We sought to evaluate whether there was variability in language used on socialmedia across different time points of pregnancy (before, during, and afterpregnancy, as well as by trimester and parity). Consenting patients sharedaccess to their individual Facebook posts and electronic medical records. Randomforest models trained on Facebook posts could differentiate first trimester ofpregnancy from 3 months before pregnancy (F1 score = .63) and from a random3-month time period (F1 score = .64). Posts during pregnancy were more likely toinclude themes about family (β = .22), food craving (β = .14), and date/times(β = .13), while posts 3 months prior to pregnancy included themes about sociallife (β = .30), sleep (β = .31), and curse words (β = .27), and 3 monthspost-pregnancy included themes of gratitude (β = .17), health appointments(β = .21), and religiosity (β = .18). Users who were pregnant for the first timewere more likely to post about lack of sleep (β = .15), activities of dailyliving (β = .09), and communication (β = .08) compared with those who werepregnant after having a child who posted about others’ birthdays (β = .16) andlife events (.12). A better understanding about social media timelines canprovide insight into lifestyle choices that are specific to pregnancy.
展开▼
