The field trial of the BPP measured the process and intermediate outcome variables that included exposure to the BPP messages, change in knowledge and practices, use of services, and reaction to emergencies. The results are detailed below.
Characteristics of the sample
The characteristics of the sample from the baseline and endline surveys are presented in . The differences between the two samples among the variables in the were not statistically significant except for the variable caste/ethnicity of mother.
Characteristics of the household survey sample
Exposure to BPP messages
Two measures of exposure to the BPP messages were defined: (a) formal exposure, i.e. a respondent who received a key chain and/or was counselled using a BPP flip-chart; and (b) exposure to specific BPP messages, i.e. a respondent who reported having heard a specific message promoted through the BPP. The results of the endline survey revealed that, after one and a half years of the implementation of the BPP, 49% of the respondents received a key chain, and 54% were formally exposed to the BPP. An additional 18% of the respondents reported no formal exposure but did report indirect exposure to the BPP messages through conversations with family members or other community members ().
Direct and indirect exposures of mothers to BPP messages, Siraha district: endline survey estimates, September 2004
Results of logistic regression analysis revealed that the variables (a) mother works outside the home (odds ratio [OR]=0.6; p=0.04) and (b) age of the newborn (one month increase in age: OR=1.1; p=0.01) were associated with formal exposure. Home-based mothers presumably had more time to interact with the FCHVs and mothers’ groups, where formal exposure takes place. It is not clear why mothers of older infants were more likely to be formally exposed to the BPP. Mothers of older infants might have more time to be exposed to the BPP as some exposure might have occurred post-delivery. Alternatively, this variable might be a proxy for the variable intervention period, suggesting that more mothers were formally exposed to the BPP in early 2004 than in mid-to-late 2004—perhaps due to shortages of key chains or other supplies during the latter stages of the implementation of the Programme. The following variables were included in the model and were non-significant: birth registered, age of mother, and literacy of mother.
The respondents of the endline survey were asked if they had heard the following three key BPP messages: (a) “A pregnant women should make four antenatal care visits with a trained health worker”; (b) “A newborn should be breastfed for the first time immediately after birth”; and, (c) “A mother and newborn should have their health checked by a trained health worker within days after birth”. The respondents who answered ‘yes’ were asked how they had been exposed. Exposure to the three messages ranged from 45% to 76%. Of respondents reporting exposure, 30–48% were exposed through the key chain, while 58–64% were exposed through a health worker.
Peer key-informant monitoring was used for gauging the acceptability and understanding of the BPP messages. Key-informant mothers-in-law and pregnant women reported that most ‘women like us’ understood the messages and found them to be acceptable. These informants confirmed the survey results by reporting that key chains—FCHVs, trained TBAs, and other health workers—were important channels for receiving the BPP messages. They noted that some pregnant women prefer trained TBAs to FCHVs as Mobilizers because they also deliver services.
Challenges to coverage
The Programme intended to supply a key chain to every pregnant woman. The number of key chains provided to the DHO did not account for women who were already pregnant at the start of the Programme, contributing to an insufficient supply of key chains.
The modes of inter-personal communication used for promoting the BPP messages also limited the coverage. The mothers’ groups represent the principal mechanism through which the FCHVs counsel mothers; however, many pregnant women do not participate in mothers’ groups for cultural and logistical reasons. Although the FCHVs were encouraged to promote the BPP messages through home-visits, these visits are not part of the job description of the FCHV, and some FCHVs were reluctant to perform this task. The coverage may have been influenced by the armed conflict between the government security forces and the insurgents.
Changes in essential newborn care
The endline estimates for essential newborn practices promoted through the BPP increased by 20–30% compared to the baseline (). The use of a clean home delivery-kit or new/boiled blade to cut the umbilical cord was high at baseline (96%) and, thus, not intensely promoted during the Programme. [Contents of the kit included: razor blade, cord ties, soap, plastic sheet, plastic disc (to cut cord on), and instruction sheet.]
Changes in essential newborn care practices in BPP programme, Siraha district: baseline and endline survey estimates (September 2002 and September 2004)
The birth-preparedness index
The SUMATA project originally conceptualized and developed the birth-preparedness index (BPI). The Programme adopted the SUMATA definition of BPI to facilitate the comparison of the results of different BPP-related efforts in Nepal. The BPI is composed of seven discrete, equally-weighted variables that measure different aspects of the birth-preparedness process. The BPI is calculated at the level of the individual as the percentage of the following components that the mother reports regarding her most recent pregnancy/delivery: (a) received antenatal care at least once from a trained provider; (b) names prolonged labour as a danger sign during delivery; (c) names excessive bleeding as a danger sign during delivery; (d) made financial preparations for emergencies during pregnancy; (e) made preparations for emergency transportation during pregnancy; (f) delivery attended by a SBA; and, (g) received postpartum care from a trained provider within six weeks of delivery.
The BPI increased from 33% at baseline to 54% at endline. Increases in six of the seven components of the BPI were statistically and practically significant (). The use of a SBA did not increase.
Changes in birth-preparedness index in BPP programme, Siraha district: baseline and endline surveys estimates (September 2002 and September 2004)
Skilled birth attendance
The use of a SBA at endline remained unchanged from baseline at 17%. Of 89% (266/300) of the endline survey respondents who stated that, using a SBA was ‘important’ or ‘very important’, only 18% (48/266) reported that a SBA attended their delivery. The 218 respondents who stated that the use of a SBA was important but did not use one were asked, “why did not a trained health worker attend your delivery?” Fifty-nine percent (129/218) of the respondents cited “do not think it was necessary” as the reason. Cost was the second most commonly-cited reason (16%; 36/218). Limited knowledge of who provides local skilled birth attendance services and a preference for home-deliveries likely contributed to the low use of SBAs, as “no service available nearby” and “no practice in the community” were frequently cited.
Key-informants in Siraha reported that many community members believe that SBAs are only necessary if an emergency occurs during delivery and, thus, only contact them in the event of a crisis. This study did not specifically explore the rationale underlying community preference for home-deliveries; however, it does not appear to be rooted in negative attitudes towards the use of health facilities. Both mothers-in-law and pregnant women reported that fathers-in-law were influential in taking decisions for issues relating to finance and transport—both related to the use of SBAs.
Use of antenatal and postnatal care
A comparison of the baseline and endline estimates of antenatal and postnatal care services with a trained provider showed an increase for both the services. Attendance at two or more antenatal care visits increased from 49% to 73% (p=0.001). Attendance at four or more antenatal care visits was not measured at baseline and was 31% at endline. The use of postnatal care services within one week of delivery increased from 11% to 25% (p=0.01), while the use within six weeks of delivery doubled from 17% to 34% (p=0.02).
Maternal tetanus toxoid, postpartum vitamin A and iron usage
The FCHVs in Siraha do not distribute vitamin A and iron to pregnant/postpartum women, while the coverage of baseline tetanus toxoid (TT) was moderately high, thus limiting the potential of the Programme to increase their use. The survey results showing no appreciable baseline-to-endline change in mothers’ use of postpartum iron and vitamin A were likely biased by the questionnaire design. The unbiased survey-based estimates of TT coverage showed no change between baseline (54%) and endline (57%). The health post staff reported intermittent shortages of commodities, a factor that may have contributed to these results.
Care-seeking during emergencies
Of women who reported emergencies, the percentage who received treatment at a health facility remained constant at baseline and endline. For example, of women who reported emergencies during pregnancy, the percentage who received treatment changed little from 83% at baseline to 85% at endline (p=0.32). The corresponding indicator for care-seeking following emergencies during delivery and the postpartum period increased slightly from 59% to 62% (p=0.95) and from 78% to 85% (p=0.56) respectively.
Association among exposure to messages, knowledge, and behaviour
The positive associations between the exposure to the BPP materials/messages and the knowledge or practice of the content of the BPP message can strengthen conclusions regarding the impact of the Programme. Logistic regression analysis was used for exploring the degree of association at endline between the exposure to the BPP (independent variable) and the BPP-related knowledge or service use/behaviour (outcome variables) for two messages: (a) made 4+ antenatal care visits and (b) practised immediate breastfeeding. Two measures of exposure—formal exposure and exposure to specific BPP messages—were employed. The analysis controlled for the demographic variables and the cluster sample design.
The results presented in suggest a strong link between (a) the exposure to the messages and the correct knowledge and (b) the exposure to the messages and the correct behaviour/use of service. The level of influence of different sources of exposure appeared to be mixed. While both the definitions of exposure are positively associated with the variable “made 4+ antenatal care visits”, only one definition of exposure—exposure to message on immediate breastfeeding—was positively associated with practised immediate breastfeeding.
Levels of association among exposure to messages, knowledge, and practice/use at endline