Quality of Maternal and Newborn Health Services in Zanzibar, 2010

Quality of Maternal and Newborn
Health Services in Zanzibar, 2010
Findings from Selected Health Facilities in
Unguja and Pemba
Marya Plotkin
Christina Lulu Makene
Asma Ramadhan Khamis
Sheena Currie
Gaudiosa Tibaijuka
Maryjane Lacoste
Molly O’Bryan
Authors:
Marya Plotkin, M&E Advisor
Christina Lulu Makene, M&E Officer
Asma Ramadhan Khamis, Midwifery Advisor
Sheena Currie, Senior Maternal Health Advisor
Gaudiosa Tibaijuka, Senior Technical Manager
Maryjane Lacoste, Country Director
Molly O’Bryan, Program Officer
The Maternal and Child Health Integrated Program (MCHIP) is the U.S. Agency for
International Development’s Bureau for Global Health flagship maternal, neonatal and child
health (MNCH) program. MCHIP supports programming in MNCH, immunization, family
planning, malaria, nutrition and HIV/AIDS, and strongly encourages opportunities for
integration. Cross-cutting technical areas include water, sanitation, hygiene, urban health
and health systems strengthening.
This study is made possible by the generous support of the American people through the
United States Agency for International Development (USAID), under the terms of the
Leader with Associates Cooperative Agreement GHS-A-00-08-00002-00. The contents are the
responsibility of the Maternal and Child Health Integrated Program (MCHIP) and do not
necessarily reflect the views of USAID or the United States Government.
Cover photo credits:
Sheena Currie
Published by:
Jhpiego
Brown’s Wharf
1615 Thames Street
Baltimore, Maryland 21231-3492, USA
www.jhpiego.org
March 2012
Table of Contents
ACRONYMS ........................................................................................................................................ VII
ACKNOWLEDGEMENTS .................................................................................................................... VIII
EXECUTIVE SUMMARY........................................................................................................................ IX
1. INTRODUCTION/BACKGROUND...................................................................................................... 1
2. STUDY DESIGN ................................................................................................................................ 2
2.1 Sample and Sampling Strategies ...................................................................................................... 2
2.2 Data Collection ................................................................................................................................... 3
2.3 Data Entry, Quality Control and Analysis........................................................................................... 3
2.4 Ethical Considerations ....................................................................................................................... 4
3. HEALTH FACILITY OVERVIEW .......................................................................................................... 5
4. FINDINGS ........................................................................................................................................ 6
4.1 Case Volume and Self-Reported Infrastructure for ANC and L&D .................................................. 6
4.2 ANC Inventory ..................................................................................................................................... 7
4.3 Findings from ANC Consultation Observations ................................................................................ 8
5. LABOUR AND DELIVERY SERVICES ..............................................................................................12
5.1 Presence of Skilled Personnel......................................................................................................... 12
5.2. Availability of Essential Maternal and Newborn Supplies ............................................................ 12
6. FINDINGS FROM LABOUR AND DELIVERY OBSERVATIONS .........................................................13
6.1 Description of Clients in Labour and Delivery Observations ......................................................... 13
6.2. Initial Client Assessment ................................................................................................................ 13
6.3. Woman-Friendly Care during Labour and Delivery (Interpersonal Communication) ................... 14
6.4. Care during the Second and Third Stage of Labour ...................................................................... 15
6.5 Immediate and Essential Newborn Care ........................................................................................ 16
6.6. Harmful and Un-Indicated Practices .............................................................................................. 16
6.7. Infection Prevention ........................................................................................................................ 17
6.8. Use of a Partograph to Monitor Labour ......................................................................................... 18
7. PREVENTION AND MANAGEMENT OF SELECTED MATERNAL AND NEWBORN HEALTH
COMPLICATIONS .........................................................................................................................19
7.1 Prevention and Management of Postpartum Haemorrhage ......................................................... 19
7.2 Prevention and Management of Pre-Eclampsia/ Eclampsia, Including Screening ...................... 21
7.3 Complicated Cases Observed during Study ................................................................................... 22
8. HEALTH WORKER KNOWLEDGE ...................................................................................................23
9. DISCUSSION ..................................................................................................................................25
9.1 Prevention and Management of PE/E ............................................................................................ 25
9.2 Prevention and Management of PPH.............................................................................................. 26
9.3 Essential Newborn Care .................................................................................................................. 27
9.4 Other Issues of Note ........................................................................................................................ 27
9.5 Comparing Key Findings from Zanzibar and Mainland Tanzania.................................................. 28
10. CONCLUSION AND RECOMMENDATIONS ..................................................................................30
Quality of Maternal and Newborn Health Services in Zanzibar
iii
APPENDIX A. PHARMACY STOCK ......................................................................................................31
APPENDIX B. EQUIPMENT INVENTORY .............................................................................................32
APPENDIX C. PROVIDERS’ PE/E KNOWLEDGE ASSESSMENT SCORES ..........................................33
REFERENCES.....................................................................................................................................35
iv
Quality of Maternal and Newborn Health Services in Zanzibar
List of Tables
Table 4.1 ANC and L&D Clients Observed................................................................................................... 6
Table 4.2 Cadre of Health Worker Providing ANC Services ........................................................................ 7
Table 4.3 Essential Supplies for Basic ANC, from ANC Inventory .............................................................. 7
Table 4.4 Key Services Provided to ANC Clients ......................................................................................... 8
Table 4.5 Screening for Pre-Eclampsia ....................................................................................................... 9
Table 4.6 Counselling on Preventive Treatments during ANC Consultation ............................................. 9
Table 4.7 Counselling on Birth Preparedness during ANC Visit ............................................................... 10
Table 4.8 Basic Content of First ANC Visit ................................................................................................ 11
Table 4.9 Discussion of Previous Pregnancies with Multigravida, First-Visit ANC Clients...................... 11
Table 5.1 Cadres of Health Care Providers Attending Observed Deliveries ............................................ 12
Table 6.1 Key Steps in Initial Client Assessment...................................................................................... 13
Table 6.2 Assessment of Previous Complications among Multiparous Clients ...................................... 14
Table 6.3 Woman-Friendly Care Components Observed during the Initial Assessment and
First Stage of Labour ....................................................................................................................... 15
Table 6.4 Practice of Immediate and Essential Newborn Care ............................................................... 16
Table 6.5 Harmful and Un-Indicated Practices ......................................................................................... 16
Table 6.6 Infection Prevention Measures for L&D Clients ....................................................................... 17
Table 6.7 Partograph Use during Labour .................................................................................................. 18
Table 7.1 Uterotonics Administered for AMTSL ........................................................................................ 20
Table 7.2 Screening for Pre-Eclampsia during Labour, Initial Assessment ............................................ 22
Table 8.1 Provider Knowledge Scores on Maternal Health Topics.......................................................... 23
Table 8.2 Provider Performance on Newborn Resuscitation Simulation ................................................ 24
Table 9.1 Comparison of Study Findings: Zanzibar and Mainland Tanzania .......................................... 28
Quality of Maternal and Newborn Health Services in Zanzibar
v
List of Figures
Figure 2.1 Study Sites: Unguja and Pemba................................................................................................. 3
Figure 4.1 Counselling on Danger Signs (n=57) ....................................................................................... 10
Figure 6.1 Assessment of Danger Signs in Initial Assessment ................................................................ 14
Figure 6.2 Tasks for Management of Second and Third Stages of Labour............................................. 15
Figure 7.1 Proportion of Births Observed in Which AMTSL Tasks Were Performed Correctly ............... 20
Figure 7.2 Proportion of Deliveries with Correct Provision of AMTSL with Oxytocin ............................... 21
Figure 7.3 Proportion of Deliveries with Correct Provision of AMTSL with Any Uterotonic ..................... 21
vi
Quality of Maternal and Newborn Health Services in Zanzibar
Acronyms
AMTSL
Active management of the third stage of labour
ANC
Antenatal care
BEmONC
Basic emergency obstetric and newborn care
CCT
Controlled cord traction
CS
Caesarean section
EDL
Essential drug list
EmOC
Emergency obstetric care
FIGO
International Federation of Gynecology and Obstetrics
HMIS
Health management information system
ICM
International Confederation of Midwives
IM
Intramuscular
IP
Infection prevention
IPTp
Intermittent preventative treatment of malaria in pregnancy
IU
International units
IV
Intravenous
L&D
Labour and delivery
MAISHA
Mothers and Infants Safe Healthy, Alive
MCHIP
Maternal Child Health Integrated Program
MDGs
Millennium Development Goals
MOH
Ministry of Health
PE/E
Pre-eclampsia/eclampsia
PMTCT
Prevention of mother-to-child transmission
PPFP
Postpartum family planning
PPH
Postpartum haemorrhage
SP
Sulphadoxine pyrimethamine
TT
Tetanus toxoid
USAID
United States Agency for International Development
WHO
World Health Organization
Quality of Maternal and Newborn Health Services in Zanzibar
vii
Acknowledgements
This study is part of a multi-country assessment of the quality of maternal and newborn health
services. In Zanzibar, the study was conducted by Jhpiego with assistance from national lifesaving skills trainers. The data collectors (both Jhpiego and non-Jhpiego) included:
Sheena Currie, Scholastica Chibehe, Hilda Nyerembe, Asma Khamis Ramadhan, John
Ndombaro, Rita Nakua, Edna Ngoli, Douglas Maro, Khadija Mohamed, Emiliyan Mmakasa,
Mary Mwakyusa, Eva Joseph Hongoli, Flora Lyimo, Wanu Bakari, Mary Mlay, Gertrude
Anderson, Neema Kasembe, Agata Liviga, Hamida Mkata, Lydia Joseph Maro and Ndeshi
Massawe
We would like to express our appreciation to Dr. Mohammed Saleh Jidawwi (Principal
Secretary) from the Zanzibar Ministry of Health (MOH) for his technical support and advice
on data collection. In addition, grateful thanks to the health facility directors and health
care providers for their participation and cooperation during the data collection.
Thank you to Barbara Rawlins and David Cantor for outstanding support on the technical
aspects of the study and the use of mobile phones. And finally, our appreciation to USAID
for supporting this study through both the MAISHA and MCHIP programs.
ABOUT THE MAISHA PROGRAM
MAISHA, meaning “life” in Swahili, promotes the philosophy that building solid foundations
for quality services will empower providers at all levels of the health care system across the
country to deliver targeted interventions that will make a real difference in keeping mothers
and their newborn infants, safe, healthy and alive.
The USAID/Tanzania-funded MAISHA program is assisting the MOH with strengthening
the platforms of focused antenatal care (FANC) and basic emergency obstetric and neonatal
care (BEmONC) for addressing the prevention and treatment of postpartum haemorrhage
and other key contributors to maternal mortality, and essential newborn care (ENC),
including newborn resuscitation, prevention and treatment of sepsis and immediate
warming and drying. MAISHA is supporting the MOH in developing national and district
resources (guidelines, training package, trainers and supervision tools) for FANC and
BEmONC and in advocating and coordinating with district health management teams,
donors and other key stakeholders to ensure that funding is allocated for implementing
quality FANC and BEmONC, including training service providers at district level (using the
resources developed at national and district levels) throughout the country. MAISHA is also
strengthening the platform of prevention of mother-to-child transmission (PMTCT) of HIV to
address gaps in integrating maternal and newborn health (MNH) services for HIV-positive
women and children.
ABOUT MCHIP
The Maternal and Child Health Integrated Program (MCHIP) is the USAID Bureau for
Global Health flagship maternal, neonatal and child health (MNCH) program. MCHIP
supports programming in maternal, newborn and child health, immunization, family
planning, malaria and HIV/AIDS, and strongly encourages opportunities for integration.
Cross-cutting technical areas include water, sanitation, hygiene, urban health and health
systems strengthening.
This program and report was made possible by the generous support of the American people
through the United States Agency for International Development (USAID), under the terms
of the Leader with Associates Cooperative Agreement GHS-A-00-08-00002-000. The contents
are the responsibility of the MAISHA program and do not necessarily reflect the views of
USAID or the United States Government.
viii
Quality of Maternal and Newborn Health Services in Zanzibar
Executive Summary
The MAISHA Quality of Maternal and Newborn Health Services study in Zanzibar,
conducted in November–December 2010, was an observational study conducted in nine
health facilities in Pemba and Unguja. The aim of the study, which combined observations of
service delivery with inventories, record reviews and health worker knowledge assessments,
was to provide strong information on the quality of maternal and newborn health care in
these facilities, as an indication of current practice in maternal and newborn care in
Zanzibar. The results serve as a baseline for the MAISHA program and as an important
source of information on quality of maternal and newborn care for policymakers and
stakeholders in Zanzibar.
KEY FINDINGS FROM ANTENATAL CARE
Blood pressure was taken in 81% of the ANC observations, and a urine test for the presence of
protein was administered in 86% of observations. Fifty-four per cent of ANC observations
included counselling and testing for HIV and 38% of clients observed were counselled on
postpartum family planning (PPFP). Among first-visit ANC observations, 48% included
provision of intermittent preventive treatment of malaria in pregnancy (IPTp) and 52%
included provision of iron and folic acid. Screening for pre-eclampsia/eclampsia (PE/E) took
place in 55% of observed ANC consults. The overall mean percentage score for counselling on
danger signs in pregnancy was 65%. The overall average for discussion of previous pregnancy
complications was 36%.
The three main preventive treatments given to ANC clients (IPTp, iron and folic acid and
tetanus toxoid injection) were administered fairly uniformly, with a mean score of 39%. Half
of facilities assessed had iron or iron folate available at the time of the assessment, and
sulphadoxine pyrimethamine (SP) was available in 63% of facilities at the time of
assessment.
The findings on ANC bring to the forefront areas of relative strength (provision of key
services for ANC) as well as some significant weaknesses (counselling on danger signs and
HIV, syphilis testing, anaemia testing, counselling for HIV-positive ANC clients and PPFP).
Improving the quality of ANC in health centres and dispensaries should be a special focus,
as these facilities typically have the highest ANC caseloads.
KEY FINDINGS FROM LABOUR AND DELIVERY
Assessment of danger signs of a woman in labour during the initial assessment was low, at
21%, while 60% of clients had their blood pressure was measured during their initial
assessment. Among multigravida clients, there was a mean score of 29% who were assessed
for complications during pregnancy. However, only 8% were assessed for a previous history
of prolonged labour and 10% for a previous history of convulsions.
Approximately 59% of the observed clients received woman-friendly care.Overall, 60% of
deliveries observed had AMTSL performed (any uterotonic, within 1 minute of birth) and
91% of deliveries had AMTSL with a relaxed definition (any uterotonic, within 3 minutes of
birth). However, AMTSL was provided according to the World Health Organization (WHO)
definition (oxytocin, given within 1 minute) at only 20% of deliveries.
Infection prevention measures were encouraging overall. Handwashing was observed 71% of
the time during initial assessments and 64% of the time during the first stage of labour.
Sharps disposal, decontamination and waste disposal were performed correctly 86%, 97%
and 97% of the time, respectively.
In the majority of cases the partograph was filled in at the beginning of labour (78%) and
after delivery (100%). However, use of the partograph was very low for recording maternal
Quality of Maternal and Newborn Health Services in Zanzibar
ix
pulse, foetal heart tones and frequency and duration of contractions. Blood pressure was
recorded every 4 hours on the partograph for less than half of the women observed (44%).
Only 34% of newborns observed were dried immediately, although this is a key step in
keeping the newborn warm. Initiation of breastfeeding within one hour also was low,
observed in only 20% of births.
Only two of the nine facilities had MgSO4 in stock in the pharmacy, and there was a severe
deficiency of antihypertensive drugs: only three facilities had nifedipine (a recommended
antihypertensive) in the pharmacy and none had hydralazine in the labour ward.
The findings on labour and delivery illustrate that there is still much work to be done in
order to strengthen the quality of service provision in Zanzibar. Specifically, particular
attention must be given to proper and consistent use of the partograph, assessment of
danger signs, use of AMTSL as per WHO guidelines, provision of immediate essential
newborn care and acquisition of commodities to effectively manage PE/E.
x
Quality of Maternal and Newborn Health Services in Zanzibar
1. Introduction/Background
Improving the quality of obstetric and newborn care in facilities is an essential strategy for
reducing maternal and newborn deaths (Van den Broek and Graham 2009). The study
described in this report measured key aspects of antenatal care (ANC), labour and delivery
care and newborn care in selected health facilities in Zanzibar.
The overall goal of the study was to collect sound information on maternal and newborn care
in selected facilities by observing key maternal and newborn health interventions. The
definition of “quality” used in the study is that services are correctly performed per globally
and nationally accepted evidence-based guidelines.
This study also serves as the baseline measurement for the USAID/Tanzania-funded
MAISHA (Mothers and Infants Safe Healthy Alive) program in Zanzibar. The MAISHA
program is a national service delivery strengthening program that has been working since
2008 to improve the quality of maternal and newborn health services by training health care
providers, providing equipment and supplies and implementing quality improvement
initiatives. MAISHA is being implemented in both Pemba and Unguja.
There are known effective interventions for screening, preventing and treating obstetric and
newborn complications in health care facilities. Improving the quality of facility-based care
to prevent and treat frequent maternal and newborn complications is a critical component in
the effort to reduce maternal and newborn deaths globally, and in helping countries meet
their targets for Millennium Development Goals (MDGs) 4 and 5. Facility based deliveries in
Zanzibar range from 24% in parts of Pemba to 70% in parts of Unguja (Tanzania DHS 2010),
so improving maternal and newborn care in health facilities could potentially dramatically
reduce mortality.
The most frequent cause of maternal mortality in Zanzibar is postpartum haemorrhage (26%
of maternal deaths), followed by hypertensive disorders in pregnancy (16%), and rupture of
uterus (12%). Sepsis and anaemia are both estimated to cause 3% of maternal deaths, and
coagulopathies are noted to be a cause of maternal deaths (Zanzibar Ministry of Health 2010).
These percentages are similar to those in developing countries generally (Khan et al. 2006).
This study looked specifically at life-saving practices related to the major causes of maternal
death, including postpartum haemorrhage and hypertensive disorders in pregnancy. The
frequency of use and correct performance of interventions for pre-eclampsia/eclampsia
(PE/E), postpartum haemorrhage, prolonged/obstructed labour and newborn sepsis and birth
asphyxia are documented.
The specific interventions assessed include screening and management of PE/E, partograph
use, use of AMTSL to prevent postpartum haemorrhage (PPH), management of PPH, IP,
and essential newborn care, including resuscitation.
The methods used in the study were (1) observations of health care providers during ANC
consultations and when conducting care in labour and deliveries; (2) health worker
knowledge and skill assessments (including a demonstration of newborn resuscitation on a
model); and (3) inventories of the ANC, maternity and general facility pharmacies. The
results of this assessment will be used as a baseline from which to measure the progress and
success of the MAISHA program in Zanzibar. The results will also inform national
programmes and policies that address quality in ANC and maternity services in Zanzibar.
Quality of Maternal and Newborn Health Services in Zanzibar
1
2. Study Design
The study used a combination of approaches, including observations of ANC and maternity
clients; inventories in the ANC, maternity and general facility pharmacies; and knowledge
and skills assessments of providers, including a simulated resuscitation of a newborn using
a model.
The following tools were used at the facility level:
•
Facility inventory: The facility inventory included documentation of infrastructure
conditions and verification of the availability and storage conditions of medications, supplies
and equipment. The inventory is conducted once per facility and also includes inventories of
the general pharmacy and the ANC and maternity wards.
•
Record review: This tool captured the number of ANC consultations, births (live and
stillborn), and maternal and newborn deaths at each facility for the last year, based on the
routine data collection tools. Up to 24 individual patient charts from the past three months
were also reviewed for partograph use and completeness.
•
Clinical practice observation of ANC and labour and deliveries: Structured clinical
observation checklists were used for observation of ANC consults and vaginal deliveries in
the facilities. The content of the checklists was developed based on international (WHOapproved) protocols for screening for PE/E in ANC; management of PE/E and PPH in labour
and delivery (L&D); and other interventions in L&D (routine and correct use of partograph,
routine and correct use of active management of the third stage of labour (AMTSL), infection
prevention behaviours, provider-client interaction/communication, immediate essential
newborn care and newborn resuscitation). Minor revisions were made to ensure that the
tools were tailored to Zanzibar policies.
•
Health care worker interviews: Health workers were asked a series of questions to test
their knowledge of how to identify, manage and treat common maternal and newborn health
complications. A simulated resuscitation using a newborn model was used to measure the
health workers’ newborn resuscitation skills.
2.1 SAMPLE AND SAMPLING STRATEGIES
All nine health facilities in Pemba and Unguja that are supported by the MAISHA program
in Zanzibar were included in the sample.
The study was powered on the number of deliveries to be observed. In order to achieve the
power desired for analysis, it was determined that 214 deliveries should be observed. The
number of deliveries to be observed was then used to establish a quota for each health facility,
with lower-volume sites allocated fewer deliveries to be observed. The quotas were developed
with the knowledge that weights would be applied to the values to adjust for the differences in
volume.
2
Quality of Maternal and Newborn Health Services in Zanzibar
Figure 2.1 Study Sites: Unguja and Pemba
Only data from the maternity observations were weighted. The findings on ANC presented
in this report are not weighed, which means that the findings might be slightly biased
towards the higher volume facilities.
2.2 DATA COLLECTION
The 21 data collectors for the study were health care providers who were also national Life
Saving Skills trainers. All data collectors had been part of data collection for the same
study, using the same tools, for mainland Tanzania. Before the mainland data collection
effort, the data collectors were given a two-day technical update in basic emergency
obstetric and newborn care (BEmONC), which was followed by seven days of training in
data collection, including two days of practice. The training covered research ethics and
consent, familiarization with all of the tools, familiarization with the mobile phone
technology (all data collection was conducted using smartphones), simulations with
scoring, an inter-rater reliability exercise, and two days of practical application using the
smartphone technology at hospitals in Dar es Salaam.
Data collectors worked in teams of two to three people, depending on the size of the health
facility. Fieldwork was conducted from November to December 2010.
2.3 DATA ENTRY, QUALITY CONTROL AND ANALYSIS
The data collectors recorded survey data on smartphones using customized data entry
programmes, which were developed with a package called PocketPC Creations running on
Windows mobile. Logic, skip and consistency checks were built into the programmes. The
data collectors were trained to review records for missing and/or inconsistent answers before
they submitted the data. Depending on whether phone coverage was available at the study
site, the data from each handheld device was either uploaded directly to a central database
at the end of each day or backed up to a secure digital card to be uploaded when the data
collectors returned from the field. Data was uploaded from the phones into a database on a
secure network. Once in the database, data was entered into tables and made available for
study team members via a website that was accessible only with a password. Analysis was
conducted both by the study’s principal investigator and team in the United States and by
the study team in Tanzania. Analyses were conducted using SPSS.
Quality of Maternal and Newborn Health Services in Zanzibar
3
2.4 ETHICAL CONSIDERATIONS
The study protocol was submitted to and approved by the Zanzibar Research Council in
Zanzibar and the institutional review board of Johns Hopkins Bloomberg School of Public
Health (JHSPH) in the United States. The JHSPH institutional review board ruled the
protocol exempt from review under 45 CFR 46.101(b), Category (5). Informed consent was
obtained from all participating health providers and patients as well as from facility
directors. If a woman was incapacitated, consent was to be obtained from next of kin or a
guardian. However, this circumstance did not occur in the course of the study.
4
Quality of Maternal and Newborn Health Services in Zanzibar
3. Health Facility Overview
The health facilities in the sample consisted of five hospitals and four health centres (n=9).
Two of the health facilities were in Pemba and the other seven in Unguja. Six of these
facilities provide comprehensive emergency obstetric and newborn care (CEmONC) services,
three provide BEmONC services, and all provide ANC services.
Table 3.1 displays basic infrastructure characteristics of the health facilities in the sample.
Table 3.1 Facility Infrastructure Characteristics of the Sample
HEALTH FACILITIES IN SAMPLE
(n=9)
FACILITY FEATURE
n
%
24-hour staff coverage (schedule observed or staff live onsite)
8
89
Emergency transport
7
78
Communication equipment
3
33
Patient room for ANC with auditory and visual privacy
8
89
Safe water source within 500 metres of facility
9
100
Electric power (grid or functioning generator with fuel)
9
100
Functional improved-type toilet
8
89
Ability to conduct surgery with general anaesthesia
4
44
< 50
4
44
51–99
3
33
100+
2
22
Mean number of overnight beds per facility
All eight out of nine facilities had 24-hour staff coverage, and seven facilities had emergency
transport. Only four out of the five hospitals provided caesarean sections (CS); one hospital
had not been upgraded to provide CS. The other four facilities were health centres that did
not offer CS as a service.
Quality of Maternal and Newborn Health Services in Zanzibar
5
4. Findings
4.1 CASE VOLUME AND SELF-REPORTED INFRASTRUCTURE FOR ANC
AND L&D
The average annual client volume for ANC was 1,843 (with a range of 46–2,213), and the
average annual client volume for labour and delivery was 1,345 (range of 137–11,831). The
annual number of caesarean sections ranged from 11–564.
ANC observations were conducted in all nine health facilities. The resulting sample of
clients comprised 57 observations (range 1–10 clients observed per facility). 217 L&D clients
were observed (range 1–6 clients per facility). Further details on the participants in the
study (both ANC and L&D) are provided in Table4.1.
Table 4.1 ANC and L&D Clients Observed
OBSERVATIONS
(n=57)
ANC CLIENTS OBSERVED
n
%
Unguja
31
54
Pemba
26
46
Total
57
100
Hospital
36
63
Health centres
21
37
Total
57
100
First visit
27
47
Follow-up visit
30
53
≤ 20 weeks
17
30
21–36 weeks
34
60
≥ 37 weeks
4
7
Unknown
2
4
Primagravida
13
23
Multigravida
44
77
Client goes home
32
57
Referred within facility
22
39
Admitted to facility
1
2
Referred to another facility
2
4
Unguja
139
64
Pemba
78
36
Total
217
100
Hospital
185
85
Cottage hospital
32
15
Type of ANC visit observed
Gestational age at visit
Gravida
Outcome of visit
L&D client observed
Number of deliveries
6
Quality of Maternal and Newborn Health Services in Zanzibar
OBSERVATIONS
(n=57)
ANC CLIENTS OBSERVED
n
%
Primagravida
50
23
Multigravida
167
77
Gravida
The mean time of an ANC consultation was 46 minutes for the first visit and 23 minutes for
the subsequent visits.
Nurse-midwives provided the majority of ANC services observed. However, in 16% of the
ANC consultations observed, a maternal and child health aide (MCHA) provided the ANC
services. MCHAs are a cadre that is not authorized by the MOH to provide ANC services at
the facility level, although they may support staff in specific tasks such as weighing a
patient.
Table 4.2 Cadre of Health Worker Providing ANC Services
OBSERVATIONS
(n=57)
n
%
Nurse-midwife
44
77
Maternal and child health aide
9
16
Other*
2
4
Unknown
2
4
Total
57
100
CADRE OF HEALTH PROVIDER
*Cadres falling in the “other” category were not specified in the tool.
4.2 ANC INVENTORY
Inventories of ANC clinic supplies and equipment and of pharmacy supplies were conducted.
Table 4.3 Essential Supplies for Basic ANC, from ANC Inventory
FACILITIES WITH FUNCTIONAL EQUIPMENT
ESSENTIAL SUPPLIES FOR ANC
n=8*
%
Blood pressure apparatus
8
100
Foetal stethoscope
8
100
Adult weighing machine
7
88
Vaginal speculum
8
100
Guidelines/protocols for ANC
6
75
Guidelines/protocols for management of PE/E
3
38
Guidelines/protocols for STIs
5
63
Disinfectant not yet mixed
5
50
Waste receptacle with lid and plastic liner
3
38
Visual aids for client
6
75
Availability of iron
4
50
Availability of SP
5
63
Availability of rapid plasma reagent (RPR) kits
8
100
*One facility had missing data for the ANC inventory.
Quality of Maternal and Newborn Health Services in Zanzibar
7
Although one facility’s data were missing for the inventory, all eight of the facilities assessed
had working blood pressure equipment, foetal stethoscopes and speculums. Availability of
iron or iron folate was low; only half of the facilities had them at the time of the assessment.
SP was available at 63% of health facilities. Although many of the facilities had protocols
and guidelines for ANC (75%) and STIs (63%), only three facilities had guidelines for
management of PE/E.
4.3 FINDINGS FROM ANC CONSULTATION OBSERVATIONS
Key Services in ANC
Key services in ANC include weighing clients; measuring blood pressure; testing urine for
protein and glucose (sugar); testing blood for anaemia, syphilis and HIV; administering
intermittent preventive treatment of malaria during pregnancy (IPTp) and tetanus toxoid; and
counselling for birth planning, danger signs and family planning. IPTp should be given twice,
at least four weeks apart, after 20 weeks of gestation—once during the second trimester and
once during the third trimester. The two doses of IPTp are referred to as IPT1 and IPT2.
Counselling on postpartum family planning (PPFP) is recommended at the third and fourth
ANC visits, whereas birth preparedness counselling should occur (with updates as necessary)
at every ANC visit.
Table 4.4 Key Services Provided to ANC Clients
OBSERVATIONS
(n=57)
KEY SERVICE
n
%
Weight taken
51
89
Blood pressure taken
46
81
Urine test for protein
49
86
Blood test for anaemia
51
89
Blood test for syphilis*
19
70
SP for IPTp *
13
48
Iron and folic acid for first visit*
14
52
Counselling for family planning
21
38
Counselling and testing for HIV
31
54
Offered tetanus toxoid
37
65
Mean per cent score on provision of key ANC services
68%
* Denominator was number of first visit clients (n=27)
Blood pressure was taken at 81% of the ANC consultations observed, and urine was tested
for protein occurred at 86% of the observed ANC visits. Performance was weaker on PPFP
(38%), counselling and testing for HIV (54%), provision of IPTp (48% of first visit clients) and
provision of iron and folic acid given (52% of first visit clients).
Pre-eclampsia Screening
Screening for pre-eclampsia can translate into saving women’s lives. Scores for provision of
the components of screening for pre-eclampsia are presented below, individually and as a
composite indicator.
8
Quality of Maternal and Newborn Health Services in Zanzibar
Table 4.5 Screening for Pre-Eclampsia
OBSERVATIONS
(n=55)
COMPONENTS OF SCREENING
n
%
Ask the client about headache or blurred vision
28
51
Ask the client about swollen hands or face
19
35
Take the client's blood pressure
46
81
Composite indicator for screening for pre-eclampsia*
55%
*The composite indicator includes screening for either one of the first two danger signs and correctly taking the client’s
blood pressure.
Using the composite indicator, 55% of the observed ANC consults included screening for
PE/E. Seven out of the nine facilities (78%) reported that urine is checked for protein as part
of routine ANC. In the other two facilities reagent was out of stock. Although not part of the
composite indicator, screening urine for protein is an essential component of detecting PE/E.
Preventive Treatments in ANC
Intermittent preventive treatment of malaria (IPTp), provision of iron/folate pills, and
tetanus toxoid injections are three important components of preventive ANC services in
Zanzibar. During the ANC visit, clients are supposed to receive 90 tablets of iron or iron
folate, SP for IPTp after 20 weeks of gestation (first or second dose), and tetanus toxoid
injections. Unlike mainland Tanzania, Zanzibar has no voucher system for insecticidetreated nets, in part because of the lower prevalence of malaria on the islands. The scores for
the three preventive treatments were fairly uniform across the treatments, with a mean
score of 39%.
Availability of iron or iron folate was low, with only half of the facilities having these at the
time of the assessment. However, testing blood for anaemia (measuring haemoglobin) was
high. Many women are given iron folate supplementation as prophylaxis in pregnancy, so in
the event that a woman is diagnosed with anaemia (Hb < 10.5gm/dl), it is not clear how or
where the needed therapeutic iron supplementation would be given. SP was available at
63% of health facilities.
In addition to providing preventive treatments, clients must be counselled on how and why
to take these treatments. Table 4.6 presents findings on counselling on the preventive
treatments.
Table 4.6 Counselling on Preventive Treatments during ANC Consultation
OBSERVATIONS
(n=57)
COUNSELLING TOPICS
n
%
Explain the purpose of the treatment
21
57
Explain how to take
22
60
Explain possible side effects
1
3
14
48
Explain the purpose of the treatment
22
82
Explain how to take
26
96
Explain possible side effects
3
11
Counselling on iron/folic acid
Counselling on tetanus toxoid injection
Explain the purpose of the treatment
IPTp counselling for first visit clients
Quality of Maternal and Newborn Health Services in Zanzibar
9
There was a notable lack of explanation about side effects of both iron/folic acid and IPTp.
Counselling on Danger Signs and Birth Preparedness
Counselling on danger signs during pregnancy is an important part of ANC services, because
it provides clients with relevant information on when to seek care urgently and can help
prevent maternal, foetal or newborn deaths. All ANC clients should be checked for key
danger signs at each visit. They should also be counselled on which danger signs (i.e.,
vaginal bleeding, swollen face and hands, severe headaches, convulsions, decrease in foetal
movement) should prompt them to seek care at the nearest health facility. In addition,
clients should be told to return for persistent cough. ANC counselling should also address
birth preparedness, including deciding where to deliver and the importance of delivering
with a skilled birth attendant, having the necessary supplies at home, and having some
money available in case of emergencies. Figure 4.1 shows the proportion of clients counselled
on danger signs.
Figure 4.1 Counselling on Danger Signs (n=57)
Return if vaginal bleeding
74%
Return if swollen hands and face
54%
Return if severe headache or blurred vision
74%
Return if persistency cough
34%
Return if severe abdominal pain
75%
Mean pervcent score for counselling on…
0%
65%
20%
40%
60%
80%
100%
There was an overall mean score of 65% for counselling on danger signs for the ANC clients.
While roughly three-quarters of clients were told to return if they experienced severe
abdominal pain, headache, blurred vision or vaginal bleeding, only 54% were told to return if
they had swollen hands and face and 34% were told to return if they developed a persistent
cough.
In addition to counselling on danger signs, ANC clients should be counselled on birth
preparedness during their ANC consult.
Table 4.7 Counselling on Birth Preparedness during ANC Visit
OBSERVATIONS
(n=57)
COUNSELLING ACTION
n
%
Ask client where she will deliver
34
60
Advise client to prepare for delivery (i.e., set aside money, arrange for emergency
transport)
39
70
Advise client to use skilled health worker during delivery
35
61
Discuss with client what items to have on hand at home for emergencies (e.g., sterile
blade)
35
61
Counsel on postpartum family planning
21
38
Mean per cent score on birth preparedness counselling
58%
While 70% of ANC clients were advised to prepare for delivery, less than half (38%) of the
clients observed were counselled about PPFP.
10
Quality of Maternal and Newborn Health Services in Zanzibar
First ANC Visits
The first ANC visit includes important services such history taking, examination, counselling
and preventative/corrective treatments. Consultations with 27 first-visit ANC clients were
observed for the study. Table 4.8 shows the basic information gathered from the client at the
first visit.
Table 4.8 Basic Content of First ANC Visit
OBSERVATIONS
(n=27)
CLIENT INFORMATION COLLECTED
n
%
Client’s age
22
82
Medications client is taking
6
22
Date of client’s last menstrual period
23
85
Number of prior pregnancies
22
82
History of Previous Pregnancies
Documenting the history of any complications during previous pregnancies and deliveries is
important because it can identify clients who may need special care. Nineteen of the 27 firstvisit ANC clients had previous pregnancies. Table 4.9 shows the discussions that took place
during their first ANC visit.
Table 4.9 Discussion of Previous Pregnancies with Multigravida, First-Visit ANC Clients
OBSERVATIONS
(n=19)
TOPICS HEALTH WORKER ASKED ABOUT OR CLIENT MENTIONED
n
%
Prior stillbirth(s)
11
58
Heavy bleeding during or after delivery*
10
56
Previous caesarean section(s)
13
68
Previous abortion(s)
11
58
Previous multiple pregnancies
3
16
Previous prolonged labour
2
11
Previous pregnancy-related hypertension
2
11
Previous pregnancy-related convulsions
1
5
Previous assisted deliveries (forceps, ventouse)
8
42
Anaemia
2
11
Prior newborn death(s)
11
58
Overall average of previous pregnancy complications discussed
36%
Although the numbers of multigravida first-visit clients were relatively low, there was
nevertheless a very low level of discussion of complications during previous pregnancies. For
example, only 11% of multigravida clients were asked if they had previously had pregnancyrelated hypertension. This is an area of concern.
Quality of Maternal and Newborn Health Services in Zanzibar
11
5. Labour and Delivery Services
5.1 PRESENCE OF SKILLED PERSONNEL
The availability of human resources for health and, in this context, the attendance of a
skilled health care provider are critical factors for improving maternal and newborn health
and reducing maternal and newborn mortality. In Zanzibar, all facility deliveries are
supposed to be attended by a skilled birth attendant. Table 5.1 shows the distribution of
health care provider cadres attending the deliveries observed in this study.
A total of 217 deliveries were observed at the nine facilities (five hospitals and four health
centres). The vast majority of deliveries (94%) were attended by nurses and midwives; four
deliveries (2%) were attended by a medical attendant, which is a cadre not authorized to
provide L&D services.
Table 5.1 Cadres of Health Care Providers Attending Observed Deliveries
OBSERVATIONS
(n=217)
CADRE OF L&D SERVICE PROVIDER
n
%
203
94
Medical attendant
4
2
Physician
2
1
Trainee
2
1
Other
6
2
Nurse/midwife
5.2. AVAILABILITY OF ESSENTIAL MATERNAL AND NEWBORN
SUPPLIES
Facilities were assessed for supplies and equipment using a standardized inventory tool. The
resulting inventories of supplies and equipment for L&D are included in Appendixes A and
B. Of note:
•
Only two health facilities had MgSO4 in stock at the pharmacy, but it was available in eight
of the labour wards.
•
There was a deficiency of antihypertensive drugs: only three facilities had nifedipine in the
pharmacy and none had hydralazine in the labour ward.
•
Although all facilities had the capacity for sterilization, only one facility had written
protocols or guidelines for sterilization or disinfection.
The lack of antihypertensive drugs has major implications for the management of
hypertensive disorders (i.e., PE/E). The implications are addressed in the discussion section
of this report.
12
Quality of Maternal and Newborn Health Services in Zanzibar
6. Findings from Labour and Delivery
Observations
6.1 DESCRIPTION OF CLIENTS IN LABOUR AND DELIVERY
OBSERVATIONS
Observations of women in labour were conducted in the maternity wards at all nine of the
health facilities in the study. The resulting sample of 217 clients included 139 in Unguja
(64%) and 78 (36%) in Pemba. Because clients were observed at different stages in their
labour, the number observed at each stage varies.
6.2. INITIAL CLIENT ASSESSMENT
When a woman in labour is admitted the provider has to undertake a full assessment to
ensure that care is planned according to the woman’s needs and to detect and manage any
problems. The initial client assessment is critical to identifying problems, especially danger
signs that require immediate attention. Table 6.1 details some of the key steps in the initial
client assessment and indicates whether the steps were conducted in the 102 deliveries in
which the initial client assessment was observed.
Table 6.1 Key Steps in Initial Client Assessment
OBSERVATIONS
(n=102*)
INITIAL CLIENT ASSESSMENT ACTION
n
%
Checks client card or asks for age, length of pregnancy, parity
87
85
Checks blood pressure
69
68
Takes temperature
70
69
Takes pulse
58
57
Asks client and/or notes amount of urine output
15
15
Checks fundal height
80
78
Checks fetal presentation with palpation of abdomen
91
89
Performs vaginal examination (cervical dilation, fetal descent, position, membranes,
meconium)
90
88
Mean per cent score for initial client assessment
69%
*Labour and delivery clients were not all observed at the same stage. Only 102 observations included the initial client
assessment; the other observations were made at a later stage of labour or later in the service delivery process.
Eighty-nine per cent (89%) of the initial assessments included checking the fetal
presentation with palpation of the abdomen, and 88% included performing a vaginal
examination. Only 57% included taking the client’s pulse, and only 15% included noting
urine output.
Assessment of Danger Signs
Assessment of danger signs during the initial client assessment is critical in order to identify
potential obstetric emergencies and, if necessary, triage clients to urgent care. Figure 6.1
details some of the key steps in the initial client assessment and whether they were
conducted in the deliveries observed for the study. Most of the danger signs were assessed
infrequently. Whereas 60% of clients were assessed for vaginal bleeding, only 2% were
assessed for shortness of breath.
Quality of Maternal and Newborn Health Services in Zanzibar
13
Figure 6.1 Assessment of Danger Signs in Initial Assessment
100%
90%
80%
70%
60%
60%
50%
40%
26%
30%
20%
24%
11%
10%
10%
15%
2%
0%
Fever
Foul-smelling Headache or Swollen hand Convulsion/ Shortness of
breath
discharge blurred vision
and face
loss of
counsiousness
Vaginal
bleeding
Assessment of Previous Complications in Multiparous Clients
For multiparas, the initial assessment should also determine whether the client has
experienced any previous complications (such a previous caesarian section), which might
affect the management of her current labour. Table 6.2 details the assessment of previous
complications for multiparous clients admitted into the maternity ward. The initial
assessments of 52 multiparous women were observed in the study.
Table 6.2 Assessment of Previous Complications among Multiparous Clients
MULTIPAROUS CLIENTS
ASSESSED (n=52)
COMPLICATIONS ASSESSED BY PROVIDER
n
%
High blo, lkiuod pressure
11
22
Convulsions
5
10
Heavy bleeding during or after delivery/haemorrhage
23
44
Caesarean section
27
52
Stillbirth
20
38
Prolonged labour
4
8
Newborn death
14
27
Abortion
23
44
Assisted delivery
9
17
Mean score for assessment of previous complications
15
29
The low assessment of clients for previous history of prolonged labour (8%) and convulsions
(10%) is particularly troubling.
6.3. WOMAN-FRIENDLY CARE DURING LABOUR AND DELIVERY
(INTERPERSONAL COMMUNICATION)
The MAISHA program provides training, quality improvement initiatives and supervisory
support to health care providers on improved interpersonal skills with clients. Provider
attitudes are one of the most important determinants of a woman accessing facility-based care;
it is critical to women’s and communities’ perceptions of quality of care (Kruk et al. 2009).
Table 6.3 presents the findings from observations of components of woman-friendly care
during the initial assessment and the first stage of labour.
14
Quality of Maternal and Newborn Health Services in Zanzibar
Table 6.3 Woman-Friendly Care Components Observed during the Initial Assessment and First
Stage of Labour
INITIAL ASSESSMENT
n=104
%
Greets client
91
88
Encourages presence of support person
23
22
Asks for questions
22
21
Explains procedures before performing them
65
65
Informs client of findings
66
66
FIRST STAGE OF LABOUR
n=116
%
Explains what happens in labour
52
45
Encourages client to consume fluids/food
73
63
Encourages/assists client with ambulating
84
72
Supports client in a friendly way
106
91
Mean percent score for woman friendly care
59%
Overall, approximately 59% of the observed clients received woman-friendly care. The
components observed most frequently were greeting the client (88%) and supporting the
client in a friendly way (91%), while those observed infrequently included encouraging the
presence of a support person (22%) and asking for questions (21%).
6.4. CARE DURING THE SECOND AND THIRD STAGE OF LABOUR
Because the majority of maternal deaths occur during labour, childbirth and immediately
postpartum (60% of maternal deaths occur in the first 48 hours following birth), it is vital
that care provided during this time is optimal (WHO 2006a). Figure 6.2 shows the
observation of essential tasks during the second and third stage of labour.
Figure 6.2 Tasks for Management of Second and Third Stages of Labour
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
95%
91%
78%
52%
Supports
perineum as
baby's head is
delivered
45%
Takes mother’s Palpates uterus
Assesses for
Assesses
15 minutes after
vital signs 15
perineal and
completeness of
delivery of
the placenta and vaginal lacerations minutes after birth
placenta
membranes
Although the perineum was examined for vaginal tears in 95% of the labours observed, vital
signs were taken 15 minutes after birth in 52% of the labours and palpation of the uterus
was performed in only 45% of the labours observed.
Information on active management of the third stage of labour (AMTSL) is included in
section 7.1.
Quality of Maternal and Newborn Health Services in Zanzibar
15
6.5 IMMEDIATE AND ESSENTIAL NEWBORN CARE
At the time of birth, the newborn has to make a rapid transition to extra-uterine life. Key
care practices can facilitate this transition and, if they are performed effectively, minimize
the complications that lead to newborn death and morbidity. These practices focus on clean
childbirth and cord care, thermal protection through warming, and early and exclusive
breastfeeding.
All of the immediate newborn care interventions observed in this study are simple to
perform and use minimal resources; yet the findings were variable. Table 6.4 presents
findings on immediate newborn care practices.
Table 6.4 Practice of Immediate and Essential Newborn Care
LABOUR AND DELIVERY CLIENTS
(n=203)
NEWBORN CARE PRATICE
n
%
Place newborn on the mother’s abdomen
146
72
Immediately dry baby with towel
69
34
Discard wet towel and cover newborn with dry towel
69
34
Cut cord with clean blade
203
100
Help initiate breastfeeding within one hour
41
20
Score for essential newborn care (all items correct)
12
6
The universal adherence to the practice of cord cutting with a clean blade is expected in all
facility births, and this standard was met in all of the deliveries observed.
Helping the woman initiate breastfeeding was not typically practiced (only 20% of
deliveries). Initiation of breastfeeding is a key lifesaving intervention for newborns. Without
it, both the baby and the mother are deprived of benefits, because early breastfeeding may
also reduce postpartum blood loss.
6.6. HARMFUL AND UN-INDICATED PRACTICES
A number of harmful and un-indicated practices were observed during the deliveries.
Harmful practices are those that have been shown to have no benefit or to cause harm. Unindicated practices are those that should be used only with specific indications and that may
otherwise be harmful or unnecessary. Table 6.5 shows the harmful and un-indicated
practices observed in the study.
Table 6.5 Harmful and Un-Indicated Practices
NUMBER OF
L&D CASES
(n=205)
% OF
DELIVERIES
Pubic shaving
1
<1
Applying fundal pressure
4
2
Slapping newborn
2
1
Holding newborn upside down
6
3
Stretching the perineum
27
13
No harmful practices observed
174
85
HARMFUL PRACTICES
16
Quality of Maternal and Newborn Health Services in Zanzibar
NUMBER OF
L&D CASES
(n=205)
% OF
DELIVERIES
Manual exploration of the uterus after delivery
1
<1
Routine use of episiotomy
8
4
Aspiration of newborn mouth and nose at birth
2
1
No un-indicated practices observed
196
96
No harmful or un-indicated practices observed
168
82
HARMFUL PRACTICES
Un-indicated practices
In 82% of the labours observed, no harmful or un-indicated practices were found. The most
frequently observed harmful practice was stretching of the perineum (13% of cases).
6.7. INFECTION PREVENTION
Infection Prevention Measures
Use of standard infection prevention and hygiene measures is a core concept for the
prevention of infection transmission in health care settings. Standard infection prevention
practices recommended during delivery care are aimed not only at preventing maternal and
newborn infections, but also infection of the health worker, other workers and the public.
Table 6.6 presents infection prevention measures observed in the L&D cases. The number of
observations for each stage of L&D varies based on the starting time of the observation of
each client.
Table 6.6 Infection Prevention Measures for L&D Clients
OBSERVED CLIENTS
INITIAL ASSESSMENT
n=99
%
70
71
n=113
%
Washes hands before examination during labour
72
64
Wears disinfected or sterile gloves for vaginal examination
110
96
Wears clothing to protect face, hands and body
31
28
n=205
%
Safely disposes of all sharps
176
86
Decontaminates all reusable instruments in 0.5% chlorine solution
200
97
Safely disposes of all containment waste
198
97
Removes apron and wipes with 0.5% chlorine solution
40
20
156
76
Washes hands before any examination
First stage of labour
Immediate newborn and postpartum care
Washes hands thoroughly with soap and water and dries them
Mean per cent score for infection prevention
73%
Correct handwashing practices were observed among 71% of providers at initial client
assessments, 64% of providers during the first stage of labour and 76% of providers for
immediate newborn care.
Sharps disposal, decontamination and waste disposal were performed correctly most of the
time (86%, 97% and 97%, respectively). Wearing an apron and/or other protective clothing
was not commonly practiced.
Quality of Maternal and Newborn Health Services in Zanzibar
17
6.8. USE OF A PARTOGRAPH TO MONITOR LABOUR
Partograph use is critical when monitoring maternal and foetal well-being and the progress
of labour, as it allows providers to make appropriate decisions on when to take actions to
save a woman’s and/or newborn’s life. The World Health Organization (WHO) recommends
using a partograph to help birth attendants make better decisions for the diagnosis and
management of prolonged and obstructed labour and to help detect foetal distress and other
complications of labour (WHO 2000; WHO 2006a).
Table 6.7 Partograph Use during Labour
MATERNITY CLIENTS
OBSERVED (n=208)
PARTOGRAPH USE
n
%
120
58
Old WHO partograph*
35
77
New WHO partograph**
25
33
92
77
Frequency and duration of contractions correctly filled
46
38
Foetal heart tones correctly filled
72
60
Maternal pulse correctly filled
14
12
All three items filled in at least every 30 minutes during labour
8
7
Blood pressure recorded every 4 hours
62
52
Birth time correctly filled
120
100
Delivery method correctly filled
120
100
Partograph used during labour
Partograph use by type (n=120)
Correct completion of the partograph (among those who used partograph)
Partograph initiated at the right time
Partograph filled in every half hour with
Partograph filled in after delivery with
* Partograph includes latent phase of labour
** Partograph has no latent phase and active phase labour noted to begin 4cm cervical dilatation
The partograph was used for 58% of the maternity clients observed. The majority of the time
the partograph was filled at the beginning of the observation (77%) and after delivery
(100%). However, providers did not consistently use the partograph for recording maternal
pulse, foetal heart tones or frequency and duration of contractions every 30 minutes.
18
Quality of Maternal and Newborn Health Services in Zanzibar
7. Prevention and Management of Selected
Maternal and Newborn Health Complications
7.1 PREVENTION AND MANAGEMENT OF POSTPARTUM
HAEMORRHAGE
Postpartum haemorrhage is the main cause of maternal death globally, and in mainland
Tanzania and Zanzibar, and many efforts to improve maternal health are focused on
reducing mortality due to PPH. The most common cause of PPH is uterine atony, or failure
of the uterus to contract after delivery. In Zanzibari health care facilities, AMTSL is
recommended for all births as the standard practice to prevent PPH, and oxytocin is the
drug of choice for AMTSL, followed next by ergometrine and then misoprostol.
Active Management of the Third Stage of Labour
Large-scale efforts in many countries have focused on prevention of PPH with AMTSL,
which has three components: (1) administration of a uterotonic within one minute of birth
(relaxed definition is a uterotonic within three minutes of birth); (3) delivery of the placenta
by controlled cord traction (CCT); and (3) uterine massage (ICM/FIGO 2006).
The uterotonic of choice for PPH globally is oxytocin, with a recommended dose of 10 IU
administered intramuscularly (WHO 2006b). For optimal effect, oxytocin requires
refrigeration, intramuscular injection, and administration by skilled providers. If oxytocin is
not available, intramuscular ergometrine or syntometrine or oral misoprostol is
recommended.
The practice of AMTSL was assessed at the Zanzibar facilities based on the use of the three
AMTSL criteria. The criteria were further defined to include both provision of a uterotonic
(which could be oxytocin, ergometrine, syntometrine or misoprostol) and the timing of the
provision (within one minute or within three minutes), controlled cord traction and uterine
massage immediately follow placenta delivery.
Overall, 60% of the deliveries observed included AMTSL with any uterotonic given within
one minute, and 88% of deliveries observed included AMTSL with any uterotonic given
within three minutes. AMTSL with oxytocin within one minute was correctly performed in
20% of deliveries observed.
Figure 7.1 shows the proportion of births observed in which the individual components of
AMTSL were performed correctly, not taking into account whether the other components
were performed correctly. A uterotonic was administered within one minute of birth in 63%
of births; CCT was performed in 81% of births and uterine massage was conducted following
the delivery of the placenta in 62% of observed deliveries.
Quality of Maternal and Newborn Health Services in Zanzibar
19
Figure 7.1 Proportion of Births Observed in Which AMTSL Tasks Were Performed Correctly
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
88%
81%
63%
62%
Provision of uterotonics Provision of uterotonics Controlled cord traction
within 3 minutes*
within 1 minute*
Uterine massage
*Uterotonics include oxytocin, ergometrine or misoprostol.
A uterotonic was given during the third or fourth stage of labour for every delivery observed
for this study. In 60% of the cases, the uterotonic was oxytocin, and in 40% of cases women
were given misoprostol (prostaglandins). Ergometrine was used in only one delivery.
Interestingly, the use of misoprostol is significantly more common in Zanzibar than on the
mainland, where it was used in only 6% of observations.
A uterotonic (oxytocin, ergometrine or misoprostol) was given within three minutes of
delivery to 88% of women observed in the study.
Table 7.1 Uterotonics Administered for AMTSL
UTEROTONICS ADMINSITERED
n
%
120
60
Ergometrine
1
0
Misoprostol
80
40
Oxytocin
Figure 7.2 shows the proportion of births observed in which different components of AMTSL
were performed correctly, with a focus on the use of oxytocin rather than other uterotonics.
The items are additive as the bars move from left to right, and the denominator changes for
each bar as cases that do not meet the criteria are dropped. While uterotonics were used in
100% of the deliveries observed, only 59% of the clients received oxytocin IM (the correct
route). Among them, 35% received it within one minute and 47% received it within three
minutes.
Using the globally accepted definition (with oxytocin), AMTSL was correctly performed in
20% of all deliveries observed.
20
Quality of Maternal and Newborn Health Services in Zanzibar
Figure 7.2 Proportion of Deliveries with Correct Provision of AMTSL with Oxytocin
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
100%
88%
60%
59%
63%
56%
29%
20%
Provision of
any
uterotonic
within 3
minutes of
delivery*
Oxytocin
used
(+) Correct
route
(+) Correct
dose and
units
(+) Correct
timing
(within 3
minutes)
(+) Uterine
(+)
(+) Correct
Controlled massage =
timing
(within 1 cord traction FIGO/ICM
standard
minute)
AMTSL
*Any uterotonic includes oxytocin, ergometrine and misoprostol.
Figure 7.3 shows the use of any type of uterotonic during labour and delivery. The
percentage of women receiving any uterotonic within three minutes of birth (relaxed
definition) is much higher (50%) than the percentage receiving oxytocin within three
minutes (47%). The low result for correct AMTSL with oxytocin given within one minute
suggests a need to promote the use of oxytocin as the preferred drug for AMTSL, particularly
at the lower-level facilities.
Figure 7.3 Proportion of Deliveries with Correct Provision of AMTSL with Any Uterotonic
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
100%
59%
50%
36%
30%
20%
Any uterotonic*
(+) Correct
route
(+) Controlled
(+) Correct
(+) Correct
timing (within 3 timing (within 1 cord traction
minute)
minutes)
(+) Uterine
massage =
FIGO/ICM
standard
AMTSL
*Any uterotonic includes oxytocin, ergometrine and misoprostol.
7.2 PREVENTION AND MANAGEMENT OF PRE-ECLAMPSIA/
ECLAMPSIA, INCLUDING SCREENING
Preeclampsia and eclampsia (PE/E) are among the most dangerous complications of
pregnancy. Eclampsia, the advanced stage of this disorder, is a major cause of maternal
deaths. To date there is no conclusive evidence on interventions that can prevent the
occurrence of PE/E. However, early detection and treatment of pre-eclampsia is beneficial as
it permits clinical monitoring and prompt therapeutic intervention for severe pre-eclampsia
and eclampsia, should they occur.
Quality of Maternal and Newborn Health Services in Zanzibar
21
Table 7.2 Screening for Pre-Eclampsia during Labour, Initial Assessment
OBSERVATIONS
(n=99)
PROVIDER SCREENING ACTIONS
n
%
Asks about signs of PE/E*
60
60.
Checks blood pressure (initial assessment)
85
85.
Conducts both PE/E screening elements
59
59.
Tests urine for presence of protein
12
12
Records BP on partograph at least once every four hours during
labour** (n=208)
60
29
*Asks about at least one of the following: headaches/blurred vision, swollen face/hands, convulsions/loss of
consciousness
**As long as diastolic is less than 90 mm Hg
BP measurement at initial assessment was carried out for the majority of women admitted
in labour (68%), but less than half (44%) of the women had their BP recorded on the
partograph every four hours. Two key screening elements (asking about danger signs and
checking BP) were performed for 59% of the women observed. In contrast to ANC, where
86% of women had their urine tested for protein, urine testing was only conducted for 10% of
women during their initial assessment in labour.
Magnesium sulfate (for prevention and treatment of PE/E) was available in the pharmacies of
only two of the facilities studied. Eight of the facilities had magnesium sulfate in the labour
ward.
7.3 COMPLICATED CASES OBSERVED DURING STUDY
Three types of complications at birth were observed during the study: PE/E, PPH and birth
asphyxia (newborn resuscitation). Only one case of pre-eclampsia/eclampsia was observed,
and no treatment was administered. Mother and baby both survived the spontaneous
vaginal delivery. One case of PPH was observed, and no treatment was administered.
Twelve cases of newborn asphyxia were observed, and five of them were managed with a bag
and mask. No newborns died, but one was referred to another facility.
22
Quality of Maternal and Newborn Health Services in Zanzibar
8. Health Worker Knowledge
Health care providers in ANC and maternity wards were given a knowledge assessment (by
interview) after they gave informed consent. Table 8.1 presents the characteristics of the
health care providers who participated in the knowledge assessment. A total of 51 providers
(75% nurse/midwives, 8% MCHAs and 18% others) participated, of whom 25 were noted to
be working in the delivery room.
Scores were highest on procedures during labour and delivery (93%) and recording of
observations and monitoring (96%). The providers achieved much lower knowledge scores on
actions, tests and interventions to manage a woman with postpartum malaise (25%), actions
to reduce mother-to-child transmission of HIV (PMTCT) (33%), and actions, tests and
interventions for retained placenta/products of conception (33%).
Table 8.1 Provider Knowledge Scores on Maternal Health Topics
SIGNS AND MANAGEMENT OF LABOUR AND DELIVERY
MEAN SCORE
(%)
Observations and monitoring during labour and delivery
54
Recording of observations and monitoring
96
Routine procedures during labour and delivery
93
Actions to reduce PMTCT during labour and delivery
33
Signs to assess in woman with heavy postpartum bleeding
48
Likely location of tears and lacerations
49
Actions, tests and interventions for heavy postpartum bleeding from
atonic/poorly contracted uterus
38
Actions, tests and interventions for retained placenta/products of conception
33
Signs of obstructed labour
35
Actions, tests and interventions for obstructed labour
39
Tests or evaluations for woman who presents 72 hours postpartum with general
malaise
45
Actions, test and, interventions for woman who presents 72 hours postpartum
with general malaise
25
Of concern are the low scores on actions, tests and interventions for heavy bleeding
postpartum, which only 33% of providers answered correctly. Knowledge of signs and actions
around obstructed labour were similarly low.
Scores related to the provider’s knowledge of PE/E are presented in detail in Appendix C.
Instead of a taking a knowledge test on newborn resuscitation, providers were observed
conducting a simulated resuscitation on a model. Findings from the providers’ performance
on the simulation are presented in Table 8.2.
Quality of Maternal and Newborn Health Services in Zanzibar
23
Table 8.2 Provider Performance on Newborn Resuscitation Simulation
STEPS IN SIMULATED RESUSCITATION (ALL MUST BE CORRECTLY
PERFORMED)
MEAN PERCENT SCORE (%)
Stimulation procedures: drying the newborn; placing the newborn on
warm clean surface with head in slightly extended position; suction with
bulb or catheter in mouth or nose (all items)
60
Ventilation procedures: placing correct size mask over newborn’s chin,
mouth and nose; squeezing bag appropriately with two fingers or hand;
ventilating at 40 breaths/minute (all items)
40
Adjustment: Checking neck position, checking seal, repeating suction,
squeezing harder (any proper adjustment)
80
Overall mean score for simulation
60
Provider performance on newborn resuscitation was low in the area of ventilation (40%).
More than half knew how to stimulate (60%), and adjustment was observed in 80% of the
simulations observed. However, it appears that updating in competencies for newborn
resuscitation is needed.
24
Quality of Maternal and Newborn Health Services in Zanzibar
9. Discussion
Quality in antenatal care and labour and delivery is at the core of improving MNH services
everywhere, and this study has identified both strengths and gaps in the provision of
essential lifesaving maternal and newborn care in the facilities observed in Zanzibar. The
MOH in Zanzibar is committed to ensuring improved access to quality care (Zanzibar
Ministry of Health 2008). Some urgency is needed in making improvements in care, as 2015,
the year for review of the MDGs, looms. Zanzibar’s policy proposes that deliveries should be
conducted at health facilities with the assistance of skilled birth attendants, but evidence
shows that about 37% of deliveries occur at home with traditional birth attendants, family
members or relatives (Zanzibar Ministry of Health 2010). Where the quality of care is poor,
women are less likely to access such care, despite its availability (Raven et al. 2011). To
encourage women to access care and deliver in facilities, various aspects of the quality of
care need to be addressed in Zanzibar.
Zanzibar’s “Road Map to Accelerate the Reduction of Maternal, Newborn and Child
Mortality” highlights the need to address the three delays linked to maternal deaths
(Zanzibar Ministry of Health 2008). The third delay (the delay in receiving appropriate care
once at the health facility) is attributable to:
•
Inadequate facilities, infrastructure, medical equipment, drugs, supplies and trained
personnel
•
Poor training and poor attitudes amongst health personnel
•
Lack of finances
These factors all surfaced in the findings of this study in Zanzibar to varying degrees.
Postpartum haemorrhage and PE/E are the two main maternal complications reported in
health facilities in Zanzibar (Zanzibar Ministry of Health 2010). The focus of this discussion
will be on these main areas as well as on newborn care. Although this study was small in
scope, it is notable that neither of the complicated cases observed—one case of PE/E and one
case of PPH—received care specific to the complication.
9.1 PREVENTION AND MANAGEMENT OF PE/E
Hypertensive disorders of pregnancy, including PE/E, are characterized by hypertension and
proteinuria, from the twentieth week of pregnancy until 42 days after delivery. Hypertensive
disorders in pregnancy, especially severe pre-eclampsia and eclampsia, are a major
contributor to maternal mortality worldwide. They are the second leading cause of maternal
deaths in Zanzibar. The majority of deaths due to PE/E are avoidable through timely and
effective care, so optimizing health care to prevent and treat women with hypertensive
disorders is a necessary step toward achieving the Millennium Development Goals (WHO
2011).
Clinical practice policies and guidelines for ensuring that cases of PE/E are managed
according to the best and most recent evidence are in place in Zanzibar. However, the
findings from this study suggest that provider practice is not at an appropriate level or in
line with clinical guidelines, due to both supply issues in supplies and the gaps in knowledge
and skills found in the provider scores on the PE/E case study.
Antenatal Care
ANC is an important factor in reducing maternal and newborn deaths, and since the
development of Zanzibar’s road map to accelerate reduction of maternal mortality, ANC
services have been given special focus. The findings of this study bring to the forefront areas of
relative strength in ANC (such as provision of key ANC services) as well as some significant
weaknesses (such as counselling on danger signs). One of the main goals of ANC is screening
Quality of Maternal and Newborn Health Services in Zanzibar
25
for PE/E, but only 55% of observed ANC consultations included screening for PE/E (using a
composite indicator). One positive finding was the high number of women (86%) whose ANC
visit included a test for protein in the urine (Tanzania MOH/MCHIP 2011).
Care in Labour
Screening at the time of admission is secondary prevention of PE/E, and tertiary prevention
includes recognition and effective management of complications such as rising blood pressure.
There was an extremely low rate of assessment of danger signs when clients first arrived at
the hospital. Both assessment for danger signs in the current pregnancy and asking about
complications in previous pregnancies were both very low. This finding indicates that
assessment of dangers signs, although it may be somewhat underreported, is practiced far less
commonly than desired standards. Assessment of danger signs is extremely important in
identifying women who need urgent attention and triage into care, and it contributes to
addressing the “third delay” that leads to maternal deaths (Mbakuru 2009).
Urine testing for protein during labour was virtually non-existent—only 3% of clients
received a urine test. In addition, only about a third of the clients observed had their blood
pressure recorded at least every four hours on the partograph. These are missed
opportunities for screening and history-taking for PE/E danger signs.
The partograph can be used both for monitoring the progress of labour and for monitoring
the mother’s well-being, including detecting rapid onset of hypertension. Although providers
have a good working knowledge of the partograph, it is nevertheless underused. Completing
a partograph retrospectively (possibly with made-up data) defeats the partograph’s purpose
and potential as a decision-making tool and also indicates poor accountability. A shortage of
staff, especially midwives, can also contribute to poor use of the partograph (UNFPA 2011).
A recent global review of the key interventions related to maternal, newborn and child
health finds that antihypertensive drugs (to treat high blood pressure) and magnesium
sulfate for eclampsia are high-impact best practices (Partnership for Maternal, Newborn &
Child Health 2011). Magnesium sulfate, especially, is a lifesaving drug that should be
available in facilities throughout the health system (WHO 2011). Magnesium sulfate was
available in the labour ward in eight facilities and in the pharmacy of only two of the
facilities overall, and it was also lacking in health centres/dispensaries (35%).
Antihypertensives also were not widely available: only three facilities had nifedipine in the
pharmacy and none had hydralazine in the labour ward.
This finding suggests that there is a bottleneck in the supply chain management for both
essential drugs. Magnesium sulfate is the drug of choice for the prevention and treatment of
eclampsia, but it is underused as a result of barriers on multiple levels (Firoz et al. 2011).
9.2 PREVENTION AND MANAGEMENT OF PPH
Postpartum haemorrhage is the main cause of maternal deaths globally and in Zanzibar,
and many efforts are focused on reducing it. WHO recommends that all women receive
AMTSL, including administration of a uterotonic within one minute after birth, CCT to
deliver the placenta, and uterine massage (WHO 2006b). Zanzibar instituted this practice
some years ago. Although other factors can contribute to PPH, this discussion focuses on
prevention of PPH using AMTSL.
One of the strongest PPH findings was that all women received a uterotonic, which offers
benefits even in the absence of other components of AMTSL. A uterotonic was given to the
woman in the third or fourth stage of labour during all of the deliveries observed for this
study. In 60% of the cases, the uterotonic was oxytocin, and in 40% of cases women were
given misoprostol. Ergometrine was hardly used at all (provided in only one delivery
observed), which is reassuring given its side effects. Misoprostol is more widely available in
26
Quality of Maternal and Newborn Health Services in Zanzibar
Zanzibar than on mainland Tanzania because of the advantage that it does not require
refrigeration.
Overall, 60% of the deliveries in which AMTSL was observed included a uterotonic given
within one minute of birth, and 88% included a uterotonic given within three minutes of
birth. AMTSL with oxytocin within one minute was correctly performed in 20% of deliveries.
All facilities were noted to have oxytocin available, which is an important finding for both
prevention and management of PPH.
Data from health worker knowledge assessments indicate significant gaps in how to
manage PPH.
9.3 ESSENTIAL NEWBORN CARE
Simple interventions in the first few minutes and first hour of life can significantly improve
newborn health (Partnership for Maternal, Newborn & Child Health 2011). These
interventions include the provision of thermal care (immediate drying, warming, skin-toskin contact and delayed bathing) for all newborns to prevent hypothermia. Only one-third
of newborns observed in the study were dried immediately after birth, and it is not clear
whether this was the result of a gap in knowledge and skills or a lack of materials with
which to dry the baby (e.g., clean towels and kangas).
Initiation of breastfeeding within one hour was infrequent, observed in only 20% of births.
This is a key lifesaving intervention for newborns, providing benefits to both the baby and
mother, as early breastfeeding may also reduce postpartum blood loss (UNICEF UK 2008).
It is worth noting that according to the Zanzibar’s Road Map, 54% of babies are breastfed
within the first hour of life; it may be that the observation of breastfeeding is underreported
in this study (Zanzibar Ministry of Health 2008).
9.4 OTHER ISSUES OF NOTE
Supportive care is highly valued by women and their families and increasingly is being
recognized as a woman’s right. The importance of this issue is reflected in the recently
released charter from the White Ribbon Alliance, Respectful Maternity Care: The Universal
Rights of Childbearing Women, which includes the right to the woman’s “choice of
companionship during maternity care” (White Ribbon Alliance 2011). During the initial
client assessment in labour, a relatively small percentage of women observed in the study
were encouraged to have a support person with them (22%). There is sound evidence that
the presence of a support person during labour improves birth outcomes, and this presence is
more important when staffing levels are insufficient to provide continuous care in labour
(Partnership for Maternal, Newborn & Child Health 2011).
Prevention of Mother-to-Child Transmission of HIV
HIV/AIDS is increasing as a major cause of maternal death in sub-Saharan Africa (WHO
2010). However, the prevalence of HIV among pregnant women in Zanzibar is low and has
dropped from 0.29% in 2009 to 0.16% in 2010. That said, only 54% of ANC observations
included counselling and testing for HIV. Nearly half of all pregnant women are not being
tested. Syphilis prevalence has risen from no cases in 2009 to 0.27% in 2010, indicating that
sexually transmitted infections, including HIV, should be addressed in ANC and maternity
services (Zanzibar Ministry of Health 2010).
Postpartum Family Planning
The combination of a high fertility rate and a low contraceptive prevalence rate in Zanzibar
increases the lifetime risk of maternal death. Zanzibar has a good family planning service
infrastructure, with most of the population having access to services. However, use of these
services is not increasing substantially (Zanzibar Ministry of Health 2010). Counselling on
family planning is a more recent component of focused antenatal care, but only 38% of the
Quality of Maternal and Newborn Health Services in Zanzibar
27
ANC clients observed in this study were counselled on PPFP. Thus, this also is a missed
opportunity.
A recently released review of essential MNH interventions reinforces the need to scale up
evidence-based, low cost, effective interventions (Partnership for Maternal, Newborn &
Child Health 2011). However, scale-up of these interventions can only be effective if
providers work within an enabling environment. Facilitating the enabling environment,
including improving the integration of essential interventions and services such as family
planning and PMTCT, should be a priority in Zanzibar.
9.5 COMPARING KEY FINDINGS FROM ZANZIBAR AND MAINLAND
TANZANIA
In July–August 2010, a study using the same tools and most of the same data collectors as
those used in Zanzibar was conducted in 52 health facilities in 11 regions in mainland
Tanzania. Comparative findings on a few key indicators are presented in Table 9.1.
Table 9.1 Comparison of Study Findings: Zanzibar and Mainland Tanzania
Items in bold are statistically significant
ZANZIBAR
MAINLAND
P-VALUE
n
%
n
%
PE/E screening
30
55%
90
24%
0.000 (t= 4.80)
Urine tested for protein
48
86%
153
40%
0.000 (t= 6.4)
Preventative treatment for malaria
27
49%
160
68%
0.06 (t=1.84)
Assessment of danger signs
20
21%
33
11%
0.01 (t=2.5)
Assessment of prolonged labour in
previous pregnancy among multiparas
14
30%
29
16%
0.03 (t=2.22)
Assessment of hypertension in
previous pregnancy among multiparas
8
16%
21
11%
0.33 (t=0.9)
Provision of AMTSL (WHO definition,
with oxytocin within one minute)
70
35%
170
41%
0.15 (t=1.43)
Provision of AMTSL (with oxytocin
within three minutes)
94
47%
278
67%
0.000 (t=4.76)
Provision of AMTSL (with any
uterotonic within three minutes)
174
88%
305
76%
0.0006 (t=3.45)
Drying and wrapping of the baby
69
34%
377
90%
0.001 (t=14.9)
Clean cord care
198
100%
419
100%
0
Breastfeeding initiated within one
hour
42
20%
182
44%
0.000 (t=5.18)
Handwashing (initial assessment)
70
71%
165
54%
0.0031 (t=2.98)
Decontamination
200
97%
383
88%
0.0002 (t=3.69)
ANC
Maternity
Newborn care
Infection prevention
The comparison shows that Zanzibari women were getting significantly more PE/E
screening and urine testing for protein during their ANCE visits than women in mainland
Tanzania, but provision of IPTp was less frequent in Zanzibar than on the mainland (not
statistically significant).
28
Quality of Maternal and Newborn Health Services in Zanzibar
During their maternity care, Zanzibari women more often were assessed for danger
signs and multiparas were more often assessed for previous complications (though still
low, at 21% and 30%, respectively) compared to mainland clients. The provision of
AMTSL with any uterotonic using the relaxed definition of three minutes was
significantly more common in Zanzibar than on mainland, while provision of AMTSL
using oxytocin was less common (statistically significant); this is likely due to a higher
prevalence of misoprostol use in Zanzibar than in mainland health facilities.
In newborn care, drying and wrapping of the baby appears to be an area of real concern.
This practice was observed three times more often in mainland health facilities than in
Zanzibar facilities. Initiation of breastfeeding was very low in both Zanzibar and
mainland health facilities.
Infection prevention practices were more often observed in Zanzibar than on the
mainland, with both decontamination and handwashing performed correctly more
frequently (statistically significant).
Quality of Maternal and Newborn Health Services in Zanzibar
29
10. Conclusion and Recommendations
Increasing women’s access to quality maternity services is a focus of global efforts to reduce
maternal and newborn mortality. Quality of care was recognized as a key element for
improved health outcomes and efficiency in WHO’s widely adopted framework for health
system strengthening in resource-poor countries (WHO 2007).
This quality of care study in Zanzibar found major gaps in coverage and performance of key
competencies in routine care in pregnancy, labour and delivery, and the management of
some complications (notably, PE/E and PPH). The presence of a “skilled birth attendant”
does not necessarily mean skilled care is being provided. To perform effectively, skilled
providers need to work within an enabling environment. Shortages of skilled staff,
equipment and supplies were observed in this study. Many of the critical challenges in
reducing maternal, newborn and child morbidity and mortality are health system issues that
need urgent attention.
Features of quality of care that are specifically important for maternal and newborn health
include a rights-based approach and evidence-based practices—themes that are echoed in
Zanzibar’s Road Map (Zanzibar Ministry of Health 2008). It is now urgent that MOH
leadership in Zanzibar ensure that resources are mobilized and actions taken to provide
quality services to women and their families.
The following recommendations can be made based on the findings from this study:
•
Encourage providers to perform a “quick check” at every contact with a pregnant or
postpartum woman to identify complications quickly and thereby initiate appropriate care.
•
Ensure that all supplies for routine delivery care and management of complications are
available—for example, magnesium sulfate and antihypertensive drugs for managing PE/E.
All facilities should have emergency trays of available drugs, and these should be checked at
least daily. Ward pharmaceutical assistants, hospital pharmacists and central medical
stores should work closely and have a clear protocol for avoiding unnecessary stock-outs on
the wards.
•
Improving the quality of ANC in health centres and dispensaries should be a special focus.
This does not mean improving the frequency of visits but rather the quality of interventions
such as counselling and preventative care
•
Allow for competency-based in-service/refresher training that can be conducted offsite, as
well as more flexible on-the-job training. Training should include clinical simulations and
other activities to encourage teamwork and improve efficiency in dealing with clinical
emergencies.
•
Scale up quality improvement approaches to enable providers to apply and become confident
in implementing best practices, especially after training
•
Ensure that up-to-date job aids are disseminated to all facilities, especially those linked to
emergency situations such as managing eclampsia, newborn resuscitation and PPH.
•
Strengthen pre-service education of all health care providers to ensure competency-based
approaches that lead to stronger performance and retention of knowledge and skills as well
as evidence-based practices that respond to Zanzibar’s priority health care needs.
•
Address provider accountability in relation to their performance, delays in providing care
and recording/reporting of information (e.g., use of the partograph). The professional
associations and regulatory bodies (e.g., Zanzibar Nurses and Midwives Council and
Zanzibar Nurses Association) have a key role to play.
30
Quality of Maternal and Newborn Health Services in Zanzibar
Appendix A. Pharmacy Stock
# PRESENT
IN PHARMACY
# PRESENT
IN L&D
Ampicillin, injectable
0
2
Gentamicin, injectable
5
6
Magnesium sulfate
2
8
Oxytocin
6
9
Hydalazine or apresoline
4
0
Nifedipine
3
0
Labetolol
0
0
Ergometrine
1
9
Misoprostol
7
9
PHARMACY STOCK AND CONDITIONS
# OF FACILITIES
STOCK CARDS MATCH FOR KEY DRUGS
No expired medicines
2
FEFO system
7
Physical conditions adequate
Off the floor and protected from water
8
Protected from sun
9
Room clean of evidence of rodents/pests
7
Adequate control of stock
Received routine supply within past 3 months
7
Always receive accurate orders (in past 3 months)
2
Quality of Maternal and Newborn Health Services in Zanzibar
31
Appendix B. Equipment Inventory
n=9
% OF
FACILITIES
Soap for handwashing
9
100
Water for handwashing
9
100
Piped water or bucket with tap
9
100
Soap and piped water/bucket with tap
9
100
Sharps container
9
100
Already mixed decontaminating solution
8
89
Clean (or sterile) gloves
9
100
Functioning electric autoclave
6
67
Functioning non-electric autoclave
1
11
Functioning electric dry heat sterilizer
5
56
Functioning electric boiler or steamer
0
0
Non-electric pot with cover AND functioning heat source
0
0
Functioning automatic timer
3
33
TST indicator strips
2
25
Functioning electric or non-electric equipment for sterilization
9
100
Functioning automatic timer or TST indicator strips
3
33
Written protocols or guidelines for sterilization or disinfection
1
11
Private delivery room with visual and auditory privacy
3
33
Functioning spotlight for pelvic exam (or flashlight/torch or exam light)
6
67
Table or bed for delivery
8
89
24-hour coverage for deliveries (staff present or on-call, schedule observed)
7
78
Guidelines for normal delivery
6
67
Guidelines for emergency obstetric care
8
89
Blank partographs
9
100
Private room with visual and auditory privacy
3
33
Shared room with audio/visual privacy
1
11
Visual privacy only
1
11
No privacy
4
44
INFECTION CONTROL ITEMS
Sterilization items
Capacity for sterilization
Delivery room infrastructure and furnishings
Other elements to support quality delivery
Delivery room privacy
32
Quality of Maternal and Newborn Health Services in Zanzibar
Appendix C. Providers’ PE/E Knowledge
Assessment Scores
PE/E Case Study Scores
SECTION 1: EXAMINATION ACTIONS
MEAN SCORE (%)
Determine time of onset of present symptoms
25
Assess level of consciousness
23
Assess for any convulsions
38
Check vitals
81
Listen to/assess fetal heart tones
42
Check urine protein
58
Mean per cent score (exam actions)
44
Correct working diagnosis (severe pre-eclampsia)
80
Mean score for assessment/diagnosis
49
SECTION 2: INITIAL INTERVENTIONS
Action to take
Stabilize with magnesium sulfate and antihypertensive drug
67
Action to take if presented with convulsion
Administer oxygen at 4–6 L per minute
17
Place in side-lying position
54
Protect from injury
48
Give magnesium sulfate
83
Provide antihypertensive drug
33
Mean per cent score (actions for convulsions)
47
Percent who answered correctly for all items listed above
6
Wrong answer: Give intravenous diazepam
27
Wrong answer: Actively restrain
2
Mean score for initial interventions
50
SECTION 3: ESSENTIAL EQUIPMENT AND SUPPLIES AT THE REFERRAL FACILITY
IV with normal saline or Ringer’s lactate
72
Urinary catheter and urinary bag
72
Patellar hammer
13
Suction machine and catheter
60
Oxygen and adult mask
49
Injectable magnesium sulfate
79
Calcium gluconate
0
Injectable antihypertensive drug
38
Mean per cent score (equipment and supplies)
48
SECTION 4: ACTION TO TAKE ONE HOUR LATER
Repeat magnesium sulfate 4 hours after last dose if reflexes and respiration
are normal
58
Maintain diastolic blood pressure between 90 and 100 through
antihypertensive
27
Monitor labour and begin partograph
44
Quality of Maternal and Newborn Health Services in Zanzibar
33
Ausculate lungs hourly
0
Record fluid intake and output hourly
50
Get and record respirations, reflexes and patellar reflexes hourly
10
Mean per cent score (actions to take one hour later)
32
Wrong answer: Arrange for immediate caesarean section
42
Wrong answer: Induce labour immediately
29
Case study score
Mean per cent score (assessment/diagnosis, initial interventions, equipment
and supplies, actions to take one hour later)
34
45
Quality of Maternal and Newborn Health Services in Zanzibar
References
Campbell O et al. 2006. Strategies for reducing maternal mortality: Getting on with what
works. Lancet Maternal Survival Series. DOI:10.1016/S0140-6736(06)69381-1. Accessed on
March 7, 2012, at:
http://www.amddprogram.org/conference/assets/Resources/Article%20Series/The%20Lancet
%27s%20Maternal%20Health%20Series/Campbell%20et%20al_Strategies%20for%20Reduci
ng%20MM_Lancet_2006.pdf
Firoz T et al. 2011.Pre-eclampsia in low and middle income countries. Best Practice &
Research Clinical Obstetrics and Gynaecology 25: 537–548.
International Confederation of Midwives/International Federation of Gynaecology and
Obstetrics (ICM/FIGO). 2006. Prevention and Treatment of Post-partum Haemorrhage: New
Advances for Low Resource Settings—Joint Statement. Accessed on March 12, 2012, at:
http://www.pphprevention.org/files/FIGO-ICM_Statement_November2006_Final.pdf
Khan K et al. 2006. WHO analysis of causes of maternal death: A systematic review. Lancet
(367): 1066–1074.
Kruk M et al. 2009. Women’s preferences for place of delivery in rural Tanzania: A
population-based discrete choice experiment. American Journal of Public Health 99 (9):
1666–1672.
Mbakuru G et al. 2009. Perinatal audit using the 3-delays model in western Tanzania.
International Journal of Gynecology and Obstetrics 106(1): 85–88.
Partnership for Maternal, Newborn & Child Health. 2011. A Global Review of the Key
Interventions Related to Reproductive, Maternal, Newborn and Child Health. PMNCH:
Geneva. Accessed on March 12, 2012, at:
http://www.who.int/pmnch/topics/part_publications/essentialinterventions14_12_2011low.pdf
Raven JH et al. 2011. What is quality in maternal and neonatal health care? Midwifery
doi:10.1016/ j.midw.2011.09.003.
Tanzania MOH/MCHIP. 2011. Quality of Maternal and Newborn Health Care in Tanzania:
A Survey of the Quality of Maternal and Newborn Health in 12 Regions in Tanzania. Report
1: Findings on Antenatal Care. Accessed on March 12, 2012, at:
http://www.mchip.net/sites/default/files/TanzaniaQoCStudyReportANC_formatted_0.pdf
UNICEF UK. Baby Friendly Initiative. 2008. The Seven Point Plan for Sustaining
Breastfeeding in the Community. Accessed March 19, 2012, at:
http://www.unicef.org.uk/BabyFriendly/Health-Professionals/Going-BabyFriendly/Community/Seven-Point-Plan-for-Sustaining-Breastfeeding-in-the-Community/
United Nations Population Fund (UNFPA). 2011. The State of The World’s Midwifery 2011:
Delivering Health, Saving Lives. Accessed March 19, 2011, at:
http://www.unfpa.org/sowmy/resources/docs/main_report/en_SOWMR_Full.pdf
Van den Broek NR and Graham WJ. 2009. Quality of care for maternal and newborn health:
The neglected agenda. BJOG 116 (Suppl 1): 18–21.
White Ribbon Alliance. 2011. Respectful Maternity Care: The Universal Rights of
Childbearing Women. Accessed March 13, 2012, at:
http://www.whiteribbonalliance.org/WRA/assets/File/Final_RMC_Charter.pdf
Quality of Maternal and Newborn Health Services in Zanzibar
35
World Health Organization (WHO). 2011. WHO Recommendations for Prevention and
Treatment of Pre-eclampsia and Eclampsia. Accessed on March 12, 2012, at:
http://www.who.int/reproductivehealth/publications/maternal_perinatal_health/9789241548
335/en/index.html
WHO. 2010. Trends in Maternal Mortality: 1990–2008: Estimates Developed by WHO,
UNICEF, UNFPA and the World Bank. Geneva: WHO.
WHO. 2007. Everybody’s Business: Strengthening Health Systems to Improve Health
Outcomes—WHO’s Framework for Action. Geneva: WHO.
WHO. 2006a. Pregnancy, Childbirth, Postpartum and Newborn Care. Geneva: WHO.
WHO. 2006b. Recommendations for the Prevention of Postpartum Haemorrhage. Geneva:
WHO.
WHO. 2000. Managing Complications in Pregnancy and Childbirth. Geneva: WHO.
Zanzibar Ministry of Health. 2010. Annual Health Information Bulletin.
Zanzibar Ministry of Health. 2008. Road Map to Accelerate the Reduction of Maternal,
Newborn and Child Mortality in Zanzibar, 2008–2015.
36
Quality of Maternal and Newborn Health Services in Zanzibar
`