| Objective. Adolescent parenthood is a complicated problem that seems to defy efforts to reduce its rates. While many studies point to sociodemographic factors as the cause, professionals have had difficulty in reducing the number of adolescent births. This study was designed to test the viability of Geographic Information Systems (GIS), a potentially powerful diagnostic tool, on a social problem about which key factors are already known. Our purpose was first, to validate the current model in our study area, and second, to discover additional variables that may contribute to high birth rates among adolescents.; Methods. The locations of all adolescent women between the ages of 12 and 19 who gave birth within our study area in San Bernardino County in 1995 were plotted on a basemap so we could determine where patterns of high and low birth rates were, for comparison purposes. Because we knew from the literature that race/ethnicity, income, and education were important factors in predicting adolescent parenthood, we extracted variables representing those factors from census data, and analyzed our 444 block groups in relation to four categories of risk derived from birth rates. Viewing birth patterns on a map enabled us to see large differences in physical size among the block groups, leading us to calculate a new variable, population density, for inclusion in the model. We used Multinomial Logistic Regression to calculate odds ratios, and our referent was the block groups with no births reported.; Results. The sociodemographic factors most often attributed to high birth rates among adolescents proved to be statistically significant in our univariate analysis. When we added population density to the full model, however, population density became the preeminent predictor of adolescent parenthood.; Conclusion. Findings suggest that population density may be a more important predictor for adolescent parenthood than the factors consistently cited in the literature. While our definition of population density is relatively straightforward, the concept of density is very complex, and simple conclusions are therefore inadvisable. This study lends credence to the use of GIS as a potent tool of discovery. |
