JMIR Publications

ABSTRACT

Background:

Wet markets are critical for food security and sustainable development in their respective regions but are also associated with health risks. Due to their cultural significance, they attract numerous visitors and generate tourist-geared information on the Web (i.e. on social networks as TripAdvisor). These unexploited data can be used to create an internationally-comprehensive wet market inventory to support epidemiological surveillance and control in these settings, which to our knowledge, does not yet exist.

Objective:

Using social network data, we aim to: assess the level of wet markets’ touristic importance online; produce the first distribution map of wet markets of touristic interest; and identify common diseases facing visitors in these settings.

Methods:

TripAdvisor was selected as the data source of this study following an analysis of food markets’ touristic relevance on the web. A web scraping tool (ParseHub) was used to extract wet market names, locations, and reviews from TripAdvisor. The latter were analyzed and when possible, assigned GeoSentinel diagnosis codes. This syndromic information was overlaid onto a mapping of wet market locations.

Results:

89 of the first 150 Google Search results (59.3%) for “wet market” (July 2017) were tourism-related. Of the 1,090 hits on TripAdvisor for this keyword, 393 (36%) were confirmed wet markets; syndromic information was available for 57 of these (14.5%). The confirmed wet markets were heterogeneously distributed: Asia concentrated 246 (62.6%) of them, Europe 76 (19.3%), North America 31 (7.9%), Oceania 20 (5.1%), Africa 12 (3.1%), and South America 8 (2.0%). Analysis of reviews corresponding to these wet markets revealed the most frequently occurring disease among visitors was food poisoning, accounting for 51 of 95 diagnoses (54%). This proved most prevalent among those visiting South American markets (18 of 51 food poisoning incidents [35%]) but less for Asian markets (6 of 51 food poisoning incidents [12%]) when normalizing for wet market number.

Conclusions:

To our knowledge, this study is first to map the global distribution of wet markets of touristic importance and adverse health events experienced by their visitors, highlighting the potential of social network data in global epidemiological surveillance.