Time-dependent covariates were handled dynamically to allow for changes during follow-up. These dynamic covariates included maternal income (four groups), education level (four groups), and civil status (five groups); this information was obtained from Statistics Denmark.34 Information on paternal depression, which was defined by the same criteria as maternal depression (four groups), was obtained from the Danish Psychiatric Central Research Register. Maternal and paternal mental comorbidities, including drug use, alcoholism, eating disorders, schizophrenia, and dementia, and bipolar disorder as a dichotomous variable (any/none), were identified by an algorithm developed by Prior et al.35
Time-independent covariates could not change during follow-up; these included: parity (indicators capped at 4), sex of the child, maternal age at birth (continuous), and gestational age (three groups). This information was obtained from the Danish Civil Registration System27 and the Danish National Patient Register.36
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.