HHS Public Access

Author manuscript
J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.
Author Manuscript

Published in final edited form as:

J Psychoactive Drugs. 2019 ; 51(5): 441–452. doi:10.1080/02791072.2019.1654151.

Prevalence and correlates of depressive symptomology among

young adults who use prescription opioids non-medically
Benjamin A. Bouvier, ScBa, Elizabeth N. Kinnard, MScb, Jesse L. Yedinak, MPAa, Yu Li, MD,
PhDa, Beth Elston, MSa, Traci C. Green, PhD, MSca,c,d, Scott E. Hadland, MD, MPHe,f,
Brandon D.L. Marshall, PhDa,*
a.Departmentof Epidemiology, Brown University School of Public Health, 121 South Main Street,
Author Manuscript

Box G-S-121-2, Providence, RI, 02912, USA.

b.Division
of Epidemiology, University of California Berkeley School of Public Health, Berkeley, CA,
United States, 94720.
c.Boston
Medical Center Injury Prevention Center and Boston University School of Medicine,
Department of Emergency Medicine, 771 Albany Street, Room 1332, Boston, MA, 02118, USA.
d.The Warren Alpert School of Medicine of Brown University, Rhode Island Hospital, 55 Claverick
Street, Providence, RI, 02903, USA.
e.Boston Medical Center, Department of Pediatrics, 850 Harrison Avenue, Boston, MA, 02118,
USA.
f.BostonUniversity School of Medicine, Department of Pediatrics, Division of General Pediatrics,
Author Manuscript

88 East Newton Street, Vose Hall Room 322, Boston, MA, 02118, USA.

Abstract
Non-medical prescription opioid (NMPO) use and depression frequently co-occur and are
mutually reinforcing in adults, yet NMPO use and depression in younger populations has been
under-studied. We examined the prevalence and correlates of depressive symptomology among
NMPO-using young adults. The Rhode Island Young Adult Prescription Drug Study (RAPiDS)
recruited young adults in Rhode Island who reported past 30-day NMPO use. We administered the
Center for Epidemiologic Studies Short Depression Scale (CES-D 10), and used modified Poisson
regression to identify the independent correlates of depressive symptomology (CES-D 10 score
≥10). Over half (59.8%, n = 119) screened positive for depressive symptomology. In modified
Poisson regression analysis, diagnostic history of depressive disorder and childhood verbal abuse
Author Manuscript

were associated with depressive symptomology. Participants with depressive symptomology were
more likely to report using prescription opioids non-medically to feel less depressed or anxious, to
avoid withdrawal symptoms, and as a substitute when other drugs are not available. Among young
adult NMPO users, depressive symptomology is prevalent and associated with distinct motivations
for engaging in NMPO use and represents a potential subgroup for intervention. Improving

*
Corresponding Author: Brandon D.L. Marshall, PhD, Associate Professor of Epidemiology, Department of Epidemiology, Brown
University School of Public Health, 121 South Main Street, Box G-S-121-2, Providence, RI, 02912, Tel: 1-401-863-6427 Fax:
1-401-863-317, brandon_marshall@brown.edu.
Disclosure of Interest Statement
The authors report no conflict of interest.
 Bouvier et al. Page 2

guidelines with tools such as screening for depressive symptomology among young adult NMPO
Author Manuscript

users may help prevent NMPO-related harms.

Keywords
prescription opioids; depressive symptomology; depression; young adults; adolescents;
motivations

Introduction
Non-medical prescription opioid (NMPO) use is an ongoing public health problem in the
United States (National Institute on Drug Abuse 2011). Increasing prevalence of NMPO use
in the last two decades has resulted in significant public health, social, and economic
consequences, including a 5.6-fold increase in the number of prescription opioid-attributable
Author Manuscript

fatal overdose deaths from 1999 to 2015 (Kolodny et al. 2015, Chen, Hedegaard, and Warner
2014). The prevalence of NMPO use is highest among young adults aged 18–25, with 6
million (1 in 5) reporting lifetime NMPO use and 2.7 million (1 in 12) reporting NMPO use
in the previous year (Center for Behavioral Health Statistics and Quality 2016, Martins et al.
2017). Young people who engage in NMPO use have an elevated risk of adverse health
consequences, such as accidental overdose (Frank et al. 2015, Lankenau, Teti, Silva, Bloom,
et al. 2012), infectious disease transmission (Hadland and Wood 2012, Surratt, Kurtz, and
Cicero 2011), transition to heroin use (Jones et al. 2015, Cerda et al. 2015, McCabe et al.
2007), and initiation of injection drug use (Lankenau, Teti, Silva, Jackson Bloom, et al.
2012, Green et al. 2011).

NMPO use and depressive symptomology frequently co-occur (Goldner et al. 2014, Becker
Author Manuscript

et al. 2008, Mackesy-Amiti, Donenberg, and Ouellet 2015, Ali et al. 2015, Amari et al.
2011, Fischer et al. 2012, Fink et al. 2015, Edlund et al. 2015). Moreover, the co-occurrence
of NMPO use and depressive symptomology is associated with greater severity and
persistence of both conditions (Kessler 2004, Rowe et al. 2004). This co-occurrence can
result from one or more non-mutually exclusive and often reinforcing pathways. NMPO use
may lead to depressive symptomology (“precipitation” hypothesis), depressive
symptomology may lead to NMPO use (“self-medication” hypothesis), and/or a third factor
might influence the development of both (“shared vulnerability” hypothesis) (Martins et al.
2012, Khantzian 1997).

In order to more fully elucidate the relationship between these two outcomes, understanding
young adult motivations for substance use is crucial, particularly since psychiatric
Author Manuscript

comorbidities may impact these motivations, as well as impact the opportunity for
appropriate intervention (Dow and Kelly 2013, Goodman, Peterson-Badali, and Henderson
2011). Multiple studies have reported on motivations for engaging in NMPO use among
young adults (Young, Glover, and Havens 2012, Boyd et al. 2006, Young et al. 2012,
Drazdowski 2016); however, to date, no study has explored whether young adult NMPO
users with depressive symptomology report distinct motivations for engaging in NMPO use.
Exploring these motivations may also provide insight into the validity of the self-medication
hypothesis of comorbid depressive symptomology and NMPO use (Khantzian 1997).

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 3

Co-occurring depressive symptomology has been studied in the context of heroin use, with
Author Manuscript

many studies reporting a high prevalence among adults who use heroin. Sociodemographic
factors such as female sex, homelessness, adverse childhood experiences, and other factors
were associated with more severe depressive symptomology (Hadland et al. 2011, Wang et
al. 2012, Wu et al. 2016, Tobin and Latkin 2003, Sordo et al. 2012, Chahua et al. 2014).
However, no studies have compared motivations for engaging in heroin use between those
who screen positive versus negative for depressive symptomology (Cornford, Umeh, and
Manshani 2012). More research is needed to understand the co-occurrence of depressive
symptomology and NMPO use among young people, especially in the context of existing
research on the co-occurrence of depressive symptomology and heroin use.

Using data from the Rhode Island Young Adult Prescription Drug Study (RAPiDS), we
explored the prevalence and correlates of depressive symptomology among NMPO-using
young adults. We also report motivations for engaging in NMPO use among participants
Author Manuscript

who screened positive for depressive symptomology compared to those who screened
negative.

Methods
RAPiDS recruiting and enrollment procedures have been previously described (Marshall et
al. 2018). In brief, young adult NMPO users were recruited through targeted canvassing
(e.g., bus advertisements, flyers), internet-based recruitment (e.g., posting to online
classifieds such as Craigslist), and word of mouth, and were invited to participate in
RAPiDS between January 2015 and February 2016 if they: 1) lived in Rhode Island; 2) were
between 18 and 29 years of age; 3) were able to provide informed consent; 4) were able to
speak and feel comfortable completing a survey in English; and 5) reported NMPO use
Author Manuscript

(defined as using prescription opioids without a prescription or not as a doctor directed) in

the past 30 days. If eligible, participants completed a 45-minute, interviewer-administered
survey at the Brown University School of Public Health or a public location of their
choosing. The survey was administered as a computer-assisted interview by a trained
interviewer. Sensitive portions of the survey (e.g., sexual behavior, injection drug use) were
self-administered using a computer. Informed consent was obtained from each participant at
the time of survey. Participants were compensated $25 USD. RAPiDS was approved by the
Brown University Institutional Review Board. A total of 340 persons contacted the study,
230 screened eligible, and 199 completed the survey.

The primary outcome for this analysis was depressive symptomology, assessed using the 10-
item Center for Epidemiologic Studies Short Depression Scale (CES-D 10) (Andresen et al.
1994, Carpenter et al. 1998, Kohout et al. 1993). RAPiDS was a pilot study that used brief
Author Manuscript

assessments; thus, we used the CES-D 10 instead of full diagnostic criteria. The validity and
reliability of the CES-D 10 are established, and the 10-item version shows good predictive
accuracy compared to the 20-item version (Andresen et al. 1994, Carpenter et al. 1998,
Kohout et al. 1993). Moreover, the CES-D 10 has been used to measure depressive
symptomology among drug-using populations (Risser et al. 2010). The CES-D 10 includes
ten statements about how participants have felt in the past week: eight assess negative mood
(e.g. “I felt lonely”) and two assess positive mood (e.g. “I was happy”). Response options

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 4

are: “Rarely or none of the time (0 days),” “Some or little of the time (1–2 days),”
Author Manuscript

“Moderate amount of time (3–4 days),” and “Most or all of the time (5–7 days).” These
response choices were given a score of 0, 1, 2, and 3, respectively, with the two positive
mood items reverse-scored. Scores were considered invalid if more than one item was
missing. If only one item was missing, its value was imputed as the mean of the participant’s
other nine item scores. Total scores were the sum of each item score, with total possible
scores ranging from 0 to 30. Consistent with prior studies, a score of 10 or greater was
considered as screening positive for depressive symptomology (Andresen et al. 1994, Risser
et al. 2010). While some studies have used higher CED-D 10 cut-offs, most studies,
including those involving homeless men on parole and HIV-positive people, use a cut-off of
10 (Nyamathi et al. 2011, Kilburn et al. 2018, Zhang et al. 2012, Lima et al. 2008).
Accordingly, we used a cut-off of 10 for our primary analyses. We also conducted a series of
sensitivity analyses in which results were re-analyzed using a higher cut-off of 15, which
Author Manuscript

was found to result in the most balanced combination of sensitivity (0.76) and specificity
(0.75) in a prior study of patients enrolled in psychiatric partial hospitalization program
(Bjorgvinsson et al. 2013).

Selection of factors associated with depressive symptomology was guided by previous

research on depressive symptomology and NMPO use (Becker et al. 2008, Fink et al. 2015,
Young, Glover, and Havens 2012). Overall, we included participant demographics, drug use
behaviors, mental health history, and adverse childhood experiences. Specifically, we
examined age; sex at birth (male vs. female); race (white, black, or mixed/other); Hispanic
or Latino descent; sexual orientation (straight vs. lesbian, gay, bisexual, queer, or something
else [LGBQ]); past six months homelessness (e.g. living in a shelter because of nowhere else
to go, living in a place not ordinarily used for sleeping); ever incarcerated; ever overdosed by
accident; ever injected or snorted drugs; using prescription opioids non-medically alone; and
Author Manuscript

ever used heroin. We asked participants how often they used prescription opioids non-
medically (never, once or a couple of times, about once per month, at least every week,
every day). We used these answer choices to create a dichotomous variable: at least weekly
NMPO use vs. less than weekly NMPO use. We also determined if and how often
participants used molly/MDMA/ecstasy, mushrooms, GHB, ketamine, crystal
methamphetamine, and cocaine in the last six months (never, once or a couple of times,
about once per month, at least every week, every day). We created a poly-substance use
variable: participants were considered to have engaged in at least monthly polysubstance use
in the last six months if they reported using opioids and at least one of these substances at
least once per month. We asked participants the first age they non-medically used the
prescription opioid they use most regularly. We then created a variable representing years of
non-medical use of this prescription opioid by subtracting this age from their current age.
Author Manuscript

Next, mental health measures were ascertained. Participants were asked: “Have you been
told that you have one or more of these diagnoses?” and were asked to check all the answers
that applied: depressive disorder, bipolar disorder, anxiety disorder, ADHD or ADD, OCD,
eating disorder, and psychosis. Then participants were asked if they had ever been
hospitalized for a mental illness or depression. Finally, adverse childhood experience
questions included ever being insulted or sworn at by a parent before the age of 18; ever
being hit or injured by a parent before the age of 18; while growing up, ever living with

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 5

someone who had a mental illness; while growing up, ever living with someone who was
Author Manuscript

using street drugs; and while growing up, ever living with someone who went to jail or
prison. We also included sexual assault or abuse before the age of 18.

Fisher’s exact test and the Wilcoxon rank sum test were used to determine the bivariate
associations of these variables with depressive symptomology. Next, we developed
multivariate stepwise regression models for a common outcome (screening positive for
depressive symptomology) consistent with established protocols (Lima et al. 2008, Harrell
2015). First, a preliminary model was constructed using all variables significant in bivariate
analyses with a standard cut-off of p < 0.05. In order to see if a more parsimonious model
had a better model fit, we then subjected these variables to a sequential backwards selection
procedure based on QIC value and Type III p-values. The variable with the highest p-value
was removed sequentially until a final model with the lowest QIC was reached. We used
modified robust Poisson regression to estimate standard error in the coefficients and
Author Manuscript

calculate 95% confidence intervals and adjusted prevalence ratios (Zou 2004).

We also asked participants about motivations for NMPO use: “Thinking now about
[prescription opioids], what are your most important reasons for using them without a
doctor’s orders or not as a doctor directed?” The following response options were
randomized, and the interviewer was instructed to read out the list, with participants
encouraged to check any that applied: to feel good or get high, to feel less depressed or
anxious, to relieve physical pain, to get a good sleep, to avoid withdrawal symptoms, to have
a good time with friends, because I was pressured into it, or as a substitute when other drugs
are not available. We conducted exploratory bivariate analyses using Fisher’s exact test with
responses to this question and screening positive vs. negative for depressive symptomology
as measured by the CES-D 10. All analyses were conducted in SAS version 9.3, and all p-
Author Manuscript

values were two-sided.

Results
Almost all participants (99.5%, n = 199) had valid CES-D 10 scores. Valid scores ranged
from 0 to 29, and the mean was 12.43 (SD = 7.07). Corrected item-total correlation was
greater than 0.80 for each of the 10 items, and the overall Cronbach’s alpha was 0.86. Over
half of participants (59.8%, n = 119) had a score greater than or equal to 10, and were thus
considered as screening positive for depressive symptomology. A total of 77 participants
(38.7%) screened positive at the higher cutoff of 15.

The median age of participants was 25 (IQR = 22–27), and 65.3%( n= 130) were assigned
male sex at birth. The majority (61.3%) were white, while 16.6% were black, and 20.6%
Author Manuscript

were mixed, bi-racial, multi-racial, or reported “other.” Twenty-seven participants (13.6%)

reported Hispanic or Latino descent (see Table 1).. When asked how often in the last six
months participants used these prescription opioids they used most, 39 (20.7%) answered
once or a couple of times, 45 (23.9%) answered about once a month, 67 (35.6%) answered
at least every week, and 26 (13.8%) answered every day. Fifty-six (28.1%) engaged in at
least monthly illicit polysubstance use in the last six months.

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 6

As shown in Table 1, compared to participants who screened negative for depressive

Author Manuscript

symptomology, participants who screened positive were more likely to: be female; be
LGBQ; have been homeless in the preceding six months; engage in at least monthly illicit
polysubstance use in the preceding six months; have ever been told they had a depressive
disorder, bipolar disorder, and/or anxiety disorder; have ever been hospitalized for a mental
illness or depression; lived with someone who had a mental illness growing up; and have
experienced adverse childhood experiences, including being sexually assaulted or abused,
being hit or injured by a parent, and/or being insulted or sworn at by a parent before the age
of 18.

In Table 2, we show our final exploratory multivariate model, which includes all the
variables with significant associations (p < 0.05) in bivariate analyses. We explored whether
removing variables with sequential backwards selection resulted in a more parsimonious
model with better fit, but removing variables resulted in models with higher QIC values. We
Author Manuscript

found that this first multivariable model with all the variables with significant bivariate
associations had the lowest QIC, and we decided that this was our final multivariable model.
In this model, ever being told about having a depressive disorder (adjusted prevalence ratio =
1.51, 95% CI: 1.14–1.99, p < 0.01) and being insulted or sworn at by a parent before the age
of 18 (adjusted prevalence ratio = 1.50, 95% CI: 1.05–2.14, p = 0.02) remained significant
when adjusting for the other covariates. The p-values for several variables shown in Table 2
were greater than the 0.05 cut-off for backward selection, but we decided to keep these
variables in the final model because removing them increased the QIC.

As shown in Table 3, participants who screened positive for depressive symptomology were
more likely to report using prescription opioids non-medically to feel less depressed or
anxious, to avoid withdrawal symptoms, and as a substitute when other drugs are not
Author Manuscript

available.

Results were broadly similar in a series of sensitivity analyses which used a higher cutoff of
15. However, several drug-related behaviors were associated with depressive
symptomatology in bivariate sensitivity analyses, including history of injection drug use
(39.5% vs. 23.8% screening positive among those with and without a history of injection
drug use, respectively, p < 0.02); history of sniffing or snorting an opioid (67.5% vs. 53.3%,
p < 0.048); and history of heroin use (52.0% vs. 36.9%, p < 0.036). The multivariable results
were similar, although recent homelessness attained statistical significance (adjusted
prevalence ratio = 1.44; CI: 1.01 – 2.06; p = 0.049). Motivations for using prescription
opioids non-medically were also similar, although persons who screened positive for
depressive symptomatology at the higher cutoff were more likely to report NMPO use to get
Author Manuscript

a good sleep (64.9% vs. 47.5%, p = 0.017).

Discussion
Among our sample of young NMPO users in Rhode Island, almost six in ten screened
positive for depressive symptomology according to the CES-D 10, confirming the findings
of two studies that have reported a high prevalence of co-occurring depressive
symptomology among young adult NMPO users (Goldner et al. 2014, Fischer et al. 2012).

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 7

Diagnostic history of depressive disorder and being insulted or sworn at by a parent before
Author Manuscript

the age of 18 were independent correlates of current depressive symptomology in our

modified Poisson regression analysis. While the prior diagnosis may be a reasonable
predictor of current symptomology, the relationship between adverse childhood experiences
and the severity or persistence of such symptomology among this population warrants
further exploration. To our knowledge, our study is the first to report that among young
adults who engage in NMPO use, depressive symptomology is associated with being LGBQ.
This finding emphasizes the need to understand the co-occurring, syndemic factors and
disparities affecting the population of young adult NMPO users and subpopulations such as
young LGBQ NMPO users (Pachankis et al. 2018, Branstrom and Pachankis 2018, Wang,
Burton, and Pachankis 2018, White Hughto et al. 2017). The finding that LGBQ status was
not independently associated with depressive symptomology also requires further
investigation. It is possible that other variables in the model (such as sex at birth and recent
Author Manuscript

homelessness) confound the relationship between LGBQ status and depressive symptoms.
Nonetheless, we believe that sexual orientation should be considered when conceptualizing
the development of co-occurrence of NMPO use, depressive symptomology, and other
outcomes.

Our finding that depressive symptomology was associated with using prescription opioids
non-medically to feel less depressed or anxious may support the self-medication hypothesis
(i.e. depressive symptomology may lead to NMPO use). Depressive symptomology was also
associated with using prescription opioids non-medically to avoid withdrawal symptoms.
This finding may be complicated by the fact that anxious or depressive symptoms can result
from opioid withdrawal (World Health Organization 2009). Additional longitudinal research
is needed to elucidate the onset and directionality of associations between mental health
distress, NMPO use, and withdrawal symptoms. Furthermore, future research should
Author Manuscript

differentiate self-medication of anxious or depressive symptoms with self-medication of

withdrawal symptoms among young adults who use prescription opioids non-medically.
Overall, these findings demonstrate that young adults with depressive symptomology have
distinct motivations for NMPO use.

We did not find that age of NMPO use onset and years of NMPO use were associated with
depressive symptomology. Given that our sensitivity analyses revealed such associations
with history of injection drug use, history of sniffing or snorting an opioid, and history of
heroin use, attention should be paid to the dynamics and trajectories of young adult NMPO
use—specifically, route of administration of any opioid, and transitions to heroin use or
injection. These measures have a bearing on substance use and mental health trajectories and
may provide insight into the co-occurrence of NMPO use, mental health symptoms, and
Author Manuscript

other outcomes (Twombly and Holtz 2008, Surratt, Kurtz, and Cicero 2011, McCabe et al.
2007, Marshall et al. 2016, Lankenau, Teti, Silva, Bloom, et al. 2012, Green et al. 2011,
Goodman, Peterson-Badali, and Henderson 2011, Frank et al. 2015, Drazdowski 2016, Dow
and Kelly 2013). As reasons for using prescription opioids non-medically often change over
time, research should explore mental health symptomology and drug using patterns over
time to capture key transition points in NMPO use trajectories (Boyd et al. 2006, Dow and
Kelly 2013, Drazdowski 2016, Goodman, Peterson-Badali, and Henderson 2011). All of
these dynamics have implications for addressing the rapidly increasing opioid use epidemic

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 8

among young adults (Martins et al. 2017, Marshall et al. 2016, Young, Glover, and Havens
Author Manuscript

2012), and to target such efforts to sub-populations such as young LGBQ people who use
drugs.

We recommend screening for both NMPO use and depressive symptomology among young
adults because NMPO use is highest among this population (Center for Behavioral Health
Statistics and Quality 2016), and screening provides an opportunity for early intervention
against adverse outcomes (Marshall et al. 2016), as well as informing appropriate treatment
options (Dow and Kelly 2013). Nationally, self-medication trends are leading to changes in
prescribing patterns: the latest National Survey on Drug Use and Health study on NMPO
suggests addressing prescribing guidelines due to the high prevalence of self-medicating for
pain (Han et al. 2017). This study’s findings suggest the need for better screening in various
settings to identify subgroups at risk for harms related to depressive symptomology and/or
NMPO use. It is important to note that even when young people are screened for depressive
Author Manuscript

symptomology, a low proportion actually receives services after screening (Young et al.
2012, Mackesy-Amiti, Donenberg, and Ouellet 2015, Allgaier et al. 2014). Furthermore,
many people are more likely to seek mental health services than they are to seek substance
use services (Khantzian 1997, Fischer et al. 2012, Allgaier et al. 2014). These points stress
the need to not only increase screening, but to implement protocols for ensuring individuals
are connected to the care they need, to integrate mental health and substance use treatment,
and to reduce the stigma associated with receiving such services (Young et al. 2012, Wang,
Burton, and Pachankis 2018, Marshall et al. 2016).

Our study is consistent with research on the co-occurrence of depressive symptomology and
heroin use, in that we similarly found a high prevalence of depressive symptomology;
moreover, variables including female sex, homelessness, polysubstance use, and adverse
Author Manuscript

childhood experiences were associated with depressive symptomology. However, we are

unable to contextualize our findings on the distinct motivations for NMPO use among young
adults with depressive symptomology, due to the fact that similar research on motivations for
heroin use among young adults has not yet been conducted. More research on motivations
for using prescription opioids and other drugs is needed to fully contextualize this research
and understand substance use trajectories.

If depressive symptomology does indeed lead to NMPO use (i.e., the self-medication
pathway), screening young adults for both depressive symptomology and NMPO use in a
variety of healthcare settings (e.g. medical office, emergency department) may help to
identify those at increased risk and offer the opportunity to intervene before overdose or
transition to heroin use occur (Twombly and Holtz 2008). This may be particularly relevant
Author Manuscript

for young adults who are prescribed opioids for pain (Simoni-Wastila and Tompkins 2001).
We were unable to test for evidence of the precipitation pathway (where NMPO use leads to
depressive symptomology), given the cross-sectional nature of the study. Finally, our study
suggests there may be an opportunity to explore the relationship between adverse childhood
experiences and the presence or severity of depressive symptomology among young adults
who use opioids nonmedically (Green et al. 2009, Salas et al. 2016, Garland, Pettus-Davis,
and Howard 2013, Young, Glover, and Havens 2012).

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 9

Our study has several limitations. First, although the CES-D 10 is a validated screening
Author Manuscript

instrument which has been used among substance-using populations, some degree of
misclassification is possible (Andresen et al. 1994, Carpenter et al. 1998, Kohout et al.
1993). While we decided to use a cut-off of 10 for the CES-D 10 to be consistent with prior
research, a higher cut-off could have been used, given that such a large proportion of our
sample screened positive for depressive symptomology with the cut-off of 10. Future
research should continue to explore what cut-off should be used for the CES-D 10 among
young adults who use prescription opioids non-medically. Second, we had participants
choose their primary motivations for using prescription opioids non-medically from a list,
and it is possible that we did not accurately capture participants’ motivations for NMPO use
(Drazdowski 2016). While motivations for prescription drug use change over time in young
adulthood (Garnier-Dykstra et al. 2012, Hartung et al. 2013), we were unable to capture such
evolving trajectories due to the cross-sectional nature of our study. Third, because our study
Author Manuscript

was cross-sectional, we could not identify any causal relationships.. Longitudinal studies are
needed to better understand causal relationships among depressive symptomology, NMPO
use, and other variables. Fourth, while our selection of variables for our bivariate analyses
was driven by prior literature, our multivariable analysis was based on backwards selection
from these variables, rather than being exclusively theory-based. We decided that since our
analysis was exploratory in nature, a data-driven variable selection procedure was
appropriate. Future research investigating the etiology of depressive symptomology among
young adult NMPO users should be more theory-based. Fifth, when we asked about
participants’ psychiatric diagnoses, we asked if they had “been told” that they had one or
more of the listed diagnoses without specifically asking if they had been told so by a medical
professional. Future research should be more specific when asking about diagnostic history,
and use standardized instruments to capture mental health diagnoses. Sixth, we did not use a
standardized scale or comprehensive set of adverse childhood experiences measures, so we
Author Manuscript

decided to analyze each adverse childhood experience individually. Future research should
employ a standardized (and more comprehensive) adverse childhood experiences scale.
Seventh, although all participants were considered NMPO users according to the study’s
eligibility criteria, we did not obtain a uniform sample in terms of types of use. For example,
a participant who misused their own prescription opioids that were previously prescribed for
a legitimate injury, and a participant who bought prescription opioids from the illicit market
would both be considered NMPO users. Thus, our findings cannot be extrapolated to a
specific subpopulation of NMPO-using young adults. Due to the fact that we did not use
verification strategies such as urine drug screens, it is possible that inaccurate reporting
existed regarding participants NMPO users. Finally, although we used diverse recruitment
strategies, our findings may not be generalizable to all young adults who engage in NMPO
use.
Author Manuscript

Conclusion
Our findings emphasize the need to understand motivations for NMPO use among young
adults with comorbid depressive symptomology, so we can better identify those at higher
risk of health consequences by increasing screening, and provide opportunities for
appropriate intervention. Our research demonstrates possible support for the self-medication

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 10

hypothesis of co-occurring NMPO use and depressive symptomology, but emphasizes the
Author Manuscript

need to further investigate all potential hypotheses of the development of this co-occurrence
through longitudinal design (Kessler 2004). We hope that this research continues to inform
the development of improved guidelines and other national strategies to effectively reduce
the harms of co-occurring depressive symptomology and NMPO use among young adults.

Acknowledgments
We thank RAPiDS study participants and staff for their contribution to the research.

Funding details

This work was supported by the US National Institute on Drug Abuse under Grant R03-DA037770.

REFERENCES
Author Manuscript

Ali MM, Dean D Jr., Lipari R, Dowd WN, Aldridge AP, and Novak SP. 2015 “The mental health
consequences of nonmedical prescription drug use among adolescents.” J Ment Health Policy Econ
18 (1):3–15. [PubMed: 25862204]
Allgaier AK, Krick K, Opitz A, Saravo B, Romanos M, and Schulte-Korne G. 2014 “Improving early
detection of childhood depression in mental health care: the Childrens Depression Screener (ChilD-
S).” Psychiatry Res 217 (3):248–52. doi: 10.1016/j.psychres.2014.03.037. [PubMed: 24742687]
Amari E, Rehm J, Goldner E, and Fischer B. 2011 “Nonmedical prescription opioid use and mental
health and pain comorbidities: a narrative review.” Can J Psychiatry 56 (8):495–502. doi:
10.1177/070674371105600808. [PubMed: 21878161]
Andresen EM, Malmgren JA, Carter WB, and Patrick DL. 1994 “Screening for depression in well
older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression
Scale).” Am J Prev Med 10 (2):77–84. [PubMed: 8037935]
Becker WC, Sullivan LE, Tetrault JM, Desai RA, and Fiellin DA. 2008 “Non-medical use, abuse and
dependence on prescription opioids among U.S. adults: psychiatric, medical and substance use
Author Manuscript

correlates.” Drug Alcohol Depend 94 (1–3):38–47. doi: 10.1016/j.drugalcdep.2007.09.018.

[PubMed: 18063321]
Bjorgvinsson T, Kertz SJ, Bigda-Peyton JS, McCoy KL, and Aderka IM. 2013 “Psychometric
properties of the CES-D-10 in a psychiatric sample.” Assessment 20 (4):429–36. doi:
10.1177/1073191113481998. [PubMed: 23513010]
Boyd CJ, McCabe SE, Cranford JA, and Young A. 2006 “Adolescents’ motivations to abuse
prescription medications.” Pediatrics 118 (6):2472–80. doi: 10.1542/peds.2006-1644. [PubMed:
17142533]
Branstrom R, and Pachankis JE. 2018 “Validating the Syndemic Threat Surrounding Sexual Minority
Men’s Health in a Population-Based Study With National Registry Linkage and a Heterosexual
Comparison.” J Acquir Immune Defic Syndr 78 (4):376–382. doi: 10.1097/QAI.
0000000000001697. [PubMed: 29940602]
Carpenter JS, Andrykowski MA, Wilson J, Hall LA, Rayens MK, Sachs B, and Cunningham LL. 1998
“Psychometrics for two short forms of the Center for Epidemiologic Studies-Depression Scale.”
Issues Ment Health Nurs 19 (5):481–94. [PubMed: 9782864]
Author Manuscript

Center for Behavioral Health Statistics and Quality. 2016 2015 National Survey on Drug Use and
Health: Detailed Tables. Rockville, MD: Substance Abuse and Mental Health Services
Administration.
Cerda M, Santaella J, Marshall BD, Kim JH, and Martins SS. 2015 “Nonmedical Prescription Opioid
Use in Childhood and Early Adolescence Predicts Transitions to Heroin Use in Young Adulthood:
A National Study.” J Pediatr 167 (3):605–12 e1–2. doi: 10.1016/j.jpeds.2015.04.071. [PubMed:
26054942]
Chahua M, Sordo L, Barrio G, Domingo-Salvany A, Brugal MT, Molist G, de la Fuente L, Bravo MJ,
and Itinere Project Group. 2014 “Non-fatal opioid overdose and major depression among street-

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 11

recruited young heroin users.” Eur Addict Res 20 (1):1–7. doi: 10.1159/000346787. [PubMed:
23921233]
Author Manuscript

Chen LH, Hedegaard H, and Warner M. 2014 “Drug-poisoning Deaths Involving Opioid Analgesics:
United States, 1999–2011.” NCHS Data Brief (166):1–8.
Cornford CS, Umeh K, and Manshani N. 2012 “Heroin users’ experiences of depression: a qualitative
study.” Fam Pract 29 (5):586–92. doi: 10.1093/fampra/cms014. [PubMed: 22357579]
Dow SJ, and Kelly JF. 2013 “Listening to youth: Adolescents’ reasons for substance use as a unique
predictor of treatment response and outcome.” Psychol Addict Behav 27 (4):1122–31. doi:
10.1037/a0031065. [PubMed: 23276316]
Drazdowski TK 2016 “A systematic review of the motivations for the non-medical use of prescription
drugs in young adults.” Drug Alcohol Depend 162:3–25. doi: 10.1016/j.drugalcdep.2016.01.011.
[PubMed: 26851986]
Edlund MJ, Forman-Hoffman VL, Winder CR, Heller DC, Kroutil LA, Lipari RN, and Colpe LJ. 2015
“Opioid abuse and depression in adolescents: Results from the National Survey on Drug Use and
Health.” Drug Alcohol Depend 152:131–8. doi: 10.1016/j.drugalcdep.2015.04.010. [PubMed:
25981310]
Author Manuscript

Fink DS, Hu R, Cerda M, Keyes KM, Marshall BD, Galea S, and Martins SS. 2015 “Patterns of major
depression and nonmedical use of prescription opioids in the United States.” Drug Alcohol
Depend 153:258–64. doi: 10.1016/j.drugalcdep.2015.05.010. [PubMed: 26026492]
Fischer B, Lusted A, Roerecke M, Taylor B, and Rehm J. 2012 “The prevalence of mental health and
pain symptoms in general population samples reporting nonmedical use of prescription opioids: a
systematic review and meta-analysis.” J Pain 13 (11):1029–44. doi: 10.1016/j.jpain.2012.07.013.
[PubMed: 23040158]
Frank D, Mateu-Gelabert P, Guarino H, Bennett A, Wendel T, Jessell L, and Teper A. 2015 “High risk
and little knowledge: overdose experiences and knowledge among young adult nonmedical
prescription opioid users.” Int J Drug Policy 26 (1):84–91. doi: 10.1016/j.drugpo.2014.07.013.
[PubMed: 25151334]
Garland EL, Pettus-Davis C, and Howard MO. 2013 “Self-medication among traumatized youth:
structural equation modeling of pathways between trauma history, substance misuse, and
psychological distress.” J Behav Med 36 (2):175–85. doi: 10.1007/s10865-012-9413-5. [PubMed:
22454227]
Author Manuscript

Garnier-Dykstra LM, Caldeira KM, Vincent KB, O’Grady KE, and Arria AM. 2012 “Nonmedical use
of prescription stimulants during college: four-year trends in exposure opportunity, use, motives,
and sources.” J Am Coll Health 60 (3):226–34. doi: 10.1080/07448481.2011.589876. [PubMed:
22420700]
Goldner EM, Lusted A, Roerecke M, Rehm J, and Fischer B. 2014 “Prevalence of Axis-1 psychiatric
(with focus on depression and anxiety) disorder and symptomatology among non-medical
prescription opioid users in substance use treatment: systematic review and meta-analyses.” Addict
Behav 39 (3):520–31. doi: 10.1016/j.addbeh.2013.11.022. [PubMed: 24333033]
Goodman I, Peterson-Badali M, and Henderson J. 2011 “Understanding motivation for substance use
treatment: the role of social pressure during the transition to adulthood.” Addict Behav 36 (6):660–
8. doi: 10.1016/j.addbeh.2011.01.011. [PubMed: 21295918]
Green TC, Black R, Grimes Serrano JM, Budman SH, and Butler SF. 2011 “Typologies of prescription
opioid use in a large sample of adults assessed for substance abuse treatment.” PLoS ONE 6
(11):e27244. doi: 10.1371/journal.pone.0027244. [PubMed: 22087270]
Author Manuscript

Green TC, Grimes Serrano JM, Licari A, Budman SH, and Butler SF. 2009 “Women who abuse
prescription opioids: Findings from the Addiction Severity Index-Multimedia Version Connect
prescription opioid database.” Drug Alcohol Depend 103 (1–2):65–73. doi: 10.1016/j.drugalcdep.
2009.03.014. [PubMed: 19409735]
Hadland SE, Marshall BD, Kerr T, Qi J, Montaner JS, and Wood E. 2011 “Depressive symptoms and
patterns of drug use among street youth.” J Adolesc Health 48 (6):585–90. doi: 10.1016/
j.jadohealth.2010.09.009. [PubMed: 21575818]

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 12

Hadland SE, and Wood E. 2012 “Commentary on Bruneau et al. (2012): injection of prescription
opioid pain relievers and infectious disease risk.” Addiction 107 (7):1328–9. doi: 10.1111/j.
Author Manuscript

1360-0443.2012.03885.x. [PubMed: 22672378]

Han Beth, Compton Wilson M, Blanco C, Crane Elizabeth, Lee Jinhee, and Jones CM. 2017
“Prescription Opioid Use, Misuse, and Use Disorders in U.S. Adults: 2015 National Survey on
Drug Use and Health “ Annals of Internal Medicine 167 (5).
Harrell FJ. 2015 Regression modeling strategies: Springer International Publishing.
Hartung CM, Canu WH, Cleveland CS, Lefler EK, Mignogna MJ, Fedele DA, Correia CJ, Leffingwell
TR, and Clapp JD. 2013 “Stimulant medication use in college students: comparison of appropriate
users, misusers, and nonusers.” Psychol Addict Behav 27 (3):832–40. doi: 10.1037/a0033822.
[PubMed: 24059834]
Jones CM, Logan J, Gladden RM, and Bohm MK. 2015 “Vital Signs: Demographic and Substance Use
Trends Among Heroin Users - United States, 2002–2013.” MMWR Morb Mortal Wkly Rep 64
(26):719–25. [PubMed: 26158353]
Kessler RC 2004 “The epidemiology of dual diagnosis.” Biol Psychiatry 56 (10):730–7. doi: 10.1016/
j.biopsych.2004.06.034. [PubMed: 15556117]
Author Manuscript

Khantzian EJ 1997 “The self-medication hypothesis of substance use disorders: a reconsideration and
recent applications.” Harv Rev Psychiatry 4 (5):231–44. doi: 10.3109/10673229709030550.
[PubMed: 9385000]
Kilburn K, Prencipe L, Hjelm L, Peterman A, Handa S, and Palermo T. 2018 “Examination of
performance of the Center for Epidemiologic Studies Depression Scale Short Form 10 among
African youth in poor, rural households.” BMC Psychiatry 18 (1):201. doi: 10.1186/
s12888-018-1774-z. [PubMed: 29914413]
Kohout FJ, Berkman LF, Evans DA, and Cornoni-Huntley J. 1993 “Two shorter forms of the CES-D
(Center for Epidemiological Studies Depression) depression symptoms index.” J Aging Health 5
(2):179–93. doi: 10.1177/089826439300500202. [PubMed: 10125443]
Kolodny A, Courtwright DT, Hwang CS, Kreiner P, Eadie JL, Clark TW, and Alexander GC. 2015
“The prescription opioid and heroin crisis: a public health approach to an epidemic of addiction.”
Annu Rev Public Health 36:559–74. doi: 10.1146/annurev-publhealth-031914-122957. [PubMed:
25581144]
Lankenau SE, Teti M, Silva K, Bloom JJ, Harocopos A, and Treese M. 2012 “Patterns of prescription
Author Manuscript

drug misuse among young injection drug users.” J Urban Health 89 (6):1004–16. doi: 10.1007/
s11524-012-9691-9. [PubMed: 22684424]
Lankenau SE, Teti M, Silva K, Bloom J. Jackson, Harocopos A, and Treese M. 2012 “Initiation into
prescription opioid misuse amongst young injection drug users.” Int J Drug Policy 23 (1):37–44.
doi: 10.1016/j.drugpo.2011.05.014. [PubMed: 21689917]
Lima VD, Harrigan R, Murray M, Moore DM, Wood E, Hogg RS, and Montaner JS. 2008
“Differential impact of adherence on long-term treatment response among naive HIV-infected
individuals.” AIDS 22 (17):2371–80. doi: 10.1097/QAD.0b013e328315cdd3
00002030-200811120-00017 [pii]. [PubMed: 18981777]
Mackesy-Amiti ME, Donenberg GR, and Ouellet LJ. 2015 “Prescription opioid misuse and mental
health among young injection drug users.” Am J Drug Alcohol Abuse 41 (1):100–6. doi:
10.3109/00952990.2014.940424. [PubMed: 25105884]
Marshall BDL, Green TC, Elston B, Yedinak JL, Hadland SE, and Clark MA. 2018 “The Effectiveness
of Internet- and Field-Based Methods to Recruit Young Adults Who Use Prescription Opioids
Author Manuscript

Nonmedically.” Subst Use Misuse 53 (10):1688–1699. doi: 10.1080/10826084.2018.1425725.

[PubMed: 29364768]
Marshall BDL, Green TC, Yedinak JL, and Hadland SE. 2016 “Harm reduction for young people who
use prescription opioids extra-medically: Obstacles and opportunities.” Int J Drug Policy 10.1016/
j.drugpo.2016.01.022. doi: 10.1016/j.drugpo.2016.01.022.
Martins SS, Fenton MC, Keyes KM, Blanco C, Zhu H, and Storr CL. 2012 “Mood and anxiety
disorders and their association with non-medical prescription opioid use and prescription opioid-
use disorder: longitudinal evidence from the National Epidemiologic Study on Alcohol and

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 13

Related Conditions.” Psychol Med 42 (6):1261–72. doi: 10.1017/s0033291711002145. [PubMed:

21999943]
Author Manuscript

Martins SS, Segura Luis E., Santaella-Tenorio Julian, Perlmutter Alexander, Fenton Miriam C., Cerdá
Magdalena, Keyes Katherine M., Ghandour Lilian A., Storr Carla L., and Hasin Deborah S.. 2017
“Prescription opioid use disorder and heroin use among 12–34 year-olds in the United States from
2002 to 2014.” Addictive Behaviors 65 (Supplement C):236–241. doi: 10.1016/j.addbeh.
2016.08.033. [PubMed: 27614657]
McCabe SE, Cranford JA, Boyd CJ, and Teter CJ. 2007 “Motives, diversion and routes of
administration associated with nonmedical use of prescription opioids.” Addict Behav 32 (3):562–
75. doi: 10.1016/j.addbeh.2006.05.022. [PubMed: 16843611]
National Institute on Drug Abuse. 2011 NIDA Research Report Series: Prescription Drug Abuse.
Bethesda, MD: National Institutes of Health.
Nyamathi A, Leake B, Albarran C, Zhang S, Hall E, Farabee D, Marlow E, Marfisee M, Khalilifard F,
and Faucette M. 2011 “Correlates of depressive symptoms among homeless men on parole.”
Issues Ment Health Nurs 32 (8):501–11. doi: 10.3109/01612840.2011.569111. [PubMed:
21767252]
Author Manuscript

Pachankis JE, Hatzenbuehler ML, Wang K, Burton CL, Crawford FW, Phelan JC, and Link BG. 2018
“The Burden of Stigma on Health and Well-Being: A Taxonomy of Concealment, Course,
Disruptiveness, Aesthetics, Origin, and Peril Across 93 Stigmas.” Pers Soc Psychol Bull 44 (4):
451–474. doi: 10.1177/0146167217741313. [PubMed: 29290150]
Risser J, Cates A, Rehman H, and Risser W. 2010 “Gender differences in social support and depression
among injection drug users in Houston, Texas.” Am J Drug Alcohol Abuse 36 (1):18–24. doi:
10.3109/00952990903544802. [PubMed: 20141392]
Rowe CL, Liddle HA, Greenbaum PE, and Henderson CE. 2004 “Impact of psychiatric comorbidity
on treatment of adolescent drug abusers.” J Subst Abuse Treat 26 (2):129–40. doi: 10.1016/
S0740-5472(03)00166-1. [PubMed: 15050090]
Salas J, Scherrer JF, Lustman PJ, and Schneider FD. 2016 “Racial differences in the association
between nonmedical prescription opioid use, abuse/dependence, and major depression.” Subst
Abus 37 (1):25–30. doi: 10.1080/08897077.2015.1129523. [PubMed: 26675823]
Simoni-Wastila L, and Tompkins C. 2001 “Balancing diversion control and medical necessity: the case
of prescription drugs with abuse potential.” Subst Use Misuse 36 (9–10):1275–96. [PubMed:
Author Manuscript

11592473]
Sordo L, Chahua M, Bravo MJ, Barrio G, Brugal MT, Domingo-Salvany A, Molist G, De la Fuente L,
and Itinere Project Group. 2012 “Depression among regular heroin users: the influence of gender.”
Addict Behav 37 (1):148–52. doi: 10.1016/j.addbeh.2011.09.009. [PubMed: 21968230]
Surratt H, Kurtz SP, and Cicero TJ. 2011 “Alternate routes of administration and risk for HIV among
prescription opioid abusers.” J Addict Dis 30 (4):334–41. doi: 10.1080/10550887.2011.609805.
[PubMed: 22026525]
Tobin KE, and Latkin CA. 2003 “The relationship between depressive symptoms and nonfatal
overdose among a sample of drug users in Baltimore, Maryland.” J Urban Health 80 (2):220–9.
doi: 10.1093/jurban/jtg025. [PubMed: 12791798]
Twombly EC, and Holtz KD. 2008 “Teens and the misuse of prescription drugs: evidence-based
recommendations to curb a growing societal problem.” J Prim Prev 29 (6):503–16. doi: 10.1007/
s10935-008-0157-5. [PubMed: 19015990]
Wang K, Burton CL, and Pachankis JE. 2018 “Depression and Substance Use: Towards the
Author Manuscript

Development of an Emotion Regulation Model of Stigma Coping.” Subst Use Misuse 53 (5):859–
866. doi: 10.1080/10826084.2017.1391011. [PubMed: 29125383]
Wang P, Wu HC, Yen CN, Yeh YC, Chung KS, Chang HC, and Yen CF. 2012 “Predictors of the
severity of depressive symptoms among intravenous heroin users receiving methadone
maintenance treatment in Taiwan: an 18-month follow-up study.” Psychol Addict Behav 26 (1):
145–50. doi: 10.1037/a0025175. [PubMed: 21859169]
White Hughto JM, Pachankis JE, Willie TC, and Reisner SL. 2017 “Victimization and depressive
symptomology in transgender adults: The mediating role of avoidant coping.” J Couns Psychol 64
(1):41–51. doi: 10.1037/cou0000184. [PubMed: 28068130]

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 14

World Health Organization. 2009 Clinical Guidelines for Withdrawal Management and Treatment of
Drug Dependence in Closed Settings. Geneva.
Author Manuscript

Wu Y, Yan S, Bao Y, Lian Z, Qu Z, and Liu Z. 2016 “Cross-sectional study of the severity of self-
reported depressive symptoms in heroin users who participate in a methadone maintenance
treatment program.” Shanghai Arch Psychiatry 28 (1):35–41. doi: 10.11919/j.issn.
1002-0829.215127. [PubMed: 27688642]
Young AM, Glover N, and Havens JR. 2012 “Nonmedical use of prescription medications among
adolescents in the United States: a systematic review.” J Adolesc Health 51 (1):6–17. doi: 10.1016/
j.jadohealth.2012.01.011. [PubMed: 22727071]
Young AM, McCabe SE, Cranford JA, Ross-Durow P, and Boyd CJ. 2012 “Nonmedical use of
prescription opioids among adolescents: subtypes based on motivation for use.” J Addict Dis 31
(4):332–41. doi: 10.1080/10550887.2012.735564. [PubMed: 23244552]
Zhang W, O’Brien N, Forrest JI, Salters KA, Patterson TL, Montaner JS, Hogg RS, and Lima VD.
2012 “Validating a shortened depression scale (10 item CES-D) among HIV-positive people in
British Columbia, Canada.” PLoS One 7 (7):e40793. doi: 10.1371/journal.pone.0040793.
[PubMed: 22829885]
Author Manuscript

Zou G 2004 “A modified poisson regression approach to prospective studies with binary data.” Am J
Epidemiol 159 (7):702–6. [PubMed: 15033648]
Author Manuscript
Author Manuscript

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 15

Table 1:

Factors associated with depressive symptomology among young adult non-medical prescription opioid users in
Author Manuscript

Rhode Island (N = 199)

Characteristic Total n (%) n=199 Screened Positive for Screened Negative for p - value
Depressive Symptomology n Depressive Symptomology n
(%) n=119 (%) n=80
a 25 (22–27) 25 (22–27) 24 (22–27) 0.79
Age (median, IQR)
Sex at birth
Female 69 (34.7) 48 (40.3) 21 (26.3) 0.05
Male 130 (65.3) 71 (59.7) 59 (73.8)
Race
Black, African, Haitian, or Cape 33 (16.6) 22 (18.5) 11 (13.8) 0.51
Verdean
White 122 (61.3) 69 (58.0) 53 (66.3)
Author Manuscript

Aggregated “Mixed/Other” 41 (20.6) 26 (21.8) 15 (18.8)

Hispanic or Latino
Yes 27 (13.6) 20 (16.8) 7 (8.8) 0.14
No 172 (86.4) 99 (83.2) 73 (91.3)
Sexual orientation
Lesbian, Gay, Bisexual, Queer,
or something else 27 (13.6) 21 (17.6) 6 (7.5) 0.05
Straight 171 (85.9) 97 (81.5) 74 (92.5)
Been homeless in the last six
months
Yes 50 (25.1) 40 (33.6) 10 (12.5) <0.01
No 149 (74.9) 79 (66.4) 70 (85.5)
Ever incarcerated
Author Manuscript

Yes 93 (46.7) 59 (49.6) 34 (42.5) 0.39

No 106 (53.3) 60 (50.4) 46 (57.5)
Ever overdosed by accident
Yes 53 (26.6) 34 (28.6) 19 (23.8) 0.51
No 146 (73.4) 85 (71.4) 61 (76.3)
Ever injected drugs
Yes 59 (29.6) 41 (34.5) 18 (22.5) 0.08
No 139 (69.8) 77 (64.7) 62 (77.5)
Ever sniffed or snorted an opioid
Yes 116 (58.3) 73 (61.3) 43 (53.8) 0.24
No 81 (40.7) 44 (37.0) 37 (46.3)
Typically use prescription opioids non-medically alone
Author Manuscript

Yes 146 (73.4) 89 (74.8) 57 (71.3) 0.51

No 51 (25.6) 28 (23.5) 23 (28.8)
Ever used heroin
Yes 85 (42.7) 55 (46.2) 30 (37.5) 0.24
No 114 (57.3) 64 (53.8) 50 (62.5)

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 16

Characteristic Total n (%) n=199 Screened Positive for Screened Negative for p - value
Depressive Symptomology n Depressive Symptomology n
(%) n=119 (%) n=80
Author Manuscript

At least weekly non-medical prescription opioid use

Yes 93 (46.7) 59 (49.6) 34 (42.5) 0.24
No 95 (47.7) 52 (43.7) 43 (53.8)
At least monthly illicit polysubstance use in the last six months
Yes 56 (28.1) 40 (33.6) 16 (20.0) 0.04
No 143 (71.9) 79 (66.4) 64 (80.0)
a
Age of first non-medical use of the prescription opioid used most regularly (median, IQR)
19 (17–22) 19 (17–22) 19 (17–21) 0.22
a
Years of non-medical use of the prescription opioid used most regularly (median, IQR)
5 (2–8) 4 (2–7) 5 (1–8) 0.46
Been diagnosed with a depressive
disorder
Author Manuscript

Yes 95 (47.8) 72 (60.5) 23 (28.8) <0.01

No 101 (50.8) 45 (37.8) 56 (70.0)
Been diagnosed with a bipolar
disorder
Yes 49 (24.6) 39 (32.8) 10 (12.5) <0.01
No 147 (73.9) 78 (65.5) 69 (86.3)
Been diagnosed with an anxiety
disorder
Yes 98 (49.2) 66 (55.5) 32 (40.0) 0.04
No 98 (49.2) 51 (42.9) 47 (58.8)
Been diagnosed with ADHD or
ADD
Yes 78 (39.0) 48 (40.3) 30 (37.5) 0.77
Author Manuscript

No 118 (59.0) 69 (58.0) 49 (61.3)

Been diagnosed with OCD
Yes 27 (13.5) 20 (16.8) 7 (8.8) 0.14
No 169 (84.5) 97 (81.5) 72 (90.0)
Been diagnosed with an eating
disorder
Yes 13 (6.5) 11 (9.2) 2 (2.5) 0.08
No 183 (91.5) 106 (89.1) 77 (96.3)
Been diagnosed with psychosis
Yes 11 (5.5) 9 (7.6) 2 (2.5) 0.20
No 185 (92.5) 108 (90.8) 77 (96.3)
Ever been hospitalized for a mental illness or depression
Yes 65 (32.7) 50 (42.0) 15 (18.8) <0.01
Author Manuscript

No 134 (67.3) 69 (58.0) 65 (81.3)

Sexually assaulted or abused before the age of 18
Yes 55 (27.6) 40 (33.6) 15 (18.8) 0.02
No 138 (69.3) 75 (63.0) 63 (78.8)

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 17

Characteristic Total n (%) n=199 Screened Positive for Screened Negative for p - value
Depressive Symptomology n Depressive Symptomology n
(%) n=119 (%) n=80
Author Manuscript

Hit or injured by a parent before the age of 18

Yes 73 (36.7) 55 (46.2) 18 (22.5) <0.01
No 119 (59.8) 60 (50.4) 59 (73.8)
Insulted or sworn at by a parent before the age of 18
Yes 137 (68.8) 94 (79.0) 43 (53.8) <0.01
No 58 (29.1) 22 (18.5) 36 (45.0)
While growing up lived with someone who had a mental illness
Yes 95 (47.7) 67 (56.3) 28 (35.0) <0.01
No 97 (48.7) 46 (38.7) 51 (63.8)
While growing up lived with someone who was using street drugs
Yes 101 (50.5) 64 (53.8) 37 (46.3) 0.18
No 90 (45.0) 48 (40.3) 42 (52.5)
Author Manuscript

While growing up lived with someone who went to jail or prison

Yes 86 (43.0) 52 (43.7) 34 (42.5) 0.77
No 106 (53.0) 61 (51.3) 45 (56.3)

Not all columns add to 100% due to missing values

Significance ascertained using Fisher’s exact test unless otherwise noted

a
Significance ascertained using the Wilcoxon rank sum test
Author Manuscript
Author Manuscript

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 18

Table 2:

Modified Poisson regression analysis of factors associated with depressive symptomology among young adult
Author Manuscript

non-medical prescription opioid users in Rhode Island (N = 199)

Characteristic Adjusted prevalence ratio (95% CI) p – value

Female vs. male sex at birth 1.24 (0.97–1.58) 0.08
Lesbian, gay, bisexual, queer, or something else vs. straight 1.23 (0.90–1.69) 0.19
Been homeless in the last six months (yes vs. no) 1.22 (0.97–1.53) 0.09
Monthly illicit polysubstance use in the last six months (yes vs. no) 1.20 (0.95–1.52) 0.13
Been diagnosed with a depressive disorder (yes vs. no) 1.51 (1.14–1.99) <0.01
Been diagnosed with a bipolar disorder (yes vs. no) 1.08 (0.86–1.35) 0.50
Been diagnosed with an anxiety disorder (yes vs. no) 0.78 (0.60–1.02) 0.07
Been hospitalized for a mental illness or depression (yes vs. no) 1.10 (0.88–1.38) 0.40
Hit or injured by a parent before the age of 18 (yes vs. no) 1.20 (0.94–1.53) 0.14
Author Manuscript

Insulted or sworn at by a parent before the age of 18 (yes vs. no) 1.50 (1.05–2.14) 0.02
Sexually assaulted or abused before the age of 18 (yes vs. no) 0.92 (0.69–1.22) 0.55
While growing up lived with someone with a mental illness (yes vs. no) 1.03 (0.78–1.36) 0.83
Author Manuscript
Author Manuscript

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.

 Bouvier et al. Page 19

Table 3:

Motivations for using prescription opioids non-medically associated with depressive symptomology among
Author Manuscript

young adult non-medical prescription opioid users in Rhode Island (N = 199)

Motivation for using prescription Total n (%) n=199 Screened Positive for Screened Negative for p - value
opioids non-medically Depressive Symptomology n Depressive Symptomology
(%) n=119 n (%) n=80

To feel good or get high

Yes 157 (78.9) 95 (79.8) 62 (77.5) 0.59
No 40 (20.1) 22 (18.5) 18 (22.5)
To feel less depressed or anxious
Yes 111 (55.8) 83 (69.7) 28 (35.0) <0.01
No 86 (43.2) 34 (28.6) 52 (65.0)
To relieve physical pain
Yes 150 (75.4) 91 (76.5) 59 (73.8) 0.61
Author Manuscript

No 47 (23.6) 26 (21.8) 21 (26.3)

To get a good sleep
Yes 107 (53.8) 66 (55.5) 41 (51.3) 0.56
No 90 (45.2) 51 (42.9) 39 (48.8)
To avoid withdrawal symptoms
Yes 82 (41.2) 56 (47.1) 26 (32.5) 0.04
No 115 (57.8) 61 (51.3) 54 (67.5)
To have a good time with friends
Yes 98 (49.2) 61 (51.3) 38 (47.5) 0.56
No 99 (49.7) 56 (47.1) 42 (52.5)
Because I was pressured into it
Yes 14 (7.0) 10 (8.4) 4 (5.0) 0.41
Author Manuscript

No 183 (92.0) 107 (89.9) 76 (95.0)

As a substitute when other drugs are not available
Yes 84 (42.2) 60 (50.4) 24 (30.0) <0.01
No 113 (56.8) 57 (47.9) 56 (70.0)

Not all columns add to 100% due to missing values

Significance ascertained using Fisher’s exact test

Author Manuscript

J Psychoactive Drugs. Author manuscript; available in PMC 2020 November 01.