The women’s group community mobilisation intervention is delivered in three districts in Bangladesh and is described in detail elsewhere . Essentially, the intervention is comprised of monthly women’s group meetings whereby communities prioritise problems that threaten the health of mothers and their babies and plan, implement and evaluate strategies to overcome these problems. Local women, who were trained in participatory methods and provided with basic information about maternal and newborn health, facilitate a participatory learning and action cycle.
The target population for the intervention is pregnant women and so all ever-married women of reproductive age are invited to participate. Every woman’s group meeting, however, is open to participation by anybody in the community, including men. Open participation is crucial to the community’s acceptance of the women’s group intervention and the dissemination of key health facts. Specific strategies were employed to attract women of reproductive age, pregnant or newly married women to participate in the groups. These strategies included actively approaching pregnant and newly married women in the community, providing them with information about the nature and objectives of the women's group intervention and encouraging them to participate.
A prospective surveillance system is in place to record and monitor birth outcomes in intervention areas. Traditional Birth Attendants (TBAs) act as key informants to notify a project monitoring and evaluation (M&E) team of all births and deaths to women of reproductive age in the intervention areas. Monitors verify births and deaths, and all women who gave birth are approached to be interviewed. Structured interviews gather detailed data on birth outcomes and care before, during and after delivery and are conducted between 42 days and one year after delivery using a survey. Detailed quantitative and qualitative process evaluation information on intervention processes, context and participation, particularly of pregnant women, is also collected throughout the intervention period. The third source of primary data gathered by the intervention implementation team is an annual household census, which records the total number of women of reproductive age and children living in the intervention areas. Publicly available secondary population data sources in Bangladesh include the 2001 Bangladesh Maternal Health Services and Maternal Mortality Survey (BMMS) , the 2007 Bangladesh Demographic and Health Survey (BDHS)  and the Bangladesh Bureau of Statistics (BBS) national census data from 2001 .
Four methods have been developed to estimate intervention coverage among pregnant women, each using at least one of the primary and/or secondary data sources mentioned above. Confidence intervals for all estimates are calculated using the confidence interval calculator in STATA version 12. Coverage estimates that are based on secondary data and assumptions about underlying fertility levels are presented with ranges that reflect extreme values and uncertainty in these assumptions. Estimates derived from each method are compared and critically discussed in terms of these inherent assumptions, methodological strengths and weaknesses and differing data and resource demands. All estimates are for the period of October 2009 to May 2010.
Method 1: Direct measurement of a proxy indicator
Recognising that direct measurement of the pregnancy status of all women in the study area is unfeasible, the proportion of all deliveries in which the mother participated in the women’s group intervention can be measured using survey methods and subsequently used as a proxy-measure for coverage among pregnant women. Method 1 uses data from the community monitoring and evaluation (M&E) survey whereby all deliveries in the study areas are recorded and all women are invited to participate in a survey interview. Women’s participation in the women’s group intervention is ascertained through these interviews. The proportion of women reporting participation during pregnancy approximates the exposure among pregnant women in the population and can be presented with 95% confidence intervals.
Method 2: Direct measurement among participants and modelled extrapolation based on routine surveillance of births
During each monthly woman’s group meeting, group facilitators ask women to volunteer their pregnancy status. This count provides the numerator in the calculation of coverage among pregnant women at a particular point in time. Obtaining the same information for the denominator, i.e. the total number of currently pregnant women in the whole population is more complex and arguably not feasible within a short time frame and without considerable resources or ethical implications. However, assuming that fertility levels remain fairly constant in relatively small geographic areas over a nine-month period, one can assume that the total number of babies born is approximately equal to the total number of pregnancies conceived and, therefore, that for every pregnancy that ends, there are nine others that are in gestation. The count of all births occurring in the study area from the prospective birth surveillance can therefore be considered an approximation of the number of new pregnancies arising, which, if multiplied by nine (for the average duration of pregnancy in months) gives an approximation of the total number of pregnant women in the population. In theory, therefore, the minimum intervention coverage among pregnant women can be estimated by dividing the number of pregnant women participating in women’s groups by the total number of deliveries that month multiplied by nine.
In practice, however, few women are aware of their pregnancy status from the moment of conception and fewer are willing to share their pregnancy status during the early stages of pregnancy. This is certainly the case in rural Bangladesh where cultural norms and beliefs usually prevent women from publicly disclosing their pregnancy status until after approximately 3.5 months. Rather than multiplying the denominator by nine, therefore, it is arguably more sensible to multiply by 5.5 (i.e. 9 minus 3.5). So, to estimate intervention coverage among pregnant women, the monthly count of currently pregnant women participating in women’s groups is divided by the total number of births captured multiplied by 5.5.
To test the sensitivity of this measure to varying assumptions of the duration of pregnancy concealment, plausible minimum coverage can be estimated by multiplying the denominator by 8 to represent early disclosure after just one month of pregnancy whilst maximum coverage may be estimated by multiplying the denominator by 4, representing delayed pregnancy status disclosure until pregnancy may be noticeable by others at approximately 5 months. These extremes therefore represent a range of coverage estimates depending on underlying assumptions of duration of pregnancy concealment, each of which can be presented with 95% confidence intervals.
Method 3: Direct measurement among participants and modelled extrapolation based on cross-sectional measurements and national data
As in Method 2, monthly counts of currently pregnant women participating in the women’s group intervention provides the numerator in calculating coverage among pregnant women at a particular point in time. However, rather than basing the denominator on counts of births, which demands prospective community surveillance or birth registration systems, Method 3 uses the intervention’s household census data and national BDHS data to estimate the total number of pregnant women in the study area. An annual household census, implemented by the study team, records the total number of households and their inhabitants, by age and sex, residing in the study area. To estimate the denominator (i.e. the total number of currently pregnant women in the study area), the total number of married women in reproductive age identified through the intervention’s household census is multiplied by the 2007 BDHS estimate of the proportion of women of reproductive-age who reported being pregnant at the time of the survey, which is 5.6% in rural areas . This method assumes that overall fertility rates have not changed substantially since 2007. However, fertility rates do vary by administrative divisions and the proportion of currently pregnant women ranges from 4.9% in Dhaka and Khulna divisions to 6.9% in Sylhet division. This range in the proportion of currently pregnant women can be used to calculate plausible minimum and maximum coverage by deriving the denominator from the lower and upper estimates of proportions of pregnant women, respectively, each of which can be presented with 95% confidence intervals.
Method 4: Direct measurement among participants and modelled extrapolation based on published national data
Often, public health interventions may not have the resources or capacity to collect and process data. Nevertheless, available data that have been previously published on a regional or national level can be used to provide estimates of intervention coverage. The count of currently pregnant women participating in the women’s group intervention forms the numerator in calculating coverage among pregnant women at a particular point in time. The data capture demands for this are minimal and it is realistic to expect most services or interventions to keep a record of the number of recipients. The denominator is derived from the total number of married women in reproductive age as identified through the 2001 BBS national census multiplied by the 2001 Bangladesh Maternal Health Services and Maternal Mortality Survey estimate of the proportion of women of reproductive-age who reported being pregnant in the same year (2001), which is 5.5 in rural areas . As in method 3, the proportion of currently pregnant women varies across administrative divisions. Khulna division had the lowest proportion currently pregnant in 2001 (4.6%), whereas the highest proportion was reported for Sylhet division (6.4%). A plausible minimum coverage can therefore be estimated by multiplying the denominator by 6.4 whilst maximum coverage may be estimated by multiplying the denominator by 5 and 95% confidence intervals can be calculated for each estimate.
The women’s group intervention trial and all monitoring and evaluation activities have received ethical approval from the University College London Research Ethics Committee (ID Number: 1488/001) and by the Ethical Review Committee of the Diabetic Association of Bangladesh. Informed verbal consent is obtained from all survey respondents before any data are collected.