndicator4

(Met need for EmOC), the number of women giving

birthinhealthfacilitiesisalmostalwaysavailable.Use

ofthesedatacangiveadministratorsaroughideaof

the extent to which pregnant women are using the

healthsystem,especiallywhencombinedwith infor-

mationonwhichfacilitiesprovideEmOC.

Theoptimallong-termobjectiveisthatallbirthstake

place in (or very near to) health facilities in which

obstetric complications can be treated when they

arise. Many countries have made having 100% of

deliveriesininstitutionstheirmainstrategyforreduc-

ing maternal mortality. As they move closer to that

objective, other problems arise. In many countries,

health systems are unable to cope with the added

patientloadwithoutmajorexpansioninfacilitiesand

staff,andmanagershavelimitedinformationonhow

healthfacilitiesarefunctioning.Givingbirthinahealth

facilitydoesnotnecessarilyequatewithhigh-quality

careorfewermaternaldeaths.Smallerhealthfacilities

may not have adequately trained staff, or staff may

nothavetheequipmentor theauthority to treat life-

threateningcomplications.Manyfacilitiesdonotfunc-

tionwellbecauseofpoormanagement,whichshould

beremediedbeforethenumberofbirthsinthefacil-

ityisincreasedgreatly(107,108).Forthesereasons,

the EmOC status of health facilities is included in

Indicator3(ProportionofallbirthsinEmOCfacilities),

andwerecommendthat this indicatorbecalculated

andinterpretedwiththeotherindicators.


Although the name of the indicator is ‘Proportion of

birthsinEmOCfacilities,’inpracticethenumeratoris

thenumberofwomengivingbirthandnotthenumber

ofinfantsborn.Werecognizethatthenumberofbirths

willbeslightlyhigherthanthenumberofwomengiv-

ingbirth,becauseofmultiplebirths;however,theextra

effortneededtocountbirthsratherthanwomengiving

birthmightnotbenecessary,norisitlikelytochange

theconclusionsdrawnfromtheresults.Thenumbers

of women giving birth in facilities are obtained from

healthfacilityrecordsystemsandareoftencollected

for monthly reports to the government. The EmOC

statusofthehealthfacilityinwhichthedeliverytakes

placeisavailablefromtheresultsofroutinemonitoring

orneedsassessmentsunderIndicator1.

Thetotalexpectednumberofbirthsinanareaisbased

oninformationaboutthepopulationandthecrudebirth

rate.Nationalstatisticsofficestendtobasepopulation

projectionsontheresultsoftheirmostrecentcensus.

Theymayalsohaveregionalcrudebirthrates.Ifnot,

the crude birth rate is often available from national

population-basedsurveys,suchasDemographicand

HealthSurveys.Whenpossible,estimatesforthespe-

cific geographical area should be used rather than

applying the national crude birth rate to all regions.

Regions are often selected for interventions or pro-

grammes because of special needs and therefore

tendtohavepoorer indicatorsthanatnational level.

Usually,thebirthrateinpoorerareasishigherthanthe

nationalaverage,sothatuseofthenationalaverage

wouldresultinanunderestimateoftheexpectednum-

berofbirths,andtheproportiondeliveredinfacilities

wouldthereforebeoverestimated.

Parallelanalysisoftheproportionofallbirthsinallthe

facilitiessurveyedallowscomparisonoftheproportion

ofbirthsinEmOCfacilitieswiththeproportionofbirths

inallfacilities.Thisindicatestheextenttowhichother

facilities provide delivery services. Figure 3 shows

that,forexampleinChadallthebirthsinfacilitieswere

inEmOCfacilities,while inBolivia,Mozambiqueand

Senegal,theproportionsofbirthsinnon-EmOCfacili-

tiesadded9–22%.Incontrast,inBenin,onlyasmall

proportionof institutionalbirthsoccurred in facilities

wheremostobstetriccomplicationscouldbetreated.

Thisindicatorcanalsobeanalysedbyleveloffacility

(hospitalandnon-hospital),byownershipormanage-

ment(publicandprivate)andbysubnationalarea,in

order todeterminewherewomenaredelivering.Are

women more likely to deliver in private or govern-

mentfacilities?Aretheremoreinstitutionaldeliveries


Figure 3. Proportions of all births in EmOC facilities and all surveyed facilities

Fromreferences(7,37,56).


Overall, this indicator shows the volume of mater-

nity services provided by facilities. If there appears

tobeunder-use,thereasonsshouldbeexplored.To

increaseuse,emphasisshouldbeplacedonenabling

womenwithcomplicationstouseEmOCfacilities.The

firstgoalofprogrammestoreducematernalmortality

shouldbetoensurethat100%ofwomenwithobstet-

ric complications have access to functioning emer-

gencyfacilities.


Atthelocallevel,additionalstudiestounderstandthe

useofservicesbetterarealmostalwaysuseful.Which

groups of women are using the services? Which

women are not, and why? Clearly, the answers to

thesequestionshaveimportantimplicationsforpublic

healthandhumanrights.

Whichwomenarenotcomingtothefacilities?

Even if the use of health facilities (including EmOC

facilities) is fairly high, it is worthwhile investigating

which women are not using them. Certain factors

strongly affect use of services in a particular area,

suchasdistance to the facility,prevalenceofethnic

orreligiousminoritygroups, levelofeducation(often

an indication of social status), the reputation of the

facilityandpoverty.Informationonsomeofthesefac-

tors, suchas residence,mayalreadybeavailable in

health facility records, and records can be reviewed

todeterminewhetherwomencomefromallpartsof

thecatchmentareaoronlyfromthetowninwhichthe

facilityislocated.Forfactorsforwhichinformationis

notroutinelyrecorded,astudycanbeconducted.For

example,studentsorstaffmemberscanbepostedin

amaternitywardforafewweeksoramonthtorecord

relevant information.Itwouldbeimportant,however,

9% 10%

24%

12% 13%

19%

9%

22%

59%

0%

20%

40%

60%

80%

Chad (2002) Senegal (2001) Bolivia (2003) Mozambique (2000) Benin (2002)

Inst

itutio

nal b

irth

EmOC Facilities Non-EmOC facilities

in certain subnational areas? Disaggregating data in

thiswaycanprovidemorespecificinformationabout

whichinterventionsaremostneeded,andwhere.


totrainandsupervisethesedatacollectorstoensure

thattheyfollowconfidentialityrules,treatpatientsand

theirfamiliesrespectfullyandaskforinformationinan

unbiasedmanner.

Ideally,theprofileofthewomenwhousetheservices

canbecomparedwiththatofwomeninthepopulation

(nationalpopulation-basedsurveys)inordertodeter-

minethecharacteristicsofthewomenwhoareunder-

representedasusersofthefacility(109).

Whydosomewomennotusethefacility?

Oncethegroupsofwomenwhoareunderrepresented

in the facility have been identified, it is important to

findoutwhy.Oneshouldnotassumethattheyknow

thereason,eveniftheyhavegrownupintheregion.If

theassumptioniswrong,any‘correctiveaction’taken

willprobablynotwork(110).Womenshouldbeques-

tioned, either through interviews or in focus groups;

orstudiestocomparesubpopulationscouldbecon-

ducted,afteradjustmentforneedorstatisticalcontrol

forconfoundingfactors.

Variousactivitiescanbeusedtoimproveuse,depend-

ingonthefactorsthatdiscourageit.

• Iffocusgroupdiscussionsshowthatpeoplelack

basicinformationaboutobstetriccomplications,

acommunityeducationprogrammewouldbein

order.Thepreciseformoftheprogrammewould

bedeterminedbylocalcircumstances,butit

shouldbeaimednotonlyatpregnantwomenbut

alsoatthepeoplewhoinfluencetheirdecisionto

seekcare,suchasotherwomenofreproductive

age,partners,mothers-in-lawandtraditionalbirth

attendants.

• IftransportfromavillagetotheEmOCfacilityis

aproblem,thecommunitycouldbemobilizedto

coordinatetheuseofexistingvehicles.

• Ifpoorroadsareabarriertocare,thelocal

governmentshouldbeapproachedtoimprove

them.Ifshortagesofsuppliesorpooroverall

qualityofcaremakepeoplefeelthatgoingtothe

hospitalisnotworththetrouble,solutionstothe

problemsshouldbesought.

• Ifwomenarereluctanttousetheservices

becauseofpracticestheyhavepreviously

experiencedorhaveheardabout,thosepractices

canbediscussedwithstaffatthefacilityto

determinehowthefacilitynormscanbeadapted

tolocalcustomsordesires.

• Ifthecostofservicesisanobstacle,medical

emergencyfundsorinsuranceschemeshave

provensuccessfulinsomeplaces(111).

Whoattendsbirthsinfacilities?

Deliveries in institutionsarenotnecessarilyattended

by skilled birth attendants (112). Therefore, a study

couldbecarriedouttoseewhichcadresofworkers

are involved in deliveries and their level of compe-

tence. Providers could be interviewed to determine

their understanding; observational studies would

allowon-siteverificationofpractices;andretrospec-

tivechartreviewswouldallowanassessmentofthose

aspectsofcarethatshouldbedocumentedoncharts

orpatientrecords.

2.4Indicator4:MetneedforEmOC

Description

‘Met need’ is an estimate of the proportion of all

womenwithmajordirectobstetriccomplicationswho

aretreatedinahealthfacilityprovidingEmOC(basic

or comprehensive). The numerator is the number of

women treated for direct obstetric complications at

emergencycarefacilitiesoveradefinedperiod,divided

bytheexpectednumberofwomenwhowouldhave

major obstetric complications, or 15% of expected

births,duringthesameperiodinaspecifiedarea.The

directobstetriccomplicationsincludedinthisindica-

tor are: haemorrhage (antepartum and postpartum),

prolongedandobstructedlabour,postpartumsepsis,

complicationsofabortion,severepre-eclampsiaand

eclampsia, ectopic pregnancy and ruptured uterus.

(Fortheoperationaldefinitionsofthesedirectobstet-

riccomplications,refertoBox2.)

AswedidforIndicator3,westronglyrecommendthat

metneedbecalculatedatallhealthfacilitiesaswellas

atEmOCfacilities,toprovideamorecompletepicture

oftheuseofthehealthsystemandwherewomenare

beingtreated.


Box 2. Operational definitions of major direct obstetric complications

HaemorrhageAntepartum• severebleedingbeforeandduringlabour:placentapraevia,placentalabruption

Postpartum(anyofthefollowing)• bleedingthatrequirestreatment(e.g.provisionofintravenousfluids,uterotonicdrugsorblood)• retainedplacenta• severebleedingfromlacerations(vaginalorcervical)• vaginalbleedinginexcessof500mlafterchildbirth• morethanonepadsoakedinbloodin5minutes

Prolongedorobstructedlabour(dystocia,abnormallabour)(anyofthefollowing)• prolongedestablishedfirststageoflabour(>12h)• prolongedsecondstageoflabour(>1h)• cephalo-pelvicdisproportion,includingscarreduterus• malpresentation:transverse,broworfacepresentation

Postpartumsepsis• Atemperatureof38°Corhighermorethan24hafterdelivery(withatleasttworeadings,aslabouralone

cancausesomefever)andanyoneofthefollowingsignsandsymptoms:lowerabdominalpain,purulent,offensivevaginaldischarge(lochia),tenderuterus,uterusnotwellcontracted,historyofheavyvaginalbleeding.(Ruleoutmalaria)

Complicationsofabortion(spontaneousorinduced)• haemorrhage due to abortion which requires resuscitation with intravenous fluids, blood transfusion

oruterotonics• sepsisduetoabortion(includingperforationandpelvicabscess)

Severepre-eclampsiaandeclampsia• Severepre-eclampsia:Diastolicbloodpressure≥110mmHgorproteinuria≥3after20weeks’gestation.

Varioussignsandsymptoms:headache,hyperflexia,blurredvision,oliguria,epigastricpain,pulmonaryoedema

• Eclampsia• Convulsions;diastolicbloodpressure≥90mmHgafter20weeks’gestationorproteinuria≥2.

Signsandsymptomsofseverepre-eclampsiamaybepresent

Ectopicpregnancy• Internalbleedingfromapregnancyoutsidetheuterus;lowerabdominalpainandshockpossiblefrom

internalbleeding;delayedmensesorpositivepregnancytest

Ruptureduterus• Uterine rupture with a history of prolonged or obstructed labour when uterine contractions suddenly

stopped.Painfulabdomen(painmaydecreaseafterruptureofuterus).Patientmaybeinshockfrominternalorvaginalbleeding




Asthegoalisthatallwomenwhohaveobstetriccom-

plicationswillreceiveEmOC,theminimumacceptable

level is100%.Governmentsmaywishtoset interim

targets once they have a baseline and they have

embarkedoninterventionstoimprovetheavailability

anduseofsuchcare.

Background

Met need is a more refined measure of the use of

EmOC than Indicator 3 (Proportion of all births in

EmOC health facilities), as it addresses whether the

women who really need life-saving obstetric care

receiveit.

In order to estimate met need for EmOC, one must

firstestimate the totalneed,and thencompare it to

thenumberofwomenwithseriousobstetriccompli-

cationswhoreceiveemergencycareinsuchfacilities.

Thetotalneed forEmOCisestimatedtobe15%of

allbirths,although therehasbeenconsiderabledis-

cussionabouttheexpectednumberofcomplications.

Studieshaveproducedarangeofresults:

• Areviewofstudiesinvariousgeographicalregions

basedonvariousdefinitionsandmethodshave

shownlevelsofmetneedaslowas1%(113).

• Oneprospectivepopulation-basedstudyinsix

WestAfricancountriesshowedthat6%ofpreg-

nantwomenhadseveredirectobstetriccomplica-

tions(114).Theauthorsreportedthattheirfindings

werelikelytobeunderestimatesbecausethedefi-

nitionsofthecomplicationsthattheyusedwere

linkedtomedicalinterventionsthatmightnothave

beenavailableatalltheparticipatingfacilities.In

addition,theyincludedonlydirectobstetriccom-

plicationsoccurringinlatestagesofpregnancy

andomittedcomplicationsofabortionandectopic

pregnancies.

• Asystematicreviewoftheprevalenceofsevere

acutematernalmorbidity(‘nearmiss’events)

basedondisease-specificcriteriashowedaprev-

alenceof0.8–8.2%(113).Reviewedstudiesvaried

intermsoftherangeandseverityofobstetric

complicationsincludedandthetimingofcompli-

cations(intrapartumandpostpartumperiods).

• AprospectivestudyofdeliveriesinIndiashoweda

17.7%incidenceofdirectobstetriccomplications

duringlabour.Thisstudydidnotincludecom-

plicationsoccurringduringpregnancy(suchas

complicationsofabortion),sotheactualpercent-

ageofwomenwithdirectcomplicationswasprob-

ablyhigher.Theauthorsconcludedthat15.3%of

womenneededEmOC,and24%moreneeded

non-emergencymedicalattention(115).

• AsecondstudyinIndiashowedthat14.4%of

deliverieswereassociatedwithseriouscomplica-

tions,butthisstudytoowasrestrictedtocompli-

cationsaroundthetimeofchildbirth(116).

• Astudyofnationaldatafor1991–1992inthe

UnitedStates,acountrywithlowmaternalmortal-

ity,showedatotalof18hospitalizationsforobstet-

ricandpregnancylossper100births(117).These

findingswereconfirmedbymorerecentdata(66).

• Althoughtheresultsvary,thetechnicalconsulta-

tiondecidedtomaintain15%asanaverageesti-

mateofthefrequencyofseriousdirectcomplica-

tionsforthepurposesofestimatingtheneedfor

EmOC.


To calculate met need, information is needed on

women in these facilities who are treated for the

major obstetric complications listed in Box 2. The

definitionswerederived fromWHO (Managingcom-

plicationsinpregnancyandchildbirthandPregnancy,

childbirth, postpartum and newborn care) (95,98)

andtheInternationalFederationofGynaecologyand

ObstetricsSavetheMothersProject.Standardization

ofdefinitionscanbeimprovedbytrainingandsuper-

vision.Thesedefinitionsarecriticalfortraininghealth

workers, enumerators or interviewers who collect

suchdataeitherroutinelyoraspartofanEmOCneeds

assessment.


Routinematernity recordsystems inmanycountries

maynotregisterthe‘reasonforadmission’or‘mater-

nal complications’, although complications can lead

to maternal deaths. Appendix B gives a list of the

informationneededtocalculatetheindicatorsandthe

typesofregistersthatshouldbeconsulted.Italsolists

itemsthatgoodregistersmight include,suchas the

timeofadmissionandthetimeofdefinitiveinterven-

tion,whichareusefulforstudyingtheintervalbetween

admission and emergency caesarean section as an

indicatorofhospitalefficiency(118).

Itislikelythatincompleteorpoorrecordswillbefound

whendataforcalculatingmetneedandsomeofthe

otherindicatorsarecollected,especiallythefirsttime.

Asperiodiccollectionofsuchdatabecomespartof

routineprogrammemonitoring,recordkeepingshould

improve.Thequestioniswhattodowhendatacollec-

tionproblemsareencountered?

Poor recordsusuallybias findings, leading tounder-

estimatesofcomplicationsinfacilities,andthismust

be taken into account in interpreting the data. In

manysituations,thelevelofEmOCbeingprovidedis

so low that, allowing for substantial under-counting,

theresultsdonotchangeverymuch.Figure4shows

actualmeasurementsofmetneedoverseveralyears.

Iftherecordsshowthatonly6%oftheneedforEmOC

is being met in an area and the true proportion is

assumedtobetwiceashigh,themetneedisstillonly

12%.Thissortofchangewillnotalterprogramming.

As recordkeeping improves,however,metneedwill

increaseandthechallengewillbetounderstandthe

attribution:Istheincreaseinmetneedatrueincrease

orisitafunctionofbetterdatacollection?Improved

datacollectionisasuccessinitself,andlongerpro-

grammemonitoringshouldhelpdetermineifthemet

needisreallyincreasing.

Figure 4. Increases in met need for EmOC during AMDD-supported projects (2000–2004)

FromBailey,P.EvaluatingAMDDPhase1:PolicyandServiceImprovements.InDeliveringSaferMotherhoodSymposium-SharingtheEvidence.2007.London,UK:Unpublisheddata.

Countries in Latin America

Countries in North Africa

Countries in sub-Saharan Africa

Countriesin Asia


The most variable element in estimating met need

for EmOC is likely to be complications of abortion.

While it is difficult to gather information on the inci-

denceofunsafeabortions(becausetheyaregenerally

clandestine),theWHOreportUnsafeabortion:global

and regional estimates of incidence of unsafe abor-

tionandassociatedmortalityin2003showedthatthe

frequencyofunsafeabortionsvariesbygeographical

area,fromthreeper100livebirthsinEuropeto29per

100livebirthsinAfrica(119).

Moreover,recordingofabortioncomplicationsishighly

variable, including inaccuracies inwhether theabor-

tion was merely incomplete (which could eventually

lead toacomplicatedabortion)or trulycomplicated

(withhaemorrhageorsepsis)atthetimeoftreatment

oradmission.Insomesettings,noattemptismadeto

distinguishbetween the two.Thus,complicationsof

abortionmightactuallybeover-reported.Thedefini-

tiongiveninBox2coversonlythoseabortioncompli-

cationsthatincludehaemorrhageorsepsis.

Itwouldnotbeappropriate,however,toexcludeabor-

tionsfromthecalculationofmetneed,ascomplica-

tionsofabortionareamajorcauseofmaternaldeath

insomecountriesandregions.Forexample,inLatin

AmericaandtheCaribbean,12%ofmaternaldeaths

areattributabletocomplicationsfromabortion(120).

Given the reporting difficulties, analysts presenting

dataonmetneedshouldstateexplicitlywhat types

ofabortiontheyhaveincludedandconsiderconduct-

ingstudiestoexaminethesubjectingreaterdetail.If

itissuspectedthatabortionswithoutseriouscompli-

cations(i.e.withouthaemorrhageorsepsis)arebeing

recordedas‘obstetriccomplications’,itmightbeuse-

fultocalculateandreportmetneedwithandwithout

abortions,forcomparison(88).

Afrequentlyaskedquestionisthepossibilityofover-

reportingdueto‘double-counting’ofwomenwhoare

admittedtomorethanonefacility,asinthecaseofa

referral,orwhoareadmittedtothesamefacilitymore

than once during a pregnancy. We recommend that

referralsbecountedatthefacilityatwhichthewomen

receive definitive treatment. A study in Thailand

showedthatmetneedwasinflatedby16%because

ofdoublecountinganddroppedto96%onceithad

beenadjustedfor(90).Ifthereisconcernaboutdou-

blecountingand itseffectonmetneed,we recom-

mendthatastudybedesignedtomeasuretheeffect.

Theresultsofthisspecialstudycanthenbetakeninto

accountwheninterpretingthegeneralfindings.

Manyhealthfacilities,ofcourse,performsomebutnot

allofthebasicEmOCsignalfunctions.Asthesefacili-

tiesmaywellavertsomematernaldeaths,werecom-

mendthatmetneedinbothEmOCfacilitiesandinall

thefacilitiessurveyedbecalculated.Evenwhenmany

facilitiesdonotperformafewsignalfunctions,itisstill

important to find out how many obstetric complica-

tionstheymanage.


Iftheminimumacceptablelevelforthisindicatorisnot

met,i.e.islessthan100%,somewomenwithcompli-

cationsarenotreceivingthemedicalcaretheyneed.

This is likely tobe thenormwherematernalmortal-

ity is high. If there are adequate numbers of EmOC

facilities,womengivebirth inthosefacilitiesandthe

metneedislessthan100%,thenationalprioritymust

be to improve use of the facilities by women with

complications. Depending on the situation, strate-

giesformeetingthisobjectivecouldincludeimprov-

ingthequalityofcareatfacilities,eliminatingbarriers

toseekingcare(e.g.transportorcost)andeducating

the community to recognize complications and the

importanceofseekingcare.Metneedmayalsobelow

becauseobstetriccomplicationsarepoorlyrecorded

inregisters.Inthiscase,itisadvisabletostudyrecord

keepingatthefacility(seediscussionaboveand‘sup-

plementarystudies’below).

If the met need is close to 100%, one might ask

whatdefinitionofabortion isused,because it isnot

uncommonformetneedtoexceed100%ifallabor-

tions(incomplete,missed,spontaneous,induced)are

includedinthenumerator.Ifthatisnotthecase,itis

reasonable toconclude thatmostwomenwhoneed

EmOCservicesarereceivingthem.Asdiscussedear-

lier, since the true incidence of complications in the

populationmightbegreater than15%, it ispossible

thateven ifmetneed is100%therearestillwomen

whoarenot receivingthe life-savingEmOCservices

theyneed.Forthisreasonalone,thelevelofmetneed

mightbegreaterthan100%.Thisshouldnotbeinter-


pretedasbeingduetofaultydata,e.g.over-diagnosis

ofcomplications; it ispossible that thegeographical

distribution of EmOC facilities is uneven, and met

need exceeds 100% because women from outside

thecatchmentareacometothefacility.Liketheques-

tionofdoublecounting,astudyofwhousesthefacil-

itycouldhelpexplainametneedhigherthan100%.

Wheninterpretingtheindicators,itishelpfultolookat

indicators3and4atthesametime.


Whilemetneed forEmOC isagaugeof the levelof

suchcareinanarea,itdoesnotshowwhatisrequired,

andalowmetneedcannotindicatewheretheprob-

lemlies.Itmightbeduetounder-recordingofcompli-

cationsortooneofmanyfactorsthataffecttheuseof

services,andfurtherinvestigationisrequired.

Itisimportanttoensurethatwomenfromallthecom-

munitiesintheareaaretreatedatthefacility.(Seethe

section on additional studies under Indicator 3 for

morewaysofexploringthistopic.)Studiestoaddress

twoquestionswouldprovideadeeperunderstanding

ofwho is included inmetneedandhow theyaffect

thisindicator:

• Howmanywomenhavecomplicationsafterthey

wereadmittedtohospital,andwhichcomplica-

tionswerethey?

• Howmanywomenareadmittedwithsigns

andsymptomsofcomplications,andwhich

complicationswerethey?

When women with complications are stabilized at a

lower-level facilitybefore referral to ahigher level of

care, we suggest that they be counted only at the

facilitywheretheyreceivedefinitivetreatment.There

is no easy mechanism for finding out whether a

referred woman reaches her destination. A study of

the women referred, their treatment before referral,

theircompliancewithreferralandtheirdefinitivetreat-

mentwouldelucidatetheeffectofdoublecountingon

metneedandwouldalsoshowhowwellthereferral

systemfunctions.Inthefield,staffatlowerlevelshas

argued in favourofcounting thesewomen twice,as

they claim that they too have treated them, usually

bystabilization.Toraisemorale,programmemanag-

ers might consider counting them twice, and with a

studyof referrals theycanalsodocument theeffect

ofdoublecountingonmetneedandmakeanyneces-

saryadjustments.

Several typesofstudycouldbeused toexplore the

qualityofrecordkeepingatafacility:

• Examinehowrecordsarekept.Doessomeone

entercomplicationsintotheregister24h/day,or

doestheseniornursedocumentthemonlyonce

adayfromverbalreportsbyotherstaff?This

practicecouldleadtoseriousunderreporting.

Discussionswithstaffaboutrecentcasescanpro-

videinsightintohowrecordsarekept.

• Comparethecomplicationsrecordedinthemater-

nityregisterwithpatientcharts,operatingtheatre

registersoremergencyadmissionslogbooks.

Whatproportionofseriouscomplicationsisnot

reportedintheregisterthatisusuallyusedforcal-

culatingmetneed?Whichcomplicationsappear

tobemostunderreported?Howdoyourfindings

changewhenyoucorrectforthisunderreport-

ing?Howoftendoesadiagnosisofcomplication

changebetweentheadmissionsregisterandthe

operatingtheatreregister?

• Examinehowabortioncomplicationsarerecorded

bydiscussingtherecordsandcasenotes

withstaff.Areminorcomplications,orevenall

incompleteabortions,countedas‘complications’?

Remember,forcalculatingmetneed,onlyserious

complications,suchascomplicationsofabortion

withsepsisandhaemorrhage,arecounted.

• Formoredetailedmonitoringofabortion

complications,werecommendasetof‘process

indicatorsforsafeabortion’,whichinclude

11signalfunctionsthatdefinebasicand

comprehensivecare.LiketheEmOCindicators,

thesafeabortionindicatorsmeasurethe

availability,distribution,useandqualityofsafe

abortionservices(121-123).

• Knowingmoreabouthowwellandhowcompletely

logbooksarekeptupcanidentifyproblems.

Investigatewhetherstafftrainingorsupervisionof

recordkeepingreducesunderreportingovertime,

andthendisseminateyourresults.


2.5Indicator5:Caesareansectionsasaproportionofallbirths

Description

Theproportionofalldeliveriesbycaesareansection

inageographicalareaisameasureofaccesstoand

use of a common obstetric intervention for averting

maternalandneonataldeathsandforpreventingcom-

plicationssuchasobstetric fistula.Thenumerator is

thenumberofcaesareansectionsperformedinEmOC

facilities for any indication during a specific period,

and thedenominator is theexpectednumberof live

births(inthewholecatchmentarea,notjustininstitu-

tions)duringthesameperiod.

Occasionally, hospitals in which caesarean sections

areperformedlackoneofthebasicsignalfunctionsof

EmOCanddonotqualifyassuchafacility.Therefore,

asforindicators3and4,werecommendthatthisindi-

cator be calculated for both EmOC facilities and all

facilities.

Minimumandmaximumacceptablelevels

Bothverylowandveryhighratesofcaesareansection

canbedangerous,buttheoptimumrateisunknown.

Pending further research, users of this handbook

mightwanttocontinuetousearangeof5–15%orset

theirownstandards.

Background

The proportion of births by caesarean section was

chosenastheindicatorofprovisionoflife-savingser-

vicesforbothmothersandnewborns,althoughother

surgicalinterventions(suchashysterectomyforarup-

tureduterusorlaparotomyforanectopicpregnancy)

can also save maternal lives. Of all the procedures

used to treat major obstetric complications, caesar-

eansectionisoneofthecommonest,andreportingis

relativelyreliable(124).

Earliereditionsofthishandbooksetaminimum(5%)

and a maximum (15%) acceptable level for caesar-

eansection.AlthoughWHOhasrecommendedsince

1985 that the rate not exceed 10–15% (125), there

isnoempiricalevidence foranoptimumpercentage

or rangeofpercentages,despite agrowingbodyof

research that shows a negative effect of high rates

(126-128).Itshouldbenotedthattheproposedupper

limitof15%isnotatargettobeachievedbutrathera

thresholdnottobeexceeded.Nevertheless,therates

inmostdevelopedcountriesandinmanyurbanareas

of lesser-developedcountriesareabove that thresh-

old.Ultimately,whatmattersmost is thatallwomen

whoneedcaesareansectionsactuallyreceivethem.

Thetechnicalconsultationfortheseguidelinesnoted

thedifficultyofestablishingalowerorupperlimitfor

theproportionofcaesareansectionsandsuggested

that a lower limit of 5% is reasonable for caesar-

eansperformedforbothmaternalandfetal reasons.

If elective or planned caesarean sections and those

performedforfetalindicationswereexcluded,alower

rangewouldbeindicated;however,therecordsystem

maynotalwaysregistertheindicationfortheoperation

andsuchprecisionisusuallynotavailable.Adetailed

analysisofthereasonsforcaesareansectioninahos-

pitalwouldbeworthwhile.

Wherematernalmortalityishigh,therateofcaesarean

sectionstendstobe low,especially in ruralareas.A

recentreviewofglobal,regionalandnationalratesof

caesareansectionshowedthatthelowestrate(3.5%)

wasinAfrica;inthe49least-developedcountries,the

ratesrangedfrom0.4%inChadto6%inCapeVerde

(oranaverageof2%)(129).Figure5showshowlow

ratesofcaesareansectioninseveralcountriesofAsia

andinsub-SaharanAfricachangedafterseveralyears

ofinterventionstoimproveEmOC.

Despite the clear inverse relation between very high

maternal mortality and low rates of caesarean sec-

tion,thisprocedure(likeanymajorsurgery)carriesa

risk for surgical or anaesthetic accident, postopera-

tiveinfection,andevendeathforthepatient(129).A

uterine scar increases the risk for uterine rupture in

futurepregnancies.Whereconditions inafacilityare

particularly precarious, the case fatality rate among

womenwhoundergocaesareansectionscanbeunac-

ceptably high, as found by the Network for Unmet

ObstetricNeedinBenin,BurkinaFaso,Haiti,Maliand

Nigerin1998and1999.1Therisksshouldbeweighed

against the potential benefits of the surgery. In the

1deBrouwereV.Personalcommunicationaboutcasefertilityratesforcaesareans,2006.


Figure 5. Caesarean sections as a proportion of births in AMDD-supported projects (2000–2005)

FromBailey,P.EvaluatingAMDDPhase1:PolicyandServiceImprovementsinDeliveringSaferMotherhoodSymposium-SharingtheEvidence.2007.London,UK:Unpublisheddata.

caseoftransversefetallie,whenexternalversionfails

or isnotadvisable, thebenefitsof surgerydefinitely

outweightherisks.Withoutacaesareansection,most

women with obstructed labour will either die or be

severelymaimed(130).Acaesareansectionisthekey

interventionforpreventingobstetricfistulacausedby

prolongedorobstructedlabour,makingthisindicator

animportantmeansformeasuringprogressinthepre-

ventionofthiscondition.

Manyobserversconsiderthatweareexperiencinga

worldwideepidemicofoveruseofcaesareansection

(131)and that the rateswillcontinue to rise, inview

ofpractitioners’andadministrators’ fearof litigation,

localhospitalcultureandpractitionerstyleaswellas

increasingpressure fromwomen inhighly industrial-

izedcountriestoundergocaesareansectionsfornon-

medicalreasons(132,133).Atthesametime,evidence

forthenegativeconsequencesofcaesareansectionis

increasing:recentstudiesincountrieswithhighrates

suggestthatcaesareansectioncarriesincreasedrisks

for maternal and neonatal morbidity and mortality

(126-128).


Whiledataontherateofcaesareansectionscanbe

collectedinpopulationsurveys,suchasdemographic

and health surveys, data for this indicator are col-

lectedfromhospitalrecords(134),asratesbasedon

service statistics are considered more precise than

population-based rates,which tend tobemarginally

higher than those based on health facility records

(124).Facilitydataarecollectedroutinelyfromoperat-

ingtheatrelogbooks,whichareoftenthemostcom-

pleterecordsavailable.

Thenumeratorforthisindicatorcoverscaesareansec-

tionsperformedforallindications,includingthosefor

maternalandneonatalreasons,aswellascaesarean

sectionsperformedinemergenciesandthosethatare

plannedorscheduled.

Throughoutthediscussionoftheindicators,wehave

stressed the importance of including data from all

types of facilities. In countries or regions where the

privatesectorplaysamajorroleindeliveringobstet-

s


ricservices,therateofcaesareansectionwillbepar-

ticularly sensitive to inclusion of such hospitals. For

instance,inLatinAmericaandAsia,theproportionof

caesareansectionsishigherinprivatethaninpublic

facilities.InElSalvador,roughlyone-halfofallcaesar-

eansectionsareperformedoutsidethepublicsector,

throughtheprivatesectorandsocialsecurityhospi-

tals(135).Thisraisesthepossibilitythatsomeofthese

operationsareperformed(ornot)forfinancial,rather

thanmedical,reasons.

Acommonmisunderstandingofthisindicatoristhatit

referstotheproportionofdeliveriesinahospitalthat

are performed by caesarean section, i.e. the ‘insti-

tutional caesarean section rate’ or the proportion of

deliveries in the facility that are done by caesarean

section.Theinstitutionalcaesareanrateisdifficultto

interpret, because it depends on the patients in the

hospital(Isthehospitalaregionalreferralhospitalthat

receivesmanycomplicatedcases?Or is it adistrict

hospital,wheremostcomplicatedcasesarereferred

further?)aswellastheskills,preferencesandhabitsof

theproviders.Thepopulation-basedindicatorrecom-

mendedheregivesanoverviewofthelevelofprovi-

sionofthiscriticalserviceinageographicalregion.

Toreducethepossibilitythatthis indicatorwillmask

inequitiesinaccesstoanduseofcaesareansection,

we strongly encourage authorities to look closely at

their data. For instance, in Morocco, Peru and Viet

Nam, the national rates of caesarean section are

5–15%,but thenationaldatamaskthehighrates in

majorcitiesandtheverylowratesinruralareas.The

rangeofpatternsisshowninTable7.

Table 7. Population rates of caesarean section from Demographic and Health Surveys among women who gave birth within three years of the survey.

Region Country YearRateofcaesareansection

Total Urban Rural

LatinAmerica DominicanRepublicPeru

20022000

33.112.9

36.221.0

27.23.2

South-EastAsia BangladeshNepalVietNam

200420012002

4.51.09.9

13.75.022.9

2.20.77.2

Africa EthiopiaKenyaMoroccoZambia

20002003

2003–20042001–2002

0.64.35.62.2

5.29.59.34.4

0.13.01.91.2

Fromreference(134).

Anotherexampleof inequitableaccesstocaesarean

sectionispresentedinFigure6.Ronsmansetal.used

demographic and health survey data to show the

rangeofratesbywealthquintilein13countrieswith

nationalratesof2.0–4.9%(136).Thisanalysisshows

thatthepoorestwomenhavelessaccesstothislife-

savingprocedure.


Figure 6. Rates of caesarean section by wealth quintile in 13 countries with national rates between 2.0%

and 4.9%

Reproduced,withpermission,fromreference(136).


Whenlessthan1–2%ofbirthsarebycaesareansec-

tion, there is littledoubt thatpregnantwomenhave

pooraccesstosurgicalfacilities.Ratesinthisrange

arecommoninruralsub-SaharanAfricaandinsome

countriesofSouthAsia(Figure6andTable7).Where

caesareansectionratesareverylow,mostareprob-

ably done for maternal emergencies; as the rates

increase,agreatersharemaybefor fetalemergen-

cies.Asthenumberofcaesareansectionsincreases,

the uncertainty between these classifications also

increases(137).


Whohascaesareansectionandwhere?

Studiesoncaesareansectionsshouldincludethepro-

portionsofbirthsinurbanandruralareas,aswellasin

smalleradministrativeorgeographicalunits.Variables

that are used to measure equity, such as economic

Rwanda 00 Bangladesh 04

Côte d’lvoire 98 Uganda 00 Pakistan 90

Mozambique 02 Kenya 03

Indonesia 02 Tanzania 99

15

16

17

18

19

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0Poorest quintile Poor quintile Middle quintile Richer quintile Richest quintile

Per

cent

of d

eliv

erie

s b

y ca

esar

ean

Malawi 00 Ghana 03 Benin 01

Comoros 96

Key: Country and year data


quintiles, ethnicity and education, can be used to

reveal where access to services is limited. Another

method for understanding data on caesarean sec-

tionsisinvestigatingthetypeofhospital(e.g.publicor

private)wherecaesareansectionsareperformed,as

thiscanindicatehowthevariouscomponentsofthe

healthsysteminteract.

Indicationsforcaesareansection

The final responsibility for ensuring that caesarean

sectionisperformedonlywhennecessaryiswithcli-

nicians. The chief medical officer or the head of an

obstetricsandgynaecologydepartmentinahospital

shouldreviewtheindicationsforthecaesareansthat

are performed. One approach is to look at the pro-

portion performed for absolute maternal indications,

which would almost certainly lead to the woman’s

deathifuntreated,includingsevereantepartumhaem-

orrhageduetoplacentapraeviaorplacentalabruption,

majorcephalo-pelvicdisproportion,transverselieand

browpresentation(138).Anotherapproachistoiden-

tifycaesareansectionsthatareperformedformater-

nalandforfetalindications,andathirdapproachisto

usetheRobsonclassificationsystem,whichrelieson

thecharacteristicsofwomenwhohavehadcaesarean

sections(139).Theclassificationsortswomeninto10

mutuallyexclusivegroupsonthebasisofparity,previ-

ousobstetrichistory,thecourseoflabouranddelivery

andgestationalage (140). It canbeused to identify

womenwhohavehadcaesareansectionsforreasons

otherthanasaresponsetoanimminentemergency.

Whoperformscaesareansections?

When the level of Indicator 5 is under the recom-

mendedminimum,poorlyfunctioninghealthfacilities

maybeacontributing factor.Thisoften results from

factorssuchaspostingsandtransfersofkeystaffor

arealshortageofhealthprofessionalstrainedtoper-

formthis life-savingservice.Studiescanbedoneto

investigatewhether this indicator isaffectedby lack

ofhumanresources.Forexample,ananalysisofwho

is trained and authorized to provide caesarean sec-

tionsmaybe informative. Incountrieswhereasmall

groupofhealthprofessionals,primarilybasedatfacili-

ties in largeurbancentres,are theonlypractitioners

abletoprovidecaesareansection,astrategymustbe

devisedtoaddressshortagesofhealthprofessionals

inruralareas.Onestrategythathasbeensuccessfully

usedinMalawi,MozambiqueandtheUnitedRepublic

ofTanzaniaistotrainmid-levelproviders(e.g.clinical

officers, assistant medical officers) to perform cae-

sarean sections (141–144). Similarly, in India, a new

programme under the auspices of the Government

and the obstetrics society is training doctors with a

Bachelor’sdegreeinmedicineandsurgeryincompre-

hensiveEmOC,includingcaesareansection(145).

Qualityofcare

Training,supervisionand leadershipbyseniorphysi-

ciansareimportantinmaintainingstandards.National

societies of obstetrics and gynaecology should

encourage the use of evidence-based protocols. In

facilitiesatalllevels,routineclinicalauditscanbeused

tomonitor change, improvepracticeandmaintaina

good quality of care; several tools exist to facilitate

this process (146–148). The infection rate in women

whohaveundergoneobstetricsurgeryisanotherindi-

catorofthequalityofcare.

Unmetobstetricneed

The indicator ‘Unmet obstetric need’ is unrelated to

Indicator4(MetneedforEmOC).Itdescribestheneed

forobstetricsurgeryforabsolutematernalindications,

whileIndicator4encompassesallthedirectobstetric

complicationstreatedwiththeEmOCsignalfunctions,

whicharebothsurgicalandnonsurgical (e.g.paren-

teralanticonvulsants,uterotonicdrugs).Theindicator

forunmetobstetricneedreferstotheneedforobstet-

ric surgery, including hysterectomy or laparotomy,

in addition to caesarean section. Caesarean section

constitutes most obstetric surgical procedures. This

indicator focusesstrictlyonmaternal life-threatening

conditions for which major obstetric surgery is per-

formed.Itisintendedtohelphealthpersonnelanswer

thequestions:

• Arepregnantwomenreceivingthemajorsurgical

obstetricinterventionstheyneed?

• Howmanywomen’sneedsareunmet?

• Wherearethosewomenwhoseneedsareunmet?

Box3providesdetailedinformationonthisindicator.


Box 3. Indicator of unmet obstetric need

Unmetobstetricneedisanestimateofthenumberofwomenneedingamajorobstetricinterventionforlife-threateningcomplicationswhodidnothaveaccesstoappropriatecare.Thisindicatorisparticularlyappropriateforidentifyinggeographicalorsocialinequityinaccesstohospitalcare.

Theconcept

Theconceptofunmetobstetricneedisthedifferencebetweenthenumberofwomenwhoneedobstetricsurgeryandthenumberofwomenwhoareinfactcoveredbyhealthservices.

Theindicatorisrestrictedtoabsolute(life-threatening)obstetricindicationsthatrequireobstetricsurgery(caesareansection,hysterectomy,laparotomy)orinternalversionandcraniotomy.Astandardlistofsuchindicationswasdrawnuponthebasisofthedegreeofseverityoftheindication,therelativestabilityofitsincidenceandrelativelyreproduciblediagnosis.Itcomprises:

• antepartumhaemorrhageduetoplacentapraeviaorabruptioplacenta;

• abnormalpresentation(transverselieorshoulderpresentation,facewithpersistentmento-

posteriorpositionorbrowpresentation);

• majorfeto-pelvicdisproportion(e.g.mechanicalcephalo-pelvicdisproportion,smallpelvis

includingpre-ruptureandruptureofuterus);and

• uncontrollablepostpartumhaemorrhage.

Inmostsituations,theincidenceofobstetricneedisnotknownprecisely.Abenchmarkcanbeusedtoestimatethenumberofwomenwithabsolutematernalindications,whichis1.4%(95%confidenceinterval,1.27–1.52),themedianforfivesub-SaharanAfricancountries,Haiti,MoroccoandPakistan(http://www.uonn.org/uonn/pdf/engintc00.pdf).Multipliedbythenumberofexpectedbirthsinanarea,thisgivestheestimatednumberofwomenwithabsolutematernalindicationsinthearea.Thesecondelementoftheequation—thenumberofmajorobstetricinterventionsactuallyperformedforabsolutematernalindications—isthesumofallsuchinterventionsperformedinthepopulationofwomeninthearea,wherevertheinterventiontookplace(privateorpublicsector,inoroutsidethedefinedarea).Thedifferencebetweenthenumberofwomenwithabsolutematernalindicationsandthenumberofmajorobstetricinterventionsactuallyperformedforthoseindicationsistheunmetneed.

Example:IntheruralpartofdistrictX,20000birthsareexpectedin2007.Thenumberofmajorobstetricinterventionsforabsolutematernalindicationsisestimatedtobe1.4%(benchmark)x20000=280interventions.Whenallpublicandprivatecomprehensiveemergencyobstetriccarefacilitieshadbeenvisited,thetotalnumberofmajorobstetricinterventionsperformedforabsolutematernalindicationswas84.Theunmetneedwasthus280–84=196,oranunmetneedof70%.Thismeansthat196womendidnothaveaccesstonecessarylife-savingsurgery.

Foradditionalinformationandformsusedtoconstructthisindicator,seethewebsiteoftheunmetobstetricneedsnetwork,www.uonn.org.

Observednumber

ofmajorobstetric

interventions

performedfor

absolutematernal

indications

Estimated

numberof

absolutematernal

indications

Estimatednumber

ofwomenwho

neededbutdidnot

receivelife-saving

care

=–


2.6Indicator6:Directobstetriccasefatalityrate

Description

Thedirectobstetriccasefatalityrateistheproportion

of women admitted to an EmOC facility with major

direct obstetric complications, or who develop such

complications after admission, and die before dis-

charge.Weincludeallsevenmajorobstetriccompli-

cationslistedinBox2.

Thenumeratoristhenumberofwomendyingofdirect

obstetric complications during a specific period at

anEmOCfacility.Thedenominator is thenumberof

womenwhoweretreatedforalldirectobstetriccom-

plicationsatthesamefacilityduringthesameperiod.

In general, the denominator for the direct obstetric

casefatalityrateisthenumeratorformetneed.

Like indicators 3–5, the direct obstetric case fatal-

ityrateshouldbecalculatedforall facilities,not just

EmOC facilities. It is usually calculated at individual

facilitiesandacross facilities,especially thoseof the

sametype,suchasdistricthospitals.

Maximumacceptablelevel

Themaximumacceptablelevelislessthan1%.

Background

Afterdeterminingtheavailabilityanduseofservices,

thenextconcernisqualityofcare,whichisthesubject

ofagrowing,complexliterature.ThesetofEmOCindi-

catorsincludesthedirectobstetriccasefatalityrateas

a relativelycrude indicatorofquality.Thisshouldbe

supplementedwithmoredetailedassessments.

Intheearliereditionsofthispublication,thisindicator

wassimplycalledthe‘casefatalityrate’.Ithasbeen

renamed ‘Direct obstetric case fatality rate’ for the

sakeofclarityandbecauseanewindicatorhasbeen

addedforindirectobstetriccomplications.

Researchershavegainedsubstantialexperiencewith

this indicator in thepast10years.Periodicmonitor-

ing(every6–12months)hasbeenthenormwhenthe

EmOCindicatorsareusedroutinely(15,36,65).The

available data, an example of which is presented in

Table8,indicatesthatsubstantialreductionsarepos-

sible within 3–5 years, if not sooner, with improved

quality of obstetric care. The direct obstetric case

fatality rate in thesestudies ranged fromalmost2%

to 10%, whereas an analysis of application of the

EmOC indicators to data from the United States in

2000 showed a direct obstetric case fatality rate of

0.06%(66).

Table 8. Direct obstetric case fatality rates before and after interventions to improve emergency obstetric care

Setting Beforeinterventions Afterinterventions Reduction

Ayacucho,Peru(2000–2004,fivefacilities)

1.7% 0.1% 94%

Gisarme,Rwanda(2001–2004,threefacilities)

2.0% 0.9% 55%

Mwanza,UnitedRepublicofTanzania(2000–2004,fourfacilities)

3.0% 1.9% 37%

Sofala,Mozambique(2000–2005,12facilities)

3.5% 1.7% 51%

Oromiya,Ethiopia(2000–2004,threefacilities)

10.4% 5.2% 50%



Giventherange,1%wouldappeartobeareasonable

maximumacceptable level, fallingbetweentherates

for less and more developed countries. The post-

interventionratesinTable8showthatitispossibleto

reduceahigh rate tobelow1%;however,countries

thatreachthisbenchmarkshouldstrivetoreducethe

rateevenfurther.Sometimes,circumstancesbeyond

thecontrolofhospitalmanagersmaymakeitdifficult

to achieve a rate below 1%. If few facilities provide

basicandcomprehensiveEmOC,womenwithcom-

plicationsarelikelytoarriveatthehospitalafteralong

journey, jeopardizing their survival. There are never-

theless low-costwaysto improvethequalityofcare

and to reduce the direct obstetric case fatality rate

progressively.


The direct obstetric case fatality rate can be calcu-

lated forany facility that treatscomplications,expe-

riences maternal deaths and has adequate records

on both these events. The same issues in collect-

ing data on major direct obstetric complications for

met need apply, although new issues arise for the

collection of information on the number of maternal

deaths. Maternal deaths are notoriously underesti-

mated because of misclassification or underreport-

ing,sometimesoutoffearofrebukeorreprisal(149).

Bothdeathsandcomplicationsshouldbethoroughly

soughtinallwardswhereadultwomenareadmitted,

notonlytheobstetricward.

Weencouragecalculationofseparatecause-specific

fatality rates for eachof themajor causesofmater-

naldeath.Treatmentofsomecomplications,suchas

obstructed labour, may improve more rapidly than

others,suchaseclampsia.Cause-specificcasefatal-

ityratesindicatewhereprogresshasbeenmadeand

whereithasnot(36).Thenumberofmaternaldeaths

in a given facility or aggregate of facilities is, how-

ever,oftentoosmall(e.g.fewerthan20)tocalculate

astablerateforeachcomplication.Therefore,inmost

facilities,onlyanaggregatedirectobstetriccasefatal-

ityratewillbecalculated.

There are good reasons for using this indicator for

individualfacilities,forallfacilitiestoreflectthestate

of the health system, or for a subset of facilities in

that system (see Table 11 in section 2.9). Averaging

theratesforallfacilitiesisonecrudemonitoringmea-

sure,but itdoesnotshowwhichfacilitiescontribute

mostheavily to thedirectobstetriccase fatality rate

and therefore where interventions are most needed.

Toidentifythosefacilitiesorregionsthatneedgreater

attention, data from various types of facilities (or in

differentareas)canbeanalysedseparatelyandthen

combined.


Direct obstetric case fatality rates do not take into

accountdeathsoutsidethehealthsystem.Thisdoes

notaffectthevalueoftheindicator,becauseitisused

onlytomeasuretheperformanceoftheEmOCfacility.

Iftheindicatorsoftheavailabilityoffacilities,thepro-

portionofbirthsinfacilitiesandmetneed(indicators

1–5)showthatEmOCservicesarewelldistributedand

well usedand thedirectobstetriccase fatality rates

are low, it issafe tosay that thematernalcaresys-

tem in thecountry isworking fairlywell. If,however,

thedirectobstetriccasefatalityrateisacceptablebut

EmOCcoverageormetneedisinsufficient,theimpli-

cation is that women who deliver in EmOC facilities

arelikelytosurvivebutmaternaldeathsoutsidehealth

facilitiesmightstillbecommon.

Comparisons of direct obstetric case fatality rates

amongindividualfacilitiescanbedifficulttointerpret

whenthefacilitiesarenotcomparable.Forexample,it

maynotbevalidtocomparetherateinadistricthos-

pitalwiththat inateachinghospital,aswomenwith

themostseriouscomplicationsmaybereferredtothe

teachinghospitalatthelastmoment,wheretheydie.

Thisdifferencewould lowerthedirectobstetriccase

fatality rateat thedistricthospitalandraise itat the

teachinghospital.

Thedirectobstetriccasefatalityrate inafacilitycan

exceedthemaximumacceptablelevelforseveralrea-

sons.Inmanycases,thequalityofcareisinadequate;

however,theremaybeotherexplanations.Forexam-

ple,longdelaysinreachingEmOCfacilitiescanresult

inapoorstatusonarrival;orafacilitywithahighdirect

obstetriccasefatalityratemightbetheend-pointof


thelocalreferralchain,sothatwomenwiththemost

seriouscomplicationsaresentthere.Itisalsoimpor-

tant to consider the number of women counted in

calculatingthedirectobstetriccasefatalityrate.Ifthe

rate is based on a small number of women, even a

singledeathcancreateadeceptively large increase.

Giventheproblemsofinterpretingsmallnumbers,the

directobstetriccasefatalityrateismostusefulatdis-

trict levelorathigh-volumefacilitieswherethereare

manymaternaldeaths.Therefore,theseratestendto

becalculatedonlyatcomprehensiveEmOCfacilities.

Theoccurrenceofsomematernaldeathsinafacility

canindicatethatwomengotherefortreatmentofcom-

plications;conversely,theabsenceofmaternaldeaths

mightindicatethatwomenwithseriouscomplications

arenotbroughtthereorareroutinelyreferredon,even

when they should be treated on site. The absence

of reported deaths could also suggest that deaths

are not being reported. In addition, the numbers of

deaths and direct obstetric case fatality rates may

increase when efforts are made to improve hospital

servicesandmorewomencome for treatment, from

further away. Thus, the direct obstetric case fatality

ratemustbe interpreted in thecontextof theprevi-

ous indicators,andstudiesshouldbeconductedfor

deeperunderstanding.Bynomeansshouldthedirect

obstetriccasefatalityratebeacauseforadministra-

tivesanctions.Thatwouldjustincreasethelikelihood

thatwomenwithseriouscomplicationsarereferredto

anotherfacilityratherthantreated,orthatdeathsthat

occuronsitearenotreported.

Bar charts or scatter plots can effectively highlight

variationsindirectobstetriccasefatalityratesatdif-

ferent levels or in different types of health facility or

geographicalregion.Eachtypeoffacilityorregioncan

bedepictedasaseparategraph,ordifferentcolours

and shading can highlight differences in the same

graph.


Highdirectobstetriccasefatalityratesindicateprob-

lems but do not, by themselves, identify corrective

actions.Theyare,however,agoodbeginningforfur-

therstudies.

Casestudiesofwomen’sconditiononadmission

Information on the condition of women with major

complications at the time of admission (e.g. pulse,

blood pressure, and temperature) can be collected,

forwomenwhosurviveandthosewhodonot.Better

understanding of patients’ condition on admission

would help differentiate the effect of condition on

arrivalfromthequalityofcareafterarrival.

Delaysindiagnosisortreatment

Therearemanypossiblereasonsfordelayeddiagno-

sisortreatmentonceawomanhasreachedafacility.

Forexample,patients’familiesmayhavetobuydrugs

andmedicalsuppliesfromlocalpharmaciesbecause

the hospitals do not have enough. The causes of

delays can vary from back-ups in the emergency

room,toagatekeeperwhodemandsatip,toelectric-

ityfailures(150).

Studiesof‘thethirddelay’(oncewomenhavereached

healthfacilities)andthe‘clientflowanalysis’exercise

intheToolbookfor improvingthequalityofservices

(150) are useful models for this type of supplemen-

tarystudy;theysystematizetheobservationandmea-

surementofdelaysandallowresearchers to identify

atwhatstage theyaremost frequent.Theexercises

are based on evidence-based standards and expert

opiniontodeterminewhatconstitutesadelay.Another

approachistocollectdataontheintervalbetweenthe

time a woman with a complication is admitted and

whenshereceivesdefinitive treatment.Good-quality

monitoring revealswhichdelaysare the longestand

mostdangerous,andthedirectobstetriccasefatality

ratecanbeloweredbyreducingthosedelays.

In the university hospital of Zaria, Nigeria, the inter-

valbetweenadmissionandtreatmentwasreducedby

57%(from3.7to1.6hours)between1990and1995.

During this time, the case fatality rate (combining

directand indirectcauses)decreasedby21%, from

14%to11%(151).

Reviewingmaternaldeaths

When a direct obstetric case fatality rate is high or

fails to decrease, a study should be conducted.

Maternal deaths can be reviewed in health facilities

and at district, regional or national level (sometimes


referred to as ‘confidential enquiry’) to identify gaps

inmanagementorclinicalservicedelivery.TheWHO

publicationBeyondthenumbers—reviewingmaternal

deaths and complications to make pregnancy safer

(148)describestwotypesofreview:

• Afacility-basedreviewisadetailedstudyofthe

systemiccausesofandcircumstancessurround-

ingmaternaldeathsatthefacility.Thegoalisto

determinewhichofthefactorsthatcontributedto

maternaldeathswereavoidableandwhatcould

bechangedtoimprovethequalityofEmOCatthe

facility.

• Aconfidentialenquiryintomaternaldeathsisan

anonymous,systematicstudyofallorarandom

sampleofmaternaldeathsoccurringinaspecified

area(urban,district,regionornational).The

researcherslookatissuessuchassubstandard

care,women’saccesstocareandtheavailability

ofmedicinesanddrugs.Byaggregatingthe

causesandfactorsthatcontributetomaternal

deathsinawiderarea,evidencecanbegenerated

tohelpdecision-makersdesignandimplement

systematicsolutionsforimprovingEmOC.

Reviewingcasesofwomenwhosurvivelife-

threateningcomplications(‘nearmisses’)

An alternative, more positive and sometimes less

threateningapproachtoimprovingqualityistostudy

systematically the care given to women with life-

threateningobstetriccomplicationswhoaresavedby

thehealth facility (‘nearmisses’).Onebenefitof this

methodisthatnearmissesoccurmorefrequentlythan

maternaldeathsandthereforeprovidemoreopportu-

nitiesforstudyingthequalityofcare.Anotherbenefit

isthatsuchareviewprovidesanoccasiontolookat

what health professionals did correctly to save the

womanratherthanfocusontheproblems.Thishelps

tocreateamoresupportiveenvironment inwhichto

discussaspectsofcarethatcouldbeimproved.The

WHO publication Beyond the numbers (148) gives

moredetailedinformation,includingoperationaldefi-

nitionsofnearmissesandastandardsetof criteria

with which a near-miss case is identified is being

developedbyWHO(1,52).

2.7Indicator7:Intrapartumandveryearlyneonataldeathrate

Description

Indicator 7 is the proportion of births that result in

a very early neonatal death or an intrapartum death

(freshstillbirth) inanEmOCfacility.Thisnew indica-

torhasbeenproposedtoshedlightonthequalityof

intrapartumcareforfoetusesandnewbornsdelivered

atfacilities(153).Thenumeratoristhesumofintrapar-

tumandveryearlyneonataldeathswithinthefirst24

hoursoflifeoccurringinthefacilityduringaspecific

period,and thedenominator isallwomenwhogave

birthinthefacilityduringthesameperiod.

Becausetheobjectiveofthisindicatoristomeasure

thequalityofintrapartumandnewborncare,itisrec-

ommendedthatnewbornsunder2.5kgbeexcluded

from the numerator and the denominator whenever

thedatapermit,aslowbirthweightinfantshaveahigh

fatalityrateinmostcircumstances.

As for the previous indicators, the intrapartum and

veryearlyneonataldeathrateshouldbecalculatedfor

allfacilities,notjustEmOCfacilities.

Maximumacceptablelevel

No standard has been set; a maximum acceptable

levelmaybedeterminedaftertheindicatorhasbeen

testedinvariouscircumstances.

Background

Globally,nearly2millioninfantsdieeachyeararound

thetimeofdelivery:900000neonataldeaths,or23%

of all neonatal deaths, and 1.02 million intrapartum

stillbirths,or26%ofallstillbirths(154).Good-quality

intrapartum care is therefore crucial for both the

motherandherinfant.Whenappropriate,timelycare

isprovided,mostmaternalandneonataldeathscan

beprevented.

Amajorcauseof fetaldeath intrapartumor immedi-

ately postpartum is birth asphyxia, which can result

from poorly managed obstetric complications, such

as obstructed or prolonged labour, ruptured uterus,


eclampsia or antepartum haemorrhage, and the

absenceofneonatalresuscitation(155).Birthasphyxia

canalsobearesultofpretermbirthorcongenitalmal-

formation, conditions that are not directly related to

thequalityofcaregivenintrapartum.Aswearecon-

cernedhereprimarilywiththehealthsystem’sability

to provide good-quality intrapartum and immediate

postpartumcare,thisindicatorfocusesonthosestill-

birthsandveryearlyneonataldeathsthatcouldhave

beenavertedbytheavailabilityanduseofgood-qual-

ityobstetriccareandneonatalresuscitation.

Datacollectionandquality

The operational definitions for this indicator include

the following components, as defined by Lawn and

colleagues(154):

• Stillbirthsoccurringintrapartumorfreshstillbirths:

infantsborndeadaftermorethan28weeksof

gestationwithoutsignsofskindisintegration

ormaceration;thedeathisassumedtohave

occurredlessthan12hoursbeforedelivery;

excludesthosebornwithsevere,lethalcongenital

abnormalities.

• Earlyneonataldeathsrelatedtointrapartum

events:neonatesbornattermwhocouldnotbe

resuscitated(orforwhomresuscitationwasnot

available)orwhohadaspecificbirthtrauma.

Thedeathmusthaveoccurredwithin24hoursof

delivery.

These two subgroups should not be equated with

perinatal deaths. The universally accepted definition

ofperinataldeathisdeathintheuterusafterthe28th

weekofpregnancyplusdeathsofallliveborninfants

upto7daysoflife.Thisnewindicatorexcludesmac-

eratedstillbirthsandnewbornswhodieafterthefirst

24h,becausemothersandtheirinfantsareoftendis-

chargedat24h,ifnotearlier.

Atthetechnicalconsultationin2006,itwassuggested

thatthisindicatorincludeonlystillbirthsandneonates

weighing≥2.5kg,whichistheinternationalstandard;

however,countriesmayprefer touse2.0kgastheir

threshold.Manysmallfacilitiesinpoorcountriesmight

nothavedataonbirthweight,especiallyofstillbirths.

Accuraterecordingofstillbirths(freshandmacerated)

andveryearlyneonataldeathsmaybeanaspectof

currentinformationsystemsthatalsowillrequiremore

attention.

Onewayofdeterminingwhetheranintrapartumdeath

occurredduringlabouristoascertainwhetherthefetal

heartbeat is recordedontheadmission log. Inprac-

tice,infacilitieswithhighturnoverandwheremothers

staylessthan24hoursafterdelivery,itmaybewise

torestrictneonataldeathstothoseoccurringintheir

first6–12hours(ratherthan24hours),becausedeaths

occurringafterdischargewillgoundetected.

Thedenominatorforthisindicatoris‘allwomengiving

birthintheEmOCfacility’,whichisthesamenumera-

torasforIndicator3(ProportionofallbirthsinEmOC

facilities). This denominator was chosen to facilitate

datacollectionand is recommended for thesakeof

international comparability. As information systems

improve,thedenominatormaybecomebirths,andthe

indicatorwillbecomeatruerate.


Testingtheindicator

This indicatorshouldbe tested,and the resultswith

and without the birth weight restriction should be

comparedtodeterminewhether2.0kgor2.5kgisthe

better threshold. If the birth weight restriction is too

onerousintermsofdatacollection,studiesareneeded

todeterminewhethernobirthweightrestrictionwould

affectthedeathrate.Additionally,amaximumaccept-

ablelevelfortheindicatorshouldbeexploredandset,

ifappropriate.

Refiningthedata

Other studies that would improve understanding of

intrapartumandearlyneonatalcareincludeinvestiga-

tionsofwhether the fetalheartbeat is recorded rou-

tinelyatadmissionandwhetherstillbornsareroutinely

weighedanddocumented.Itcouldalsobeimportant

tostudytheexacttimeofearlyneonataldeath,which

israrelyrecordedwithprecision.

Infacilitieswithhighearlyneonatalandstillbirthrates,

it might be useful to conduct perinatal death audits

togainabetterunderstandingofhowtoimprovethe

qualityofcare(156).


2.8Indicator8:ProportionofdeathsduetoindirectcausesinEmOCfacilities

Description

The numerator of this new indicator is all maternal

deathsduetoindirectcausesinEmOCfacilitiesduring

aspecificperiod,anditsdenominatorisallmaternal

deathsinthesamefacilitiesduringthesameperiod.

Directcausesofdeatharethose‘resultingfromobstet-

ric complications of the pregnant state (pregnancy,

labour, and puerperium), from interventions, omis-

sions, incorrect treatment,or fromachainofevents

fromanyoftheabove’.Indirectcausesofdeathresult

from‘previousexistingdiseaseordiseasethatdevel-

oped during pregnancy and which was not due to

directobstetriccauses,butwhichwasaggravatedby

thephysiologiceffectsofpregnancy’(157).

Other categories of maternal death (death after 42

dayspostpartum,fortuitous,coincidentalorincidental

deaths)aregenerallynotincludedinthecalculationof

maternaldeathratesorratios,andtheyareexcluded

forthepurposesofthisindicator.

Acceptablelevel

This indicatordoesnot lend itselfeasily toa recom-

mendedorideallevel.Instead,ithighlightsthelarger

socialandmedicalcontextofacountryorregionand

hasimplicationsforinterventionstrategies,especially

inadditiontoEmOC,whereindirectcauseskillmany

womenofreproductiveage.

Background

A substantial proportion of maternal deaths in most

countriesaredue to indirectcauses.This isparticu-

larly true where HIV and other endemic infections,

suchasmalariaandhepatitis,areprevalent.Toooften,

whereinfectiousandcommunicablediseaseratesare

high, the number of maternal deaths due to direct

causes is also high. The causes of maternal deaths

areoftenmisclassifiedinsuchcases;forexample,the

death of an HIV-positive woman might be classified

asdue toAIDSeven if itwasdue toadirectcause

suchashaemorrhageorsepsis.Mostmaternaldeaths

fallintothecategorieslistedinTable9;weknoweven

lessabout ‘accidentalor incidental’causesofdeath

forwomeninpoorcountries.

The most recent systematic study of the causes of

maternal death was published in 2006 by research-

ers atWHO,who reviewed the literature since1990

(120).Table10summarizes theproportionsofdirect

andindirectcausesofdeathbyworldregion.

Table 9. Main conditions leading to maternal death

Directcauses Indirectcauses

Haemorrhage Infections(e.g.malaria,hepatitis)

Hypertensivediseases Cardiovasculardisease

Abortion Psychiatricillnesses,includingsuicideandviolence

Sepsisorinfections Tuberculosis

Obstructedlabour Epilepsy

Ectopicpregnancy Diabetes

Embolism

Anaesthesia-related


Table 10. Estimates of direct and indirect causes of maternal death by region

RegionMaternaldeaths(%)

Duetoindirectcauses

Duetodirectcauses Unclassified

Developedcountries 14.4 80.8 4.8

Africa 26.6 68.0 5.4

Asia 25.3 68.6 6.1

LatinAmericaandtheCaribbean 3.9 84.4 11.7

Fromreferences(120).

Datacollectionandquality

Thereportingofmaternaldeathsandtheircausesvar-

ieswidelyandisassociatedwithacountry’sstatisti-

caldevelopment;nevertheless,alltendtofollowsome

versionof the InternationalClassificationofDiseases

(157).Incountrieswithwell-developedstatisticalsys-

tems,thesourceofthisinformationisthevitalregis-

trationsystem,but,asstatedabove,misclassification

results inseriousunder-recording inofficialstatistics

in virtually all countries.Wherevital registrationsys-

temsareweak,omissionandmisclassificationleadto

under-recordingandproblemsofattributionofcause.

Deathcertificatesmayneverbefilledout,ortheymay

failtoindicatewhetherpregnancywasarecentoccur-

rence;therefore,thefactthatthedeathwasamater-

naldeathgoesundetected.Multiplecausesofdeath

may be listed, but an underlying cause may not be

registered.

ThisislikelytobethecasewithregardtoHIVinfec-

tion. In many countries with a high prevalence of

HIVinfection,thenumberofmaternaldeathsamong

HIV-positivewomenwillbeunderreported,untilthere

is universal HIV testing, serological status is reliably

recordedandreported,anddiscriminationandstigma

donot inhibit testingor reporting.On theonehand,

HIV infection might be an underreported cause of

maternaldeath.Ontheother,whenthewoman’sHIV

statusisknown,thecauseofdeathmaybereported

as AIDS even though the actual cause was a direct

obstetriccondition.

Althoughofficialstatistics inresource-poorcountries

arelikelytoincludeunderreportingofindirectcauses

of death, industrialized countries also underreport.

InareviewofWHOdatabasesonmaternalhealth in

1991–1993,ofthe60countriesreportingvitalregistra-

tionfiguresforcausesofmaternaldeaths,33reported

noindirectdeaths(158).

Collectingdataforthisnewindicatorwillbedifficult;

however, the technical consultation considered that

itwouldbeuseful forgovernmentsand international

agencies. Inafewyears,weshallreviewexperience

withthesenewindicatorstoseewhethertheyareuse-

fulandwhethertheyshouldbemodified.


Agreatdealofresearchremainstobedoneinthearea

of indirectmaternaldeaths, includingonthemecha-

nisms by which indirect conditions cause maternal

death and programmes that could reduce them. As

with the recording of obstetric complications, train-

ing staff to complywithnational standardsofdeath

certificatecompletioncanresultinmoreaccurateand

completerecording.Reviewsofalldeathsofwomen

ofreproductiveageinfacilities,especiallythosewho

donotdieonthematernityward,couldleadtomore

completerecording.Asdiscussedunder Indicator6,

itmightbeusefultoreviewmaternaldeathsandnear

missestolearnhowtoimprovethequalityofcare.


2.9Summaryandinterpretationofindicators1–8

Table11providesasummaryof the indicators,how

they are calculated, and acceptable levels, when

appropriate.Oneofthebenefitsofusingtheseindica-

torsisthat,whenusedasaset,theygiveafullpicture

ofahealthsystem’s response toobstetricemergen-

cies.Below,wediscussissuesthataffecttheinterpre-

tationofmostoftheindicators,includingdistinguish-

ing between ‘minimum or maximum’ and ‘optimum’

levels, assessing the generalizability of results and

working with incomplete or poor data. The section

alsoprovidesexamplesofinterpretingsetsofindica-

torsandendswithanexerciseininterpretingtheindi-

catorstogether.

Minimumormaximumandoptimumlevels

An important distinction that applies to most of the

indicators is the difference between minimum or

maximumandoptimumlevels.Bynecessity,themini-

mumormaximumacceptablelevelsproposedinthis

manualareapproximations.Therefore, if theaccept-

able level is met for a particular indicator, this does

not imply that theoptimum level hasbeen reached.

Forinstance,akeyassumptioninsettingacceptable

levelsisthatapproximately15%ofpregnantwomen

experience serious obstetric complications. If this

is an underestimate—as recent studies indicate it

maybe—themaximum level for Indicator5 (15%of

expectedbirthsaredeliveredbycaesareansectionin

EmOCfacilities)maybelow(159,160).Anumberof

studieshaveshown,however,thatitisdifficultoreven

impossibletomeasuremorbidityaccuratelyfromsur-

veys(161).Therefore,weassume(onthebasisofthe

evidenceusedthroughoutthismanual)thatacountry

thatachievesacceptablelevelsforeachindicatorhas

astrongprogrammeforreducingmaternaldeaths.

Even if the minimum acceptable level for an indica-

torismetatthenationallevel,however,theremaybe

problemsinspecificareas.Whenthelevelfallsbelow

theminimumacceptable,onecanconcludethat the

needforEmOCisnotbeingmetinmostareasofthe

country.Thegeneralprincipleisthatfavourablefind-

ings, while reassuring, do not justify complacency;

unfavourable findings clearly indicate that action is

needed.

Generalizabilityofresults

When subnational areas or facilities are selected for

study, the generalizability of the findings may be a

concern.Visitingallthefacilitiesinanarea,whenpos-

sible,canhavestrongprogrammaticimplications,as

health managers will be able to design site-specific

changes. Insection3.2,onpreparationfordatacol-

lection,theselectionoffacilitiesforstudycomprises

twosteps:selectionofareasand,withintheseareas,

selectionof facilities. If thesestepsare followed(i.e.

theworksheetisused),biasisminimized.

If it appears that, due to chance, random selec-

tion has produced a bias (for example, most of the

facilities selected are concentrated in one area of a

certainregion),thisshouldbenoted,asevenbiased

dataareuseful if thedirectionof thebias isknown.

Forinstance,supposethattheEmOCfacilitiesinthe

studywerenotrandomlyselectedandweretherefore

muchmorelikelytobelocatedonamajorroadthan

a randomlyselectedgroupwouldhavebeen. In this

case, it is possible to say with reasonable certainly

thathospitalsfarfrommajorroadsarelesslikelythan

hospitalsonmajor roads toperformcaesareansec-

tions.Therefore,theestimatederivedfromthebiased

sample probably presents an unrealistically favour-

ablepictureof Indicator5,andthesituationisprob-

ablyworsethanthedataindicate.Iftheinformationis

stillnotusefulforgeneralization,e.g. if it isnotclear

whichwaythebiasworks,thedatamaynevertheless

beuseful formanagingorevaluatinghealthservices

inthearea.Tousetheexampleabove,thedatamay

showthatsomehospitalsarenotprovidinglife-saving

servicessuchascaesareansection,eventhoughgov-

ernmentstandardsstatethattheyshould.This infor-

mation, by itself, can be used to guide activities to

reducematernaldeaths.

Incompleteorpoordata

Routinematernity recordsystems inmanycountries

do not facilitate the collection of data on obstetric

complications, maternal deaths, stillbirths and very

earlyneonataldeaths.Often,staffhasfallenoutofthe

habitoffillinginsomeofthecolumnsofthematernity

registerortheadmissionsanddischargeregisters.This

isamanagementproblemthatrequiresattentionover

timetoensurecomplete,accuraterecordkeeping.


Indicator Description Numerator Denominator Acceptablelevel

1&2* AvailabilityofEmOC(nationalorsubnational)

RatioofEmOCfacilitiestopopulationandgeographicaldistributionoffacilities

No.offacilitiesinareaprovidingbasicorcomprehensiveEmOC

Populationofareadividedby500000

≥5EmOCfacilitiesper500000population

No.offacilitiesinareaprovidingcomprehensiveEmOC

Populationofareadividedby500000

≥1comprehensivefacilityper500000population

3 ProportionofallbirthsinEmOCfacilities

ProportionofallbirthsinpopulationinEmOCfacilities

No.ofwomengivingbirthinEmOCfacilitiesinspecifiedperiod

Expectedno.ofbirthsinareainsameperiod

Recommendedleveltobesetlocally

4 MetneedforEmOC

ProportionofwomenwithmajordirectobstetriccomplicationstreatedatEmOCfacilities

No.ofwomenwithmajordirectobstetriccomplicationstreatedinEmOCfacilitiesinspecifiedperiod

Expectedno.ofwomenwithseveredirectobstetriccomplicationsinareainsameperiod**

100%

5 Caesareansectionasaproportionofallbirths

ProportionofallbirthsinpopulationbycaesareansectioninEmOCfacilities

No.ofcaesareansectionsinEmOCfacilitiesinspecifiedperiod

Expectedno.ofbirthsinareainsameperiod

5–15%

6 Directobstetriccasefatalityrate

ProportionofwomenwithmajordirectobstetriccomplicationswhodieinEmOCfacilities

No.ofmaternaldeathsduetodirectobstetriccausesinEmOCfacilitiesinspecifiedperiod

No.ofwomentreatedfordirectobstetriccomplicationsinEmOCfacilitiesinsameperiod

<1%

7 Intrapartumandveryearlyneonataldeathrate

Proportionofbirthsthatresultinanintrapartumoraveryearlyneonataldeathwithinthefirst24hinEmOCfacilities

No.ofintrapartumdeaths(freshstillbirths;≥2.5kg)andveryearlyneonataldeaths(<24h;≥2.5kg)inEmOCfacilitiesinspecifiedperiod

No.ofwomengivingbirthinEmOCfacilitiesinsameperiod

Tobedecided

8 Proportionofmaternaldeathsduetoindirectcauses

PercentageofallmaternaldeathsinEmOCfacilitiesduetoindirectcauses

No.ofmaternaldeathsduetoindirectcausesinEmOCfacilitiesinspecifiedperiod

Allmaternaldeaths(fromdirectandindirectcauses)inEmOCfacilitiesinsameperiod

Noneset

Table 11. Emergency obstetric care indicators

*Indicators1and2involvethesamecalculations,withdataonthecorrespondingregionalpopulationandfacilityinsteadofaggregatednationaldata.

**Equalto15%ofexpectedbirthsinthesameareaandperiod.


As stated earlier, in many countries maternity regis-

tersdonothaveacolumnfor‘reasonforadmission’

or ‘maternalcomplications’.Whenproviderswant to

recordmaternalcomplications,therefore,theyhaveto

makeanoteinanothercolumn,suchas‘remarks’,or

inthemargin.Whilethismayappeartobeanadminis-

trativedetail,itisastrongindicationofcommitmentto

improvingmaternalhealth.Thereisoftenroominreg-

isterstoaddsuchacolumn,perhapsbyreplacinga

columnusedforuncommonevents,suchasmultiple

births.Persuadingministriesofhealth(andfunders)to

addthiscolumnisanimportantstepinmakingthese

indicatorspartofhealthmanagementinformationsys-

tems.(AppendixBliststheitemsthatshouldappear

in facility registers.) As periodic collection of these

databecomespartofroutineprogrammemonitoring,

recordkeepingshouldimprove.

Dataonmaternaldeaths,stillbirthsandveryearlyneo-

nataldeathsareoftendifficult tocollect forsomeof

thesamereasonsstatedabove.Inaddition,because

of the sensitive nature of these events, health staff

maynotrecordthemforfearofreprisal.Interventions

geared to improve the working environment should,

over time, help health staff feel more comfortable

aboutaccuratelyrecordingdeaths.

Asrecordkeepingofcomplications,maternaldeaths,

stillbirths and very early neonatal deaths improves,

the reportednumberofcomplicationsanddeaths in

the facilitywill increase. It iscritical to reassurestaff

that these temporary increases will be appropriately

interpreted; that theywill notbeassumed tobe the

resultofpoorordeterioratingpatientcare.Oneway

ofidentifying‘recordingbias’istouseotherindicators

inthesetasbenchmarks,especiallythoseindicators

basedonservicesthatarereportedoftenandarefairly

reliable,suchas thenumbersofwomengivingbirth

andcaesareansectionsinthefacility.Usingtheindi-

catorsasasetcanhelpclarifywhethertheapparent

increase in complications or deaths is due to better

reportingorifitisarealincrease.Forexample,ifthe

reportednumberofwomenwithmajorcomplications

treatedinthefacilityincreasesby150%over3years,

but the number of women giving birth in the facility

increasesby75%andthenumberofcaesareansec-

tionsperformedincreasesby50%,itcanbeassumed

thatsomeofthereportedincreaseincomplicationsis

duetobetterreporting(probably intherangeofone

halftotwothirds).Asthecommunity’sconfidencein

thequalityofcareimprovesandwomenwithcompli-

cations are more likely to be brought for treatment,

manyofthewomenwillrequireacaesareansection;

therefore, thenumbersofcomplicationsandofcae-

sareansections should rise together, unless there is

a problem that limits the availability of surgery. This

exampleillustratesthekindofexplorationofthedata

thatcanbeusefulatlocallevel.

RelationofEmOCindicatorstomaternalmortality

Asnotedearlierinthishandbook,metneedforEmOC

and caesarean section as a proportion of all births

are closely correlated with maternal mortality ratios,

and it is logical that as met need goes up and the

direct obstetric case fatality rate declines, the num-

berofdeathsinthepopulationduetodirectobstetric

complicationswilldeclineaswell.Maternalmortality

ratios,however,aredifficulttomeasure,especiallyin

arelativelysmallarea(suchasaprojectarea)orover

a short period. Nevertheless, methods for capturing

theeffectofmaternalhealthprogrammesarecontinu-

ingtoimprove.Forexample,amethodforestimating

deathsaverted,basedon theEmOC indicators,has

beenproposed,althoughitmustbetested(162).Aset

of tools is available at: http://www.immpact-interna-

tional.org/index.php?id=67&top=60.

Anexerciseininterpretingtheindicatorsasaset

Table 12 shows three very different scenarios for

EmOCindicators.Thisexerciseshowsthatsuchdata

aredirectlyapplicableforprogramming.Examinethe

setsofindicatorsinthethreescenariosasifyouwere

anofficialoftheministryofhealthincountryX,look-

ing at data from various districts of the country. On

thebasisofthehypotheticaldataandtheacceptable

levels,identifyprioritiesforimprovingthesituationfor

womenwithobstetriccomplications.


Indicator Level

Scenario1

Population 950000

NumberoffunctioningEmOCfacilities:

•basic 2•comprehensive 1

GeographicaldistributionofEmOCfacilities Mostlyindistrictcapital

ProportionofallbirthsinbasicandcomprehensiveEmOCfacilities 10%

MetneedforEmOC 8%

Caesareansectionsasapercentageofallbirths 0.7%

Directobstetriccasefatalityrate 5%

Scenario2

Population 950000

NumberofEmOCfacilities:


GeographicaldistributionofEmOCfacilities Someurban,somerural


MetneedforEmOC 8%

Caesareansectionsasapercentageofallbirths 2%


Scenario3

Population 950000

NumberofEmOCfacilities


GeographicaldistributionofEmOCfacilities Someurban,somerural


MetneedforEmOC 65%

Caesareansectionsasapercentageofallbirths 12%


InScenario1,therearefartoofewfunctioningEmOC

facilities. For a population of nearly 1 million, there

should be 10 such facilities, at least two of which

are comprehensive, rather than the existing three.

Furthermore, the functioning facilities are mostly in

urbanareas.Theother indicatorsarenotverygood

either(e.g.thedirectobstetriccasefatalityrateistoo

highat5%),butclearlythefirstpriorityistoseewhich

healthfacilitiescanbeupgradedtoprovideappropri-

atecare,especiallyinruralareas.

InScenario2,thenumberoffunctioningEmOCfacili-

tiesismuchhigher:therearenine;twoofthesepro-

videcomprehensivecare,andsomeareinruralareas.

Table 12. Three scenarios for emergency obstetric care (EmOC) indicators and levels


Theproportionofdeliveries that takeplace in these

facilities is, however, low (10%), as is the met need

(8%).Thedirectobstetriccasefatalityrateisnotvery

high(at2%),butthisisnotareasonforcomplacency,

becausesofewwomenarecaredforat thesefacili-

ties.Thehighestpriorityherewouldbetofindoutwhy

use is so low, by using a variety of methods: com-

munityfocusgroups,discussionswithstaff,observa-

tionoftheservicesandareviewoftherecord-keeping

system.

InScenario3,thereismorethantheminimumnum-

ber of EmOC facilities (13); three of these are com-

prehensive(ratherthantheminimumoftwo),andthey

seemtobewelldistributedintermsofurbanandrural

areas. The proportion of births in the facilities (25%

ofallbirths)andmetneed(65%)arefairlyhigh.The

proportionofdeliveriesbycaesareansection(12%)is

towardsthehighendoftheacceptablerange(5–15%),

andthedirectobstetriccasefatalityrateisveryhighat

15%(withamaximumacceptablelevelof1%).Inthis

situation,thequalityofcareintheEmOCfacilities is

thefirstconcern.Clinicalauditsanddirectobservation

ofserviceswouldbeappropriate.Asmetneedandthe

directobstetriccasefatalityratearebothhighinthis

scenario,itisimportanttoanalysewhy.Forinstance,

women may present at the health facility very late,

whichisnotrelatedtothequalityofthehealthfacility.

Maternal death audits and verbal autopsies present

opportunities forhealthmanagers tounderstand the

relevantissues.


3.1Typesofdatarequired

Constructing the EmOC indicators proposed in this

documentrequiresdataonthepopulation,birthrate,

andhealthfacility.Table13showshowtheindicators

arecomposedofsuchdata.

Information on population and birth rates is avail-

ableinmostcountriesatcentrallevel(e.g.thecentral

statisticaloffice).Gatheringinformationonthesignal

functions,modeofchildbirth,obstetriccomplications

andmaternaldeaths,however,meansvisitinghealth

facilitiesandreviewingfacilityregisters.Theemphasis

ison theEmOCservices that a facility actuallypro-

videsratherthanonwhatitissupposedtobeableto

provide.

Thissectionlaysoutthestepsforcollectingthedata

necessaryfortheindicatorsofEmOC.Table14givesa

summaryofthesteps,andeachisdiscussedindetail

below.Sampledatacollectionformsaretobefoundin

AppendixAandarediscussedhere.Inaddition,sug-

gestionsaregivenaboutadditionaldatathatcanbeof

useinareamonitoring.

3.2Preparation

Mostofthedatanecessaryforcalculatingtheseindi-

catorswillbecollectedinfacilities.Inarelativelysmall

country,visitingeveryhospitalshouldnotbetoodif-

ficult,butinalargecountryitmightnotbepossible.

VisitingeveryhealthcentrethatmightprovideEmOC,

althoughidealfromaprogrammeviewpointwouldbe

difficult even in some small countries. Therefore, in

mostcountries,asubsetofpotentialEmOCfacilities

willhavetobeselectedforreview.

Wehope that ina fewyears thekindof information

requiredfortheseindicatorswillbereportedroutinely

toministriesofhealth,inwhichcasedataforallfacili-

tieswouldbecompiledandavailable.Ifthisinforma-

tionisavailableinaregularhealthmanagementinfor-

mationsystem,itiseasiertoassesstheavailabilityof

servicesandmakechangesandimprovementsinthe

healthsystem.

Thestepsdescribed inthissectionandthenextwill

helpinidentifyingagroupoffacilitiesthatgivesarea-

sonablyaccuratepictureofthesituation,whileatthe

same timenot requiringanunreasonable amountof

work.Incountrieswherefinancialandhumanresources

are constrained, the approach described below will

suffice to yield informative data about the maternity

care system. Ensuring that the facilities selected for

review give a fairly accurate picture of the situation

depends largely on avoiding two major pitfalls: sys-

tematicbiasandtheeffectsofchancevariation.

Systematicbiascanoccurwhenconsciousoruncon-

sciousfactorsaffecttheselectionoffacilitiesforstudy.

Forexample,thepeopleselectingthefacilitiesmight

want topresent thesituation in themost favourable

light possible, or they might select facilities that are

easilyaccessible(e.g.onapavedroadornearalarge

town).Ineithercase,thedatacollectedmightgivean

overly favourable impression. The effects of chance

are, of course, unpredictable, but they do tend to

diminishasthenumberoffacilitiesstudiedincreases.

Selection isdone intwostages:selectingareasofa

country for study and then selecting facilities within

thoseareas.Sections3.2.1and3.2.2presentaguide

forselectingareasforstudyatnationallevel.Facilities

within those areas are selected at the area level, as

describedinsections3.3.1and3.3.2.

3.2.1Determinethenumberofareastobestudied

Considera levelsmallerthan‘national’.Thetermfor

thisadministrativelevelwillvarybycountry,e.g.state,

province,butisreferredtohereasan‘area’.Inafew

countrieswheretheadministrativeunitsof‘provinces’

or ‘states’ are exceptionally large, it may be prefer-

able to define smaller areas, e.g. district or county,

for selection into the study. Alternatively, it may be

logisticallybetter toselect theoriginaladministrative

unitseveniftheyarelarge,butthenselectsubareas

forstudyatasecondstage.Asa roughguide, ifan

areahasmorethan100hospitals(publicandprivate),

subareas may be selected; the number of subareas

3.Collectingdatafortheindicators


Typ

eo

fd

ata

Ind

icat

or

1A

vaila

bili

tyo

fE

mO

C(b

asic

and

co

mp

rehe

nsiv

e)

Ind

icat

or

2G

eog

rap

hica

ld

istr

ibut

ion

of

Em

OC

faci

litie

s

Ind

icat

or

3P

rop

ort

ion

ofa

llb

irth

sin

Em

OC

fa

cilit

ies

Ind

icat

or

4M

etn

eed

for

Em

OC

Ind

icat

or

5C

aesa

rean

se

ctio

nsa

sp

rop

ort

ion

of

allb

irth

s

Ind

icat

or

6D

irect

ob

stet

ric

case

fata

lity

rate

Ind

icat

or

7In

trap

artu

m

and

ver

yea

rly

neo

nata

ld

eath

rat

e

Ind

icat

or

8P

rop

ort

ion

ofm

ater

nal

dea

ths

due

to

ind

irect

ca

uses

Exi

stin

ges

timat

es:

Pop

ulat

ion

size

xx

xx

x

Cru

de

bir

thr

ate

ofa

rea

xx

x

Hea

lthf

acili

tyd

ata:

Em

OC

sig

nalf

unct

ions

xx

xx

xx

xx

No.

ofw

om

eng

ivin

gb

irth

xx

No.

ofw

om

enw

ith

ob

stet

ricc

om

plic

atio

nsx

x

No.

ofc

aesa

rean

s

ectio

nsx

No.

ofm

ater

nald

eath

sd

uet

od

irect

ob

stet

ric

cau

ses

xx

No.

ofm

ater

nald

eath

sd

uet

oin

dire

ctc

ause

sx

Int

rap

artu

md

eath

s(

fres

hst

illb

irth

s;

≥2

.5k

g)a

ndv

ery

early

neo

nata

ldea

ths

(≤

24h

;≥2

.5k

g)i

nfa

cilit

y

x

Tab

le 1

3. T

ypes

of d

ata

used

to

co

nstr

uct

emer

gen

cy o

bst

etri

c ca

re in

dic

ato

rs


Table 14. Guide to data collection and forms

Activity Action Refertooruse

Sampleselection 1. Selectareasforstudy,ifnotnational.2. Determineasingle12–monthperiodtostudy

andenteronform2(Facilitycasesummaryform).

3. ListallpossiblefacilitiesintheareathatmightprovideEmOC.

4. Ifsamplingisnecessary,selectfacilitiestobevisited.

Sections3.2.1,3.2.2.Section3.2.3.

Sections3.3–3.3.2,form1andworksheets1aand1b

Datacollection 5. Conductsitevisitstofacilities. Section3andform2

Datapreparation 6. Ifasampleoffacilitieswasvisited,separatethemintohealthcentres(orotherlower-levelfacilities)andhospitalsbyareaandthenadjustthedataforareaestimates.

7. Ifallfacilitiesinanareaweresurveyed,separatethemintothreegroupsbyarea:• actualcomprehensiveEmOCfacilities• actualbasicEmOCfacilities• non-EmOCfacilities

8. Summarizefindingsforallindicatorsdisaggregatedbyclassifiedleveloffacility(i.e.basicandcomprehensiveandallsurveyedfacilities).

Section3.5,form3andworksheets3a,3band3corworksheets3d,3eand3f

Calculationandinterpretationofindicators

9. Calculateindicatorsfor(each)area(forEmOCfacilitiesandforallfacilities).

10.Interpret.11.Consolidateforms1–4(withworksheets)forall

studyareasifnational.12.Calculateindicatorsforentirecountry.13.Interpret.

Section3.6andform4Section3.1andtextoneachindicator(section2)Section3.7,form5andworksheet5aSection3.1andtextoneachindicator(section2)

Anareaistheadministrativelevelorgeographicareainthecountryincludedinthefacilitysurvey;e.g.,district,

state,province.

studiedshouldrepresentatleast30%ofthetotal.For

the purposes of the forms, each subarea should be

consideredan‘area’.Professionalhelpfromastatisti-

cianshouldbesoughtinobtainingnationalestimates

incountrieswheresubareasareselected.

Thefollowingguidelinesshouldbeusedtodetermine

whethertostudyallareasofacountry:

• Ifacountryhas100orfewerhospitals(publicand

private),thenstudyallareas.

• Ifacountryhasmorethan100hospitals(pub-

licandprivate),thenasubsetofareasmaybe

selectedforstudy.Selectasmanysubnational

areasaspossible,butthenumberselectedshould

beatleast30%ofthetotalnumberofsubnational

areasinthecountry.

Inselectingasubsetofareas, theaimshouldbe to

study as many areas as possible, without compro-

misingthequalityofthedatacollected.Forexample,

if there are 21 administrative areas in a country, 10

might be selected for study. Fewer can be studied

if resources are scarce, but the proportion selected

shouldnotbelessthan30%oraminimumofseven

administrativeareas.


3.2.2Randomselectionofareas

Toavoidbias,describedabove,theselectionofareas

withineachtypemustberandom.Theprocedurefor

randomselectionisasfollows:

Step1:Makealistofallareasinthecountry.Thelist

shouldbeinalphabeticalorder,tominimizethepos-

sibilityofbias.

Step2:Assigneachareaaconsecutivenumber,start-

ingwith1forthefirstareaonthelist.

Step3:Calculatethe‘samplinginterval’,whichwilltell

you toselecteveryntharea,once the firstareahas

beenselectedatrandom.Usethefollowingformula:

Samplinginterval=

totalnumberofareasinthecountry

dividedby

numberofareasselected

CountryWhasatotalof21areas,ofwhich10areto

beselectedforstudy,givingasamplingintervalof2

(21/10=2.1).Samplingintervalsshouldberoundedto

thenearestwholenumber.If,forexample,ithadbeen

decidedthat15ofthe21areaswouldbestudied,the

samplingintervalwouldbe1.4,whichwouldtherefore

rounddownto1,anindicationthateitherfewerareas

should be selected for study or all areas should be

includedinthesample.

Step 4: Identify the first area to be included in the

samplebygeneratinga randomnumber that is less

thanorequaltothesamplingintervalbutgreaterthan

zero.Thiscanbedonewitha randomnumber table

(Appendix C). To use the table, look away from the

pageandtouchitwiththepointofapencil.Thedigit

closest to where the pencil touches the page is the

randomnumber.Ifthedigitislessthanorequaltothe

samplingintervalandgreaterthanzero,useit;ifnot,

read from left to right until a digit that satisfies this

conditionisreached.Thisnumberwillbethefirstarea

selected.

For country W, the sampling interval is 2. Using the

randomnumbertable,ourpencilpointfallsonthedigit

7,atrow22,column5.Thisislargerthanoursampling

interval,sowereadfromlefttoright,passingthedigits

0,7and0,untilwecometo2.Thus,area2onthelist

willbethefirstareaselected.

Step5: Identifyallotherareas tobe included in the

samplebyaddingthesamplingintervaltothenumber

ofthefirstareaandcontinuetoselectareasuntilthe

desirednumberhasbeenreached.Asthefirstselected

areais2onthelistofareas,thenextonewouldbe2

plus2,or4,andthenext6,andsoon,until10areas

havebeenselected.

3.2.3Determineanationallyuniform12–monthperiodtobestudied

Thedatacollectedfromfacilitieswillberetrospective,

butthe12-monthperiodselectedshouldbearecent

one,toensurethatthedatawillstillbeavailable.For

comparabilityofdata,itisimportantthatalldatacol-

lected throughout the country be for the same 12–

monthperiod.Adecisionaboutwhichperiodtouse

shouldbemadeatnationallevel,anditshouldthenbe

enteredonthetopofthefacilitycasesummaryformof

form2beforeitisduplicatedforuse.Thiswillensure

that data collected at all facilities refer to the same

period.The12–monthperiodcanbeeither acalen-

daryear(e.g.1January2010–31December2010)or

anyother12-monthperiod(e.g.1June2012–31May

2013).

Onceareashavebeenselectedforstudy, forms1–4

and all the worksheets should be duplicated and a

complete set given to the person coordinating the

researchineacharea.

3.3Form1:AllpotentialEmOCfacilitiesinselectedareas

Thefirststepingatheringtherequireddataistomake

anexhaustive,up-to-datelistofallthefacilitiesineach

selected areas that may be providing delivery and

EmOCservices(basicorcomprehensive),asdefined

bythesignalfunctions(Table4).Afacilitythatmaybe

providingEmOCservicesisonethatis:

• ontheministryofhealth’slistofhospitalsand

lower-levelfacilitiesthatshouldbeprovidingdeliv-

eryservices;


• onalistofprivatehospitalsandlowerlevelfacili-

tiesthatmightbeprovidingatleastsomedelivery

services;or

• knownbytheareamedicalofficeraspossiblypro-

vidingdeliveryservices.

Thelistshouldbeascompleteaspossiblesothatno

EmOCfacilityisoverlooked;however,careshouldbe

takentoavoiddoublecounting.Worksheets1aand1b

canbeusedforthispurposeandshouldbeusedto

listallofthevariousfacilities—hospitals,maternities,

healthcentres,clinicsandhealthposts—thatmaybe

providingbasicorcomprehensiveEmOCinthearea.

Aseachworksheethasspacetolistonly10facilities,it

islikelythatthelistsofeachtypeoffacilitywillbesev-

eralpageslong.Itisrecommendedthattheselistsbe

inalphabeticalordertoreduceanybiasintheselec-

tionprocess(see3.3.2below).Form1summarizesthe

numberoffacilitieslistedonworksheets1aand1b.

3.3.1Determinethenumberoffacilitiestobereviewed

In a relatively small area, it may be possible to visit

every hospital, while in larger areas it will not. Even

in small areas, it will often be difficult to visit every

lower-levelfacilitythatprovidesdeliveryservicesand

mightbeprovidingbasicEmOC.Thus,inmostareas,

a subset of facilities may be selected for review. To

avoidbias,thissecondstageofselectionshouldalso

berandom.Thecriteriabelowcanbeusedtodecide

whethertostudyallfacilitiesortoselectasubsetfor

review.

Itisimportanttoincludeprivatesectorfacilitiesinthis

exercise. Therefore, countries may want to conduct

thefollowingexerciseseparatelyforpublicandprivate

facilities.

Hospitals(e.g.regional,district,rural,maternity):

• Ifthereare25orfewer,studyallofthem.

• Iftherearemorethan25,asubsetcanbe

selected.Selectasmanyaspossible,butthe

numbershouldrepresentatleast30%andthere

shouldnotbefewerthan20facilities.

Lower-levelfacilities(e.g.healthcentres,healthposts,

clinics):

• Ifthereare100orfewer,studyallofthem.

• Iftherearemorethan100,asubsetcanbe

selected.Selectasmanyaspossible,butthe

numbershouldrepresentatleast30%.

Example:InareaX,thereare48hospitalsofdifferent

levelsandtypes.Although48isgreaterthan25,itis

decidedthatitisfeasibletovisitallofthem.Thereare

also390healthcentresandhealthposts,butitwould

be too difficult and costly to visit all of them and a

subsetofthesefacilitiesmustbeselectedforreview.

Ifasubsetofeithertypeoffacilityistobeselected,the

numbertobevisitedmustbedecided.Asdescribed

above,thisnumbershouldbeaslargeaspossiblein

ordertominimizetheeffectsofchancevariation,and

shouldbeatleast30%ofallfacilitiesofeachtype.In

determiningthenumberoffacilitiestovisit,itisimpor-

tanttostrikeagoodbalancebetweenthenumberof

facilitiesand thequalityof thedata thatwill becol-

lectedfromthem.Inotherwords,thenumberoffacili-

tiesselectedshouldbeaslargeaspossiblewhilestill

allowingforcarefuldatacollectionateachfacility.

Example:InareaX,all48hospitalswillbevisited,and

40%ofthehealthcentresandpostswillbeselected

for review. Thus, 156 (0.4 x 390) health centres and

posts will be selected. The percentages of selected

hospitalsandlower-levelfacilitiesineachareashould

be recorded, so that thiscanbe taken intoaccount

whencombining the information fromall areas.This

stepisnotneededifthesamepercentageisselected

inallareas.

3.3.2Randomselectionoffacilities

Once thenumberof facilities tobevisitedhasbeen

decided,thenextstepistoselecttheactualfacilities.

Tominimizethechanceofbias,thisshouldbedone

randomly, inaproceduresimilar to that followed for

selecting areas. If all facilities are to be visited, this

stepwillnotbenecessary.Ifasubsetofbothhospi-

talsandlower-levelfacilitiesistobeselected,random

selection should be carried out separately for each

level.Theprocedureisoutlinedbelow.Randomselec-


tionwillbedonewithall the lists inworksheet1aor

1bthathavebeenfilledoutforthegeographicalarea

inquestion.

Step 1: Assign each facility a consecutive number.

In order to minimize the possibility of bias, facilities

should be listed in alphabetical order before being

numbered.

Step2:Calculatethesamplinginterval,whichwilltell

youtoselecteverynthfacilityoncethefirstfacilityhas

beenselectedatrandom.Usethefollowingformula:

Samplinginterval=

numberoffacilitiesinthearea

dividedby

numberoffacilitiestobeselected

Example: InareaX,a totalof390healthcentres,of

which156are tobeselected for review,producesa

samplingintervalof3(390/156=2.5).Samplinginter-

valsareroundedtothenearestwholenumber.

Step3: Identify the first facility tobe included in the

samplebygeneratinga randomnumber that is less

thanorequaltothesamplingintervalbutgreaterthan

zero.Thiscanbedoneusingarandomnumbertable

(Appendix C). To use the table, look away from the

pageandtouchitwiththepointofapencil.Thedigit

closest to where the pencil touches the page is the

random number. If the digit is less than or equal to

thesampling intervalandgreater thanzero,use it; if

not, read from left to right until a digit that satisfies

thisconditionisreached.Thisnumberwillbethefirst

facilityselected.

Example:Forlower-levelfacilitiesinareaX,thesam-

plingintervalis3.Usingtherandomnumbertable,our

pencilpointfallsonthedigit4, inrow12,column2.

This is larger thanoursampling interval,sowe read

from left to right,passing thedigits0,9and6,until

wecometo1.Thus,facility1onthelistoflower-level

facilitieswillbethefirstareaselected.

Step 4: Identify all other facilities to be studied by

addingthesamplingintervaltothenumberofthefirst

facility. Continue to select facilities until the desired

numberhasbeenreached.Ifyoucometotheendof

the list in theselectionprocess, return to thebegin-

ning,butdonotcountthosefacilitiesthathavealready

beenselected.

Example: Since the first selected facility is 1 on the

list, the next one would be 1 plus 3, or 4, and the

next7,andsoon.Facility388willbethe129thfacility

selected,andfacility3willbethe130th(sincefacility1

hasalreadybeenselectedandshouldnotbecounted

inthesecondpassthroughthelist).Everythirdfacil-

itywillcontinuetobeselectedinthiswayuntilall156

havebeenselected.

Oncethefacilitiestobereviewedhavebeenselected,

sitevisitstocollectdataateachfacilitycanbegin.

3.4Form2:ReviewofEmOCatfacilities

Acopyof form2 shouldbeusedat each facility to

recordthetypeandamountofservicesprovided.The

informationcompiledonthisformwillenableresearch

staff to determine whether a given facility is actu-

allyprovidingEmOCservicesand, if it is,whether it

is functioning at the basic or comprehensive level.

Exceptfordataonpopulationsizeandthecrudebirth

rate,alltheinformationneededtoconstructtheindi-

catorsiscontainedinform2.

EmOCsignalfunctions

TodeterminewhethertheEmOCsignalfunctionswere

performedinthepast3months,reviewfacilityregis-

ters,observeandifnecessaryinterviewhealthwork-

ersinthematernitywardandotherdepartments.

• Recordwhetherthesignalfunctionhasbeenper-

formedinthepast3monthsand,ifnot,whyithas

notbeenperformed.

• Considerallthefollowingwhendetermining

whetheraparticularsignalfunctionwasavailable:

– Isstaffatfacilitytrainedtoprovidetheservice?

– Aretherequisitesuppliesandequipmentpres-

ent?Istheequipmentfunctioning?

– Weretherecasesforwhichtheuseofa

particularsignalfunctionwasindicated?

– Arethecadresofstaffworkingatthefacility

authorizedtoperformtheservice?


• Ifasignalfunctionwasnotperformedinthepast

3months,indicatewhynot,usingthefollowing

definitions:

– Trainingissues:

- Authorizedcadreisavailable,butnottrained;

- Providerslackconfidenceintheirskills.

– Suppliesandequipmentissues:

- Suppliesorequipmentarenotavailable,not

functionalorbroken;

- Neededdrugsareunavailable.

– Managementissues:

- Providersdemandcompensationtoperform

thisfunction;

- Providersareencouragedtoperform

alternativeprocedures;

- Providersareuncomfortableorunwillingto

performtheprocedureforreasonsunrelated

totraining.

– Policyissues:

- Therequiredlevelofstaffisnotpostedtothis

facilityinadequatenumbers(oratall);

- Nationalorhospitalpoliciesdonotallowthe

functiontobeperformed.

– Noindication:

- Nowomanneedingthisprocedurecameto

thefacilityduringtheperiod.(Beforemarking

‘Noindication’,considertheprevious

options;forexample,ifasitedoesnothave

someonetrainedtoprovideaprocedureor

equipmentanddrugs,womenwillnotcome

fortheprocedure.).

Numberofwomengivingbirth• Thisisthenumberofwomenwithnormalvaginal

births+thenumberofwomenwithassisted

vaginaldeliveries+thenumberofcaesarean

sectionsinthefacility.

• Ifbreechdeliveriesarerecordedseparately,add

theseaswell,butremembertocheckthatthey

arenotalreadyincludedinnormaldeliveriesor

caesareansections.

• Remembertocountthenumberofwomenandnot

thenumberofbirths(i.e.infants).

Numberofcaesareansections• Remembertocountallemergencycaesarean

sectionsandallplannedorscheduledcaesarean

sections.

• Countcaesareansectionsperformedforneonatal

aswellasmaternalreasons.

Numberofwomenwithdirectobstetriccomplications• Inordertobeconsideredacaseandtobe

includedinthedata,awomanmustbepregnant

atthetimeofadmission,recentlydeliveredor

aborted.

• Includeonlyeventsofsufficientseveritythat

shouldbetreatedwithalife-savingprocedureor

arestabilizedandthenreferredtoanotherfacility.

• Thepatienthasacleardiagnosisofanyoneofthe

obstetriccomplications(seeBox2).

• Treatmentwasstartedbeforereferraltoanother

facility(includingstabilization).

• Whendiagnosisofcomplicationsisnotavailable,

usethefollowingcriteriaforinclusion:

– Recordsindicateclearsignsorsymptomssuch

asbleeding,highbloodpressure,feverwith

dischargeandconvulsions.

– Recordsindicatedefiniteinterventionssuchas

caesareansection,vacuumorforcepsdelivery,

bloodtransfusion,manualremovalofplacenta,

injectionofanticonvulsantorinjectionof

oxytocin.

• Excludewomenwhowereadmittedwithoutany

diagnosis(orcluesleadingtoadiagnosisas

mentionedabove)andwhoreceivednotreatment

beforebeingreferredtoanotherfacility.

• Ifonepatienthastwodiagnoses,selectthemore

seriousone.Forexample,ifapregnantwoman

wasadmittedforhaemorrhageandruptured

uterus,themaindiagnosisisruptureduterus.Ifthe

interviewerisunsureaboutthediagnosis,heor

sheshouldconsultthestaffworkinginthehealth


facility.Remembertocountthenumberofwomen

withobstetriccomplicationsandnotthenumber

ofobstetriccomplications.

• Abortioncomplicationsincludeonlythosewith

infectionorhaemorrhage(seecasedefinitionsin

Box2).

• Complicationsofabortioncanresultfromeither

inducedorspontaneousabortion.

• Whensearchingforcomplicationsofabortion,the

teamshouldlookinfemalewardregisters,emer-

gencyregistersandmaternity,labour,delivery,or

wardregisters.

Numberofmaternaldeathsduetodirectobstetriccauses• TheWHOdefinitionof‘maternaldeath’shouldbe

used:“Thedeathofawomanwhilepregnantor

within42daysofterminationofpregnancy,irre-

spectiveofthedurationorsiteofthepregnancy,

fromanycauserelatedtooraggravatedbythe

pregnancyoritsmanagement,butnotfromacci-

dentalcauses.”

• Countonlymaternaldeathsthatoccurredinthe

facilitybeingstudied.

• ThedefinitionsofobstetriccauseslistedinBox2

shouldbereferredtowhenfillinginthissection.

• Maternaldeathscanbedifficulttofindinsome

facilityregisters.Therefore,itisveryimportantto

lookatasmanysourcesaspossible(e.g.mater-

nitywardregisters,morguerecordbooks,emer-

gencyroomrecords).

• Maternaldeathscanbeasensitiveissuetodis-

cusswithhealthworkers.Sometimesitmightbe

helpfultoexplainthatthereviewisnotanaudit.

Inordertomakestafffeelmoreatease,onecan

pointoutsomethingpositiveabouttheirfacility(for

example,howmanywomentheyhavebeenable

totreat).

Numberofindirectmaternaldeaths• Beforefillingintheform,listthemajorindirect

causesofmaternaldeathsthatarerelevanttothe

countryunderreview,e.g.HIVinfection,severe

anaemiaandmalaria.

Numberoffreshstillbirthsandveryearlyneonataldeaths≥2.5kg• Refertothedefinitionsoffreshstillbirthsandvery

earlyneonataldeathsabove.

• Omitveryearlyneonataldeathswhenmothers

gavebirthoutsidehealthfacilities(i.e.inthecom-

munityorathome).

• Whenthebirthweightisunavailable,recordthe

deathandstatethatthebirthweightwasunknown.

Collectingcasesummarydata

Dependingon thesizeofeach facilityand thequal-

ityofitsrecords,itmaybetoodifficulttocollectthe

necessary information for the entire year directly on

form2.Therefore,twoplansarepresented.

Plan 1 should be followed whenever possible. This

entailscompletingthegridonform2(i.e.recordingthe

numberofwomengivingbirth,eachtypeofcomplica-

tion,caesareansection,maternaldeaths,intrapartum

deathsandveryearlyneonataldeaths)at thefacility

duringeachofthe12monthsbeingstudied.

Plan2canbefollowedifthefacility’spatientvolume

is so large that collecting this information for all 12

monthswouldbetootime-consuming(e.g.ifthereare

morethan10000deliveriesperyear). Inthisplan,a

sampleof4monthsdistributedthroughouttheyearis

usedandthenmultipliedbythreetoestimatethetotal

numberfortheyear.Incountrieswheretherearevast

seasonaldifferencesindeliveries,itmaybeimportant

tochoose4monthsdistributedthroughouttheyearto

accountforthisvariation.

3.5Form3:SummaryofdataonEmOCfacilitiesinanarea

Iftheanalysisistobeconductedmanuallyandnotby

computer,afterall thesectionsof form2havebeen

completed,theformsshouldbecollectedandsorted

bygeographicalarea.Thenextstepistosummarize

thefindingsforeacharea.Form3isusedforthispur-

poseandhastwosections,AandB,onlyoneofwhich

shouldbecompleted.


Section Awhichrequestsastraightforwardsummary

of the data collected from facilities, should be used

only if all facilities in the area were visited (that is,

therewasno selectionof facilities). Facilities should

besortedintothreegroupsonthebasisoftheentries

in the box entitled ‘Determination of EmOC status’

onform2.Thethreegroupsarefacilitiesthatprovide

comprehensive EmOC, facilities that provide basic

EmOC and facilities that do not fully provide either

basicorcomprehensiveservices.Worksheets3a,3b

and3carethenusedtopreparethesummary.

Section Bshouldbeusedifasampleoffacilitieswas

chosen.Itincludesanintermediatestepforadjusting

thedatacollectedintoestimatesforallfacilitiesinthe

area.Worksheets3d,3eand3fareneededtoprepare

thissummary.

Thus,onecopyofform3willbefilledoutforeacharea

included in thestudy,completingeithersectionAor

sectionB(deletethepartyoudonotuse).

3.6Form4:Calculationofindicatorsforeacharea

Oncethefindingsfromsitevisitshavebeensumma-

rized,form4canbeusedtocalculatetheindicators

foreacharea.This form laysout thesteps forusing

theinformationsummarizedinform3andincludesa

summarychecklisttodeterminewhethereachindica-

tormeetsanacceptablelevel.

While,ultimately, thedataon facilitieswillbeaggre-

gatedinordertocalculatetheindicatorsforthewhole

country,thearea-levelindicatorsprovideusefulinfor-

mation for setting programme priorities at the area

level,andanentiresetofcompletedforms1–4should

bemaintainedintheareaforthispurpose.Secondly,

theseindicatorsallowcomparisonsamongstudyareas

at the national level. Using the information obtained

foreachselectedarea,researcherscanexaminedif-

ferencesintheavailabilityofEmOCservices,useand

performanceindifferentareasofthecountry.Thiscan

haveimportantimplicationsforpolicyandsettingpro-

grammepriorities.

3.7Form5:Calculationofindicatorsforthecountry

In order to calculate the EmOC indicators for the

countryasawhole, researchersmustcollectcopies

ofallforms3and4(includingworksheets)fromeach

studyarea.Theinformationneededforthisfinalstep

issummarizedonform5andworksheet5a.Thelat-

tersummarizes informationonthenumberofEmOC

facilities, women giving birth, women with obstetric

complications, caesarean sections, maternal deaths

(direct and indirect) and intrapartum and very early

neonataldeathsinalltheareasselected.

The indicators for thecountryasawholearedeter-

minedonform5.Similarlytoform4forthecalculation

ofindicatorsatthearealevel,asummarychecklistof

acceptablelevelsforeachindicatorisprovided.

Oncetheindicatorshavebeencalculated,thelaststep

isinterpretation.Generalnotesontheinterpretationof

EmOCindicatorsareincludedunderthedescriptionof

eachindicatorinthefirstsectionofthishandbook.

3.8Monitoringatthearealevel

Area officials and planners may be interested in

greaterdetailthanisrequiredfornationalmonitoring.

Therefore, further questions might be added during

site visits to facilities. This can be done by attach-

inganextrasheettoform2(ReviewofEmOCfacili-

ties). Some questions that might be of interest are

discussedbelow.Itisimportant,however,thatallthe

data required for thecalculationof the indicatorsbe

collecteduniformlyforthewholecountry.Whileques-

tions may be added to form 2, none of the existing

questions should be modified or deleted. Additional

modules useful for conducting a more extensive

needsassessmentareavailableat:http://www.amd-

dprogram.org

3.8.1Leveloffunctioningoffacilities

Forthepurposesofmonitoring,it iscrucialthatonly

facilities that provide full basic or comprehensive

EmOC(i.e.facilitiesthatperformedallthedesignated

signal functions inTable4 in thepast 3months)be

includedinthefirstanalysis.Areaplannersmightalso


be interested in knowing what signal functions the

otherfacilitiesintheareahaveperformed,andwhich

of them could potentially function as basic or com-

prehensive EmOC facilities. Tables can be prepared

to determine how many facilities did not perform

oneormoresignalfunctions,andwhichsignalfunc-

tionsfacilitiestheydidordidnotprovideinthepast3

months.Understandingwhysignalfunctionswerenot

performedisimportant.Theseinvestigationswouldbe

particularlyuseful if theanalysisofEmOC indicators

revealsashortageoffacilities.Inthatcase,information

aboutwhichfacilitiesareclosetoprovidingsuchcare

canbeusedinplanningwhichfacilitiestoupgrade.If

aparticularsignal function,suchasassistedvaginal

delivery,isoftennotperformed,apolicyreviewmight

becalledforinordertoascertainwhoistrainedtodo

what,atwhatlevelofthehealthsystem.

3.8.2Timeavailabilityofservices

Anotherfactorthatareaofficialsmightwishtoexam-

ine is whether obstetric services are available 24 h/

day, 7 days/week at facilities that are already fully

functioning.Forexample,aquestiononthehoursper

dayanddaysperweekthatsignalfunctionsareactu-

allyavailablemightbeaddedtothefacilityreviewform

(form2).Asobstetriccomplicationsareunpredictable,

itisimportantthatwomenhaveaccesstolife-saving

EmOCaroundtheclock.Analysesoflocalpatternsin

theavailabilityofsignalfunctionsmightshowthatthe

EmOCcoverageisactuallylowerthanthenumberof

facilities would imply. In such cases, expanding the

hourswhenservicesareavailable isstronglyrecom-

mended.

3.8.3Geographicaldistributionofserviceswithinareas

ThegeographicaldistributionofEmOCfacilitiesalso

affects the accessibility of services. Although the

numberoffacilities inanareamightmeetorexceed

the minimum acceptable level, smaller geographical

regionsmayhavetoofewornofacilities.Atthearea

level,therefore,itmaybedesirabletolocatefacilities

onamapinordertoidentifylocalareaswherewomen

donothaveaccesstoEmOC,eitherbecausefacilities

donotexistorbecausetheexistingfacilitiesarenot

accessible,e.g.becauseofpoorornonexistentroads

andbridges.

3.8.4Differencesbetweenpublic-andprivate-sectorfacilities

Health planners may be interested in examining dif-

ferencesbetweenfacilitiesthataregovernment-oper-

ated and those that are managed by religious insti-

tutions, nongovernmental organizations or for-profit

organizations. Such differences can have important

implicationsforprogramming.Forexample,onemight

want to know the proportions of women with com-

plications who are receiving EmOC in public and in

private facilities, or which types of facilities perform

moreEmOCsignal functions.Onemightalsoexam-

inedifferencesincasefatalityratesinhospitalbytype

offacility.Insomesituations,accesstoservicesand

issues of equity can be related to facility ownership

andcostofservices.

3.8.5Qualityofcareatfacilities

As discussed earlier, case fatality rates are a crude

indicatorof the levelofperformanceatEmOCfacili-

ties.Area researchersoradministratorsmight there-

forewishtocollectadditionalinformationtogainmore

insight into the quality of care provided at selected

localfacilities.Oneapproachistocollectdataonthe

intervalbetweenthetimeawomanisadmittedtoan

EmOCfacilityandthetimesheactuallyreceivestreat-

ment,asdiscussedunder‘Supplementarystudies’in

thesectionondirectobstetriccasefatalityrates.

Detailed case reviews or audits of both maternal

deaths and ‘near misses’ can also provide valuable

information about the quality of care. Case reviews

andauditshavetheadvantageofidentifyingproblem

areas within facilities and suggesting possible rem-

edies.Someresourcesthatcanbeusedforstudiesof

thequalityofcareare:

• EngenderHealthandAMDD.Qualityimprovement

forEmOC:leadershipmanualandtoolbook

(http://www.engenderhealth.org/pubs/maternal/

qi-emoc.php)(163).Thispublicationcanhelp

health-careproviderstoidentifyandsolvetheir

ownproblems.Itoutlinesacontinuous,four-step


qualityimprovementprocessbasedonparticipa-

toryprinciples,withstaffinvolvementandowner-

shipandfocusingonclients’rightsandneeds.It

alsocontainsinstrumentsforcollectinginforma-

tionandinstructionsfortheiruse.

• AMDD.ImprovingEmOCthroughcriterion-

basedaudit,2002(http://www.amddprogram.

org/resources/CriterionBased%20AuditEN.pdf).

Thismanualdescribes‘criterion-basedaudit’as

acomparisonofactualpracticewithevidence-

basedstandardsofcare.Itisusedtoimproveclin-

icalandmanagerialpractice,tomakemoreratio-

naluseofscarceresourcesandtoimprovestaff

morale.Theauditcycleincludesdatacollection,

analysis,andaplanofactiontocorrectdeficien-

cies,implementationofthatplanandrepetition

ofthecycletomeasurechange.Criterion-based

auditcanalsobeusedtoexaminemanagementor

theorganizationofservicesandhumanrightsina

clinicalsetting.

• WHO.Beyondthenumbers.Reviewingmaternal

deathsandcomplicationstomakepregnancy

safer,2004(http://www.who.int/reproductive-

health/publications/btn).Thisbookisdirectedat

healthprofessionals,health-careplannersand

managersworkingonmaternalandnewborn

healthwhowishtoimprovethequalityofcare

provided.Theyshouldbeinapositionandwilling

totakeremedialactiononthebasisofthefindings

ofthesereviews.Theinformationcanbeusedto

improvematernalhealthoutcomesbyencouraging

healthprofessionalstoevaluatecurrentpractices

criticallyandtochangethemifnecessary.As

actionistheultimategoalofthesereviews,itis

importantthatpeoplewhocanimplementtherec-

ommendedchangesparticipateactively.

3.8.6Qualityoffacilityrecords

Area-level officials should examine the method by

which the number of women with complications is

derivedinthefacilityreviewforms(form2).Theform

offers twoplans forarrivingat thisnumber (seedis-

cussion in section 3.4). Some facilities are probably

treating more women with obstetric complications

thantheirrecordsindicate,andthefinalquestionson

theformaskthereviewertogiveaninformedopinion

aboutthecompletenessofthefacility’srecords.Area-

level officials might be interested in examining the

repliestothisquestionforfacilitiesintheirarea.Ifthe

recordsforanumberoffacilitiesappeartobeincom-

plete,aworkshoponfacilityrecordkeepingcouldbe

conducted.


References1. UNICEF, WHO, UNFPA. Guidelines for monitoring the

availability and use of obstetric services. New York, United Nations Children Fund, 1997.

2. Freedman LP et al. Practical lessons from global safe motherhood initiatives: Time for a new focus on implementation. Lancet, 2007, 370(9595):1383-1391.

3. Graham WJ. Now or never: The case for measuring mater-nal mortality. Lancet, 2002, 359(9307):701-704.

4. Smith JB et al. Estimates of the maternal mortality ratio in two districts of the Brong-Ahafo region, Ghana. Bulletin of the World Health Organization, 2001, 79:400-408.

5. Stanton C, Abderrahim N, Hill K. An assessment of DHS maternal mortality indicators. Studies in Family Planning, 2000, 31(2):111-123.

6. Ministère de la Santé Publique du Bénin. Evaluation des besoins en soins obstétricaux d’urgence au Bénin – rapport final [Needs assessment for emergency obstetric care in Benin – final report]. Cotonou, Benin, 2003.

7. AMDD Working Group on Indicators. Program note: Using UN process indicators to assess needs in emergency obstetric services: Benin and Chad. International Journal of Gynecology and Obstetrics, 2004, 86:110-120.

8. Konate S, Traore, A., Gueye, A. Evaluation de la disponibilité et de l’utilisation des services obstétricaux au Cameroun [Evaluation of the availability and use of obstetric care in Cameroon]. UNFPA, IFORD, 2001.

9. AMDD Working Group on Indicators. Program note: Using UN process indicators to assess needs in emergency obstetric services: Bhutan, Cameroon and Rajasthan, India. International Journal of Gynecology and Obstetrics, 2002, 77(3):277-284.

10. UNFPA. Making safe motherhood a reality in West Africa - using indicators to programme for results. New York, 2003.

11. Ministère de la Santé Publique du Tchad. Evaluation des besoins en soins obstétricaux d’urgence au Tchad [Needs assessment for emergency obstetric care in Chad]. N’Djamena, 2003.

12. Affane S. Rapport d’étude sur les indicateurs de processus pour le suivi et la surveillance de la mortalité maternelle [Report of a study on process indicators for follow-up and monitoring of maternal mortality]. Moroni, Ministère de la Santé, 2005.

13. Ministère de la Santé Publique et al. Disponibilité, utilisation et efficacité des soins obstétricaux d’urgence (SOU) en Côte d’Ivoire [Availability, use and eficcacity of emergency obstetric care in Côte d’Ivoire]. Abidjan, 2001.

14. Gottlieb P, Lindmark G. WHO indicators for evaluation of maternal health care services, applicability in least devel-oped countries: A case study from Eritrea. African Journal of Reproductive Health, Aug 2002, 6(2):13-22.

15. Kayongo M et al. Making emergency obstetric care a reality—CARE’s experiences in areas of high maternal mortality in Africa. International Journal of Gynecology and Obstetrics, 2006, 92(3):308-319.

16. Ministry of Public Health of the République Gabonaise. Evaluation de la disponibilité, de l’utilisation et de la qualite des soins obstétricaux d’urgence au Gabon [Evaluation of the availability, use and quality of emergency obstetric care in Gabon]. Libreville, 2003.

17. Fauveau V. Program note: Using UN process indicators to assess needs in emergency obstetric services: Gabon, Guinea-Bissau, and The Gambia. International Journal of Gynecology and Obstetrics, 2007, 96(3):233-240.

18. Njie I, Kah-Janneh M. Evaluation of availability, utilisation and quality of emergency obstetric care in The Gambia: Draft report. Banjul, UNFPA (The Gambia), Department of State for Health, 2004.

19. Ministry of Health, UNFPA. Accelerating improvements in maternal and newborn health in northern Ghana. Accra, 2005.

20. Ministère de la Santé Publique, Programme National de Maternité Sans Risques, UNICEF. Rapport de l’evaluation nationale des besoins en soins obstétricaux d’urgence en République de Guinee [Report on the national evaluation of needs for emergency obstetric care in Guinea-Bissau]. Conakry, 2004.

21. Nogueira J et al. Avaliacao dos cuidados obstetricos de ur-gencia na Guine-Bissau [Evaluation of emergency obstetric care in Guinea-Bissau]. Guinea Bissau, Ministerio de Saude Publica, UNFPA, AMDD, 2003.

22. Reproductive Health Section, Ministry of Health, UNICEF Kenya Country Office. Report: Emergency obstetric care service, North Eastern Province, Kenya. Nairobi, 2003.

23. Pearson L, Shoo R. Availability and use of emergency obstetric services: Kenya, Rwanda, Southern Sudan, and Uganda. International Journal of Gynecology and Obstetrics, 2005, 88(2):208-215.

24. National Coordinating Agency for Population and Develop-ment et al. Kenya service provision assessment survey 2004. Nairobi, 2005.

25. Lesotho Ministry of Health and Social Welfare. Emergency obstetric care assessment report Lesotho, March-May 2005. Maseru, 2005.


26. Rasolofomanana J, Ralisimalala A. Evaluation des besoins en matière de soins obstétricaux ex néo-natals d’urgence à Madagascar [Evaluation of needs for emergency obstetric care in Madagascar]. Tananarive, 2004.

27. Ministry of Health. Emergency obstetric care services in Malawi, report of a nationwide assessment. Lilongwe, 2005.

28. Goodburn EA et al. Monitoring obstetric services: Putting the UN guidelines into practice in Malawi. I: Developing the system. International Journal of Gynecology and Obstetrics, 2001, 74(2):105-117.

29. Hussein J et al. Monitoring obstetric services: Putting the ‘UN guidelines’ into practice in Malawi: 3 years on. International Journal of Gynecology and Obstetrics, 2001, 75:63-73.

30. Leigh B et al. Emergency obstetric care: How do we stand in Malawi? International Journal of Gynecology and Obstetrics, 2008, 101(1):107-111.

31. Ministère de la Santé, Division Santé de la Reproduction. Evaluation de l’offre des besoins obstétricaux d’urgence au Mali [Evaluation of the availability of emergency obstetric needs in Mali]. Bamako, 2003.

32. Otchere SA, Kayo A. The challenges of improving emergen-cy obstetric care in two rural districts in Mali. International Journal of Gynecology and Obstetrics, 2007, 99(2):173-182.

33. Faire de la Maternité Sans Risque une Réalité, UNFPA. Résultat des etudes pays sur les soins obstétricaux d’urgence [Results of country studies on emergency obstetric care - Mauritania]. Nouakchott; 2001.

34. Jamisse L, Eduardo M, Derveeuw M. Needs assessment on availability and use of emergency obstetric care services in Sofala Province, Mozambique. Final report of the needs assessment, December, 2000. Ministério da Saúde, Direcção Nacional de Saúde, Departamento de Saúde da Comunidade, 2000.

35. Jamisse L et al. Assessment of the capacity of the health services to provide essential obstetric care in Tete province, Mozambique. Final report of the needs assessment, December, 2000. Tete, Ministério da Saúde, Direcção Nacional de Saúde, Departamento de Saúde da Comunidade, 2000.

36. Santos C et al. Improving emergency obstetric care in Mozambique: The story of Sofala. International Journal of Gynecology and Obstetrics, 2006, 94(2):190-201.

37. Bailey PE, Paxton A. Program note: Using UN process indicators to assess needs in emergency obstetric services. International Journal of Gynecology and Obstetrics, 2002, 76(3):299-305.

38. Ministry of Health and Social Services. Report of needs as-sessment for emergency obstetrics care. Windhoek, 2006.

39. AMDD Working Group on Indicators. Program note. Using UN process indicators to assess needs in emergency obstetric services: Niger, Rwanda and Tanzania. International Journal of Gynecology and Obstetrics, Oct 2003, 83(1):112-120.

40. UNICEF. Report on emergency obstetric care assessment: Gakoma, Kabaya, Muhororo and Nyanza health districts. Kigali, 2004.

41. Kayongo M et al. Improving availability of emergency obstetric care services in Rwanda—CARE’s experiences and lessons learned at Kabgayi referral hospital. International Journal of Gynecology and Obstetrics, 2006, 92(3):291-298.

42. National Institute of Statistics of Rwanda. Rwanda service provision assessment survey 2007. Ongoing.

43. Moreira I et al. Evaluation de la disponibilité, de l’utilisation et de la qualité des soins obstétricaux d’urgence au Senegal. Rapport final (draft) [evaluation of the availability, use and quality of emergency obstetric care in Senegal. Final report (draft)]. Dakar, Ministerè de la Santé, Service National de la Santé de la Reproduction, FNUAP, University of Colombia, 2001.

44. The Sierra Leone Prevention of Maternal Mortality Program, and WHO. Assessment of district hospitals in Sierra Leone for the delivery of safe motherhood and reproductive health services: A needs assessment survey report. Freetown, 2002.

45. Uganda Ministry of Health, UNICEF. Status of emergency obstetric care in Uganda, a national needs assessment of emergency obstetric care process indicators. Kampala, 2004.

46. Orinda V et al. A sector-wide approach to emergency ob-stetric care in Uganda. International Journal of Gynecology and Obstetrics, 2005, 91(3):285-291.

47. Malecela M et al. Situation analysis of emergency obstetric care for safe motherhood in public health facilities in Tanzania mainland. Dodoma, National Institute for Medical Research, Ministry of Health and Social Welfare, Reproductive and Child Health Section, 2006.

48. National Bureau of Statistics, Macro International Inc. Tanzania service provision assessment survey 2006. Dodoma, 2007.

49. Olsen Ø, Ndeki S, Norheim OF. Availability, distribution and use of emergency obstetric care in northern Tanzania. Health Policy and Planning, 2005, 20(3):167-175.

50. Urasra DP et al. Are process indicators adequate to assess essential obstetric care at district level? A case study from Rufiji district, Tanzania. African Journal of Reproductive Health, 2005, 9(3):100-111.


51. Olsen OE, Ndeki S, Norheim OF. Complicated deliveries, critical care and quality in emergency obstetric care in northern Tanzania. International Journal of Gynecology and Obstetrics, 2004, 87(1):98-108.

52. Ministry of Health and Central Board of Health. A national assessment of emergency obstetric care services in Zambia. Lusaka, 2005.

53. Ministry of Health and Child Welfare and Reproductive Health Unit. Assessment of coverage and utilisation of essential obstetric care in Zimbabwe. Harare, 2000.

54. Ministry of Health and Child Welfare. Assessment of ma-ternal and neonatal health services in Zimbabwe (working draft). Harare, Ministry of Health and Child Welfare, UNICEF, UNFPA, WHO, 2005.

55. UNFPA Bolivia. Informe final de la evaluacion de necesi-dades para los cuidados obstétricos de emergencia [final report on the needs assessment for emergency obstetric care]. La Paz, 2004.

56. Bailey P. Program note: Using UN process indicators to assess needs in emergency obstetric services: Bolivia, El Salvador and Honduras. International Journal of Gynecology and Obstetrics, 2005, 89(2):221-230.

57. Canales de Calderón R, Hernández M, Morales Velado M. Evaluación de disponibilidad y uso de cuidados obstétricos de emergencia en El Salvador [evaluation of the availability and use of emergency obstetric care in El Salvador]. San Salvador, 2003.

58. Ministerio de Salud Publica y Asistencia Social. Linea de base de mortalidad materna en El Salvador: Junio 2005 - Mayo 2006 [baseline maternal mortality in El Salvador: June 2005 - May 2006]. San Salvador, Sistema Activo de Vigilancia Epidemiologica de la Mortalidad Materna, 2006.

59. Bailey P et al. Monitoring utilization and need for obstetric care in the highlands of Guatemala. International Journal of Gynecology and Obstetrics, 2005, 89(2):209-217.

60. Deras V, Discua F. Informe final sobre: Disponibilidad y uso de cuidados obstétricos de emergencia [final report on the availability and use of emergency obstetric care]. Tegucigalpa, 2003.

61. Managua F. Maternidad segura evaluación de necesidades en los servicios de emergencia obstétrica [safe motherhood evaluation of the need for emergency obstetric care services]. Managua, FNUAP Managua and MINSA, 2001.

62. AMDD Working Group on Indicators. Program note: using the UN process indicators to assess needs in emergency obstetric services: Morocco, Nicaragua and Sri Lanka. International Journal of Gynecology and Obstetrics, 2003, 80:222-230.

63. FEMME. Reporte de la línea de base evaluación de necesidades en emergencias obstétricas [report on the baseline evaluation of needs for emergency obstetric care]. Ayacucho, CARE, 2000.

64. AMDD Working Group on Indicators. Program note: using UN process indicators to assess needs in emergency obstetric services: Pakistan, Peru and Vietnam. International Journal of Gynecology and Obstetrics, 2002, 78:275-282.

65. Kayongo M et al. Strengthening emergency obstetric care in Ayacucho, Peru. International Journal of Gynecology and Obstetrics, 2006, 92(3):299-307.

66. Lobis S, Fry D, Paxton A. Program note: Applying the UN process indicators for emergency obstetric care to the United States. International Journal of Gynecology and Obstetrics, 2005, 88(2):203-207.

67. Mouniri H. Project collaboration agreement to support maternal mortality reduction in Afghanistan January 2003-March 2005: End of the project report. Kabul, UNICEF Afghanistan Country Office and Columbia University/AMDD, 2005.

68. Programme Santé de la Reproduction. Evaluation des besoins en soins obstétricaux et néonataux d’urgence en République de Djibouti [needs assessment for emergency obstetric and neonatal care in the Republic of Djibouti]. Djibouti, 2005.

69. Ministère de la Santé, Direction de la Population, UNFPA. Evaluation des ressources, des besoins et monitorage des soins obstetricaux d’urgence au Maroc [evaluation of resources, needs and monitoring of emergency obstetric care in Morocco]. Rabat, 2002.

70. UNICEF Karachi. Needs assessment for 9 districts in Sindh Province. UNICEF Karachi and Pakistan Medical Association, 2000.

71. Ali M et al. Emergency obstetric care in Pakistan: Potential for reduced maternal mortality through improved basic emergency obstetric care facilities, services, and access. International Journal of Gynecology and Obstetrics, 2005, 91(1):105-112.

72. Ali M et al. Emergency obstetric care availability, accessibility and utilization in eight districts in Pakistan’s Northwest Frontier Province. Journal of the Ayub Medical College of Abbottabad, 2006, 18(4):10-15.

73. Ali M, Kuroiwa C. Accurate record keeping in referral hospi-tals in Pakistan’s Northwest Frontier Province and Punjab: A crucial step needed to improve maternal health. Journal of the Pakistan Medical Association, 2007, 57(9):443-446.

74. Reproductive Health Program and UNICEF. Draft report of the needs assessment for emergency obstetrics care Somaliland (NWZ). Mogadishu, 2007.

75. Federal Ministry of Health and Reproductive Health Direc-torate. Report on: National emergency obstetric care needs assessment: October - December 2005. Khartoum, 2005.

76. Kashmiry A. Final report: Baseline assessment for the FEMME project. Dushanbe, CARE Tajikistan, 2000.


77. Associates for Community and Population Research. Review of availability and use of emergency obstetric care services in Bangladesh. Dhaka, UNICEF, 2001.

78. Reproductive Health Program et al. Emergency obstetric care services: Inventory report. Dhaka, 2006.

79. Islam MT et al. Improvement of coverage and utilization of emergency obstetric care services in southwestern Bangladesh. International Journal of Gynecology and Obstetrics, 2005, 91(3):298-305.

80. Wangmo D et al. Report on first phase needs assessment in emergency obstetric care facilities and service delivery in the district hospitals of Bhutan. Thimphu, Ministry of Health and Education, Department of Health Services, 2000.

81. Khamgaonkar M. Women’s right to life and health project, State of Maharashta, needs assessment phase. Final report. Mumbai, UNICEF, 2000.

82. Pal P, Joshi S. Women’s right for life and health annual report - Rajasthan. New Delhi, UNICEF, 2000.

83. AMDD, UNFPA. Facility survey reports IPD districts of Rajasthan. Jaipur, 2000.

84. Government of Rajasthan. Facility survey for emergency obstetric services. Jaipur, 2000.

85. Biswas AB et al. Availability and use of emergency obstetric care services in four districts of West Bengal, India. Journal of Health Population and Nutrition, 2005, 23(3):266-274.

86. Family Health Division of Ministry of Health and UNICEF. Needs assessment on the availability of emergency obstetric care services in eastern, western and mid-western regions of Nepal. Final report. New era. Katmandu, 2000.

87. Rana TG et al. Strengthening emergency obstetric care in Nepal: The women’s right to life and health project (WRLHP). International Journal of Gynecology and Obstetrics, 2007, 98(3):271-277.

88. Pathak LR et al. Process indicators for safe motherhood programmes: Their application and implications as derived from hospital data in Nepal. Tropical Medicine and International Health, 2000, 5(12):882-890.

89. Family Health Bureau, Sri Lanka College of Obstetricians and Gynaecologists, UNICEF. Needs assessment study women’s right to life and health project Sri Lanka. Colombo, UNICEF, 2001.

90. Liabsuetrakul T et al. Emergency obstetric care in the southernmost provinces of Thailand. International Journal for Quality in Health Care, 2007, 19(4):250-256.

91. Save the Children. Needs assessment reports for the provinces of Quang Tri and Thanh Hoa. Hanoi, 2000.

92. Otchere SA, Binh HT. Strengthening emergency obstetric care in Thanh Hoa and Quang Tri provinces in Vietnam. International Journal of Gynecology and Obstetrics, 2007, 99(2):165-172.

93. Technical consultation on Guidelines for monitoring the availability and use of obstetric services. World Health Organization/UNFPA/UNICEF/AMDD, Geneva, 2006.

94. Complications of abortion: Technical and managerial guidelines for prevention and treatment. World Health Organization, Geneva, 1994.

95. Managing complications in pregnancy and childbirth: A guide for midwives and doctors. (WHO/RHR/00.7). World Health Organization, Geneva, 2003.

96. Managing newborn problems: A guide for doctors, nurses, and midwives. Geneva, World Health Organization, 2003.

97. Paxton A, Maine D, Hijab N. Using the UN process indicators of emergency obstetric care: Questions and answers. AMDD Workbook. New York, AMDD Program, Heilbrunn Department of Population and Family Health, 2003.

98. Pregnancy, childbirth, postpartum and newborn care: A guide to essential practice. World Health Organization, Geneva, 2003.

99. JHPIEGO. Basic maternal and newborn care: A guide for skilled providers (BMNC) http://www.jhpiego.org/scripts/pubs/category_detail.asp?category_id=24. Accessed: December 2008.

100. Maine D et al. The design and evaluation of maternal mor-tality programs. New York, Center for Population and Family Health School of Public Health Columbia University, 1997.

101. Campbell OMR, Graham WJ. Strategies for reducing ma-ternal mortality: Getting on with what works. Lancet, 2006, 368(9543):1284-1299.

102. Paxton A et al. Global patterns in availability of emergency obstetric care. International Journal of Gynecology and Obstetrics, 2006, 93(3):300-307.

103. Ipas and IHCAR. Deciding women’s lives are worth saving: Expanding the role of midlevel providers in safe abortion care. In: Issues in abortion care. Vol 7. Chapel Hill, Ipas, 2002.

104. Paxton A, Bailey P, Lobis S. The United Nations process in-dicators for emergency obstetric care: Reflections based on a decade of experience. International Journal of Gynecology and Obstetrics, 2006, 95(2):192-208.

105. UNFPA. Faire de la maternité sans risque une réalité. Ré-sultats de l’enquête sur les soins obstetricaux d’urgence en Mauritanie [making safe motherhood a reality: Results of a survey of emergency obstetric care in Mauritania]. Nouak-Nouak-chott, 2000.

106. Fauveau V, Donnay F. Can the process indicators for emergency obstetric care assess the progress of maternal mortality reduction programs? An examination of UNFPA projects 2000-2004. International Journal of Gynecology and Obstetrics, 2006, 93(3):308-316.


107. Miller S et al. Quality of care in institutionalized deliver-ies: The paradox of the Dominican Republic. International Journal of Gynecology and Obstetrics, 2003, 82(1):89-103.

108. Koblinsky M et al. Going to scale with professional skilled care. Lancet, 2006, 368(9544):1377-1386.

109. Pitchforth E et al. Development of a proxy wealth index for women utilizing emergency obstetric care in Bangladesh. Health Policy and Planning, 2007, 22(5):311-319.

110. Wilson JB et al. The maternity waiting home concept: The Nsawam, Ghana experience. International Journal of Gynecology and Obstetrics, 1997, 59 (Suppl. 2):S 165-172.

111. Renaudin P et al. Ensuring financial access to emergency obstetric care: Three years of experience with obstetric risk insurance in Nouakchott, Mauritania. International Journal of Gynecology and Obstetrics, 2007, 99(2):183-190.

112. Harvey SA et al. Skilled birth attendant competence: An initial assessment in four countries, and implications for the safe motherhood movement. International Journal of Gynecology and Obstetrics, 2004, 87(2):203-210.

113. Say L, Pattinson R, Gulmezoglu AM. WHO systematic review of maternal morbidity and mortality: The prevalence of severe acute maternal morbidity. Reproductive Health, 2004, 1(3):1-5.

114. Prual A et al. Severe maternal morbidity from direct obstetric causes in West Africa: Incidence and case fatality rates. Bulletin of the World Health Organization, 2000, 78(5):882-890.

115. Bang RA et al. Maternal morbidity during labour and the puerperium in rural homes and the need for medical at-tention: A prospective observational study in Gadchiroli, India. British Journal of Obstetrics and Gynecology, 2004, 111(3):231-238.

116. McCord C et al. Efficient and effective emergency obstetric care in a rural Indian community where most deliveries are at home. International Journal of Gynecology and Obstetrics, 2001, 75(3):297-307.

117. Bennett TA et al. Pregnancy-associated hospitalizations in the United States in 1991 and 1992: A comprehensive view of maternal morbidity. American Journal of Obstetrics and Gynecology, 1998, 178(2):346-354.

118. The Prevention of Maternal Mortality Network. Situation analyses of emergency obstetric care: Examples from eleven operations research projects in West Africa Social Science and Medicine, 1995, 40(5):657-667.

119. Unsafe abortion: Global and regional estimates of incidence of unsafe abortion and associated mortality in 2003. World Health Organization, Geneva, 2007.

120. Khan KS et al. WHO analysis of causes of maternal death: A systematic review. Lancet, 2006, 367(9516):1066-1074.

121. Healy J, Otsea K, Benson J. Counting abortions so that abortion counts: Indicators for monitoring the availability and use of abortion care services. International Journal of Gynecology and Obstetrics, 2006, 95:209-220.

122. Otsea K. Workbook for monitoring safe abortion care (SAC) service provision. Chapel Hill, North Carolina, Ipas, 2007.

123. Otsea K, Tesfaye S. Monitoring safe abortion care service provision in Tigray, Ethiopia: Report of a baseline assess-ment in public-sector facilities. Chapel Hill, North Carolina, Ipas, 2007.

124. Stanton CK et al. Reliability of data on caesarean sections in developing countries. Bulletin of the World Health Organi-zation, 2005, 83(6):449-455.

125. WHO. Appropriate technology for birth. Lancet 1985, 2(8452 ):436-437.

126. Deneux-Tharaux C et al. Postpartum maternal mortality and caesarean delivery. Obstetrics and Gynecology, 2006, 108(3 Part 1):541-548.

127. MacDorman MF et al. Infant and neonatal mortality for primary caesarean and vaginal births to women with “No indicated risk,” United States, 1998-2001 birth cohorts. Birth, 2006, 33(3):175-182.

128. Villar J et al. Caesarean delivery rates and pregnancy outcomes: The 2005 WHO global survey on maternal and perinatal health in Latin America. Lancet, 2006, 367(9525):1819-1829.

129. Betrán AP et al. Rates of caesarean section: Analysis of global, regional and national estimates. Paediatric and Perinatal Epidemiology, 2007, 21(2):98-113.

130. Essential elements of obstetric care at first referral level. World Health Organization, Geneva, 1991.

131. Flamm BL. Cesarean section: A worldwide epidemic? Birth 2000, 27(2):139-140.

132. What is the right number of caesarean sections? (Editorial) Lancet, March 22 1997, (349):815.

133. Lavender T et al. Caesarean section for non-medical reasons at term. Cochrane Database of Systematic Reviews, 2006, 3:CD004660.

134. DHS Datasets and Macro International Inc. MEASURE DHS - Stat compiler created map.www.measuredhs.com. Accessed July 24, 2007.

135. Encuesta Nacional de Salud Familiar FESAL 2002/03. Informe final San Salvador, Asociación Demográfica Salvadoreña, USAID 2004.

136. Ronsmans C, Holtz S, Stanton C. Socioeconomic differen-tials in caesarean rates in developing countries: A retro-spective analysis. Lancet, 2006, 368(9546):1516-1523.


137. Kalish RB, McCullough LB, Chervenak FA. Decision-making about caesarean delivery. Lancet, 2006, 367(9514):883-885.

138. Unmet Obstetric Need Network. Tackling unmet obstetric needs. Comparative study in Benin, Burkina Faso, Haiti, Mali, Morocco, Niger, Pakistan and Tanzania. Antwerp, Institute of Tropical Medicine, 2002.

139. Stanton C, Ronsmans C, the Baltimore Group on Caesar-ean. Recommendations for routine reporting on indication for cesarean delivery in developing countries. Birth, 2008, 35:201-211.

140. Robson MS. Classification of caesarean sections. Fetal and Maternal Medicine Review, 2001, 12:23-39.

141. Bergström S. Who will do the caesareans when there is no doctor? Finding creative solutions to the human resource crisis. British Journal of Obstetrics and Gynaecology, 2005, 112(9):1168-1169.

142. Chilopora G et al. Postoperative outcome of caesarean sections and other major emergency obstetric surgery by clinical officers and medical officers in Malawi. Human Resources for Health, 2007, 5(1):17.

143. Mbaruku G, Bergström S. Reducing maternal mortality in Kigoma, Tanzania. Health Policy and Planning, 1995, 10(1):71-78.

144. Pereira C et al. Meeting the need for emergency obstetric care in Mozambique: Work performance and histories of medical doctors and assistant medical officers trained for surgery. BJOG: An International Journal of Obstetrics and Gynaecology, 2007, 114(12):1530-1533.

145. Ministry of Family Health and Welfare. Annual report 2006-2007. New Delhi, Government of India, 2006.

146. AMDD. Improving emergency obstetric care through criterion-based audits. New York City, Mailman School of Public Health, 2002.

147. Wagaarachchi PT, et al. Conducting criterion-based clinical audit of obstetric care: A practical field guide. Aberdeen, Dugald Baird Centre for Research on Women’s Health, 2001.

148. Beyond the numbers: Reviewing maternal deaths and complications to make pregnancy safer. World Health Organization, Geneva, 2004.

149. Deneux-Tharaux C, Berg C. Underreporting of pregnancy-related mortality in the United States and Europe. Obstetrics and Gynecology, 2005, 106:684-692.

150. Edson W et al. Safe motherhood studies--timeliness of in-hospital care for treating obstetric emergencies: Results from Benin, Ecuador, Jamaica, and Rwanda. Operations research results. Washington, D.C., Quality Assurance Project, United States Agency for International Development, 2006.

151. Ifenne D et al. Improving the quality of obstetric care at the teaching hospital, Zaria, Nigeria. International Journal of Gynecology and Obstetrics, 1997, 59 (Suppl 2):S.37-46.

152. Say L, Souza J, Pattinson R. Severe acute maternal mortal-ity or near miss - towards a standard tool for monitoring quality of maternal health care. Best Practice and Research Clinical Obstetrics and Gynaecology, in press.

153. Fauveau V. New indicator of quality of emergency obstetric and newborn care. Lancet, 2007, 370(9595):1310-1310.

154. Lawn J, Shibuya K, Stein C. No cry at birth: Global estimates of intrapartum stillbirths and intrapartum-related neonatal deaths Bulletin of the World Health Organization, 2005, 83(6):409-417.

155. Lawn J. 4 million neonatal deaths: When? Where? Why? Lancet, 2005, 365:891-900.

156. Pattinson R, ed. Saving babies: A perinatal care survey of South Africa 2000. Pretoria: Medical Research Council; 2000.

157. International Statistical Classification of Diseases and Related Health Problems. Tenth Revision (ICD-10). Geneva: World Health Organization, 1992.

158. AbouZahr C. Maternal mortality overview. In: Murray CJ, Lopez AD, eds. Health dimensions of sex and reproduction Geneva, WHO, 1998:111-164.

159. Bhatia JC, Cleland J. Obstetric morbidity in south India: Results from a community survey. Social Science and Medicine, 1996, 43(10):1507-1516.

160. Koblinsky MA. Beyond maternal mortality - magnitude, interrelationship and consequences of women’s health, pregnancy-related complications and nutritional status on pregnancy outcomes. International Journal of Gynecology and Obstetrics, 1995, 48(Suppl 1):S21-S32.

161. Fortney J, Smith J. The base of the iceberg: Prevalence and perceptions of maternal morbidity in four developing coun-tries. Research Triangle Park, North Carolina, Family Health International, 1996.

162. Maine D et al. Research note: Estimating maternal deaths averted: A field-based methodology. International Journal of Gynecology and Obstetrics, 2005, 89(2):218-220.

163. EngenderHealth and AMDD. Quality improvement for emer-gency obstetric care: Leadership manual and toolbook. New York City, EngenderHealth, 2003.

Appendix A: Forms and worksheets for data collection and calculation of

EmOC indicators

Form 1 Possible EmOC facilities

Worksheet 1a List of health centres, health clinics and health posts

Worksheet 1b List of hospitals

Form 2 Review of potential EmOC facilities

Form 3 Summary of data on EmOC facilities in the area

Worksheet 3a Summary of reviews of basic EmOC facilities

Worksheet 3b Summary of reviews of comprehensive EmOC facilities

Worksheet 3c Summary of reviews of non-EmOC facilities

Worksheet 3d Summary of health centres and other lower-level facilities

Worksheet 3e Summary of hospitals

Worksheet 3f Area-wide estimates of EmOC

Form 4 Calculation of indicators for geographic area

Form 5 Calculation of indicators for a country

Worksheet 5a Amount of EmOC services

These forms are useful for collecting information. The format can be adapted if necessary. It is important that all the data be

collected in order to have a complete picture of the services available and services needed.

61Monitoringemergencyobstetriccare:ahandbook


Form 1. Possible EmOC facilities

1. Name of area

2. Population of area

3. Crude birth rate of area

4. Form completed by (list name and title)

5. Form completed on (date)

Worksheets 1a–1b need to be completed before filling in the total below.

6. Total number of health centres, health clinics and health posts

7. Total number of hospitals


Wor

kshe

et 1

a. L

ist o

f hea

lth c

entr

es, h

ealth

clin

ics

and

heal

th p

osts

Area

(pro

vinc

e, re

gion

, dis

trict

, etc

.) __

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

houl

d be

use

d to

list

all

faci

litie

s th

at p

rovi

de s

ome

mat

erni

ty c

are

but a

re n

ot h

ospi

tals

, inc

ludi

ng h

ealth

cen

tres,

hea

lth c

linic

s an

d he

alth

pos

ts.

The

easi

est w

ay to

org

aniz

e th

is in

form

atio

n is

to c

reat

e a

tabl

e in

Exc

el o

r ano

ther

sof

twar

e pa

ckag

e.

Faci

lity

nam

eLo

catio

nTy

pe o

f fac

ility

(hea

lth c

entre

, clin

ic)

Owne

rshi

p (g

over

nmen

t, pr

ivat

e, m

issi

on)

Tota

l num

ber o

f fac

ilitie

s th

at a

re n

ot h

ospi

tals

but

offe

r som

e m

ater

nity

car

e


Wor

kshe

et 1

b. L

ist o

f hos

pita

ls

Area

(pro

vinc

e, re

gion

, dis

trict

): __

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

houl

d be

use

d to

list

all

hosp

itals

that

pro

vide

mat

erni

ty c

are.

The

eas

iest

way

to o

rgan

ize

this

info

rmat

ion

is to

cre

ate

a ta

ble

in E

xcel

or a

noth

er s

oftw

are

pack

age.

Faci

lity

nam

eLo

catio

nTy

pe o

f fac

ility

(dis

trict

hos

pita

l, re

gion

al h

ospi

tal)

Owne

rshi

p (g

over

nmen

t, pr

ivat

e, m

issi

on)

Tota

l num

ber o

f hos

pita

ls o

fferin

g m

ater

nity

car

e


Form 2. Review of possible EmOC facilities

Identification

Facility name District name (or other subnational area)

Region name (or other subnational area)

Date of data collection Interviewer

Day Month Year Name

Adapt the following lists of options to the local situation.

Type of facility: (circle one)

1. National hospital 2. Regional hospital 3. District hospital 4. Maternity

5. Health centre 6. Clinic 7. Other: specify__________________________

Type of operating agency: (circle one)

1. Government 2. Private 3. Nongovernmental organization 4. Religious mission

5. Other: specify__________________

EmOC signal functions

Answer the following questions about EmOC signal functions by reviewing facility registers, through observation and if necessary

interviewing health workers in the maternity ward and other departments. Record whether the function has been performed in the past

3 months, and if not, why it has not been performed.

Consider all of the following when determining whether a particular signal function was performed:

Are staff at the facility trained to provide the service?

Are the requisite supplies and equipment present? Is the equipment functioning?

Were there no cases for which the use of a particular signal function was indicated?

Are the cadres of staff working at the facility authorized to perform the service?


Performance of signal functions

Item Performed in past

3 months?

If not performed in past 3 months, why?

(a) Administer parenteral antibiotics 0. No

1. Yes

1. Training issues

2. Supplies, equipment, drugs issue

3. Management issue

4. Policy issues

5. No indication

(b) Administer uterotonic drugs

(i.e. parenteral oxytocin)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(c) Administer parenteral

anticonvulsants for pre-eclampsia and

eclampsia (i.e. magnesium sulfate)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(d) Perform manual removal of placenta 0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(e) Perform removal of retained

products (e.g. manual vacuum

aspiration, dilation and curettage)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(f) Perform assisted vaginal delivery

(e.g. vacuum extraction, forceps

delivery)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication


Item Performed in past

3 months?

If not performed in past 3 months, why?

(g) Perform newborn resuscitation

(e.g. with bag and mask)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(h) Perform blood transfusion 0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

(i) Perform surgery

(e.g. caesarean section)

0. No

1. Yes

1. Training issues


3. Management issue

4. Policy issues

5. No indication

Training issues: Authorized cadre is available but not trained, or there is lack of confidence in providers’ skills.

Supplies, equipment issue: Supplies or equipment are not available, not functional or broken, or needed drugs are unavailable.

Management issues: Providers desire compensation to perform this function, providers are encouraged to perform alternative

procedures, or providers uncomfortable or unwilling to perform procedure for reasons unrelated to training.

Policy issues: Required level of staff is not posted to this facility in adequate numbers (or at all), or national or hospital policies do

not allow function to be performed.

No indication: No client needing this procedure came to the facility during this period.

Determination of EmOC status

Use the questions above on the performance of signal functions. Check only one category below.

If all questions a–i = Yes, tick _____ comprehensive EmOC

If all questions a–g = Yes, tick _____ basic EmOC

If any questions a–g = No, tick _____ non-EmOC


Form

2 (c

ontin

ued)

Faci

lity

case

sum

mar

y

Faci

lity

nam

e

Com

plet

e th

e fo

llow

ing

char

t by

rev

iew

ing

faci

lity

regi

ster

s an

d co

untin

g th

e nu

mbe

r of

wom

en g

ivin

g bi

rth, t

he n

umbe

r of

cae

sare

an s

ectio

ns, t

he n

umbe

r of

wom

en w

ith o

bste

tric

com

plic

atio

ns, t

he n

umbe

r of m

ater

nal d

eath

s an

d th

e nu

mbe

r of i

ntra

partu

m a

nd v

ery

early

neo

nata

l dea

ths.

Data

col

lect

ion

plan

1: F

or fa

cilit

ies

in w

hich

few

er th

an 1

0 00

0 w

omen

giv

e bi

rth p

er y

ear,

data

for a

ll 12

mon

ths

are

the

mos

t use

ful.

Data

col

lect

ion

plan

2: I

f m

ore

than

10

000

wom

en g

ive

birth

per

yea

r, se

lect

4 m

onth

s th

at p

rovi

de a

goo

d di

strib

utio

n of

dat

a th

roug

hout

the

yea

r (e

.g. m

onth

s 1,

4, 7

and

10)

.

For m

ore

info

rmat

ion

on p

lan

2, re

fer t

o se

ctio

n 3.

4 of

the

hand

book

.

EmOC

indi

cato

r dat

a

Mon

th (w

rite

the

nam

e of

a

mon

th a

bove

eac

h nu

mbe

r)

Year

12

34

56

78

910

1112

Tota

l1

No. o

f wom

en g

ivin

g bi

rth (n

orm

al

vagi

nal,

assi

sted

vag

inal

, bre

ech

and

caes

area

n)

No. o

f cae

sare

an s

ectio

ns

Dire

ct o

bste

tric

com

plic

atio

ns tr

eate

d

Haem

orrh

age

(ant

e- a

nd

post

partu

m)

Obst

ruct

ed o

r pro

long

ed la

bour

Rupt

ured

ute

rus

Post

partu

m s

epsi

s

76 This page has been left blank Monitoringemergencyobstetriccare:ahandbook 77

76 This page has been left blank Monitoringemergencyobstetriccare:ahandbook 77

Mon

th (w

rite

the

nam

e of

a

mon

th a

bove

eac

h nu

mbe

r)

Year

12

34

56

78

910

1112

Tota

l1

Seve

re p

re-e

clam

psia

or

ecla

mps

ia

Com

plic

atio

ns o

f abo

rtion

(with

hae

mor

rhag

e or

sep

sis)

Ecto

pic

preg

nanc

y

Tota

l no.

of d

irect

obs

tetr

ic

com

plic

atio

ns tr

eate

d (a

dd e

ach

colu

mn)

Othe

r dire

ct o

bste

tric

com

plic

a-

tions

that

wer

e tre

ated

but

are

not

liste

d ab

ove

or n

ot s

peci

fied;

list

som

e of

the

caus

es, i

f spe

cifie

d,

in th

e sp

ace

belo

w:


Year

12

34

56

78

910

1112

Tota

l1

Mat

erna

l dea

ths

from

di

rect

obs

tetr

ic c

ause

s

Haem

orrh

age

(ant

e- a

nd p

ostp

artu

m)

Obst

ruct

ed o

r pro

long

ed

labo

ur

Rupt

ured

ute

rus

Post

partu

m s

epsi

s

Seve

re p

re-e

clam

psia

or

ecla

mps

ia

Com

plic

atio

ns o

f abo

rtion

(with

hae

mor

rhag

e or

sep

sis)

Ecto

pic

preg

nanc

y

Tota

l no.

of m

ater

nal

deat

hs fr

om d

irect

ob

stet

ric c

ause

s (a

dd e

ach

colu

mn)

Othe

r mat

erna

l dea

ths

due

to d

irect

obs

tetri

c ca

uses

, ot

her t

han

thos

e lis

ted

abov

e or

not

spe

cifie

d; li

st s

ome

of

the

caus

es, i

f spe

cifie

d, in

th

e sp

ace

belo

w:


Year

12

34

56

78

910

1112

Tota

l1

Indi

rect

mat

erna

l dea

ths:

Lis

t cau

ses

of in

dire

ct o

bste

tric

com

plic

atio

ns a

nd m

ater

nal d

eath

s th

at a

re re

leva

nt fo

r the

loca

l con

text

(e.g

. HIV

, sev

ere

anae

mia

, mal

aria

)

Indi

rect

mat

erna

l dea

th

(put

rele

vant

cau

se h

ere)

Indi

rect

mat

erna

l dea

th

(put

rele

vant

cau

se h

ere)

All o

ther

indi

rect

mat

erna

l

deat

hs

Still

birt

hs a

nd n

eona

tal d

eath

s

Intra

partu

m d

eath

s (fr

esh

still

birth

s) ≥

2.5

kg

Very

ear

ly n

eona

tal d

eath

s

(< 2

4 h)

≥ 2

.5 k

g

1 If p

lan

2 w

as s

elec

ted

(i.e.

onl

y 4

mon

ths

of d

ata

wer

e co

llect

ed),

mul

tiply

the

tota

l of 4

mon

ths

by 3

to e

stim

ate

the

data

for 1

2 m

onth

s.


Quality of information

Item Responses

In your informed opinion (e.g. from talking to staff, looking at the

record system), what proportion of complications treated in the

facility are recorded on this form?

(tick one)

None

Some (less than half)

Most (more than half)

All

In your informed opinion (from talking to staff, looking at the record

system, etc.), what proportion of the maternal deaths that occurred

in the facility are recorded on this form?

(tick one)

None

Some

Most

All

Type of register used Yes No

Maternity ward register

Delivery register or book

General admissions register

Operating theatre register

Female ward register

Discharge register

Other:

Other:

What sources of data were used to complete this form? (e.g. maternity ward register, delivery book, general admissions

register, operating theatre register, female ward register, discharge register).


Form 3. Summary of data on EmOC facilities in the area

This form summarizes all the data on facilities within the geographical area that have been entered in all sections of form 2.

One copy of form 3 should be completed for each area.

Name of area

Population size of area

Crude birth rate (no. of births per 1000 population) of area

Expected births in area

[(crude birth rate of area ÷ 1000) x Population size of area]

Complete either section A or section B on the following page. The other section can then be deleted.

If all facilities in the area were visited, complete section A only (and delete section B).

If a subset of facilities in the area were selected, complete section B only (and delete section A).


Section A:

Use worksheets 3a–c on the following pages to complete the table below.

In a 12-month period Column 1

Basic EmOC facilities

Column 2

Comprehensive EmOC facilities

Column 3

Total no. from EmOC

facilities

(column 1+

column 2)

Column 4

Non-EmOC facilities

Column 5

Total from all facilities

surveyed

(column 3 +

column 4)

No. of facilities

No. of women giving birth

No. of women with direct

obstetric complications treated

No. of caesarean sections

No. of maternal deaths from

direct obstetric causes

No. of maternal deaths from

indirect causes

No. of intrapartum deaths

(fresh stillbirths; ≥ 2.5 kg)

+ No. of very early neonatal

deaths (≤ 24 h; ≥ 2.5 kg)


Section B:

Use worksheets 3d–f to complete the table below.

In 12-month period: Column 1

Basic EmOC facilities

Column 2

Comprehensive EmOC facilities

Column 3

Total no. from EmOC facilities

(column 1+

column 2)

Column 4

Non-EmOC facilities

Column 5

Total from all facilities

surveyed

(column 3 +

column 4)

No. of facilities

No. of women giving birth

No. of women with direct

obstetric complications

treated

No. of caesarean sections

No. of maternal deaths

from direct obstetric

causes

No. of maternal deaths

from indirect causes

No. of intrapartum deaths

(fresh stillbirths; ≥ 2.5 kg)

+ No. of very early neonatal

deaths (≤ 24 h; ≥ 2.5 kg)


Wor

kshe

et 3

a. S

umm

ary

of re

view

s of

bas

ic E

mOC

faci

litie

s

Area

: ___

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

umm

ariz

es th

e da

ta c

olle

cted

on

form

2 fr

om a

ll ba

sic

EmOC

faci

litie

s. U

se fo

rm 2

to id

entif

y al

l bas

ic E

mOC

faci

litie

s.

Atta

ch a

dditi

onal

she

ets

if ne

cess

ary.

Exce

l or a

noth

er s

oftw

are

pack

age

can

be u

sed

for t

his

sum

mar

y.

Colu

mn

1Co

lum

n 2

Colu

mn

3Co

lum

n 4

Colu

mn

5Co

lum

n 6

Colu

mn

7

Faci

lity

No. o

f wom

en g

ivin

g

birth

No. o

f wom

en w

ith

dire

ct o

bste

tric

com

plic

atio

ns tr

eate

d

No. o

f cae

sare

an

sect

ions

No. o

f mat

erna

l dea

ths

from

dire

ct o

bste

tric

caus

es

No. o

f mat

erna

l dea

ths

from

indi

rect

cau

ses

No. o

f int

rapa

rtum

deat

hs (f

resh

stil

l-

birth

s; ≥

2.5

kg)

+ N

o.

of v

ery

early

neo

nata

l

deat

hs

(≤ 2

4 h;

≥ 2

.5 k

g)

Colu

mn

tota

ls*

*If m

ore

than

one

she

et w

as u

sed,

add

she

et to

tals

to o

btai

n th

e ov

eral

l tot

al.

Tota

l num

ber o

f bas

ic E

mOC

faci

litie

s lis

ted

in c

olum

n 1

=


Wor

kshe

et 3

b. S

umm

ary

of re

view

s of

com

preh

ensi

ve E

mOC

faci

litie

s

Area

: ___

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

umm

ariz

es th

e da

ta c

olle

cted

on

form

2 fr

om a

ll co

mpr

ehen

sive

Em

OC fa

cilit

ies.

Use

form

2 to

iden

tify

all c

ompr

ehen

sive

Em

OC fa

cilit

ies.

Atta

ch a

dditi

onal

she

ets

if ne

cess

ary.

Exce

l or a

noth

er s

oftw

are

pack

age

can

be u

sed

for t

his

sum

mar

y.

Colu

mn

1Co

lum

n 2

Colu

mn

3Co

lum

n 4

Colu

mn

5Co

lum

n 6

Colu

mn

7

Faci

lity

No. o

f wom

en g

ivin

g

birth

No. o

f wom

en w

ith

dire

ct o

bste

tric

com

plic

atio

ns tr

eate

d

No. o

f cae

sare

an

sect

ions

No. o

f mat

erna

l dea

ths

from

dire

ct o

bste

tric

caus

es

No. o

f mat

erna

l dea

ths

from

indi

rect

cau

ses

No. o

f int

rapa

rtum

deat

hs (f

resh

stil

l-

birth

s; >

2.5

kg)

+ N

o.

of v

ery

early

neo

nata

l

deat

hs (≤

24

hour

s;

> 2

.5 k

g)

Colu

mn

tota

ls*

*If m

ore

than

one

she

et w

as u

sed,

add

she

et to

tals

to o

btai

n th

e ov

eral

l tot

al.

Tota

l num

ber o

f com

preh

ensi

ve E

mOC

faci

litie

s lis

ted

in c

olum

n 1

=


Wor

kshe

et 3

c. S

umm

ary

of re

view

s of

non

-Em

OC fa

cilit

ies

Area

: ___

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

umm

ariz

es th

e da

ta c

olle

cted

on

form

2 fr

om a

ll no

n-Em

OC fa

cilit

ies.

Use

form

2 to

iden

tify

all n

on-E

mOC

faci

litie

s.

Atta

ch a

dditi

onal

she

ets

if ne

cess

ary.

Exce

l or a

noth

er s

oftw

are

pack

age

can

be u

sed

for t

his

sum

mar

y.

Colu

mn

1Co

lum

n 2

Colu

mn

3Co

lum

n 4

Colu

mn

5Co

lum

n 6

Colu

mn

7

Faci

lity

No. o

f wom

en g

ivin

g

birth

No. o

f wom

en w

ith

dire

ct o

bste

tric

com

plic

atio

ns tr

eate

d

No. o

f cae

sare

an

sect

ions

No. o

f mat

erna

l dea

ths

from

dire

ct o

bste

tric

caus

es

No. o

f mat

erna

l dea

ths

from

indi

rect

cau

ses

No. o

f int

rapa

rtum

deat

hs (f

resh

stil

l-

birth

s; ≥

2.5

kg)

+ N

o.

of v

ery

early

neo

nata

l

deat

hs

(≤ 2

4 h;

≥ 2

.5 k

g)

Colu

mn

tota

ls*

* If

mor

e th

an o

ne s

heet

was

use

d, a

dd s

heet

tota

ls to

obt

ain

the

over

all t

otal

.

Tota

l num

ber o

f non

-Em

OC fa

cilit

ies

liste

d in

col

umn

1 =


Wor

kshe

et 3

d. S

umm

ary

of h

ealth

cen

tres

and

oth

er lo

wer

-lev

el fa

cilit

ies

Area

: ___

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

umm

ariz

es d

ata

colle

cted

on

form

2 fr

om a

ll he

alth

cen

tres

and

othe

r low

er-le

vel f

acili

ties

in a

geo

grap

hica

l are

a. U

se fo

rm 2

to id

entif

y th

e Em

OC s

tatu

s of

the

heal

th

cent

res

and

othe

r low

er-le

vel f

acili

ties

visi

ted.

Atta

ch a

dditi

onal

she

ets

if ne

cess

ary.

Exce

l or a

noth

er s

oftw

are

pack

age

can

be u

sed

for t

his

sum

mar

y.

Colu

mn

1Co

lum

n 2

Colu

mn

3Co

lum

n 4

Colu

mn

5Co

lum

n 6

Colu

mn

7Co

lum

n 8

Faci

lity

EmOC

sta

tus

(bas

ic,

com

preh

ensi

ve o

r

none

)

No. o

f wom

en

givi

ng b

irth

No. o

f wom

en w

ith

dire

ct o

bste

tric

com

plic

atio

ns

treat

ed

No. o

f cae

sare

an

sect

ions

No. o

f mat

erna

l

deat

hs fr

om d

irect

obst

etric

cau

ses

No. o

f mat

erna

l

deat

hs fr

om

indi

rect

cau

ses

No. o

f int

rapa

rtum

deat

hs (f

resh

stil

l-

birth

s; (≥

2.5

kg)

+ N

o. o

f ver

y ea

rly

neon

atal

dea

ths

(≤ 2

4 h;

≥ 2

.5 k

g)

Colu

mn

tota

ls*

* If

mor

e th

an o

ne s

heet

was

use

d, a

dd s

heet

tota

ls to

obt

ain

the

over

all t

otal

Tota

l num

ber o

f bas

ic E

mOC

faci

litie

s lis

ted

in c

olum

n 2

=

Tota

l num

ber o

f com

preh

ensi

ve E

mOC

faci

litie

s lis

ted

in c

olum

n 2

=

Tota

l num

ber o

f non

-Em

OC fa

cilit

ies

liste

d in

col

umn

2 =


Wor

kshe

et 3

e. S

umm

ary

of h

ospi

tals

Area

: ___

____

____

____

____

____

____

____

____

____

____

___

This

wor

kshe

et s

umm

ariz

es d

ata

colle

cted

on

form

2 fr

om a

ll ho

spita

ls in

a g

eogr

aphi

cal a

rea.

Use

form

2 to

iden

tify

the

EmOC

sta

tus

of th

e ho

spita

ls v

isite

d.

Atta

ch a

dditi

onal

she

ets

if ne

cess

ary.

Exce

l or a

noth

er s

oftw

are

pack

age

can

be u

sed

for t

his

sum

mar

y.

Colu

mn

1Co

lum

n 2

Colu

mn

3Co

lum

n 4

Colu

mn

5Co

lum

n 6

Colu

mn

7Co

lum

n 8

Faci

lity

EmOC

sta

tus

(bas

ic, c

ompr

ehen

-

sive

or n

one)

No. o

f wom

en

givi

ng b

irth

NNo.

of w

omen

with

dire

ct o

bste

tric

com

plic

atio

ns

treat

ed

No. o

f cae

sare

an

sect

ions

No. o

f mat

erna

l

deat

hs fr

om d

irect

obst

etric

cau

ses

No. o

f mat

erna

l

deat

hs fr

om

indi

rect

cau

ses

No. o

f int

rapa

rtum

deat

hs (f

resh

stil

l-

birth

s; ≥

2.5

kg)

+

No. o

f ver

y ea

rly

neon

atal

dea

ths

(≤ 2

4 h;

≥ 2

.5 k

g)

Colu

mn

tota

ls*

*If m

ore

than

one

she

et w

as u

sed,

add

she

et to

tals

to o

btai

n th

e ov

eral

l tot

al.

Tota

l num

ber o

f bas

ic E

mOC

faci

litie

s lis

ted

in c

olum

n 2

=

Tota

l num

ber o

f com

preh

ensi

ve E

mOC

faci

litie

s lis

ted

in c

olum

n 2

=

Tota

l num

ber o

f non

-Em

OC fa

cilit

ies

liste

d in

col

umn

2 =


Worksheet 3f. Area-wide estimates of EmOC

Area: ______________________________________________

This worksheet allows conversion of the data from the subset of facilities that were selected for site visits into estimates

for the entire area.

If a subset of health centres (and other lower-level facilities) were selected for study:

No. of health centres (or other) visited in area

Total no. of health centres (or other) in area

Proportion of health centres (or other) for which data were collected (No. of health centres

visited in area ÷ Total no. of health centres in area)

Use worksheet 3d for the health centres (and other lower-level facilities) studied.

Totals from facilities visited

÷ Proportion of health centres

visited

(see chart above)

= Estimate for area

Estimated no. of basic EmOC facilities ÷ =

Estimated no. of comprehensive EmOC facilities ÷ =

Estimated no. of non-EmOC facilities ÷ =

Estimated no. of women giving birth in facilities

classified as basic and comprehensive facilities

÷ =


classified as non-EmOC facilities

÷ =

Estimated no. of women with direct obstetric com-

plications treated in facilities classified as basic

and comprehensive facilities

÷ =

Estimated no. of women with direct obstetric

complications treated in facilities classified as non-

EmOC facilities

÷ =

Estimated no. of caesarean sections in facilities

classified as basic and comprehensive facilities

÷ =


classified as non-EmOC facilities

÷ =

Estimated no. of maternal deaths from direct

obstetric causes in facilities classified as basic and

comprehensive

÷ =



÷ Proportion of health centres

visited

(see chart above)

= Estimate for area


obstetric causes in facilities classified as non-

EmOC

÷ =

Estimated no. of maternal deaths from indirect

causes in facilities classified as basic and

comprehensive

÷ =


causes in facilities classified as non-EmOC

÷ =

Estimated no. of intrapartum deaths (fresh

stillbirths; ≥ 2.5 kg) and very early neonatal deaths

(≤ 24 h; ≥ 2.5 kg) in facilities classified as basic

and comprehensive

÷ =


stillbirths; ≥ 2.5 kg) and very early neonatal deaths

(≤ 24 h; ≥ 2.5 kg) in facilities classified as

non-EmOC

÷ =

If a sub-set of hospitals was selected for study:

No. of hospitals visited in area

Total no. of hospitals in area

Proportion of hospitals for which data were collected above (No. of hospitals visited in area ÷

Total no. of hospitals in area)


Use worksheet 3e for the hospitals studied.


÷ Proportion of hos-pitals visited (see

chart above)

= Estimate for area

Estimated no. of basic EmOC facilities ÷ =

Estimated no. of comprehensive EmOC facilities ÷ =

Estimated no. of non-EmOC facilities ÷ =


classified as basic and comprehensive

÷ =


classified as non-EmOC

÷ =


complications treated in facilities classified as

basic and comprehensive

÷ =


complications treated in facilities classified as non-

EmOC

÷ =


classified as basic and comprehensive

÷ =


classified as non-EmOC

÷ =


obstetric causes in facilities classified as basic and

comprehensive

÷ =


obstetric causes in facilities classified as non-

EmOC

÷ =


causes in facilities classified as basic and

comprehensive

÷ =


causes in facilities classified as non-EmOC

÷ =


stillbirths; ≥ 2.5 kg) and of very early neonatal

deaths (≤ 24 h; ≥ 2.5 kg) in facilities classified as

basic and comprehensive

÷ =


stillbirths; ≥ 2.5 kg) and of very early neonatal

deaths (≤ 24 h; ≥ 2.5 kg) in facilities classified as

non-EmOC

÷ =


Form

4.

Calc

ulat

ion

of in

dica

tors

for g

eogr

aphi

c ar

ea

Use

form

3, s

ectio

n A

or B

, to

calc

ulat

e th

e in

dica

tors

bel

ow.

Area

: ___

____

____

____

____

_

Indi

cato

r 1: A

vaila

bilit

y of

Em

OC

Is a

ccep

tabl

e le

vel m

et?

Tota

l no.

of b

asic

+

com

preh

ensi

ve E

mOC

faci

litie

s in

are

a

Popu

latio

n of

are

a

Indi

cato

r 1a

No. o

f Em

OC fa

cilit

ies

per 5

00 0

00

popu

latio

n

Min

imum

acc

epta

ble

leve

l

≥ 5

per 5

00 0

00 p

opul

atio

n

(÷

)X

500

000

=o

Met

o N

ot m

et

Tota

l no.

of c

ompr

ehen

sive

EmOC

faci

litie

s in

are

aPo

pula

tion

of a

rea

Indi

cato

r 1b

No. o

f com

preh

ensi

ve E

mOC

faci

litie

s

per 5

00 0

00 p

opul

atio

n

Min

imum

acc

epta

ble

leve

l

≥ 1

per 5

00 0

00 p

opul

atio

n

(÷

)X

500

000

=o

Met

o N

ot m

et


Indicator 2: Geographical distribution of EmOC facilities

This indicator is generally intended for use at the national level. In large areas (e.g. with millions of inhabitants), it is reasonable

to calculate the distribution of EmOC facilities for subareas. This can be done by repeating the steps above (in Indicator 1), and

then calculating the percentage of subareas meeting the minimum acceptable levels. The minimum acceptable level for this

indicator is 100%.

Another option is to lay the facilities in the area on a map that shows roads and topographic areas, to identify problems of access

and showing referral systems. This can be done with a geographical information system or another mapping method.


Indi

cato

r 3:

Prop

ortio

n of

all

birt

hs in

Em

OC fa

cilit

ies

and

all s

urve

yed

faci

litie

s

Tota

l no.

of w

omen

giv

ing

birth

in

EmOC

faci

litie

s in

are

aEx

pect

ed b

irths

in a

rea

Indi

cato

r 3a

Prop

ortio

n of

birt

hs in

Em

OC fa

cilit

ies

Min

imum

acc

epta

ble

leve

l: ta

rget

s to

be

set

loca

lly

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of w

omen

giv

ing

birth

in

all s

urve

yed

faci

litie

s in

are

aEx

pect

ed b

irths

in a

rea

Indi

cato

r 3b

Prop

ortio

n of

birt

hs in

all

surv

eyed

faci

litie

s

Min

imum

acc

epta

ble

leve

l: ta

rget

s to

be

set

loca

lly

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et


Indi

cato

r 4:

Met

nee

d fo

r Em

OC

No. o

f wom

en w

ith d

irect

obse

tric

com

plic

atio

ns

treat

ed in

Em

OC fa

cilit

ies

in a

rea

No. o

f exp

ecte

d bi

rths

in a

rea

Indi

cato

r 4a

Prop

ortio

n of

wom

en e

stim

ated

to h

ave

obst

etric

com

plic

atio

ns w

ho a

re tr

eate

d in

EmOC

faci

litie

s

Acce

ptab

le le

vel =

100

%

oM

et

oNo

t met

÷(

X0.

15*

)=

____

__ x

100

= _

____

_ %

No. o

f wom

en w

ith d

irect

obse

tric

com

plic

atio

ns in

all s

urve

yed

faci

litie

s

in a

rea

No. o

f exp

ecte

d bi

rths

in a

rea

Indi

cato

r 4b

Prop

ortio

n of

wom

en e

stim

ated

to h

ave

obst

etric

com

plic

atio

ns w

ho a

re tr

eate

d in

all s

urve

yed

faci

litie

s

Acce

ptab

le le

vel =

100

%

oM

et

oNo

t met

÷(

X0.

15*

)=

____

__ x

100

= _

____

_ %

* Ex

pect

ed b

irths

are

mul

tiplie

d by

0.1

5 to

est

imat

e th

e to

tal o

bste

tric

com

plic

atio

ns in

the

popu

latio

n.


Indi

cato

r 5:

Caes

area

n se

ctio

ns a

s a

prop

ortio

n of

all

birt

hs

Tota

l no.

of c

aesa

rean

sec

tions

in

EmOC

faci

litie

s in

are

a

Expe

cted

birt

hs in

are

aIn

dica

tor 5

a

Caes

area

n se

ctio

ns in

Em

OC fa

cilit

ies

as a

prop

ortio

n of

all

birth

s

Acce

ptab

le le

vel:

5–15

%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of c

aesa

rean

sec

tions

in

all s

urve

yed

faci

litie

s in

are

aEx

pect

ed b

irths

in a

rea

Indi

cato

r 5b

Caes

area

n se

ctio

ns in

all

surv

eyed

faci

litie

s as

a

prop

ortio

n of

all

birth

s

Acce

ptab

le le

vel:

5–15

%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et


Indi

cato

r 6:

Dire

ct o

bste

tric

cas

e fa

talit

y ra

te

Tota

l no.

of m

ater

nal d

eath

s fro

m

dire

ct o

bste

tric

caus

es in

Em

OC

faci

litie

s in

are

a

Tota

l no.

of w

omen

with

obst

etric

com

plic

atio

ns in

EmOC

faci

litie

s in

are

a

Indi

cato

r 6a

Dire

ct o

bste

tric

case

fata

lity

rate

in E

mOC

faci

litie

s

Acce

ptab

le le

vel:

≤ 1%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of m

ater

nal d

eath

s fro

m

dire

ct o

bste

tric

caus

es in

all

sur-

veye

d fa

cilit

ies

in a

rea

Tota

l no.

of w

omen

with

obst

etric

com

plic

atio

ns in

all

surv

eyed

faci

litie

s in

are

a

Indi

cato

r 6b

Dire

ct o

bste

tric

case

fata

lity

rate

in a

ll su

rvey

ed

faci

litie

s

Acce

ptab

le le

vel:

≤ 1%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

In a

dditi

on to

the

aggr

egat

ed c

alcu

latio

ns, t

he d

irect

obs

tetri

c ca

se fa

talit

y ra

te s

houl

d be

cal

cula

ted

for e

ach

hosp

ital.

The

resu

lts c

an b

e pr

esen

ted

as a

bar

cha

rt: th

e ho

rizon

tal a

xis

shou

ld b

e la

belle

d w

ith th

e fa

cilit

y na

mes

, and

the

verti

cal a

xis

shou

ld b

e la

belle

d “D

irect

obs

tetri

c ca

se fa

talit

y ra

te (%

)”. A

noth

er w

ay o

f pre

sent

ing

faci

lity-

base

d re

sults

is to

giv

e th

e

rang

e of

dire

ct o

bste

tric

case

fata

lity

rate

s fro

m d

iffer

ent h

ospi

tals

as

wel

l as

the

aggr

egat

e di

rect

cas

e fa

talit

y ra

te.


Indi

cato

r 7:

Intr

apar

tum

and

ver

y ea

rly n

eona

tal d

eath

rate

Tota

l no.

of i

ntra

partu

m d

eath

s

(≥ 2

.5 k

g) +

ver

y ea

rly n

eona

-

tal d

eath

s (≤

24

h; ≥

2.5

kg)

in

EmOC

faci

litie

s in

are

a

Tota

l no.

of w

omen

giv

ing

birth

in E

mOC

faci

litie

s in

are

a

Indi

cato

r 7a

Intra

partu

m a

nd v

ery

early

neo

nata

l dea

th ra

te in

EmOC

faci

litie

s

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %

Tota

l no.

of i

ntra

partu

m d

eath

s

(≥ 2

.5 k

g) +

ver

y ea

rly n

eona

tal

deat

hs (≤

24

h; ≥

2.5

kg)

in a

ll

surv

eyed

faci

litie

s in

are

a

Tota

l no.

of w

omen

giv

ing

birth

in a

ll su

rvey

ed fa

cilit

ies

in a

rea

Indi

cato

r 7b

Intra

partu

m a

nd v

ery

early

neo

nata

l dea

th ra

te in

all

surv

eyed

faci

litie

s

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %


Indi

cato

r 8:

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s

Tota

l no.

of m

ater

nal d

eath

s fro

m

indi

rect

cau

ses

in E

mOC

faci

litie

s

in a

rea

Tota

l no.

of m

ater

nal d

eath

s

from

all

caus

es in

Em

OC

faci

litie

s in

are

a

Indi

cato

r 8a

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s

in E

mOC

faci

litie

sAc

cept

able

leve

l:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %

Tota

l no.

of m

ater

nal d

eath

s fro

m

indi

rect

cau

ses

in a

ll su

rvey

ed

faci

litie

s in

are

a

Tota

l no.

of m

ater

nal d

eath

s

from

all

caus

es in

all

surv

eyed

faci

litie

s in

are

a

Indi

cato

r 8b

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s

in a

ll su

rvey

ed fa

cilit

ies

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %


Form

5.

Calc

ulat

ion

of in

dica

tors

for a

cou

ntry

Com

plet

e w

orks

heet

5a

befo

re c

alcu

latin

g th

e in

dica

tors

bel

ow.

Indi

cato

r 1:

Avai

labi

lity

of E

mOC

ser

vice

s

Is a

ccep

tabl

e le

vel m

et?

Tota

l no.

of b

asic

+

com

preh

ensi

ve E

mOC

faci

litie

s

Tota

l pop

ulat

ion

(wor

kshe

et 5

a; c

olum

n 3

tota

l)

Indi

cato

r 1a

No. o

f Em

OC fa

cilit

ies

per 5

00 0

00

popu

latio

n

Min

imum

acc

epta

ble

leve

l

≥ 5

per 5

00 0

00 p

opul

atio

n

(÷

)X

500

000

=o

Met

o N

ot m

et

Tota

l no.

of c

ompr

ehen

sive

EmOC

faci

litie

sTo

tal p

opul

atio

n

Indi

cato

r 1b

No. o

f com

preh

ensi

ve E

mOC

faci

litie

s

per 5

00 0

00 p

opul

atio

n

Min

imum

acc

epta

ble

leve

l

≥ 1

per 5

00 0

00 p

opul

atio

n

(÷

)X

500

000

=o

Met

o N

ot m

et


Indi

cato

r 2:

Geog

raph

ical

dis

trib

utio

n of

Em

OC fa

cilit

ies

No. o

f are

as in

cou

ntry

mee

ting

min

imum

leve

ls (i

.e. a

t lea

st 5

faci

litie

s pe

r 500

000

pop

ulat

ion

incl

udin

g at

leas

t 1 c

ompr

ehen

sive

faci

lity)

No. o

f are

as in

cou

ntry

Indi

cato

r 2

Prop

ortio

n of

are

as w

ith th

e m

inim

um a

ccep

tabl

e

num

ber o

f Em

OC fa

cilit

ies

Acce

ptab

le le

vel:

100%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et


Indi

cato

r 3: P

ropo

rtio

n of

all

birt

hs in

Em

OC fa

cilit

ies

and

all s

urve

yed

faci

litie

s

Tota

l no.

of w

omen

giv

ing

birth

in a

ll

EmOC

faci

litie

sTo

tal e

xpec

ted

birth

sIn

dica

tor 3

a

Prop

ortio

n of

all

birth

s in

Em

OC fa

cilit

ies

Min

imum

acc

epta

ble

leve

l:

targ

ets

to b

e se

t loc

ally

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of w

omen

giv

ing

birth

in a

ll

surv

eyed

faci

litie

sTo

tal n

o. o

f exp

ecte

d bi

rths

Indi

cato

r 3b

Prop

ortio

n of

all

birth

s in

all

surv

eyed

faci

litie

s

Min

imum

acc

epta

ble

leve

l:

targ

ets

to b

e se

t loc

ally

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et


Indi

cato

r 4:

Met

nee

d fo

r Em

OC

Tota

l no.

of w

omen

with

dire

ct o

bste

tric

com

plic

atio

ns

treat

ed in

all

EmOC

faci

litie

s

Tota

l no.

of e

xpec

ted

birth

s

Indi

cato

r 4a

Prop

ortio

n of

wom

en e

stim

ated

to h

ave

obst

etric

com

plic

atio

ns w

ho a

re tr

eate

d in

EmOC

faci

litie

s

Acce

ptab

le le

vel =

100

%

oM

et

oNo

t met

÷(

x0.

15*

)=

____

__ x

100

= _

____

_ %

Tota

l no.

of w

omen

with

dire

ct o

bste

tric

com

plic

a-

tions

trea

ted

in a

ll su

rvey

ed

faci

litie

s

Tota

l no.

of e

xpec

ted

birth

s

Indi

cato

r 4b

Prop

ortio

n of

wom

en e

stim

ated

to h

ave

obst

etric

com

plic

atio

ns w

ho a

re tr

eate

d in

all

surv

eyed

faci

litie

s

Acce

ptab

le le

vel =

100

%

oM

et

oNo

t met

÷(

x0.

15*

)=

____

__ x

100

= _

____

_ %

* Ex

pect

ed b

irths

are

mul

tiplie

d by

0.1

5 to

est

imat

e th

e to

tal o

bste

tric

com

plic

atio

ns in

the

popu

latio

n.


Indi

cato

r 5:

Caes

area

n se

ctio

ns a

s a

prop

ortio

n of

all

birt

hs

Tota

l no.

of c

aesa

rean

sec

tions

in a

ll

EmOC

faci

litie

sTo

tal n

o. o

f exp

ecte

d bi

rths

Indi

cato

r 5a

Caes

area

n se

ctio

ns in

Em

OC fa

cilit

ies

as a

pro

porti

on

of a

ll bi

rths

Acce

ptab

le le

vel:

5–15

%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of c

aesa

rean

sec

tions

in a

ll

surv

eyed

faci

litie

sTo

tal n

o. o

f exp

ecte

d bi

rths

Indi

cato

r 5b

Caes

area

n se

ctio

ns in

all

surv

eyed

faci

litie

s as

a p

ro-

porti

on o

f all

birth

s

Acce

ptab

le le

vel:

5–15

%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et


Indi

cato

r 6:

Dire

ct o

bste

tric

cas

e fa

talit

y ra

te

Tota

l no.

of m

ater

nal d

eath

s fro

m

dire

ct o

bste

tric

caus

es in

all

EmOC

faci

litie

s

Tota

l no.

of w

omen

with

dire

ct

obst

etric

com

plic

atio

ns in

all

EmOC

faci

litie

s

Indi

cato

r 6a

Dire

ct o

bste

tric

case

fata

lity

rate

in E

mOC

faci

litie

s

Acce

ptab

le le

vel:

≤ 1%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

Tota

l no.

of m

ater

nal d

eath

s fro

m

dire

ct o

bste

tric

caus

es in

all

surv

eyed

faci

litie

s

Tota

l no.

of w

omen

with

dire

ct

obst

etric

com

plic

atio

ns in

all

surv

eyed

faci

litie

s

Indi

cato

r 6b

Dire

ct o

bste

tric

case

fata

lity

rate

in a

ll su

rvey

ed

faci

litie

s

Acce

ptab

le le

vel:

≤ 1%

÷=

____

____

_ x

100

= _

____

___

%o

Met

o N

ot m

et

In a

dditi

on, t

he d

irect

obs

tetri

c ca

se fa

talit

y ra

te s

houl

d be

cal

cula

ted

for a

ll ho

spita

ls in

eac

h su

bare

a. T

he re

sults

can

be

pres

ente

d as

a b

ar c

hart:

the

horiz

onta

l axi

s sh

ould

be

labe

lled

with

the

suba

rea

nam

es, a

nd th

e ve

rtica

l axi

s sh

ould

be

labe

lled

“Dire

ct o

bste

tric

case

fata

lity

rate

(%)”

. Ano

ther

way

of p

rese

ntin

g th

e re

sults

is to

giv

e th

e ra

nge

of c

ase

fata

lity

rate

s

from

dire

ct o

bste

tric

caus

es fo

r sub

area

s.


Indi

cato

r 7:

Intr

apar

tum

and

ver

y ea

rly n

eona

tal d

eath

rate

Tota

l no.

of i

ntra

partu

m d

eath

s

(≥ 2

.5 k

g) +

ver

y ea

rly n

eona

tal

deat

hs (≤

24

h; ≥

2.5

kg)

in a

ll

EmOC

faci

litie

s

Tota

l no.

of w

omen

giv

ing

birth

in a

ll Em

OC fa

cilit

ies

Indi

cato

r 7a

Intra

partu

m a

nd v

ery

early

neo

nata

l dea

th ra

te in

EmOC

faci

litie

s

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %

Tota

l no.

of i

ntra

partu

m d

eath

s

(≥ 2

.5 k

g) +

ver

y ea

rly n

eona

tal

deat

hs (≤

24

h; ≥

2.5

kg)

in a

ll

surv

eyed

faci

litie

s

Tota

l no.

of w

omen

giv

ing

birth

in a

ll su

rvey

ed fa

cilit

ies

Indi

cato

r 7b

Intra

partu

m a

nd v

ery

early

neo

nata

l dea

th ra

te in

all

surv

eyed

faci

litie

s

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %


Indi

cato

r 8:

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s

Tota

l no.

of m

ater

nal d

eath

s fro

m

indi

rect

cau

ses

in a

ll Em

OC fa

cilit

ies

Tota

l no.

of m

ater

nal d

eath

s

from

all

caus

es in

all

EmOC

faci

litie

s

Indi

cato

r 8a

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s in

EmOC

faci

litie

sAc

cept

able

leve

l:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %

Tota

l no.

of m

ater

nal d

eath

s fro

m

indi

rect

cau

ses

in a

ll su

rvey

ed

faci

litie

s

Tota

l no.

of m

ater

nal d

eath

s

from

all

caus

es in

all

surv

eyed

faci

litie

s

Indi

cato

r 8b

Prop

ortio

n of

mat

erna

l dea

ths

due

to in

dire

ct c

ause

s in

all f

acili

ties

Acce

ptab

le le

vel:

oNo

sta

ndar

d ha

s be

en s

et

oNo

t app

licab

le÷

=__

____

___

x 10

0 =

___

____

_ %


Worksheet 5a. Amount of EmOC services

Use forms 3 and 4 to fill in the information below.

Name of area No. of basic

EmOC facilities

in area

No. of comp-

rehensive EmOC

facilities in area

Population of

area

Has the minimum level of

EmOC been met?

If yes, please tick in column.

Column totals*

* If more than one sheet is used, add sheet totals to obtain the overall column total.


Wor

kshe

et 5

a (c

ontin

ued)

Nam

e of

are

aTo

tal n

o. o

f

expe

cted

birt

hs

Tota

l no.

of

wom

en g

ivin

g

birth

in a

ll

EmOC

faci

litie

s

Tota

l no.

of

wom

en g

ivin

g

birth

in a

ll

surv

eyed

faci

litie

s

Tota

l no.

of w

omen

with

com

plic

atio

ns in

all E

mOC

faci

litie

s

Tota

l no.

of w

omen

with

com

plic

atio

ns in

all s

urve

yed

faci

litie

s

Tota

l no.

of

caes

area

n

sect

ions

in a

ll

EmOC

faci

litie

s

Tota

l no.

of

caes

area

n

sect

ions

in

all s

urve

yed

faci

litie

s

Colu

mn

tota

ls*

*If m

ore

than

one

she

et is

use

d, a

dd s

heet

tota

ls to

obt

ain

the

over

all c

olum

n to

tal.


Nam

e of

are

aTo

tal n

o. o

f

mat

erna

l dea

ths

from

dire

ct

obst

etric

cau

ses

in

all E

mOC

faci

litie

s

Tota

l no.

of

mat

erna

l dea

ths

from

dire

ct

obst

etric

cau

ses

in a

ll su

rvey

ed

faci

litie

s

Tota

l no.

of

mat

erna

l dea

ths

from

indi

rect

caus

es in

all

EmOC

faci

litie

s

Tota

l no.

of

mat

erna

l dea

ths

from

indi

rect

caus

es in

all

surv

eyed

faci

litie

s

Tota

l no.

of

intra

partu

m d

eath

s

(≥ 2

.5 k

g) a

nd v

ery

early

neo

nata

l dea

ths

(≤ 2

4 h;

≥ 2

.5 k

g) in

all E

mOC

faci

litie

s

Tota

l no.

of

intra

partu

m d

eath

s

(≥ 2

.5 k

g) a

nd v

ery

early

neo

nata

l dea

ths

(≤ 2

4 h;

≥ 2

.5 k

g) in

all s

urve

yed

faci

litie

s

Colu

mn

tota

ls*

*If m

ore

than

one

she

et is

use

d, a

dd s

heet

tota

ls to

obt

ain

the

over

all c

olum

n to

tal.


Appendix B: Information on registers and data collection

Signal functions:

To determine whether a facility offers each of the signal functions, data collectors should:

• observe the availability of requisite drugs, supplies, and equipment;

• interview health workers in the maternity ward and other departments; and

• review facility registers (see below).

It is important to consider all the following when determining whether a particular signal function was provided:

• Is staff at the facility trained to perform the service?

• Do the requisite supplies and equipment exist? Are they functioning?

• Were there any cases for which a particular signal function was indicated?

• Are the cadres of staff working at the facility authorized to perform the service?

Other variables:

To collect the data necessary to calculate the EmOC indicators, data from registers in many different rooms or departments at the facility must be reviewed and abstracted. The table below provides an overview of where to look for different variables.


Regi

ster

s an

d ot

her s

ourc

es to

be

used

to c

olle

ct d

ata

for t

he E

mOC

indi

cato

rs

Info

rmat

ion

Regi

ster

s in

mat

erni

ty w

ard

(incl

udin

g th

ose

foun

d in

: lab

our,

deliv

ery,

pre-

and

post

partu

m a

nd

neon

atal

room

s)

Regi

ster

s in

ope

ratin

g

thea

tre (i

nclu

ding

thos

e fo

r maj

or a

nd

min

or s

urge

ry)

Regi

ster

s in

the

fem

ale

or

gyna

ecol

ogic

al

war

ds (i

nclu

ding

post

-abo

rtion

car

e

regi

ster

s)

Regi

ster

s in

the

outp

atie

nt d

epar

tmen

t

Regi

ster

s fo

r

inpa

tient

s an

d

adm

issi

ons

Over

all a

dmin

istra

tion

(incl

udin

g re

cord

s

and

regi

ster

s in

the

mor

gue,

reco

rds

offic

e, h

ead

heal

th

wor

kers

’ offi

ce)

No. o

f wom

en g

ivin

g bi

rthX

No. o

f wom

en w

ith o

bste

tric

com

plic

atio

nsX

XX

XX

X

No. o

f cae

sare

an s

ectio

nsX

X

No. o

f mat

erna

l dea

ths

due

to d

irect

obs

tetri

c ca

uses

XX

XX

X

No. o

f mat

erna

l dea

ths

due

to in

dire

ct o

bste

tric

caus

esX

XX

XX

No. o

f fre

sh s

tillb

irths

and

intra

partu

m d

eath

s ≥

2.5

kgX

XX

No. o

f ver

y ea

rly n

eona

tal

deat

hs (1

st 2

4 h)

≥ 2

.5 k

gX

X


As can be seen from the table above, the registers in maternity departments should, in theory, contain a lot of the data necessary

to calculate the EmOC indicators; however, it is likely that they will not have all of the data needed. Monitoring should help facility

managers to perceive the need for maintaining good quality, complete records and will help them to improve record-keeping

systems.

Some of the most important columns that should be included in maternity registers are:

• admission time and date;

• mode of delivery (normal vaginal, assisted vaginal, caesarean section);

• obstetric complications (e.g. antepartum haemorrhage, postpartum haemorrhage, obstructed labour, prolonged labour,

pre-eclampsia, eclampsia, ruptured uterus, postpartum sepsis, complications of abortion, ectopic pregnancies) (Cases

of complications of abortion and ectopic pregnancies will usually be found in other departments in the facility, such as

the female or gynaecology ward, operating theatres or outpatient registers.);

• treatment or intervention provided to woman, including time of intervention (e.g. magnesium sulfate administered,

oxytocin provided, manual removal of the placenta);

• treatment or intervention provided to newborn, including time of intervention (e.g. resuscitated);

• outcome of mother (e.g. discharged, with time and date, referred to X facility, death); and

• outcome of infant (e.g. discharged, referred to X facility, fresh stillbirth, macerated stillbirth, very early neonatal death).

Note: Cases of complications of abortion and ectopic pregnancies are often found in other departments of the hospital than the

maternity, such as the female or gynaecology ward, operating theatres or outpatient or emergency departments.


Appe

ndix

C.

Rand

om n

umbe

r tab

le

12

34

56

78

910

1112

1314

110

480

1501

101

536

0201

181

647

9164

669

179

1419

462

590

3620

720

969

9957

091

291

9070

0

222

368

4657

325

595

8539

330

995

8919

827

982

5340

293

965

3409

552

666

1917

439

615

9950

5

324

130

4836

022

527

9726

576

393

6480

915

179

2483

049

340

3208

130

680

1965

563

348

5862

9

442

167

9309

306

243

6168

007

856

1637

639

440

5353

771

341

5700

484

974

917

9775

816

379

537

570

3397

581

837

1665

606

121

9178

260

468

8130

549

684

6067

214

110

0692

701

263

5461

3

677

921

0690

711

008

4275

127

756

5349

818

602

7065

990

655

1505

321

916

8182

544

394

4288

0

799

562

7290

556

420

6999

498

872

3101

671

194

1873

844

013

4884

063

213

2106

910

634

1295

2

896

301

9197

705

463

0797

218

876

2092

294

595

5686

969

014

6004

518

425

8490

342

508

3230

7

989

579

1434

263

661

1028

117

453

1810

357

740

8437

825

331

1256

658

678

4494

755

8556

941

1085

475

3685

753

342

5398

853

060

5953

338

867

6230

008

158

1798

316

439

1145

818

593

6495

2

1128

918

6957

888

231

3327

670

997

7993

656

865

0585

990

106

3159

501

547

8559

091

610

7818

8

1263

553

4096

148

235

0342

749

626

6944

518

663

7269

552

180

2084

712

234

9051

133

703

9032

2

1309

429

9396

952

636

9273

788

974

3348

836

320

1761

730

015

0827

284

115

2715

630

613

7495

2

1410

365

6112

987

529

8568

948

237

5226

767

689

9339

401

511

2635

885

104

2028

529

975

8986

8

1507

119

9733

671

048

0817

877

233

1391

647

564

8105

697

735

8597

729

372

7446

128

551

9070

7

1651

085

1276

551

821

5125

977

452

1630

860

756

9214

449

442

5390

070

960

6399

075

601

4071

9

1702

368

2138

252

404

6026

889

368

1988

555

322

4481

901

188

6525

564

835

4491

905

944

5515

7

1801

011

5409

233

362

9490

431

273

0414

618

594

2985

271

585

8503

051

132

0191

592

747

6495

1

1952

162

5391

646

369

5858

623

216

1451

383

149

9873

623

495

6435

094

738

1775

235

156

3574

9

2007

056

9762

833

787

0999

842

698

6691

7698

813

602

5185

146

104

8891

619

509

2562

558

104

2148

663

9124

585

828

1434

609

172

3016

890

229

0473

459

193

2217

830

421

6166

699

904

3281

2

2254

164

5849

222

421

7410

347

070

2530

676

468

2638

458

151

0664

621

524

1522

796

909

4459

2

2332

639

3236

305

597

2420

013

363

3800

594

342

2872

835

806

0691

217

012

6416

118

296

2285

1

2429

334

2700

187

637

8730

858

731

0025

645

834

1539

846

557

4113

510

367

0768

436

188

1851

0

2502

488

3306

228

834

0735

119

731

9242

060

952

6128

050

001

6765

832

586

8667

950

720

9495

3

Abrid

ged

from

Bey

er W

H, e

d. H

andb

ook

of ta

bles

for p

roba

bilit

y an

d st

atis

tics,

2nd

ed.

Boc

a Ra

ton,

Flo

rida,

The

Che

mic

al R

ubbe

r Co.

, 196

8. U

sed

with

the

perm

issi

on o

f CRC

Pre

ss, I

nc.

152 This page has been left blank 152

This handbook is an update of an earlier publication on monitoring the availability and use of obstetric services, issued by UNICEF, WHO and UNFPA in 1997. The indicators defined within the publication have been used by ministries of health, international agencies and programme managers in over 50 countries around the world. This revision incorporates changes based on monitoring and assessment conducted worldwide and the emerging evidence on the topic over the years, and has been agreed by an international panel of experts. It includes two new indicators and an additional signal function, with updated evidence and new resources.

This handbook aims to describe the indicators and to give guidance on conducting studies to people working in the field. It includes a list of life-saving services, or ‘signal functions’, that define a health facility with regard to its capacity to treat obstetric emergencies. The emphasis is on actual rather than theoretical functioning. The emergency obstetric care indicators described in this handbook can be used to measure progress in a programmatic continuum: from the availability of and access to emergency obstetric care to the use and quality of those services.

ISBN 978 92 4 154773 4

monitoring emergency obstetric care

Documents