Asian Journal of Psychiatry 6 (2013) 544–547

The efficacy of atomoxetine as adjunctive treatment for co-morbidsubstance use disorders and externalizing symptoms§

Vivek Benegal a, Biju Viswanath a,*, Janardhanan C. Narayanaswamy a, Sam P. Jose b,Vaskar Chakraborty a, Deepa Sankar a, Thennarasu Kandavel a, Muralidharan Kesavan a

a Department of Psychiatry, National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore 560029, Indiab Chalmeda Anadarao Institute of Medical Sciences, Karimnagar, Andhra Pradesh, India

A R T I C L E I N F O

Article history:

Received 17 August 2012

Received in revised form 22 July 2013

Accepted 24 July 2013

Keywords:

Atomoxetine

ADHD

Substance

Diathesis

Comorbid

Externalizing

A B S T R A C T

Background: We examined the effect of atomoxetine supplementation in treated-as-usual patients with

alcohol, tobacco and other drug dependence (ATOD) and co-morbid externalizing symptoms (ES).

Method: Subjects were selected from a substance dependence treatment-cohort and assessed for: (a)

high ES counts, (b) maximum prior period of abstinence, (c) quality of life during that period, and (d)

shortest time from prior relapse to restarting treatment. Subjects were prescribed atomoxetine and

followed up to their first relapse.

Results: Out of 262 subjects screened during the study period (March–April 2008), 18 subjects who

fulfilled eligibility criteria were recruited. All subjects were male, with early onset of substance

dependence to at least two substances. Atomoxetine treatment led to significant treatment benefits: ES

reduction, longer abstinence, shorter turnaround time and better quality of life.

Conclusions: Atomoxetine has a potential role in the treatment of early onset ATOD patients with ES, as

an adjuvant to the standard treatment.

� 2013 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Asian Journal of Psychiatry

jo u rn al h om epag e: ww w.els evier .c o m/lo cat e/a jp

1. Narrative

1.1. Background

Alcohol, tobacco and other drug dependence (ATOD), childhoodconduct disorders (CD), attention-deficit hyperactivity disorders(ADHD), and adult antisocial behavior commonly occur incombination (Carroll and Rounsaville, 1993; Biederman et al.,1998; Schubiner et al., 2000; Sullivan and Rudnik-Levin, 2001;Barkley et al., 2004; Knop et al., 2009). Various estimates havereported between 20% and 50% co-morbidity in ATOD and ADHD(Johann et al., 2003; Ohlmeier et al., 2008). Studies suggest thatcomorbid ATOD and ADHD forms a distinct clinical phenotypecharacterized by an increased severity of substance-relatedsymptoms and behavioral/emotional problems, longer course,and greater difficulty in achieving abstinence and to remain in

§ An earlier version of this paper was presented as a poster in the Research

Society of Alcoholism Annual Conference 2009 at San Antonio, Texas, USA. Author

BV received a junior investigator travel award to present the poster at the

conference.

* Corresponding author. Tel.:+91 80 26995250; fax: +91 80 26564822.

E-mail addresses: bijuv1@gmail.com, bijuv1@rediffmail.com (B. Viswanath).

1876-2018/$ – see front matter � 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.ajp.2013.07.007

treatment. ADHD is also associated with a higher risk of earlydevelopment of alcohol and other drug dependence (Wilens et al.,1997; Johnson et al., 2000; Johann et al., 2003; Ohlmeier et al.,2008).

Available evidence, from twin/family studies, electrophysio-logical and neuro-imaging studies, suggest that a shared geneticneurobiological diathesis, which manifests as a behavioral-temperamental trait, characterized by ADHD symptoms (inatten-tion, impulsivity), oppositional behavior and/or conduct problems,constitutes a vulnerability for ATOD, and has been called theexternalizing psychopathology (Silva et al., 2007; Dick et al., 2008).Intuitively, this evidence should make a strong case for concurrenttreatment of at least the ADHD symptoms in persons withcomorbid ATOD and externalizing psychopathology. Yet, there is apaucity of literature on such concurrent treatment strategies(Wilson and Levin, 2005). A meta-analysis of pharmacotherapy inADHD youth found a 1.9-fold reduction in risk for ATOD in thosetreated with stimulants, compared with youth who did not receivepharmacotherapy (Wilens et al., 2003), while another randomizedcontrol trial of stimulant treatment in persons with ATOD and co-morbid ADHD showed no significant difference between medica-tion and control groups (Schubiner et al., 2002).

Atomoxetine is a noradrenergic agent approved for thetreatment of children and adults with ADHD (Michelson et al.,

V. Benegal et al. / Asian Journal of Psychiatry 6 (2013) 544–547 545

2001, 2003). A selective inhibitor of presynaptic norepinephrinetransporter in the central nervous system, it was found to be moreefficacious than placebo in reducing ADHD symptoms, equallyefficacious in head-to-head trials with stimulants and had lowdiscontinuation rates even in doses up to 120 mg (Spencer et al.,1998; Michelson et al., 2003). Atomoxetine, unlike stimulants, isfree of abuse potential (Heil et al., 2002) and therefore, of specialinterest in comorbid ATOD (Wilens, 2004). Adverse effects includeappetite suppression, initial weight loss, initial tachycardia andincrease in blood pressure which stabilizes later, drowsiness,dizziness, light headedness and fainting (Unni, 2006; Adler, 2007).A 3-month double blind placebo controlled study of adult alcoholuse disorder with comorbid ADHD found that atomoxetinesignificantly reduced ADHD symptoms compared to placebo, butwith no significant differences between treatment groups in time-to-relapse of heavy drinking. However, cumulative heavy drinkingdays were reduced by 26% in the atomoxetine group (Wilens et al.,2008). In addition, ADHD score reductions were found to correlatewith decrease in alcohol craving; this was more notable withatomoxetine than placebo (Wilens et al., 2011). Open-label studiesfound treatment with atomoxetine to be of limited utility in thetreatment of cannabis and cocaine dependence (Tirado et al., 2008;Levin et al., 2009).

We sought to examine the effect of atomoxetine supplementa-tion in subjects with co-morbid ATOD and externalizing symptoms(ES), belonging to a long-term treatment cohort, currently intreatment (with pharmacological and behavioral interventions).We used a dimensional score for ES using the World HealthOrganization Adult ADHD Self-Report Scale (ASRS) (Kessler et al.,2005) rather than using a categorical diagnosis of ADHD.

2. Methodology

2.1. Sample

The study used a within-subject retrospective design withnaturalistic follow-up. Subjects were recruited from the outpatientservices of the Deaddiction Center, National Institute of MentalHealth and Neurosciences (NIMHANS), Bangalore, India, during themonths of March and April 2008. During this period, allconsecutive, who presented with a relapse during this time-window were screened, and included in the study if they: (1)satisfied dependence criteria for at least one substance (other thannicotine) according to the International Classification of Diseases,version 10 [ICD-10] (WHO, 1992), (2) had available records of pasttreatment and outcome in the database, (3) had score of 24 or moreon the World Health Organization Adult ADHD Self-Report Scale–version 1.1 (ASRS) (Kessler et al., 2005), (4) were never treated forADHD, (5) were above 18 years and, (6) did not have a history ofcomorbid psychosis, mood or anxiety disorders and seizures. Thestudy was conducted in accordance with the Declaration ofHelsinki and subjects were recruited after written informedconsent.

2.2. Methodology

Subjects were assessed using the following measures

(A) ASRS for the measurement of ES.(B) A life chart for each patient constructed from the treatment

records and corroborated with the subjects and significantfamily members. The previous period of maximum abstinence(from any substance other than nicotine) on the ongoingtreatment regime was identified and quantified. Nicotine wasuniversally used by all study subjects, and abstinence hadnever been achieved with past or current treatment. Also the

minimum turnaround time (defined as the minimum timerequired to re-engaging in treatment after any relapse on anyprevious treatment in the past 2 years) was quantified.

(C) The World Health Organization Quality of Life—WHOQOLBREFscale (WHO, 1998) was used with individual subjects to deriveretrospective information about quality of life during theabstinence period.

(D) Subjects were also asked to make a subjective evaluation ofeffectiveness of treatment on a visual analog scale of 0–10 fromnot effective at all to extremely effective.

(E) Side effects checklist—constructed from known symptoms andsigns of adverse effects of atomoxetine available in theliterature (Appendix A) was applied for subsequent monitoringduring out-patient visits.

Assessments were done during a brief (3 weeks) inpatientadmission after the subject was re-engaged in the treatmentprocess. After completion of the inpatient detoxification for controlof withdrawal symptoms (usually completed in 7–10 days), ratingson measures A–D were performed. This was followed by additionof atomoxetine to the treatment regime which was in use prior tothe current relapse (combination of long-term pharmacotherapyand relapse prevention therapy). All subjects were required tocome thereafter for out-patient monitoring visits, to facilitateincremental changes in the dosage of atomoxetine, up to amaximum of 50 mg/day, and assessment of emergent adverseeffects, rated on the side effects checklist. Subjects continued toreceive ‘treatment as usual’ (TAU), which consisted of pharmaco-logical treatments prescribed previously as well as relapseprevention counseling.

All measures were reapplied after the baseline measurementsduring the outpatient visit in the sixth week. The subjects werethen followed up every 2 months till their next relapse and re-engagement into treatment, or for a period of 1 year, whicheveroccurred earlier. Relapse was ascertained by interviewing thepatient and at least one family member living with the patient. Incase of doubt, urine testing was carried out to detect the substance.

2.3. Statistical analysis

The paired ‘t’-test was used to compare (1) the period ofabstinence with atomoxetine with past maximum period ofabstinence (on TAU without atomoxetine), (2) the shortestprevious turnaround time with current turnaround time whereapplicable, (3) the ES scores at baseline and 6 weeks, (4) theWHOQOL BREF scores during past abstinence with scores duringcurrent abstinence and (5) well as the subjective ratings ontreatment efficacy. A p value of <0.05 was considered statisticallysignificant.

3. Results

3.1. Sample characteristics (Table 1)

Of 262 subjects screened during the study period, 18 fulfilledeligibility criteria and were recruited after obtaining writteninformed consent. All subjects were male, with early onset (<25years) of substance dependence and were dependent on at leasttwo substances. In addition to having high ES scores, clinicalinterview by an experienced psychiatrist (VB) showed thatmajority (13/18, 72%) of them also had a DSM-IV diagnosis ofADHD. Clinical interviews also revealed that the ES were ofchildhood onset in all patients.

The mean dose of atomoxetine was 41.9 � 13.7 mg/day. Four ofthe subjects did not return after initial assessment during inpatientcare (no follow-ups) and hence, were excluded from the analysis of

Table 1Socio-demographic and clinical variables.

Variable Mean � SD N (%)

Age (years) 27.2 � 5.9

Number of years of education 13.4 � 3.1

Age at dependence (years) 17.9 � 3.5

Positive family history of ATOD (first degree) 13 (72.2)

Primary substance of abuse

Alcohol 8 (44.4)

Opioids 6 (33.3)

Cannabis 3 (16.7)

Solvents 1 (5.6)

Doses of previous drugs (in milligrams) and number of subjects who used them

Baclofen 46.3 � 5.2 7 (38.9)

Naltrexone 43.8 � 12.5 4 (22.2)

Buprenorphine 7.5 � 1.0 4 (22.2)

Disulfiram 500 � 0.0 3 (16.7)

V. Benegal et al. / Asian Journal of Psychiatry 6 (2013) 544–547546

study results. The mean duration of current follow up (N = 14) ontreatment with atomoxetine was 17 � 13.9 weeks (minimumduration was 7 weeks). The drugs used previously were continuedat the same doses.

3.2. Comparison of atomoxetine � TAU period with the past period of

TAU (Table 2)

ES scores reduced significantly post-atomoxetine between thefirst and second assessments. There were significant differences inthe maximum period of abstinence, turnaround time, quality of lifescores and the subjective rating of treatment efficacy scores.Another important finding was that the number of cigarettessmoked per day had also dropped significantly during this time.There was no difference between subjects with and without aclinical diagnosis of ADHD in terms of the maximum period ofabstinence (t = �0.63, p = 0.54).

3.3. Adverse effects

None of the patients reported any adverse effect requiringdiscontinuation of treatment with atomoxetine.

4. Discussion

We examined the effect of atomoxetine supplementation toTAU in subjects with co-morbid ATOD and ES. Subjects with add-on atomoxetine, had a better treatment outcome (increased periodof abstinence and decreased turnaround time). They also reporteda better subjective quality of life than with earlier treatment.Interestingly, nicotine use reduced significantly in subjects withadd-on atomoxetine.

Although it seems straightforward to assume that the improvedtreatment can be attributed to atomoxetine, many limitations ofthis study need to be considered before results are interpreted. This

Table 2Comparison of post atomoxetine period with the past period of well being.

Variable Mean � SD (TAU + atomox

Maximum period of abstinence (in days) 229.11 � 119.37

Turnaround time (in days) 53.91 � 51.90

ADHD scores 29.32 � 9.60

Quality of life domain 1 (physical) 28.67 � 3.76

Quality of life domain 2 (psychological health) 24.11 � 2.30

Quality of life domain 3 (social relationships) 11.33 � 2.03

Quality of life domain 4 (environment) 33.39 � 3.97

Subjective rating 4.56 � 1.95

Cigarette smoking (number per day) 11.00 � 13.34

study used a small sample with a within-subject retrospectivestudy design, which is inferior to the usual case-control design. Theselection of subjects was not subject to randomization procedures.The follow-up was not systematic. The assessment of patients’quality of life during previous treatment periods was doneretrospectively, and subject to recall bias. Another major limitationis that psychosocial factors which could have contributed to theabstinence/relapse were not assessed or controlled for. It is wellknown that psychosocial/lifestyle factors play a major role in theinitiation and maintenance of substance use. These factors couldhave had a more important role in subjects who already have ES.The ASRS measures only ADHD symptoms and does not measureoppositional/conduct symptoms which are also ES. Although it beargued that treatment options are available only for ADHDsymptoms, measurement of other ES (oppositional and conductsymptoms) was not performed. Personality disorders were also notassessed. In spite of these limitations, this preliminary study in anaturalistic real world setting using a dimensional measure of ESindicates the potential clinical utility of atomoxetine in difficult-to-treat patients with ATOD.

The externalizing behavior spectrum diathesis, underlying thevulnerability to ATOD, is hypothesized to be mediated by a state ofcentral nervous system hyperexcitability, due to disturbedhomeostasis between the inhibitory and excitatory brain systems(Begleiter and Porjesz, 1999), which in turn has been linked tohypofunction of both the dopamine and the norepinephrineneurotransmitter systems. Substance use initially raises theconcentration of dopamine in the brain reward circuit, temporarilyameliorating the ES, encouraging repeated use, but ultimatelyresulting in a dopamine deficit state, which characterizesdependence/addiction (Volkow et al., 2007). This leads to a viciouscycle, wherein the externalizing diathesis promotes substancedependence, and in turn gets worsened by it. We hypothesize thatatomoxetine, via noradrenergically mediated augmentation ofprefrontal cortical function, acts on this diathesis, leading toreduction in factors perpetuating substance use. We furtherspeculate that atomoxetine may have some direct effects onsubstance use through improving executive functioning. Atomox-etine has been shown to have cognition-enhancing effects inADHD, especially in improving response inhibition (Chamberlainet al., 2007), and could exert the same effect in reducing impulsesto use substances (Brown et al., 2007). In mice, atomoxetine hasbeen found to reverse nicotine withdrawal-associated deficits incontextual fear conditioning and thereby hypothesized to preventrelapse (Davis and Gould, 2007).

In contrast to a previous study (Wilens et al., 2008) whichobserved a reduction in the number of heavy drinking days but nodifferences in time to relapse in alcohol dependent subjects onatomoxetine alone compared to a placebo-control group; thecurrent study differed in using atomoxetine as a supplement totreatment-as-usual with pharmacological anti-craving agents andrelapse prevention therapy. This could account for the significantlyimproved outcome in time-to-relapse.

etine) Mean � SD (TAU only) t p

38.89 � 48.21 6.51 <0.001

603.55 � 614.51 �3.10 0.011

52.68 � 11.48 �12.91 <0.001

18.11 � 4.76 8.64 <0.001

14.50 � 3.40 11.28 <0.001

8.00 � 2.61 5.56 <0.001

21.78 � 6.22 9.33 <0.001

7.44 � 0.92 7.31 <0.001

20.18 � 12.81 �4.21 0.002

V. Benegal et al. / Asian Journal of Psychiatry 6 (2013) 544–547 547

5. Conclusion

Our results from a preliminary study indicate that atomoxetinehas a potential role in the treatment of early onset ATOD patientswith co-morbid ADHD, as an adjuvant to the standard treatmentsfor substance dependence. Since this population represents thegroup with more severe problems, greater recidivism and pooreroutcomes among the treatment seekers, these results, subject tovalidation in larger populations, holds promise for a change in thetreatment strategy for substance dependence.

Author disclosure

Contributors: Author BV and VB wrote the first draft of themanuscript. Authors BV, JCN, SPJ, DS and VC did the data collectionand edited the manuscript. Authors VB and MK managed theliterature search. Author TK was involved in statistical analysis. Allauthors contributed to and have approved the final manuscript.

Conflict of interest

None.

Appendix A. Checklist of side effects of atomoxetine

Appetite suppression

Drowsiness

Dizziness/light headedness

Fainting

Weight loss

Elevated blood pressure

Tachycardia

Heart burn

Stomach upset

Vomiting

Loss of appetite

Constipation

Dry mouth

Excessive tiredness

Initial insomnia or difficulty staying asleep

Headache

Fever

Chills

Muscle pain

Hot flushes

References

Adler, L.A., 2007. Non-stimulant trials of adult ADHD. CNS Spectr. 12 (4 Suppl.) 11–13.Barkley, R.A., Fischer, M., et al., 2004. Young adult follow-up of hyperactive

children: antisocial activities and drug use. J. Child Psychol. Psychiatry 45(2) 195–211.

Begleiter, H., Porjesz, B., 1999. What is inherited in the predisposition towardalcoholism? A proposed model. Alcohol. Clin. Exp. Res. 23 (7) 1125–1135.

Biederman, J., Wilens, T.E., et al., 1998. Does attention-deficit hyperactivity disorderimpact the developmental course of drug and alcohol abuse and dependence?Biol. Psychiatry 44 (4) 269–273.

Brown, T.E., Holdnack, J., et al., 2007. Atomoxetine Alleviates Executive FunctionImpairments in Adults with ADHD. American Psychiatric Association, SanDiego, CA.

Carroll, K.M., Rounsaville, B.J., 1993. History and significance of childhood attentiondeficit disorder in treatment-seeking cocaine abusers. Compr. Psychiatry 34 (2)75–82.

Chamberlain, S.R., Del Campo, N., et al., 2007. Atomoxetine improved responseinhibition in adults with attention deficit/hyperactivity disorder. Biol. Psychia-try 62 (9) 977–984.

Davis, J.A., Gould, T.J., 2007. Atomoxetine reverses nicotine withdrawal-associateddeficits in contextual fear conditioning. Neuropsychopharmacology 32 (9)2011–2019.

Dick, D.M., Aliev, F., et al., 2008. Using dimensional models of externalizingpsychopathology to aid in gene identification. Arch. Gen. Psychiatry 65 (3)310–318.

Heil, S.H., Holmes, H.W., et al., 2002. Comparison of the subjective, physiological,and psychomotor effects of atomoxetine and methylphenidate in light drugusers. Drug Alcohol Depend. 67 (2) 149–156.

Johann, M., Bobbe, G., et al., 2003. Comorbidity of alcohol dependence withattention-deficit hyperactivity disorder: differences in phenotype with in-creased severity of the substance disorder, but not in genotype (serotonintransporter and 5-hydroxytryptamine-2c receptor). Alcohol. Clin. Exp. Res.27 (10) 1527–1534.

Johnson, B.A., Cloninger, C.R., et al., 2000. Age of onset as a discriminator betweenalcoholic subtypes in a treatment-seeking outpatient population. Am. J. Addict.9 (1) 17–27.

Kessler, R.C., Adler, L., et al., 2005. The World Health Organization Adult ADHD Self-Report Scale (ASRS): a short screening scale for use in the general population.Psychol. Med. 35 (2) 245–256.

Knop, J., Penick, E.C., et al., 2009. Childhood ADHD and conduct disorder asindependent predictors of male alcohol dependence at age 40. J. Stud. AlcoholDrugs 70 (2) 169–177.

Levin, F.R., Mariani, J.J., et al., 2009. Atomoxetine treatment for cocaine abuse andadult attention-deficit hyperactivity disorder (ADHD): a preliminary open trial.J. Dual Diagn. 5 (1) 41–56.

Michelson, D., Adler, L., et al., 2003. Atomoxetine in adults with ADHD:two randomized, placebo-controlled studies. Biol. Psychiatry 53 (2)112–120.

Michelson, D., Faries, D., et al., 2001. Atomoxetine in the treatment of children andadolescents with attention-deficit/hyperactivity disorder: a randomized, pla-cebo-controlled, dose-response study. Pediatrics 108 (5) E83.

Ohlmeier, M.D., Peters, K., et al., 2008. Comorbidity of alcohol and substancedependence with attention-deficit/hyperactivity disorder (ADHD). Alcohol Al-cohol. 43 (3) 300–304.

Schubiner, H., Saules, K.K., et al., 2002. Double-blind placebo-controlled trial ofmethylphenidate in the treatment of adult ADHD patients with comorbidcocaine dependence. Exp. Clin. Psychopharmacol. 10 (3) 286–294.

Schubiner, H., Tzelepis, A., et al., 2000. Prevalence of attention-deficit/hyperactivitydisorder and conduct disorder among substance abusers. J. Clin. Psychiatry 61(4) 244–251.

Silva, M.C., Benegal, V., et al., 2007. Cognitive deficits in children of alcoholics: atrisk before the first sip! Indian J. Psychiatry 49, 182–188.

Spencer, T., Biederman, J., et al., 1998. Effectiveness and tolerability of tomoxetinein adults with attention deficit hyperactivity disorder. Am. J. Psychiatry 155 (5)693–695.

Sullivan, M.A., Rudnik-Levin, F., 2001. Attention deficit/hyperactivity disorder andsubstance abuse. Diagnostic and therapeutic considerations. Ann. N. Y. Acad.Sci. 931, 251–270.

Tirado, C.F., Goldman, M., et al., 2008. Atomoxetine for treatment of marijuanadependence: a report on the efficacy and high incidence of gastrointestinaladverse events in a pilot study. Drug Alcohol Depend. 94 (1–3) 254–257.

Unni, J.C., 2006. Atomoxetine. Indian Pediatr. 43 (7) 603–606.Volkow, N.D., Wang, G.J., et al., 2007. Depressed dopamine activity in caudate and

preliminary evidence of limbic involvement in adults with attention-deficit/hyperactivity disorder. Arch. Gen. Psychiatry 64 (8) 932–940.

WHO, 1992. The ICD-10 classification of mental and behavioral disorders; clinicaldescription and diagnostic guidelines. World Health Organization, Geneva.

WHO, 1998. Development of the World Health Organization WHOQOL-BREF qualityof life assessment. Psychol. Med. 28 (3) 551–558.

Wilens, T.E., 2004. Impact of ADHD and its treatment on substance abuse in adults. J.Clin. Psychiatry 65 (Suppl. 3) 38–45.

Wilens, T.E., Adler, L.A., et al., 2011. Correlates of alcohol use in adults withADHD and comorbid alcohol use disorders: exploratory analysis of aplacebo-controlled trial of atomoxetine. Curr. Med. Res. Opin. 27 (12) 2309–2320.

Wilens, T.E., Adler, L.A., et al., 2008. Atomoxetine treatment of adults with ADHDand comorbid alcohol use disorders. Drug Alcohol Depend. 96 (1–2) 145–154.

Wilens, T.E., Biederman, J., et al., 1997. Attention deficit hyperactivity disorder(ADHD) is associated with early onset substance use disorders. J. Nerv. Ment.Dis. 185 (8) 475–482.

Wilens, T.E., Faraone, S.V., et al., 2003. Does stimulant therapy of attention-deficit/hyperactivity disorder beget later substance abuse? A meta-analytic review ofthe literature. Pediatrics 111 (1) 179–185.

Wilson, J.J., Levin, F.R., 2005. Attention-deficit/hyperactivity disorder andearly-onset substance use disorders. J. Child Adolesc. Psychopharmacol. 15(5) 751–763.