Accepted Manuscript Sexual orientation disparities in prescription drug misuse among a nationally representative sample of adolescents: Prevalence and correlates
Dennis H. Li, Blair C. Turner, Brian Mustanski, Gregory L. Phillips PII: DOI: Reference:
S0306-4603(17)30366-0 doi:10.1016/j.addbeh.2017.09.021 AB 5308
To appear in:
Addictive Behaviors
Received date: Revised date: Accepted date:
4 May 2017 14 September 2017 26 September 2017
Please cite this article as: Dennis H. Li, Blair C. Turner, Brian Mustanski, Gregory L. Phillips , Sexual orientation disparities in prescription drug misuse among a nationally representative sample of adolescents: Prevalence and correlates. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Ab(2017), doi:10.1016/j.addbeh.2017.09.021
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.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
1
Sexual Orientation Disparities in Prescription Drug Misuse Among a Nationally Representative Sample of Adolescents: Prevalence and Correlates
1
IP
CR
Phillips, II, PhD, MS1
T
Dennis H. Li, PhD, MPH,1 Blair C. Turner, MPH,1 Brian Mustanski, PhD1 and Gregory L.
Institute for Sexual and Gender Minority Health and Wellbeing, Northwestern University, 625
AN
US
N. Michigan Avenue, Suite 1400, Chicago, IL 60611
AC
CE
PT
ED
M
Please address all correspondence to Dennis H. Li at [email protected].
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
2
Abstract Objective: Sexual minority adolescents (SMA) may be at disproportionate risk for misusing prescription psychotropic medications compared to their heterosexual peers. However, generalizable studies specific to this age group are lacking. The current study aimed to describe
T
the prevalence of sexual orientation disparities in prescription drug misuse among a nationally
IP
representative sample of adolescents as well as to examine key correlates of misuse.
CR
Method: Using data from the National Youth Risk Behavior Survey, we conducted stepwise multivariable weighted logistic regressions, sequentially controlling for demographics,
US
experiences of victimization, mental health, and other illicit substance use. Results: Adjusting
AN
for grade and race/ethnicity, female SMA and gay and unsure males had significantly elevated odds of ever misusing a prescription drug compared to heterosexual adolescents (ORs from 1.7–
M
2.5). Most sexual orientation disparities among females remained significant with the addition of
ED
victimization and mental health covariates but attenuated completely after controlling for other illicit drug use. The effect for unsure males attenuated when victimization variables were
PT
included, but the effect for gay males remained significant through the final model. Controlling
CE
for other illicit drug use, mental health variables remained significant correlates for females whereas only forced sex was significant for males. Conclusion: These results suggest
AC
experiences of victimization and mental health partially account for the disparities in prescription drug misuse between SMA and heterosexual adolescents, and their effects may differ by sex. A combination of structural, individual coping, and universal drug prevention approaches should be used to make the largest impact on reducing these disparities.
Keywords: sexual minority; prescription drugs; gay; lesbian; bisexual; adolescent; victimization
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
3
Introduction Prescription psychotropic medications are the second most prevalent type of drug used illicitly among adolescents in the United States (Johnston, O’Malley, Miech, Bachman, & Schulenberg, 2016; Martins et al., 2017). The majority of high school youth perceive using
IP
T
prescription stimulants, pain relievers, tranquilizers, and sedatives without medical supervision
CR
as low risk (Miech, Johnston, O’Malley, Bachman, & Schulenberg, 2016). As such, over 18% of twelfth- grade students report ever having misused a prescription drug, and 13% report having
US
misused in the past year (Miech et al., 2016). Like other substances, prescription drug misuse can have short- and long-term negative social and health effects, including family and peer
AN
relationship problems, poor school performance, impaired brain functioning, increased risk for
M
mental health and substance use disorders, and overdose death (Ali et al., 2015; McCabe, Veliz, Boyd, & Schulenberg, 2017; National Institute on Drug Abuse, 2016). Thus, early prevention,
ED
identification, and treatment of prescription drug misuse is of critical public health concern.
PT
Limited evidence suggests emerging and young adults who identify as sexual minorities
CE
may be at increased risk of prescription drug misuse compared to those who identify as heterosexual (Corliss et al., 2010; McCabe, 2005; Rosario et al., 2014; Shadick, Dagirmanjian,
AC
Trub, & Dawson, 2016). Even less research has examined this relation among sexual minority adolescents (SMA, loosely defined here as those under 18 years), who may be less out and have less support for their sexual identity (Savin-Williams & Cohen, 2015) as well as different access to substances than older individuals (Johnston et al., 2016). In the only study of sexual orientation differences in prescription drug misuse that examined adolescents separately from emerging adults, Corliss et al. (2010) found that the elevated risk of misuse among sexual minorities was larger for gay, lesbian, and bisexual youth aged 12–17 than for those aged 18–23.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
4
In addition, whereas the estimates for emerging adults reflected other studies of sexual minority young adults that suggest the disparity is largely driven by bisexuals, questioning individuals, and females (McCabe, 2005; Rosario et al., 2014; Shadick et al., 2016), among the younger group, the risks for mostly heterosexual males and females were substantially attenuated relative
T
to other SMA subgroups, and the adjusted effect for gay males remained significantly elevated.
IP
These age group discrepancies as well as the general lack of studies among younger individuals
CR
highlight a need for more targeted research examining prescription drug misuse specifically
US
among SMA.
In addition to establishing the prevalence of prescription drug misuse among SMA, it is
AN
important to identify factors that contribute to heightened risk. School-based experiences of
M
victimization are particularly salient to this population and may be one such factor (D’Augelli, 1998). SMA experience higher levels of both general peer victimization and homophobic teasing
ED
than heterosexual youth (Birkett, Espelage, & Koenig, 2009). In a national survey of 10,528
PT
sexual and gender minority students, over 85% reported experiencing verbal harassment, largely based on sexual orientation, and 58% and 27% respectively reported feeling unsafe and being
CE
physically harassed in the past year because of their sexual orientation (Kosciw, Greytak, Giga,
AC
Villenas, & Danischewski, 2016). Per the minority stress model (Meyer, 2003), such stressors not only impact individuals acutely but can lead to and interact with internalizations of stigma and negative self-worth, which may subsequently foster risk behaviors and negative physical and mental health outcomes (Frost, Lehavot, & Meyer, 2015; Newcomb & Mustanski, 2010). SMA who experience victimization may turn to prescription drugs to cope, which has been suggested by some studies of emerging and young adults (Kecojevic, Corliss, & Lankenau, 2015; Rosario, Hunter, & Gwadz, 1997). Relatedly, having a positive school environment has been shown to be
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
5
protective against alcohol and marijuana use, closing the disparity between gay and lesbian students and their heterosexual peers (Birkett et al., 2009). Heck et al. (2014) found that sexual and gender minority students attending high schools without a gay–straight alliance had greater odds of using several licit and illicit substances, including twice the odds of misusing attention-
T
deficit/hyperactivity disorder and pain medications, than students attending schools with such an
IP
organization, but no other studies of peer victimization and prescription drug misuse in this age
CR
group have been identified.
US
SMA also experience higher levels of other types of victimization, namely forced sexual intercourse (McLaughlin, Hatzenbuehler, Xuan, & Conron, 2012). In a meta-analysis (2011),
AN
SMA were found to be 3.8 times more likely to have experienced sexual abuse than heterosexual
M
youth. Early-life adverse events, including sexual abuse and sexual violence, have been linked to more tobacco, marijuana, and illicit drug use; greater odds of alcohol use; and higher levels of
ED
prescription opioid misuse among sexual minority emerging and young adults (Kecojevic, Wong,
PT
Corliss, & Lankenau, 2015; McLaughlin et al., 2012) as well as to earlier initiation and greater odds
of prescription drug misuse among similarly aged youth in the general population (Kecojevic et
CE
al., 2012; A. Young, Grey, Boyd, & McCabe, 2011) . Sexual abuse and sexual violence have also
AC
been linked to greater odds of depression and suicidality among sexual minority youth (McLaughlin et al., 2012). These findings in related populations suggest early-life sexual victimization may serve as another driver of prescription drug misuse disparities. However, no direct evidence has yet been reported for adolescents, and there has been limited exploration into differences by sexual orientation. The current study aimed to estimate the prevalence of prescription drug misuse among a nationally representative sample of adolescents and to identify disparities in misuse between
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
6
male and female SMA and heterosexual adolescents. We also sought to determine whether victimization experiences contribute to these disparities and to explore how they might do so through impacting mental health outcomes and other illicit substance use. We hypothesized that SMA would be at greater risk for misusing prescription drugs than heterosexual students and that
IP
T
victimization would account for much of the disparity between the groups.
CR
Methods Design and Sample
US
We analyzed data from the 2015 National Youth Risk Behavior Survey (YRBS), a crosssectional survey of priority health risk behaviors among public and private high school students
AN
in grades 9 through 12 conducted biennially by the Centers for Disease Control and Prevention
M
(Centers for Disease Control and Prevention, 2016). National representativeness was achieved
ED
using a three-stage cluster sample design, with random selection of schools and classes probabilistically weighted by student enrollment in the target grades. All students in sampled
PT
classes were eligible to participate. School and student response rates were 69% and 86%, respectively; 15,624 usable questionnaires were obtained from 125 schools (Brener et al., 2013;
CE
Kann et al., 2016). Records were weighted based on student sex, grade, and race/ethnicity;
AC
school and student nonresponse; and oversampling of Hispanic and Black students. All current study activities were classified as non-human subjects research by the Institutional Review Board at [institution blinded for review]. Measures All variables in the YRBS are assessed via self-report. The YRBS codebook as well as a discussion of YRBS reliability and validity can be found elsewhere (Brener et al., 2013; 2016).
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
7
Sexual orientation. Self-identified sexual orientation was assessed with the question, “Which of the following best describes you?” Respondents could endorse identifying as heterosexual (straight), gay or lesbian, bisexual, or not sure. Demographics. Sex was coded as either female or male. Grade in school ranged between
IP
T
9th and 12th grades. Two items for ethnicity and race were collapsed into a single variable with five categories: non-Hispanic White, non-Hispanic Black or African American, Hispanic/Latino,
CR
non-Hispanic multiracial, and non-Hispanic other.
US
Prescription drug misuse. Respondents were asked how many times in their life they had taken a prescription drug without a doctor’s orders and were provided examples of
AN
prescription drugs (i.e., OxyContin, Percocet, Vicodin, codeine, Adderall, Ritalin, or Xanax).
M
Any endorsement greater than zero was coded as ever having misused.
ED
Victimization. Two items asking respondents whether they had experienced bullying on school property and electronically (through, e.g., websites, texting) in the past 12 months were
PT
collapsed into one dichotomous bullying variable. One item asked how many times in the past 12
CE
months respondents were threatened or injured by someone with a weapon and was dichotomized between ever and never threatened. Forced sexual intercourse was assessed with
AC
one dichotomous item asking respondents if they had ever been physically forced to have sexual intercourse when they did not want to. Mental health. One dichotomous item asked respondents if, in the past 12 months, they had ever felt so sad or hopeless almost every day for two weeks or more in a row that they stopped doing some usual activities. Another dichotomous item asked if they had ever seriously considered attempting suicide in the past 12 months.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
8
Marijuana and other illicit substance use. Respondents were asked a series of items about their use of marijuana, cocaine, inhalants, heroin, methamphetamines, ecstasy, injection drugs, and hallucinogenic drugs. Endorsement of any marijuana use was coded as lifetime use. Endorsement of any other drug was coded as lifetime use of a non-marijuana illicit drug.
IP
T
Analysis
We excluded students from analysis if they were missing any primary demographic
CR
variables (2.3% were missing race/ethnicity; 0.8%, sex; and 0.8%, grade, not mutually
US
exclusive), sexual orientation (5.9%), or prescription drug misuse (1.7%). The final analytic sample contained 14,477 individuals.
AN
Descriptive statistics were calculated for prescription drug misuse and other variables,
M
stratified by sex. Bivariate associations between prescription drug misuse and other variables
ED
were examined using sample-weighted Pearson’s χ2 test of independence. We then conducted stepwise multivariable weighted logistic regressions, starting with sexual orientation and
PT
sequentially adding blocks of covariates: (1) demographics, (2) victimization, (3) mental health,
CE
and (4) other illicit substance use. Data cleaning was done in SAS, Version 9.4 (Cary, NC), and analyses were conducted in SAS-callable SUDAAN, Version 11, to adjust for survey weighting
AC
(2016). Akaike information and Schwartz criteria were inspected at each step to ensure appropriate model selection. Results Univariate analyses Table 1 contains descriptive statistics of the analytic sample. Overall, 16.7% of students reported ever misusing prescription drugs, with a higher prevalence among males (17.8%) than
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
9
females (15.5%). The largest proportion of those who misused prescription drugs reported doing so only once or twice (45.6% of females, 33.1% of males), though 29.2% of females and 40.7% of males reported misusing at least 10 times. About 15.5% of females (n = 1184) and 6.8% of males (n = 522) identified as sexual minorities. Covariate estimates can be found in the table.
AC
CE
PT
ED
M
AN
US
CR
IP
T
Table 1. Descriptive statistics for students in the 2015 National YRBS ( N=14,477) by sex a Total Female (n=7248) Male (n=7229) Variable n % n % n % Demographics: Sexual orientation Heterosexual 12771 88.96 6064 84.52 6707 93.17 Gay/lesbian 316 1.98 165 1.99 151 1.97 Bisexual 900 5.93 725 9.76 175 2.30 Not sure 490 3.13 294 3.73 196 2.57 Grade 9th grade 3687 26.89 1876 26.10 1811 27.64 10th grade 3642 25.84 1822 26.88 1820 24.86 11th grade 3707 24.01 1834 23.65 1873 24.35 12th grade 3364 23.26 1684 23.38 1680 23.15 Race/ethnicity White 6386 54.46 3222 55.03 3164 53.92 Black/African American 1525 13.52 768 13.37 757 13.67 Hispanic/Latino 4822 22.68 2424 22.82 2398 22.54 Multiple 684 4.50 363 4.61 321 4.39 Other 802 4.84 365 4.16 437 5.48 Prescription drug misuse: Ever used prescription drugs Yes 2501 16.70 1184 15.51 1317 17.82 No 11976 83.30 6064 84.49 5912 82.18 Number of times misused b 1 or 2 times 3 to 9 times 10 to 19 times 20 to 39 times 40 or more times Victimization: Bullied or cyber bullied Yes No Threatened at school
934 680 360 187 340
38.77 25.76 14.09 7.80 13.58
514 324 149 91 106
45.61 25.21 12.86 7.95 8.37
420 356 211 96 234
33.12 26.22 15.10 7.68 17.88
3509 10968
25.70 74.30
2129 5119
32.10 67.90
1380 5849
19.62 80.38
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
10
M
Percentages are out of those who indicated misusing prescription drugs.
ED
b
AN
US
CR
IP
T
Yes 847 5.62 330 4.54 517 6.65 No 13043 94.38 6616 95.46 6427 93.35 Forced to have sex Yes 1056 6.53 791 10.29 265 2.96 No 13256 93.47 6377 89.71 6879 97.04 Mental health: Felt sad or hopeless Yes 4456 29.81 2959 39.81 1497 20.31 No 9902 70.19 4236 60.19 5666 79.69 Considered suicide Yes 2604 17.58 1737.0 23.38 867 12.09 No 11730 82.42 5442.0 76.62 6288 87.91 Other illicit substance use: Smoked marijuana Yes 5675 38.82 2781 37.42 2894 40.16 No 8461 61.18 4350 62.58 4111 59.84 Used another illicit drug Yes 2065 13.21 945 11.96 1120 14.40 No 12412 86.79 6303 88.04 6109 85.60 a Unweighted frequencies (n) are provided; percentages (%) reflect adjusted sampling weights
Bivariate analyses
PT
Table 2 presents bivariate analyses between prescription drug misuse and other variables.
CE
Female and male students who reported misusing prescription drugs comprised significantly greater proportions of SMA than those who reported no misuse. Students who misused also
AC
differed significantly from those who did not on demographic and other covariates save race/ethnicity among males. Relative to heterosexuals, SMA groups reported higher prevalence of prescription drug misuse; all pairwise comparisons were statistically significant except for bisexual males, which approached significance (Supplemental Table A).
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
11
AC
CE
PT
ED
M
AN
US
CR
IP
T
Table 2. Bivariate differences between students who did and did not misuse prescription drugs in the 2015 National YRBS (N=14,477) by sex a Female (n=7248) Male (n=7229) Misused Did not misuse Misused Did not misuse n=1184 n=6064 n=1317 n=5912 Variable n % n % pb n % n % pb Demographics: Sexual orientation
Dennis H. Li, Blair C. Turner, Brian Mustanski, Gregory L. Phillips PII: DOI: Reference:
S0306-4603(17)30366-0 doi:10.1016/j.addbeh.2017.09.021 AB 5308
To appear in:
Addictive Behaviors
Received date: Revised date: Accepted date:
4 May 2017 14 September 2017 26 September 2017
Please cite this article as: Dennis H. Li, Blair C. Turner, Brian Mustanski, Gregory L. Phillips , Sexual orientation disparities in prescription drug misuse among a nationally representative sample of adolescents: Prevalence and correlates. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Ab(2017), doi:10.1016/j.addbeh.2017.09.021
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.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
1
Sexual Orientation Disparities in Prescription Drug Misuse Among a Nationally Representative Sample of Adolescents: Prevalence and Correlates
1
IP
CR
Phillips, II, PhD, MS1
T
Dennis H. Li, PhD, MPH,1 Blair C. Turner, MPH,1 Brian Mustanski, PhD1 and Gregory L.
Institute for Sexual and Gender Minority Health and Wellbeing, Northwestern University, 625
AN
US
N. Michigan Avenue, Suite 1400, Chicago, IL 60611
AC
CE
PT
ED
M
Please address all correspondence to Dennis H. Li at [email protected].
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
2
Abstract Objective: Sexual minority adolescents (SMA) may be at disproportionate risk for misusing prescription psychotropic medications compared to their heterosexual peers. However, generalizable studies specific to this age group are lacking. The current study aimed to describe
T
the prevalence of sexual orientation disparities in prescription drug misuse among a nationally
IP
representative sample of adolescents as well as to examine key correlates of misuse.
CR
Method: Using data from the National Youth Risk Behavior Survey, we conducted stepwise multivariable weighted logistic regressions, sequentially controlling for demographics,
US
experiences of victimization, mental health, and other illicit substance use. Results: Adjusting
AN
for grade and race/ethnicity, female SMA and gay and unsure males had significantly elevated odds of ever misusing a prescription drug compared to heterosexual adolescents (ORs from 1.7–
M
2.5). Most sexual orientation disparities among females remained significant with the addition of
ED
victimization and mental health covariates but attenuated completely after controlling for other illicit drug use. The effect for unsure males attenuated when victimization variables were
PT
included, but the effect for gay males remained significant through the final model. Controlling
CE
for other illicit drug use, mental health variables remained significant correlates for females whereas only forced sex was significant for males. Conclusion: These results suggest
AC
experiences of victimization and mental health partially account for the disparities in prescription drug misuse between SMA and heterosexual adolescents, and their effects may differ by sex. A combination of structural, individual coping, and universal drug prevention approaches should be used to make the largest impact on reducing these disparities.
Keywords: sexual minority; prescription drugs; gay; lesbian; bisexual; adolescent; victimization
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
3
Introduction Prescription psychotropic medications are the second most prevalent type of drug used illicitly among adolescents in the United States (Johnston, O’Malley, Miech, Bachman, & Schulenberg, 2016; Martins et al., 2017). The majority of high school youth perceive using
IP
T
prescription stimulants, pain relievers, tranquilizers, and sedatives without medical supervision
CR
as low risk (Miech, Johnston, O’Malley, Bachman, & Schulenberg, 2016). As such, over 18% of twelfth- grade students report ever having misused a prescription drug, and 13% report having
US
misused in the past year (Miech et al., 2016). Like other substances, prescription drug misuse can have short- and long-term negative social and health effects, including family and peer
AN
relationship problems, poor school performance, impaired brain functioning, increased risk for
M
mental health and substance use disorders, and overdose death (Ali et al., 2015; McCabe, Veliz, Boyd, & Schulenberg, 2017; National Institute on Drug Abuse, 2016). Thus, early prevention,
ED
identification, and treatment of prescription drug misuse is of critical public health concern.
PT
Limited evidence suggests emerging and young adults who identify as sexual minorities
CE
may be at increased risk of prescription drug misuse compared to those who identify as heterosexual (Corliss et al., 2010; McCabe, 2005; Rosario et al., 2014; Shadick, Dagirmanjian,
AC
Trub, & Dawson, 2016). Even less research has examined this relation among sexual minority adolescents (SMA, loosely defined here as those under 18 years), who may be less out and have less support for their sexual identity (Savin-Williams & Cohen, 2015) as well as different access to substances than older individuals (Johnston et al., 2016). In the only study of sexual orientation differences in prescription drug misuse that examined adolescents separately from emerging adults, Corliss et al. (2010) found that the elevated risk of misuse among sexual minorities was larger for gay, lesbian, and bisexual youth aged 12–17 than for those aged 18–23.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
4
In addition, whereas the estimates for emerging adults reflected other studies of sexual minority young adults that suggest the disparity is largely driven by bisexuals, questioning individuals, and females (McCabe, 2005; Rosario et al., 2014; Shadick et al., 2016), among the younger group, the risks for mostly heterosexual males and females were substantially attenuated relative
T
to other SMA subgroups, and the adjusted effect for gay males remained significantly elevated.
IP
These age group discrepancies as well as the general lack of studies among younger individuals
CR
highlight a need for more targeted research examining prescription drug misuse specifically
US
among SMA.
In addition to establishing the prevalence of prescription drug misuse among SMA, it is
AN
important to identify factors that contribute to heightened risk. School-based experiences of
M
victimization are particularly salient to this population and may be one such factor (D’Augelli, 1998). SMA experience higher levels of both general peer victimization and homophobic teasing
ED
than heterosexual youth (Birkett, Espelage, & Koenig, 2009). In a national survey of 10,528
PT
sexual and gender minority students, over 85% reported experiencing verbal harassment, largely based on sexual orientation, and 58% and 27% respectively reported feeling unsafe and being
CE
physically harassed in the past year because of their sexual orientation (Kosciw, Greytak, Giga,
AC
Villenas, & Danischewski, 2016). Per the minority stress model (Meyer, 2003), such stressors not only impact individuals acutely but can lead to and interact with internalizations of stigma and negative self-worth, which may subsequently foster risk behaviors and negative physical and mental health outcomes (Frost, Lehavot, & Meyer, 2015; Newcomb & Mustanski, 2010). SMA who experience victimization may turn to prescription drugs to cope, which has been suggested by some studies of emerging and young adults (Kecojevic, Corliss, & Lankenau, 2015; Rosario, Hunter, & Gwadz, 1997). Relatedly, having a positive school environment has been shown to be
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
5
protective against alcohol and marijuana use, closing the disparity between gay and lesbian students and their heterosexual peers (Birkett et al., 2009). Heck et al. (2014) found that sexual and gender minority students attending high schools without a gay–straight alliance had greater odds of using several licit and illicit substances, including twice the odds of misusing attention-
T
deficit/hyperactivity disorder and pain medications, than students attending schools with such an
IP
organization, but no other studies of peer victimization and prescription drug misuse in this age
CR
group have been identified.
US
SMA also experience higher levels of other types of victimization, namely forced sexual intercourse (McLaughlin, Hatzenbuehler, Xuan, & Conron, 2012). In a meta-analysis (2011),
AN
SMA were found to be 3.8 times more likely to have experienced sexual abuse than heterosexual
M
youth. Early-life adverse events, including sexual abuse and sexual violence, have been linked to more tobacco, marijuana, and illicit drug use; greater odds of alcohol use; and higher levels of
ED
prescription opioid misuse among sexual minority emerging and young adults (Kecojevic, Wong,
PT
Corliss, & Lankenau, 2015; McLaughlin et al., 2012) as well as to earlier initiation and greater odds
of prescription drug misuse among similarly aged youth in the general population (Kecojevic et
CE
al., 2012; A. Young, Grey, Boyd, & McCabe, 2011) . Sexual abuse and sexual violence have also
AC
been linked to greater odds of depression and suicidality among sexual minority youth (McLaughlin et al., 2012). These findings in related populations suggest early-life sexual victimization may serve as another driver of prescription drug misuse disparities. However, no direct evidence has yet been reported for adolescents, and there has been limited exploration into differences by sexual orientation. The current study aimed to estimate the prevalence of prescription drug misuse among a nationally representative sample of adolescents and to identify disparities in misuse between
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
6
male and female SMA and heterosexual adolescents. We also sought to determine whether victimization experiences contribute to these disparities and to explore how they might do so through impacting mental health outcomes and other illicit substance use. We hypothesized that SMA would be at greater risk for misusing prescription drugs than heterosexual students and that
IP
T
victimization would account for much of the disparity between the groups.
CR
Methods Design and Sample
US
We analyzed data from the 2015 National Youth Risk Behavior Survey (YRBS), a crosssectional survey of priority health risk behaviors among public and private high school students
AN
in grades 9 through 12 conducted biennially by the Centers for Disease Control and Prevention
M
(Centers for Disease Control and Prevention, 2016). National representativeness was achieved
ED
using a three-stage cluster sample design, with random selection of schools and classes probabilistically weighted by student enrollment in the target grades. All students in sampled
PT
classes were eligible to participate. School and student response rates were 69% and 86%, respectively; 15,624 usable questionnaires were obtained from 125 schools (Brener et al., 2013;
CE
Kann et al., 2016). Records were weighted based on student sex, grade, and race/ethnicity;
AC
school and student nonresponse; and oversampling of Hispanic and Black students. All current study activities were classified as non-human subjects research by the Institutional Review Board at [institution blinded for review]. Measures All variables in the YRBS are assessed via self-report. The YRBS codebook as well as a discussion of YRBS reliability and validity can be found elsewhere (Brener et al., 2013; 2016).
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
7
Sexual orientation. Self-identified sexual orientation was assessed with the question, “Which of the following best describes you?” Respondents could endorse identifying as heterosexual (straight), gay or lesbian, bisexual, or not sure. Demographics. Sex was coded as either female or male. Grade in school ranged between
IP
T
9th and 12th grades. Two items for ethnicity and race were collapsed into a single variable with five categories: non-Hispanic White, non-Hispanic Black or African American, Hispanic/Latino,
CR
non-Hispanic multiracial, and non-Hispanic other.
US
Prescription drug misuse. Respondents were asked how many times in their life they had taken a prescription drug without a doctor’s orders and were provided examples of
AN
prescription drugs (i.e., OxyContin, Percocet, Vicodin, codeine, Adderall, Ritalin, or Xanax).
M
Any endorsement greater than zero was coded as ever having misused.
ED
Victimization. Two items asking respondents whether they had experienced bullying on school property and electronically (through, e.g., websites, texting) in the past 12 months were
PT
collapsed into one dichotomous bullying variable. One item asked how many times in the past 12
CE
months respondents were threatened or injured by someone with a weapon and was dichotomized between ever and never threatened. Forced sexual intercourse was assessed with
AC
one dichotomous item asking respondents if they had ever been physically forced to have sexual intercourse when they did not want to. Mental health. One dichotomous item asked respondents if, in the past 12 months, they had ever felt so sad or hopeless almost every day for two weeks or more in a row that they stopped doing some usual activities. Another dichotomous item asked if they had ever seriously considered attempting suicide in the past 12 months.
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
8
Marijuana and other illicit substance use. Respondents were asked a series of items about their use of marijuana, cocaine, inhalants, heroin, methamphetamines, ecstasy, injection drugs, and hallucinogenic drugs. Endorsement of any marijuana use was coded as lifetime use. Endorsement of any other drug was coded as lifetime use of a non-marijuana illicit drug.
IP
T
Analysis
We excluded students from analysis if they were missing any primary demographic
CR
variables (2.3% were missing race/ethnicity; 0.8%, sex; and 0.8%, grade, not mutually
US
exclusive), sexual orientation (5.9%), or prescription drug misuse (1.7%). The final analytic sample contained 14,477 individuals.
AN
Descriptive statistics were calculated for prescription drug misuse and other variables,
M
stratified by sex. Bivariate associations between prescription drug misuse and other variables
ED
were examined using sample-weighted Pearson’s χ2 test of independence. We then conducted stepwise multivariable weighted logistic regressions, starting with sexual orientation and
PT
sequentially adding blocks of covariates: (1) demographics, (2) victimization, (3) mental health,
CE
and (4) other illicit substance use. Data cleaning was done in SAS, Version 9.4 (Cary, NC), and analyses were conducted in SAS-callable SUDAAN, Version 11, to adjust for survey weighting
AC
(2016). Akaike information and Schwartz criteria were inspected at each step to ensure appropriate model selection. Results Univariate analyses Table 1 contains descriptive statistics of the analytic sample. Overall, 16.7% of students reported ever misusing prescription drugs, with a higher prevalence among males (17.8%) than
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
9
females (15.5%). The largest proportion of those who misused prescription drugs reported doing so only once or twice (45.6% of females, 33.1% of males), though 29.2% of females and 40.7% of males reported misusing at least 10 times. About 15.5% of females (n = 1184) and 6.8% of males (n = 522) identified as sexual minorities. Covariate estimates can be found in the table.
AC
CE
PT
ED
M
AN
US
CR
IP
T
Table 1. Descriptive statistics for students in the 2015 National YRBS ( N=14,477) by sex a Total Female (n=7248) Male (n=7229) Variable n % n % n % Demographics: Sexual orientation Heterosexual 12771 88.96 6064 84.52 6707 93.17 Gay/lesbian 316 1.98 165 1.99 151 1.97 Bisexual 900 5.93 725 9.76 175 2.30 Not sure 490 3.13 294 3.73 196 2.57 Grade 9th grade 3687 26.89 1876 26.10 1811 27.64 10th grade 3642 25.84 1822 26.88 1820 24.86 11th grade 3707 24.01 1834 23.65 1873 24.35 12th grade 3364 23.26 1684 23.38 1680 23.15 Race/ethnicity White 6386 54.46 3222 55.03 3164 53.92 Black/African American 1525 13.52 768 13.37 757 13.67 Hispanic/Latino 4822 22.68 2424 22.82 2398 22.54 Multiple 684 4.50 363 4.61 321 4.39 Other 802 4.84 365 4.16 437 5.48 Prescription drug misuse: Ever used prescription drugs Yes 2501 16.70 1184 15.51 1317 17.82 No 11976 83.30 6064 84.49 5912 82.18 Number of times misused b 1 or 2 times 3 to 9 times 10 to 19 times 20 to 39 times 40 or more times Victimization: Bullied or cyber bullied Yes No Threatened at school
934 680 360 187 340
38.77 25.76 14.09 7.80 13.58
514 324 149 91 106
45.61 25.21 12.86 7.95 8.37
420 356 211 96 234
33.12 26.22 15.10 7.68 17.88
3509 10968
25.70 74.30
2129 5119
32.10 67.90
1380 5849
19.62 80.38
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
10
M
Percentages are out of those who indicated misusing prescription drugs.
ED
b
AN
US
CR
IP
T
Yes 847 5.62 330 4.54 517 6.65 No 13043 94.38 6616 95.46 6427 93.35 Forced to have sex Yes 1056 6.53 791 10.29 265 2.96 No 13256 93.47 6377 89.71 6879 97.04 Mental health: Felt sad or hopeless Yes 4456 29.81 2959 39.81 1497 20.31 No 9902 70.19 4236 60.19 5666 79.69 Considered suicide Yes 2604 17.58 1737.0 23.38 867 12.09 No 11730 82.42 5442.0 76.62 6288 87.91 Other illicit substance use: Smoked marijuana Yes 5675 38.82 2781 37.42 2894 40.16 No 8461 61.18 4350 62.58 4111 59.84 Used another illicit drug Yes 2065 13.21 945 11.96 1120 14.40 No 12412 86.79 6303 88.04 6109 85.60 a Unweighted frequencies (n) are provided; percentages (%) reflect adjusted sampling weights
Bivariate analyses
PT
Table 2 presents bivariate analyses between prescription drug misuse and other variables.
CE
Female and male students who reported misusing prescription drugs comprised significantly greater proportions of SMA than those who reported no misuse. Students who misused also
AC
differed significantly from those who did not on demographic and other covariates save race/ethnicity among males. Relative to heterosexuals, SMA groups reported higher prevalence of prescription drug misuse; all pairwise comparisons were statistically significant except for bisexual males, which approached significance (Supplemental Table A).
ACCEPTED MANUSCRIPT Running head: YRBS PRESCRIPTION DRUGS
11
AC
CE
PT
ED
M
AN
US
CR
IP
T
Table 2. Bivariate differences between students who did and did not misuse prescription drugs in the 2015 National YRBS (N=14,477) by sex a Female (n=7248) Male (n=7229) Misused Did not misuse Misused Did not misuse n=1184 n=6064 n=1317 n=5912 Variable n % n % pb n % n % pb Demographics: Sexual orientation
