Effects of a four-year health systems intervention on the use of maternal and infant health services: results from a programme evaluation in two districts of rural Chad.

BACKGROUND: Attendance of maternal and infant care services in rural Chad are consistently low. Our study aimed to assess the use of antenatal (ANC) and postnatal care (PNC) services, health facility delivery and infant health services after 4 years of a health systems intervention for improving the infrastructure, supplies, training and sensitization for maternal and infant health in two districts of rural Chad. METHODS: Data from a repeated cross-sectional household survey conducted in Yao and Danamadji in 2015 and in 2018 were analyzed. A stratified two-stage cluster sampling methodology was applied to achieve a representative sample of the rural settled and mobile population groups in the study area. A generalized linear model was applied to determine the health care utilization rates. Multivariate regression models were used to assess the association between the programme intervention and utilization outcomes of selected maternal and infant health services. RESULTS: Complete datasets were available for 1284 households at baseline. The endline analysis included 1175 households with complete survey data. The use of at least one ANC amongst pregnant women increased in both settled communities (from 80% in 2015 to 90% in 2018) and amongst mobile pastoralist communities (from 48% in 2015 to 56% in 2018). The rate of home delivery among settled communities and mobile pastoralists changed little between baseline and endline and remained high for both population groups. Individuals that were covered by the health systems intervention were however significantly more likely to attend ANC and less likely to give birth at home. PNC services only showed improvements amongst the settled communities (of 30%). Infants' reported health outcomes and vaccination coverage considerably improved; the latter especially among mobile pastoralist (from 15% in 2015 to 84% in 2018). CONCLUSION: A combination of health systems strengthening interventions was associated with an increased use of certain maternal and infant health services. However, to facilitate equitable access to and use of health care services in particular in times of increased vulnerability and by certain population groups in hard-to-reach areas, reinforced health education and culturally adapted communication strategies, including gender-specific messaging will be needed over a sustained period.

Bottlenecks in the provision of antenatal care: rural settled and mobile pastoralist communities in Chad.

OBJECTIVE: To assess antenatal care (ANC) coverage and analyse constraining factors for service delivery to rural settled and mobile populations in two districts in Chad. METHOD: Data from cross-sectional household and health facility surveys in the two Chadian rural health districts were analysed. First, contact coverage of ANC services in the study area was estimated from household data as the proportion of women who visited health facilities to obtain ANC during their last pregnancy. Second, bottlenecks in the provision of this service were explored by calibrating a multiplicative model of ANC contact coverage to household and health facility data. The model allowed quantification of the magnitude by which coverage decreased as it progressed through the health system. Sensitivity analysis was applied to account for uncertainty around the estimated coverage factors. RESULTS: Direct estimates revealed that ANC contact coverage decreased as the number of required visits increased: 79% of rural settled mothers and 46% of mobile pastoralist mothers visited a health facility to obtain ANC at least once (ANC 1). Among mobile pastoralists, only 20% of pregnant women attended ANC at least three times compared to 63% of rural settled women. Availability, accessibility, affordability and acceptability contributed to reductions in service coverage in both populations. For mobile pastoralists, acceptability was clearly the most important factor. ANC 1 contact coverage resulting from the model is 50% for rural settled and 30% for mobile pastoralists. CONCLUSION: Antenatal care coverage was low in rural districts of Chad, particularly for mobile pastoralists. Acceptability largely explained the prevailing difference between the two population groups.