Clustering of characteristics associated with level of pregnancy intention: a latent class analysis of an urban cross-sectional sample in the Netherlands

Abstract Background Despite evidence that pregnancy intentions are complex, unintended pregnancy remains studied using binary measures and few studies have examined combinations of factors contributing to pregnancy intention. This study aimed to identify groups of pregnant people with similar combinations of characteristics and exposures, and study the association between these groups and level of pregnancy intention. Methods This study, part of a population-based cohort study of pregnant people and partners (RISE UP study), uses cross-sectional surveys linked to routine data. Latent class analysis identified groups of pregnant people in The Hague distinguished by shared socioeconomic and reproductive characteristics and positive and negative exposures. Linear regression assessed the association between class membership and level of pregnancy intention, adjusted for recruitment location, gestational age at survey, and survey year. Results In the final sample of 560 pregnant people, four classes were identified. Two reflected general stability, differing by gravidity: high stability, multigravida and high stability, primigravida . Two reflected greater socioeconomic adversity and fewer positive exposures, differing by presence of negative exposures: partial stability and cumulative adversity . Compared to the high stability, multigravida class, pregnancy intention was significantly lower in the partial stability class (adjusted beta = -0.71 95%CI (-1.31 – -0.11)) and the cumulative adversity class (adjusted beta = -1.81 95%CI (-2.31 – -1.30)). Discussion Cumulative adversity and partial stability were associated with lower pregnancy intention, supporting suggestions that pregnancy intention may indicate underlying systemic inequalities. Policy and care providers, including midwives, should address these inequalities and tailor support to individual needs.