Choosing if and when to have a child is important from a human right's perspective and represents a benchmark for promoting a person's health and wellbeing [1,2]. Despite widespread use of contraception, unintended pregnancies remain frequent and represent a significant public health concern. Studies and reviews, past and recent, report on associations between unintended pregnancy and poor maternal health, particularly postpartum depressive symptoms [3,4], as well as adverse perinatal outcomes including low birthweight and preterm birth [5,6]. While the causal link between pregnancy intentions and maternal and child health is a matter of debate, studies consistently indicate women and other people with unintended pregnancies receive less adequate prenatal care [7] and have less healthy behaviors during pregnancy [3,8,9].

Unintended pregnancy is a combined measure of pregnancy wantedness and timing, assessing whether a person wanted any more children at the time of conception and if so, if the pregnancy came too soon or at the right time. It was originally developed by demographers to characterize fertility trends and to monitor unwanted fertility [10,11]. Researchers in perinatal health often use this standard indicator [[3], [4], [5],7], which only requires one or two questions, a pragmatic advantage in long multi-thematic surveys. However, the use of this demographic measure raises concerns when predicting individual maternal behaviours and perinatal outcomes, because the question leaves little space for complexity in pregnant person's intentions and emotional response to pregnancy [12], [13], [14]. The addition of a response option for women who were unsure of their desire for pregnancy was introduced in 2012 in the US Pregnancy Risk Assessment Monitoring System and showed that 15% of pregnant women were uncertain about their pregnancy intentions [15]. A number of researchers have attempted to account for the multidimensionality of person's persectives on pregnancy, including the emotional response to the pregnancy, the extent the person was trying to conceive, pre-conception behaviours (folic acid use, seeking health advice, etc), and the partner's intendness of the pregnancy [11,16,17], but these additional indicators have yet to be considered in perinatal research.

Using the 2016 French national perinatal survey [18], we examined how information on pre-conception contraceptive behaviors could complement the traditional pregnancy time-based preference indicator (wanted pregnancy now, sooner, later or not at all). We explored how a composite indicator including pregnancy timing preference and preconception contraception vary according to women's sociodemographic characteristics and examined how this composite indicator correlated with prenatal care seeking behaviors.