Measuring Marijuana Use and Craving via Text Messaging as a Form of Ecological Momentary Assessment
|
If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL). |
Abstract
Background:
Substance use behavior and craving can be difficult to assess, with many studies focusing on retrospective self-report. Past research examining the association between craving and substance use have not always found the two to be related, or if they are related, often the relationship is not particularly strong. In recent years, increased attention has been placed on using ecological momentary assessment (EMA) to learn more about substance use behavior in the moment, though much of this work has focused almost exclusively on tobacco and alcohol use, with less focus on illicit drugs. Several recent EMA studies have shown that craving for marijuana is related to marijuana use.
Objective:
The goal of the current study was to examine whether marijuana craving would predict marijuana use when measured in two different ways through EMA. In addition, we aimed to examine how marijuana use assessed through EMA compared to a commonly used and validated retrospective substance use measure – Timeline Followback (TLFB).
Methods:
College student marijuana users (n = 57) in Colorado (prior to legalization) were recruited to participate in a baseline assessment, two-week EMA, and brief follow-up. To help protect participants' confidentiality, only first names were collected and used throughout the study. Participants averaged 20.05 (SD = 2.60) years of age and were predominantly Caucasian (77%) and Latino (11%). Most were heavy marijuana users, smoking on average 25 days out of the last 30. Participants were sent text messages randomly during three time blocks throughout the day for a two-week period. Overall EMA response rate was 89%. Each Short Message Service (SMS) text included the same nine questions (three questions used for the current analyses). Marijuana craving was assessed on a 1-10 scale (low to high), while marijuana use was assessed by the number of times participants reported using marijuana since they were last texted and the number of minutes they spent smoking since the last text message. A 30-day TLFB calendar was completed at the follow-up.
Results:
We conducted two time-lagged multilevel models to examine whether craving at one assessment point predicted marijuana use at the next EMA instance. Model 1 showed that craving significantly predicted the amount of time (in minutes) participants spent smoking at the next time assessment (F[1,1771] = 1869.35, p < 0.001), with a positive relationship (β = 0.13). Similarly, Model 2 showed that craving positively predicted the number of times participants smoked (frequency) at the next time assessment (F[1,1704] = 11.69, p < 0.001; β = 0.11). A Pearson correlation between the total number of EMA and TLFB marijuana use instances was significant (r = .851, p = .01), but the percentage agreement for direct daily comparisons between the TLFB and EMA showed that only 29% of responses were exact matches. Most of the inconsistencies (66%) showed that participants reported smoking less on the TLFB.
Conclusions:
Our findings suggest that craving positively predicted marijuana use, measured in two different ways (time and frequency). Though there was a relationship between EMA and TLFB marijuana use, our data suggests that TLFB and EMA are probably not equivalent in daily instances reported. Relationships between retrospective and EMA methods should be further evaluated, as there are limitations to assessing past substance use following EMA measurement. Future researchers should consider incorporating innovative EMA methods to determine if findings differ from retrospective report.
Substance use behavior and craving can be difficult to assess, with many studies focusing on retrospective self-report. Past research examining the association between craving and substance use have not always found the two to be related, or if they are related, often the relationship is not particularly strong. In recent years, increased attention has been placed on using ecological momentary assessment (EMA) to learn more about substance use behavior in the moment, though much of this work has focused almost exclusively on tobacco and alcohol use, with less focus on illicit drugs. Several recent EMA studies have shown that craving for marijuana is related to marijuana use.
Objective:
The goal of the current study was to examine whether marijuana craving would predict marijuana use when measured in two different ways through EMA. In addition, we aimed to examine how marijuana use assessed through EMA compared to a commonly used and validated retrospective substance use measure – Timeline Followback (TLFB).
Methods:
College student marijuana users (n = 57) in Colorado (prior to legalization) were recruited to participate in a baseline assessment, two-week EMA, and brief follow-up. To help protect participants' confidentiality, only first names were collected and used throughout the study. Participants averaged 20.05 (SD = 2.60) years of age and were predominantly Caucasian (77%) and Latino (11%). Most were heavy marijuana users, smoking on average 25 days out of the last 30. Participants were sent text messages randomly during three time blocks throughout the day for a two-week period. Overall EMA response rate was 89%. Each Short Message Service (SMS) text included the same nine questions (three questions used for the current analyses). Marijuana craving was assessed on a 1-10 scale (low to high), while marijuana use was assessed by the number of times participants reported using marijuana since they were last texted and the number of minutes they spent smoking since the last text message. A 30-day TLFB calendar was completed at the follow-up.
Results:
We conducted two time-lagged multilevel models to examine whether craving at one assessment point predicted marijuana use at the next EMA instance. Model 1 showed that craving significantly predicted the amount of time (in minutes) participants spent smoking at the next time assessment (F[1,1771] = 1869.35, p < 0.001), with a positive relationship (β = 0.13). Similarly, Model 2 showed that craving positively predicted the number of times participants smoked (frequency) at the next time assessment (F[1,1704] = 11.69, p < 0.001; β = 0.11). A Pearson correlation between the total number of EMA and TLFB marijuana use instances was significant (r = .851, p = .01), but the percentage agreement for direct daily comparisons between the TLFB and EMA showed that only 29% of responses were exact matches. Most of the inconsistencies (66%) showed that participants reported smoking less on the TLFB.
Conclusions:
Our findings suggest that craving positively predicted marijuana use, measured in two different ways (time and frequency). Though there was a relationship between EMA and TLFB marijuana use, our data suggests that TLFB and EMA are probably not equivalent in daily instances reported. Relationships between retrospective and EMA methods should be further evaluated, as there are limitations to assessing past substance use following EMA measurement. Future researchers should consider incorporating innovative EMA methods to determine if findings differ from retrospective report.
Medicine 2.0® is happy to support and promote other conferences and workshops in this area. Contact us to produce, disseminate and promote your conference or workshop under this label and in this event series. In addition, we are always looking for hosts of future World Congresses. Medicine 2.0® is a registered trademark of JMIR Publications Inc., the leading academic ehealth publisher.
This work is licensed under a Creative Commons Attribution 3.0 License.