Annotated bibliography and position paper on the correlation between mood and substance abuse disoreders. PLEASE READ ENTIRE ASSIGNMENT

S d

A a

b

a

A R R A A

K A A S M D B

1

u a G v e W s ( B a D

e T

0 h

Drug and Alcohol Dependence 133 (2013) 338– 343

Contents lists available at ScienceDirect

Drug and Alcohol Dependence

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / d r u g a l c d e p

ubstance use disorders increase the odds of subsequent mood isorders

ileen Kenneson a,∗, Jennifer S. Funderburk a,b, Stephen A. Maisto a,b

Center for Integrated Healthcare, Veterans Affairs Medical Center, Syracuse, NY 13210, USA Department of Psychology, Syracuse University, Syracuse, NY 13244, USA

r t i c l e i n f o

rticle history: eceived 17 January 2013 eceived in revised form 6 June 2013 ccepted 8 June 2013 vailable online 29 July 2013

eywords: lcohol abuse lcohol dependence ubstance use disorder ood disorders epression ipolar disorder

a b s t r a c t

Background: There is a well-known association between mood disorders and substance use disorders (SUD), but little research has been conducted on SUDs as risk factors for the development of subsequent mood disorders. Methods: We analyzed data from the National Comorbidity Survey Replication study. Diagnoses were determined using DSM-IV criteria. Odds ratios (aORs) of subsequently developing mood disorders were adjusted for age, sex and race/ethnicity. Results: Data from 5217 individuals were included (6.6% male; mean age 45.3 years; 72.6% White, 11.2% Black, 12.5% Hispanic and 3.7% other). Subsequent mood disorders developed in 26.4% of individuals with primary adolescent-onset SUD (12–17 years), 21.7% of those with SUD onset at 18–25 years, and 14.0% of those with SUD onset between the ages of 26 and 34 years. The mean lagtime between SUD onset and development of a mood disorder was about 11 years. Controlling for demographic variables, the aORs of developing a mood disorder in these three age groups were 2.44, 3.65, and 3.25. Substance dependence was associated with higher odds of mood disorders than was abuse. Among the specific mood disorders,

the increased odds of developing bipolar disorder were particularly high among individuals with drug dependence. Conclusions: Individuals with adolescent and young adult-onset SUD had increased odds of develop- ing a secondary mood disorder. This indicates that adolescents and young adults with SUD should be closely monitored for both positive and negative mood symptoms. SUD treatment and aftercare offer opportunities for the early identification of secondary mood disorders.

. Introduction

Individuals with mood disorders have high rates of substance se disorders (SUD; Conway et al., 2006; Grant et al., 2005) and re over-represented among people with SUD (Conner et al., 2009; rant et al., 2004; Lai and Huang, 2009). About 30–50% of indi- iduals with bipolar disorder (BPD) have a primary SUD (Fossey t al., 2006; Strakowski and DelBello, 2000; Strakowski et al., 2005; inokur et al., 1995), and 30–55% of individuals with major depres-

ive disorder (MDD) have a primary alcohol use disorder (AUD) Boschloo et al., 2012; Falk et al., 2008; Swendsen et al., 1998).

ecause the dual diagnosis of both SUD and a mood disorder is ssociated with very high rates of morbidity (Blanco et al., 2012; utta et al., 2007; Gao et al., 2008) and mortality (Oquendo et al.,

∗ Corresponding author at: Center for Integrated Healthcare, Mailstop 116C, Vet- rans Affairs Medical Center, 800 Irving Avenue, Syracuse, NY 13210, USA. el.: +1 315 425 4400; fax: +1 315 425 4871.

E-mail address: [email protected] (A. Kenneson).

376-8716/$ – see front matter. Published by Elsevier Ireland Ltd. ttp://dx.doi.org/10.1016/j.drugalcdep.2013.06.011

Published by Elsevier Ireland Ltd.

2010; Sublette et al., 2009; Waller et al., 1999), understanding the trajectory of mood disorders among individuals with SUD will help clinicians improve the quality of life among patients with this dual diagnosis.

Primary SUD might be a result of self-medication of prodro- mal mood disorder symptoms (Strakowski and DelBello, 2000) or may be a causal factor for mood disorders perhaps by triggering an underlying susceptibility (Fergusson et al., 2011; Strakowski and DelBello, 2000; Boschloo et al., 2012). Thus, a better understanding of the connection between primary SUD and the development of subsequent mood disorders may lead to opportunities for the early detection, intervention, and perhaps even prevention, of mood dis- orders among people with SUD.

Unfortunately, only a few studies have assessed primary SUD as a potential risk factor for the development of a subsequent mood disorder. In longitudinal studies of depression, nonmedical opioid

dependence was a predictor of subsequent MDD (adjusted Hazard Ratio = 5.2l; Martins et al., 2009); alcohol or illicit drug use in child- hood or adolescence predicted MDD in one’s late 20s (Brook et al., 2002); AUD in adolescence predicted depression in early adulthood

ohol D

( o e i R 2 S

s s h r p d i a a n e a d b m

o l R m y c d d t h a S s s

2

2

a t a S c m a a r p i u s

2

s O v a t

A. Kenneson et al. / Drug and Alc

OR = 2.3; Rohde et al., 2001); and binge drinking increased the risk f depressive symptoms five years later by about two-fold (Paljarvi t al., 2009). In studies of BPD, opioid dependence from nonmed- cal use increased risk of subsequent BPD type I (adjusted Hazard atio = 5.0; Martins et al., 2009), and adolescents with SUD were .8 times more likely to have BPD than were adolescents without UD (Wilens et al., 1997).

Although comparisons among studies are hindered by different tudy populations and outcome variables, the available research uggests that individuals with specific primary SUDs may be at igher risk of developing mood disorders. For instance, the higher isk of mood disorders associated with opioid dependence as com- ared to binge-drinking or combined SUD suggests that substance ependence may be a greater risk factor for mood disorders than

s substance abuse. Consistent with this, 34.0% of individuals with lifetime diagnosis of drug abuse had a lifetime mood disorder, s compared to 61.7% of individuals with drug dependence in a ational survey (Conway et al., 2006). On the other hand, Falk t al. (2008) found that alcohol abuse, but not dependence, was ssociated with an increased frequency of secondary mood disor- ers. Further study is needed to elucidate the potential relationship etween substance use disorder diagnosis and the development of ood disorders. To further explore the relationship between the development

f mood disorders among individuals with primary SUD, we ana- yzed data from the population-based National Comorbidity Survey eplication (NCS-R) study. The purpose of our analysis was to esti- ate risk of secondary mood disorders among adolescents and

oung adults with SUD, including the relationships between spe- ific types of primary SUDs and specific types of secondary mood isorders. We also assessed the frequency of alcohol and specific rug use prior to the onset of mood disorder in this popula- ion. Given the association between SUD and mood disorders, we ypothesized that (a) adolescents and young adults with SUD have

higher risk of subsequent mood disorders than do those without UD, and (b) the risk of mood disorders associated with primary ubstance dependence is greater than risk associated with sub- tance abuse.

. Methods

.1. Data source/participants

The National Comorbidity Survey Replication (NCS-R) study was cross-sectional study that collected data on symptoms of men- al disorders of 9282 individuals age 18 years and older from

nationally-representative population in the continental United tates (US) in 2001 through 2003. Part I of the interview was onducted on all participants and included a core diagnostic assess- ent. A second part asked about additional disorders and was

dministered to 5692 individuals who met the lifetime criteria for core disorder in Part I, as well as a probability subsample of other espondents. Additional details of the study procedures have been ublished elsewhere (Kessler and Ustun, 2004). After excluding

ndividuals with SUDs secondary to mood disorders and individ- als who developed SUDs after the age of 34 years, the final sample ize for our analysis was 5217.

.2. Procedures

The NCS-R DSM-IV diagnoses and ages of onset were based on elf-report of symptoms using Version 3.0 of the World Health

rganization’s (WHO) Composite International Diagnostic Inter- iew (CIDI), a fully-structured lay-administered interview (Kessler nd Ustun, 2004). A mood disorder diagnosis was defined as a life- ime diagnosis of MDD, dysthymia or BPD. With regard to BPD, we

ependence 133 (2013) 338– 343 339

included type I, type II and subthreshold. We included subthreshold in the bipolar category, because it more closely resembles bipolar disorder than MDD in outcome (Nusslock and Frank, 2011), and that other research has found that the risk of lifetime (Merikangas et al., 2007) and secondary (Kenneson et al., 2013) SUD is also increased among subthreshold bipolar disorder in addition to type I and type II. Subthreshold bipolar disorder was defined as (a) recurrent subthreshold hypomania and intercurrent MDE, or (b) recurrent hypomania without recurrent MDE, or (c) recurrent subthreshold hypomania without intercurrent MDE. Subthreshold hypomania was defined as meeting two or more criterion B symptoms and all other criteria for hypomania. Cases with plausible organic causes were excluded. Additional details regarding the categorization of bipolar disorder in the NCS-R study have been previously published (Merikangas et al., 2007).

SUD was defined as lifetime alcohol abuse or dependence, or drug (cocaine, cannabis, prescription, or other) abuse or dependence. The “other” category included sedatives, tranquilizers, stimulants, analgesics, inhalants, hallucinogens, and heroin. Among individuals with more than one of the above four lifetime SUD diag- noses, the one that occurred first was used to determine the individ- ual’s age of SUD onset. Primary SUD was defined as SUD onset occur- ring at an earlier age than mood disorder onset. Within the NCS-R dataset, age at onset is reported in years; therefore, we excluded individuals who had the same age of onset for both conditions (n = 64), because it was impossible to determine sequence of onset of the two conditions. We categorized individuals by age at SUD onset as adolescents (age 12–17 years), emerging adults (18–24 years) and young adults (25–34 years). We have chosen these age categories because SUDs among children are uncommon, and because as the age of SUD onset increases the percent of people with mood disorders that develop after, compared to before, the onset of SUD decreases. As a consequence, the sample sizes are small for people with childhood (n = 23) or older adult onset (n = 76) primary SUD and a subsequent mood disorder in this study population.

2.3. Analysis

All statistical analyses were conducted using Statistical Analy- sis Software (SAS) version 9.2. To account for the complex design structure and weighting of NCS-R, we used the survey applica- tions, which utilize the Taylor series linearization method (An, 2002). Chi-square analysis was used to compare the frequency of mood disorders between groups with different ages of SUD onset, with p-values of less than 0.0083 considered to be statis- tically significant (based on the Bonferroni method of correction for multiple comparisons). Logistic regression, controlling for cur- rent age, race/ethnicity (White, Black, Hispanic or Other) and sex, was used to determine adjusted odds ratios (aORs) of developing a subsequent mood disorder. The comparison group for each analy- sis was comprised of individuals who did not have lifetime SUD or mood disorder at the age in question. Thus, each comparison group included individuals who did not have a lifetime SUD at the time of the survey, as well as those that developed an SUD after the upper limit of the age group in that analysis (e.g., when analyzing the 12–17 year age group, the comparison group included individuals who develop an SUD at the age of 18 or later). Reported p-values are not corrected for multiple comparisons. We calculated 16 aORs per age group; therefore, aORs with p-values < 0.0015 were considered to be significant in this study.

2.4. Approvals

The study procedures were approved by the institiutional review board of the Syracuse Veterans Affairs Medical Center.

340 A. Kenneson et al. / Drug and Alcohol Dependence 133 (2013) 338– 343

Table 1 Frequency of SUD in entire sample and by mood disorder.

SUD Entire sample BPD MDD DYS Any mood disorder

Alcohol use disorder 10.0% 19.3% 13.6% 10.4% 14.5%

3

t y o i w 3 ( s D i

i t o A d y t o o a m

3

o p S h r w r ( o

Table 3 aORs of developing specific secondary mood disorders among individuals with pri- mary SUD (no mood disorder) compared to individuals without SUD or a mood disorder in the same age group.

Disorder adjOR 95% CI

12–17 years n = 455 Any mood disorder 2.438* 1.889–3.145

MDD 1.769* 1.288–2.430 Dysthymia 2.116 1.534–2.918 BPD 3.742* 2.550–5.490

18–25 years n = 505 Any mood disorder 3.653* 2.689–4.963

MDD 3.046* 2.234–4.154 Dysthymia 4.802* 3.359–6.866 BPD 4.174* 2.308–7.548

26–34 years n = 105 Any mood disorder 3.245* 2.085–5.049

MDD 3.009* 1.886–5.093 Dysthymia 3.343 1.125–9.927

T D

Drug use disorder 6.2% 15.0% 9.3% 7.1% 10.1% Any SUD 16.0% 24.2% 19.8% 16.6% 20.5%

. Results

Within the NCS-R sample, there were 1122 people with a life- ime diagnosis of SUD. Of these, 455 had adolescent-onset (12–17 ears) SUD, 505 had onset at age 18 through 25 years, and 105 had nset at age 26 through 34 years. The NCS-R also includes 4152 ndividals without a lifetime disgnosis of SUD. The study sample

as 46.6% male and 72.6% White, 11.2% Black, 12.5% Hispanic and .7% other. The mean age at the time of the survey was 45.3 years se = 0.47; range 18–98 years). The frequency of SUDs in the entire ample and by type of mood disorder are presented in Table 1. emographics of our sample by age at SUD onset are presented

n Table 2. Subsequent mood disorders developed in 26.4% (se = 2.23) of

ndividuals with primary adolescent-onset SUD, 21.7% (se = 2.45) of hose with SUD onset at 18 through 25 years, and 14.0% (se = 2.29) f those with SUD onset between the ages of 25 and 34 years. mong those without a lifetime diagnosis of SUD, 17.0% (se = 0.63) eveloped a lifetime diagnosis of a mood disorder. People in the oungest group had a significantly higher rate of mood disorders han did those in the 25–34 age group (p = 0.0034) and those with- ut SUD (p < 0.001). The mean lagtime between onset of SUD and nset of a mood disorder was 11.6 (se = 1.33) years, 11.2 (se = 1.00), nd 11.0 (se = 0.75) years for the three groups, respectively. The ean lagtimes did not differ significantly.

.1. Risk of developing secondary mood disorders

For each age at SUD onset group, we calculated aORs of devel- ping specific secondary mood disorders among individuals with rimary SUD compared to individuals who did not have a lifetime UD or mood disorder at that age (see Table 3). Individuals who ad primary onset of SUD in adolescence (12–17) had increased isks of developing BPD (aOR = 3.742) or MDD (aOR = 1.769). Those

ith onset of SUD at the age of 18–25 years also had increased

isk of mood disorders, which was more pronounced for dysthymia aOR = 4.802) and BPD (aOR = 4.174) than MDD (aOR = 3.046). On the ther hand, those with a later age at SUD onset (26–34 years) were

able 2 emographics by age at primary SUD onset.

Age at onset of SUD N Age at time of surv mean (se)

12–17 years 455 34.5 years (0.57)

18–25 years 505 41.5 years (0.74)

26–34 years 105 47.7 years (1.29)

None 4152 46.4 years (0.535)

BPD 2.607 0.688–9.879

* p-Value ≤ 0.0001.

not at an increased risk of BPD or dysthymia, although they did have an increased risk of MDD (aOR = 3.009).

3.2. Risks associated with specific types of SUD

We next examined the effect of specific types of SUD on the risk of developing a subsequent mood disorder (see Table 4). Among those with adolescent-onset SUD, either alcohol or drug abuse or dependence was related to an increased risk for BPD, with aORs ranging from 3.888 for alcohol abuse to 10.098 for alcohol depend- ence. Increased risk of BPD was also associated with SUD in the 18–25 year age of onset group, with the highest aOR (13.794) due to drug dependence. MDD was not associated with specific SUDs in the 12–17 year age group, but was associated with alcohol abuse, drug abuse and drug dependence in the 18–24 year group, and with alcohol abuse or dependence in the 26–34 age group. Dys- thymia was associated with SUDs only in the 18–25 year group, with aOR ranging from 3.598 for drug abuse to 7.833 for drug dependence.

3.3. Alcohol and drug use prior to development of secondary

mood disorder

Finally, we assessed the frequency of alcohol and specific drug use among dually-diagnosed individuals prior to the development

ey Male % (se)

Race/ethnicity % (se)

69.4% (2.98)

76.0% (2.61) White 14.9% (2.23) Black 5.2% (1.42) Hispanic 3.9% (1.06) Other

72.1% (1.95)

77.4% (2.61) White 10.7% (1.85) Black 8.2% (1.33) Hispanic 3.7% (1.17) Other

66.2% (4.71)

69.6% (5.33) White 12.0% (3.41) Black 17.1 (3.13) Hispanic 1.3% (0.70) Other

42.2% (1.26)

71.9% (1.94) White 11.0% (1.27) Black 13.4% (1.16) Hispanic 3.7% (0.46) Other

A. Kenneson et al. / Drug and Alcohol Dependence 133 (2013) 338– 343 341

Table 4 SUD-specific aORs of developing subsequent mood disorders among individuals with primary SUD.

Type of SUD Any Mood MDD DYS BPD

SUD with onset at 12–17 years Alcohol abuse 2.353*

(1.591–3.479) 1.475 (0.900–2.418)

1.679 (0.926–3.044)

4.299*

(2.924–6.321) Alcohol dependence 4.571*

(2.396–8.516) 1.774 (0.934–3.369)

1.911 (0.530–6.896)

10.098*

(3.791–24.223) Drug abuse 2.225*

(1.662–2.979) 1.420 (0.970–2.079)

1.578 (0.726–3.431)

3.888*

(2.360–6.405) Drug dependence 2.460

(1.168–5.181) 1.246 (0.660–2.355)

1.292 (0.388–4.302)

6.483*

(3.148–13.350)

SUD with onset at 18–25 years Alcohol abuse 4.305*

(3.199–5.794) 3.172*

(2.277–4.418) 4.726*

(2.994–7.461) 5.950*

(3.240–10.926) Alcohol dependence 2.840**

(1.566–5.150) 1.824 (0.988–3.366)

5.937*

(2.763–12.758) 6.279*

(3.392–11.624) Drug abuse 4.643*

(3.123–6.904) 3.692*

(2.375–5.741) 3.598*

(3.785–8.575) 4.933*

(2.794–8.712) Drug dependence 9.491*

(4.454–20.225) 4.627*

(1.827–11.719) 7.833*

(3.903–15.705) 13.794*

(7.002–27.171)

SUD with onset at 26–34 years Alcohol abuse 2.603**

(1.450–4.671) 2.379 (1.217–4.650)

1.617 (0.362–7.217)

2.177 (0.510–9.293)

Alcohol dependence 4.524**

(1.993–10.268) 3.402**

(1.420–8.148) 2.041 (0.278–14.988)

3.548 (0.751–16.760)

Drug abuse 5.472**

(2.047–14.626) 4.394 (1.704–11.330)

11.201 (02.337–53.689)

9.117*

(2.934–28.328) Drug dependence 1.422

(0.298–6.787) 0.598 (0.080–4.500)

5.822 (0.687–49.366)

5.478 (0.723–41.493)

o o u o s t m 1 C m a

4

f s t r i T o i d d a p p t t t

d a f O

* p-Value ≤ 0.0001. ** p-Value > 0.0001 and ≤0.0015.

f the secondary mood disorder. Among those with adolescent- nset primary SUD, 100% drank alcohol, 92.2% used cannabis, 33.4% sed cocaine, and 55.6% used another drug sometime before the nset of the secondary mood disorder. Likewise, alcohol was con- umed before the onset of a mood disorder by 100% of those in he other two age-at-onset groups. Cocaine use prior to develop-

ent of a secondary mood disorder was more common among the 8–25 year (66.9%) and 26–34 year (47.9%) age at SUD onset groups. annabis and other drug use was less common among those in the iddle group (87.8% and 29.9%, respectively) and the 26–34 year

ge group (73.7% and 16.9%) than in the adolescent onset group.

. Discussion

Little research has been conducted on primary SUD as a risk actor for the development of secondary mood disorder. In this tudy, we found that people who develop SUD at the ages of 12 hrough 34 years have a frequency of subsequent mood disorders, anging from 26% in individuals with adolescent-onset SUD to 14% n individuals with SUD onset between the ages of 25 and 34 years. he explanation of the finding of a decreasing frequency of sec- ndary mood disorder with age probably is that as age at SUD onset ncreases, the chance of already having developed a primary mood isorder increases, consequently reducing the proportion of dually- iagnosed individuals with secondary mood disorders. This would lso explain the lower frequency of subsequent mood disorders in eople who develop SUDs at age 25 through 34 years than among eople without a lifetime diagnosis of SUD. Consequently, moni- oring of mood symptoms in individuals with a SUD, particularly hose in the younger age groups, may lead to early recognition of he onset of mood disorders.

As we predicted, individuals with a SUD had increased odds of

eveloping a mood disorder. After adjusting for sex, race, ethnicity nd current age, individuals in all three age groups had about a two- old increase in odds of developing a subsequent mood disorder. ur results are consistent with those of other studies, in which risk

of mood symptoms or mood disorders ranged from about 2 to 5 fold, depending on the variables used for alcohol or drug use and mood. Furthermore, in our analysis, substance dependence was associated with higher odds of secondary mood disorders than was substance abuse. This severity effect supports the hypothesis that substance use may trigger the development of mood disorders, although it is also possible that SUD is a prodromal symptom of a mood disor- der or is the result of a common underlying susceptibility of some people to both SUDs and mood disorders. The odds of developing BPD were particularly high, especially among those with alcohol or other substance dependence, compared to other mood disorders. These results indicate that adolescents and young adults with sub- stance use disorders should be closely monitored for both positive and negative mood symptoms.

The association between SUDs and subsequent mood disorders was strongest among the group of individuals with onset of SUD at the age of 18–25 years. Individuals who develop primary SUD at age 26–34 years do not appear to have as high of a risk for sec- ondary mood disorder, with the exception of those with drug abuse (aOR = 9.117) for BPD. The risk of BPD was also high among people with adolescent-onset SUD; in fact, the risk of any mood disorders appears to be primarily due to the risk for BPD.

There are several possible reasons for higher odds of subsequent mood disorders among emerging adults compared to adolescents or adults with a SUD. First, individuals who develop primary SUD at different ages have different frequencies of marijuana, cocaine and other drug use. Second, there may be differences in etiology of early and late onset BPD, which might account for fact that SUD onset among those ages 26–34 does not coincide with increased risk of BPD. Third, later age at onset of SUD may be due to differ- ent etiology, which may also affect its role in subsequent mood disorders.

The limitations of this analysis are primarily related to the data collection instruments. The CIDI protocol collects data based on recall and calculates age at onset retrospectively. Although this approach is limited by recall bias, test-retest reliability of age at

3 ohol D

o s ( fi a w m i a o

d n r e b t o h t a

a a a m o a m a n q p

R

H l r

C

m i o r t

C

R

A

B

B

42 A. Kenneson et al. / Drug and Alc

nset in the CIDI were high (ICC between 0.70 and 0.80) for depres- ive episodes and dysthymia (Barkow et al., 2002). Chengappa et al. 2003) cross-checked responses to different questions about age of rst mood episode in two separate parts of a SCID-based interview nd reported that recall bias in BPD is limited to three to six months, ell within the one-year time frame that we used to distinguish pri- ary and secondary mood disorders. Recall bias may increase with

ncreasing age. This may have contributed to the observation that ge of onset of primary SUD increased with the mean age at time f the interview.

In addition, when the analysis focused on specific mood disor- ers or specific types of SUD, the sample sizes became small, so on-significant results are not necessarily evidence of no increased isk. Finally, it is difficult to diagnosis mood disorders in the pres- nce of SUDs. This may explain the lack of significant association etween substance dependence and subsequent mood disorders in he 26–34 year age group. Also, we cannot rule-out the inclusion f individuals with substance-induced mood disorders, which may ave affected the results. However, we minimized the chance of heir inclusion by excluding cases with onset reported at the same ge.

The results of this study have implications for the clinical care of dolescents and young adults with SUDs. The SUD treatment and ftercare offer opportunities for the early identification, assessment nd treatment of secondary mood disorders. In addition, under the odel that SUDs may trigger underlying susceptibility to mood dis-

rders, it is possible that early treatment of adolescents and young dults with SUD might decrease the odds of developing subsequent ood disorders. Further research on the relationship between SUD

nd subsequent mood disorders has the potential to improve prog- osis for adolescents and young adults with SUDs, and thus improve uality of life and decrease morbidity and mortality among this opulation.

ole of funding source

Funding for this study was provided by the Center for Integrated ealthcare. The CIH had no further role in study design; in the col-

ection, analysis and interpretation of data; in the writing of the eport; or in the decision to submit the paper for publication.

ontributors

Author AKA designed the study, wrote the first draft of the anuscript and completed statistical analyses. JSF and SAM were

nvolved in conceptualization of the study and the completion f the final manuscript. All authors contributed to addressing eviewer comments. All authors contributed to and have approved he final manuscript.

onflict of interest

All authors declare that they have no conflicts of interest.

eferences

n, A.B.,2002. Performing logistic regression on survey data with the new survey- logistic procedure. In: Proceedings of the Twenty-Seventh Annual SAS® Users Group International Conference 2002, (paper 258–327). SAS Institute Inc., Cary, NC.

arkow, K., Heun, R., Ulstun, T.B., 2002. Test-retest reliability of self-reported age at onset of selected psychiatric diagnoses in general health care. Acta Psychiatry Scand. 106, 117–125.

lanco, C., Alegría, A.A., Liu, S.M., Secades-Villa, R., Sugaya, L., Davies, C., Nunes, E.V., 2012. Differences among major depressive disorder with and without co- occurring substance use disorders and substance-induced depressive disorder: results from the National Epidemiologic Survey on Alcohol and Related Condi- tions. J. Clin. Psychiatry 73, 865–873.

ependence 133 (2013) 338– 343

Boschloo, L., Vogelzangs, N., van den Brink, W., Smit, J.H., Veltman, D.J., Beekman, A.T., Penninx, B.W., 2012. Alcohol use disorders and the course of depressive and anxiety disorders. Br. J. Psychiatry 200, 476–484.

Brook, D.W., Zhang, C., Cohen, P., Whiteman, M., 2002. Drug use and the risk of major depressive disorder, alcohol dependence, and substance use disorders. Arch. Gen. Psychiatry 59, 1039–1044.

Chengappa, K.N., Kupfer, D.J., Frank, E., 2003. Relationship of a birth cohort and early age at onset of illness in a bipolar disorder case registry. Am. J. Psychiatry 160, 1636–1642.

Conner, K.R., Pinquart, M., Gamble, S.A., 2009. Meta-analysis of depression and sub- stance use among individuals with alcohol use disorders. J. Subst. Abuse Treat. 37, 127–137.

Conway, K.P., Compton, W., Stinson, F.S., Grant, B.F., 2006. Lifetime comorbidity of DSM-IV mood and anxiety disorders and specific drug use disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. J. Clin. Psychiatry 67, 247–257.

Dutta, R., Boydell, J., Kennedy, N., Van Os, J., Fearon, P., Murray, R.M., 2007. Suicide and other causes of mortality in bipolar disorder: a longitudinal study. Psychol. Med. 37, 839–847.

Falk, D.E., Yi, H.-Y., Hilton, M.E., 2008. Age of onset and temporal sequencing of lifetime DSM-IV alcohol use disorders relative to comorbid mood and anxiety disorders. Drug Alcohol Depend. 94, 234–245.

Fergusson, D.M., Boden, J.M., Horwood, L.F., 2011. Structural models of the comorbidity of internalizing disorders and substance use disorders in a longitudinal birth cohort. Soc. Psychiatry Psychiatr. Epidemiol. 46, 933–942.

Fossey, M.F., Otto, M.W., Yates, W.R., Wisniewski, S.R., Gyulai, L., Allen, M.H., Mik- lowitz, D.J., Coon, K.A., Ostacher, M.J., Neel, J.L., Thase, M.E., Sachs, G.S., Roger, D., Weiss, R.D., for STEP-BD Investigators, 2006. Validity of the distinction between primary and secondary substance use disorder in patients with bipo- lar disorder: data from the first 1000 STEP-BD participants. Am. J. Addict., 138–143.

Gao, K., Verduin, M.L., Kemp, D.E., Tolliver, B.K., Ganocy, S.J., Elhaj, O., Bilali, S., Brady, K.T., Findling, R.L., Calabrese, J.R., 2008. Clinical correlates of patients with rapid- cycling bipolar disorder and a recent history of substance use disorder: a subtype comparison from baseline data of 2 randomized, placebo-controlled trials. J. Clin. Psychiatry 69, 1057–1063.

Grant, B.F., Stinson, F.S., Dawson, D.A., Chou, P., Dufour, M.C., Compton, W., Pickering, R.P., Kaplan, K., 2004. Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders: results from the National Epi- demiologic Survey on Alcohol and Related Conditions. Arch. Gen. Psychiatry 61, 807–816.

Grant, B.F., Stinson, F.S., Hasin, D.S., Dawson, D.A., Chou, P., Ruan, W.J., Huang, B., 2005. Prevalence, correlates, and comorbidity of bipolar I disorder and axis I and II disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. J. Clin. Psychiatry 66, 1205–1215.

Kenneson, A., Funderburk, J.S., Maisto, S.A., 2013. Risk factors for secondary sub- stance use disorders in people with childhood and adolescent-onset bipolar disorder: opportunities for prevention. Compr. Psychiatry, 439–446.

Kessler, R.C., Ustun, T.B., 2004. The World Mental Health (WMH) survey initiative version of the World Health Organization (WHO) Composite International Diag- nostic Interview (CIDI). Int. J. Methods Psychiatry Res. 13, 93–121.

Lai, H.M., Huang, Q.R., 2009. Comorbidity of mental disorders and alcohol- and drug- use disorders: analysis of New South Wales inpatient data. Drug Alcohol Rev. 28, 235–242.

Martins, S.S., Keyes, K.M., Storr, C.L., Zhu, H., Chilcoat, H.D., 2009. Pathways between nonmedical opioid use/dependence and psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Drug Alcohol Depend. 103, 16–24.

Merikangas, K.R., Akiskal, H.S., Angst, J., Greenberg, P.E., Hirschfeld, R.M.A., Petukhova, M., Kessler, R.C., 2007. Lifetime and 12-month prevalence of bipo- lar spectrum disorder in the national comorbidity survey replication. Arch. Gen. Psychiatry 64, 543–552.

Nusslock, R., Frank, E., 2011. Subthreshold bipolarity: diagnostic issues and chal- lenges. Bipolar Disord. 13, 587–603.

Oquendo, M.A., Currier, D., Liu, S., Kim, J., Choi, J.S., Blanco, C., 2010. Increased risk for suicidal behavior in comorbid bipolar disorder and alcohol use disorders. J. Clin. Psychiatry 71, 902–909.

Paljarvi, T., Koskenvuo, M., Poikolainen, K., Kauhanen, J., Sillanmaki, L., Makela, P., 2009. Binge drinking and depressive symptoms: a 5-year population-based cohort study. Addiction 104, 1168–1178.

Rohde, P., Lewinsohn, P.M., Kahler, C.W., Seeley, J.R., Brown, R.A., 2001. Natural course of alcohol use disorders from adolescence to young adulthood. J. Am. Acad. Child Adolesc. Psychiatry 40, 83–90.

Strakowski, S.M., DelBello, M.P., 2000. The co-occurrence of bipolar and substance use disorders. Clin. Psychol. Rev. 20, 191–206.

Strakowski, S.M., DelBello, M.P., Fleck, D.E., Adler, C.M., Anthenelli, R.M., Keck Jr., P.E., Arnold, L.M., Amicone, J., 2005. Effects of co-occurring alcohol abuse on the course of bipolar disorder following a first hospitalization for mania. Arch. Gen. Psychiatry, 851–858.

Sublette, M., Carballo, J.J., Moreno, C., Galfalvy, H.C., Brent, D.A., Birmaher, B., John

Mann, J., Oquendo, M.A., 2009. Substance use disorders and suicide attempts in bipolar subtypes. J. Psychiatry Res. 43, 230–238.

Swendsen, J.D., Merikangas, K.R., Canino, G.J., Kessler, R.C., Rubio-Stipec, M., Angst, J., 1998. The comorbidity of alcoholism with anxiety and depressive disorders in four geographic communities. Compr. Psychiatry 39, 176–184.

ohol D

W

W

A. Kenneson et al. / Drug and Alc

aller, S.J., Lyons, J.S., Constantini-Ferrando, M.F., 1999. Impact of comorbid affec- tive and alcohol use disorders on suicide ideation and attempts. J. Clin. Psychol. 55, 585–595.

ilens, T.E., Biederman, J., Abrantes, A.M., Spencer, T.J., 1997. Clinical characteristics of psychiatrically referred adolescent outpatients with

ependence 133 (2013) 338– 343 343

substance use disorder. J. Am. Acad. Child Adolesc. Psychiatry 36, 941–947.

Winokur, G., Coryell, W., Akiskal, H.S., Maser, J.D., Keller, M.B., Endicott, J., Mueller, T., 1995. Alcoholism in Manic-depressive (bipolar) illness: Familial illness, course of illness, and the primary-secondary distinction. Am. J. Psychiatry, 365–372.