Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 1.
Published in final edited form as: Addict Behav. 2020 Dec 2;114:106759. doi: 10.1016/j.addbeh.2020.106759

The relationship between alcohol and marijuana use with nonsuicidal self-injury among adolescent inpatients: Examining the 90 days prior to psychiatric hospitalization

Christina M Sellers 1,2, Antonia Díaz-Valdés 3, Michelle M Oliver 1,2, Kevin M Simon 1,4,5, Kimberly H McManama O’Brien 1,4
PMCID: PMC7792986  NIHMSID: NIHMS1651298  PMID: 33338906

Abstract

Background:

This study examined the trajectories of alcohol use, cannabis use, suicide planning (SP), and nonsuicidal self-injury (NSSI) prior to hospitalization and examined the role of alcohol and cannabis use, independently and jointly, in predicting NSSI on a daily level and over time.

Methods:

Participants included 71 adolescents hospitalized for suicide risk (75% female; 25% male; Mage = 15.79). All participants drank alcohol at least once in the prior 90-days. We conducted mixed effect models to assess the trajectories of alcohol use, cannabis use, and NSSI over the 90-days prior hospitalization. To test the effect of SP, alcohol use, and cannabis use on NSSI, we conducted logistic random effect models, while controlling for demographics.

Results:

SP (OR=4.47, p<0.001) and SI (OR=10.09, p<0.001) significantly increased the odds of engaging in NSSI. Neither cannabis nor alcohol use independently predicted the odds of engaging in NSSI, however, the co-occurrence of alcohol and cannabis use increased the odds of engaging in NSSI on a given day (OR=30.5, p<0.05).

Conclusions:

Study findings extend current knowledge about the longitudinal and day-to-day relationships between alcohol and cannabis use and NSSI. Results underscore the importance of developing interventions that address polysubstance use among suicidal adolescents engaging in NSSI.

Keywords: nonsuicidal self-injury, suicide attempts, adolescents, alcohol, cannabis

1. Introduction

Adolescents psychiatrically hospitalized following engagement of self-injurious behaviors are at high risk for reengagement of dangerous behaviors or escalation towards suicide attempts (SA) and/or death. Additionally, substance use is consistently implicated as a risk factor for risky behaviors. Accidents and suicide are the first and second leading causes of death in adolescents (CDC, 2018). Despite this knowledge, there have been few studies that have investigated the longitudinal relationship of substance use, nonsuicidal self-injury (NSSI), and psychiatric hospitalization specifically. Improving assessment of risk factors for adolescent self-injurious behaviors, including NSSI and substance use and their relationship to psychiatric hospitalization is a critical public health priority particularly as there has been a steady rise in the use of emergency room visits related to mental health emergencies since 2007 (Lo, Bridge, Shi, Ludwig, & Stanley, 2020).

1.1. NSSI

Nonsuicidal Self-Injury (NSSI) is an emergent problem in adolescents (Cunningham, Goldstick, & Carter, 2019; Hawton, Saunders, & O’Connor, 2012; Heron, 2019). NSSI is defined as direct and deliberate destruction of body tissue without suicidal intent (Lloyd-Richardson, Perrine, Dierker, & Kelley, 2007; Nock, 2010). The most common forms of NSSI are cutting, burning, biting, or head banging (Plener et al., 2018). Adolescent rates of NSSI are twice that of adult NSSI rates (E. D. Klonsky, 2011; Nock, 2010). While roughly 5% of the adult population engage in NSSI, over 18% of non-clinical adolescents engage in NSSI (E. D. Klonsky, 2011; Nock, 2010; Swannell, Martin, Page, Hasking, & St John, 2014). In clinically referred populations NSSI is more frequent and severe than in community samples. In addition, in clinically referred populations NSSI is associated with more severe psychopathology with upwards of 50% of the adolescent clinic population engaging in NSSI (Asarnow et al., 2011). NSSI appears to be cross-sectionally associated with suicide attempts (SA) and also predicts future SA and suicide in adulthood (Tuisku et al., 2013). Evidence suggests that NSSI is one of the strongest predictors of SA and suicide deaths, as it increases ones capability of suicide by lowering pain sensitivity and fear of death (Whitlock et al., 2013; Willoughby, Heffer, & Hamza, 2015). Recent reports highlight emergency room visits for NSSI increased 329% nationally between 2007 - 2016 (Lo et al., 2020). Because NSSI is typically associated with emotional distress and increases risk for suicide, it is crucial to establish accurate conceptual and clinical models of this behavior.

1.2. Substance Use

Identifying risk factors for NSSI is key to helping clinicians better identify at risk individuals and for the development of interventions (Fox et al., 2016). Substance use is a risk factor in adolescents engaging in NSSI (Brausch & Boone, 2015;). Among adolescents, the top three substances (in rank order) are alcohol, cannabis, and tobacco use (including e-cigarettes) (Chadi, Schroeder, Jensen, & Levy, 2019; Johnston et al., 2019; Levy, Campbell, Shea, & Du Pont, 2018; Wang et al., 2019). Recent evidence suggests that nationally over the last decade (2006-2017), among youth who visited pediatric emergency departments, there was a 75% increase in substance use disorders (Lo et al., 2020).

1.2.1. Alcohol Use

By senior year of high school, more than half (60%) of students reported having ever tried alcohol in their life, 43% of whom reported ever being drunk (Johnston et al., 2019). Nearly a quarter (24%) reported having initiated alcohol use by the end of middle school (Johnston et al., 2019). Evidence has established that alcohol consumption is a robust risk factor for NSSI (Rossow et al., 2007). Notably, one third of adolescents on a psychiatric unit with a history of NSSI carry a diagnosis of alcohol use disorder (Nock et al., 2006), and alcohol use is prospectively associated with NSSI among adolescents (Tuisku et al., 2013).

1.2.2. Cannabis Use

Cannabis continues to be the most used illicit drug among adolescents in the US and the leading reason for entering SUD treatment, with use typically initiated in middle to late adolescence (Johnson, 2018). Cannabis can be used by smoking, vaping, eating or dabbing (i.e., smoking or inhaling marijuana in the form of hash oil or wax). Nationally, 14% of 8th grade youth and 33% of 10th grade youth have tried cannabis and 6% of 8th grade youth and 17% of 10th grade youth have used in the past 30 days (Johnston et al., 2019). Lifetime use of cannabis (44%) has eclipsed cigarette use (24%) among high school seniors (Johnston et al., 2019; Levy et al., 2018). Further, there is a strong relationship between alcohol use and cannabis use, in that adolescents who use cannabis are more likely to use alcohol compared to teens who do not use cannabis (Patrick, Schulenberg, O’Malley, Johnston, & Bachman, 2011). In a study of psychiatrically hospitalized adolescents, over 12% of those engaging in NSSI qualified for a cannabis abuse diagnosis and almost 30% met criteria for cannabis dependence (Nock et al., 2006). Evidence suggests that cannabis use affects judgment in the adolescent population that can lead to impairments in learning, decision making, cognitive functioning and increase risky behaviors (Morin et al., 2019; S. L. Moss, Santaella-Tenorio, Mauro, Keyes, & Martins, 2019). Additionally concerning, there has been a steady rise in pediatric emergency room visits related to cannabis use and self-injurious behaviors (Chen & Klig, 2019).

1.3. Polysubstance Use

Adolescent polysubstance use may signal increased vulnerability to a variety of negative health outcomes. Specifically, persons who engage in polysubstance use are more likely than their peers who engage in single-substance use to engage in problematic drug use, relapse from substance use treatment, engage in violence, and unsafe sexual behaviors, and experience serious mental illness in adulthood (Ahmadi-Montecalvo et al., 2019; Felton, Kofler, Lopez, Saunders, & Kilpatrick, 2015; Hopfer, Tan, & Wylie, 2014). The prevalence of polysubstance use in teens has been suggested to range between 15%-39% of the teenagers who engage in substance use depending on study sample and how polysubstance is defined (Conway et al., 2013; Coulter, Ware, Fish, & Plankey, 2019). A latent class analysis of a nationally representative group of 10th graders found 8% were “predominantly polysubstance users” (Conway et al., 2013). More recent data from a 2013 nationally representative sample confirmed an elevated prevalence (34%) of early adolescent polysubstance use most strongly associated with a spectrum of young adult substance use challenges and DSM criteria defining substance use disorders (H. B. Moss, Chen, & Yi, 2014). Then another national study found 3.6% of teens to be “medium frequency users” of 3 or more substances, while 4.1% were “high frequency users” (Coulter et al., 2019). One study conducted with adolescents in outpatient psychiatric treatment in Turkey, found that polysubstance use was associated with a greater likelihood of NSSI and history of suicide attempt compared to non-polysubstance use (Guvendeger Doksat, Zahmacioglu, Ciftci Demirci, Kocaman, & Erdogan, 2017). While it is not entirely clear, what types of polysubstance use are predominant among adolescents, and how they can be predicted and their relationship to NSSI, Schneider et. al. (2020) most recently characterized lifetime polysubstance use among high school students, in Baltimore, Maryland using latent class analysis demonstrating polysubstance use was common among adolescent who use substances and three primary patterns of substance use: marijuana and alcohol use, polysubstance use, and alcohol, pain medication, and inhalant use (Schneider et al., 2020). From a prevention standpoint, it seems therefore indispensable to investigate the role of single and polysubstance use in relation to NSSI.

1.4. Conceptual Framework

An emotion dysregulation model provides the framework for our conceptualization of NSSI and substance use among adolescents. Emotion dysregulation is defined as the failure to regulate negative emotional states, particularly with regards to anger, sadness, rage, and anxiety (Beauchaine, 2015; Linehan, 1993). A recent systematic review and meta-analysis found that greater emotion dysregulation is associated with a higher risk of NSSI (Wolff et al., 2019). NSSI can have an emotion regulating effect by relieving intolerable feelings (Linehan, 1993). Similarly, substance use can aid in the regulation of unwanted negative emotions (Gratz & Tull, 2010; Linehan, 1993). Individuals with substance use disorders have been shown to have increased emotion dysregulation. In particular, individuals with co-occuring disorders, such as substance use and NSSI, can have greater difficulty regulating emotions (Buckholdt et al., 2015; Gratz & Tull, 2010).

1.5. Purpose of Study

Although prior studies have identified the cross-sectional relationship between alcohol and cannabis use with NSSI among adolescents, research on the longitudinal relationship is sparse and even less research exists on the relationship between polysubstance use and NSSI. Additional research is needed given these phenomena are increasing in prevalence among adolescents. In addition, past studies are limited in a variety of ways requiring additional inquiry into this area. First, prior research has been limited to cross-sectional associations. Second, research has not assessed the effects of alcohol and cannabis use on NSSI prior to inpatient psychiatric hospitalization, which represents a vulnerable and risky time period among adolescents and is critical to study in order to inform prevention efforts. Third, no research to our knowledge has examined these relationships through fine-grained analyses at the daily level allowing scholars to consider the dynamic nature of alcohol and cannabis use as risk factors for increased NSSI. This study tested the daily relationships among adolescents’ alcohol, cannabis, and polysubstance use and NSSI by retrospectively assessing the presence of these variables in the three months prior to inpatient psychiatric hospitalization.

The purpose of this study was to first, examine the retrospective trajectories of alcohol use, cannabis use, and NSSI leading up to psychiatric hospitalization, and second, to examine the role of alcohol and cannabis use, independently and jointly, in predicting NSSI through the use of retrospective data. Given our use of the emotion dysregulation model, we hypothesized that alcohol use, cannabis use, and NSSI would increase leading up to inpatient psychiatric hospitalization. We also hypothesized that alcohol and cannabis use would predict NSSI on a given day, and over time.

2. Materials and Methods

2.1. Participants

This study used baseline data from two larger randomized control trials testing the feasibility and acceptability of a brief alcohol and suicide intervention and mHealth booster for adolescents psychiatrically hospitalized following a suicidal event (BLIND FOR REVIEW). All participants were recruited from the inpatient psychiatric unit of an urban general hospital in the northeastern United States. Eligibility criteria included: hospitalization following a suicidal event, history of alcohol use, age 13-17, and the ability to speak and understand English. Exclusion criteria included psychosis or state custody. The study was approved by the overseeing hospital’s Institutional Review Board and all participants provided informed assent/consent.

Participants in this study included 71 adolescents (75% female; Mage = 15.79 SD = 1.00). The majority of participants identified as White (66%) and Non-Hispanic/Latinx (77%). Table 1 provides demographic information for the sample.

Table 1.

Descriptive Statistics for Time-Invariant and Time Varying Variables

Time-Invariant Variables % (n) / M(SD)
Gender
 Male 25% (18)
 Female 75% (53)
Sexual Orientation
 Sexual Minority 32% (23)
 Heterosexual 68% (48)
Race
 White 66% (47)
 Non-White 34% (24)
Ethnicity
 Non-Hispanic 77% (55)
 Hispanic 23% (16)
 Age 15.79 (1.00)
Time-Varying Variables Between % Within %
 Suicidal Ideation 98.55% (68) 32.68%
 Suicide Planning 59.42% (41) 4.23%
 NSSI 56.34% (40) 10.81%
 Marijuana Use 77.46% (55) 24.85%
 Alcohol Consumption 97.18% (69) 7.81%
 Hospitalization 87.32% (62) 4.68%
Open in a new tab

2.2. Measures

Face-to-face interviews were used to collect study data during the adolescents’ inpatient psychiatric hospitalization. All data was collected between the first 2-7 days of the adolescents’ hospitalization. We used a modified version of the Timeline Followback Calendar (TLFB; Sobell & Sobell, 1992) to retrospectively record alcohol use (1=drank that day, 0=did not drink that day), cannabis use (1=used cannabis that day, 0=did not use cannabis that day), and suicide ideation (SI) (1=SI on that day, 0=no SI on that day), suicide planning (SP) (1=SP that day, 0=no SP that day), suicide attempts (SA) (1=SA on that day, 0=no SA on that day), and NSSI (1=NSSI on that day, 0=no NSSI on that day) over the 90-day period prior to inpatient hospitalization. The TLFB calendar provides a retrospective report over the past three months by using a calendar as way to remember days in which substances were use and suicide related thoughts and behaviors occurred. Specifically, a trained research assistant works collaboratively with the participant in order to identify anchors on the calendar to assist with retrospective recall prior to the start of data collection. These anchors include holidays, birthdays, and any other significant dates that are specific to the participant (i.e., school vacations, anniversaries with significant others, parties attended etc.). Previous studies have shown that the TLFB has yielded high test-retest reliability for alcohol use (Sobell & Sobell, 1992), good convergent validity with adolescent treatment samples(Waldron, Slesnick, Brody, Turner, & Peterson, 2001), and good test-retest reliability for adolescents self-report specifically for the prior 90 days with intra class correlations ranging from 0.83-0.92 (Levy et al., 2004).

For this study, we defined a SA as “a self-inflected, potentially injurious behavior with a nonfatal outcome for which there is evidence (either explicit or implicit) of intent to die” (Silverman, Berman, Sanddal, O’Carroll, & Joiner, 2007, p. 273). We defined SP as “a proposed method of carrying out a design that will lead to a potentially self-injurious outcome” (Silverman et al., 2007, p. 268). NSSI was operationalized as “a self-inflected, potentially injurious behavior for which there is evidence (either implicit or explicit) that the person did not intend to kill himself/herself (i.e., had no intent to die)” (Silverman et al., 2007, p. 272). Examples of NSSI within this sample included cutting, headbanging, and burning.

Age in years, gender (1=male, 0=female; no participants identified as non-binary, transgender participants are coded with the gender they specified), race (1=White, 0=Non-White), ethnicity (1=Hispanic or Latino, 0=Non-Hispanic or Latino), sexual orientation (1=Heterosexual, 0=Sexual Minority), and hospitalizations (1=hospitalized on that day, 0=not hospitalized on that day) were included as control variables. We controlled for hospitalizations as some participants may have been hospitalized for other reasons prior to the current inpatient psychiatric hospitalization. While hospitalized, participants are less able to use substances or attempt suicide. Age was grand mean centered.

2.3. Analytic Approach

Data were analyzed with a longitudinal multilevel model using Stata SE 15.1 given the nested structure of the data with days nested within person. First, univariate descriptive statistics for time-variant and time-invariant variables were calculated. Second, mixed effect models were estimated to evaluate the trajectory of SP, NSSI, alcohol use, and cannabis use, over the 90-days prior to inpatient psychiatric hospitalization. Additionally, random effect models were conducted to test the effect of SP, alcohol use, and cannabis use on NSSI. As indicated by the Hausman tests, random effect models are appropriate as there were no significant differences on the between and within effects (Chi2=5.39, p>0.05). To test study hypotheses, first a model including only time-variant variables; SP, SI, cannabis use, alcohol use, and hospitalization, to predict NSSI, was conducted. Second a model including time-variant variables along with control variables (time-invariant variables) – age and gender, was conducted. A third model with all time-variant and time-invariant variables, along with an interaction between cannabis and alcohol use was conducted to measure polysubstance use. Finally, all significant variables were included in a parsimonious model. All models were estimated using LaPlace method of integration, Bernoulli distribution, full maximum likelihood method of estimation and logit link function (Azevedo-Filho & Shachter, 1994; Kassahun, Neyens, Molenberghs, Faes, & Verbeke, 2012; Kim, Choi, & Emery, 2013). Unit-specific models are reported in table 2 (Kim et al., 2013; Rabe-Hesketh & Skrondal, 2012).

Table 2.

Odds Ratios and standard Error of Random Effect Models of SP, alcohol and marijuana use on NSSI, controlling by gender, race, age, sexual orientation, SI and hospitalization.

Model 1 Model 2 Model 3 Model 4
Suicidal Ideation 10.13*** (1.97) 9.99 *** (1.94) 9.92*** (1.93) 10.09*** (1.95)
Suicide Planning 4.45*** (1.44) 4.52*** (1.46) 4.55*** (1.479 4.47*** (1.43)
Marijuana Use 1.23 (0.34) 1.33 (0.37) 1.05 (0.33) 1.04 (0.3)
Alcohol Use 0.96 (0.24) 0.94 (0.239 0.70 (0.21) 0.70 (0.21)
Alcohol*Marijuana 3.03* (1.73) 3.05* (1.75)
Hospitalization 0.99 (0.30) 0.98 (0.30) 0.95 (0.29)
Gender (1=female) 3.98 (3.00) 3.76 (2.84) 4.49* (3.13)
Sexual orientation (1=heterosexual) 0.74 (0.49) 0.75 (0.49)
Race (1=White) 0.89 (0.60) 0.87 (0.59)
Ethnicity(1=Hispanic) 2.17 (1.80) 2.14 (1.76)
 Age 0.96 (0.30) 0.97 (0.30)
Goodness-of-fit Statistics
AIC 1879.71 1883.97 1882.19 1875.84
BIC 1927.04 1965.11 1970.10 1929.94
ICC 0.62 0.59 0.59 0.59
Open in a new tab
*

p<0.05

**

p<.01

***

p<.001

3. Results

3.1. Univariate Results

As depicted in table 1, 56.34% (n=40) of the participants in this study endorsed NSSI at least one day in the 90 days prior to hospitalization. Participants reported NSSI on average, 10 of the 90 days (10.83% of the time). During that same 90-day period, 98.59% (n=70) endorsed SI- and 60.56% of participants (n=43) made a SP- on at least one day. With respect to substance use, 77.46% of participants (n=55) used cannabis at least once, and on average, those who used cannabis used on an average of 23 of the 90-days (24.85% of the time). For alcohol, 97.10% (n=69) drank at least one day, and those who used alcohol reported using on an average of 7 of the 90-days (7.81% of the time). Additionally, 55% of adolescents (n=38) consumed alcohol and marijuana on the same day, and those who used both alcohol and marijuana reported using an average of 2.61 days of the 90-days period (2.87%). Lastly, 87.32% of participants (n=62) reported they were psychiatrically hospitalized for at least one of the 90-days leading up to their current inpatient psychiatric hospitalization, for an average of 4 of the 90-days (4.68% of the time).

3.2. Unconditional Models and Intraclass Correlations

Prior to testing hypotheses, an unconditional model was examined to calculate the intraclass correlation (ICC) of 0.62, indicating that about 62% of the variance in odds of attempting suicide was due to between-person differences. Results indicated a good reliability, and that a multilevel model was appropriate.

3.3. Trajectories

The average odds of NSSI (OR=1.007, p=0.06), alcohol use (OR=1.005, p=0.09), and cannabis use (OR=1.00, p=0.72) remained stable across the 90 days leading up to psychiatric hospitalization. However, the average odds of SP increased significantly, though modestly, across the same time period (OR=1.05, p=0.000). For each day, there was a 4% increase in the odds of developing or significantly altering a SP. Although the trajectories of NSSI and alcohol use did not statistically significantly change over the 90 days, the trajectories were trending upward, and were marginally significant at the 90% level of significance.

3.4. Effects of SI and SP on Odds of Same-Day NSSI

Consistently, across all four models, SI and SP were significant predictors of NSSI. Both were associated with increased odds of NSSI on a given day. As depicted in model 4, those who experienced SI were 9 times more likely (OR=10.09, p<0.001) to engage in NSSI than those who did not have SI, while controlling for cannabis use, alcohol use, and gender. Similarly, those who made a SP were almost 3.5 times more likely to have NSSI on a given day (OR=4.47, p<0.001), relative to those who did not make a SP on that same day.

3.5. Effects of Alcohol and Cannabis Use on Odds of Same-Day NSSI

As depicted in Table 2, the independent use of alcohol and cannabis were not significantly predictive of NSSI (ps>0.05). However, their combined use increased the odds (OR=30.5, p<0.05) of NSSI on a given day, while controlling for other variables, such that when respondents used alcohol and cannabis on the same day, they were 29 times more likely to engage in NSSI on that same day.

4. Discussion

This study investigated trajectories of alcohol use, cannabis use, SP, and NSSI in the 90 days prior to inpatient hospitalization of suicidal adolescents and aimed to understand the role of alcohol and cannabis use in predicting NSSI. Results from this study partially supported our hypotheses. First, alcohol use, cannabis use, and NSSI remained stable across the 90 days leading up to psychiatric hospitalization. Although the trajectories of NSSI and alcohol use did not statistically significantly change over the 90 days, the trajectories were trending upward. Second, neither cannabis nor alcohol use were independently associated with increased odds of engaging in NSSI, however, the combined use of alcohol and cannabis was associated with increased the odds of engaging in NSSI. Specifically, on days when adolescents used both alcohol and cannabis, they were 29 times more likely to engage in NSSI that same day.

Although NSSI and alcohol use remained relatively stable across the 90 days prior to inpatient psychiatric hospitalization, they were trending upward. This finding is not surprising because prior research has indicated that among youth with depression, NSSI tends to be relatively stable in nature (Barrocas, Giletta, Hankin, Prinstein, & Abela, 2014). The nature of the upward trajectory, however, indicates a need to take these factors into account clinically when a gradual increase occurs as it may point to an increased risk for SA and a possible need for an earlier clinical intervention.

The independent effects of alcohol and cannabis use were not associated with increased risk for NSSI on the same day or over time. Although previous research has indicated cross-sectionally that alcohol and cannabis use are risk factors for NSSI (Nock et al., 2006), our sample is unique in that it includes adolescents psychiatrically hospitalized who reported drinking alcohol at least once in the prior month upon admission. It is possible that our findings may not have to do solely with alcohol or cannabis in and of themselves, but rather be better explained as the use of cannabis as an additional substance combined with a history of alcohol use. This possible explanation is consistent with our findings on the combined effect of alcohol and cannabis on a given day in predicting NSSI that same day.

Moreover, although research has demonstrated the long-term effects of cannabis use on suicidal and nonsuicidal behavior, limited research has examined the acute effects of cannabis use making our understanding of the acute effects of cannabis on NSSI less comprehensive and clear. In a recent meta-analysis the authors identified only a single study that linked acute marijuana use with SI (Guilherme Borges, Bagge, & Orozco, 2016). However, NSSI is conceptually different from SI as there is a lack of suicidal intent. Consequently, it is difficult to draw hypotheses based on literature examining the effects of cannabis on suicidal thoughts and it is possible that the effects of cannabis differ for NSSI.

It is also possible that the relationship between cannabis and NSSI was not found in this study because it may be that for some of the sample cannabis increased risk for NSSI, while for others it buffered against NSSI. Although we did not explicitly test this hypothesis, in addition to past research indicating cannabis as a risk factor for NSSI (Nock et al., 2006), other research has identified that NSSI can be used to effectively alleviate individuals’ negative emotional states (Brausch & Muehlenkamp, 2018; E. David Klonsky, Glenn, Styer, Olino, & Washburn, 2015; Nock & Prinstein, 2004; Taylor et al., 2018). Both NSSI and cannabis use can be used to cope with negative emotions. When faced with negative emotions NSSI is often used to help regulate or relieve negative feelings. Cannabis similarly can be used to escape or numb overwhelming negative feelings (Zimmermann et al., 2017). Future research should consider inquiry regarding the functions of NSSI and cannabis. Importantly, attempting to draw direct relationships between substance use and NSSI is challenging due to genetic and/or familial environments, whereas person-specific factors that may or may not be of a causal nature contributed to the correlation between substance use and NSSI (Few et al., 2016).

Lastly, our study found that on days when youth used both alcohol and cannabis, they were 29 times more likely to engage in same day NSSI. This is interesting given the independent effects of alcohol and cannabis were not significant predictors, however, it is consistent with prior research which has indicated that risk taking is often maximized by using alcohol and cannabis simultaneously (Li et al., 2012; Metrik, Caswell, Magill, Monti, & Kahler, 2016; Volkow, Baler, Compton, & Weiss, 2014). It is possible that adolescents who use both alcohol and cannabis may be higher risk in general when compared to those who use a single substance. The use of multiple substances on the same day may be indicative of participants’ level of distress on that day leading to the increased likelihood of engaging in NSSI. If their level of distress is so high that multiple substances do not achieve the desired relief the next step may be to use NSSI to regulate their negative state. It is also possible that the combined mood-altering effects of alcohol and cannabis results in a greater dysphoric mood or lower distress tolerance, increasing the likelihood of engaging in NSSI. In fact, prior research has found this to be true when looking at substance use and suicide related variables. Specifically, one study found when individuals use a single substance (compared to none), they were 2.6 times more likely to attempt suicide, and 1.6 times more likely to endorse SI, and when they use two substances, the odds of SI and SA significantly increase (Borges, Walters, & Kessler, 2000). Other research has suggested a “developmental cascade” where symptoms of depression promote symptoms of substance misuse, particularly with polysubstance use, in a reciprocal fashion overtime (Felton et al., 2015). These research findings help us to understanding why polysubstance use is associated with higher risk of NSSI among youth.

It is important to note that our findings should be interpreted with caution given our use of a small and high-risk sample of adolescents. Consequently, the results of this study are not generalizable to the larger population of suicidal adolescents and results must be interpreted specifically to a high-risk clinical sample of suicidal adolescents with a history of alcohol use. A second limitation is that we retrospectively examined substance use and NSSI leading up to the psychiatric hospitalization. Our use of retrospective data and lack of temporal ordering of alcohol use, cannabis use, and NSSI on a given day, limit our ability to make causal statements despite the longitudinal nature of the data. Despite these limitations, this study utilized daily level data to assess the relationship among distinct substance- and self-injury-related variables on a fine-grained level. This is a strength of our study, as researchers frequently aggregate data which fails to allow for analysis of day-today variability.

To conclude, our findings extend prior research by focusing on the risk that alcohol and cannabis use confer on NSSI on a given day. The potential elevation of risk of co-occurring alcohol and cannabis use suggests the importance of practitioners’ awareness of adolescents’ ongoing polysubstance use behaviors and their understanding of how they relate to NSSI and thus confer risk for suicide. Obtaining information that an adolescent uses both alcohol and cannabis should signal to the practitioner that more stringent safety measures and perhaps a higher level of intervention may be warranted, depending on the frequency and severity of the NSSI.

Highlights:

  • Suicide planning and suicidal ideation significantly increased the odds of engaging in same day nonsuicidal self-injury (NSSI).

  • Neither cannabis nor alcohol use independently predicted the odds of same day NSSI.

  • The co-occurrence of alcohol and cannabis use increased the odds of same day NSSI.

Acknowledgments

Funding: This work was supported by the American Foundation for Suicide Prevention [YIG-1-097-13, PI: O’Brien, Mentor: Spirito] and the National institute on Alcohol Abuse and Alcoholism [R34AA025763, PI: O’Brien]

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Ahmadi-Montecalvo H, Lilly CL, Zullig KJ, Jarrett T, Cottrell LA, & Dino GA (2019). A latent class analysis of the co-occurrence of risk behaviors among adolescents. American Journal of Health Behavior, 43(3), 449–463. 10.5993/AJHB.43.3.1 [DOI] [PubMed] [Google Scholar]
  2. Asarnow JR, Porta G, Spirito A, Emslie G, Clarke G, Wagner KD, … Brent DA. (2011). Suicide attempts and nonsuicidal self-injury in the treatment of resistant depression in adolescents: Findings from the TORDIA study. Journal of the American Academy of Child and Adolescent Psychiatry. 10.1016/j.jaac.2011.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Azevedo-Filho A, & Shachter RD (1994). Laplace’s Method Approximations for Probabilistic Inference in Belief Networks with Continuous Variables. [Google Scholar]
  4. Bachman JG, Johnston LD, & O’malley PM (n.d.). Explaining Recent Increases in Students’ Marijuana Use: Impacts of Perceived Risks and Disapproval, 1976 through 1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barrocas AL, Giletta M, Hankin BL, Prinstein MJ, & Abela JRZ (2014). Nonsuicidal Self-Injury in Adolescence: Longitudinal Course, Trajectories, and Intrapersonal Predictors. Journal of Abnormal Child Psychology, 43(2), 369–380. 10.1007/s10802-014-9895-4 [DOI] [PubMed] [Google Scholar]
  6. Beauchaine TP (2015). Future Directions in Emotion Dysregulation and Youth Psychopathology. Journal of Clinical Child and Adolescent Psychology, 44(5), 875–896. 10.1080/15374416.2015.1038827 [DOI] [PubMed] [Google Scholar]
  7. Borges G, Walters EE, & Kessler RC (2000). Associations of Substance Use, Abuse, and Dependence with Subsequent Suicidal Behavior. American Journal of Epidemiology, 151(8), 781–789. 10.1093/oxfordjournals.aje.a010278 [DOI] [PubMed] [Google Scholar]
  8. Borges Guilherme, Bagge CL, & Orozco R (2016). A literature review and meta-analyses of cannabis use and suicidality. Journal of Affective Disorders, 195, 63–74. 10.1016/J.JAD.2016.02.007 [DOI] [PubMed] [Google Scholar]
  9. Brausch AM, & Boone SD (2015). Frequency of nonsuicidal self-injury in adolescents: Differences in suicide attempts, substance use, and disordered eating. Suicide and Life-Threatening Behavior. 10.1111/sltb.12155 [DOI] [PubMed] [Google Scholar]
  10. Brausch AM, & Muehlenkamp JJ (2018). Perceived effectiveness of NSSI in achieving functions on severity and suicide risk. Psychiatry Research, 265, 144–150. 10.1016/j.psychres.2018.04.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buckholdt KE, Parra GR, Anestis MD, Lavender JM, Jobe-Shields LE, Tull MT, & Gratz KL (2015). Emotion Regulation Difficulties and Maladaptive Behaviors: Examination of Deliberate Self-harm, Disordered Eating, and Substance Misuse in Two Samples. Cognitive Therapy and Research, 39(2), 140–152. 10.1007/s10608-014-9655-3 [DOI] [Google Scholar]
  12. Chadi N, Schroeder R, Jensen JW, & Levy S (2019). Association between Electronic Cigarette Use and Marijuana Use among Adolescents and Young Adults: A Systematic Review and Meta-analysis. JAMA Pediatrics, 173(10). 10.1001/jamapediatrics.2019.2574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen Y-C, & Klig JE (2019). Cannabis-related emergencies in children and teens. Current Opinion in Pediatrics, 31(3), 291–296. 10.1097/MOP.0000000000000752 [DOI] [PubMed] [Google Scholar]
  14. Conway KP, Vullo GC, Nichter B, Wang J, Compton WM, Iannotti RJ, & Simons-Morton B (2013). Prevalence and patterns of polysubstance use in a nationally representative sample of 10th graders in the United States. Journal of Adolescent Health, 52(6), 1–18. 10.1038/jid.2014.371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Coulter RWS, Ware D, Fish JN, & Plankey MW (2019). Latent Classes of Polysubstance Use Among Adolescents in the United States: Intersections of Sexual Identity with Sex, Age, and Race/Ethnicity. LGBT Health, 6(3), 116–125. 10.1089/lgbt.2018.0149 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cunningham RM, Goldstick J, & Carter PM (2019). The Major Causes of Death in Children. In The New England journal of medicine (Vol. 380). 10.1056/NEJMc1901264 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Felton JW, Kofler MJ, Lopez CM, Saunders BE, & Kilpatrick DG (2015). The emergence of co-occurring adolescent polysubstance use and depressive symptoms: A latent growth modeling approach. Development and Psychopathology, 27(4), 1367–1383. 10.1017/S0954579414001473 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Few LR, Grant JD, Nelson EC, Trull TJ, Grucza RA, Bucholz KK, … Agrawal A. (2016). Cannabis involvement and nonsuicidal self-injury: A discordant twin approach. Journal of Studies on Alcohol and Drugs, 77(6), 873–880. 10.15288/jsad.2016.77.873 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fox KR, Franklin JC, Ribeiro JD, Kleiman EM, Bentley KH, & Nock MK (2016). Meta-analysis of risk factors for nonsuicidal self-injury. Clinical Psychology Review, 42, 156–167. 10.1016/j.cpr.2015.09.002.Meta-analysis [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Giletta M, Scholte RHJ, Engels RCME, Ciairano S, & Prinstein MJ (2012). Adolescent non-suicidal self-injury: A cross-national study of community samples from Italy, the Netherlands and the United States. Psychiatry Research, 197(1-2), 66–72. 10.1016/j.psychres.2012.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gratz KL, & Tull MT (2010). The relationship between emotion dysregulation and deliberate self-harm among inpatients with substance use disorders. Cognitive Theory Research, 34(6), 1544–1553. 10.1007/s10608-009-9268-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Guvendeger Doksat N, Zahmacioglu O, Ciftci Demirci A, Kocaman GM, & Erdogan A (2017). Association of Suicide Attempts and Non-Suicidal Self-Injury Behaviors With Substance Use and Family Characteristics Among Children and Adolescents Seeking Treatment for Substance Use Disorder. Substance Use and Misuse, 52(5), 604–613. 10.1080/10826084.2016.1245745 [DOI] [PubMed] [Google Scholar]
  23. Hawton K, Hall S, Simkin S, Bale L, Bond A, Codd S, & Stewart A (2003). Deliberate self-harm in adolescents: A study of characteristics and trends in Oxford, 1990-2000. Journal of Child Psychology and Psychiatry and Allied Disciplines, 44(8), 1191–1198. 10.1111/1469-7610.00200 [DOI] [PubMed] [Google Scholar]
  24. Hawton K, Saunders KEA, & O’Connor RC (2012). Self-harm and suicide in adolescents. The Lancet, 379(9834), 2373–2382. 10.1016/S0140-6736(12)60322-5 [DOI] [PubMed] [Google Scholar]
  25. Heron M (2019). National Vital Statistics Reports Volume 68, Number 6, June 24, 2019, Deaths: Leading Causes for 2017. [PubMed] [Google Scholar]
  26. Hopfer S, Tan X, & Wylie JL (2014). A social network-informed latent class analysis of patterns of substance use, sexual behavior, and mental health: Social network study iii, winnipeg, Manitoba, Canada. American Journal of Public Health, 104(5), 834–839. 10.2105/AJPH.2013.301833 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Johnson JK (2018, August). Elucidating the Impact of Adolescent Marijuana Use. Journal of Adolescent Health, Vol. 63, pp. 129–130. 10.1016/j.jadohealth.2018.06.002 [DOI] [PubMed] [Google Scholar]
  28. Johnston LD, Miech RA, O’Malley PM, Bachman JG, Schulenberg JE, & Patrick ME (2019). Monitoring the Future: National survey results on drug use 1975-2018. 2018 Overview: Key findings on adolescent drug use. The National Institute on Drug Abuse at The National Institutes of Health, 1–104. [Google Scholar]
  29. Kassahun W, Neyens T, Molenberghs G, Faes C, & Verbeke G (2012). Modeling overdispersed longitudinal binary data using a combined beta and normal random-effects model. Archives of Public Health, 70(1). 10.1186/0778-7367-70-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kim Y, Choi YK, & Emery S (2013). Logistic regression with multiple random effects: A simulation study of estimation methods and statistical packages. American Statistician, 67(3), 171–182. 10.1080/00031305.2013.817357 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Klonsky ED (2011). Non-suicidal self-injury in United States adults: Prevalence, sociodemographics, topography and functions. Psychological Medicine, 41 (9), 1981–1986. 10.1017/s0033291710002497 [DOI] [PubMed] [Google Scholar]
  32. Klonsky E. David, Glenn CR, Styer DM, Olino TM, & Washburn JJ (2015). The functions of nonsuicidal self-injury: converging evidence for a two-factor structure. Child and Adolescent Psychiatry and Mental Health, 9(1), 44 10.1186/s13034-015-0073-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Levy S, Campbell MD, Shea CL, & Du Pont R (2018, August). Trends in abstaining from substance use in adolescents: 1975–2014. Pediatrics, Vol. 142 10.1542/peds.2017-3498 [DOI] [PubMed] [Google Scholar]
  34. Levy S, Sherritt L, Harris K, Gates EC, Holder DW, Kulig JW, & Knight JR (2004). Test-Retest Reliability of Adolescents’ Self-Report of Substance Use. 10.1097/01.ALC.0000134216.22162.A5 [DOI] [PubMed] [Google Scholar]
  35. Li M-C, Brady JE, Dimaggio CJ, Lusardi AR, Tzong KY, & Li G (2012). Epidemiologic Reviews Marijuana Use and Motor Vehicle Crashes. 34 10.1093/epirev/mxr017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Linehan MM (1993). Cognitive-Behavioral treatment of borderline personality disorder. New York, NY: Guilford Press. [Google Scholar]
  37. Lloyd-Richardson EE, Perrine N, Dierker L, & Kelley ML (2007). Characteristics and functions of non-suicidal self-injury in a community sample of adolescents. Psychological Medicine, 37(8), 1183–1192. 10.1017/S003329170700027X [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lo CB, Bridge JA, Shi J, Ludwig L, & Stanley RM (2020). Children’s Mental Health Emergency Department Visits: 2007–2016. Pediatrics, e20191536 10.1542/peds.2019-1536 [DOI] [PubMed] [Google Scholar]
  39. Metrik J, Caswell AJ, Magill M, Monti PM, & Kahler CW (2016). Sexual Risk Behavior and Heavy Drinking Among Weekly Marijuana Users. In JOURNAL OF STUDIES ON ALCOHOL AND DRUGS (Vol. 77). [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Morin JFG, Afzali MH, Bourque J, Stewart SH, Séguin JR, O’Leary-Barrett M, & Conrod PJ (2019). A population-based analysis of the relationship between substance use and adolescent cognitive development. American Journal of Psychiatry, 176(2), 98–106. 10.1176/appi.ajp.2018.18020202 [DOI] [PubMed] [Google Scholar]
  41. Moss HB, Chen CM, & Yi H. ye. (2014). Early adolescent patterns of alcohol, cigarettes, and marijuana polysubstance use and young adult substance use outcomes in a nationally representative sample. Drug and Alcohol Dependence, 136(1), 51–62. 10.1016/j.drugalcdep.2013.12.011 [DOI] [PubMed] [Google Scholar]
  42. Moss SL, Santaella-Tenorio J, Mauro PM, Keyes KM, & Martins SS (2019). Changes over time in marijuana use, deviant behavior and preference for risky behavior among US adolescents from 2002 to 2014: testing the moderating effect of gender and age. Addiction, 114(4), 674–686. 10.1111/add.14506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Nock MK (2010). Self-Injury. Annual Review of Clinical Psychology, 6(1), 339–363. 10.1146/annurev.clinpsy.121208.131258 [DOI] [PubMed] [Google Scholar]
  44. Nock MK, Joiner TE, Gordon KH, Lloyd-Richardson E, & Prinstein MJ (2006). Non-suicidal self-injury among adolescents: Diagnostic correlates and relation to suicide attempts. Psychiatry Research. 10.1016/j.psychres.2006.05.010 [DOI] [PubMed] [Google Scholar]
  45. Nock MK, & Prinstein MJ (2004). A Functional Approach to the Assessment of Self-Mutilative Behavior. Journal of Consulting and Clinical Psychology, 72(5), 885–890. 10.1037/0022-006X.72.5.885 [DOI] [PubMed] [Google Scholar]
  46. Patrick ME, Schulenberg JE, O’Malley PM, Johnston LD, & Bachman JG (2011). Adolescents’ reported reasons for alcohol and marijuana use as predictors of substance use and problems in adulthood. Journal of Studies on Alcohol and Drugs, 72(1), 106–116. 10.15288/jsad.2011.72.106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Plener PL, Kaess M, Schmahl C, Pollak S, Fegert JM, & Brown RC (2018). Non-suicidal self-injury in adolescents. Deutsches Arzteblatt International, 115(3), 23–30. 10.3238/arztebl.2018.0023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Rabe-Hesketh S, & Skrondal A (2012). Multilevel and Longitudinal Modeling Using Stata, Volumes I and II, Third Edition (Third) StataCorp LP. [Google Scholar]
  49. Rossow I, Ystgaard M, Hawton K, Madge N, Van Heeringen K, Jan De Wilde E, … Morey C (2007). Cross-National Comparisons of the Association between Alcohol Consumption and Deliberate Self-Harm in Adolescents. 10.1521/suli.2007.37.6.605 [DOI] [PubMed] [Google Scholar]
  50. Schneider KE, Brighthaupt S-C, Winiker AK, Johnson RM, Musci RJ, & Linton, Sabriya L (2020). Characterizing profiles of polysubstance use among high school students in Baltimore, Maryland: A latent class analysis. Drug and Alcohol Dependence, 211(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Sobell LC, & Sobell MB (1992). Timeline Follow-Back In Measuring Alcohol Consumption (pp. 41–72). 10.1007/978-1-4612-0357-5_3 [DOI] [Google Scholar]
  52. Swannell SV, Martin GE, Page A, Hasking P, & St John NJ (2014). Prevalence of nonsuicidal self-injury in nonclinical samples: Systematic review, meta-analysis and meta-regression. Suicide and Life-Threatening Behavior, 44(3), 273–303. 10.1111/sltb.12070 [DOI] [PubMed] [Google Scholar]
  53. Taylor PJ, Jomar K, Dhingra K, Forrester R, Shahmalak U, & Dickson JM (2018). A meta-analysis of the prevalence of different functions of non-suicidal self-injury. Journal of Affective Disorders, 227, 759–769. 10.1016/J.JAD.2017.11.073 [DOI] [PubMed] [Google Scholar]
  54. Tuisku V, Kiviruusu O, Pelkonen M, Karlsson L, Strandholm T, & Marttunen M (2013). Depressed adolescents as young adults – Predictors of suicide attempt and nonsuicidal self-injury during an 8-year follow-up. Journal of Affective Disorders, 152–154, 313–319. 10.1016/j.jad.2013.09.031 [DOI] [PubMed] [Google Scholar]
  55. Volkow ND, Baler RD, Compton WM, & Weiss SRB (2014). Adverse Health Effects of Marijuana Use. New England Journal of Medicine, 370(23), 2219–2227. 10.1056/NEJMra1402309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Waldron HB, Slesnick N, Brody JL, Turner CW, & Peterson TR (2001). Treatment outcomes for adolescent substance abuse at 4- and 7-month assessments. Journal of Consulting and Clinical Psychology, 69(5), 802–813. 10.1037/0022-006X.69.5.802 [DOI] [PubMed] [Google Scholar]
  57. Wang TW, Gentzke AS, Creamer MR, Cullen KA, Holder-Hayes E, Sawdey MD, … Neff LJ (2019). Tobacco Product Use and Associated Factors Among Middle and High School Students — United States, 2019. MMWR. Surveillance Summaries, 68(12), 1–22. 10.15585/mmwr.ss6812a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Whitlock J, Muehlenkamp J, Eckenrode J, Purington A, Baral Abrams G, Barreira P, & Kress V (2013). Nonsuicidal self-injury as a gateway to suicide in young adults. Journal of Adolescent Health, 52(4), 486–492. 10.1016/j.jadohealth.2012.09.010 [DOI] [PubMed] [Google Scholar]
  59. Willoughby T, Heffer T, & Hamza CA (2015). The link between nonsuicidal self-injury and acquired capability for suicide: A longitudinal study. Journal of Abnormal Psychology, 124(4), 1110–1115. 10.1037/abn0000104 [DOI] [PubMed] [Google Scholar]
  60. Wolff J (2020). Emotion dysregulation an non-suicidal self-injury: A systematic review and meta-analysis. 25–36. 10.1016/j.eurpsy.2019.03.004.Emotion [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zimmermann K, Walz C, Derckx RT, Kendrick KM, Weber B, Dore B, … Becker B (2017). Emotion regulation deficits in regular marijuana users. Human Brain Mapping, 38(8), 4270–4279. 10.1002/hbm.23671 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES