JMIR Publications

ABSTRACT

Background:

Technology-based computational strategies that leverage social network site (SNS) data to detect substance use are promising screening tools but rely on the presence of sufficient data to detect risk if it is present. A better understanding of the association between substance use and SNS participation may inform the utility of these technology-based screening tools. Most studies examining substance use and SNS participation have focused on the association between drinking and Facebook use among young adults.

Objective:

To examine associations between substance use – including both alcohol and drug use – and Instagram posts among adults of all ages, and to test whether such associations differ as a function of age, gender, and race/ethnicity.

Methods:

As detailed in a recent report from Hassanpour and colleagues, participants with an Instagram account were recruited primarily via Clickworker (N = 3117). With participant permission and Instagram’s approval, participants’ Instagram photo posts were downloaded with an Application Program Interface (API). The assessment of participants’ past-year substance use was adapted from the NIDA Quick Screen. At-risk drinking was defined as having “more than a few alcoholic beverages”, drug use was defined as any use of non-prescription drugs such as cannabis, heroin, and cocaine, and prescription drug use was defined as any non-medical use of prescription medications such as opioid painkillers, benzodiazepines, and stimulants. We used logistic regression to examine the associations between substance use and any Instagram posts, and negative binomial regression to examine the associations between substance use and number of Instagram posts. We examined whether age (18-25; 26-38; 39+ years), gender, and race/ethnicity moderated associations in both logistic and negative binomial models. All differences noted were significant at the .05 level.

Results:

Compared to no at-risk drinking, any at-risk drinking was associated with both a higher likelihood of any Instagram posts (OR = 1.516; P = .027) and a higher number of posts (OR = 1.881; P < .001), except among Hispanic/Latino individuals where at-risk drinking was associated with a similar number of posts. Compared to no drug use, any drug use (e.g., cannabis, cocaine, heroin, etc.) was associated with a higher likelihood of any posts (OR = 1.771; P = .021) but was associated with a similar number of posts. Compared to no prescription drug use, any prescription drug use (e.g., non-medical use of opioids, benzodiazepines, stimulants, etc.) was associated with a similar likelihood of any posts, and was associated with a lower number of posts only among those ages 39 and older (OR = .242; P < .001). Of note, main effects of gender and race/ethnicity were significant. Being female was significantly associated with a greater likelihood of any posts (OR = 2.677; P < .001) and greater number of posts (OR = 2.967; P < .001) compared to males. Compared to White identification, Black identification was significantly associated with a greater likelihood of any posts (OR = 1.676; P = .002), while Hispanic/Latino identification was significantly associated with a greater likelihood of any posts (OR = 1.653; P = .022) and greater number of posts (OR = 2.005; P < .001).

Conclusions:

Individuals with at-risk drinking may provide enough data to reliably develop and test SNS-based computational risk detection models. If aiming to detect prescription drug use among middle and older-aged adults, however, such models may have fewer posts with which to work. As more is learned about SNS behaviors among those who use substances, researchers may be better positioned to successfully design and interpret these innovative risk detection approaches.