April - UNICEF...programs in Cambodia and in other countries was conducted. This was followed by...

:»-wp^^ttfgBg^sEsjgra —* -a r^s^F-^auSparfit ; 4 .

— * - sr M**ma: w, >~«w,i •>-

StungTreng:: PI?O|«tfcr f"T e

April:K - < rrW^ ^'^w^ -v ^ v ^ *** r * "". . - * Jtfl*S.V«*. *»•.-•-. ^^^ jf ,n. . . . .

Evaluation of a Traditional Birth

Attendant Programme in the

Remote North Eastern Province

of Stung Treng, Cambodia

April 2003

, :i: iii w«iii 'iC i

"The Value of a TEA"

" We tell our children about the TBA who delivered them. It is ourcustom, we never forget the gratitude we owe the TBA for bringingour babies into the world. When our son or daughter are oldenough we make sure they know the TBA who delivered them

Our children show their grati tude to the TBA by carrying water,helping with household chores or giving the TBA small gifts.The TBA can call 'her' children to help whenever she needs them."

Quoted from village women's discussions in Siem Bok district,Stung Treng province, March 2003.

Traditional Birth Attendant Programme Stune Trene, Cambodia

ii £*},i£w*^»tV-'.-''.O£Oi ^» i>J&i* iiC«Lftl»>:»>--X».

Map of Cambodia.

Traditional Birth Attendant Programme Stung Treng, Cambodia

Acknowledgements

The evaluators wish to thank the many people who contributed their time andenergy towards making this evaluation a valuable and interesting experience.

Special thanks to:

The village leaders, development committees and families, especially the manymothers and traditional birth attendants who so warmly welcomed us, and openlyshared their experiences and ideas with us. We cannot forget you.

Dr Hengnur -Chief of PHD, Stung Treng.

Ms Boo Vatha -Deputy chief of PHD Stung Treng.

Ms San Channy -vice chief of MCH Stung Treng.

Ms Chan Heang- MCH staff and TEA programme manager

Dr Keo Parin- UNICEF Provincial Advisor, Stung Treng.

Joy Scott, YWAM Project Co-ordinator, Stung Treng.

Philip Scott- Country Director, YWAM.

The staff of the referral hospital obstetric ward, Stung Treng.

Dr Rosikor, UNICEF, Phnom Penh.

The health centre staff of Jamgar Leu, Stung Treng and Srair Krasang health centres.

Mr Chan Vitharin for the beautiful photographs.

Om Mate for the warm hospitality and delicious food.

Traditional Birth Attendant Programme Stung Treng, Cambodia iii

CONTENTS

Page

Acronyms———————— vi

Definition of terms vii

Executive summary— viii

1. Introduction 1

1.1 Present situation concerning maternal child health in Cambodia 1

1.2 TBA programmes in other countries and in Cambodia 2

1.3 Geography and demographics of Stung Treng 3

1.4 The health service situation in Stung Treng 4

1.5 The work of YWAM in Stung Treng 5

1.6 The history of the TBA programme in Stung Treng 5

2. Methodology of the evaluation 10

2.1 Purpose of the evaluation 10

2.2 The evaluation team 10

2.3 Terms of reference 11

2.4 Methodology 12

2.5 Steps in evaluation 12

2.6 Constraints 12

3. Findings 18

3.1 Antenatal care 18

3.2 Safe delivery 25

3.3 Post natal care 31


3.4 Health education 32

3.5 Identification of high risk and referral 33

3.6 Maternal deaths 36

3.7 TEA curriculum 41

3.8 Reporting system 41

3.9 TBA programme costs 44

3.10 Program sustainability 45

4. Discussion 49

5. Conclusions 55

6. Recommendations 58

7. General conclusions 61

8. Appendices 63

A. Programme of evaluation/List of people interviewed 63

B. Job Description of TBA in Stung Treng 67

C. TBA kit resupply costs 67

D. MCH data- 68

E. Suggested additional TBA programme indicators 70

F. References 71


Acronyms and abbreviations

AIDS

A/N

CDHS

CS

DFID

HOC

EmOC

FS

HIV

MCH

MMR

MOH

MSF

NCHADS

NGO

PHD

P/N

TBA

UNICEF

UNFPA

VDC

WHO

YWAM

Acquired Immune Deficiency Syndrome

Antenatal care

Cambodian Demographic and Health Survey

Caesarean Section

Department of International Development

Essential Obstetric Care

Emergency Obstetric Care

Family Spacing

Human Immune- deficiency Virus

Maternal Child Health

Maternal Mortality Ratio

The Ministry of Health

Medicin Sans Frontiere

National Centre for HIV/AIDS, Dermatology and STD's

Non government organization

Provincial Health Department

Post natal care

Traditional Birth Attendant

United Nations Children Fund

United Nations Population Fund

Village Development Committee

The World Health Organization

Youth With a Mission

Traditional Birth Attendant Programme Stung Treng/ Cambodia VI

Definition of terms

Maternal Mortality Ratio

Maternal death

Still birth

Neonatal death

Perinatal death

Infant mortality

Child mortality

Under five mortality

Traditional Birth Attendant

Primary midwife

Secondary midwife

The number of maternal deaths per100,000 live births

A death that occurs during pregnancy, childbirthor within two months after the birth ortermination of pregnancy.

A baby bom after 22 weeks gestation who showsno signs of life

A death of an infant that occurs within thefirst month of life.

A death occurring during late pregnancy(at 22 weeks gestation and over, during childbirthand up to seven completed days of Life.

An infant death between birth and the first yearof life

The death of an infant between the age of one yearand five years.

A death between birth and the fifth birthday

Is a person who assists the mother duringchildbirth and initially acquires her/his skills bydelivering babies herself/ himself or through anapprenticeship to other traditional birthattendants.

A government trained midwife who hascompleted one year of training.

A government trained midwife that hascompleted three years of training

Traditional Birth Attendant Programme Stung Treng, Cambodia vn

EXECUTIVE SUMMARY

YWAM programmes in Stung Treng

Youth With a Mission (YWAM) commenced working in Stung Treng in 1991.Their present activities include providing technical and financial support to the TBAprogramme in four of the five districts; a HIV/AIDS prevention programmespecially focused on youth using peer educators; an English teaching programmefor health staff and community members; a malaria prevention programmeproviding health education and the distribution of insecticide treated bednets, in thedistricts of Siem Pang and Stung Treng and a prison ministry programme.The Siem Pang Community development programme is a long term activity in 12remote villages. A smaller project supports the Kravet minority group non-formaleducational programme. YWAM also responds to emergency situations within theprovince. They currently have four expatriate staff and ten national staff working inStung Treng province.

The Stung Treng TBA Programme

The TBA programme began in 1993. An assessment conducted by YWAMconfirmed that TBA's were conducting the majority of deliveries at village level.The safe practice of these TBA's was considered to be a priority. At that time nodeliveries were conducted at health centre level and reporting of maternal childhealth statistics from village level was non existent. Following a Training ofTrainers' course for seven provincial level midwives a series of six day TBAtraining's were conducted in all five districts.

The programme's priority was on training as many TBA's as possible to performclean safe deliveries, detect complications and refer. Following training, TBA's weresupplied with a TBA kit and bag. During a four-year period more than 480 TBA'sreceived training. Following training regular three monthly TBA meetings wereheld, facilitated by the provincial trainers and attended by the health centremidwives. Technical and financial support was provided by YWAM. In the earlyyears of the programme YWAM supplied 100% of supplies and equipment for theTBA's, as MoH supplies became more plentiful and distribution systems improvedthe provincial health department assumed responsibility for supplying gentianviolet, compresses, cord ties, gloves and iron tablets. During three monthlyfollow-up sessions, TBA's are able to buy at a reduced price re-supplies such asforceps for clamping the cord, brushes, plastic sheets and scissors.This equipment isprovided by YWAM.

In 1997 MSF started a health project in Stung Treng and assumed responsibility forthe TBA program in Sesan district, hi 2000 MSF discontinued its work in StungTreng province.

Traditional Birth Attendant Programme Stung Treng, Cambodia viii

During the three years that MSF worked in Sesan no TEA meetings were held andthe TBA's didn't receive refresher training. Due to budget and staffing restraintsYWAM were unable to resume support for Sesan when MSF withdrew.

In 1998 follow-up support and training was discontinued for TBAs who wereconducting less than ten deliveries per year. At the end of 2002 there were 394trained and active TBA's included in the program in the five districts of StungTreng.

In 2000 YWAM employed two national staff to work as TEA monitors. Theirresponsibility was to follow up activities of TBAs at village level and interviewmothers who had been delivered by TBAs. Their monitoring activities werediscontinued at the end of 2002 when the TEA programme managers started followup at village level.

In 2001 UNICEF assumed responsibility for providing the budget to the PHD forfollow up activities of TBA's in Sesan, including TEA meetings, follow up visits atvillage level and refresher training. YWAM continues to provide money to SesanTBAs to cover transport costs when they refer women to the provincial hospital.

In 2001 the TEA programme was renamed ' The Bridge of Friendship'.The provincial health department assumed complete responsibility for managementof the programme in July 2001. With the integration of "The Bridge of Friendship'programme into the PHD activities, it was planned that the provincial MCH staffsrole would change from direct programme implementers to program managers.YWAM continues to provide some financial and technical support.

Ongoing activities of the TEA program are: follow up support and on the jobtraining both at meetings and during village visits; three monthly TEA reportreviews and re-supply of materials and equipment; two yearly refresher training forTBA's who have already been trained and six day TEA training courses for TBAs invillages where there is a shortage of trained birth attendants; re-imbursement oftransport costs for emergency referrals by TBAs and capacity building of the TEAprogramme managers through formal courses and on the job training.

YWAMS strategy is to gradually withdraw both financial and technical supportwith PHD eventually assuming complete responsibility for the 'Bridge ofFriendship'.

The purpose of the evaluation

The purpose of the evaluation was to provide an objective assessment of progressmade in meeting the objectives of the TEA programme, document lessons learnedand to make recommendations for the future direction of the programme.

ix Traditional Birth Attendant Programme Stung Treng, Cambodia

The overall objective of the TEA evaluation was to assess the impact the TBA'sprogramme has had on strengthening maternal and child health services in thetarget areas and improving the health of mothers and infants.

This was to be achieved by:

Assessing changes during the program period in the knowledge attitudesand practices of TBA's who participate in the program and identifying thestrengths and weaknesses of their work.

. Assessing the strengths and weaknesses of the various components of theprogramme including the training course, reporting system, managementsystem, monitoring and referral system.

Assessing whether project objectives were achieved at a reasonable cost.

Methodology

A literature review of relevant documents concerning Safe Motherhood and TBAprograms in Cambodia and in other countries was conducted. This was followed byseven days field work in Stung Treng province (see appendices of details of fieldwork) Study sites were selected independently by the evaluators and based on thegeography and location, distance from the nearest health centre/referral hospitaland time available for field work. The evaluators decided to spend quality time infewer villages rather than rushing to cover many villages and districts.Eight villages in three districts were visited with overnight stays in two villages.Districts not visited during the evaluation were Sesan, where YWAM has not beeninvolved since 1997 and Siem Pang, because of its remote location and time neededfor travel (See map with locations visited marked) Although these two districtswere not visited, interviews with key informants provided information concerningTBA's activities in these areas, and available MCH data from all districts wasreviewed and considered.

A triangulation of methods was used including review and analysis of relevant dataand several qualitative methods including TBA skills observations, semi -structuredinterviews with key stakeholders including PHD, UNICEF, YWAM staff, MCHtrainers/ health centre managers/ referral hospital and health centre mid wives;individual interviews with TBAs, families, mothers, community leaders and villagedevelopment committee members and TBA and mother focus discussion groupdiscussions. A total of 101 key stakeholders participated in the evaluation.

Constraints

There was no available baseline information concerning the knowledge skills andpractices of TBAs in the programme area. To address this the evaluators comparedthe knowledge attitudes and skills of Stung Treng TBAs with the results of severalother TBA baseline surveys conducted in other Cambodian provinces. Also duringTBA interviews and discussion groups the evaluators recorded TBA's self-assess-ment of changes following training.

Traditional Birth Attendant Programme Stung Treng, Cambodia x

Mothers interviewed were able to identify and describe, without prompting,differences between TBAs who had been trained and those who had not receivedtraining.

There was a lack of, and inconsistency of MCH data in Stung Treng. The evaluatorsfocused the analysis of data on the three years preceding the evaluation from2000-2002. More complete and accurate data was available for this time period.

The evaluation focused on three of the five districts in Stung Treng and involvedonly 21 (5.3%) out of the 394 trained TBAs currently active at village level. Thereforethe assessment of TEA skills in this report cannot be taken as representative of theskills of all the TBAs in the programme areas. A wide range of views andinformation was collected from other key informants.

Focus group discussions at village level were difficult to conduct in a quietenvironment. Such was the interest of village women that many more womenjoined the discussions than was originally planned. This proved useful in obtaininga wide range of views and also assessing the general health situation of villagewomen, but affected the quality and depth of information obtained.

Findings

The key question the evaluation sought to answer was:

" Has the TBA programme in Stung Treng contributed to strengthening of the MCHservices in the province and improving the health of mothers and infants?"

The evaluators found that the provincial health department, health centre staff,YWAM and more recently UNICEF, through a strong collaborative effort insupporting the TBA programme have achieved an significant improvement in thequality of MCH services available to women in remote villages of Stung Treng.

With difficult access and often impossible travel conditions, 394 TBAs (with medianage 45yrs- 55yrs) in villages of all five districts have been trained and are active.They are provided regular follow up support and supervision. Equipment andsupplies are re- supplied in a timely manner and information from village levelabout births, deaths and transfers are reported at regular intervals. The evaluationteam found an impressive level of knowledge and skills among TBAs observed andinterviewed, concerning how to conduct a clean and safe delivery and identifyhigh-risk women that need referral. TBAs reported several changes that had takenplace in TBA practices such as cleaning and sterilising of equipment, wearing ofgloves, putting of gentian violet on the cord, waiting for signs of placentalseparation before attempting delivery of the placentas and drying the baby andputting to the mothers breast instead of giving an immediate bath. Interviews anddiscussion with village women who had delivered with a TBA confirmed thesepractices.

xi Traditional Birth Attendant Programme Stung Treng, Cambodia

Statistics and follow up investigation show that neonatal tetanus is probably not aproblem in Stung Treng.This is an important indicator of success for the TBAprogram, that has very likely contributed to this achievement by increasingthe number of clean deliveries conducted by TBAs and achieving appropriate careof the cord. Tetanus toxoid vaccination rates remain too low to contribute towardsthis achievement (15% T2 in 2002).

Stillbirths and neonatal deaths remain high but there are some signs ofimprovement. As there is no record of the age at death of the neonate it isimpossible to estimate perinatal mortality rates. Neonatal mortality statistics show adecrease from 4.0% of total reported deliveries in 2000 to 3.7% in 2001 and 3.2% in2002. There is very little reporting or recording of infant or under five mortality soit's impossible to assess if the health of under five children has improved. Fromcomments during village level discussions it appears that in both groups deathsrates are high.

Immunisation coverage rates for under one year old children have improved in thelast three years (66% of children receiving DTC 3 and Polio 3 in 2002 as compare to53% in 2000) These improvement have occurred due to increased outreach activitiesby health centre staff. Antenatal care with government midwives has increaseddramatically in 2002 (from 28% in 2001 to 58% in 2002) and this is attributed toincreased outreach activities by health centre midwives and the participation ofTBAs during outreach.

The TBA's are to be commended for their ever increasing involvement intransferring high risk women. The number of transfers of emergency obstetric casesto the referral hospital is increasing yearly (49 in 2000, 57 in 2001 and 74 in 2002) Areview of the cases transferred in 2002 showed that TBAs competently andsometimes under very difficult conditions accompanied women to the hospital.TBAs often stayed with the woman several days until her condition stabilised. Fromthe review of transferred cases it is also clear that without the prompt action of theTBA concerned some of these women may have been added to the maternalmortality statistics.

Observed difference in maternal death statistics in a small population province suchas Stung Treng is not a good indicator by which to measure the success of an MCHprogramme. The maternal mortality statistics are too few in number and differencesobserved may be due to changes in the reporting system or may be subject to awide random variation resulting from a small number of events. There may be largechanges in the burden of morbidities before this is reflected in the MMR.

It is positive that maternal death audits have been conducted since 2002 and thatTBAs in Stung Treng are active in reporting of maternal deaths. In 2002 TBAsreported the majority of maternal deaths that were investigated. In a review of thecases that TBAs were involved in it appears that the TBAs responded appropriatelyaccording to the situation. It is Likely that all maternal deaths are reported.

The programme is judged as very cost effective at 30 US $ per TBA per year (2.50US$ per month per TBA), 32% less expensive compared to a TBA programmeimplemented in a more easily accessible province. Seventy four percent of the StungTreng TBA budget is spent on capacity building of both TBAs and government staff.

Traditional Birth Attendant Programme Stung Treng, Cambodia xii

Issues that need to be addressed:

1. Some inconsistency and lack of clarity in implementation of the TEAprogramme, created by having two agencies, with some differences inphilosophy providing financial and technical support to the programme fordifferent districts.

2. The need for a comprehensive provincial TBA policy.Although the Bridge ofFriendship TBA program has a clear TBA policy, there are some differencesbetween it and the UNICEF policy concerning TBAs in Sesan district.

3. Dependence on three provincial TBA managers/ trainers to implement themajority of TBA program activities.

4. A lack of confidence by village people in the quality of services offered at thereferral hospital.

5. Sole dependence on the village TBA concerning the emergency referralsystem and the dependence on YWAM to re-imburse travel expenses.

6. Some weakness in the TBA reporting system because of the prolongedinterval between conducting of activities to the recording and reporting ofthem.

7. The need for improvement of TBA skills concerning supporting theestablishment of exclusive breast feeding, identification and referral fortreatment of anaemia both antenatal and post natal, identification of high riskinfants and mother in the post natal period, and health education skills topromote antenatal care and tetanus vaccine, exclusive breast feeding and theintroduction of appropriate timely complementary feeds.

8. The existing system for documenting of TBA activities and statistics needs tobe strengthened and standardised.

9. Village women's lack of knowledge concerning prevention of commondiseases, family spacing methods, the benefits of antenatal care, importanceof iron supplementation, the benefits of exclusive breast feeding, theintroduction of appropriate and timely supplementary feeds and dangersigns during pregnancy, labour and in the postnatal period.

10. The reluctance of TBAs to promote or use the individual birth kits. The kitshave some important benefits including the active participation of women inensuring their delivery is as clean as possible, and guaranteeing that eachmother will have her own disposable gloves and plastic sheets.

11. The refusal of women in remote rural villages, who are identified as high riskduring their pregnancy to go to the referral centre.

xiii Traditional Birth Attendant Programme Stung Treng, Cambodia

RECOMMENDATIONS IN ORDER OFIMPORTANCE

Recommendations to PHD, UNICEF and YWAM

1. A decision should be made as soon as possible concerning the future of thepresently divided Stung Treng TEA programme. PHD in consultation withYWAM and UNICEF should decide which agency is in the best position tooffer long term (at least two years) continuing technical and financial supportto the TEA programme in all five districts. The evaluator's assessment is thatYWAM because of its history with the TEA programme in the province, itsexperienced staff, its present support to four of the five districts and its longterm commitment to the Stung Treng health programmes is in the bestposition to do this. If all parties involved are in agreement and YWAM iswilling and able to agree to this, they should re-assume responsibility for thesupport of TBAs in Sesan district. If funds are not immediately available todo this UNICEF should be requested to provide funds to YWAM as aninterim measure until further funds can be identified.

2. PHD should play the lead in organising and conducting a workshopinvolving key stakeholders in the TEA programme to develop acomprehensive TEA policy for the province, addressing issues such asmanagement structure at each level of the programme, a standard TEA jobdescription, relationships, role and responsibility of health centre and provincial level staff concerning the TEA programme, transfer of emergency cases,TEA perdiem levels and the resupply of TEA supplies and equipment.

3. An extra full time midwife should be assigned to work at provincial MCHlevel.The present MCH staffing level is too low to adequately cover theamount of activities currently being implemented.

Recommendation to TEA programme managers at provincial level.

4. With the support of the YWAM TEA programme technical advisor, and incollaboration with health centre staff a plan should be developed to ensurethe smooth and gradual decentralisation of TEA programme activities tohealth centre level. It is recommended that ultimately health centre staffshould be responsible for all TEA training, follow-up meetings andmonitoring of TEA activities at village level. Provincial level staff shouldincrease their involvement in on the job training of health centre staff andco-ordination and monitoring of MCH activities at health centre level.

Recommendation to PHD and UNICEF

5. PHD with the support of UNICEF and possibly an external consultant shouldconduct an assessment at the referral hospital, addressing the urgent need forimprovement of provision of quality emergency obstetric care. Funds shouldbe identified to follow through on subsequent recommendations.

Traditional Birth Attendant Programme Stung Treng, Cambodia xiv


6. The provincial TBA programme managers should work with health centrestaff, health centre management committees, village developmentcommittees, TBA's and other organisations and groups working at villagelevel to develop and strengthen the referral system at village level. The TBAspresently lack support concerning organising emergency transfers. Theyrequire strong backup support from village leaders or committees whenthe decision is made to make an emergency transfer. The development of arealistic achievable village emergency transfer fund is also important Thisemergency fund scheme should be started on a pilot basis in a few villageswhich have active motivated development committees. Following a review oflesson learned the programme could be adapted and introduced to othervillages.

7. The accuracy of the TBA reporting system would be improved by morefrequent collection of TBA statistics. Statistics may be more reliable if theywere collected monthly, as the TBAs would then be encouraged to recordtheir activities closer to the time they were conducted. The logistics involvedin doing this will vary according to the area and the human resourcesavailable. It might be possible that VDC members could collect monthlystatistics from TBAs in their respective villages and send them to the healthcentre level.

8. Future refresher training for TBAs should concentrate on improving the skillsof TBA in the areas identified during the evaluation. These are: thepromotion and support to mothers in the establishment of exclusive breastfeeding, identification and referral for treatment of anaemic women bothantenatally and post natally; the prompt identification of problems in infantsand mother in the post natal period; improved health education skills topromote antenatal care, tetanus vaccine, exclusive breast feeding and theintroduction of appropriate timely complementary feeding. If funding isavailable it would be helpful to develop and pre-test a TBA flip chart withcolourful pictures, key messages and minimal script, that TBAs could use asa tool when discussing health topics with village women.

9. The system of documentation and recording of TBA programme activitiesand statistics should be reviewed. If possible a user friendly centralisedaccessible data base should be designed that standardises how importantTBA programme indicators are recorded. The TBA programme provides a lotof important information. If the data is easy to access, it can help identify andrespond to problems promptly and facilitate easier monitoring of theprogramme.

10. There is a need for effective IEC materials and health education sessions forvillage people concerning important health topics and prevention commondiseases. The TBA cannot be expected to take complete responsibility forvillage health education, human resources at village level should beidentified that can participate in conducting these activities.

11. Despite the reluctance of TBAs to promote or use the individual home birthkit, the kit has some important benefits including the active participation ofwomen in planning for their delivery and ensuring it is as clean as possible.

xv Traditional Birth Attendant Programme Stung Treng, Cambodia

The kit also guarantees that each mother will have her own disposable glovesand plastic sheets. The kit should be continued to be promoted at villagelevel. Women might be more enthusiastic about buying it if they understoodbetter the advantages the kit offers. A flexible approach should be used withTBAs who have been proficiently and safely using scissors and clamps formany years. They should be encouraged to use as many of the disposableitems in the kits as possible.

12. The TBA programme managers should strongly advocate that the NationalMCH Centre review the adequacy of the materials supplied in the individualhome birth kit and make recommendations for adding additional materialsaccording to field experiences in Stung Treng.

Recommendation to PHD, UNICEF and YWAM

13. The possibility of 'waiting homes' at provincial level for high-risk pregnantwomen should be explored. They have found to be successful in severalother countries, especially in countries that developed simple realistic planswith active involvement of communities (WHO, 2002). Stung Treng transferstatistics show that very few high-risk women are electively transferredduring pregnancy. Most are transferred when they are already an obstetricemergency. This report has mentioned some of the reasons for this situation.

General conclusions

The TBA programme in Stung Treng has achieved commendable success inincreasing women's1 access to clean deliveries, strengthening the skills of traditionalbirth attendants and decreasing the use of harmful practices. The main reason forthe success is the close follow-up support and continuing training opportunities thathave been provided to TBAs. Considering the geography of Stung Treng this is nosmall achievement.

Despite this success much remains to be done to improve women and children'shealth. The MoH' strengthening of district health services' is progressing.The quality of health care available in rural areas is slowly improving, but it isunrealistic to expect that the MoH objective to provide at least one competentfemale health provider trained in Basic EmOC at each health centre in rural areas,and to increase the provision of quality essential maternity care at this level will beachieved within the next few years. Until this happens and women have increasedaccess to and confidence in the available health services, the TBAs will continue tofill the gap, providing an important and affordable service to rural women.

Provincial health department staff have confidently taken over the management ofthe TBA programme. They strongly believe in the value of the TBA and theimportance of her role in providing a link between the community and governmenthealth services. It is important to continue to build and strengthen the relationshipbetween the TBAs and the health centres. Due to budget restraints and lack ofhuman resources PHD will require future financial and technical support for theTBA programme.

Traditional Birth Attendant Programme Stung Treng, Cambodia xvi

1. INTRODUCTION

1.1 The Maternal Child Health Situation in Cambodia

The current population of Cambodia is approximately 11 million with an annualgrowth rate of 2.5%. Eighty four percent of the population lives in rural areas.Cambodia remains one of the poorest countries in South -East Asia with the grossdomestic product per capita estimated at approximately $ 238 in 2000. Almost 40%of Cambodia's population living below the poverty line of 0.50 US$ cents per day.The average life expectancy is 54.4 yr. for men and 56 yr. for women (World Bank,2001). Common diseases are diarrhoea, acute respiratory infection, denguehaemorrhagic fever, malaria, malnutrition, TB and other vaccine preventablediseases (CDHS, 2000). HIV/AIDS also is a major problem in Cambodia with 2.8%of the population HIV positive and with up to 18,000 people dying of AIDS in 2001.It is estimated that approximately 3% of pregnant women are HIV+(NCHAD, 2001).

Research conducted in 2002 to rate access to maternal and neonatal health servicesin forty nine developing countries rated Cambodia in the 'very weak' category witha score of 33 out of 100 (WHO Bulletin, 2002). Maternal and child mortality arehigh with a maternal mortality ratio: 473 per 100,000 live births (4 deaths per 1,000live births). The main causes of maternal mortality are abortion complications,pre-eclampsia and haemorrhage. The Cambodian Demographic and Health Survey2000 found the main problem encountered in accessing health care was not havingenough money (CHDS, 2000). Two in five women reported not knowing where togo for health care. Nationally fewer than 38% of women receive A/N care fromtrained personnel. In rural areas 59% of women receive no antenatal care. Tetanustoxoid coverage (received two tetanus toxoid injections) is 30%. Eighty nine per centof births are at home, In rural areas 70% are attended by TBAs. Neo-natal mortalityis 37 per 1,000 live births, infant mortality is 89 per 1,000 live births and under fivemortality is 124 per 1,000 live births.

Although 96% of mother's breast feed their infants only 18% of infants less than twomonths old are exclusively breast fed. By age four to five months only 5% of infantsare exclusively breast fed (CDHS, 2000). Chronic malnutrition amongst under fivechildren in Cambodia is high with 45% of children moderately stunted and 21%severely stunted. Sixty three per cent of children 6-59mths are anaemic and 58% ofCambodian women are anaemic. Immunisations rates in under one year oldchildren are BCG 71%, measles 55% and DPT 3 49% with a high drop out rate.

The low status of women in Cambodia is recognised as a major barrier to improvingwoman and child health. Literacy levels are low with 42% of women literate andonly 13% of rural women studying to secondary level. (CDHS, 2000). Violenceagainst women is high with 16% of women reporting physical violence by theirpartners within the previous 12mths (CDHS, 2000).


A Cambodian birth spacing policy was formulated in 1994 and a birth spacingprogramme was introduced in 1995. A National Safe Motherhood Policy andStrategy document was developed in 1997. The goal of the second five year plan2001-2005 is: to improve the quality of Life of Cambodian women and children byimproving pregnancy outcomes and reducing maternal and infant morbidity andmortality. Some of the objectives of the plan are to improve standards andguidelines to ensure quality and consistency in implementation of safe motherhoodactivities; to provide at least one competent female health provider trained in BasicEmOC at each health centre; to increase the provision of quality essential maternitycare, particularly essential obstetric care at health centre level and to increase accessto safe motherhood services, especially emergency obstetric care by educatingwomen, families and communities about pregnancy, childbirth and birth spacing.

Progress has been made in implementing the Safe Motherhood Plan especially theintroduction of standard guidelines and upgrading the skills of health centre staffproviding MCH services. Despite this the reality is that only a small percentage ofhealth centres have competent female health providers. Utilisation of governmenthealth services remains extremely low, for example utilisation of governmentoutpatients services in Stung Treng for 2002 was 0.38 per person per year.(Stung Treng Provincial Health Department, 2002)

1.2 History of TBA's in other countries and in Cambodia

Millions of women in developing countries do not have access to modern healthcare services or trained staff. Sixty to eighty per cent of births in developingcountries occur outside health facilities. The number and distribution ofprofessional midwives are often inadequate. Many countries have decided to traintraditional birth attendant because these practitioners have already been chosen bymothers and have respect and authority in the community. Because of the currentshortage of professional midwives and institutional facilities to provide prenatalcare and clean safe deliveries several UN agencies involved in reproductive healthprograms also promote the training of TBA's. This is seen as a temporary strategy tobridge the gap until all women and children have access to acceptable, professionaland modern health care services. (WHO, UNICEF, UNFPA Joint Statement, 1992)

" TBA's are recognised as wise intelligent women who have been chosen by thewomen in their village for their practical approach or experience. Though oftenilliterate, they speak the language and are part of the religious and cultural systemof the community they serve. TBA's are private practitioners who negotiate theirown compensation with their clients. Compensation usually includes favouredstatus in the community" (WHO, 1992).

Despite the acceptance that TBAs are still necessary in developing countries, therecontinues to be controversy and debate surrounding the role of training them.Opponents of TBA training consider that training TBA's reveals a lack ofcommitment to providing quality maternity care.


It is also argued that scarce resources are used for TEA training rather thanprioritising training and deployment of qualified midwives and providing qualityaccessible health care services. One major question raised concerning TEA trainingis whether TBAs' practices change after training.

The Safe Motherhood Policy Cambodia (MOH 1997) states that;" While the MOHrecognises that the TBA is the maternity care provider mostly used in the village (insome remote areas TBA's conduct up to 90% of deliveries) the MOH commits itselfto train and post professional midwives at health centre level. This professionalmidwife will be the lynchpin in the maternity services between the community andthe referral hospital. The goal is to place trained midwives in all areas of Cambodiabut this will not be able to become a reality for many more years."

The Cambodian MoH policy describes the TBA as a traditional practitioner who isnot officially integrated into the Cambodian health service workforce, and there isno plan to do so. As the TBA is not a member of the health service, she is notremunerated for the work that she does, therefore it is important that TBAs arerespected for the work they do especially assisting women doing childbirth andacting as a link between the health service and the village. In the National SafeMotherhood Plan 2001-2005 training of TBAs is a key objective to strengthen andextend service delivery at community level.

1.3 Geography and demographics of Stung Treng Province

Stung Treng province covers an area of 11,092 square kms and has a population of80,217. It has five administrative districts with thirty-four communes and onehundred and twenty eight villages. Situated in the Northeast of Cambodian and 481kms from Phnom Penh, it is one of Cambodia's remotest and poorest provinces witha population density of eight people to one square Km. Stung Treng is bordered bythe provinces of Ratanakiri to the east province, Kratie to the south, KompongThorn and Preah Vihear to the west and the country of Laos in the north.The majority of the population is Khmer or Lao with smaller groups of fourteenethnic minorities. There are four major rivers in the province, the Sekong, Mekong,Srair Pok and Sesan. Access to Stung Treng is difficult, there are three main roadsin the province, "Road 71 which runs from Kratie province through Stung Treng andonto the Laos border, 'Road 78' which runs from Stung Treng to Ratanakiri provinceand Road 214 which runs from Thalaborivat district in Stung Treng to Preah Vihearprovince. These roads are in very poor condition and impassable in the rainyseason, when there is often extensive flooding. The majority of people live along theriver and boats are the main mode of transport within and out of the province.Travel on the river is expensive and can be hazardous, especially during the dryseason when water levels are low and boats have to manoeuvre numerous juttingrocks, making it almost impossible to travel by night. There are commercial flightsfrom Phnom Penh three times per week but the cost of air travel puts it beyond thereach of the majority of the population. The main livelihood of the population is ricefarming, vegetable gardening, fishing, wood cutting for charcoal production and asmall number of handicraft enterprises and small businesses.


1.4 Current health situation in Stung Treng

Stung Treng has been implementing the 'MoH Reform Coverage Plan' since 1995.Stung Treng province has one operational district with a referral hospital situated inStung Treng town that provides a ' Complementary Package of Activities'. There areeight health centres providing a 'Minimum Package of Activities', with a plan tobuild a further two health centres when funds and staff are available. Nationallyonly 30 % of the new health coverage plan has been implemented, and it's difficultto assess how successful it is at this stage. Utilisation of public health services inCambodia is extremely low with only 0.3 visits per person per year. Rural healthexpenditure varies from 11% to 28% of yearly income. Per capita health expenditureis approximately 20US$ with the household contributing most of this. (DFID, 2001).In low-density provinces like Stung Treng providing adequate health servicecoverage is difficult and challenging. In 2002 only 28% of health centre staff inStung Treng had received MPA training. The provincial health departmentidentifies one of its main problems as its inability to maintain adequate staffinglevels both at the referral hospital and health centres. Staff coming from distantprovinces and other areas finds it difficult to live in remote areas with fewincentives for their work. The number of health staff in 1998 was 260. In 2002 withan increasing workload and scaling up of outreach activities the number of staff isreduced to 225. (conversation with PHD chief and vice chief 2003). Anotherimportant problem identified by the provincial health department is health servicefinancing. Up until the end of 2002 only four HC have formed management andfeedback committees. A large percentage of the rural population live below thepoverty line and find it difficult to pay for HC services. Health centres are short ofqualified staff including midwives and only four of the present eight health centreshave midwives with experience conducting deliveries. The health coverage planallows for small health posts in remote villages (more than 15kms from the nearesthealth centre and with a population of 2-3,000,but Stung Treng is unable to providestaff to run health posts. Stung Treng has numerous remote villages and travel isexpensive and hindered by rivers mountains and poor roads. Large numbers of therural population are mobile especially during the planting season when they moveto live temporarily in small huts near their fields. An increased number of MCHoutreach activities in 2002, have dramatically boosted the number of A/N careexams, tetanus toxiod vaccine and childhood vaccinations. Malaria is endemic inthis province with a incidence of 23% of the population, hi 2002 there were 3,209cases of malaria reported. Other common health problems are TB and dengue fever.In 2000 schistosomiasis had an incidence of 8.58% which through aggressiveidentification and treatment programme activities reduced to 1.21% in 2002.

UN agencies and health related NGO's working in Stung Treng are: YWAM(activities described in this report) UNICEF has worked with the provincial healthdepartment since 2000 and provides a provincial health advisor who supports PHDto implement the ' Strengthening of District Health Services' with a priority oncapacity building of PHD, referral hospital and health centre staff.


UNICEF also support immunisation activities, community participation in healthservice management, the Safe Motherhood Program, maternal death audits andHIV/AIDS programme support. The UNICEF supported 'Seth Kuma1 integratedrural development program is active in several villages in three districts.Other NGO's working in Stung Treng are 'Partners for Development' who supportthe work of malaria prevention and distribution of impregnated mosquito nets inthe commune of Preah Romkel in Thalborivat district and HIV/AIDS preventionactivities in Sesan district. Community Action Abroad supports HIV/AIDS andbirth spacing health education in Srair Krasang commune, Siem Bok district. SWDCa local NGO provides support to poor chronic patients in Stung Treng referralhospital and health education for in-patients. Mekong Eye Doctors provide training,supplies, equipment and a new building for the provincial eye department withinthe referral hospital complex. Pharmaciens San Frontieres has just commenced aone-year training programme working with private drug sellers

1.5 History of YWAM in Stung Treng Province

YWAM commenced work in Stung Treng in 1991. Initially they were involved in avariety of programmes including the strengthening of health centre services in fivedistricts of Stung Treng, Siem Bok, Thalaborivat, Siem Pang and Sesan districts.Technical and financial support included training staff, repair of buildings, buildingof staff accommodation and three health centres, supply of equipment and drugsandproviding the budget for outreach activities. Input to health centre level wasgradually phased down and the provincial health department assumedresponsibility for this at the end of 2002. Other YWAM programmes in Stung Trengare a HIV/AIDS prevention programme specially focused on youth using peereducators, an English teaching programme for health staff and communitymembers; a malaria prevention programme providing health education and thedistribution of insecticide treated bednets, in the districts of Siem Pang and StungTreng; a prison ministry programme provides basic toiletries and health education,literacy skills, leisure activities and spiritual support. The Siem Pang Communitydevelopment programme is a major activity in 12 villages of Siem Pang district andinvolves community health development. A smaller but important activity is pro-viding technical and financial support to the Kravet minority group non-formaleducational programme. YWAM also responds to emergency situations within theprovince providing seeds for planting following floods and interventions for chil-dren at risk. YWAM currently have five expatriate staff and ten national staff work-ing in Stung Treng province.

1.6 The Stung Treng TEA Programme

The Aim and Objectives of the Stung Treng TEA programme

Aim: That by strengthening the Stung Treng rural maternity services withTraditional Birth Attendants (TBA's) the health of mothers and children will beimproved.


Objectives

1. To promote safe motherhood and child survival through training TBAs andprovide regular updating of previously trained TBAs. The updatedknowledge and skills of the TBAs will enable them to be safe practitioners.

2. To see a decrease in infant mortality from 87 per 1000 live births to 60 per1000 live births. There will be a decrease by 20% in maternal death rate from473 per 100,000 births (government projected figures)

3. To train new TBAs in villages where there is insufficient trained TBAs4. To provide refresher/update training every two years for TBAs who have

already been trained5. To make the programme sustainable in the long run6. In co-operation with the VDC's establish health insurance schemes for

patient referrals.

History of the TBA programme in Stung Treng

A village health assessment conducted by YWAM staff in 1993 confirmed thatTBA's were conducting the majority of deliveries at village level. YWAM consideredthe safe practice of these midwives to be a priority. At that time no deliveries wereconducted at health centre level and reporting of maternal child health statisticsfrom village level was non existent. At the beginning of the programme TBA'sfound to be conducting five or more deliveries a year were selected to attendtraining to upgrade their skills. There were some exceptions in villages wherevillagers selected TBA's to attend training.

YWAM conducted a Training of Trainers course' for seven midwives from theprovincial Maternal Child Health Department. Later the trainers attended anadditional TOT at national level. Three of the original trainers are still working withthis programme, six-day TBA training's then commenced and were conducted asclose to the TBA's village as possible, sometimes at the health centre. Health centremidwives and YWAM national staff also participated in the training and wereresponsible for teaching some of the topics.

During the first three years of the project the priority was on training as manyTBA's as possible to perform clean safe deliveries, detect complications early andrefer. Following training, TBA's were supplied with a TBA kit. Later a special TBAbag was given at a presentation ceremony conducted at village level to inform tothe community about TBA's who had received training and what services theycould provide. During a four-year period more than 480 TBA's received training.Following training regular three monthly TBA meetings were held facilitated by theprovincial trainers,attended by the health centre midwife and with technical and financial supportfrom YWAM. At the TBA meetings, TBA's report on the activities of the last threemonths and submit a completed report form (see picture of report form in thisreport) Problems encountered are discussed, training points are reviewed andsupplies refilled.


Up until 2001 YWAM was supplying 100% of supplies and equipment for theTBA's, but as MoH supplies became more plentiful and distribution systemsimproved the provincial health department has assumed responsibility forsupplying gentian violet, compresses, cord ties, gloves and iron tablets. YWAM stillre-supplies forceps for clamping the cord, brushes, plastic and scissors as needed.The TEA contributes towards the cost of any equipment she requires(see picture of TEA kit).

In 1997 MSF started a health project in Stung Treng and the two organisationsagreed that YWAM would discontinue supporting Sesan TBA's. This responsibilitywould be assumed by MSF. In 2000 MSF discontinued its work in Stung Treng.During the three years that MSF worked in Sesan no TBA meetings were held andthe TBA's didn't receive refresher training.

In 1998 the TBA program staff reviewed the activities of the trained TBA's and adecision was made to continue follow-up support and training for only for thoseTBA's who were conducting more than ten deliveries per year. At the end of 2002there were 394 active trained TBA's included in the program in the five districts ofStung Treng.

Table 1: No of trained TBAs active in Stung Treng 2002

District

Sesan

Siem Bok

Thala

Siem Pang

Stung Treng

Total

No of TBAs

68

56

133

93

44

394

In 2000 YWAM employed two national staff to work as TBA monitors. Theirresponsibility was to follow up the activities of TBAs at village level and interviewmothers who had been delivered by TBAs, in order to monitor TBA practices. Oneof the monitors was a former TBA the other was a younger woman without medicaltraining but with literacy skills to complement the skills of her TBA partner. Themonitors worked for two years. Their activities were discontinued at the end of2002 when the TBA programme managers started follow up at village level.


Due to budget and staffingrestraints YWAM was unable to resume support forSesan when MSF withdrew. In 2001 UNICEF assumed responsibility for providingthe budget to the PHD for follow up activities of TBA's in Sesan, including TEAmeetings, follow up visits at village level and refresher training. YWAM continuesto provide money to Sesan TBAs to cover transport costs when they refer women tothe provincial hospital.

In 2001, YWAM and the PHD TBA programme managers recognised that a strongnetwork of trained TBA's had been developed. To ensure the TBA programmewould be identified as an important component of the PHD maternal child healthprogramme, and to promote its sustainability a decision was made to rename theprogramme' The Bridge of Friendship1. The provincial health department assumedcomplete responsibility for management of this programme in July 2001. With theintegration of 'The Bridge of Friendship1 programme, it was planned that the MCHprovincial staff role would change from one of direct programme implementers toprogram managers. YWAM continues to provide some financial and technicalsupport and planned to slowly withdraw with PHD assuming completeresponsibility for the 'Bridge of Friendship1 by the end of 2002.

Ongoing activities of the TBA program

(a) Follow up support and on the job trainingIn early 2002 the regular three monthly TBA meetings were changed to six monthly.Between these six monthly meeting TBAs are visited at village level. It was foundthat follow up at village level was more effective in providing on the job trainingaccording to the needs identified. This new strategy meant that TBAs still haveregular three monthly follow-up support, alternating between group meetings andfieldwork. TBA's arere- supplied with equipment and medical supplies on a three monthly basis.Three monthly reports are submitted and discussed at TBA follow up. Equipmentneeded to be replaced is provided by YWAM and is sold to TBA's at a lower thancost price (See appendix for cost of re-supplies) Medical supplies such as gloves,gentian violet, cord ties, compresses and iron and folic acid are supplied by thehealth centre. Health centre midwives attend the TBA meetings and have theopportunity to discuss problems or review important topics with TBAs asnecessary. Attendance at the one day TBA meeting is reported to be very good withover 90% of TBAs regularly attending. YWAM supports the travel costs of the TBAsand a perdiem of 7,000 Riel per person (1.75 US$) for lunch.

(b) Refresher training for TBA's who have already been trained.A three-day refresher training is held two yearly in each district and TBAs aretaught the same curriculum as their initial training. Topics are updated as the needarises. The training's are held in or as close as possible to the TBA's villages.

In 2001 all literate TBAs in the three districts of Siem Bok, Thala and Stung Trengreceived a special five-day training about birth spacing methods.


(c) Training for already practising TBAs who have not yet received training invillages that are short of TB As. In villages that are short of trained TB As and wherethere are already practising TB As, six day TBA training course are held for selectedTBAs.

Table 2 Training Courses and Refresher Training

Year

2000

2001

2002

Totals

No TBA

New Training

Thala

Siem Bok

Stung Treng

Siem Pang

Thala

Siem Bok

Stung Treng

Siem Pang

Thala

Siem Bok

Stung Treng

Siem Pang

18000

00

90

000

0

27

No TBARefresher training

Thala

Siem Bok

Stung Treng

Siem Pang

Thala

Siem Bok

Stung Treng

Siem Pang

Thala

Siem Bok

Stung Treng

Siem Pang

8294290

5

0

0

71

124

47

0

0

452

(d) Support of transport costs for referrals by TBAs to referral hospitalYWAM reimburses the transport costs of TBAs referring high-risk women to thereferral hospital. A system is in place where TBAs who transfer to the referralhospital sign a transfer book and can collect money from the YWAM provincialoffice. A small budget is also available to support poor high-risk women duringtheir stay in the hospital and transport costs to return home. The TBA receives a giftof a sarong, soap and candle for each transfer she makes.

(e) Upgrading the skills of the TBA programme managersA YWAM midwife provides ongoing technical support to the TBA programmemanagers. YWAM supported the cost of the TBA programme manager to attend amanagement course in Phnom Penh during 2002.


H

W«^*l«*

Map of Stung Treng Province ,*~\

i '-'' ^^ 7"Hf

SIEM

>-"i/

V

/'PANG \

/'/

ys

51>2>*

52

KRATIEEvaluation sites

2 METHODOLOGY OF THE EVALUATION

2.1 Purpose of the evaluation:

To provide an objective assessment of progress made in meeting the objectives ofthe TEA programme, document lessons learned and to make recommendations forthe future direction of the programme

2.2 Evaluation team members:

Dr Chin Lan (MD, MPH) Infant and Young Child Feeding Programme Co-ordinatorNational Maternal Child Health Centre, Phnom Penh. Presently co-ordinatmg theimplementation of the Baby Friendly Hospital Initiative in several provinces ofCambodia. In the process of planning for a 'Baby Friendly Community Initiative' inrural areas of Cambodia. A member of the National IMCI Committee.

10 Traditional Birth Attendant Programme Stung Treng, Cambodia

Mary Dunbar (Nurse/Midwife, MSc Social Development, Planning andManagement). Reproductive Health Consultant with twenty years experience ofworking with Cambodian MCH programmes including implementation of TBAtraining programmes in three provinces of Cambodia.

2.3 Terms of Reference

Overall objective of the TBA evaluationAssess the impact the TBA's programme has had on strengthening maternal andchild health services in the target areas and improving the health of mothers andinfants

This will be achieved by:

Specific objectives:

1. Assess changes during the program period in the knowledge attitudes andpractices of TBA's who participate in the program. Identify strengths andweaknesses of their work

2. Assess strengths and weaknesses of the following components of theprogramme:

programme design /sustainabihty/relevancy to MOH plans and objectivesfor MoHmanagement structuretraining (trainers / TBA curriculum/follow up training

. monitoring/ follow-up support/ Smthly meetingsreporting systems

. referral system

. supplies and equipment/re-supply systemnetworking/ collaboration with other agencies/participation/collaboration of stakeholders

3. Assess whether project objectives are achieved at a reasonable cost. Wherecould activities been more cost effective? Are resources used in an efficientmanner, which minimises project costs?

4. Make recommendations concerning any change in strategy.

5. Make recommendations as to mechanism for transferring programme tocomplete PHD responsibility.

6. Produce a report for funding partners and MOH to assess the effectiveness ofthe programme and its replicability in other provinces.

Traditional Birth Attendant Programme Stung Treng, Cambodia 11

2.4 Methodology of the evaluation

A triangulation of methods was used which included a literature review, ananalysis of relevant MCH data and several qualitative methods. Information wasgathered during skill observation visits with TB As to the homes of pregnant andnewly delivered women;semi -structured interviews with key stakeholders including PHD, UNICEF, YWAMstaff, MCH trainers/ health centre managers/ referral hospital and health centremidwives; individual interviews with TBAs, families, mothers, village/communityleaders and village development committee members, and TEA and mother focusdiscussion groups. A total of 101 key stakeholders participated in the evaluation(see table 3).

2.5 Steps in the evaluation

1. A review of literature concerning the Cambodian Safe Motherhood program andpolicies; the present activities of MCH programs in Cambodia and documentsrelating to the present health situation and health programmes in Stung Treng.

2. Seven days field work in Stung Treng province (see appendices of details offield work) Study sites were selected independently by the evaluators and basedon the geography and location, distance from the nearest health centre/referralhospital and time available for field work. One inland location Jamgar Ler inThalaborivat district was selected for its remoteness and because it is onlyaccessible by road. Villages in Siem Bok were selected because of their access byriver and also their remote distance from the referral hospital. Stung Trengvillages were selected to provide a perspective of the situation for pregnantwomen and TBA's who live in more accessible locations. The evaluators decidedto spend quality time in fewer villages rather than rushing to cover manyvillages and districts. Eight villages in the three districts were visited, threevillages in Thalaborivat district, three villages in Siem Bok district and twovillages in Stung Treng district. Two days and one night were spent in JamgarLeu commune, Thalaborivat district and two days and one night in villages inSiem Bok district. Districts not visited during the evaluation were Sesan whereYWAM has not been involved since 1997 and Seam Pang, because of its remotelocation and time needed for travel (See map with locations visited marked)Although these two districts were not visited, interviews with key informantsprovided information concerning TBA's activities in these areas, and availableMCH data from all districts was reviewed and considered.

2.6 Constraints encountered during the evaluation

There was no baseline information available concerning the knowledge skills andpractices of TBAs in the programme area. To address this the evaluators comparedthe knowledge attitudes and skills of Stung Treng TBAs with the results of TBAbaseline surveys conducted in several other Cambodian provinces (MOH, 2001;RACHA, 2001; Parco and Jacobs, 2000). Also during TBA interviews and discussiongroups the evaluators were able to elicit the TBAs former knowledge and practicesand record TBA's self-assessment of changes following training. Mothersinterviewed were able to identify and describe without prompting, differencesbetween TBAs who had been trained and those who had not received training.


There was a lack of and inconsistency of available MCH data in Stung Treng.The HIS has improved since 2000. Since 2002 the PHD with the support of UNICEFhave been conducting maternal death audits, which provide more completeinformation of events leading up to maternal deaths. The evaluators focused theevaluation on the TBA programme during the three years preceding the evaluationfrom 2000-2002 where more complete and accurate data is available.

The evaluation focused on three of the five districts in Stung Treng and involvedonly 21 (5.3%) out of the 394 trained TBAs currently active at village level. Thereforethe assessment of TEA skills in this report cannot be taken as representative of theskills of all TBAs in the programme areas. A wide range of views and informationwas collected from other key informants such as village women to complement theinformation obtained from TBAs. It was not possible to observe a deliveryconducted by a TEA. Assessment of TBA delivery skills was based upon verbalcommunication with the TBAs and feed back from mothers and community,members at village level. TBAs were selected for skill observations by the evaluatorsfollowing individual interviews with TBAs.

Focus group discussions at village level were difficult to conduct in a quietenvironment. Such was the interest of village women that many more womenjoined the discussions than was originally planned. This proved useful in obtaininga wide range of views and also assessing the general health situation of villagewomen, but affected the quality and depth of information obtained.

Table 3. Information was obtained from the following key informants

No of interviews

OneOneOneOneTwoOneThree focus groupSevenFive skills observation

ThreeTwo group discussionsNineteenTwoTwoOne interviewTOTALS

Group

Provincial Health DeptUNICEFYWAMTBA monitorsMCH trainers/ managersMidwives referral hospTBAsIndividual TBAsIndividual TBAs

Health centre midwivesVillage womenIndividual mothersVDCVillage leadersFamily of maternal death

Total number involved

3 people1 person1 person2 people3 people5 people14 TBAs7 TBAs5 TBAs MB. (samemidwives as above)8 midwives30 women19 women5 people2 people2 people101 people


Travelling in Stung Treng

*%Pt *•. : ?;••**: ' :^%


Gathering information during the evaluation


Village women at work


3. Findings concerning the specific objectives ofthe evaluation:

Assess the capacity of the TBA's in the programme areas to deliver ante-natal care,safe delivery post natal care, health education and identify high-risk women andrefer. Identify strengths and weaknesses of their work.

3.1 Antenatal care : >

Ante- natal coverage in Stung Treng is very low but in 2002 has improveddramatically for first A/N visits. This is thought to be due to health centre staffincreasing village outreach activities in 2002 and TBA's participating in the outreachactivities by calling pregnant women to attend. There is still a low return index. TT2 has also improved but remains low (15% in 2002).

Chart 1 Stung Treng Province Antenatal 2000-2002

Year

200020012002

Total no. of pregnant women

2,8873,3043,027

1st visit

721936

1,654

%

25%28%55%

2nd visit

350391887

%

12%12%29%

Return

2.501.561.77

2000

| 1500

! 1000

500

0

56%

28%25%

29%

12% 12%

200020012002

1st AN visit 2nd AN visit

The MOH policy for TBAs describes the TBA role during the antenatal period asencouraging the pregnant woman to attend A/N care and receive at least two dosesof tetanus vaccine. Other tasks identified as TBAs responsibility during theantenatal period are to recognise anything abnormal and refer to the nearest healthfacility and to give advice and information about the importance of good nutrition.The Stung Treng job description is a slightly expanded version of this and describesthe TBAs role in antenatal care as: give antenatal care at least two times duringpregnancy by making a physical exam of the abdomen, checking for anaemia,giving ferrous folic 30 tablets per month, checking for oedema and asking otherrelevant questions concerning the mothers health, give advice about diet duringpregnancy, detect any problems/risks and refer as soon as possible.


The ability of the TBAs to deliver ante-natal care according to the Stung Treng jobdescription was assessed by observation and assessment using a skills checklist ofTBAs conducting ante-natal care in villages and individual TEA interviewsconcerning knowledge, attitudes and practices related to ante-natal care. Nineteenmothers with infants less than one year old were also interviewed about theirantenatal care experience during pregnancy. Focus group discussions with TBAsand village women also provided additional information.

Findings:

TBA monitors employed for a two year period by YWAM who visited 129 village ofStung Treng over an 18mth period and interviewed 182 mothers who had deliveredtheir baby with a TBA, reported that 73% of women interviewed said that A/N carewas given by a TBA. The quality of antenatal care given is not assessed or reported,and the number of AN visits per woman is not specified. Fifty five percent ofwomen interviewed by the TBA monitors said they received iron tablets from aTBA, but it is not reported if the woman knew the benefits of the iron or if they tookit regularly.

The TBAs interviewed during the evaluation stated their role in ante-natal care wasto check the position of the baby and to give information about nutrition to themother. If women do not request a visit the TBA will not initiate one. Womenusually only called them if they felt there was a problem. Some women especiallyprimigravida called the TBA to confirm early pregnancy. TBA's said they did thisby feeling for the enlarged uterus. TBAs said women only feel they need antenatalcare if they have a problem. All TBA's interviewed could name at least four dangersigns during pregnancy (oedema, anaemia, bleeding and abnormal position of thebaby after seven months of pregnancy). If they identified a problem they stated theywould tell the women to go to the nearest health facility. TBAs said women rarelyfollowed this advice and in the discussion groups TBAs mentioned the problem ofat risk pregnant women not following their advice as causing them the greatestfrustration.

TBAs reported distributing iron tablets only one time (30 tablets), which would notbe enough to improve the haemoglobin of a pregnant woman. It is recommendedthat each woman should take 90 iron/folic acid tablets during her pregnancy.All the TBAs knew that taking iron could help to prevent anaemia duringpregnancy, but were unclear how long a woman should take the iron for and howmany times per day.

During observation of TBA ante-natal skills at village level the TBA's observed wereable to make an accurate abdominal palpation and correctly identify the position ofthe baby. All TBAs checked for oedema of the legs and hands and gave adviceabout appropriate food to eat to stay healthy. TBAs also gave women advice aboutnot excluding certain foods, which some women believe are harmful or may causethem to have a large baby. If the woman was close to delivery the TBA advisedabout preparing clean washed materials for the baby after delivery and also aboutpreparing the place of delivery by cleaning a mat and putting it and the sparepieces of cloth in the sun to dry.


Only one of the TBAs was observed to check the eyes and hands for signs ofanaemia and TBAs observed did not give advice about tetanus immunisation or theimportance of taking iron tablets.

Pregnant and newly delivered mothers interviewed showed little understanding ofthe benefits or importance of antenatal care, even though several of them mentionedthat the TEA had advised them to go to the health centre. All the womeninterviewed had received at least one tetanus injection, usually during outreachactivities but didn't know the benefit of this vaccination or how many times theyshould receive it. When asked where they would go if they wanted advice duringtheir pregnancy all of them mentioned the TBA. The unanimous reason given forthis choice was that the TEA was close to the woman's home and that they trustedthe TEA. Pregnant women who had a white antenatal card didn't know what thiscard meant or what benefit it might be.

Case Study One, Jamgar Ler Village/Thalaborivat district

Sarom the village TEA happily guided us through her village, which was situatedclose to the health centre. She wanted to show us the village tradition concerningchildbirth. Families in this village build a special delivery hut of bamboo and thatch.Woman don't deliver in the main family house, they believe that child birth is a messybusiness with lots of blood. They also want to make it easy to follow the custom ofroasting for three days following delivery. We arrived at the small bamboo hut on stilts.A large stack of firewood was already neatly chopped and ready under the house.Inside the hut there was a small room with a kerosene lamp, candles, a new rush mat, amosquito net and a pillow. The expectant mother eighteen year old Sophy rushed outof her mothers nearby house to greet us. Sophy a primigravida was extremely thin andanaemic. When asked if she has received ante-natal care she produced her white antenatalcard from the delivery hut Sophy was nine months pregnant, 142cms in height andweighed 34kgs. She had attended the health centre twice and there was no mention onthe card of her anaemia or any note that identified her as high risk. Her nineteen year oldhusband joined us and proudly explained how he built the delivery house. Sophy hadnot taken the iron tablets given her by the health centre. She had not liked the taste ofthem and stopped after the first two tablets. Sarom the TEA explained that Sophy's familyis very poor, her husband doesn't work and they don't eat well. Sophy, her husband andSophys mother (who is also anaemic) sit down as Sarom explains to them that it wouldbe safer if Sophy delivered her baby at the referral hospital. Sophys husband explainedthey could not make a decision as he must discuss with his father in law who was not inthe village, and would return later that day. Sophys husband also expressed concernthat they didn't have money to go to the referral hospital. Sarom sat with them foralmost an hour and explained about anaemia and reassured them that they wouldn'thave to pay for transport to the hospital. They would also receive a small amount ofmoney from a local NGO to cover food costs during the hospital stay. Sophy didn't sayanything. Their faces showed that they were very reluctant to go to the referral hospital.As Sarom walked away she commented "I know they will call me when she startslabour pains and I will have to manage."


Sophy and her delivery house


Case Study No Two Koh Chrum, Siem Bok District

The six TBAs sat in a circle dearly animated by the sharing of experiences. They appearedrelaxed and at ease. Each was taking turns to speak about their work and problems theyhad encountered

Sixty year old Mou Sow expressed the wish to retire," I'm too old now and its time to takea rest. The problem with this job is we don't choose to become TBAs, village people chooseus and they keep calling us, how can we refuse?"

The other TBAs smile and nod their heads in agreement. Mou Sow continues to proudlymention that she is also an unofficial advisor to the "younger1 TBAs. " They always call mewhen they are worried because they know I have the most experienced TBA in this area ".

Yan a 47yr old TBA then shared her latest worry. A pregnant woman who has already hadsix pregnancies had a very big uterus and was feeling very tired. Yan related how she hasseen this woman twice already and advised her to go to the referral hospital or healthcentre for delivery, but the woman and her husband refused. They didn't have anyone totake care of their small children or their house..The other TBAs empathised and reminisced about similar problems they had faced.

The evaluators decided that when the discussion was over they would accompany Yanback to her village to observe as she reviewed with the woman and her husband possibleoptions to solve this problem..

As they travelled across the river in a small boat to Yan's village, Yan discussed her fearthat the woman may go into labour at night and have a problem. It was the dry season andthe water level is low, large rocks jut up from the water. Yan said at night it was verydifficult to avoid hitting them. It takes three hours by boat to the referral hospital and it isoften difficult to find a boat that will agree to go. It is also expensive. Even though the TBAwill be reimbursed travel costs at provincial level, if the family of the pregnant woman ispoor the TBA must first find the money to pay the boat.

They arrived at a small bamboo and thatch house, there were a group of small childrenplaying with the family dog under the house. Upstairs the nine month pregnant womanwas sitting pale and breathless in the corner. She appeared very anaemic and lookedexhausted. Her husband looked anxious and solemn. She had not visited the health centrefor antenatal care. Last time the health centre midwife came to the village Yan had calledher to check this woman. Yan gently palpated the woman's abdomen and said;" The uterus is so big, it seems like she has a lot of fluid. Its good that the baby's head isdown." After much discussion of possible options the husband agreed that they may go thehealth centre, but he would wait until his wife went into labour to decide. The healthcentre is a one hour journey by boat.

Yan looked anxious.. As Yan and the evaluators boarded the boat to go to another villagethey all expressed the hope that the woman's husband would reconsider and take his wifeto the health centre before she goes into labour.


TEA Yan tries to transfer a high-risk woman

23 Tradit ional Bir th Attendant Programme Stung Treng, Cambodia

TBA strengths concerning ante-natal care

TBAs involved in the evaluation were able to perform a correct abdominalpalpation and identify the position of the baby. They gave women adviceappropriate advice about the importance of good nutrition and hygiene duringpregnancy, also about attending for antenatal care with a government midwife.TBAs were able to identify at least four high-risk problems during pregnancy, andknew which problems should be referred. All TBAs reported distributing irontablets to pregnant women. It was clear during the village visits that TBAs have thetrust and confidence of the village women. TBAs are aware of which women arepregnant in their villages and know women's previous obstetric histories. AlthoughTBAs are not proactive in seeking to provide antenatal care they willing go whenrequested. Women only seek the TBA in pregnancy when they feel they have aproblem or they want to confirm pregnancy. Health centre midwives interviewedmentioned that TBAs were very helpful during outreach activities, calling pregnantwomen to come to an antenatal check at village level. The health centre midwivesgives the TBA a small financial incentive to participate in outreach activities.

Areas that need strengthening

TBAs evaluated didn't have a clear understanding of the importance of tetanusvaccine or how many times a woman should be vaccinated. Although theydistributed iron tablets they didn't realise that a pregnant woman needed to takemore than thirty tablets during her pregnancy, nor did they realise that a pregnantwoman could take more than one tablet per day if she was anaemic. Most of theTBAs observed examining pregnant women did not check for anaemia and the twopregnant women who were very anaemic were not recognised as anaemic by theTBA, although the TBAs had identified the women as being in poor health.

Pregnant women interviewed in eight villages didn't understand the necessity ofantenatal care. They felt antenatal care was only necessary if they had a problem.There is no understanding of antenatal care promoting a healthy pregnancy andallowing early recognition and treatment of problems. TBAs face various barriers inreferring women who have problems during the A/N period. One of the mostfrequent difficulties mentioned during the evaluation was that most high-riskwomen would not agree to go to a health facility when they had a problem duringtheir pregnancy. It was stated by both TBAs and village people that the referralhospital will not admit a pregnant woman until she is in labour. Another importantfactor preventing women seeking health care was the perceived low status ofwomen, who appear to have little decision making power concerning their ownhealth. In the two case studies of high-risk women, other family members wouldmake the final decision about transfer. People interviewed also stated the strongdesire of the woman and her family to stay at home and deliver with someoneknown and trusted. Poverty, lack of carers for younger children, lack of securityconcerning livestock or property were other constraints mentioned.


3.2 Safe Delivery

The Cambodian national policy for TBAs states the role of the TBA is to deliver lowrisk women at home and encourage the attention of the health centre midwife. TheStung Treng TBA job description states the main role of the TBA is to carry out aclean and safe delivery according to her training. Nationally women deliver 89% ofbirths at home and 66.3% of births are attended by a TBA, in rural areas thisincreases to 70% of births with a TBA (CDHS,2000)

Approximately ninety per cent of deliveries in Stung Treng are conducted at homeby TBAs with most of the active TBAs having received training and regularfollow-up and supplies The statistics for the last three years show there has beenlittle change in this situation although there is a slight increase in the number ofwomen from villages close to health centres delivering at health centre level(midwives at Srair Krasang health centre in Siem Bok district reported eightdeliveries at the health centre in 2002) TBAs and village women both identify themain role of the TBA as a birth attendant. The ability of the TBA to conduct a cleanand safe delivery is thus vital. During interviews with village women they statedthey prefer to call a trained TBA (they recognised her by the TBA bag she carried)

Chart 2 Total Deliveries Stung Treng 2000-2002

Year

200020012002

TBA

1,9262,2262,562

%

87.0%90.0%92.1%

Trainedmidwives

281243222

%

13.0%10.0%

8.3%

Total del.

2,2072,4692,784

Expecteddels

2,8873,3043,027

% Expectdels

reported76.0%74.0%91 .9%

2500

2000

1500

1000

500

0

92.1%

90.0%

87.0%

200020012002

1S 0.0% 8.30/0

TBA Trained midwives


TBA strengths concerning safe delivery

The evaluation team was unable to observe a TBA conducting a delivery at villagelevel. The TBA's ability to perform a clean and safe delivery was assessed byindividual TBA interviews, group discussions with TBA and village women andindividual interviews with newly delivered women and their families.

All the TBAs interviewed had meticulously kept delivery kits. Equipment was cleanand kept stored in the stainless steel box provided (see picture of TBA kit) All thekits examined were complete. All TBAs interviewed could describe correctly how toclean and sterilise the equipment. All mothers with young babies who wereinterviewed mentioned without prompting that the TBA boiled her equipmentwhen she came to the house. Mothers also reported that they liked to call the TBAwho had received training and had a delivery kit. They believed the TBA who hadreceived training was more skilled and used clean equipment. Some village womensaid it was difficult to access a trained TBA during the farming season as manyvillage women with their families moved temporarily to live close to their fieldssome distance from the village. In these situations they called an untrained birthattendant.

TBA's interviewed, accurately described the steps in caring for a woman duringlabour and conducting a delivery. All TBAs examined the woman abdomen tocheck the position of the baby, all washed their hands before delivery and claimedto wear gloves. (The TBA monitor report states that of 182 mothers interviewed,98-100% reported that TBAs wore gloves during delivery) The TBAs always askedthe mother to shower and wear a clean sarong during labour. Preparation of aclean delivery area was described and all the TBAs said they use the plastic sheetprovided in their kit to help keep the area for delivery clean. It was interesting thatmany TBAs and village women said that two TBAs attended the delivery one todeal with the mother and the other to manage the baby after delivery.This partnership was usually a mutual agreement between the two TBAs.

TBAs described waiting for the baby's head to appear, and then receiving the babyas it was expulsed. Since training they reported they had discontinued the practiceof pushing on the woman's uterus to deliver the baby. Another change theymentioned was that babies used to be washed immediately after delivery, butduring training they had learned that this is not necessary. Now they cut the cordand immediately dry and wrap the baby. All the TBAs could describe accuratelytying of the cord. About 50% of the TBAs and mothers interviewed said the TBAputs the baby to the breast after delivery. TBAs said that this was to help theplacenta to come out quickly. Only one TBA reported seeing the baby suck afterputting got the mother's breast, the other TBAs said the baby didn't suck and theyput the baby away from the mother again. No TBAs mentioned the value ofcolostrum or breast milk for the baby. All TBAs reported applying gentian violet onthe cord following delivery.

Delivery of the placenta was recognised by the TBAs as a dangerous time. Theywere able to describe accurately signs of placental separation and gently exertingpressure on the top of the uterus to help the placenta deliver.


All TBA mentioned checking to see if the placenta was complete. Two TBAs saidthey would explore the uterus if they thought a piece of placenta was missing.

No TBAs mentioned checking the mother's perineum for tears or treating the tearsby cleaning or applying gentian violet. All TBAs mentioned checking for bleedingafter delivery. Besides putting the baby to the breast, cutting the cord and applyinggentian violet no further baby care was described. One 60yr old TBA describedrubbing a piece of umbilical cord in the baby's mouth to prevent mouth infection.

All TBAs interviewed could name three danger signs during labour and delivery.The common danger signs mentioned were bleeding before and after labour, a longlabour and a retained placenta. All the TBAs said they would manage post partumhaemorrhage by massaging the uterus and putting the baby to suck on the breast.

Areas concerning labour and delivery that need strengthening

TBAs can potentially provide important support to mothers to helping themestablish early and exclusive breast-feeding. Presently they lack the knowledge andskills to do this well. Some TBAs mentioned squeezing breast milk into the baby'smouth if s/he didn't suck. Others mentioned that if they baby didn't suck theywaited until it cried.

Two TBAs mentioned that if the placenta was not complete they would perform amanual revision of the uterus, this is potentially very dangerous.

TBAs did not mention examining the newborn baby after delivery. Some TBAsreported putting eye ointment in the baby's eyes after delivery and others said theyhad never seen eye ointment.

Most TBAs reported keeping the baby in the same room but at a distance from themother following delivery. Almost all the mothers practised three days of 'roasting'and drinking of traditional herbs after delivery. TBAs reported that theyencouraged 'light roasting.'

It is likely that TBAs don't always use gloves during delivery. TBAs are suppliedtwo pairs of gloves every three months. Some health centres in Stung Treng areselling delivery kits to expectant mothers for 3,000 Kiel. This kit includes a pair ofgloves, a plastic sheet, soap powder and a razor for cutting the cord. TBAsinterviewed reported they had seen the kit but felt it was unnecessary as theyalready had equipment; they felt the razor and cord ties would be clumsy to usewhen they had become accustomed to using cord clamps and scissors. Mothersinterviewed had not seen the kit but commented that it was not necessary to spend3,000 Riel when TBAs already had equipment. No TBAs interviewed were sellingthis kit (it was originally planned that TBAs would sell the kits to expectant mothersand would receive 500 Riel of the 3,000 Riel for each kit sold. Some mothersinterviewed said they would not be able to afford 3,000 Riel for the kit. The TBApresently reuses the plastic in her kit. It has the potential to spread infection if notwell cleaned, although during the evaluation all the kits examined had a cleanplastic.


TBAs interviewed were asked to assess what was different in their practice following trainingand why:

" I'm so happy to have the TBA kit Before training I didn't know that it was important to boilthe equipment I used to use a oyster shell and piece of charcoal for cutting the cord, the shellwas clean as I used to soak it in hot water but I never boiled it."

"Before I received training I used to put insects nest on the babies cord. After training I startedto use gentian violet and I've found its better. The cord falls off quickly and doesn't becomesticky and wet"

" Now I don't put a price on my service. Before I used to charge for the cost of the equipmentand supplies. Now I have equipment and regular supplies I just accept what the family offersme and it's much easier"

" Before I received training I didn't know the danger signs during delivery. I used to just waitand hope that everything would be OK. Now I know what problems I should refer to thehospital but it's still difficult to transfer because of the distance."

" I can't eat well for three or four days following a delivery. I think its because since I receivedtraining I know about the problems that can occur and I worry more than before."

" I feel proud to have my TBA equipment and bag. It is a sign that I have received training."

"Some village people misunderstand and think that I have a salary now that I have receivedtraining. Sometimes they are reluctant to pay me for my services".

" Tine TBA bag is very important when I transfer a woman to the referral hospital.'The referralhospital staff can immediately see I have received training. Sometimes I'm able to go into thedelivery room to observe".


TEA Training

Traditional Birth Attendant Programme Stung T'reng, Cambodia 29

TBA equipment used before and after training

TBA equipment used before training

TBA equipment used after training

<--,...„„ -!-..„.,„ r^-,i-.ii-i/->,Hi

3.3 Post natal care

The MOH policy for TBAs states the role of TBAs concerning postnatal care is toprovide health education to mothers, promote early breast feeding and referproblems to the nearest health staff. Stung Treng TEA job description specifies theTEA should visit at least five times post delivery to check on the mother and babyand to promote breast feeding within one hour of birth and exclusive breast feedingthere after.

TBAs skills regarding postnatal care were assessed by observing TBAs visitingnewly delivered mothers and their babies. Also by interviewing mothers withbabies below one month, village women's discussion groups and individual TEAinterviews.Village women interviewed reported that the TEA is an important person toprovide postnatal care at village level. The TEA is central to the traditions followedin thepost natal period. All the TBAs interviewed reported visiting the mother once perday for up to three days following delivery. Although most of them mentionedputting the baby to the breast following delivery, none of them mentioned checkingthe baby's breast feeding in the postnatal period. The TBAs saw their role in thepost natal period as checking the baby's cord, the size of the mothers uterus andchecking the mother for bleeding. TBAs also said they advised the mother aboutgood hygiene arid appropriate foods to eat following delivery. Most advised not torestrict foods following delivery. All the mothers interviewed had roasted for threedays following delivery and drank traditional Khmer herbs made by relatives.The TEA played an important role in participating in the ceremony to celebrate theend of 'roasting' on the third day. It is at this time that the family shows theirgratitude to the TEA and presents her with a gift according to their means.

All the TBAs reported change of practice since training including thediscontinuation of the practice of drinking salt water to induce thirst in the newlydelivered woman. They also said that before training they had thought that heavybleeding was a good thing to clean out the uterus. Now they were more alert andworried in case there should be heavy bleeding. Both TBAs and village womenmention that if the newly delivered woman had enough money they would callhealth staff to give injections. They usually inject vitamins but also sometimes injectantibiotics.

Areas concerning post natal care that need strengthening.

The TBA is in an excellent position of trust and proximity to women at village levelto promote the early establishment of exclusive breast feeding and to supportmothers in managing common breast feeding problems like engorged breasts. To dothis they will need further practical on the job training in basic skills concerningproper attachment, counselling skills and management of common breast feedingproblems. Many of the mothers interviewed were giving pre-lateal feeds such assugar-water to their babies and were unaware that breast milk was a complete foodfor the baby.


TBAs need to be alert to early detection of potentially serious problems for mothersand babies in the post natal period. TBAs didn't mentioned any danger signs in thenew-born. Stung Treng statistics show a high neonatal mortality rate with hardlyany information about why these babies have died. When asked about possibleproblems that could occur in the post natal period the only problem mentioned washeavy bleeding from the mother. None of the TBAs reported checking the motherfor fever in the post natal period. When asked what should be done if a baby had afever following delivery two TBAs answered that it would not be a problem as themother was drinking Khmer medicine and that this would also have an effect onthe baby who was breast feeding.

Several of the newly delivered mothers interviewed at village level had clinicalsigns of anaemia. Some of them had taken iron tablets during pregnancy but nonehad taken them postnatally. TBAs interviewed were not aware that mothers neediron tablets after delivery and had never distributed iron once a woman haddelivered her baby.

3.4 TBAs role as health educators

Mothers interviewed at village level identified TBAs as their main source of healthinformation concerning pregnancy and delivery. Several mothers said they hadlearned about what to eat during pregnancy from a TBA. They had also receivedinformation from a TBA about receiving tetanus vaccination, although most womenwere not aware of what the benefits of this vaccine was or how many times theyshould be vaccinated. Village leaders mentioned that an important role of TBAs wasto give health information to pregnant women.

Women also reported that TBAs had advised them to seek the health centremidwife to obtain ante-natal care but most of them had not followed this advice asthey saw no benefit in checking their pregnancy. None of the mothers interviewedcould name problems that might occur during pregnancy though with promptingsome mothers mentioned oedema as a problem during pregnancy.TBAs reported giving information about foods that were good for pregnant womenbut there was no consistency in what they advised and each TBA had differentmessages about nutrition. Although all TBAs were distributing iron tablets mothersdid not know what the benefits of the iron tablets were or how many tablets theyshould take during pregnancy. There was no awareness among TBAs or villagewomen that iron tablets would be beneficial in the postnatal period.

TBAs reported giving women advice about what to prepare before delivery andalso how to ensure they delivery place was clean. At village level the evaluatorswere able to observe that pregnant women had prepared very well for theirdelivery, and it seemed they had all followed closely the advice of the TBA.

Apart from putting the baby to the breast immediately after delivery TBAs did notmention giving mothers advice about breast feeding or supporting the womanwhile breast feeding for the first few days.

3? Traditional Birth Attendant Programme Stung Treng, Cambodia

One TBA mentioned helping a woman when her breasts were engorged by usingwarm compresses and expressing some of the milk.TBAs did not mention the benefits of colostrum or of exclusive breast feeding.None of the mothers with babies under six months old who were interviewed wereexclusively breastfeeding.

During the post natal period mothers reported that TBAs told them about regularbathing and the importance of eating nutritious foods. There was no reportedadvice about the care of the newborn or care of the cord. TBA interviewed gavesimilar information and did not mention giving advice about care of the newborn.The only problem identified by TBAs in the post natal period was maternalhaemorrhage. Mothers interviewed could not identify any problems that mighthappen in the post natal period.

Mothers interviewed all knew about immunisations for their children and said theyhad obtained this information form health staff during outreach sessions in thevillage. TBAs interviewed did not mention giving information to the mothers aboutimmunisations, although two TBAs said they call women with young children to bevaccinated when the health staff comes to the village. Several TBAs were confusedbetween the children's immunisations and the mother tetanus irrununisation andcalled them both 'protection against the six diseases.1

Mothers reported that TBAs did not give birth spacing information. Most mothershad received information from other women in the village. TBAs interviewed didnot seem interested in giving information about family spacing methods or feel itwas something they should do, although they could all name two main methodsand knew where they could be obtained. There were many rumours at village levelconcerning birth spacing methods and it was reported that if one woman in thevillage had a problem other women would decide to stop this method even if theydidn't have a problem.

3.5 Identify high risk and the referral system

All TBAs interviewed could identify some signs of high risk during pregnancy andlabour that would need referral, and all mentioned that haemorrhage after deliverywas a problem needing referral. All TBAs expressed concerns about referral andmany of them related cases of referral that had been very difficult especially casesinvolving women haemorrhaging before or after delivery. In the villages visited thebiggest constraint faced by TBAs wanting to refer is the refusal of the woman or herfamily to be transferred. Reasons given for refusing to go to the referral hospitalincluded having no money, fear of been far from home and not knowing the staff,lack of confidence in the referral hospital services, no one to care for small children,property and possessions when away from home. Women also expressed a fatalisticattitude that it was better if they were to die that it would be at home rather than ina hospital.


£i? i/»v< Owv >l'> I<C'i6iiXjiC<>i«iC«Uii,x>C «i.1>'i,»- w

TBAs and village women all agreed that women couldn't make the decision to go tothe referral hospital, the decision will usually be made by the woman's partner or afamily elder. TBAs expressed the importance of the funds provided by YWAM toreimburse travel expenses. The majority of the women needing transfer areextremely poor and without this help they would waste valuable time trying toraise or borrow money for transport to the referral hospital.

In the last three years transfers have increased from all districts except Siem Pangwhich is very remote being lOOkms from Stung Treng town, with travel solely byriver.

Chart 3 Total Transfers by TBAs 2000-2002

TOTAL 2000TOTAL 2001TOTAL 2002

TOTAL

APH

81925

LongLabour

41016

Stillbirth

233

RetainedPlacenta

457

positionbaby

abnormal

321

Twins

133

abnormalbaby

020

oedemamother

014

otherproblems

mother

64

12

Totals

284971

148

Total Transfers by TBAs 2000-2002

fc APH~ Retained placenta61 Abnornal baby

25

20

15

10

5

0

• Long LabourH Position baby abnornalif' Oedema mother

n Stillbirth• TwinsK Other problem mother

ii r ii-TOTAL 2000 TOTAL 2001 TOTAL 2002

The reasons for referrals show that all transfers were appropriate. The breakdownof reasons for transfer also shows that the majority of women are transferred whenthey have a serious problem either during labour or in the post natal period.The majority of transfers needed prompt emergency obstetric management. Veryfew high-risk women are electively transferred during the antenatal period. Some ofthe problems concerning elective transfer of high risk women have already beenraised in this report. What was startling at village level especially in JamgarLeu,Thala borivat district, was the poor health status of the majority of the women.

•» V> /-\ /•{ i -jj

Malnutrition and anaemia were very visible and not recognised by the villagepeople as a problem. During village discussion groups almost 75% of womeninterviewed had a child that had died in the first year of life. The high birth ratewas also very visible with small children and toddlers in abundance.

TBAs experiences when transferring women to the referral hospital

1. TEA, Jamgar Ler comune, Thalaborivat district." It was a terrifying experience . They called me during the night A 30yr mother with herfourth pregnancy was in labour. Her first three children had all died in the first few days oflife. When I arrived at her house she was having weak contractions but was bleeding fromthe vagina. I thought she would die. I told her family we must go quickly to the provincialhospital. Two of her family carried her in a hammock on a pole. I walked beside them givingher sips of water all the way. We left her village at 5am and arrived at the hospital at 8pm.Itwas the rainy season and the road was very difficult When we arrived at the hospital shehad lost a lot of blood, the placenta was in front of the baby. The hospital gave her blood anddid a caesarean section. The baby was breech and died. I couldn't sleep that night, I thoughtto myself I don't want to be a TBA, its too hard. I stayed at the hospital with the woman forthree days. Women in the village continue to call me and I always go, its difficult to refuse"

2.TBA Jamgar Ler commune, Thalaborivat district" Last year I transferred a woman who was bleeding following an abortion. The ox cart cost40,000 Riel and took from 7am to 5pm.At the referral hospital they did a curettage. Thewoman's family borrowed 200,000 Riel from the village, and they are still paying it back.

3. Tbong Cla vil!age,Siem Bok district"It was the woman's sixth pregnancy and she looked very big like twins. I couldn't feel theuterus as it was so full of fluid. She ruptured the membranes before she started labour pain.When I felt the position of the baby after the membranes ruptured, 1 felt a very big head. Iadvised the woman's family that she should go to the referral hospital They refused. Afterseveral hours of labour the baby's head did not come down. I refused to stay longer with herand told her family again that she must go to the hospital. Finally they agreed. We travelledby boat for two hours and it was already evening when we arrived. There was only a youngmidwife in the hospital the other staff had gone to a wedding. The midwife asked thewoman to push, I told her the woman could not push as the baby's head was very big andhad not come done. Later she called another staff member and he came and tried to do avacuum extractor. The baby could not be born. Eventually she had a caesarean section, thebaby died, it was abnormal with a big head. The woman was sterilised during the surgeryto prevent another problem.. She needed blood and spent 350,000 Riel (approx US$ 87)during her stay in the hospital.

Yei Rim's Niece

Yei Rim sadly recalled the time she transferred her niece to the referral hospital. Her niecewas married to a soldier and pregnant with her fifth child. They were living in Kratieprovince. When she went into labour she was attended by a 'military midwife'. After a longlabour with no progress her family decided to transfer her to the care of her aunt a TBA inStung Treng province.

On arriving in Stung Treng her niece was in very poor condition. Yei Rim examined her andfound the baby was in a transverse position. Her niece's husband was very drunk. Yei Reamhad to argue strongly with him before he agreed that his wife could be transferred to thereferral hospital.

" When we arrived at the referral hospital the doctor was in Phnom Penh. I saw the hospitalstaff cut the baby in pieces like a piece of meat to remove it from the uterus. The baby wasalready dead. My niece survived but one year later she was murdered by her husband whostrangled her when he was drunk"


TB As interviewed reported that transfer of women to the referral hospital was verystressful for all concerned and that they encountered many obstacles and delays.Some of the common problems have been illustrated in the case studies and are:

Women inability to make a decision without their partner or relatives consentLack of confidence in the quality of service available at the referral centrePoverty and lack of support to care for property or younger children whilethe mother is absentDelays in arranging transport especially during the rainy season or at nightPregnant women's1 lack of information and understanding about the possiblerisks during pregnancy, labour and delivery.

TBAs appreciated the funds available from YWAM to pay transport costs, which arere embursed on arrival at the provincial town. There is a need to strengthen thereferral system and provide support to TBAs at this level. It was planned that aemergency transfer fund would be set up at village level but for various reasons thisnever materialised. This should be reconsidered and in participation with people atvillage level in motivated villages a small pilot project could be started. The TBAsalso need respected people with authority who can support them in arranging thetransport of emergency cases.

The possibility of "waiting homes' at provincial level for high risk pregnant womenshould be explored. They have been successful in several other countries, especiallyin countries that set them up with community participation. Stung Treng transferstatistics show that very few women are electively transferred during pregnancy.Most are transferred when they are already an obstetric emergency.

3.6 Maternal Deaths in Stung Treng

The low population and small number of maternal death cases involved in StungTreng province results in wide confidence intervals and difficulties in reliablydetecting changes. WHO states that the maternal mortality ratio (MMR) has thefollowing constraints: observed differences in maternal mortality ratio may not bespecific to improved maternal health status, but may be due to changes in thereporting system and ascertainment of maternal deaths, or to a wide randomvariation resulting from a small number of events (WHO, 2000) MMR sensitivity tooverall changes in maternal health status may be low and there may be largechanges in the burden of morbidity's before this is reflected in the (MMR).A national average may hide wide differentials between areas or population groups.The MMR is useful at national level as a direct measure of health status and usefulat local level where each maternal death should be reported and reviewed toprovide information for programme planning.

A Stung Treng Maternal Death Audit committee was set up in 2002 and sevenreported maternal deaths were investigated. Five of the seven cases wereprimigravida .


One case had a postpartum haemorrhage and retained placenta. She was referredby the TEA to the hospital but died at the hospital following a uterine revision. Noblood was given. The four other primigravida had premature deliveries, one wasmalaria positive ++++, one with eclampsia and two with fever of unknown originbefore delivery. Two other mothers a gravida 7 who ? had induced abortion atfive months followed by haemorrhage and a gravida 5 who had a history of chronicillness for four years with an abdominal tumour She delivered prematurely atseven months at home. The baby died 15minutes after delivery? cause of death.

One of these cases were transferred to the referral hospital by the TEA with aretained placenta and bleeding twelve hours after delivery, and one other case wasadvised by a TBA to go to hospital but refused. Relatives and traditional healersdelivered two of the cases. Only one of the seven cases had received antenatal careand her high-risk condition was not identified during her antenatal visit.

The case histories of the seven maternal deaths in 2002 illustrate clearly the poorhealth status of pregnant women, women's lack of access to health care services,women's and communities lack of awareness about danger signs during pregnancy,lack of decision making powers of women and lack of confidence by patients andtheir families in government health services. Reviewing the case studies it seemsthat in only one case a TBA may have been slow in referring the woman to thereferral hospital. In the other cases that TBAs were involved, their actions wereappropriate to the situation. It is important when investigating the events leadingup to a maternal death to identify and document any reasons causing delay inreferring so that theses constraints may be addressed.

During field work in Stung Treng village people expressed feelings about lack ofconfidence in the services available at the referral hospital, and their belief that oftenin emergencies senior staff were absent. Some village women said they knew ofwomen whose families had decided to transfer them to Kratie province (three tofour hours travel by boat from Stung Treng town), as they could be certain therewould be senior staff to manage the problem. Another frequent comment fromvillage people was that even if a woman had problems during her pregnancy, it wasimpossible to be admitted to the obstetric ward until the woman was in labour.

During interviews with hospital midwives they discussed some of the constraintsthey face. The obstetric ward has been without a light at night for several months.They previously used a battery and florescent bulb. It was taken away forre-charging and because of lack of funds not brought back. This means that from12.00 midnight to Sam there is no light in the referral hospital obstetric ward! Ahospital generator is turned on when a patient needs surgery. A woman recentlydied from post partum haemorrhage in the hospital and resuscitation wasattempted by candlelight. The blood bank has also experienced problems, in thatalthough there is equipment and supplies to cross match and transfuse, onoccasions, women needing emergency blood have not received it. Recently theblood bank did not have any staff although a new staff member is now trained andready to work.


It is estimated that from 1- 5% of expected pregnancies warrant caesarean sectionfor delivery. This illustrates clearly how in Stung Treng province the large unmetneed for HOC.

Chart 4 Caesarian Sections (CS) Stung Treng 2000-2002

Year

200020012002

Total no of pregnant women

2,8873,3043,027

CS

111213

%

0.38%0.36%0.43%

0.36%0,43%

200020012002

Caesarian Sections (CS)

- r- -r i"^~,l^^A;

As long as essential emergency services are not available or inaccessible, TBAs orhealth centre midwives will be unable to achieve significant improvements inmanagement of women with serious obstetric complications. Dr John Rooney (1999)quotes an article by Deborah Maine and Allan Rosenfield " The Safe Motherhood:Why has it stalled. "This article suggests that maternal mortality will not bereduced by a reliance on improved uptake of antenatal care nor by a generalimprovement in social and economic conditions in the community. There isevidence to show that reduction in maternal mortality is specifically linked to accessto quality Emergency Obstetric Care (EmOQ It is only when people observe theimpact of accessible and successful HOC on preventing the deaths of pregnantwomen that they will be better ready to respond to the broader messages abouthealth and nutrition provided through antenatal and other services".

In 2000 18 maternal deaths were reported, 5 in 2001 and 10 in 2002. There were fourreported in the first three months of 2003. In 2002 TBAs reported six out of sevenmaternal death cases that were investigated. Much can be learned from maternaldeath audits. It is important to ensure that audits are done in a sensitive way andpeople involved in the case are not made to feel that the investigation is seeking tofind fault or to blame them. It is also essential to ensure that any recommendationsmade are realistic and achievable. Equally important is to ensure recommendationsare actually implemented and don't just remain words on paper.

Emphasis on advertising success cases should also be considered (near misses).Communities will be very motivated to listen to and will remember case studies ofwomen in their village who were transferred and had a successful outcome. Thebest communicator is the village woman who was transferred; who with thesupport of the TEA or health centre midwife could retell her story at a villagemeeting.

Because of the small population of Stung Treng and of the low number of maternaldeaths expected, reduction in maternal mortality is not the most appropriateindicator to measure improvement in maternal and child health. As the WHOstates there may be large changes in the burden of morbidity before this is reflectedin the MMR. It may be better to select key process indicators that would moreaccurately indicate an improvement in Women and child health.

It is clear that same of the mothers who were transfered by TBA's may have becomematernal deaths, if not for the prompt intervention by the TBA's.


Sochea

Sochea is a pretty, petite four day old baby girl. She lies on a cushion under a mosquito net.Beside her is a plastic baby bottle and a can of Alaska sweetened condensed milk. Her olderthree year old brother is crying and pulling at the arm of his father who sits forlornly staringinto space.

Sochea's mother, thirty three year old Sopheap died four-hours after Sochea was born. She bledto death.

Sochea has two older brothers; both boys' aged three and seven yearsNold. The seven year oldis the son of Sopheaps first husband who separated from many years ago. Sopheap's secondhusband is a soldier. Her children will now be cared for by their grandmother, Sopheap'smother.

Sopheap had not been well during her pregnancy and had a chronic cough. When she wasnine months pregnant her mother persuaded her to go to the hospital. She was admitted to themedical ward and treated for a chest infection. She stayed on the medical ward for ten daysbefore she started labour pains. Her mother brought her to the delivery room at 10pm thatnight.She delivered Sochea at 12 mid-night and was sent back to the main ward. Her mother noticedshe was bleeding following delivery and called the hospital staff to help. The staff made a uter-ine revision in the ward. Sopheap kept bleeding profusely. She was moved back into thedelivery room and given two IV infusions and some oxygen, another staff member did asecond uterine revision. Sopheap kept bleeding and died at 4am.

Charttha

Om Sen a TEA for fifteen years recalled the case of a young woman who died in 2000.Seventeen year old Chantha lived with her parents. She was six months pregnant when OmSen was called for advice. Chantha was very tired and also had oedema of the legs and face.Om Sen advised her parents that the swelling was a danger sign, but her parents refused to letChantha go to the health centre or be transferred to the referral hospital." They felt shy asChantha was not married."

When Chantha went into labour when eight months pregnant, her parents did not call OmSen. They conducted the delivery by themselves. Chantha died four days following delivery,her baby had died a few hours following birth.

" Chantha was unconscious since delivery, her mother and father said they had no money totake her to the hospital. When I pleaded with them to transfer her they said it better to let herdie at home. After she died her mother and father felt guilty. I felt sad, there was nothing Icould do".


3.7 The TEA curriculum

The evaluators reviewed the Stung Treng TEA training curriculum and compared itto the national TBA curriculum used in many other provinces of Cambodia (MoH,2001) It was found that the TBA training curriculum was very similar to the nationalcurriculum and contained all the relevant topics reflecting the Cambodian SafeMotherhood policies.

TBA is trained to educate women about nutrition, anaemia, breast feeding andpromotes birthspacing and immunisation services, alsotcTeneeurage pregnantwomen to attend antenatal care. They receive both theory and practise (with a largepercentage of time allotted to practice) on how to conduct a normal delivery withemphasis on cleaner delivery practices recognise and refer women with high riskpregnancy and complications during pregnancy, labour, delivery and post partumto the closest health centre or referral hospital. They also receive practice inconducting postnatal visits and identifying common problems in the newborn.

Training materials are appropriate and most pictures including signs of high riskpregnancy are ones that were developed at national level. There are lot of practicalsessions and discussions with adequate supplies and equipment for demonstrations.

3.8 Reporting system

TBA's record their monthly activities on a visual reporting form. These forms aresubmitted during the TBA meetings. They also report problems that occur at villagelevel especially maternal and peri-natal/neonatal deaths, during attendance at theTBA meetings. During fieldwork TBA's could relate cases of maternal deaths intheir areas for the last few years and give accurate details. It might be difficult forTBAs to identify all cases of maternal deaths as their definition of a maternal deathis of a mother dying during or soon after labour. There was no mention by TBAs ofwomen dying in early pregnancy or after the first week following delivery. Over90% of expected neonatal deaths have been reported by TBAs in the last three years,which confirms that TBAs are providing accurate information concerning thisindicator. It was found during field work that some TBAs were confused about thedifferent pictures on the reporting forms, and hadn't filled them in since the lastmeeting. Most TBAs reported completing them just before the meeting or even atthe meeting with help from someone else. This meant there was sometimes longintervals before the TBA recorded her previous three months activities, making thepossibility of inaccuracies likely. It appeared also that different districts haddifferent systems for collecting the reports. Some TBAs said they gave their reportto the TBA group leader in their areas and others said they delivered the reportduring the TBA meeting. Health centre midwives who attended the TBA meetingswere given summaries of the total TBA activities in their area, but the TBA reportsdid not go directly to health centre level.


It would improve accuracy and facilitate prompt response to anyxproblems, if TEAactivity reports are collected monthly. It might be possible to involve a member ofthe VDC to collect the reports at village level, and submit to the health centremidwife once per month. The health centre midwife could then include the TEAactivities in her monthly report and submit this three monthly to the provincialhealth department TEA manager. The TBAs could be given back their report formsduring the TBA meetings and have the opportunity to discuss any problems thatoccurred. 1

There is lack of information about possible reasons for peri/natal/neonatal deathsand lack of information about infant and under five mortality at village level. Itwould be very helpful if the VDC member could communicate closely with TBAsand include a monthly summary report of all infant and under five deaths in thevillage.


TBA training materials

/i-5

3.9 TEA programme costs

Item

Support to government staff

Drugs /medical supplies

Non medical supplies

Training/education

Medical equipment

Others

Total

2000

9212

191

358

286

283

888

11, 218 US $

2001\

\6740

212

133

1593

/1885'

10,563 US$

2002

7506

| 121

35

2133

9,795 US$

NB.

Support to government staff covers the cost of travel, perdiems, for the trainers andthe TBAs. Non medical supplies and equipment lines have both decreased over theyears as the government has supplied these through the MOH Central MedicalStores. This budget does not include Sesan district (UNICEF supported) exceptSesan TBA travel expenses for emergency transfers. This budget also does notinclude the cost of the YWAM expatriate advisor who is supported by private dona-tions.

If the cost of the programme for 2002 is divided by the total number of active TBAsin four districts, excluding Sesan (326 TBAs) it will be seen that the TBA programmecosts approximately 30 US$ per TBA per year or 2.50 US per month per TBA.Thetotal cost could also be divided by the total number of deliveries done by TBAs inthe four districts 2002 (2,237) each delivery by a TBA costs approximately 4.30 US$.(not including cost of supplies provided by MoH).

CARE organisation who has supported the TBA programme in Kompong Chhnang,Cambodia for the last ten years, assessed their TBA programme to be cost effectivewith each TBA costing approximately 44 US$ per year (this did not include supportfor emergency transfers). The Stung Treng TBA programme runs at 32% lower costper TBA per year than the Kompong Chhnang programme. This is a commendableachievement in a remote province with difficult access to most villages.

This programme also provides value for money in that the largest percentage of thebudget (76% in 2002) was spent on capacity building for both TBAs and govern-ment health staff.

/i/i Tr-o^i>i^,T^cil RiVtVi Af+on/Har>f Prnrrramma QtiinCT Trpncr Cambodia

3.10 Programme sustainability

The management structure of the TEA programme

There are three provincial health department staff responsible for the TEAprogramme. These three staff are also the senior TEA trainers who are veryexperienced, confident and are skilled in participatory approaches and training ofsemi-literate and illiterate students. The manager of the TEA programme isresponsible for TEA activities in the districts of Thalaborivat and Siem Pang,another MCH staff member is responsible for Siem Bok and Sesan district and thethird member of the team, who is also in charge ofihechild_irrmiujiisationprogramme is responsible for Stung Treng district. The TEA programme managercurrently spends up to 50% of her time on TEA programme activities whichincludes planning and monitoring the programme, conducting refresher training,and organising and participating in TEA meetings and TEA follow up at villagelevel. The other two members of the provincial MCH team spend about 15%-20 % oftheir time on TEA programme activities. The programme manager receives a smallsalary supplement from YWAM. Two YWAM national staff who are experiencedformer TBA's provide additional support during training and follow up activities.

The quality of programme management is good, with efficient implementation ofplans and regular follow-up support and training of the TBAs. Programmedocumentation needs improvement, presently information and statistics are kept invarious locations and recorded in different formats. Because of this informationabout the programme is not always easily accessible and achievements immediatelyidentifiable.

An important weakness of the programme is that implementation of activities in thefield are mainly conducted by the provincial TEA programme managers.This strategy was appropriate for the first few years of the programme, but it is nowtimely to accelerate the process of decentralisation. The provincial staff are keypeople at provincial level who have other responsibilities. Although health centremidwives have been involved in training and refresher courses, the provincialtrainers report that many of the health centre midwives lack delivery experienceand are not confident trainers. They are reported to be inhibited and shy whiletraining TBAs and are thus unable to gain the respect of the TBAs. It needs to bediscussed and planned how to strengthen the skills of the already scarce humanresource pool of midwives at health centre level, so that in the future they canassume responsibility for training and follow-up support of the TBAs in their area.

The quality of the TEA programme will undoubtedly suffer with hand over ofresponsibilities to less experienced and younger staff, but in the long term thereshould be increased benefits. During the evaluation the health centre staff expresseda willingness to assume more responsibility for TEA activities and in some healthcentres they have already began to increase their involvement. To achieve asuccessful transition strong support will be needed during a period of gradual handover. It is clear that some health centres will need closer support than others.


Suggested steps that could be taken to facilitate decentralisation of theprogramme.

1. Increase on the job training and supervision to health centre mid wives toimprove their technical skills and confidence. On the job training should bedone by the experienced TEA managers/trainers and'conducted at bothhealth centre level and during outreach activities^

2. Conduct a TOT for health centre midwives with selected literate TBA leaders.During this training form 'Commune TBA Training Teams' consisting of ahealth centre midwife and capable TBAs. Some of the TBAs interviewedduring the evaluation were intelligent, skilled and respected.The experienced TBA could concentrate on teaching delivery skills and thehealth centre midwife on antenatal, post natal care, immunisations andfamily spacing. It would take time to make this transition and the quality oftraining may not be maintained at the present level, but the longer termbenefits would be that it will help strengthen the link between the TBAs andhealth centre and increase trust and confidence between the two groups.The provincial trainers would remain important as technical advisors andmonitors of the programme. They would have more time to co-ordinate andsupervise the various aspects of the province wide MCH programme,promoting an integrated rather than a vertical approach to reproductivehealth services.

3. The TBA meetings should be conducted by the 'commune training teams'and follow up TBA visits at village level could be conducted during regularoutreach work by health centre staff with occasional monitoring byprovincial health department staff.

4. To ensure continued transparency in use of TBA program funds the healthcentre management committees should oversee the payments of perdiems toTBAs for meetings and training's and provide reports to PHD/ YWAM andthe health centre feedback committees on a regular basis. PHD and YWAMshould discuss and develop a policy concerning the use of and methods ofpayment of TBA programme funds and provide regular monitoring on theuse of these funds.

4A Trarh'Hnnal Rirth AHp>nHant Prncrrammp 5-ituner TrpnP. Cambodia

ix-j ^u^»^ *vXCv>t>:<ii ^>i J o :ii.oi

Mothers and babies

Mothers and babies

r-, T rx,,,,u,,,-i,

4. Discussion

The issue of sustainability of TEA programmes is a topical and contentious one.From the TBAs perspective their work as a birth attendant is sustainable and alwayswill be as long as village women continue to choose them as their favoured birthattendant. They are private practitioners whose livelihoods are supported by thewomen who request their services. But being a TBA is about much more thanmaking a living, as can be seen from some of the case studies in this report TBAsare not rich or even comfortably off members of their communities. Most of themexpressed during interviews that due to lack of income they cannot achieve theirmost important goal of educating their children. All of them do various other kindsof activities to supplement their income. Most of them spend a large percentage oftime doing heavy manual work in the fields.

One of the Cambodian MoH priorities is to improve the quality of the TBAs workand strengthen the links between them and government health staff. To achieve thismany additional responsibilities have been added to the TBAs 'job description'often without consulting TBAs or women at village level who use their services.In most provinces where TBAs have received training they are expected to attendregular follow up meetings, submit activity reports, conduct antenatal and postnatal care, give health education, join government health staff outreach activitiesand refer high risk women to government health staff.

The Cambodian MoH like many other developing countries does not recognise theTBA as government staff, therefore she is not entitled to a government salary orother conditions. There is not a government budget available for follow up supportof TBA activities. A JICA survey (2001) in ten provinces of Cambodia reveals that inall ten provinces the provincial health departments are dependant on NGO's tofinancially support the training and follow up of TBA programmes. The CAREsupported TBA programme in Kompong Chhnang which has been running for thelast ten years estimates they spend approximately 44 US $ per TBA per year (MoH,2001) (This included equipment, refresher training, follow up and perdiems forhealth centre and provincial staff). Cambodia's dependence on external funds tosupport TBA activities is likely to continue for several more years. It is appropriateand necessary to provide funds for TBA programmes as long as it is in parallel toand does not detract from the long term goal of having trained experiencedmidwives providing quality comprehensive reproductive services at everyhealth centre.

As illustrated in this evaluation TBAs are providing a much needed and valuedservice to village women at a minimal cost. They are filling a huge needs gap thatpresently the Cambodian government is unable to meet. TBAs interviewed inStung Treng province reported that without the perdiem provided for travel andfood, they would be unable to attend the refresher training or TBA meetings. TBAprogrammes are probably one of the few activities in Cambodia where tangiblebenefits such as continuing training opportunities are accessible to poor villagewomen in remote areas. The TBAs in turn are putting into practice newly learnedskills that influence the health of village women and children.


The Stung Treng TEA programme is very relevant to the National Safe MotherhoodPlan 2001-2005, which states that training of TBAs, is a key objective to strengthenand extend service delivery at community level. The original program design of theTEA programme was appropriate considering the health situation in Stung Treng in1994, a remote province with difficult access to under staffed government healthfacilities and with 90% of deliveries done by TBAs and lack of trained health centremidwives in remote areas.

Some programme adaptation has occurred over the last few years to meet the needsof a changing environment Positive changes have been the increased role of theprovincial health department staff in the planning and management of the pro-gramme; the percentage of TBA supplies provided by the government; follow up ofthe TBAs work at village level; the adaptation of the TBA training curriculumfollowing the latest developments in MCH services and the increased participationof health centre midwives in TBA programme activities. After several years ofimplementation experience, the time is right to consider further decentralisation ofthe TBA programme. More responsibility needs to be assumed by health centrelevel staff, as their ability to fulfil this role is strengthened.

Other relevant findings:

(a) Lack of confidence by village people in the quality of services offered at thereferral hospital. Village people related incidences of cases that had beentransferred to the hospital that had received inadequate care.

(b) Lack of information and awareness of village people about common healthproblems and their prevention, when and how to introduce appropriatecomplementary foods, the importance of exclusive breast feeding, the benefitsthat ante natal care could offer, danger signs in pregnancy labour anddelivery and methods and availability of services for family spacing.

(c) Changing perceptions of village people about the role of the TBA. In TBAsdiscussion groups a common problem raised by TBAs were the changes inperceptions of village people about the TBA role. There seemed to bemisunderstanding among some village people that TBAs who receivedre-training and equipment were now receiving a salary. This createdproblems of jealousy and also occasionally TBAs were not paid for theirservices. There was also the perception of some village women that 'trainedTBAs were more expensive than untrained TBAs,1 although TBAsinterviewed denied this was the case.

(d) Home delivery kit dilemma

TBAs interviewed were resistant to the use of the individual home delivery kit thatis available to sell to mothers. This kit provides among other things a blade to cutthe cord.

SO TVarHHnnal Rirth Anpnrianf- Prncrrammp Shiner Trpncr Cambodia

Most TBAs said they had become proficient in the use of scissors and cord clampsand would find itdifficult to take a "backward step1 to use a blade and cord ties. Other reasonsmentioned for not accepting the kit were that poor mothers would be unable toafford it, and that many village women mentioned that there was no need to spend3,000 Riel on a kit when the TBA already had adequate supplies and equipment.The evaluator's impression was that even if mothers start to buy the kits, TBAswould continue to use scissors and clamps, as they have done for several years. Allthe TBA equipment examined was in excellent condition and very well kept.TBAs and mothers all reported that TBAs boiled the equipment before use.Investigations in 2002 to identify possible tetanus cases proved negative, with notetanus cases found in the cases of neonatal deaths investigated, despite the lowuptake of Tetanus 2 (15%). Despite the constraint of TBAs being reluctant to acceptand use the individual TBA kit the evaluators feel it is beneficial and should bemore actively and aggressively promoted both among TBAs and villagecommunities. It provides many useful supplies especially the plastic sheet andgloves, which should not be reused on other deliveries. As /men in the villagesdecide how money should be spent they should also be involved in promotioncampaigns. Women and their families, not only TBAs need to be aware of what aclean delivery means and the many benefits of having individual delivery kits.It might be possible to promote the kit by providing it free for pregnant womenwho have attended three antenatal visits with a government midwife and receivedtwo tetanus injections. /

YWAM staff mentioned the idea of developing a new delivery kit to sell cheaply tomothers. This kit would have extra compresses and plastic sheet and also a cleancotton cloth to wrap the new born baby in. The idea of a new cloth to wrap the newborn baby is good, but it is recommended that rather than YWAM developing yetanother version of a delivery kit, they should strongly advocate at National MCHlevel to modify the present individual delivery kit.

(e) YWAM identity of the TBA programme.

The TBA's identify the TBA programme as a YWAM programme even thoughYWAM have gradually withdrawn support and now provide mostly managementand financial support at provincial health department level. This is understandableconsidering YWAM's long history of involvement in the TBA programme. It wouldbe better if TBA's recognised the TBA programme as a government approved activi-ty, supported by some external funds. This message could bere-enforced by village people seeing that health centre staff follow up and workedin close collaboration with TBAs at village level, and also by health centre andprovincial health centre staff explaining to influential village leaders the role of thevarious partners in the TBA programme, and how TBA activities are onecomponent of the province wide MCH programme. If the relationship of the TBAprogramme to government health activities was clarified it might also help toreduce the misconceptions and rumours about TBA's receiving salaries.


(f) Sesan TBA's

It was unfortunate that MSF did not continue the support of ongoing TBA activitieswhen they worked in Sesan district. It meant the TBAs of Sesan were left withoutfollow up training or support for three years. It also caused a gap in the MCHstatistics from 1997-2000. Further complexities were created when YWAM did notassume responsibility for support of the Sesan TBAs when MSF left in 2000.

The present situation concerning the TBA programme in Stung Treng is thatYWAM supports financially and technically, four of the five districts in Stung Trengprovince and UNICEF has assumed financial support for the TBA programme inSesan ~^^district since mid 2001.The goal and objectives of both YWAM and UNICEF,concerning TBA activities are almost identical. Both YWAM and UNICEF see theTBA programme as an interim measure, until there are enough trained governmentmidwives in rural areas. The difference is with the perceptions of the two agencies \in how soon this can become a reality.

Another difference is with the supply of equipment to TBAs. In Sesan district theTBAs do not receive re-supply of scissors and cord clamps but is encouraged to usethe individual delivery kits with the disposable blades. It is reported that TBAs inthis district are still using the scissors they acquired when they were trained severalyears ago without the opportunity to purchase new ones, as in the YWAMsupported districts. YWAM continues to provide re-embursement of travel costs forTBAs who transfer women to the referral hospital.

The situation of two separate but similar policies for the TBA programme in onesmall province needs to rectified. The Stung Treng provincial health departmentfinds itself in the middle, with the delicate task of facilitating two donors.Ultimately responsible for implementing health programmes in the province is PHDresponsibility. PHD need to take the lead with the participation of key stakeholdersin the TBA programme, the development of a TBA policy and strategy that isappropriate for Stung Treng.

It is certain that PHD will need continuing financial and some technical support forthe TBA programme for several years to come. It would be easier for them tomanage if the financial and technical support for the whole TBA programme camefrom the same source. This would also enable the best utilisation of availablehuman resources, ensure continuity and avoid duplication and communicationproblems.


Village people

<;•}

Village people

5. CONCLUSIONS

The key question the evaluation sought to answer was:

" Has the TEA programme in Stung Treng contributed to strengthening of theMCH services in the province and improving the health of mothers and infants?"

The evaluators found that the provincial health department, health centre staff,YWAM and more recently UNICEF, through a strong collaborative effort insupporting the TEA programme have achieved an significant improvement in thequality of MCH services available to women in remote villages of Stung Treng.

With difficult access and often impossible travel conditions, 394 TBAs (with medianage 45yrs- 55yrs) in villages of all five districts have been trained and-are^active.They are provided regular follow up support and supervision. Equipment ancsupplies are re- supplied in a timely manner and information from village levelabout births, deaths and transfers are reported at regular intervals. The evaluationteam found an impressive level of knowledge and skills among TBAs observed andinterviewed, concerning how to conduct a clean and safe delivery and identifyhigh-risk women that need referral. TBAs reported several changes that had takenplace in TEA practices such as cleaning and sterilising of equipment, wearing ofgloves, putting of gentian violet on the cord, waiting for signs of placentalseparation before attempting delivery of the placentas and drying the baby andputting to the mothers breast instead of giving an immediate bath. Interviews anddiscussion with village women who had delivered with a TBA confirmed thesepractices.

Statistics and follow up investigation show that neonatal tetanus is probably not aproblem in Stung Treng.This is an important indicator of success for the TBAprogram, that has very likely contributed to this achievement by increasingthe number of clean deliveries conducted by TBAs and achieving appropriate careof the cord. Tetanus toxoid vaccination rates remain too low to contribute towardsthis achievement (15% T2 in 2002).

Stillbirths and neonatal deaths remain high but there are some signs ofimprovement. As there is no record of the age at death of the neonate it isimpossible to estimate perinatal mortality rates. Neonatal mortality statistics show adecrease from 4.0% of total reported deliveries in 2000 to 3.7% in 2001 and 3.2% in2002. There is very little reporting or recording of infant or under five mortality soit's impossible to assess if the health of under five children has improved. Fromcomments during village level discussions it appears that in both groups deathsrates are high.


Immunisation coverage rates for under one year old children have improved in thelast three years (66% of children receiving DTC 3 and Polio 3 in 2002 as compare to53% in 2000) These improvement have occurred due to increased outreach activitiesby health centre staff. Antenatal care with government midwives has increaseddramatically in 2002 (from 28% in 2001 to 58% in 2002) and this is attributed toincreased outreach activities by health centre midwives and the participation ofTBAs during outreach.

The TEA's are to be commended for their ever increasing involvement intransferring high risk women. The number of transfers of emergency obstetric casesto the referral hospital is increasing yearly (49 in 2000, 57 in 2001 and 74 in 2002) Areview of the cases transferred in 2002 showed that TBAs competently andsometimes under very difficult conditions accompanied women to the hospital.TBAs often stayed with the woman several days until her condition stabilised. Fromthe review of transferred cases it is also clear that without the prompt action of theTB A concerned some of these women may have been added to the maternalmortality statistics.

Observed difference in maternal death statistics in a small population province suchas Stung Treng is not a good indicator by which to measure the success of an MCHprogramme. The maternal mortality statistics are too few in number and differencesobserved may be due to changes in the reporting system or may be subject to awide random variation resulting from a small number of events. There may be largechanges in the burden of morbidities before this is reflected in the MMR.

It is positive that maternal death audits have been conducted since 2002 and thatTBAs in Stung Treng are active in reporting of maternal deaths. In 2002 TBAsreported the majority of maternal deaths that were investigated. In a review of thecases that TBAs were involved in it appears that the TBAs responded appropriatelyaccording to the situation. From discussions and observations at village level itseems that everyone is aware when a maternal death occurs and it is likely that allmaternal deaths are reported.

The programme is judged as very cost effective at 30 US $ per TEA per year (2.50US$ per month per TEA), 32% less expensive compared to a TEA programmeimplemented in a more easily accessible province. Seventy four percent of the StungTreng TEA budget is spent on capacity building of both TBAs and government staff.

Issues that need to be addressed:

1. Some inconsistency and lack of clarity in implementation of the TEAprogramme, created by having two agencies, with some differences inphilosophy providing financial and technical support to the programme fordifferent districts.


2. The need for a comprehensive provincial TBA policy.Although the Bridge ofFriendship TBA program has a clear TBA policy, there are some differencesbetween it and the UNICEF policy concerning TBAs in Sesan district.

3. Dependence on three provincial TBA managers/ trainers to implement themajority of TBA program activities.

4. A lack of confidence by village people in the quality of services offered at thereferral hospital.

5. Sole dependence on the village TBA concerning the emergency referralsystem and the dependence on YWAM to re-emburse travel expenses.

6. Some weakness in the TBA reporting system because of the prolongedinterval between conducting of activities to the recording and reporting ofthem.

7. The need for improvement of TBA skills concerning supporting theestablishment of exclusive breast feeding, identification and referral fortreatment of anaemia both antenatal and post natal, identification of high riskinfants and mother in the post natal period, and health education skills topromote antenatal care and tetanus vaccine, exclusive breast feeding and theintroduction of appropriate timely complementary feeds.

8. The existing system for documenting of TBA activities and statistics needs tobe strengthened and standardised.

9. Village women's lack of knowledge concerning prevention of commondiseases, family spacing methods, the benefits of antenatal care, importanceof iron supplementation, the benefits of exclusive breast feeding, theintroduction of appropriate and timely supplementary feeds and dangersigns during pregnancy, labour and in the postnatal period.

10. The reluctance of TBAs to promote or use the individual birth kits. The kitshave some important benefits including the active participation of women inensuring their delivery is as clean as possible, and guaranteeing that eachmother will have her own disposable gloves and plastic sheets.

11. The refusal of women in remote rural villages, who are identified as high riskduring their pregnancy to go to the referral centre.


6. RECOMMENDATIONS IN ORDER OFIMPORTANCE

Recommendations to PHD, UNICEF and YWAM

1. A decision should be made as soon as possible concerning the future of thepresently divided Stung Treng TBA programme. PHD in consultation withYWAM and UNICEF should decide which agency is in the best position tooffer long term (at least two years) continuing technical and financial supportto the TBA programme in all five districts. The evaluator's assessment is thatYWAM because of its history with the TBA programme in the province, itsexperienced staff, its present support to four of the five districts and its longterm commitment to the Stung Treng health programmes is in the bestposition to do this. If all parties involved are in agreement and YWAM iswilling and able to agree to this, they should re-assume responsibility for thesupport of TBAs in Sesan district. If funds are not immediately available todo this UNICEF should be requested to provide funds to YWAM as aninterim measure until further funds can be identified.

2. PHD should play the lead in organising and conducting a workshopinvolving key stakeholders in the TBA programme to develop acomprehensive TBA policy for the province, addressing issues such asmanagement structure at each level of the programme, a standard TBA jobdescription, relationships, role and responsibility of health centre and provincial level staff concerning the TBA programme, transfer of emergency cases,TBA perdiem levels and the resupply of TBA supplies and equipment.

3. An extra full time midwife should be assigned to work at provincial MCHlevel.The present MCH staffing level is too low to adequately cover theamount of activities currently being implemented.

Recommendation to TBA programme managers at provincial level.

4. With the support of the YWAM TBA programme technical advisor, and incollaboration with health centre staff a plan should be developed to ensurethe smooth and gradual decentralisation of TBA programme activities tohealth centre level. It is recommended that ultimately health centre staffshould be responsible for all TBA training, follow-up meetings andmonitoring of TBA activities at village level. Provincial level staff shouldincrease their involvement in on the job training of health centre staff andco-ordination and monitoring of MCH activities at health centre level.

Recommendation to PHD and UNICEF

5. PHD with the support of UNICEF and possibly an external consultant shouldconduct an assessment at the referral hospital, addressing the urgent need forimprovement of provision of quality emergency obstetric care. Funds shouldbe identified to follow through on subsequent recommendations.



6. The provincial TEA programme managers should work with health centrestaff, health centre management committees, village developmentcommittees, TBA's and other organisations and groups working at villagelevel to develop and strengthen the referral system at village level.The TBAs presently lack support concerning organising emergency transfers.They require strong backup support from village leaders or committees whenthe decision is made to make an emergency transfer. The development of arealistic achievable village emergency transfer fund is also important. Thisemergency fund scheme should be started on a pilot basis in a few villageswho have active motivated development committees. Following a review oflesson learned the programme could be adapted and introduced to othervillages.

7. The accuracy of the TEA reporting system would be improved by morefrequent collection of TEA statistics. Statistics may be more reliable if theywere collected monthly, as the TBAs would then be encouraged to recordtheir activities closer to the time they were conducted. The logistics involvedin doing this will vary according to the area and the human resources available. It might be possible that VDC members could collect monthly statisticsfrom TBAs in their respective villages and send them to the health centrelevel.

8. Future refresher training for TBAs should concentrate on improving the skillsof TBA in the areas identified during the evaluation. These are: thepromotion and support to mothers in the establishment of exclusive breastfeeding, identification and referral for treatment of anaemic women bothantenatally and post natally; the prompt identification of problems in infantsand mother in the post natal period; improved health education skills topromote antenatal care, tetanus vaccine, exclusive breast feeding and theintroduction of appropriate timely complementary feeding. If funding isavailable it would be helpful to develop and pre-test a TBA flip chart withcolourful pictures, key messages and minimal script, that TBAs could use asa tool when discussing health topics with village women.

9. The system of documentation and recording of TBA programme activitiesand statistics should be reviewed. If possible a user friendly centralisedaccessible data base should be designed that standardises how importantTBA programme indicators are recorded. The TBA programme provides a lotof important information. If the data is easy to access, it can help identify andrespond to problems promptly and facilitate easier monitoring of theprogramme.


10. There is a need for effective IEC materials and health education sessions forvillage people concerning important health topics and prevention commondiseases. Important health education topics for women and their partners areantenatal care, danger signs during pregnancy labour and delivery, familyspacing, breast feeding, nutrition, immunisations and introduction ofappropriate timely complementary feeding. The TBA cannot be expected totake complete responsibility for village health education, human resources atvillage level should be identified that can participate in conducting theseactivities.

11. Despite the reluctance of TBAs to promote or use the individual home birthkit, the kit has some important benefits including the active participation ofwomen in planning for their delivery and ensuring it is as clean as possible.The kit also guarantees that each mother will have her own disposable glovesand plastic sheets. The kit should be continued to be promoted at villagelevel. Women might be more enthusiastic about buying it if they understoodbetter the advantages the kit offers. A flexible approach should be used withTBAs who have been proficiently and safely using scissors and clamps formany years. They should be encouraged to use as many of the disposableitems in the kits as possible.

12. The TBA programme managers should strongly advocate that the NationalMCH Centre review the adequacy of the materials supplied in the individualhome birth kit and make recommendations for adding additional materialsaccording to field experiences in Stung Treng.

Recommendation to PHD, UNICEF and YWAM

13. The possibility of 'waiting homes' at provincial level for high-risk pregnantwomen should be explored. They have found to be successful in severalother countries, especially in countries that developed simple realistic planswith active involvement of communities. Stung Treng transfer statistics showthat very few high-risk women are electively transferred during pregnancy.Most are transferred when they are already an obstetric emergency. Thisreport has mentioned some of the reasons for this situation (WHO, 2002).


7.GENERAL CONCLUSIONS

The TBA programme in Stung Treng has achieved commendable success inincreasing women's' access to clean deliveries, strengthening the skills of traditionalbirth attendants and decreasing the use of harmful practices. The main reason forthe success is the close follow-up support and continuing training opportunities thathave been provided to TBAs. Considering the geography of Stung Treng this is nosmall achievement.

Despite this success much remains to be done to improve women and children'shealth. The MoH' strengthening of district health services' is progressing.The quality of health care available in rural areas is slowly improving, but it isunrealistic to expect that the MoH objective to provide at least one competentfemale health provider trained in Basic EmOC at each health centre in rural areas,and to increase the provision of quality essential maternity care at this level will beachieved within the next few years. Until this happens and women have increasedaccess to and confidence in the available health services, the TBAs will continue tofill the gap, providing an important and affordable service to rural women.

Provincial health department staff have confidently taken over the management ofthe TBA programme. They strongly believe in the value of the TBA and theimportance of her role in providing a link between the community and governmenthealth services. It is important to continue to build and strengthen the importantlink between the TBAs and the health centres. Due to budget restraints and lack ofhuman resources PHD will require future financial and technical support for theTBA programme.


Village children

f" u«^4 i ' 6?

i^ii-iCiivli-I^itiviiXr^K,

APPENDICES

A, Programme of evaluation, list of people interviewed/ sites visited

The Evaluation Programme

Date Location Persons interviewed

Mon 17th March 200312.00 - Arrived inStung Treng

14.00

16.30

Provincial HealthDepartment

Provincial HealthDepartment

DrHengnur -Chief of PHDMs Boo Vatha -Deputy chief of PHDMs San Channy -vice chief of MCH

DrKeoParin UNTCEF

Tues 18th 8.00

14.00

16.00

17.00

Stung Treng HealthCentre

Obstetric wardProvincial referral hospital

YWAM office

TBA trainers

Choung Ravy- Secondary midwifeNoi Sarim - Secondary midwife

Tang Som - Secondary midwifeChief of ward

Kun Theary - Secondary midwifeBua Camnor- Primary midwifeKim Da Neth - Secondary midwife

Por Puthany - TBA monitorChan Talorn - TBA monitor

San Channy - vice chief of MCHNou Chan Hinge- provincial MCHYimSokem- EPI chief

Weds 19th 9.00

11.00

14.00

15.00

16.30

17.30

Jamgar Ler Health CentreThala Boriwiwat District

Jamgar Ler health centre

Jamgar Ler village

Jamgar Ler

Romdeang villageJamgar Ler commune

Romdeang villageJamgar Ler commune

Seang Kim Sorn -Chief of Health CentreBen Sopheap - Primary midwife

TBA discussion groupHoy SengPich SothMohem LornKeo HunNin PenRare Rah

Village women's discussion group

Village development committee

Village women's discussion group

Village leader interview


ii^^irAvc*vi^>X'C-iw-iX- :. 3ii . ^^>-<vOC' i<or-i--i-j -:-i<.Ti-'X>.ir:-

Thursday 20th 8.30

930

14.00

15.30

Friday 21st 8.00

10.00

14.00

16.00

17.30

Saturday 22nd 8.00

13.00

15.30

16.00

16.30

17.30

Jamgar Ler village

Romdeang village

Jamgar Ler village

Return to Stung Treng

Bajong villageStung Treng

Preah Bat villageStung Treng

Travel to Siem BokDistrict

Srey Greysang HC

Travel to DambongClaVillage

Omreah Village

Dabong Kla village

Dabong Qa village

Koh Jerum village

Return to Stung Treng OTreah village

Thala Borivat district

Pregnant women interview

TBA discussionCam DengKim Santania

Interview family of woman who diedfrom PPH

Interview with HC midwives

Interview with village development com-mittee leader Overnight in Dambong Clavillage

TBA skills observation andinterviewsVisit post-natal /ante natalwomen

TBA discussion group

Pregnant woman interviews

Pregnant women interview

TBA individual interviews

Village women interview


Evaluation Activities

Location

Stung Treng

Bajong village

Preah Bat village

ThalborivatDistrictJamgar Leucommune

Jamgar Ler vilage

Jamgar Ler village

Romdeang villageJamgar Lercommune

Romdeang village

Romdeang Village

Thalaborivatcommune

Otreah village

Otreah village

Person interviewed

Provincial health departmentUNICEF Advisor

Provincial MCH trainers

TEA monitors

Health centre midwives

Provincial obstetric ward staff

YWAM

TBA's

Family of recent maternaldeath

Health centre chiefHealth centre midwife

TBA's

TBA's

Village Women

Village developmentCommittee

Village women

Village leader

Pregnant/ newly deliveredwomen

TBAs

Village women

Type of interview

Semi-structured

Semi -structured

Semi structured

Semi-structured

Semi-structured

Semi-structured

Semi-structured

Discussion group

Semi-structuredinterview

Semi structured inter-view

Focus group discussion

Individual interviews

Discussion group

Semi structuredinterview

Discussion group

Semi-structuredinterview




Total no of people

3

1

3

2

2

5

1

2

2

2

6

2

15

4

12 women

1

5

2

2

2


Siem Bok districtSre Krasang

Tbong Cla village

Omreah village

Omreah village

Omreah village

Kok Jerum village

Tbong Cla village

Tbong Cla

Totals3 districts8 villages

Health centre midwives

Village developmentcommittee chief

TBAs

Village women

TBAs

TBAs

TBA's

Village women

Discussiongrouplndividualinterview



Practical skillsobservations

Practical skillsobservation

Discussion group


3 TBA discussiongroup (14 TBAs)

7 individual TBAinterview

5 TBA practical skills

19 individual villagewomen interviews

2 village leaderinterviews

2 village womendiscussion groups

2 VDC interviews

1 family interviewfollowing a maternaldeath

4

1

3

4

3

2

6

4

101 people involvedininterviews anddiscussions


B. Job Description of TEA in Stung Treng

1. Give ante-natal care to village women starting early in pregnancy and at leasttwo times during pregnancy by:Physical exam of the abdomenChecking for anaemiaGiving ferrous folk 30 tablets per monthChecking for oedema and asking relevant questions concerning the mothershealth

2. To give advice to the mother aboutDiet throughout motherhoodTetanus injection during pregnancyChild immunisationBirthspacingTreatment of diarrhoea with ORSPersonal hygieneIdentifying STD's and sicknesses needing referral

3. To carry out a clean and safe delivery according to her training4. To visit at least five times post delivery to check on mother and baby5. To promote breast feeding within one hour of birth and then exclusive

breastfeeding there after6. To give advice to the mothers about when to start weaning, what foods to

give while continuing to breast feed there after7. To detect deviations from normal through out pregnancy, labour and post

natally and refer at risk women without delay8. To make reports of births and deaths and attend Smonthly meetings

Reports to village chief and VDC, commune nurse

C. TEA kit resupply costs

ITEM

Forceps

Brush

Soap

Plastic

Scissors

Nail clippers

Cost

3000 Riel

500 Riel

Free

500 Riel

3,000 Riel

1,000 Riel


D. MCH data

Chart 5 Stung Treng Vaccinations Under 1yrs children 2000-2002

Year

2000

2001

2002

Target Children

2,887

2,965

3,027

BCGNo

2,270

2,263

2,510

%

79%76%

83%

Polio 3

1,527

1,848

2,003

%

53%

62%

66%

CTD3

1,5271,848

2,003

%

53%62%

66%

Rouvac

1,449

1,531

1,858

%

50%

52%

61%

Chart 6 MCH activities Stung Treng 2000

Total pop.EsL pregnant1st ante-natal2nd ante-natalTetanus 2Del trainedDel TEANo referralMaternal deathsStillbirthsNeonatal deathsBS.

Thala

24,1348706039

7641611340na

%

6.9%45%

Siembok

10,047422344

2758236na

%

8.1%0.9%

Siempang

13,4745687417

53214027na

%

13.0%3.0%

Sesan

11,10944816058

700

na

%

35.7%12.9%

Stung Treng

24,134899382207

35517125na

%

42.5%23.0%

HCmid

Govstaff

Provhosp

Totals

82,8983,2077103256012811,92649188781,731

%

11.45%5.10%12.00%12.70%87.20%

39 per 115.00%

XX) live births


Total pop.Est pregnant1st ante-natal2nd ante-natalTetanus 2Del trainedDel. TEANo referralMaternal deathsStill birthsNeonatal deathsBS.

Thala

24,13491011782

1,0043931229565

%

129%9.0%

18.8%

Siembok

10,0474328234

32110038317

%

19fl%7.9%

20.4%

Siempang

13,47458212051

54801942269

%

20.6%8.8%

13.1%

Sesan

11,10945911960

7

217

7

373

%

25.9%13.1%

23.1%

Stung Treng

24,134615328234

3536024221

<V/•

533%38.0%

10.2%

HCmid

6

Govstaff

52

Provhosp

317

Totals

82,8982,9987664616393752,22657526831,883

%

19.00%14.40%85.50%

33.2 per15.46%

1000 live births



Total pop.Est. pregnant1st ante-natal2nd ante-natalTetanus 2Del trainedDel. TEANo referralMaternal deathsStillbirthsNeonatal deathsB5.

Thala

24,13483336182

8982631228114

%

433%9.8%

173%

Siembok

10,0473%258145

3652136778

%

65.2%36.6%

22.0%

Siempang

13,474533237 -74 f

60760318134

%

44.5%13.9%

17.0%

Sesan

11,109420305157

32483511314

%

72.6%37.4%

30.0%

Stung Treng

24,134564467370

36713028148

%

82.8%65.6%

11.5%

HCmid

13

Govstaff

41

Provhosp

160

1611

Totals

82,8982,7461,6288281,8832142,561741034832,455

%

15.46%

31 per 119.69%

XX) live births


E. Suggestions for additional TEA programme indicators

These indicators could be monitored over a three year period during outreach workat village level,TBA meetings,village discussion groups and follow-up observatiosof TBA work at village level.

1. No of villages with established emergency transfer systems that havesuccessfully organised emergency transfers.

2. Percentage of trained TBAs attending TBA meetings.

3. Outcome of mother and baby following emergency transfers fromvillage level.

4. No of follow up visits of TBAs at village level by health centre staff.

5. No of pregnant women recieving 90 tablets of iron/folate supplement.

6. No of home delivery kits sold by TBAs at village level.

7. No of newly delivered mothers recieving post natal care.

8. No of women referred for antenatal care by TBA to government facility.

9. Participation of TBA in health centre staff outreach activities at village level.

10. Increased awareness of village women about antenatal care,exclusive breast feeding,iron supplementation,tetanus vaccination,familyspacing, danger signs during pregnancy,labour and delivery.

12. No of TBA trainings conducted by health centre midwives

13. No fo TBA purchasing re-supply of equipment such as scissors and cordclamps.

14 No of referrals of high risk women by TBAs before they become an obstetricemergency.

/O Traditional Birth Attendant Programme Stung Treng, Cambodia

F. References:

Bulatoa, RA; RossJA (2002) Rating Maternal and Neonatal Health Services inDeveloping countries. Bulletin of the World Health Organisation. Downloaded fromWHO library database 22nd October 2002.

Ministry of Health (2001) National Safe Motherhood 5 Year Action Plan2001-2005.MOH, Phnom Penh, Cambodia.

National Institute of Statistics (2000) Cambodia Demographic and Health Survey2000. Ministry of Planning, Phnom Penh, Cambodia.

National Maternal Child Health Centre (2001) A TEA case study in three provincesin Cambodia: Sway Rieng, Rattanakiri and Kampong Chhnang Provinces.Unpublished, NMCHC, Phnom Penh Cambodia.

National Maternal and Child Health Centre (2002).Evaluation report of TEA kitdistribution. MOH, Phnom Penh, Cambodia.

Parco, KJacobs, B. (2000) Knowledge, Attitudes and Practices of Traditional BirthAttendants in Maung Russey, Cambodia: scope and ways for improvement.MOVIMONDO, Battambang, Cambodia.

Provincial Health Department (2000, 2001, 2002). Annual Provincial HealthProgramme Activity Reports, Stung Treng, Cambodia.

RACHA (2001) RACHA's TEA Project. Unpublished report, Phnom Penh,Cambodia.

Rooney, J. (1999) A Review of the Safer Motherhood Pilots and Maternal MortalityAudits in the Operational Districts of Prey Veng and Kompong Chhnang,Cambodia. Consultancy report, UNFPA.Phnom Penh, Cambodia.

Scott, J. (2003) Health Care Development in Stung Treng Province, Cambodia.Annual Report. YWAM, Phnom Penh.

UN AIDS (2001) Country Profile: The HIV/AIDS/STI situation and the nationalresponse in Cambodia. UN AIDS, Phnom Penh.

White, P.M. (1997) Crossing the River: A Study of Khmer Women's Beliefs andPractices During Pregnancy, Birth and Postpartum. Unpublished thesis fordoctorate in public health.

WHO (1992) Traditional Birth Attendants: A Joint WHO/UNFPA/UNICEFStatement. WHO, Geneva Switzerland.


WHO (2000) Reproductive Health Indicators for Global Monitoring. Report of theSecond Interagency Meeting. Department of Reproductive Health and Research.WHO, Geneva.

WHO (2002) Maternity Waiting Homes: A review of Experiences. Downloadedfrom the WHO web site publications.

World Bank (2000) Cambodia Country Assistance Strategy: Building theFoundations for Sustainable Development and Poverty Reduction. East Asia andPacific Regional Office, Bangkok.


.?-~ w MSSSfelirattbaS

?m u.Wl

m

-•41

•-f-fti-/^

KTfcv****•

Jv,^w-Ui;*^:-"::-; --fifc>--S>-,v.-" --• •

.w-- *sS-i".- -•-••^•ffcift^t^isi^f ,r...- •^T*Xi.".;5i ;-J£;.'- r. - • •

-,••,,-.-., -: >a*^p~l«fgW«!Mte ,r 7 -:i; ..* -,i , •-- --..„- i-^«-.'.^^K^5' ' ^JES*s>t**»^1^ri"~"-^V-- •;-'i-^*-."""• • ' ' " .k-.-:ifc:i~ ;™;3»ite.-:ir:ai!.r5-:-.r'.r/;u->r-"'• - >; - .-- -.;--*-.^^.^-J^rriS^^r^:^-^.^,-,',.j :->—"-x- ^.,,

- . . - - -:::--'!A^±:'3x?&z£j^^^;i;?-^.:i: .-.'..• -:---.-. f,r.-_. •^^^^^sBiSl™;-; ^1!*^*- - ---.- ,: vj Vi ^s^sfe----^^^-----^^;-,;,«:•,. - • ' •;™;.S r3SigS| i! 5K jpcgS|£?;i".-?;'>

Exf /D EVU03.06.02/ CBD 2003/012

Evaluation of a Traditional BirthAttendant Programme: StungTreng Province, Cambodia

... -~---4,i;;i;Ljf,i3i£3i=K:._- - -.v .; •.vrri'.'^.v.jS3-#ftj*£3Msji?-;;?:;si'%-j:rbKSSS;ra!s""

-;•:--< .>*Vi7:.\iv-*:*g3-':saL,_•:""!isS3||iS. -,-:.-, .' --.-J •- .-"-H-.V^»V*' *—

. .--,-::^-r^r^-^«:si?a!a_'-' :.- -;:- P::-77a'S..3*'fe

:;v-^^^g. .,..„—i., .L.x;f-i- «4^c i*»ii».

- *•%'->. Js ' 'r"^^Ci iJ

--^-^mm^mAlternate UNICEF Inventory Label

unM Illlllll RCF0008A9F

Folder CF-RAI-USAA-DB01 -EV-2004-00410Owner Location Evaluation Office, UNICEF NY-HQ = 5128u~™ i ~^t;nn rP.RAP.II<;AA.nRni-9nnd-n2fi27 fin Contained

April - UNICEF...programs in Cambodia and in other countries was conducted. This was followed by...

Documents

Transcript of April - UNICEF...programs in Cambodia and in other countries was conducted. This was followed by...