Description
Hypothesis: The primary null hypothesis is that there is no significant negative global relationship between maternal mortality and percent births overseen by skilled health workers. The secondary null hypothesis is that there is no spatial clustering of countries that exhibit high or low rates of maternal mortality and variables significantly associated with maternal mortality. Methods: In addition to the independent variable of interest, percent births overseen by skilled health workers, the analyses will assess maternal mortality's relationships with the following variables: mothers receiving prenatal care, female contraceptive prevalence, adolescent births, physicians per capita, and per capita health expenditure. First, bivariate relationships with maternal mortality will be tested through simple linear regression. Second, a multiple linear regression model will be constructed from statistically significant bivariate associations with maternal mortality. Last, geospatial analyses will be conducted to determine the geographic locations associated with high and low levels of all variables included in statistical analyses. Results: In simple linear regression tests, all independent variables were significantly related to maternal mortality. Furthermore, bivariate modeling revealed that many countries were concentrated among low levels of maternal mortality. Multiple linear regression revealed that percent births overseen by skilled health workers, contraceptive prevalence, and adolescent birth rate remained significant even when controlling for other variables. Geospatial analysis showed that maternal mortality was high in Africa and southern Asia. Summary: Maternal mortality is significantly associated with percent births overseen by skilled health workers, even when controlling for several covariates
