Social media interventions and the moderation of baseline substance use: A secondary data analysis

Substance use is a significant public health problem with peak prevalence among adolescents and emerging adults (EAs). We present a secondary data analysis evaluating the moderating effect of baseline substance use on intervention outcomes from a randomized controlled trial of social media-delivered interventions for risky drinking among adolescents and EAs.

Adolescents and EAs (ages 16 to 24; N = 955, 54.5 % female, 62.6 % white) were randomized to one of three conditions: social media intervention (SMI), social media intervention + incentives (SMI + I), and control. Using negative binomial regression models, we examined the extent to which baseline substance use (an index of the sum of frequencies of use for nine substances) moderated the effect of the SMI and SMI + I conditions on 3-, 6-, and 12-month substance use outcomes.

The interaction between baseline substance use and the SMI + I treatment condition, compared to control, was significant at 3-months (estimate = -0.024; p = 0.009), wherein individuals with higher baseline substance use had lower substance use at 3-months compared to the control condition. The interaction was not statistically significant at 6-months (p = 0.055) or 12-months (p = 0.782). The interaction with baseline substance use index was not significant for the SMI condition compared to the control condition.

Individuals in the SMI + I condition with higher baseline substance use had lower substance use at the 3-month follow-up compared to the control condition, suggesting that additional incentives for interaction benefited individuals in the short-term, particularly those with higher baseline substance use. Future work should explore the moderating role of subgroup characteristics to inform future tailoring of intervention content for substance use.