Discovering Cohorts of Pregnant Women From Social Media for Safety Surveillance and Analysis

Abstract

BackgroundPregnancy exposure registries are the primary sources of information about the safety of maternal usage of medications during pregnancy. Such registries enroll pregnant women in a voluntary fashion early on in pregnancy and follow them until the end of pregnancy or longer to systematically collect information regarding specific pregnancy outcomes. Although the model of pregnancy registries has distinct advantages over other study designs, they are faced with numerous challenges and limitations such as low enrollment rate, high cost, and selection bias.ObjectiveThe primary objectives of this study were to systematically assess whether social media (Twitter) can be used to discover cohorts of pregnant women and to develop and deploy a natural language processing and machine learning pipeline for the automatic collection of cohort information. In addition, we also attempted to ascertain, in a preliminary fashion, what types of longitudinal information may potentially be mined from the collected cohort information.MethodsOur discovery of pregnant women relies on detecting pregnancy-indicating tweets (PITs), which are statements posted by pregnant women regarding their pregnancies. We used a set of 14 patterns to first detect potential PITs. We manually annotated a sample of 14,156 of the retrieved user posts to distinguish real PITs from false positives and trained a supervised classification system to detect real PITs. We optimized the classification system via cross validation, with features and settings targeted toward optimizing precision for the positive class. For users identified to be posting real PITs via automatic classification, our pipeline collected all their available past and future posts from which other information (eg, medication usage and fetal outcomes) may be mined.ResultsOur rule-based PIT detection approach retrieved over 200,000 posts over a period of 18 months. Manual annotation agreement for three annotators was very high at kappa (κ)=.79. On a blind test set, the implemented classifier obtained an overall F1 score of 0.84 (0.88 for the pregnancy class and 0.68 for the nonpregnancy class). Precision for the pregnancy class was 0.93, and recall was 0.84. Feature analysis showed that the combination of dense and sparse vectors for classification achieved optimal performance. Employing the trained classifier resulted in the identification of 71,954 users from the collected posts. Over 250 million posts were retrieved for these users, which provided a multitude of longitudinal information about them.ConclusionsSocial media sources such as Twitter can be used to identify large cohorts of pregnant women and to gather longitudinal information via automated processing of their postings. Considering the many drawbacks and limitations of pregnancy registries, social media mining may provide beneficial complementary information. Although the cohort sizes identified over social media are large, future research will have to assess the completeness of the information available through them.

Social media mining for birth defects research: A rule-based, bootstrapping approach to collecting data for rare health-related events on Twitter

“…In recent work [10], we took the first step towards exploring whether social media mining could be used to complement pregnancy exposure registries as a novel method for observing pregnancies. Considering that 21% of American adults and, more specifically, 36% of Americans between ages 18–29 use Twitter [11], the promise of valuable information directly from the population of interest motivated us to develop and deploy a natural language processing (NLP) and machine learning pipeline that automatically collects and stores the Twitter user timelines —all publicly available posts over time by that user—of women who have reported a pregnancy on Twitter.…”

An annotated data set for identifying women reporting adverse pregnancy outcomes on Twitter

“…We handcrafted 11 regular expressions to retrieve tweets that mention adverse pregnancy outcomes, from a database containing more than 400 million public tweets posted by more than 100,000 users who have announced their pregnancy on Twitter [7] . These query patterns were designed to account for the various ways adverse pregnancy outcomes may be linguistically expressed on social media—for example, reporting a miscarriage or stillbirth through the use of rainbow baby (Pattern 2) or hashtags such as #babyloss, #pregnancyloss, #iam1in4 , or #waveoflight (Pattern 9), learned through an iterative process of manually reviewing tweets matched by other query patterns [8] .…”

“… Data format Raw, analyzed Parameters for data collection Tweets were collected if they mention miscarriage, stillbirth, preterm birth/premature labor, low birthweight, or neonatal intensive care. Description of data collection Handcrafted regular expressions retrieved 22,912 tweets that mention adverse pregnancy outcomes from a database containing public tweets posted by women who have announced their pregnancy on Twitter [7] . Two professional annotators labeled 8109 of the 22,912 tweets (one random tweet per user) in a binary fashion, distinguishing those potentially reporting that the user has personally experienced the outcome from those that merely mention the outcome.…”

“…The 6487 tweets were retrieved from a database [7] using 11 handcrafted regular expressions—search patterns that define matching text strings (Supplementary Material). Table 1 presents samples of (slightly modified) tweets in the data set, and total distribution of “outcome” and “non-outcome” tweets for each of the 11 query patterns.…”

“…To advance the use of social media data as a complementary resource for epidemiology of adverse pregnancy outcomes, we present a data set of 6487 tweets that mention miscarriage, stillbirth, preterm birth or premature labor, low birthweight, neonatal intensive care, or fetal/infant loss in general. These tweets are a subset of 22,912 tweets retrieved by applying hand-written regular expressions to a database containing more than 400 million public tweets posted by more than 100,000 women who have announced their pregnancy on Twitter [7] . Two professional annotators labeled the 6487 tweets in a binary fashion, distinguishing those potentially reporting that the user has personally experienced the outcome (“outcome” tweets) from those that merely mention the outcome (“non-outcome” tweets).…”