Abstract
Objectives: to analyze the association between formal sexual education and pregnancy planning in public and private maternity hospitals in Curitiba, Brazil. Methods: cross-sectional study of 391 pregnant women (260 public hospital, 131 private) using the LMUP (cutoff 9/10). Adjusted prevalence ratios (PR) were estimated by Poisson regression with robust variance in a hierarchical model excluding partner support (main analysis). Sensitivity analyses included income and parsimonious models, the LMUP recalculated without item 5, stratified analysis, and E-value. Results: the prevalence of unplanned pregnancy (score <10) was 52.2%. Lack of formal sexual education was modestly associated with unplanned pregnancy, with a stable direction in eight specifications (PR 1.07–1.42; significant in two). The point-estimate E-value was 1.61; indicating that an unmeasured residual confounder would need to have a simultaneous association of ≥ 1.61 with both exposure and outcome, on a risk scale, to reverse the finding; the interpretation of robustness should be made with caution because the confidence interval of the main estimate includes the null value. The recalculation of the LMUP without item 5, with the concomitant exclusion of partner support as a covariate, suggested that the circularity between these dimensions did not substantially distort the core finding (RP=1.22; p=0.059). Contraceptive knowledge and prior contraceptive use were not associated with any specification of the outcome in. Conclusions: the findings are consistent with the hypothesis that formal sexual education contributes dimensions not captured by contraceptive knowledge alone. The non- association between contraceptive knowledge and pregnancy planning was the most robust finding. The hypothesis requires confirmation in longitudinal studies with more granular exposure measurement.

This publication has 18 references indexed in Scilit: