Twitter-based measures of neighborhood sentiment as predictors of residential population health.

Several studies have recently applied sentiment-based lexicons to Twitter to gauge local sentiment to understand health behaviors and outcomes for local areas. While this research has demonstrated the vast potential of this approach, lingering questions remain regarding the validity of Twitter mining and surveillance in local health research. First, how well does this approach predict health outcomes at very local scales, such as neighborhoods? Second, how robust are the findings garnered from sentiment signals when accounting for spatial effects? To evaluate these questions, we link 2,076,025 tweets from 66,219 distinct users in the city of San Diego over the period of 2014-12-06 to 2017-05-24 to the 500 Cities Project data and 2010-2014 American Community Survey data. We determine how well sentiment predicts self-rated mental health, sleep quality, and heart disease at a census tract level, controlling for neighborhood characteristics and spatial autocorrelation. We find that sentiment is related to some outcomes on its own, but these relationships are not present when controlling for other neighborhood factors. Evaluating our encoding strategy more closely, we discuss the limitations of existing measures of neighborhood sentiment, calling for more attention to how race/ethnicity and socio-economic status play into inferences drawn from such measures.

Computational Lexical Analysis of the Language Commonly Used to Describe Gout.

OBJECTIVE: To characterize the current language that is used in describing and defining gout, its symptoms, and its treatment by reviewing recent publications in rheumatology and determining how word choice may, or may not, be reflective of recent scientific developments in gout specifically. METHODS: This was a computational linguistics study, using collocations analyses and concordance analyses on a database of scientific literature related to gout. The final data set for analysis included 2,590 articles, all relating to gout and published between May 2003 and May 2013 and amounting to 12,101,036 tokens (sentence segments). Analysis was conducted by a team of linguists and social scientists. RESULTS: Our primary finding is that current disease language in gout is marked by ambiguity and imprecision, as evidenced by numerous terms that have similar but distinct meanings, but are nevertheless used interchangeably, therefore blending the slight but significant distinctions between these words. Whereas treatment language is characterized by a multitude of terms to describe a therapeutic mechanism of action, there is a relative void of terms and phrases used to describe success (treating to target) in gout. CONCLUSION: The data suggest that the language used to describe gout could be improved and updated. A transformation from an antiquated and insufficiently descript terminological set to one that reflects the recent scientific and clinical advancements made in the category would maximize opportunities for patient and physician understanding.