JMIR Publications

ABSTRACT

Background:

The popularity of electronic cigarettes (e-cigarettes) has steadily increased, prompting a considerable number of individuals to search for relevant information on them. Previous e-cigarette infodemiology studies have focused on assessing the quality and reliability of website content and quantifying the impact of policies. In reality, most low-income countries and low- and middle-income countries have not yet conducted e-cigarette use surveillance. Data sourced from internet search engines related to e-cigarettes have the potential to serve as cost-effective supplementary means to traditional monitoring approaches.

Objective:

This study aimed to analyze the spatio-temporal distribution characteristics and associated sociodemographic factors of e-cigarette searches using trends from the Baidu search engine.

Methods:

The query data related to e-cigarettes for 31 provinces in mainland China were retrieved from the Baidu index database from January 1, 2015, to December 31, 2022. Concentration ratio methods and spatial autocorrelation analysis were applied to analyze the temporal aggregation and spatial aggregation of the e-cigarette Baidu index, respectively. A variance inflation factor test was performed to avoid multicollinearity. A spatial panel econometric model was developed to assess the determinants of e-cigarette online attention.

Results:

The daily average Baidu index for e-cigarettes increased from 53234.873 in 2015 to 85416.995 in 2021 and then declined to 52174.906 in 2022. This index was concentrated in the southeastern coastal region, whereas the hotspot shifted to the northwestern region after adjusting for population size. Positive spatial autocorrelation existed in the per capita Baidu index of e-cigarettes from 2015-2022. The results of the local Moran’s I showed that there were mainly low-low cluster areas of the per capita Baidu index, especially in the central region. Additionally, the male-female ratio, the proportion of high school and above education and the per capita gross regional domestic product were positively correlated with the per capita Baidu index for e-cigarettes. A higher urbanization rate was associated with a reduced per capita Baidu index.

Conclusions:

With the increasing popularity of online searches for e-cigarettes, a targeted e-cigarette health education program for individuals in the northwest, males, rural populations, high school and above educated individuals and high-income groups is warranted.