Environmental Context and Reporting Completeness of Event-Based Ecological Momentary Assessment in Cigarette and Cannabis Use.

INTRODUCTION

Ecological momentary assessment (EMA) allows researchers to capture substance use as it occurs in daily life, but participants may not report every use event. Little is known about whether the environments where people spend time affect the completeness of these reports.

METHODS

We examined how consistently young adults reported cigarette and cannabis use through event-based EMA and whether reporting differed by the social vulnerability of places they visited. We analysed 30 days of GPS-enabled EMA and daily diary data from 34 young adults in California (730 participant days). We compared daily cigarette and cannabis use events reported through EMA with those reported in the following day's diary and classified EMA reports as lower than, matching, or higher than the diary. Using GPS data and the Social Vulnerability Index, we measured the social vulnerability of participants' daily activity spaces relative to home.

RESULTS

Bayesian multilevel multinomial logistic regression showed that higher social vulnerability in daily activity spaces relative to home was associated with greater odds of reporting fewer use events through EMA than in the daily diary for cigarettes (aOR 1.55, 95% CrI 1.25-1.93) and cannabis (aOR 1.28, 95% CrI 1.06-1.55). The association for cigarettes was also observed in a larger independent dataset.

DISCUSSION AND

CONCLUSIONS

These findings suggest that the completeness of real-time substance use reporting may depend partly on where participants spend their time. Researchers using event-based EMA should consider environmental context when interpreting incomplete reports and designing studies to improve data completeness.