Background

 

Maternal mortality is a major health challenge. Levels and trends in maternal mortality provide an important marker of societal health and development and of health system performance. A reduction in the maternal mortality ratio (MMR) was included as one of six health-related Millennium Development Goals (MDGs). Reduction of MMR is a public health priority in Bangladesh. 

 

In 2001, as part of the determined focus on MMR, the first Bangladesh Maternal Mortality and Health Care Survey (BMMS) was conducted in Bangladesh, to gain a better understanding of the MMR level and proportion of maternal deaths due to different causes. At that time the MMR was estimated at 322 maternal deaths per 100,000 live births.

 

The second BMMS 2010 indicated a substantial reduction in the MMR, an annual rate of decrease of 5.6 percent. The MMR fell from 322 deaths per 100,000 live births (95% CI 253-391) in 1998-2001 to 194 deaths per 100,000 live births (95% CI 149-238) in 2007-2010, indicating a degree of success in the health sector. But Bangladesh is committed to achieve the goal of the HPNSDP to reduce MMR to less than 143 deaths per 100,000 live births by 2016. 

 

The objective of the 2016 BMMS is to monitor progress in maternal health and use of maternal health services in Bangladesh and assist policy makers and program managers to strengthen and refocus the national effort to avert maternal deaths. 

 

Survey objectives: 

 

The specific objectives of this survey are as follows:

 

1. To estimate the national level of MMR for the period 2013-15 in Bangladesh; 
2. To identify the specific causes of maternal and non-maternal deaths in adult women; 
3. To assess the experience of and care seeking for maternal complications and changes in care seeking pattern during 2013-2015; 
4. To estimate under-five mortality, child mortality, infant mortality, neonatal and post neonatal mortality rates by division; 
5. To assess the level and use of family planning, antenatal care, post natal care, skilled birth attendant at delivery in 2013-2015; 
6. To assess newborn health care practices;  
7. To measure the level of selected maternal morbidities; and 
8. To provide district level socio-economic, demographic and health care utilization information.

 

Survey method

 

Sample

 

There were 2,316 survey clusters from the divisions of Dhaka, Chittagong and Sylhet. In each selected cluster, 65 households are randomly selected for interview.  All households in these clusters is received a short questionnaire administered at the household level and to ever-married women of reproductive age (13-49) residing in the household. The members of 23 (randomly selected) households in each cluster are also received a supplemental questionnaire. This implies a total sample size of 150,540 households drawn from the 2,316 clusters, with 97,272 households covered with the short questionnaire and 53,268 with the both long and short questionnaires.

 

The sample frame is based on the 2011 Bangladesh census frame. The design is multi-stage with wards (urban areas) and unions (rural and other urban areas) selected in the first stage, mohallas (urban) and mouzas (rural and other urban) selected at the second stage, and then clusters within mohallas and mouzas selected at the third stage. The sample design has provided by MEASURE Evaluation with assistance from the Bangladesh Bureau of Statistics (BBS). The listing and mapping work was supervised by NIPORT, ICDDR, B and MEASURE Evaluation.

 

Tool

 

There was a listing form to take complete census of the households in the cluster through house-to-house visits. Also there were six types of questionnaires; 1) a household questionnaire, (2) a short questionnaire (including only core questions ) for individual ever-married women age 13-49, (3) a long questionnaire (including both the core questions and supplemental questionnaire on maternal and child health, and family planning) for individual ever married women age 13-49 (4) Verbal autopsy questionnaire, (5)Maternal morbidity questionnaire to collect information for maternal morbidity (6) a community questionnaire.

 

Tasks carried out by Mitra and Associates

 

Preparatory works

 

The specific tasks for preparation of questionnaires, manuals and program for smart phone for face to face survey include:

• Translation and back translation of the questionnaires and manuals in Bengali and making necessary changes after pre-test talking with the technical team. 

• Preparation of a) household listing forms and a brief listing manual, b) Interviewers’ and Supervisors’  Manuals,  c) Interviewer and Supervisor Assignment Sheets for management of fieldwork, d) Monitoring and data quality check instrument (including using of smart phone by supervisors to submit key data points after completing each cluster).

1. 1. A standard, user friendly, smooth application for BMMS monitoring was built.
   2. The application was tested and upgraded by trial and error method.
   3. For the security of the application, a strong anti-virus was used in each smart phone.

• For another security measure, a strong app locker was also used in each smart phone.

 

Pre-testing

 

Pre-testing of the questionnaires in two areas (one urban and one rural) outside the clusters drawn in the sample had been done under the supervision and guidance of NIPORT, ICDDR,B, and MEASURE Evaluation. The pretesting was for checking the translation, consistency and integrity of the questionnaire.

 

Training

 

Two separate training were carried out including training for household listing and household data collection. Prior to the household data collection training, a 5 days long training were carried out for the listing teams followed by a 17 days training for the household data collection teams. Prior to start of the training of the data collection teams, a training of the trainers (ToT) was carried out by NIPORT with technical support from the MEASURE Evaluation/USA for 7 days including the first 3 for training on questionnaire , second 2 days for video shooting on questionnaire and third 2 days for video shooting on verbal autopsy. The videos were used as a guideline for conducting the main training for household teams.

 

Data collection and quality control

 

The data collection was carried out by deploying 42 interviewing teams. Each interviewing teams were consisted of one male supervisor, one female editor, 5 female interviews and 1 male logistical assistant. In addition, 10 quality control teams were also deployed. The total fieldwork took place for 6 months. MEASURE Evaluation provided us programmes to generate some field check tables on some key variables to assess the performance of the teams. The field check tables were used to asses any variation with the overall data as well as with the national level data. Female editors edit the questionnaire at field and make necessary correction and quality control checking was undertaken in both presence and absence of the interviewing.

 

In addition, smart phones with a standard application were used in monitoring purpose. In this regard, one smartphone per each team was provided. The supervisors were instructed to send some information on some key variables using the smart phone after completing each cluster. Once, the information was sent to the server, that was monitored centrally and the teams were warned if any inconsistency was found.

 

Equipment and logistics

 

1. 1. Mitra office arranged a secured storage place for the smart phones used in the survey operation. 
   2. Mitra office ensured proper maintenance and functionality of the smart phones.
   3. Mitra arranged sound internet package to send the surveys to server.

• Proper management and workability of the monitoring tool was ensured.

Data processing & reporting

 

Data from the questionnaires was entered into computer, using programs supplied by MEASURE Evaluation, written in the CS-Pro software.  Data was entered with 100% double entries. Data files were cleaned through checking of range, structure and internal consistencies. 

 

Client reference

 

On a competitive bidding Mitra and Associates won the contract for BMMS. The contract was awarded by the National Institute of Population Research and Training (NIPORT). 13/1 Sheikh Shaheb Bazar, Azimpur, Dhaka-1205, Bangladesh, Phone: 88-02-9662495, 9612193, 58611418, 58611206, Fax: 88-02-58613362, Email: info@niport.gov.bd. The contact persons are Mohammed Ahsanul Alam, Evaluation Specialist, Contact number: 01552356842 and Shahin Sultana, Sr. Research Associate, Phone (Office): 58614450, Phone (Home): 9141101, Mobile: 01715-22964, Email:ssultanabd@yahoo.comand technical contact person: Dr Mizanur Rahman, PhD, Senior Research Advisor, MEASURE Evaluation, University of North Carolina at Chapel Hill, USA, Cell : 01732-678853, email: mizanur@icddrb.org; andmizanur20wood@gmail.com.