Introduction
School truancy is defined as a failure to attend and includes behaviors such as skipping classes and unexcused and tardies. School truancy may be a precursor to negative developmental trajectories.1,2 For instance, in one study, 75% of chronically truant adolescents fail to graduate high school in comparison to only 3% of their counterparts.3 Further, risk for substance use and early sexual activity are expected to be higher among this population given that truant youth, compared to their non-truant counterparts, spend more time unsupervised, have lower levels of parental monitoring,4–7 and spend more time with deviant peers,5,6,8,9 three well supported risk factors for problem behavior.
Several studies demonstrate that substance use and sexual behaviors during adolescence often co-occur, and school truancy seems to increase the presence of these behaviors during adolescence. However, few studies have examined the co-occurrence between alcohol use, marijuana use and sexual intercourse, and even fewer have explored daily patterns of this co-occurrence. Thus, the aim of the current study was to: 1) examine the daily co-occurrence of alcohol and marijuana use among a sample of truant adolescents, and, 2) whether the use of alcohol and/or marijuana increased the odds of engaging in sexual intercourse on a given day.
Methods
Participants and procedure
The sample included 76 adolescents (30 female) with a mean age of 15.79 years (see Table 1 for demographics). Participants were recruited between September 2010 and April 2013 as part of a pilot randomized controlled trial which examined the preliminary efficacy of an adolescent and parent motivational enhancement intervention for truant adolescents’ engaging in marijuana use and related problem behaviors.10 This study was reviewed and approved by the university’s Institutional Review Board and only baseline data collected prior to randomization and intervention delivery were used for the current study.
Table 1.
Participant Characteristics and Frequencies of Problem Behaviors
| Variable | Participant Characteristics (n = 76) |
Daily Frequencies (Days = 6840) |
||
|---|---|---|---|---|
| n | % | Days | % a | |
| Age | ||||
| 13 – 14 yrs. | 13 | 17.11% | ||
| 15 – 16 yrs. | 41 | 53.95% | ||
| 17 – 19 yrs. | 22 | 28.95% | ||
| Sex | ||||
| Male | 46 | 60.53% | ||
| Female | 30 | 39.47% | ||
| Race/Ethnicity | ||||
| White (Non-Hispanic) | 34 | 44.74% | ||
| White (Hispanic) | 11 | 14.47% | ||
| Black (Non-Hispanic) | 12 | 15.79% | ||
| Black (Hispanic) | 5 | 6.58% | ||
| Multi-Racial (Non-Hispanic) | 7 | 9.21% | ||
| Multi-Racial (Hispanic) | 9 | 11.84% | ||
| Year in School | ||||
| 7th-8th | 4 | 5.26% | ||
| 9th | 23 | 30.26% | ||
| 10th | 21 | 27.63% | ||
| 11th | 13 | 17.11% | ||
| 12th | 15 | 19.74% | ||
| Truancy | ||||
| Unexcused Absences | 58 | 76.32% | ||
| Skipped Classes | 32 | 42.11% | ||
| Tardies | 54 | 71.05% | ||
| Substance Use | ||||
| Marijuana Use | 76 | 100.00% | 3135 | 45.83% |
| Alcohol Use | 63 | 82.89% | 517 | 7.56% |
| Heavy Drinking (5 or more standard drinks) | 43 | 56.58% | 269 | 3.93% |
| Sexual Intercourse | ||||
| Vaginal and/or Anal Intercourse | 49 | 64.47% | 405 | 5.92% |
| Co-occurring Problem Behaviors | ||||
| Marijuana and Alcohol Use | 63 | 82.89% | 349 | 5.10% |
| Alcohol and Sexual Intercourse | 43 | 56.58% | 66 | 0.96% |
| Marijuana and Sexual Intercourse | 49 | 64.47% | 246 | 3.60% |
| Alcohol, Marijuana, and Sexual Intercourse | 43 | 56.58% | 55 | 0.80% |
Note: Data was collected from the demographic questionnaire and the Timeline Follow-Back.
daily percentages are based on 6840 total days analyzed.
Participants were recruited from the community, including local high schools and family and truancy courts, as well as through advertisements or referrals. To be eligible to participate, adolescents must have: 1) been between the ages of 13 and 19 years old; 2) been living at home with a parent/legal guardian; 3) have used marijuana at least three times in the past 90 days; and 4) engaged in school truancy in the past school year, defined as skipping classes, unexcused absences, unexcused tardies, and/or involvement in truancy court, a family court program designed to reduce truancy rates by monitoring attendance, academic performance, and behavioral issues.
Once parental consent and adolescent assent/consent were obtained, a research assistant administered a baseline assessment to the adolescent, which took about 30 to 40 minutes.
Measures
Timeline Follow-Back (TLFB).11,12
Alcohol use, marijuana use, truancy, and sexual intercourse were assessed using the TLFB’s calendar format, with temporal cues (i.e. holidays and special occasions) to assist in recall of days when adolescents engaged in these behaviors. Alcohol use was recorded as the number of standard drinks adolescents consumed on each day. Marijuana use was recorded as yes or no on each day, as has been done in previous studies due to difficulty assessing quantity of use per occasion.13,14 The TLFB tracked the number of sexual intercourse instances per day, and whether it was unprotected or protected. Finally, truancy was recorded as yes or no on each day.
Data analysis
All analyses were conducted using SAS 9.3 (SAS Institute Inc., Cary, NC). First, frequency of truancy, alcohol use, marijuana use, and sexual intercourse across the 90-day period were examined. Second, Generalized Estimating Equations (GEE) with a logit link function were conducted to test the hypothesis that alcohol and marijuana use (both dichotomized) on a given day would be associated with increased odds of engaging in sexual intercourse on that day. A model examining whether use of both alcohol and marijuana on a given day were associated with increased odds of engaging in sexual intercourse on that day was also evaluated. An autoregressive correlation structure was used for all models as it provided superior fit for most. All models controlled for age, biological sex, and school days versus non-school days.
Models also included percent days of alcohol use, marijuana use, and sexual intercourse as covariates to control for between subject differences in absolute levels of problem behavior engagement. These models allowed us to isolate more clearly the effect of substance use on a given day. Interactions between marijuana use on a given day and frequency of marijuana use, as well as the interaction between alcohol use on a given day and the frequency of alcohol use during the 90-day period were also tested. Significant interactions were interpreted according to procedures established by Aiken and West.15
Results
Frequencies of problem behaviors occurring in the 6840 days analyzed in this study are summarized at the bottom of Table 1. Analyses examining the co-occurrence of substance use and sexual intercourse revealed that: 1) use of marijuana and alcohol co-occurred on the same day in 82.89% of participants; 2) marijuana and sexual intercourse co-occurred on the same day in 64.47% of adolescents; and, 3) alcohol co-occurred with sexual intercourse in 56.58% of adolescents. Occurrence of all three behaviors on the same day also occurred in 56.58% of adolescents. Additionally, unprotected sexual intercourse occurred among 39.47% of adolescents.
Marijuana use, alcohol use, and sexual intercourse
Preliminary GEE models with age, sex, and school days predicting each of the problem behavior variables revealed that age significantly predicted sexual intercourse (OR = 1.52, p < .001) and unprotected sexual intercourse (OR = 1.63, p = 0.002). Additionally, non-school days significantly predicted sexual intercourse (OR = 1.37, p = 0.03), unprotected sexual intercourse (OR = 1.52, p = 0.0485), alcohol use (OR = 2.45, p <.0001), and marijuana use (OR = 1.18, p = 0.0498). All other models were non-significant.
Follow-up GEE models were conducted to investigate the interactions between frequency and use of alcohol and marijuana on a given day (See Table 2). The main effect model demonstrated that alcohol use significantly increased the odds of engaging in sexual intercourse on the same day, OR = 2.62, p = 0.01. There was a significant interaction between frequency of alcohol use in the 90-day period and alcohol use on a given day, p = 0.01, indicating that the association between alcohol use and sex on a given day was moderated by how often one drank during the 90-day period. Among those who used alcohol occasionally (1 standard deviation below the mean), alcohol use on a given day was associated with almost seven times higher odds of also engaging in sexual intercourse on the same day, B = 1.91, SE = 0.41, OR = 6.76, 95% CI [3.02, 15.13], p <.0001. Among frequent alcohol users (1 standard deviation above the mean), alcohol use on a given day was associated with two times higher odds of engaging in sexual intercourse on the same day, B = 0.80, SE = 0.32, OR = 2.23, 95% CI [1.19, 4.21], p = 0.01.
Table 2.
Generalized Estimating Equation for Marijuana and Alcohol Use Predicting Sexual Intercourse
| Variable | B | SE B | OR (95% CI) | p |
|---|---|---|---|---|
| Main Effects | ||||
| Frequency of sexual intercourse | 0.10 | 0.01 | 1.11 (1.09, 1.12) | <.0001 |
| Frequency of marijuana use | −0.01 | 0.003 | 0.99 (0.99, 0.998) | 0.01 |
| Frequency of alcohol use | 0.01 | 0.01 | 1.01 (0.99, 1.04) | 0.22 |
| Time-varying marijuana use | 0.60 | 0.19 | 1.81 (1.24, 2.66) | 0.002 |
| Time-varying alcohol use | 0.96 | 0.35 | 2.62 (1.33, 5.17) | 0.01 |
| Time | 0.004 | 0.01 | 1.00 (0.99, 1.02) | 0.54 |
| Interaction Effects | ||||
| Frequency of marijuana use | −0.001 | 0.003 | 0.999 (0.99, 1.01) | 0.78 |
| Frequency of alcohol use | 0.03 | 0.01 | 1.03 (1.01, 1.06) | 0.01 |
| Time-varying marijuana use | 1.03 | 0.27 | 2.80 (1.65, 4.76) | 0.0001 |
| Time-varying alcohol use | 1.82 | 0.39 | 6.19 (2.89, 13.26) | <.0001 |
| Frequency of marijuana use by time-varying marijuana use | −0.01 | 0.004 | 0.99 (0.98, 0.998) | 0.01 |
| Frequency of alcohol use by time-varying alcohol use | −0.06 | 0.02 | 0.94 (0.90, 0.98) | 0.01 |
| Time | 0.004 | 0.01 | 1.00 (0.99, 1.02) | 0.49 |
Note: Data were collected from the Timeline Follow-Back, and the model controlled for adolescent age, sex, and school days. Time was coded as which day the use occurred on during the 90 day period ranging from 1 to 90.
Marijuana use increased the odds of engaging in sexual intercourse on the same day by almost twofold, OR = 1.81, p = 0.002. There was a significant interaction between the frequency of individual marijuana use during the 90-day period and marijuana use on a given day, p = 0.01. Among occasional marijuana users (1 standard deviation below the mean), marijuana use on a given day was associated with two and a half times higher odds of engaging in sexual intercourse on the same day, B = 0.91, SE = 0.24, OR = 2.49, 95% CI [1.57, 3.96], p = 0.0001. However, among those who used marijuana frequently (1 standard deviation above the mean), marijuana use on a given day was not significantly predictive of also engaging in sexual intercourse on the same day, B = 0.08, SE = 0.24, OR = 1.09, 95% CI [0.68, 1.75], p = 0.73. The association of these variables in the interaction model remained constant over the 90-day period, p = 0.49.
Analyses were also conducted excluding participants who were not sexually active in the 90-day period. Results among the sexually active sample were stronger, yet in the same direction as those reported above. However, due to the relatively low number of participants (n=49, 4410 observations) that were included in this model, we only focus on results for the overall sample.
Analyses were also conducted examining the odds of engaging in unprotected sexual intercourse on a given day. Unlike the results for any sexual intercourse, alcohol use did not predict unprotected sex on a given day (p = 0.15) nor did alcohol use frequency during the 90-day period (p = 0.15). However, marijuana use on a given day significantly increased the odds of engaging in unprotected sexual intercourse on the same day, B = 0.72, SE = 0.28, OR = 2.05, 95% CI [1.19, 3.58], (p =0.01). Yet, frequency of individual marijuana use during the 90-day period significantly decreased the odds of engaging in unprotected sexual intercourse on the same day, OR = 0.98, p < .0001. Results remained constant over the 90-day period, p =0.96 (See Table 3). Moreover, there were no interactions between frequency of use in the 90-day period and use on a given day for either alcohol or marijuana.
Table 3.
Generalized Estimating Equation for Marijuana and Alcohol Use Predicting Unprotected Sexual Intercourse
| Variable | B | SE B | OR (95% CI) | p |
|---|---|---|---|---|
| Main Effects | ||||
| Frequency of sexual intercourse | 0.11 | 0.01 | 1.11 (1.09, 1.13) | <.0001 |
| Frequency of marijuana use | −0.02 | 0.01 | 0.98 (0.96, 0.99) | <.0001 |
| Frequency of alcohol use | 0.05 | 0.03 | 1.05 (0.98, 1.12) | 0.15 |
| Marijuana use on a given day | 0.72 | 0.28 | 2.06 (1.19, 3.58) | 0.01 |
| Alcohol use on a given day | 0.60 | 0.42 | 1.82 (0.80, 4.12) | 0.15 |
| Time | 0.001 | 0.01 | 1.00 (0.98, 1.02) | 0.96 |
Note: Data were collected from the Timeline Follow-Back, and the model controlled for adolescent age, sex, and school days. Time was coded as which day the use occurred on during the 90 day period ranging from 1 to 90.
Discussion
The majority of truant participants, about 8 out of 10, reported they used both marijuana and alcohol. Further, marijuana and sexual intercourse co-occurred in about two-thirds of adolescents, and, alcohol co-occurred with sexual intercourse in a little over half of the adolescents. Occurrence of all three behaviors on the same day also occurred in about half of the adolescents. Interestingly, the results also indicated that these behaviors seem to occur over the weekend rather than on days when adolescents were truant from school. Results supported all of the substance use hypotheses, indicating that both alcohol and marijuana significantly increase the odds of engaging in sexual intercourse on the same day. However, analyses revealed that these associations varied by frequency of use.
Our analysis indicated that only marijuana use significantly increased the odds of engaging in unprotected sexual intercourse on the same day. However, it is important to note that our results may have been influenced by the number of adolescents who reported engaging in drinking versus marijuana use, and the even smaller number who reported engaging in unprotected sexual intercourse (n = 30, 178 observations).
Despite the small sample size in the current study and the limitations in collecting recall data, there are important implications which can inform future directions. With discussions regarding the legalization of marijuana use on the rise, the results of this study highlight the importance of future research distinguishing between risks among those who use occasionally versus those who use frequently. More importantly, as has been show before,5 the results demonstrate that problem behaviors do not occur independently, and highlight the importance of examining the underlying role other problem behaviors play in sustaining the problem behavior of concern. These findings also suggest that truant adolescents exhibiting any one of these problem behaviors may be at an increased risk for escalating and engaging in the other problem behaviors. While our study was not able to examine the temporal ordering of the problem behaviors on a given day, such analysis would be important for the development of prevention programs. Alternatively, examining whether targeting any one of these behaviors in a preventative intervention with truant adolescents reduces the occurrence of other related behaviors is also worth exploring in future studies.
Acknowledgments
Sources of Funding: National Institute on Drug Abuse, R34 DA029871
National Institute on Alcohol Abuse and Alcoholism, K01 AA021081
National Institute on Alcohol Abuse and Alcoholism, P01 AA019072
These funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
No conflicts of interest exist for any of the authors.
References
- 1.Garry EM. Truancy: First step to a lifetime of problems. OJJDP: Juvenile Justice Bulletin. 1996;October:1–7. [Google Scholar]
- 2.Baker ML, Sigmon JN, Nugent ME. Truancy reduction: Keeping students in school. Bulletin of the OJJDP. 2001;September:1–15. [Google Scholar]
- 3.Barry AE, Chaney B, Chaney JD. The impact of truant and alcohol-related behavior on educational aspirations: a study of US high school seniors. J Sch Health. 2011;81(8):485–492. [DOI] [PubMed] [Google Scholar]
- 4.Henry KL. Who’s skipping school: Characteristics of truants in 8th and 10th grade. J Sch Health. 2007;77(1):29–35. [DOI] [PubMed] [Google Scholar]
- 5.Henry KL, Thornberry KL. Truancy and escalation of substance use during adolescence. J Stud Alcohol Drugs. 2010;71(1):115–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Henry KL, Thornberry TP, Huizinga DH. A discrete-time survival analysis of the relationship between truancy and the onset of marijuana use. J Stud Alcohol Drugs. 2009; 70(7):5–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Maynard BR, Salas-Wright CP, Vaughn MG, Peters KE. Who are truant youth? Examining distinctive profiles of truant youth using latent profile analysis. J Youth Adolesc 2012;41:1671–1648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Somers CL, Gizzi TJ. Predicting adolescents’ risky behaviors: The influence of future orientatin, school involvement, and school attachment. Adolesc Fam Health. 2001;2(1):3–11. [Google Scholar]
- 9.Houck CD, Hadley W, Tolou-Shams M, Brown L. Truancy is associated with sexual risk among early adolescents. J Dev Behav Pediatr 2012;33(9):728–731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Spirito A, Hernandez L, Cancilliere MK, et al. Parent and Adolescent Motivational Enhancement Intervention for Substance-Using, Truant Adolescents: A Pilot Randomized Trial. J Clin Child Adolesc Psychol 2017:1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sobell LC, Sobell MB. Timeline Follow-back: A technique for assessing self-reported ethanol consumption In: Allen J, Litten RZ, eds. Measuring Alcohol Consumption: Psychosocial and Biological Methods Totowa, NJ: Humana Press; 1992:41–72. [Google Scholar]
- 12.Sobell LC, Sobell MB. Alcohol timeline followback (TLFB) users’ manual Toronto: Addiction Research Foundation; 1996. [Google Scholar]
- 13.Shrier LA, Walls C, Lops C, Kendalla AD, Blood EA. Substance use, sexual intercourse, and condom nonuse among depressed adolescents and young adults. J Adolesc Health. 2012;50(3):264–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hendershot CS, Magnan RE, Bryan AD. Associations of marijuana use and sex-related marijuana expectancies with HIV/STD risk behavior in high-risk adolescents. Psychol Addict Behav 2010;24(3):404–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Aiken LS, West SG. Multiple regression: Testing and interpreting interactions. Newbury Park: Sage; 1991. [Google Scholar]