[Individual and environmental characteristics associated with immunization of children in rural areas in Burkina Faso: a multi-level analysis].

PROBLEM: National and international efforts to immunize children aim to remove barriers that hinder full vaccination programs and to reinforce factors promoting it. Despite Burkina Faso's participation in all international and sub-regional initiatives to protect mothers and children from vaccination-preventable communicable diseases, vaccination coverage there remains low and has grown irregularly, from 34.7% in 1993 to 29.3% in 1998 and 43.9% in 2003. The situation is even more critical in rural than in urban areas.
OBJECTIVE: To analyze the contribution of individual and environmental characteristics associated with vaccination of children aged 12-23 months in rural areas in Burkina Faso. Study population and methods. Data from the 1998 DHS (Demographic and Health Survey) and the 1997 Health Ministry Statistical Yearbook were used with a multi-level approach. Analysis distinguished two levels corresponding to the data's hierarchical structure: characteristics of children and their family's environment (level 1) and the health system and social environment (level 2). The study included 805 children aged 12 to 23 months, living in 44 health districts. The dependent variable was the child's vaccination status and is dichotomous (completely vaccinated or not). Completely vaccinated children are those who have received the BCG, the three doses of DTCoq, oral polio, measles and yellow fever vaccines, according to either their vaccination cards or their mothers' statement.
RESULTS: The likelihood of vaccination increased with the level of household wealth (OR [well-off/poor]=1.88; [CI: 1.15-3.06] and was strongly associated with use of health services (OR [Prenatal care and assisted delivery/none of these services]=5.64; [CI: 3.16-10.05]). Nevertheless, these 2 variables did not alone explain the differences in vaccination observed between districts. More than 37% of the variation for vaccination completeness can be attributed to differences between health districts. Resources appear to play a minor role but a 1% increase in the proportion of educated women in the district increased the odds of complete vaccination by a factor of 1.14 [CI: 1.01-1.27]. Discussion. Despite universal access to free vaccination, children from poor households are less likely to receive all their vaccines than children from well-off households. This is probably due to indirect costs that stem from vaccination; the financial barrier remains one of the most significant factors preventing complete vaccination. Previous utilization of prenatal care and institutional delivery is more related to dynamics or even interaction between individuals and the health system. In addition to their direct effects, the interrelation between population and health systems may constitute a vaccination culture that may play a major role in explaining vaccination completeness. The resources of the health system bear little relation to vaccination. They are necessary but not sufficient for good health services. The organizational dynamic of health teams, the leadership of health district supervisors and staff motivation are key elements in these processes but were not measured in this study.
CONCLUSION: Adding resources to vaccination programs is always a challenge for a number of national healthcare systems. It is not, however, the only key to success. The organization of healthcare systems and the contacts and relationships they establish with their populations appear to be determinant. The local vaccination culture that results from this interaction may be a key to explaining the variations observed between the different health districts.