Background

 

It is conducted in two sub-districts—Baniachang and Nabiganj of Habiganj District with the aim of estimating the sensitivity and specificity of the BMMS 2016 maternal morbidity questions for identifying obstetric fistula (OF) and pelvic organ prolapsed (POP)cases. The study was started on July 14, 2016 and ended on December 15, 2016.

 

Survey objectives

 

The objectives of the Maternal Morbidity Validation Study – Household survey are:

 

1. To determine the sensitivity and specificity of the diagnostic survey instrument to identify obstetric fistula (OF);
2. To determine the sensitivity and specificity of the diagnostic survey instrument to identify pelvic organ prolapse (POP); 
3. To estimate prevalence of obstetric fistula, urinary incontinence, and POP in selected sub-districts.

 

Survey method

 

Sample

 

The study sample selected using simple cluster sampling, using Unions as the clusters. Out of the eight sub-districts in Habiganj district, two sub-districts – Baniachang and Nabiganj – were selected for the validation study. Out of 28 rural Unions in the two sub-districts,11 Unions were randomly selected. It is likely that the BMMS 2016 had one or more unions included in the selected clusters for this validation study. To avoid overlap with the BMMS 2016 sample, the BMMS 2016 selected clusters excluded from the selected Unions for this validation study. A census of all households in the selected Unions was carried out to provide the target sample of 47,608 households. All eligible respondents in the selected households were interviewed.

 

The Maternal Morbidity Validation Study–Household Survey was carried out with three ultimate separate samples of target cases. These are (i) a sample of 78 positive cases of obstetric fistula-like symptoms, (ii) a sample of 175 positive cases for POP like symptoms, and (iii) a sample of 234 age-matched control cases. The following steps was used to screen the ultimate samples of the target cases.

 

• Step 1, out of 28ruralUnionsinthetwosub-districts, Baniachang, and Nabiganj—11 Unions were randomly selected.
• Step 2, in order to obtain an up-to-date frame for the households from a union, household in every selected union had listed making house-to-house visit (census) in the union. A census of all households in the selected Unions had done to provide the target sample of 47,608 households.
• Step 3, after completing the household listing in a union, the interviewing teams systematically visited every listed household one after another and administered the household questionnaire for identifying the eligible respondents (ever-married women, age15-64 years, who had at least one delivery) for the individual interview, using  a short set of screening questions for identifying the positive cases of obstetric fistula-like symptoms, positive cases for POP like symptoms, and age-matched control cases. 
• Step 4, all the completed households collected from the field and brought to Mitra’s head office in Dhaka for entering data into computer with unique ID. 

 

Tool

 

Mainly two questionnaires were used in the survey—a household questionnaire and an individual questionnaire. Also a household listing operation (census) conducted to list all the households currently living in a sample union.

 

Tasks to be carried out by Mitra and Associates

 

Preparatory works

 

1. Mitra and Associates, reviewed, revised and finalized, and translated the questionnaires which were prepared by MEASURE Evaluation. 
2. Mitra and Associates also developed the following technical manuals and forms in consultation with MEASURE Evaluation: a) Interviewers’ and supervisors’ manuals; b) Household listers’ manual and listing forms; and c) Interviewers’ and supervisor’s assignment sheets for control of fieldwork.

 

Pre-test

 

Mitra and Associates carried out a pretest to check the translation, consistency and integrity of the questionnaire.

 

Training

 

Training on household listing operation conducted for 3 days. A 7 days training program on data collection operation organized for the field personnel including interviewers, supervisors and quality control officers.

 

Data collection and quality control

 

Data collection carried out over a period of 4 months. Quality control checking designed to physically verify whether the interviewer completed the questionnaires by interviewing the right respondents in the right households by asking the right questions. Quality control checking was undertaken both in presence and absence of the interviewing team in a sample union. A set of field control tables ran periodically during fieldwork to monitor and check the quality of fieldwork. Based on these tables, performances of individual supervisors and interviewers assessed and discussed with them in the debriefing session.

 

Data processing & reporting

 

MEASURE Evaluation provided the data entry, editing and tabulation programs in CS-Pro and provided assistance in setting up the entire data processing operation at Mitra and Associates. All data was entered into computer with double entries (100 percent), resolving discrepancies through referral to the original questionnaire. Mitra and Associates produced frequency distributions from the data set to see the accuracy of the data, prepared a preliminary data set and a final data set, both with appropriate documentation. They also prepared a draft report describing all data collection and data management procedures, supervisor observations/comments, and limitations/ problems encountered.

 

Client reference

 

The project was awarded by the University of North Carolina at Chapel Hill (UNC- CH), MEASURE Evaluation, Carolina Population Center, USA and funded by USAID, Washington, USA. The Contact persons are Curtis, Sian Louise (Sian), PhD, Carolina Population Center, Senior Evaluation Advisor, MEASURE Evaluation Project, Tel. +1-919-445-0434, Cell +1-919-260-9049Email: scurtis@email.unc.edu, Mizanur Rahman, PhD, Senior Research Advisor, MEASURE Evaluation, University of North Carolina at Chapel Hill, USA, Cell: 01732-678853, Email: mizanur@icddrb.org; and mizanur20wood@gmail.com, Shusmita Khan, Research Associate, MEASURE Evaluation, Carolina Population Center, University of North Carolina, Chapel Hill, Phone: +8801713209091 and Email: shusmita.h.khan@gmail.com.